From c9e8504e483edd8534083e2343e8d45328dfb99a Mon Sep 17 00:00:00 2001 From: wangxc2006 <68397007@qq.com> Date: Sun, 9 Oct 2022 15:14:24 +0800 Subject: [PATCH 01/29] test --- .../imagenet_example/vscode_launch_file.txt | 43 +++++++++++++++++++ 1 file changed, 43 insertions(+) create mode 100644 application/imagenet_example/vscode_launch_file.txt diff --git a/application/imagenet_example/vscode_launch_file.txt b/application/imagenet_example/vscode_launch_file.txt new file mode 100644 index 00000000..ca45a684 --- /dev/null +++ b/application/imagenet_example/vscode_launch_file.txt @@ -0,0 +1,43 @@ +{ + "version": "0.2.0", + "configurations": [ + { + "name": "Python: main.py", + "type": "python", + "request": "launch", + // "preLaunchTask": "clear", //??? + //"program": "${file}", + "program": "imagenet_example/main.py", + // "program": "application/imagenet_example/PTQ/ptq/ptq.py", + "console": "integratedTerminal", + "justMyCode": false, + // "env": {"CUDA_VISIBLE_DEVICES":"0,1,2,3"}, + "args": [ + // "--arch=shufflenet_v2_x0_5", "--batch-size=320", + // "--arch=mobilenet_v2", "--batch-size=64", + // "--arch=resnet18", "--batch-size=128", + // "--arch=vgg11_bn", "--batch-size=32", + // "--arch=resnet50", "--batch-size=32", + // "--arch=squeezenet1_1", "--batch-size=128", + "--arch=mobilenet_v3_small", "--batch-size=128", + "--epochs=1","--lr=1e-4", + "--gpu=0", + "--pretrained", + "--evaluate", + // "--resume=tmp_path1/squeezenet1_1/checkpoint.pth.tar_best", + "--output_path=tmp_test", + // "--output_path=/workspace/tmp_path_1008", + "--fast_test", + "--backend=sophgo_tpu", + "--optim=sgd", + // "--backend=tensorrt_nlp", + // "--deploy", + // "--cpu", + // "--pre_eval_and_export", + "--train_data=/data/imagenet/for_train_val/", + "--val_data=/data/imagenet/for_train_val/" + ] + // "args": ["--config=application/imagenet_example/PTQ/configs/qdrop/r18_2_4.yaml"] + } + ] +} \ No newline at end of file From 0491be5a83f3b6793059199530607d786ea76729 Mon Sep 17 00:00:00 2001 From: wangxc2006 <68397007@qq.com> Date: Sun, 9 Oct 2022 15:22:55 +0800 Subject: [PATCH 02/29] support sophgo_tpu backend,initial ver --- application/imagenet_example/main.py | 176 +++++-- mqbench/convert_deploy.py | 26 +- mqbench/custom_quantizer/__init__.py | 3 +- mqbench/custom_quantizer/model_quantizer.py | 27 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 74 +++ mqbench/deploy/deploy_linear.py | 44 +- mqbench/fusion_method.py | 45 +- mqbench/nn/intrinsic/qat/modules/__init__.py | 2 + .../qat/modules/conv_fused_sophgo_tpu.py | 458 ++++++++++++++++++ .../qat/modules/linear_fused_sophgo_tpu.py | 294 +++++++++++ mqbench/nn/qat/modules/__init__.py | 2 +- mqbench/nn/qat/modules/conv.py | 19 + mqbench/prepare_by_platform.py | 82 +++- 13 files changed, 1201 insertions(+), 51 deletions(-) create mode 100644 mqbench/custom_quantizer/sophgo_tpu_quantizer.py create mode 100644 mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py create mode 100644 mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index 30349712..f233e806 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -4,6 +4,8 @@ import shutil import time import warnings +import numpy as np +import copy import torch import torch.nn as nn @@ -17,13 +19,14 @@ import torchvision.transforms as transforms import torchvision.datasets as datasets import torchvision.models as models -from mqbench.convert_deploy import convert_deploy +from mqbench.convert_deploy import convert_deploy, convert_onnx from mqbench.prepare_by_platform import prepare_by_platform, BackendType from mqbench.utils.state import enable_calibration, enable_quantization, disable_all model_names = sorted(name for name in models.__dict__ if name.islower() and not name.startswith("__") and callable(models.__dict__[name])) +cali_batch_size = 10 parser = argparse.ArgumentParser(description='PyTorch ImageNet Training') parser.add_argument('--train_data', metavar='DIR', @@ -80,22 +83,33 @@ 'multi node data parallel training') parser.add_argument('--model_path', type=str, default=None) -parser.add_argument('--backend', type=str, choices=['tengine_u8', 'tensorrt', 'nnie', 'ppl', 'snpe'], default='tensorrt') +parser.add_argument('--output_path', type=str, default=None) +parser.add_argument('--backend', type=str, choices=['tengine_u8', 'tensorrt', 'nnie', 'ppl', 'snpe', 'sophgo_tpu', 'openvino', 'tensorrt_nlp'], default='sophgo_tpu') parser.add_argument('--optim', type=str, default='sgd') parser.add_argument('--not-quant', action='store_true') parser.add_argument('--deploy', action='store_true') +parser.add_argument('--fast_test', action='store_true') +parser.add_argument('--cpu', action='store_true') +parser.add_argument('--pre_eval_and_export', action='store_true') BackendMap = {'tensorrt': BackendType.Tensorrt, 'nnie': BackendType.NNIE, + 'tensorrt_nlp': BackendType.Tensorrt_NLP, 'ppl': BackendType.PPLW8A16, + 'openvino': BackendType.OPENVINO, 'snpe': BackendType.SNPE, 'vitis': BackendType.Vitis, + 'sophgo_tpu': BackendType.Sophgo_TPU, 'tengine_u8': BackendType.Tengine_u8} best_acc1 = 0 def main(): args = parser.parse_args() + if args.output_path is None: + args.output_path = './' + args.output_path=os.path.join(args.output_path, args.arch) + os.system('mkdir -p {}'.format(args.output_path)) args.quant = not args.not_quant args.backend = BackendMap[args.backend] @@ -130,6 +144,18 @@ def main(): # Simply call main_worker function main_worker(args.gpu, ngpus_per_node, args) +layer_names = [] +features_out_hook = {} +i = 0 +def hook(module, fea_in, fea_out): + global i + if i >= len(layer_names): + return None + name = layer_names[i] + i += 1 + global features_out_hook + features_out_hook[name] = fea_out.cpu().numpy() + return None def main_worker(gpu, ngpus_per_node, args): global best_acc1 @@ -155,14 +181,38 @@ def main_worker(gpu, ngpus_per_node, args): else: print("=> creating model '{}'".format(args.arch)) model = models.__dict__[args.arch]() + print('ori module:', model) # for internal cluster if args.model_path: state_dict = torch.load(args.model_path) print(f'load pretrained checkpoint from: {args.model_path}') model.load_state_dict(state_dict) + train_loader, train_sampler, val_loader, cali_loader = prepare_dataloader(args) + criterion = nn.CrossEntropyLoss().cuda(args.gpu) + if args.gpu is not None: + model = model.cuda(args.gpu) + else: + model = model.cpu() + if args.pre_eval_and_export: + validate(val_loader, model.eval(), criterion, args) + kwargs = { + 'input_shape_dict': {'data': [cali_batch_size, 3, 224, 224]}, + 'output_path': args.output_path, + 'model_name': args.arch, + 'dummy_input': None, + 'onnx_model_path': os.path.join(args.output_path, '{}_ori.onnx'.format(args.arch)), + } + module_tmp = copy.deepcopy(model) + module_tmp = module_tmp.cpu() + convert_onnx(module_tmp.eval(), **kwargs) + del module_tmp + model = model.train() # quantize model if args.quant: - model = prepare_by_platform(model, args.backend) + prepare_custom_config_dict= { + } + model = prepare_by_platform(model, args.backend, prepare_custom_config_dict) + print('prepared module:', model) if not torch.cuda.is_available(): print('using CPU, this will be slow') elif args.distributed: @@ -192,10 +242,12 @@ def main_worker(gpu, ngpus_per_node, args): model.features = torch.nn.DataParallel(model.features) model.cuda() else: - model = torch.nn.DataParallel(model).cuda() + if args.cpu: + model = model.cpu() + else: + model = model.cuda() # define loss function (criterion) and optimizer - criterion = nn.CrossEntropyLoss().cuda(args.gpu) if args.optim == 'sgd': optimizer = torch.optim.SGD(model.parameters(), args.lr, momentum=args.momentum, @@ -206,8 +258,12 @@ def main_worker(gpu, ngpus_per_node, args): weight_decay=args.weight_decay, amsgrad=False) - # prepare dataset - train_loader, train_sampler, val_loader, cali_loader = prepare_dataloader(args) + if args.quant and not args.cpu: + enable_calibration(model) + calibrate(cali_loader, model, args) + cudnn.benchmark = True + if args.quant: + enable_quantization(model) # optionally resume from a checkpoint if args.resume: @@ -218,6 +274,8 @@ def main_worker(gpu, ngpus_per_node, args): else: # Map model to be loaded to specified single gpu. loc = 'cuda:{}'.format(args.gpu) + if args.cpu: + loc = 'cpu' checkpoint = torch.load(args.resume, map_location=loc) args.start_epoch = checkpoint['epoch'] best_acc1 = checkpoint['best_acc1'] @@ -237,26 +295,23 @@ def main_worker(gpu, ngpus_per_node, args): .format(args.resume, checkpoint['epoch'], best_acc1)) else: print("=> no checkpoint found at '{}'".format(args.resume)) - elif args.quant: - enable_calibration(model) - calibrate(cali_loader, model, args) - cudnn.benchmark = True - if args.quant: - enable_quantization(model) - if args.quant and args.deploy: - convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [10, 3, 224, 224]}) - return + exit(1) - if args.evaluate: - if args.quant: + if args.evaluate: from mqbench.convert_deploy import convert_merge_bn - convert_merge_bn(model.eval()) - validate(val_loader, model, criterion, args) - return - + module_tmp2 = copy.deepcopy(model) + convert_merge_bn(module_tmp2.eval()) + validate(val_loader, module_tmp2, criterion, args) + del module_tmp2 + gen_test_ref_data(cali_loader, model, args) + convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [cali_batch_size, 3, 224, 224]}, + model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) + exit(0) + + filename= os.path.join(args.output_path, 'checkpoint.pth.tar') for epoch in range(args.start_epoch, args.epochs): if args.distributed: train_sampler.set_epoch(epoch) @@ -280,7 +335,10 @@ def main_worker(gpu, ngpus_per_node, args): 'state_dict': model.state_dict(), 'best_acc1': best_acc1, 'optimizer' : optimizer.state_dict(), - }, is_best) + }, is_best, filename=filename) + gen_test_ref_data(cali_loader, model, args) + convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [cali_batch_size, 3, 224, 224]}, + model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) def prepare_dataloader(args): traindir = os.path.join(args.train_data, 'train') @@ -306,7 +364,6 @@ def prepare_dataloader(args): train_dataset, batch_size=args.batch_size, shuffle=(train_sampler is None), num_workers=args.workers, pin_memory=True, sampler=train_sampler) - cali_batch_size = 10 cali_batch = 10 cali_dataset = torch.utils.data.Subset(train_dataset, indices=torch.arange(cali_batch_size * cali_batch)) cali_loader = torch.utils.data.DataLoader(cali_dataset, batch_size=cali_batch_size, shuffle=False, @@ -324,6 +381,12 @@ def prepare_dataloader(args): return train_loader, train_sampler, val_loader, cali_loader +def get_node_name_by_module_name(qname, model): + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + for node in nodes: + if node.target in modules and qname == node.target: + return node.name def calibrate(cali_loader, model, args): model.eval() print("Start calibration ...") @@ -336,6 +399,41 @@ def calibrate(cali_loader, model, args): print("Calibration ==> ", i+1) print("End calibration.") return +def gen_test_ref_data(cali_loader, model, args): + model.eval() + global layer_names + hook_handles = [] + input_data = {} + exclude_module = ['mqbench', 'torch.fx', 'batchnorm', 'torch.nn.modules.module.Module'] + nodes = list(model.graph.nodes) + for name, child in model.named_modules(): + if not any([i in str(type(child)) for i in exclude_module]): + print("add hook on", str(type(child)), name) + node_name = get_node_name_by_module_name(name, model) + layer_names.append(node_name) + hd = child.register_forward_hook(hook=hook) + hook_handles.append(hd) + print('layer_names:', layer_names) + if args.cpu: + model = model.cpu() + with torch.no_grad(): + for i, (images, target) in enumerate(cali_loader): + if args.gpu is not None: + images = images.cuda(args.gpu, non_blocking=True) + else: + images = images.cpu() + output = model(images) + print("gen_test_ref_data ==> ", i+1) + if i == 0: + input_data['data'] = images.cpu().numpy() + np.savez(os.path.join(args.output_path, 'input_data.npz'), **input_data) + global features_out_hook + np.savez(os.path.join(args.output_path, 'layer_outputs.npz'), **features_out_hook) + for hd in hook_handles: + hd.remove() + break + print("End gen_test_ref_data.") + return def train(train_loader, model, criterion, optimizer, epoch, args): batch_time = AverageMeter('Time', ':6.3f') @@ -383,6 +481,14 @@ def train(train_loader, model, criterion, optimizer, epoch, args): if i % args.print_freq == 0: progress.display(i) + # for param in model.named_parameters(): + # sum = torch.isnan(param[1]).sum() + # if sum > 0: + # print(param[0], 'has Nan', param[1].shape, 'sum:', sum) + + if args.fast_test: + if i % 100 == 0: + break def validate(val_loader, model, criterion, args): batch_time = AverageMeter('Time', ':6.3f') @@ -396,14 +502,21 @@ def validate(val_loader, model, criterion, args): # switch to evaluate mode model.eval() + if args.cpu: + model = model.cpu() + with torch.no_grad(): end = time.time() for i, (images, target) in enumerate(val_loader): - if args.gpu is not None: - images = images.cuda(args.gpu, non_blocking=True) - if torch.cuda.is_available(): - target = target.cuda(args.gpu, non_blocking=True) + if not args.cpu: + if args.gpu is not None: + images = images.cuda(args.gpu, non_blocking=True) + if torch.cuda.is_available(): + target = target.cuda(args.gpu, non_blocking=True) + else: + images = images.cpu() + target = target.cpu() # compute output output = model(images) @@ -421,6 +534,11 @@ def validate(val_loader, model, criterion, args): if i % args.print_freq == 0: progress.display(i) + + if args.fast_test: + if i % 100 == 0: + break + # TODO: this should also be done with the ProgressMeter print(' * Acc@1 {top1.avg:.3f} Acc@5 {top5.avg:.3f}' .format(top1=top1, top5=top5)) @@ -431,7 +549,7 @@ def validate(val_loader, model, criterion, args): def save_checkpoint(state, is_best, filename='checkpoint.pth.tar'): torch.save(state, filename) if is_best: - shutil.copyfile(filename, 'model_best.pth.tar') + shutil.copyfile(filename, filename+'_best') class AverageMeter(object): diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index c5fdb0b7..80f79bfa 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -1,3 +1,4 @@ +import json import os.path as osp import torch @@ -32,15 +33,17 @@ @register_deploy_function(BackendType.NNIE) @register_deploy_function(BackendType.Vitis) @register_deploy_function(BackendType.OPENVINO) +@register_deploy_function(BackendType.Sophgo_TPU) def convert_merge_bn(model: GraphModule, **kwargs): + print('wlog before convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) logger.info("Merge BN for deploy.") nodes = list(model.graph.nodes) modules = dict(model.named_modules()) for node in nodes: if node.op == 'call_module': - if type(modules[node.target]) in FUSED_MODULE_CONVERT_FUNCTION: + if node.target in modules and type(modules[node.target]) in FUSED_MODULE_CONVERT_FUNCTION: FUSED_MODULE_CONVERT_FUNCTION[type(modules[node.target])](model, node) - + print('wlog after convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) @register_deploy_function(BackendType.Academic_NLP) @register_deploy_function(BackendType.Tensorrt_NLP) @@ -54,8 +57,10 @@ def convert_merge_bn(model: GraphModule, **kwargs): @register_deploy_function(BackendType.NNIE) @register_deploy_function(BackendType.Vitis) @register_deploy_function(BackendType.OPENVINO) +@register_deploy_function(BackendType.Sophgo_TPU) def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_path, **kwargs): logger.info("Export to onnx.") + model = model.cpu() output_names = kwargs.get('output_names', []) dynamic_axes = kwargs.get('dynamic_axes', {}) input_names = kwargs.get('input_names', []) @@ -112,6 +117,23 @@ def deploy_qparams_openvino(model: GraphModule, onnx_model_path, model_name, **k def deploy_qparams_tensorrt(model: GraphModule, onnx_model_path, model_name, **kwargs): logger.info("Extract qparams for TensorRT.") remove_fakequantize_and_collect_params(onnx_model_path, model_name, backend='tensorrt') +@register_deploy_function(BackendType.Sophgo_TPU) +def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, **kwargs): + logger.info("Extract qparams for sophgo_tpu.") + remove_fakequantize_and_collect_params(onnx_model_path, model_name, backend='sophgo_tpu') + output_path = osp.dirname(onnx_model_path) + context_filename = osp.join(output_path, '{}_clip_ranges.json'.format(model_name)) + file_h = open(context_filename, "r") + blob_range = json.loads(file_h.read())["sophgo_tpu"] + file_h.close() + cali_table = osp.join(output_path, '{}_cali_table_from_mqbench_sophgo_tpu'.format(model_name)) + with open(cali_table, 'w') as f: + f.write("# op_name threshold min max\n") + ori_layer_names = '' + for name,value in blob_range.items(): + f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) + ori_layer_names += '{},'.format(value['ori_name']) + f.write("#{}\n".format(ori_layer_names[0:-1])) @register_deploy_function(BackendType.Vitis) diff --git a/mqbench/custom_quantizer/__init__.py b/mqbench/custom_quantizer/__init__.py index 94646a04..a8723d27 100644 --- a/mqbench/custom_quantizer/__init__.py +++ b/mqbench/custom_quantizer/__init__.py @@ -6,4 +6,5 @@ from .tensorrt_quantizer import TRTModelQuantizer, TensorrtNLPQuantizer from .tengine_u8_quantizer import TengineQuantizer from .onnx_qnn_quantizer import ONNXQNNQuantizer -from .nlp_quantizer import AcademicNLPQuantizer \ No newline at end of file +from .nlp_quantizer import AcademicNLPQuantizer +from .sophgo_tpu_quantizer import SophgoTpuQuantizer diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index 8a4a64bb..12d3203e 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -34,7 +34,7 @@ from mqbench.utils.logger import logger from mqbench.utils.registry import register_model_quantizer from mqbench.prepare_by_platform import BackendType - +import mqbench.nn.intrinsic.qat as qnniqat @register_model_quantizer(BackendType.Tensorrt) @register_model_quantizer(BackendType.NNIE) @@ -63,7 +63,7 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): def prepare(self, model: GraphModule, qconfig): model = _fuse_fx(model, self.extra_fuse_dict) model = self._weight_quant(model, qconfig) - model = self._insert_fake_quantize_for_act_quant(model, qconfig) + model = self._insert_fake_quantize_for_act_quant(model, qconfig['']) return model def _insert_fake_quantize_for_act_quant( @@ -85,25 +85,34 @@ def _insert_fake_quantize_for_act_quant( with graph.inserting_after(node): inserted_node = graph.create_node("call_module", quantizer_name, (node,), {}) for _node in nodes: - _node.args = self._fix_succ_recursivly(_node.args, node, inserted_node) + _node.args = self._fix_succ_recursivly(model, _node, fake_quantizer, _node.args, node, inserted_node) model.recompile() model.graph.lint() return model - def _fix_succ_recursivly(self, args, target_node, inserted_node): + def _fix_succ_recursivly(self, model, node, fake_quantizer, args, target_node, inserted_node): # List / Tuple if isinstance(args, (list, tuple)): _tmp = list(args) for _i, _arg in enumerate(args): if _arg == target_node: _tmp[_i] = inserted_node + modules_has_bias = (qnniqat.ConvBnReLU2d_sophgo, + qnniqat.ConvBn2d_sophgo, + qnniqat.LinearReLU_sophgo, + qnniqat.Linear_sophgo) + modules = dict(model.named_modules()) + if (node.op == "call_module" and isinstance(modules[node.target], modules_has_bias)): + setattr(modules[node.target], "input_fake_quantizer", fake_quantizer) + print('wlog:', node.target,'\'type is:', type(modules[node.target]), "add its new attr:input_fake_quantizer", ',its old pre op is:', + target_node.name, target_node.type, ',its new insert pre op is:', inserted_node.name) elif isinstance(_arg, tuple): - _tmp[_i] = self._fix_succ_recursivly(_arg, target_node, inserted_node) + _tmp[_i] = self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node) elif isinstance(_arg, list): - _tmp[_i] = list(self._fix_succ_recursivly(_arg, target_node, inserted_node)) + _tmp[_i] = list(self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node)) elif isinstance(_arg, dict): - _tmp[_i] = self._fix_succ_recursivly(_arg, target_node, inserted_node) + _tmp[_i] = self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node) return tuple(_tmp) # Dict elif isinstance(args, dict): @@ -112,7 +121,7 @@ def _fix_succ_recursivly(self, args, target_node, inserted_node): if v == target_node: _tmp[k] = inserted_node elif not isinstance(v, torch.fx.node.Node): - _tmp[k] = self._fix_succ_recursivly(v, target_node, inserted_node) + _tmp[k] = self._fix_succ_recursivly(model, node, fake_quantizer, v, target_node, inserted_node) else: _tmp[k] = v return _tmp @@ -121,7 +130,7 @@ def _fix_succ_recursivly(self, args, target_node, inserted_node): def _weight_quant(self, model: GraphModule, qconfig): logger.info("Replace module to qat module.") - flattened_qconfig_dict = get_flattened_qconfig_dict({'': qconfig}) + flattened_qconfig_dict = get_flattened_qconfig_dict(qconfig)#torch??? propagate_qconfig_(model, flattened_qconfig_dict) self._qat_swap_modules(model, self.additional_qat_module_mapping) return model diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py new file mode 100644 index 00000000..8b914030 --- /dev/null +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -0,0 +1,74 @@ +import torch +from torch.fx import GraphModule +import torch.nn.intrinsic as nni +import mqbench.nn.intrinsic.qat as qnniqat +from mqbench.utils.registry import register_model_quantizer +from mqbench.prepare_by_platform import BackendType +from mqbench.custom_quantizer import ModelQuantizer +import torch.nn as nn + + +@register_model_quantizer(BackendType.Sophgo_TPU) +class SophgoTpuQuantizer(ModelQuantizer): + """There is only INT8 calculations in the model. + We quantize the input tensors and output tensors of all layers, + except those in _passed_func_type and _passed_module_type. + For example add + relu pattern, there is no need to insert fake + quantize node between them. + """ + + def __init__(self, extra_quantizer_dict, extra_fuse_dict): + super().__init__(extra_quantizer_dict, extra_fuse_dict) + self.additional_qat_module_mapping = { + # Intrinsic modules: + nni.ConvBn2d: qnniqat.ConvBn2d_sophgo, + nni.ConvBnReLU2d: qnniqat.ConvBnReLU2d_sophgo, + # nni.ConvReLU2d: qnniqat.ConvReLU2d_sophgo, + nni.LinearReLU: qnniqat.LinearReLU_sophgo, + nn.Linear: qnniqat.Linear_sophgo + } + + @property + def module_type_to_quant_input(self) -> tuple: + return super().module_type_to_quant_input + ( + qnniqat.ConvBnReLU2d_sophgo, + qnniqat.ConvBn2d_sophgo, + qnniqat.LinearReLU_sophgo, + qnniqat.Linear_sophgo + ) + + @property + def _passed_func_type(self): + return ( + torch.nn.functional.relu, + torch.nn.functional.relu6, + torch.flatten + ) + + @property + def _passed_module_type(self): + return ( + torch.nn.ReLU, + torch.nn.ReLU6 + ) + + def _find_act_quants(self, model: GraphModule) -> list: + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + node_need_to_quantize_output = super()._find_act_quants(model) + for node in nodes: + if (node.op == "call_module" and node.target in self.exclude_module_name) or \ + ((node.op == 'call_function' or node.op == 'call_method') and + node.target in self.exclude_function_type) or \ + node.name in self.exclude_node_name: + continue + if (node.op == "call_module" and isinstance(modules[node.target], self.module_type_to_quant_input)) or \ + ((node.op == 'call_function' or node.op == 'call_method') and + node.target in self.function_type_to_quant_input): + for next_node in node.users: + if not ((next_node.op == 'call_function' and next_node.target in self._passed_func_type) or + (next_node.op == 'call_module' and isinstance(modules[next_node.target], self._passed_module_type))): + node_need_to_quantize_output.append(node) + else: + node_need_to_quantize_output.append(next_node) + return node_need_to_quantize_output \ No newline at end of file diff --git a/mqbench/deploy/deploy_linear.py b/mqbench/deploy/deploy_linear.py index 8dd4deb3..0ab923fd 100644 --- a/mqbench/deploy/deploy_linear.py +++ b/mqbench/deploy/deploy_linear.py @@ -1,5 +1,6 @@ import json import os +import copy import onnx @@ -115,10 +116,14 @@ def clip_weight(self, node, name2data, inp2node, named_initializer): new_data = numpy_helper.from_array(new_data) named_initializer[tensor_name].raw_data = new_data.raw_data - def post_process_clip_ranges(self, clip_ranges, graph, inp2node): + def post_process_clip_ranges(self, clip_ranges, graph, inp2node, out2node): def find_the_closest_clip_range(node): - if node.input[0] in clip_ranges: - return node.input[0] + input_0 = node.input[0] + tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') + if tensor_name[-1] == '_': + tensor_name = tensor_name[:-1] + if tensor_name in clip_ranges: + return tensor_name elif node.op_type in ['Flatten', 'Resize'] and node.output[0] in inp2node: return find_the_closest_clip_range(inp2node[node.output[0]][0][0]) else: @@ -128,7 +133,13 @@ def find_the_closest_clip_range(node): if node.op_type in ['Flatten', 'Resize']: tensor_name = find_the_closest_clip_range(node) if tensor_name: - clip_ranges[node.input[0]] = clip_ranges[tensor_name] + old = clip_ranges[tensor_name] + new_name = node.input[0] + new_name = '{}_{}'.format(new_name, out2node[new_name].op_type if new_name in out2node else '') + if new_name[-1] == '_': + new_name = tensor_name[:-1] + clip_ranges[new_name] = copy.deepcopy(old) + clip_ranges[new_name]['ori_name'] = new_name logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') return clip_ranges @@ -155,8 +166,8 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) redundant_nodes = self.deal_with_weight_fakequant(node, out2node, inp2node, named_initializer) nodes_to_be_removed.extend(redundant_nodes) self.clip_weight(node, name2data, inp2node, named_initializer) + tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) if backend == 'ppl': - tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) clip_ranges[tensor_name] = {'step': [float(x) for x in scale], 'zero_point': [int(x) for x in zero_point], 'min': [float(x) for x in scale * (qmin - zero_point)], @@ -191,6 +202,14 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) # fake quantize for activations self.deal_with_activation_fakequant(node, inp2node) tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) + scale_name = node.input[1] + pre_layer_name = 'none' + post_str = '_post_act_fake_quantizer.scale' + if scale_name.endswith(post_str): + pre_layer_name = scale_name[:len(scale_name)-len(post_str)] + else: + print('not _post_act_fake_quantizer') + input_0 = node.input[0] for out in graph.output: if out.name == node.output[0]: out.name = tensor_name @@ -203,6 +222,15 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) 'min': float(scale * (qmin - zero_point)), 'max': float(scale * (qmax - zero_point))} ] + elif backend == 'sophgo_tpu': + input_0 = node.input[0] + tensor_name += '_{}'.format(out2node[input_0].op_type if input_0 in out2node else '') + if tensor_name[-1] == '_': + tensor_name = tensor_name[:-1] + clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #对称量化时这个参数生效 + 'min': float(scale * (qmin - zero_point)), + 'max': float(scale * (qmax - zero_point)), + 'ori_name':pre_layer_name} if backend == 'ppl': clip_ranges[tensor_name] = {'step': float(scale), 'zero_point': int(zero_point), @@ -226,7 +254,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) continue graph.initializer.remove(initial_data) - clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node) + clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) if backend == 'tensorrt': context = {"tensorrt": {"blob_range": clip_ranges}} elif backend == 'snpe': @@ -237,6 +265,10 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) context = {'vitis': clip_ranges} elif backend == 'ppl-cuda': context = {'ppl-cuda': clip_ranges} + elif backend == 'sophgo_tpu': + context = {'sophgo_tpu': clip_ranges} + context['w_qscheme'] = '' + context['a_qscheme'] = '' output_path = os.path.dirname(onnx_path) context_filename = os.path.join(output_path, '{}_clip_ranges.json'.format(model_name)) with open(context_filename, 'w') as f: diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index 2bbf693c..e879db28 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -11,6 +11,44 @@ from mqbench.quantization.default_bias_fake_quant import bias_fake_quantizer +@register_convert_function(qnniqat.Linear_sophgo) +def convert_qnniqat_linear(model, fused_node): + modules = dict(model.named_modules()) + fused_module = modules[fused_node.target] + linear = torch.nn.Linear(fused_module.in_features, fused_module.out_features, fused_module.bias is not None) + linear.weight = fused_module.weight + if fused_module.bias is not None: + linear.bias = fused_module.bias + linear.qconfig = fused_module.qconfig + linear = torch.nn.qat.Linear.from_float(linear) + linear.weight_fake_quant = fused_module.weight_fake_quant + linear_parent_name, linear_name = _parent_name(fused_node.target) + setattr(modules[linear_parent_name], linear_name, linear) +@register_convert_function(qnniqat.LinearReLU_sophgo) +def linearert_qnniqat_linearrelu(model, fused_node): + convert_qnniqat_linear(model, fused_node) + modules = dict(model.named_modules()) + fused_module = modules[fused_node.target] + linear_parent_name, linear_name = _parent_name(fused_node.target) + relu_name = 'relu' + if not hasattr(modules[linear_parent_name], relu_name): + setattr(modules[linear_parent_name], relu_name, + torch.nn.ReLU(inplace=True).train(fused_module.training)) + modules = dict(model.named_modules()) + graph = model.graph + nodes = list(model.graph.nodes) + with graph.inserting_after(fused_node): + relu_node_name = relu_name if linear_parent_name == "" else "{}.{}".format(linear_parent_name, relu_name) + assert relu_node_name in modules and isinstance(modules[relu_node_name], torch.nn.ReLU) + inserted_node = graph.create_node("call_module", relu_node_name, (fused_node,), {}) + for _node in nodes: + for i, _arg in enumerate(_node.args): + if _arg == fused_node: + _tmp = list(_node.args) + _tmp[i] = inserted_node + _node.args = tuple(_tmp) + model.recompile() + model.graph.lint() @register_convert_function(qnni.LinearBn1d) def convert_qnni_linearbn(model, fused_node): modules = dict(model.named_modules()) @@ -21,6 +59,7 @@ def convert_qnni_linearbn(model, fused_node): @register_convert_function(qnniqat.LinearBn1d) +@register_convert_function(qnniqat.LinearBn1d_sophgo) def convert_qnniqat_linearbn(model, fused_node): modules = dict(model.named_modules()) fused_module = modules[fused_node.target] @@ -43,6 +82,7 @@ def convert_qnniqat_linearbn(model, fused_node): @register_convert_function(qnniqat.ConvFreezebn2d) @register_convert_function(nniqat.ConvBn2d) @register_convert_function(nniqat.ConvBn3d) +@register_convert_function(qnniqat.ConvBn2d_sophgo) def convert_nniqat_convbn(model, fused_node): """nniqat.ConvBn2d ----> nn.Conv2d ----> nniqat.Conv2d """ @@ -53,6 +93,8 @@ def convert_nniqat_convbn(model, fused_node): nniqat.ConvBnReLU2d: torch.nn.Conv2d, nniqat.ConvBn3d: torch.nn.Conv3d, nniqat.ConvBnReLU3d: torch.nn.Conv3d, + qnniqat.ConvBn2d_sophgo: torch.nn.Conv2d, + qnniqat.ConvBnReLU2d_sophgo: torch.nn.Conv2d, } fused_qat_module_class_map = { torch.nn.Conv2d: torch.nn.qat.Conv2d, @@ -82,6 +124,7 @@ def convert_nniqat_convbn(model, fused_node): @register_convert_function(qnniqat.ConvFreezebnReLU2d) @register_convert_function(nniqat.ConvBnReLU2d) @register_convert_function(nniqat.ConvBnReLU3d) +@register_convert_function(qnniqat.ConvBnReLU2d_sophgo) def convert_nniqat_convbnrelu(model, fused_node): convert_nniqat_convbn(model, fused_node) modules = dict(model.named_modules()) @@ -263,7 +306,7 @@ def convert_qnniqat_convbnrelu(model, fused_node): with graph.inserting_after(fused_node): relu_node_name = relu_name if conv_parent_name == "" else "{}.{}".format(conv_parent_name, relu_name) assert relu_node_name in modules and isinstance(modules[relu_node_name], torch.nn.ReLU) - inserted_node = graph.create_node("call_module", relu_node_name, (fused_node,), {}) + inserted_node = graph.create_node("call_module", relu_node_name, (fused_node,), {}, name = fused_node.name) for _node in nodes: for i, _arg in enumerate(_node.args): if _arg == fused_node: diff --git a/mqbench/nn/intrinsic/qat/modules/__init__.py b/mqbench/nn/intrinsic/qat/modules/__init__.py index 3118f9bf..c469b714 100644 --- a/mqbench/nn/intrinsic/qat/modules/__init__.py +++ b/mqbench/nn/intrinsic/qat/modules/__init__.py @@ -2,3 +2,5 @@ from .deconv_fused import ConvTransposeBnReLU2d, ConvTransposeBn2d, ConvTransposeReLU2d from .conv_fused import ConvBnReLU2d, ConvBn2d, ConvReLU2d from .freezebn import ConvFreezebn2d, ConvFreezebnReLU2d, ConvTransposeFreezebn2d, ConvTransposeFreezebnReLU2d +from .conv_fused_sophgo_tpu import ConvBnReLU2d_sophgo, ConvBn2d_sophgo, ConvReLU2d_sophgo +from .linear_fused_sophgo_tpu import LinearBn1d_sophgo, LinearReLU_sophgo, Linear_sophgo diff --git a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py new file mode 100644 index 00000000..52b161f5 --- /dev/null +++ b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py @@ -0,0 +1,458 @@ +import math + +import torch +import torch.nn as nn +import torch.nn.intrinsic as nni +import torch.nn.functional as F +from torch.nn import init +from torch.nn.intrinsic import _FusedModule +from torch.nn.parameter import Parameter +from torch.nn.modules.utils import _pair + +from typing import TypeVar + + +import mqbench.nn.qat as qnnqat +import torch.nn.qat.modules as nnqat +# from mqbench.quantization.default_bias_fake_quant import bias_fake_quantizer + +_BN_CLASS_MAP = { + 1: nn.BatchNorm1d, + 2: nn.BatchNorm2d, + 3: nn.BatchNorm3d, +} + +MOD = TypeVar('MOD', bound=nn.modules.conv._ConvNd) + + +class _ConvBnNd(nn.modules.conv._ConvNd, _FusedModule): + + _version = 2 + _FLOAT_MODULE = MOD + + def __init__(self, + # ConvNd args + in_channels, out_channels, kernel_size, stride, + padding, dilation, transposed, output_padding, + groups, + bias, + padding_mode, + # BatchNormNd args + # num_features: out_channels + eps=1e-05, momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None, + dim=2): + nn.modules.conv._ConvNd.__init__(self, in_channels, out_channels, kernel_size, + stride, padding, dilation, transposed, + output_padding, groups, False, padding_mode) + assert qconfig, 'qconfig must be provided for QAT module' + self.qconfig = qconfig + self.freeze_bn = freeze_bn if self.training else True + self.bn = _BN_CLASS_MAP[dim](out_channels, eps, momentum, True, True) + self.weight_fake_quant = self.qconfig.weight() + if bias: + self.bias = Parameter(torch.empty(out_channels)) + else: + self.register_parameter('bias', None) + # self.bias_fake_quant = bias_fake_quantizer() + self.reset_bn_parameters() + + # this needs to be called after reset_bn_parameters, + # as they modify the same state + if self.training: + if freeze_bn: + self.freeze_bn_stats() + else: + self.update_bn_stats() + else: + self.freeze_bn_stats() + + def reset_running_stats(self): + self.bn.reset_running_stats() + + def reset_bn_parameters(self): + self.bn.reset_running_stats() + init.uniform_(self.bn.weight) + init.zeros_(self.bn.bias) + # note: below is actully for conv, not BN + if self.bias is not None: + fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight) + bound = 1 / math.sqrt(fan_in) + init.uniform_(self.bias, -bound, bound) + + def reset_parameters(self): + super(_ConvBnNd, self).reset_parameters() + + def update_bn_stats(self): + self.freeze_bn = False + self.bn.training = True + return self + + def freeze_bn_stats(self): + self.freeze_bn = True + self.bn.training = False + return self + +# def _forward(self, input): +# assert self.bn.running_var is not None +# running_std = torch.sqrt(self.bn.running_var + self.bn.eps) +# scale_factor = self.bn.weight / running_std +# weight_shape = [1] * len(self.weight.shape) +# weight_shape[0] = -1 +# bias_shape = [1] * len(self.weight.shape) +# bias_shape[1] = -1 +# scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) +# # using zero bias here since the bias for original conv +# # will be added later +# if self.bias is not None: +# zero_bias = torch.zeros_like(self.bias) +# conv_bias = self.bias +# else: +# zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) +# conv_bias = torch.zeros_like(zero_bias, device=scaled_weight.device) +# if self.bn.affine: +# full_bias = (conv_bias - self.bn.running_mean) / running_std * self.bn.weight + self.bn.bias +# else: +# full_bias = (conv_bias - self.bn.running_mean) / running_std +# quant_bias = self.bias_fake_quant(full_bias) +# conv_with_bias = self._conv_forward(input, scaled_weight, quant_bias) +# conv_orig = (conv_with_bias - full_bias.reshape(bias_shape)) / scale_factor.reshape(bias_shape) + conv_bias.reshape(bias_shape) +# conv = self.bn(conv_orig) +# return conv + + + def bias_fake_quant_proc(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('error! scale has 0, scale:', scale) + scale[torch.abs(scale) < 1e-10] = 1e-10 + print('new scale:', scale) + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def _forward(self, input): + # print('xxx2') + assert self.bn.running_var is not None + running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + scale_factor = self.bn.weight / running_std + weight_shape = [1] * len(self.weight.shape) + weight_shape[0] = -1 + bias_shape = [1] * len(self.weight.shape) + bias_shape[1] = -1 + if torch.isnan(self.weight).any(): + print('weight have nan') + if self.input_fake_quantizer is not None and torch.isnan(self.input_fake_quantizer.scale).any(): + print('input_fake_quantizer.scale have nan') + if self.bias is not None and torch.isnan(self.bias).any(): + print('weight have nan') + scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + #bias浼噺鍖 + if self.weight_fake_quant.fake_quant_enabled[0] == 1: + _, fused_bias = nn.utils.fuse_conv_bn_weights(self.weight, self.bias, + self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) + in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) + diff_fused_bias = fused_bias - scale_fused_bias + # using zero bias here since the bias for original conv + # will be added later + if self.bias is not None: + zero_bias = torch.zeros_like(self.bias) + else: + zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + conv = self._conv_forward(input, scaled_weight, zero_bias) + conv_orig = conv / scale_factor.reshape(bias_shape) + if self.bias is not None: + conv_orig = conv_orig + self.bias.reshape(bias_shape) + conv = self.bn(conv_orig) + if self.weight_fake_quant.fake_quant_enabled[0] == 1: + conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 + return conv + + def extra_repr(self): + # TODO(jerryzh): extend + return super(_ConvBnNd, self).extra_repr() + + def forward(self, input): + return self._forward(input) + + def train(self, mode=True): + """ + Batchnorm's training behavior is using the self.training flag. Prevent + changing it if BN is frozen. This makes sure that calling `model.train()` + on a model with a frozen BN will behave properly. + """ + self.training = mode + if not self.freeze_bn: + for module in self.children(): + module.train(mode) + return self + + # ===== Serialization version history ===== + # + # Version 1/None + # self + # |--- weight : Tensor + # |--- bias : Tensor + # |--- gamma : Tensor + # |--- beta : Tensor + # |--- running_mean : Tensor + # |--- running_var : Tensor + # |--- num_batches_tracked : Tensor + # + # Version 2 + # self + # |--- weight : Tensor + # |--- bias : Tensor + # |--- bn : Module + # |--- weight : Tensor (moved from v1.self.gamma) + # |--- bias : Tensor (moved from v1.self.beta) + # |--- running_mean : Tensor (moved from v1.self.running_mean) + # |--- running_var : Tensor (moved from v1.self.running_var) + # |--- num_batches_tracked : Tensor (moved from v1.self.num_batches_tracked) + def _load_from_state_dict(self, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs): + version = local_metadata.get('version', None) + if version is None or version == 1: + # BN related parameters and buffers were moved into the BN module for v2 + v2_to_v1_names = { + 'bn.weight': 'gamma', + 'bn.bias': 'beta', + 'bn.running_mean': 'running_mean', + 'bn.running_var': 'running_var', + 'bn.num_batches_tracked': 'num_batches_tracked', + } + for v2_name, v1_name in v2_to_v1_names.items(): + if prefix + v1_name in state_dict: + state_dict[prefix + v2_name] = state_dict[prefix + v1_name] + state_dict.pop(prefix + v1_name) + elif prefix + v2_name in state_dict: + # there was a brief period where forward compatibility + # for this module was broken (between + # https://github.com/pytorch/pytorch/pull/38478 + # and https://github.com/pytorch/pytorch/pull/38820) + # and modules emitted the v2 state_dict format while + # specifying that version == 1. This patches the forward + # compatibility issue by allowing the v2 style entries to + # be used. + pass + elif strict: + missing_keys.append(prefix + v2_name) + + super(_ConvBnNd, self)._load_from_state_dict( + state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) + + @classmethod + def from_float(cls, mod): + r"""Create a qat module from a float module or qparams_dict + + Args: `mod` a float module, either produced by torch.ao.quantization utilities + or directly from user + """ + # The ignore is because _FLOAT_MODULE is a TypeVar here where the bound + # has no __name__ (code is fine though) + assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + \ + cls._FLOAT_MODULE.__name__ # type: ignore[attr-defined] + assert hasattr(mod, 'qconfig'), 'Input float module must have qconfig defined' + assert mod.qconfig, 'Input float module must have a valid qconfig' + qconfig = mod.qconfig + conv, bn = mod[0], mod[1] + qat_convbn = cls(conv.in_channels, conv.out_channels, conv.kernel_size, + conv.stride, conv.padding, conv.dilation, + conv.groups, conv.bias is not None, + conv.padding_mode, + bn.eps, bn.momentum, + False, + qconfig) + qat_convbn.weight = conv.weight + qat_convbn.bias = conv.bias + qat_convbn.bn.weight = bn.weight + qat_convbn.bn.bias = bn.bias + qat_convbn.bn.running_mean = bn.running_mean + qat_convbn.bn.running_var = bn.running_var + # mypy error: Cannot determine type of 'num_batches_tracked' + qat_convbn.bn.num_batches_tracked = bn.num_batches_tracked # type: ignore[has-type] + return qat_convbn + + def to_float(self): + modules = [] + cls = type(self) + conv = cls._FLOAT_CONV_MODULE( # type: ignore[attr-defined] + self.in_channels, + self.out_channels, + self.kernel_size, + self.stride, + self.padding, + self.dilation, + self.groups, + self.bias is not None, + self.padding_mode) + conv.weight = torch.nn.Parameter(self.weight.detach()) + if self.bias is not None: + conv.bias = torch.nn.Parameter(self.bias.detach()) + modules.append(conv) + + if cls._FLOAT_BN_MODULE: # type: ignore[attr-defined] + bn = cls._FLOAT_BN_MODULE( # type: ignore[attr-defined] + self.bn.num_features, + self.bn.eps, + self.bn.momentum, + self.bn.affine, + self.bn.track_running_stats) + bn.weight = Parameter(self.bn.weight.detach()) + if self.bn.affine: + bn.bias = Parameter(self.bn.bias.detach()) + modules.append(bn) + + if cls._FLOAT_RELU_MODULE: # type: ignore[attr-defined] + relu = cls._FLOAT_RELU_MODULE() # type: ignore[attr-defined] + modules.append(relu) + + result = cls._FLOAT_MODULE(*modules) # type: ignore[operator] + result.train(self.training) + return result + + + +class ConvBn2d_sophgo(_ConvBnNd, nn.Conv2d): + r""" + A ConvBn2d module is a module fused from Conv2d and BatchNorm2d, + attached with FakeQuantize modules for weight, + used in quantization aware training. + + We combined the interface of :class:`torch.nn.Conv2d` and + :class:`torch.nn.BatchNorm2d`. + + Similar to :class:`torch.nn.Conv2d`, with FakeQuantize modules initialized + to default. + + Attributes: + freeze_bn: + weight_fake_quant: fake quant module for weight + + """ + _FLOAT_MODULE = nni.ConvBn2d + _FLOAT_CONV_MODULE = nn.Conv2d + _FLOAT_BN_MODULE = nn.BatchNorm2d + _FLOAT_RELU_MODULE = None + + def __init__(self, + # ConvNd args + in_channels, out_channels, kernel_size, stride=1, + padding=0, dilation=1, groups=1, + bias=None, + padding_mode='zeros', + # BatchNorm2d args + # num_features: out_channels + eps=1e-05, momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None): + kernel_size = _pair(kernel_size) + stride = _pair(stride) + padding = _pair(padding) + dilation = _pair(dilation) + _ConvBnNd.__init__(self, in_channels, out_channels, kernel_size, stride, + padding, dilation, False, _pair(0), groups, bias, padding_mode, + eps, momentum, freeze_bn, qconfig, dim=2) + +class ConvBnReLU2d_sophgo(ConvBn2d_sophgo): + r""" + A ConvBnReLU2d module is a module fused from Conv2d, BatchNorm2d and ReLU, + attached with FakeQuantize modules for weight, + used in quantization aware training. + + We combined the interface of :class:`torch.nn.Conv2d` and + :class:`torch.nn.BatchNorm2d` and :class:`torch.nn.ReLU`. + + Similar to `torch.nn.Conv2d`, with FakeQuantize modules initialized to + default. + + Attributes: + weight_fake_quant: fake quant module for weight + + """ + # base class defines _FLOAT_MODULE as "ConvBn2d" + _FLOAT_MODULE = nni.ConvBnReLU2d # type: ignore[assignment] + _FLOAT_CONV_MODULE = nn.Conv2d + _FLOAT_BN_MODULE = nn.BatchNorm2d + _FLOAT_RELU_MODULE = nn.ReLU # type: ignore[assignment] + + def __init__(self, + # Conv2d args + in_channels, out_channels, kernel_size, stride=1, + padding=0, dilation=1, groups=1, + bias=None, + padding_mode='zeros', + # BatchNorm2d args + # num_features: out_channels + eps=1e-05, momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None): + super(ConvBnReLU2d_sophgo, self).__init__(in_channels, out_channels, kernel_size, stride, + padding, dilation, groups, bias, + padding_mode, eps, momentum, + freeze_bn, + qconfig) + + def forward(self, input): + return F.relu(ConvBn2d_sophgo._forward(self, input)) + + @classmethod + def from_float(cls, mod): + return super(ConvBnReLU2d_sophgo, cls).from_float(mod) + +class ConvReLU2d_sophgo(qnnqat.Conv2d_sophgo, _FusedModule): + r"""A ConvReLU2d module is a fused module of Conv2d and ReLU, attached with + FakeQuantize modules for weight for + quantization aware training. + + We combined the interface of :class:`~torch.nn.Conv2d` and + :class:`~torch.nn.BatchNorm2d`. + + Attributes: + weight_fake_quant: fake quant module for weight + + """ + _FLOAT_MODULE = nni.ConvReLU2d + _FLOAT_CONV_MODULE = nn.Conv2d + _FLOAT_BN_MODULE = None + _FLOAT_RELU_MODULE = nn.ReLU + + def __init__(self, in_channels, out_channels, kernel_size, stride=1, + padding=0, dilation=1, groups=1, + bias=True, padding_mode='zeros', + qconfig=None): + super(ConvReLU2d_sophgo, self).__init__(in_channels, out_channels, kernel_size, + stride=stride, padding=padding, dilation=dilation, + groups=groups, bias=bias, padding_mode=padding_mode, + qconfig=qconfig) + + def forward(self, input): + # print('xxx3') + return self.forward(input) + + @classmethod + def from_float(cls, mod): + assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + cls._FLOAT_MODULE.__name__ + assert hasattr(mod, 'qconfig'), 'Input float module must have qconfig defined' + assert mod.qconfig, 'Input float module must have a valid qconfig' + if type(mod) == cls._FLOAT_MODULE: + mod = mod[0] + qconfig = mod.qconfig + qat_conv = cls(mod.in_channels, mod.out_channels, mod.kernel_size, + stride=mod.stride, padding=mod.padding, dilation=mod.dilation, + groups=mod.groups, bias=mod.bias is not None, + padding_mode=mod.padding_mode, qconfig=qconfig) + qat_conv.weight = mod.weight + qat_conv.bias = mod.bias + return qat_conv diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py new file mode 100644 index 00000000..94f1d0bb --- /dev/null +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -0,0 +1,294 @@ +import math + +import torch +import torch.nn as nn +import torch.nn.functional as F +from torch.nn import init +from torch.nn import Linear +from torch.nn.intrinsic import _FusedModule +from torch.nn.parameter import Parameter + +from mqbench.nn.intrinsic import LinearBn1d +import torch.nn.intrinsic as nni + + +class LinearBn1d_sophgo(Linear, _FusedModule): + _version = 2 + _FLOAT_MODULE = LinearBn1d + + def __init__(self, + # ConvNd args + in_features, out_features, bias, + # BatchNormNd args + # num_features: out_channels + eps=1e-05, momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None): + Linear.__init__(self, in_features, out_features, False) + assert qconfig, 'qconfig must be provided for QAT module' + self.qconfig = qconfig + self.freeze_bn = freeze_bn if self.training else True + self.bn = nn.BatchNorm1d(out_features, eps, momentum, True, True) + self.weight_fake_quant = self.qconfig.weight() + if bias: + self.bias = Parameter(torch.empty(out_features)) + else: + self.register_parameter('bias', None) + self.reset_bn_parameters() + + # this needs to be called after reset_bn_parameters, + # as they modify the same state + if self.training: + if freeze_bn: + self.freeze_bn_stats() + else: + self.update_bn_stats() + else: + self.freeze_bn_stats() + + def reset_running_stats(self): + self.bn.reset_running_stats() + + def reset_bn_parameters(self): + self.bn.reset_running_stats() + init.uniform_(self.bn.weight) + init.zeros_(self.bn.bias) + # note: below is actully for Linear, not BN + if self.bias is not None: + fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight) + bound = 1 / math.sqrt(fan_in) + init.uniform_(self.bias, -bound, bound) + + def reset_parameters(self): + super(LinearBn1d, self).reset_parameters() + + def update_bn_stats(self): + self.freeze_bn = False + self.bn.training = True + return self + + def freeze_bn_stats(self): + self.freeze_bn = True + self.bn.training = False + return self + + def bias_fake_quant(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('Linear error! scale has 0, scale:', scale) + scale[torch.abs(scale) < 1e-10] = 1e-10 + print('new scale:', scale) + + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def _forward(self, input): + in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale + conv = F.linear(input, self.weight_fake_quant(self.weight), + self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) + return conv + + def forward(self, input): + # return F.linear(input, self.weight_fake_quant(self.weight), self.bias) + return self._forward(input) + + def extra_repr(self): + return super(LinearBn1d, self).extra_repr() + + def forward(self, input): + return self._forward(input) + + def train(self, mode=True): + """ + Batchnorm's training behavior is using the self.training flag. Prevent + changing it if BN is frozen. This makes sure that calling `model.train()` + on a model with a frozen BN will behave properly. + """ + self.training = mode + if not self.freeze_bn: + for module in self.children(): + module.train(mode) + return self + + def _load_from_state_dict(self, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs): + version = local_metadata.get('version', None) + if version is None or version == 1: + # BN related parameters and buffers were moved into the BN module for v2 + v2_to_v1_names = { + 'bn.weight': 'gamma', + 'bn.bias': 'beta', + 'bn.running_mean': 'running_mean', + 'bn.running_var': 'running_var', + 'bn.num_batches_tracked': 'num_batches_tracked', + } + for v2_name, v1_name in v2_to_v1_names.items(): + if prefix + v1_name in state_dict: + state_dict[prefix + v2_name] = state_dict[prefix + v1_name] + state_dict.pop(prefix + v1_name) + elif prefix + v2_name in state_dict: + # there was a brief period where forward compatibility + # for this module was broken (between + # https://github.com/pytorch/pytorch/pull/38478 + # and https://github.com/pytorch/pytorch/pull/38820) + # and modules emitted the v2 state_dict format while + # specifying that version == 1. This patches the forward + # compatibility issue by allowing the v2 style entries to + # be used. + pass + elif strict: + missing_keys.append(prefix + v2_name) + + super(LinearBn1d, self)._load_from_state_dict( + state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) + + @classmethod + def from_float(cls, mod): + r"""Create a qat module from a float module or qparams_dict + + Args: `mod` a float module, either produced by torch.quantization utilities + or directly from user + """ + assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + \ + cls._FLOAT_MODULE.__name__ + assert hasattr(mod, 'qconfig'), 'Input float module must have qconfig defined' + assert mod.qconfig, 'Input float module must have a valid qconfig' + qconfig = mod.qconfig + linear, bn = mod[0], mod[1] + qat_linearbn = cls(linear.in_features, linear.out_features, False, + bn.eps, bn.momentum, + False, + qconfig) + qat_linearbn.weight = linear.weight + qat_linearbn.bias = linear.bias + qat_linearbn.bn.weight = bn.weight + qat_linearbn.bn.bias = bn.bias + qat_linearbn.bn.running_mean = bn.running_mean + qat_linearbn.bn.running_var = bn.running_var + # mypy error: Cannot determine type of 'num_batches_tracked' + qat_linearbn.bn.num_batches_tracked = bn.num_batches_tracked # type: ignore[has-type] + return qat_linearbn + +class Linear_sophgo(nn.Linear): + r""" + A linear module attached with FakeQuantize modules for weight, + used for quantization aware training. + + We adopt the same interface as `torch.nn.Linear`, please see + https://pytorch.org/docs/stable/nn.html#torch.nn.Linear + for documentation. + + Similar to `torch.nn.Linear`, with FakeQuantize modules initialized to + default. + + Attributes: + weight: fake quant module for weight + """ + _FLOAT_MODULE = nn.Linear + + def __init__(self, in_features, out_features, bias=True, + qconfig=None, device=None, dtype=None) -> None: + factory_kwargs = {'device': device, 'dtype': dtype} + super().__init__(in_features, out_features, bias, **factory_kwargs) + assert qconfig, 'qconfig must be provided for QAT module' + self.qconfig = qconfig + self.weight_fake_quant = qconfig.weight(factory_kwargs=factory_kwargs) + + def bias_fake_quant(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('Linear error! scale has 0, scale:', scale) + scale[torch.abs(scale) < 1e-10] = 1e-10 + print('new scale:', scale) + + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def _forward(self, input): + # print('xxx1') + in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale + conv = F.linear(input, self.weight_fake_quant(self.weight), + self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) + return conv + + def forward(self, input): + # return F.linear(input, self.weight_fake_quant(self.weight), self.bias) + return self._forward(input) + + @classmethod + def from_float(cls, mod): + r"""Create a qat module from a float module or qparams_dict + + Args: `mod` a float module, either produced by torch.quantization utilities + or directly from user + """ + assert type(mod) == cls._FLOAT_MODULE, ' qat.' + cls.__name__ + '.from_float only works for ' + \ + cls._FLOAT_MODULE.__name__ + assert hasattr(mod, 'qconfig'), 'Input float module must have qconfig defined' + assert mod.qconfig, 'Input float module must have a valid qconfig' + if type(mod) == nni.LinearReLU: + mod = mod[0] + + qconfig = mod.qconfig + qat_linear = cls(mod.in_features, mod.out_features, bias=mod.bias is not None, qconfig=qconfig) + qat_linear.weight = mod.weight + qat_linear.bias = mod.bias + return qat_linear + + def to_float(self): + linear = torch.nn.Linear(self.in_features, self.out_features, self.bias is not None) + linear.weight = torch.nn.Parameter(self.weight.detach()) + if self.bias is not None: + linear.bias = torch.nn.Parameter(self.bias.detach()) + linear.train(self.training) + return linear + + +class LinearReLU_sophgo(Linear_sophgo): + r""" + A LinearReLU module fused from Linear and ReLU modules, attached with + FakeQuantize modules for weight, used in + quantization aware training. + + We adopt the same interface as :class:`torch.nn.Linear`. + + Similar to `torch.nn.intrinsic.LinearReLU`, with FakeQuantize modules initialized to + default. + + Attributes: + weight: fake quant module for weight + + Examples:: + + >>> m = nn.qat.LinearReLU(20, 30) + >>> input = torch.randn(128, 20) + >>> output = m(input) + >>> print(output.size()) + torch.Size([128, 30]) + """ + _FLOAT_MODULE = nni.LinearReLU + + def __init__(self, in_features, out_features, bias=True, + qconfig=None): + super(LinearReLU_sophgo, self).__init__(in_features, out_features, bias, qconfig) + + def forward(self, input): + return F.relu(Linear_sophgo._forward(self, input)) + + @classmethod + def from_float(cls, mod): + return super(LinearReLU_sophgo, cls).from_float(mod) + + def to_float(self): + linear = torch.nn.Linear(self.in_features, self.out_features, self.bias is not None) + linear.weight = torch.nn.Parameter(self.weight.detach()) + if self.bias is not None: + linear.bias = torch.nn.Parameter(self.bias.detach()) + relu = torch.nn.ReLU() + return torch.nn.intrinsic.LinearReLU(linear, relu) diff --git a/mqbench/nn/qat/modules/__init__.py b/mqbench/nn/qat/modules/__init__.py index c0109020..bb13511c 100644 --- a/mqbench/nn/qat/modules/__init__.py +++ b/mqbench/nn/qat/modules/__init__.py @@ -1,4 +1,4 @@ from .linear import Linear from .deconv import ConvTranspose2d -from .conv import Conv2d +from .conv import Conv2d, Conv2d_sophgo from .embedding import Embedding \ No newline at end of file diff --git a/mqbench/nn/qat/modules/conv.py b/mqbench/nn/qat/modules/conv.py index 08a7e0e4..acf579f7 100644 --- a/mqbench/nn/qat/modules/conv.py +++ b/mqbench/nn/qat/modules/conv.py @@ -1,4 +1,5 @@ import torch.nn.qat.modules as nnqat +import torch.nn as nn from mqbench.quantization.default_bias_fake_quant import bias_fake_quantizer @@ -9,3 +10,21 @@ def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, def forward(self, input): return self._conv_forward(input, self.weight_fake_quant(self.weight), self.bias_fake_quant(self.bias)) + +class Conv2d_sophgo(nnqat.Conv2d): + def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True, padding_mode='zeros', qconfig=None, device=None, dtype=None): + super().__init__(in_channels, out_channels, kernel_size, stride=stride, padding=padding, dilation=dilation, groups=groups, bias=bias, padding_mode=padding_mode, qconfig=qconfig) + + def bias_fake_quant_proc(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def forward(self, input): + #bias浼噺鍖 + if self.bias is not None and self.weight_fake_quant.fake_quant_enabled[0] == 1: + in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + self.bias = self.bias_fake_quant_proc(self.bias, self.weight_fake_quant.scale, in_scale) + return self._conv_forward(input, self.weight_fake_quant(self.weight), self.bias) \ No newline at end of file diff --git a/mqbench/prepare_by_platform.py b/mqbench/prepare_by_platform.py index fd850446..866475a5 100644 --- a/mqbench/prepare_by_platform.py +++ b/mqbench/prepare_by_platform.py @@ -53,6 +53,7 @@ class BackendType(Enum): Tengine_u8 = "Tengine_u8" Tensorrt_NLP = "Tensorrt_NLP" Academic_NLP = "Academic_NLP" + Sophgo_TPU = "Sophgo_TPU" ParamsTable = { @@ -121,6 +122,14 @@ class BackendType(Enum): default_act_quantize=LearnableFakeQuantize, default_weight_observer=MinMaxObserver, default_act_observer=EMAMinMaxObserver), + BackendType.Sophgo_TPU: dict(qtype='affine', # noqa: E241 + w_qscheme=QuantizeScheme(symmetry=True, per_channel=True, pot_scale=False, bit=8), + a_qscheme=QuantizeScheme(symmetry=False, per_channel=False, pot_scale=False, bit=8), + # b_qscheme=QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8), + default_weight_quantize=LearnableFakeQuantize, + default_act_quantize=LearnableFakeQuantize, + default_weight_observer=MinMaxObserver, + default_act_observer=EMAMinMaxObserver) } ParamsTable[BackendType.Tensorrt_NLP] = ParamsTable[BackendType.Tensorrt] ParamsTable[BackendType.Academic_NLP] = ParamsTable[BackendType.Academic] @@ -174,6 +183,12 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): 'a_qscheme': { same with w_qscheme. } + "object_type": [ + (torch.add, qconfig) + ], + "module_name": [ + ("conv1", qconfig) + ] } """ w_observer = extra_qparams.get('w_observer', None) @@ -210,7 +225,14 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): **w_qscheme.to_observer_params()) a_config = backend_params['default_act_quantize'].with_args(observer=a_observer, **a_qscheme.to_observer_params()) - return QConfig(activation=a_config, weight=w_config) + qconfig = {'': QConfig(activation=a_config, weight=w_config)} + object_type = extra_qparams.get('object_type', None) + if object_type is not None: + qconfig["object_type"] = object_type + module_name = extra_qparams.get('module_name', None) + if module_name is not None: + qconfig["module_name"] = module_name + return qconfig # Academic setting should specific quant scheme in config. if deploy_backend in [BackendType.Academic, BackendType.Academic_NLP]: @@ -261,8 +283,62 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): if backend_params['qtype'] == 'vitis': logger.info('Bias Qconfig:\n TqtFakeQuantize with MinMaxObserver') - return QConfig(activation=a_qconfig, weight=w_qconfig) + qconfig = {'': QConfig(activation=a_qconfig, weight=w_qconfig)} + if deploy_backend == BackendType.Sophgo_TPU: + qconfig["object_type"] = {torch.nn.Linear:createQConfig(deploy_backend)} + object_type = extra_qparams.get('object_type', None) + if object_type is not None: + if "object_type" in qconfig: + qconfig["object_type"].update(object_type) + else: + qconfig["object_type"] = object_type + + module_name = extra_qparams.get('module_name', None) + if module_name is not None: + qconfig["module_name"] = module_name + return qconfig + +#LearnableFakeQuantize, LearnableFakeQuantize, MinMaxObserver, EMAMinMaxObserver +# 'w_qscheme': { +# 'bit': bitwidth, +# 'symmetry': whether quantize scheme is symmetric, +# 'per_channel': whether quantize scheme is perchannel, +# 'pot_scale': whether scale is power of two. +# } +# 'a_qscheme': { +# 'bit': bitwidth, +# 'symmetry': whether quantize scheme is symmetric, +# 'per_channel': whether quantize scheme is perchannel, +# 'pot_scale': whether scale is power of two. +# } + +def createQConfig(deploy_backend, onlyCreate_WQconfig = True, w_fakequantize = 'LearnableFakeQuantize', a_fakequantize = 'LearnableFakeQuantize', + w_observer = 'MinMaxObserver', a_observer = 'EMAMinMaxObserver', w_qscheme = {}, a_qscheme = {}, + w_fakeq_params = {}, a_fakeq_params = {}, w_observer_extra_args = {}, a_observer_extra_args = {}): + w_observer = ObserverDict[w_observer] + w_fakequantize = FakeQuantizeDict[w_fakequantize] + if w_qscheme is not None: + w_qscheme = QuantizeScheme(**w_qscheme) + else: + w_qscheme = QuantizeScheme(symmetry=True, per_channel=True, pot_scale=False, bit=8) + if deploy_backend == BackendType.Sophgo_TPU: + w_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8) + w_qscheme.kwargs.update(w_observer_extra_args) + w_qconfig = w_fakequantize.with_args(observer=w_observer, **w_fakeq_params, **w_qscheme.to_observer_params()) + + if not onlyCreate_WQconfig: + a_observer = ObserverDict[a_observer] + a_fakequantize = FakeQuantizeDict[a_fakequantize] + if a_qscheme is not None: + a_qscheme = QuantizeScheme(**a_qscheme) + else: + a_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8) + a_qscheme.kwargs.update(a_observer_extra_args) + a_qconfig = a_fakequantize.with_args(observer=a_observer, **a_fakeq_params, **a_qscheme.to_observer_params()) + else: + a_qconfig = torch.nn.Identity + return QConfig(activation=a_qconfig, weight=w_qconfig) class CustomedTracer(Tracer): """ @@ -379,8 +455,10 @@ def prepare_by_platform( if custom_tracer is not None: tracer = custom_tracer graph = tracer.trace(model, concrete_args) + # print('trace graph:',graph) name = model.__class__.__name__ if isinstance(model, torch.nn.Module) else model.__name__ modules = dict(model.named_modules()) + # print('named_modules:',modules) graph, duplicated_modules = duplicate_reused_nodes(graph, modules) constant_nodes = prepare_constant_dict(graph, model) modules.update(duplicated_modules) From ef3f30e0b3f105c1c2d974ae3cb02dbb0c782f90 Mon Sep 17 00:00:00 2001 From: wangxc2006 <68397007@qq.com> Date: Thu, 3 Nov 2022 10:00:15 +0800 Subject: [PATCH 03/29] update, fix some bug --- application/imagenet_example/main.py | 2 + mqbench/convert_deploy.py | 13 +++- mqbench/custom_quantizer/model_quantizer.py | 5 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 15 ++++- mqbench/deploy/deploy_linear.py | 60 +++++++++++++++++-- mqbench/fake_quantize/fp16.py | 24 ++++++++ mqbench/fusion_method.py | 18 ++++++ .../qat/modules/conv_fused_sophgo_tpu.py | 2 +- mqbench/nn/qat/modules/conv.py | 5 +- mqbench/train_all_model.py | 37 ++++++++++++ 10 files changed, 167 insertions(+), 14 deletions(-) create mode 100644 mqbench/fake_quantize/fp16.py create mode 100644 mqbench/train_all_model.py diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index f233e806..c5a60b53 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -194,6 +194,7 @@ def main_worker(gpu, ngpus_per_node, args): else: model = model.cpu() if args.pre_eval_and_export: + print('原始onnx模型精度') validate(val_loader, model.eval(), criterion, args) kwargs = { 'input_shape_dict': {'data': [cali_batch_size, 3, 224, 224]}, @@ -301,6 +302,7 @@ def main_worker(gpu, ngpus_per_node, args): exit(1) if args.evaluate: + print('resume模型精度') from mqbench.convert_deploy import convert_merge_bn module_tmp2 = copy.deepcopy(model) convert_merge_bn(module_tmp2.eval()) diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index 80f79bfa..9fde92e4 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -36,6 +36,7 @@ @register_deploy_function(BackendType.Sophgo_TPU) def convert_merge_bn(model: GraphModule, **kwargs): print('wlog before convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) + print('wlog before convert_merge_bn, model.graph:', model.graph) logger.info("Merge BN for deploy.") nodes = list(model.graph.nodes) modules = dict(model.named_modules()) @@ -44,6 +45,7 @@ def convert_merge_bn(model: GraphModule, **kwargs): if node.target in modules and type(modules[node.target]) in FUSED_MODULE_CONVERT_FUNCTION: FUSED_MODULE_CONVERT_FUNCTION[type(modules[node.target])](model, node) print('wlog after convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) + print('wlog after convert_merge_bn, model.graph:', model.graph) @register_deploy_function(BackendType.Academic_NLP) @register_deploy_function(BackendType.Tensorrt_NLP) @@ -59,7 +61,7 @@ def convert_merge_bn(model: GraphModule, **kwargs): @register_deploy_function(BackendType.OPENVINO) @register_deploy_function(BackendType.Sophgo_TPU) def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_path, **kwargs): - logger.info("Export to onnx.") + logger.info("Export to onnx, onnx_model_path:{}".format(onnx_model_path)) model = model.cpu() output_names = kwargs.get('output_names', []) dynamic_axes = kwargs.get('dynamic_axes', {}) @@ -128,11 +130,16 @@ def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, * file_h.close() cali_table = osp.join(output_path, '{}_cali_table_from_mqbench_sophgo_tpu'.format(model_name)) with open(cali_table, 'w') as f: + f.write("# work_mode:QAT_all_int8 //Automatically generated, do not modify, work_mode choice:[QAT_all_int8, QAT_mix_prec]\n") f.write("# op_name threshold min max\n") ori_layer_names = '' for name,value in blob_range.items(): - f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) - ori_layer_names += '{},'.format(value['ori_name']) + if 'threshold' in value: + f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) + ori_layer_names += '{},'.format(value['ori_name']) + else: + f.write("{} {} {} {} {}\n".format(name, len(value['step']), ' '.join([str(i) for i in value['step']]), + len(value['zero_point']), ' '.join([str(i) for i in value['zero_point']]))) f.write("#{}\n".format(ori_layer_names[0:-1])) diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index 12d3203e..e33403f4 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -35,6 +35,7 @@ from mqbench.utils.registry import register_model_quantizer from mqbench.prepare_by_platform import BackendType import mqbench.nn.intrinsic.qat as qnniqat +import mqbench.nn.qat as qnnqat @register_model_quantizer(BackendType.Tensorrt) @register_model_quantizer(BackendType.NNIE) @@ -100,6 +101,8 @@ def _fix_succ_recursivly(self, model, node, fake_quantizer, args, target_node, i _tmp[_i] = inserted_node modules_has_bias = (qnniqat.ConvBnReLU2d_sophgo, qnniqat.ConvBn2d_sophgo, + qnniqat.ConvReLU2d_sophgo, + qnnqat.Conv2d_sophgo, qnniqat.LinearReLU_sophgo, qnniqat.Linear_sophgo) modules = dict(model.named_modules()) @@ -241,7 +244,7 @@ def _find_act_quants(self, model: GraphModule) -> List: if (node.op == "call_module" and isinstance(modules[node.target], self.module_type_to_quant_input)) or \ ((node.op == 'call_function' or node.op == 'call_method') and node.target in self.function_type_to_quant_input) or node.name in self.additional_node_name: - input_node_list = self._flatten_args(node.args) + input_node_list = self._flatten_args(node.all_input_nodes) #wxc # Means this is not Tensor + Tensor. if not all([isinstance(_node, torch.fx.node.Node) for _node in input_node_list]): continue diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index 8b914030..0e8603a9 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -1,3 +1,4 @@ +import operator import torch from torch.fx import GraphModule import torch.nn.intrinsic as nni @@ -6,6 +7,7 @@ from mqbench.prepare_by_platform import BackendType from mqbench.custom_quantizer import ModelQuantizer import torch.nn as nn +import mqbench.nn.qat as qnnqat @register_model_quantizer(BackendType.Sophgo_TPU) @@ -23,7 +25,8 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): # Intrinsic modules: nni.ConvBn2d: qnniqat.ConvBn2d_sophgo, nni.ConvBnReLU2d: qnniqat.ConvBnReLU2d_sophgo, - # nni.ConvReLU2d: qnniqat.ConvReLU2d_sophgo, + nn.Conv2d: qnnqat.Conv2d_sophgo, + nni.ConvReLU2d: qnniqat.ConvReLU2d_sophgo, nni.LinearReLU: qnniqat.LinearReLU_sophgo, nn.Linear: qnniqat.Linear_sophgo } @@ -32,11 +35,19 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): def module_type_to_quant_input(self) -> tuple: return super().module_type_to_quant_input + ( qnniqat.ConvBnReLU2d_sophgo, + qnniqat.ConvReLU2d_sophgo, qnniqat.ConvBn2d_sophgo, + qnnqat.Conv2d_sophgo, qnniqat.LinearReLU_sophgo, - qnniqat.Linear_sophgo + qnniqat.Linear_sophgo, ) + @property + def function_type_to_quant_input(self) -> tuple: + return super().function_type_to_quant_input + [ + torch.cat + ] + @property def _passed_func_type(self): return ( diff --git a/mqbench/deploy/deploy_linear.py b/mqbench/deploy/deploy_linear.py index 0ab923fd..8129414f 100644 --- a/mqbench/deploy/deploy_linear.py +++ b/mqbench/deploy/deploy_linear.py @@ -143,6 +143,34 @@ def find_the_closest_clip_range(node): logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') return clip_ranges + def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): + def find_the_closest_clip_range(node): + input_0 = node.input[0] + tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') + if tensor_name[-1] == '_': + tensor_name = tensor_name[:-1] + + if tensor_name in clip_ranges: + return tensor_name + elif node.op_type in ['Flatten', 'Resize'] and node.output[0] in inp2node: + return find_the_closest_clip_range(inp2node[node.output[0]][0][0]) + else: + return None + + for node in graph.node: + if node.op_type in ['Flatten', 'Resize']: + tensor_name = find_the_closest_clip_range(node) + if tensor_name: + old = clip_ranges[tensor_name] + new_name = node.input[0] + new_name = '{}_{}'.format(new_name, out2node[new_name].op_type if new_name in out2node else '') + if new_name[-1] == '_': + new_name = tensor_name[:-1] + clip_ranges[new_name] = copy.deepcopy(old) + clip_ranges[new_name]['ori_name'] = new_name + logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') + return clip_ranges + def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend): model = onnx.load(onnx_path) graph = model.graph @@ -161,6 +189,10 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) nodes_to_be_removed.append(node) nodes_to_be_removed.extend(get_constant_inputs(node, out2node)) + if node.output[0] not in inp2node: + assert node.output[0] in [l.name for l in graph.output] + inp2node[node.output[0]] = [] + next_nodes = inp2node[node.output[0]] if node.op_type in PERCHANNEL_FAKEQUANTIZER: # fake quantize for weights, suppose per-channel quantize only for weight redundant_nodes = self.deal_with_weight_fakequant(node, out2node, inp2node, named_initializer) @@ -178,19 +210,36 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) elif backend == 'vitis': logger.info("Vitis-DPU does not support per-channel quatization.") raise NotImplementedError("Vitis-DPU does not support per-channel quatization.") + elif backend == 'sophgo_tpu': - + if len(next_nodes) == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']:#?????????1???,??1??????conv?? + next_node_output = next_nodes[0][0].output[0] #???????1???????1???tensor + if next_node_output in inp2node: + print(next_node_output, 'not in inp2node') + if len(inp2node[next_node_output]) == 0: + print(next_node_output, ',users not exsit') + if inp2node[next_node_output][0][0].op_type == 'Relu':##???????1???conv????1??????Relu(fake->conv->relu) + tensor_name = '{}_{}'.format(inp2node[next_node_output][0][0].output[0], 'Relu') + else: + tensor_name = '{}_{}'.format(next_node_output, next_nodes[0][0].op_type) + tensor_name += '_{}'.format('weight' if next_nodes[0][1] == 1 else 'bias' ) + clip_ranges[tensor_name] = {'step': [float(x) for x in scale], + 'zero_point': [int(x) for x in zero_point] + } elif node.op_type in PERTENSOR_FAKEQUANTIZER: - if node.output[0] not in inp2node: - assert node.output[0] in [l.name for l in graph.output] - inp2node[node.output[0]] = [] - next_nodes = inp2node[node.output[0]] if len(next_nodes) == 1 and next_nodes[0][1] == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']: # fake quantize for weights redundant_nodes = self.deal_with_weight_fakequant(node, out2node, inp2node, named_initializer) tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) nodes_to_be_removed.extend(redundant_nodes) self.clip_weight(node, name2data, inp2node, named_initializer) + if backend == 'sophgo_tpu': + assert next_nodes[0][0].op_type == 'Gemm' + tensor_name += '{}_{}_weight'.format(inp2node[node.output[0]][0][0].output[0], inp2node[node.output[0]][0][0].op_type) + clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #??????????? + 'min': float(scale * (qmin - zero_point)), + 'max': float(scale * (qmax - zero_point)), + 'ori_name':pre_layer_name} elif len(next_nodes) == 1 and next_nodes[0][1] == 2 and next_nodes[0][0].op_type in ['Gemm', 'Conv']: # fake quantize for bias assert backend == 'vitis' @@ -266,6 +315,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) elif backend == 'ppl-cuda': context = {'ppl-cuda': clip_ranges} elif backend == 'sophgo_tpu': + clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node) context = {'sophgo_tpu': clip_ranges} context['w_qscheme'] = '' context['a_qscheme'] = '' diff --git a/mqbench/fake_quantize/fp16.py b/mqbench/fake_quantize/fp16.py new file mode 100644 index 00000000..829fb5b2 --- /dev/null +++ b/mqbench/fake_quantize/fp16.py @@ -0,0 +1,24 @@ +import torch + + +class Fp16FakeQuantize(): + def __init__(self): + pass + + @torch.jit.export + def extra_repr(self): + return 'Fp16FakeQuantize' + + def forward(self, X): + #璋冪敤鑷畾涔塼orch c++ op灏唂p32鐨刋杞负fp16鍚庡啀杞細fp32锛屽紩鍏ヨ宸 + +class BF16FakeQuantize(): + def __init__(self): + pass + + @torch.jit.export + def extra_repr(self): + return 'BF16FakeQuantize' + + def forward(self, X): + #fp32鐨刋杞负bf16鍚庡啀杞細fp32锛屽紩鍏ヨ宸 \ No newline at end of file diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index e879db28..b90593db 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -24,6 +24,24 @@ def convert_qnniqat_linear(model, fused_node): linear.weight_fake_quant = fused_module.weight_fake_quant linear_parent_name, linear_name = _parent_name(fused_node.target) setattr(modules[linear_parent_name], linear_name, linear) +@register_convert_function(qnnqat.Conv2d_sophgo) +def convert_qnniqat_conv2d(model, fused_node): + modules = dict(model.named_modules()) + fused_module = modules[fused_node.target] + # Create a Conv2d from FusedModule. + conv = torch.nn.Conv2d(fused_module.in_channels, fused_module.out_channels, fused_module.kernel_size, + fused_module.stride, fused_module.padding, fused_module.dilation, + fused_module.groups, fused_module.bias is not None, fused_module.padding_mode) + conv.weight = fused_module.weight + if fused_module.bias is not None: + conv.bias = fused_module.bias + # We need nn.qat.conv here to export weight quantize node. + conv.qconfig = fused_module.qconfig + conv = torch.nn.qat.Conv2d.from_float(conv) + # Attach weight fake quantize params. + conv.weight_fake_quant = fused_module.weight_fake_quant + conv_parent_name, conv_name = _parent_name(fused_node.target) + setattr(modules[conv_parent_name], conv_name, conv) @register_convert_function(qnniqat.LinearReLU_sophgo) def linearert_qnniqat_linearrelu(model, fused_node): convert_qnniqat_linear(model, fused_node) diff --git a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py index 52b161f5..2e6882da 100644 --- a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py @@ -439,7 +439,7 @@ def __init__(self, in_channels, out_channels, kernel_size, stride=1, def forward(self, input): # print('xxx3') - return self.forward(input) + return F.relu(qnnqat.Conv2d_sophgo.forward(self, input)) @classmethod def from_float(cls, mod): diff --git a/mqbench/nn/qat/modules/conv.py b/mqbench/nn/qat/modules/conv.py index acf579f7..6fed17eb 100644 --- a/mqbench/nn/qat/modules/conv.py +++ b/mqbench/nn/qat/modules/conv.py @@ -24,7 +24,8 @@ def bias_fake_quant_proc(self, bias, scale_w, in_scale): def forward(self, input): #bias浼噺鍖 + bias = self.bias if self.bias is not None and self.weight_fake_quant.fake_quant_enabled[0] == 1: in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale - self.bias = self.bias_fake_quant_proc(self.bias, self.weight_fake_quant.scale, in_scale) - return self._conv_forward(input, self.weight_fake_quant(self.weight), self.bias) \ No newline at end of file + bias = self.bias_fake_quant_proc(self.bias, self.weight_fake_quant.scale, in_scale) + return self._conv_forward(input, self.weight_fake_quant(self.weight), bias) \ No newline at end of file diff --git a/mqbench/train_all_model.py b/mqbench/train_all_model.py new file mode 100644 index 00000000..adf6805f --- /dev/null +++ b/mqbench/train_all_model.py @@ -0,0 +1,37 @@ +import os +import time +import argparse +from multiprocessing import Pool + +parser_auto_cali = argparse.ArgumentParser(description='uto_cali params.', conflict_handler='resolve') +parser = argparse.ArgumentParser(description='auto_cali_test params.') +parser.add_argument('--debug_cmd', type=str, default='onnx,sym', help='exclude') +opt = parser.parse_args() + +model_list_all=[ + #"--arch=shufflenet_v2_x0_5 --batch-size=320", + "--arch=mobilenet_v2 --batch-size=64", + "--arch=resnet18 --batch-size=128", + "--arch=vgg11_bn --batch-size=32", + "--arch=resnet50 --batch-size=32", + #"--arch=squeezenet1_1 --batch-size=128", + + #"--arch=mobilenet_v3_small --batch-size=128" +] + +cmd_str = "--epochs=10 --lr=1e-4 --gpu=0 --pretrained --evaluate --backend=sophgo_tpu --optim=sgd --pre_eval_and_export --train_data=/data/imagenet/for_train_val/ --val_data=/data/imagenet/for_train_val/ --output_path=/workspace/tmp_path_1024"# --fast_test" + +def worker(cmd_line): + os.system(cmd_line) + +if __name__ == "__main__": + wp = os.getcwd() + po = Pool(1) + for i,model in enumerate(model_list_all): + cmd_line = 'cd {};python3 imagenet_example/main.py {} {} >{}_log_train_{} 2>&1'.format(wp, model, cmd_str, i, model.split(' ')[0].split('=')[1]) + print('start', model) + po.apply_async(worker, (cmd_line,)) + + po.close() + po.join() + print('all end') From 96683b69e6a90b1eae0face4eb02e6f8a79f404d Mon Sep 17 00:00:00 2001 From: wangxc2006 <68397007@qq.com> Date: Thu, 3 Nov 2022 17:33:25 +0800 Subject: [PATCH 04/29] add deconv and refine SophgoTpuQuantizer --- mqbench/custom_quantizer/model_quantizer.py | 23 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 45 +- mqbench/fusion_method.py | 36 +- mqbench/nn/intrinsic/qat/modules/__init__.py | 1 + .../qat/modules/conv_fused_sophgo_tpu.py | 4 - .../qat/modules/deconv_fused_sophgo_tpu.py | 426 ++++++++++++++++++ .../qat/modules/linear_fused_sophgo_tpu.py | 25 +- mqbench/nn/qat/modules/__init__.py | 2 +- mqbench/nn/qat/modules/deconv.py | 54 +++ 9 files changed, 567 insertions(+), 49 deletions(-) create mode 100644 mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index e33403f4..aa3d0789 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -86,36 +86,25 @@ def _insert_fake_quantize_for_act_quant( with graph.inserting_after(node): inserted_node = graph.create_node("call_module", quantizer_name, (node,), {}) for _node in nodes: - _node.args = self._fix_succ_recursivly(model, _node, fake_quantizer, _node.args, node, inserted_node) + _node.args = self._fix_succ_recursivly(_node.args, node, inserted_node) model.recompile() model.graph.lint() return model - def _fix_succ_recursivly(self, model, node, fake_quantizer, args, target_node, inserted_node): + def _fix_succ_recursivly(self, args, target_node, inserted_node): # List / Tuple if isinstance(args, (list, tuple)): _tmp = list(args) for _i, _arg in enumerate(args): if _arg == target_node: _tmp[_i] = inserted_node - modules_has_bias = (qnniqat.ConvBnReLU2d_sophgo, - qnniqat.ConvBn2d_sophgo, - qnniqat.ConvReLU2d_sophgo, - qnnqat.Conv2d_sophgo, - qnniqat.LinearReLU_sophgo, - qnniqat.Linear_sophgo) - modules = dict(model.named_modules()) - if (node.op == "call_module" and isinstance(modules[node.target], modules_has_bias)): - setattr(modules[node.target], "input_fake_quantizer", fake_quantizer) - print('wlog:', node.target,'\'type is:', type(modules[node.target]), "add its new attr:input_fake_quantizer", ',its old pre op is:', - target_node.name, target_node.type, ',its new insert pre op is:', inserted_node.name) elif isinstance(_arg, tuple): - _tmp[_i] = self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node) + _tmp[_i] = self._fix_succ_recursivly(_arg, target_node, inserted_node) elif isinstance(_arg, list): - _tmp[_i] = list(self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node)) + _tmp[_i] = list(self._fix_succ_recursivly(_arg, target_node, inserted_node)) elif isinstance(_arg, dict): - _tmp[_i] = self._fix_succ_recursivly(model, node, fake_quantizer, _arg, target_node, inserted_node) + _tmp[_i] = self._fix_succ_recursivly(_arg, target_node, inserted_node) return tuple(_tmp) # Dict elif isinstance(args, dict): @@ -124,7 +113,7 @@ def _fix_succ_recursivly(self, model, node, fake_quantizer, args, target_node, i if v == target_node: _tmp[k] = inserted_node elif not isinstance(v, torch.fx.node.Node): - _tmp[k] = self._fix_succ_recursivly(model, node, fake_quantizer, v, target_node, inserted_node) + _tmp[k] = self._fix_succ_recursivly(v, target_node, inserted_node) else: _tmp[k] = v return _tmp diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index 0e8603a9..eb4253ae 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -3,6 +3,7 @@ from torch.fx import GraphModule import torch.nn.intrinsic as nni import mqbench.nn.intrinsic.qat as qnniqat +import mqbench.nn.intrinsic as qnni from mqbench.utils.registry import register_model_quantizer from mqbench.prepare_by_platform import BackendType from mqbench.custom_quantizer import ModelQuantizer @@ -28,19 +29,16 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): nn.Conv2d: qnnqat.Conv2d_sophgo, nni.ConvReLU2d: qnniqat.ConvReLU2d_sophgo, nni.LinearReLU: qnniqat.LinearReLU_sophgo, - nn.Linear: qnniqat.Linear_sophgo + nn.Linear: qnniqat.Linear_sophgo, + qnni.LinearBn1d: qnniqat.LinearBn1d_sophgo, + qnni.ConvTransposeBnReLU2d:qnniqat.ConvTransposeBnReLU2d_sophgo, + qnni.ConvTransposeReLU2d:qnniqat.ConvTransposeReLU2d_sophgo, + qnni.ConvTransposeBn2d:qnniqat.ConvTransposeBn2d_sophgo, } @property def module_type_to_quant_input(self) -> tuple: - return super().module_type_to_quant_input + ( - qnniqat.ConvBnReLU2d_sophgo, - qnniqat.ConvReLU2d_sophgo, - qnniqat.ConvBn2d_sophgo, - qnnqat.Conv2d_sophgo, - qnniqat.LinearReLU_sophgo, - qnniqat.Linear_sophgo, - ) + return super().module_type_to_quant_input + self._layers_need_scale_form_input_fake_quantizer @property def function_type_to_quant_input(self) -> tuple: @@ -63,6 +61,22 @@ def _passed_module_type(self): torch.nn.ReLU6 ) + @property + def _layers_need_scale_form_input_fake_quantizer(self): + return ( + qnniqat.ConvBnReLU2d_sophgo, #todo:add transposeConv support + qnniqat.ConvBn2d_sophgo, + qnniqat.ConvReLU2d_sophgo, + qnnqat.Conv2d_sophgo, + qnniqat.LinearReLU_sophgo, + qnniqat.Linear_sophgo, + ) + + def prepare(self, model: GraphModule, qconfig): + model = super().prepare(model, qconfig) + model = self._set_fake_quantizer_to_next_weight_layer(model) + return model + def _find_act_quants(self, model: GraphModule) -> list: nodes = list(model.graph.nodes) modules = dict(model.named_modules()) @@ -82,4 +96,15 @@ def _find_act_quants(self, model: GraphModule) -> list: node_need_to_quantize_output.append(node) else: node_need_to_quantize_output.append(next_node) - return node_need_to_quantize_output \ No newline at end of file + return node_need_to_quantize_output + + def _set_fake_quantizer_to_next_weight_layer(self, model: GraphModule): + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + for node in nodes: + if node.target in modules and "_post_act_fake_quantizer" in node.target: + fake_quantizer = getattr(model, node.target) + for user in node.users: + if (user.op == "call_module" and isinstance(modules[user.target], self._layers_need_scale_form_input_fake_quantizer)): + setattr(modules[user.target], "input_fake_quantizer", fake_quantizer) + print('wlog:', user.target,'\'type is:', type(modules[user.target]), "add input_fake_quantizer") \ No newline at end of file diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index b90593db..3c0ba997 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -24,8 +24,9 @@ def convert_qnniqat_linear(model, fused_node): linear.weight_fake_quant = fused_module.weight_fake_quant linear_parent_name, linear_name = _parent_name(fused_node.target) setattr(modules[linear_parent_name], linear_name, linear) + @register_convert_function(qnnqat.Conv2d_sophgo) -def convert_qnniqat_conv2d(model, fused_node): +def convert_qnnqat_conv2d(model, fused_node): modules = dict(model.named_modules()) fused_module = modules[fused_node.target] # Create a Conv2d from FusedModule. @@ -42,6 +43,29 @@ def convert_qnniqat_conv2d(model, fused_node): conv.weight_fake_quant = fused_module.weight_fake_quant conv_parent_name, conv_name = _parent_name(fused_node.target) setattr(modules[conv_parent_name], conv_name, conv) + +@register_convert_function(qnnqat.ConvTranspose2d_sophgo) +def convert_qnnqat_deconv2d(model, fused_node): + modules = dict(model.named_modules()) + fused_module = modules[fused_node.target] + # Create a ConvTranspose2d from FusedModule. + deconv = torch.nn.ConvTranspose2d(fused_module.in_channels, fused_module.out_channels, fused_module.kernel_size, + stride=fused_module.stride, padding=fused_module.padding, output_padding=fused_module.output_padding, + groups=fused_module.groups, bias=fused_module.bias is not None, + dilation=fused_module.dilation, + padding_mode=fused_module.padding_mode) + deconv.weight = fused_module.weight + if fused_module.bias is not None: + deconv.bias = fused_module.bias + fused_deconv = fuse_deconv_bn_eval(deconv.eval(), fused_module.bn) + # We need nn.qat.conv here to export weight quantize node. + fused_deconv.qconfig = fused_module.qconfig + fused_deconv = qnnqat.ConvTranspose2d.from_float(fused_deconv) + # Attach weight fake quantize params. + fused_deconv.weight_fake_quant = fused_module.weight_fake_quant + deconv_parent_name, deconv_name = _parent_name(fused_node.target) + setattr(modules[deconv_parent_name], deconv_name, fused_deconv) + @register_convert_function(qnniqat.LinearReLU_sophgo) def linearert_qnniqat_linearrelu(model, fused_node): convert_qnniqat_linear(model, fused_node) @@ -67,6 +91,7 @@ def linearert_qnniqat_linearrelu(model, fused_node): _node.args = tuple(_tmp) model.recompile() model.graph.lint() + @register_convert_function(qnni.LinearBn1d) def convert_qnni_linearbn(model, fused_node): modules = dict(model.named_modules()) @@ -89,10 +114,10 @@ def convert_qnniqat_linearbn(model, fused_node): # Merge Linear + BN fused_linear = fuse_linear_bn_eval(linear.eval(), fused_module.bn) # We need nn.qat.linear here to export weight quantize node. - linear.qconfig = fused_module.qconfig - linear = torch.nn.qat.Linear.from_float(linear) + fused_linear.qconfig = fused_module.qconfig + fused_linear = torch.nn.qat.Linear.from_float(fused_linear) # Attach weight fake quantize params. - linear.weight_fake_quant = fused_module.weight_fake_quant + fused_linear.weight_fake_quant = fused_module.weight_fake_quant linear_parent_name, linear_name = _parent_name(fused_node.target) setattr(modules[linear_parent_name], linear_name, fused_linear) @@ -195,6 +220,7 @@ def convert_qnni_deconvbn(model, fused_node): @register_convert_function(qnniqat.ConvTransposeFreezebn2d) @register_convert_function(qnniqat.ConvTransposeBn2d) +@register_convert_function(qnniqat.ConvTransposeBn2d_sophgo) def convert_qnniqat_deconvbn(model, fused_node): modules = dict(model.named_modules()) fused_module = modules[fused_node.target] @@ -219,6 +245,7 @@ def convert_qnniqat_deconvbn(model, fused_node): @register_convert_function(qnni.ConvTransposeFreezebnReLU2d) @register_convert_function(qnni.ConvTransposeBnReLU2d) +@register_convert_function(qnniqat.ConvTransposeBnReLU2d_sophgo) def convert_qnni_deconvbnrelu(model, fused_node): convert_qnni_deconvbn(model, fused_node) modules = dict(model.named_modules()) @@ -247,6 +274,7 @@ def convert_qnni_deconvbnrelu(model, fused_node): @register_convert_function(qnniqat.ConvTransposeFreezebnReLU2d) @register_convert_function(qnniqat.ConvTransposeBnReLU2d) +@register_convert_function(qnniqat.ConvTransposeBnReLU2d_sophgo) def convert_qnniqat_deconvbnrelu(model, fused_node): convert_qnniqat_deconvbn(model, fused_node) modules = dict(model.named_modules()) diff --git a/mqbench/nn/intrinsic/qat/modules/__init__.py b/mqbench/nn/intrinsic/qat/modules/__init__.py index c469b714..cd4dd80a 100644 --- a/mqbench/nn/intrinsic/qat/modules/__init__.py +++ b/mqbench/nn/intrinsic/qat/modules/__init__.py @@ -4,3 +4,4 @@ from .freezebn import ConvFreezebn2d, ConvFreezebnReLU2d, ConvTransposeFreezebn2d, ConvTransposeFreezebnReLU2d from .conv_fused_sophgo_tpu import ConvBnReLU2d_sophgo, ConvBn2d_sophgo, ConvReLU2d_sophgo from .linear_fused_sophgo_tpu import LinearBn1d_sophgo, LinearReLU_sophgo, Linear_sophgo +from .deconv_fused_sophgo_tpu import ConvTransposeBnReLU2d_sophgo, ConvTransposeBn2d_sophgo, ConvTransposeReLU2d_sophgo diff --git a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py index 2e6882da..6d9f3f38 100644 --- a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py @@ -129,15 +129,12 @@ def bias_fake_quant_proc(self, bias, scale_w, in_scale): scale = scale_w*in_scale if torch.nonzero(scale).size()[0] != scale.numel(): print('error! scale has 0, scale:', scale) - scale[torch.abs(scale) < 1e-10] = 1e-10 - print('new scale:', scale) bias_q = bias/scale bias = (bias_q.round()-bias_q).detach() + bias_q bias = bias*scale return bias def _forward(self, input): - # print('xxx2') assert self.bn.running_var is not None running_std = torch.sqrt(self.bn.running_var + self.bn.eps) scale_factor = self.bn.weight / running_std @@ -438,7 +435,6 @@ def __init__(self, in_channels, out_channels, kernel_size, stride=1, qconfig=qconfig) def forward(self, input): - # print('xxx3') return F.relu(qnnqat.Conv2d_sophgo.forward(self, input)) @classmethod diff --git a/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py new file mode 100644 index 00000000..717ce582 --- /dev/null +++ b/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py @@ -0,0 +1,426 @@ +import math + +import torch +import torch.nn as nn +import torch.nn.functional as F +from torch.nn import init +from torch.nn.intrinsic import _FusedModule +from torch.nn.parameter import Parameter +from torch.nn.modules.utils import _pair, _single + +from typing import TypeVar + +import mqbench.nn.intrinsic as qnni +import mqbench.nn.qat as qnnqat +from mqbench.utils.fusion import fuse_deconv_bn_weights + + +_BN_CLASS_MAP = { + 1: nn.BatchNorm1d, + 2: nn.BatchNorm2d, + 3: nn.BatchNorm3d, +} + +MOD = TypeVar('MOD', bound=nn.modules.conv._ConvTransposeNd) + + +class _ConvTransposeBnNd(nn.modules.conv._ConvTransposeNd, _FusedModule): + + _version = 2 + _FLOAT_MODULE = MOD + + def __init__( + self, + # ConvTransposeBnNd args + in_channels, + out_channels, + kernel_size, + stride, + bias, + transposed, + padding, + output_padding, + groups, + dilation, + padding_mode, + # bn args + # BatchNormNd args + # num_features: out_channels + eps=1e-05, + momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None, + dim=2): + kernel_size = _single(kernel_size) + stride = _single(stride) + padding = _single(padding) + dilation = _single(dilation) + output_padding = _single(output_padding) + nn.modules.conv._ConvTransposeNd.__init__(self, in_channels, + out_channels, kernel_size, + stride, padding, dilation, + transposed, output_padding, + groups, False, padding_mode) + assert qconfig, 'qconfig must be provided for a QAT module' + self.qconfig = qconfig + self.freeze_bn = freeze_bn if self.training else True + self.bn = _BN_CLASS_MAP[dim](out_channels, eps, momentum, True, True) + self.weight_fake_quant = self.qconfig.weight() + # ConvTranspose do per-channel quantize on output channel. + if self.weight_fake_quant.ch_axis != -1: + self.weight_fake_quant.ch_axis = 1 + self.weight_fake_quant.activation_post_process.ch_axis = 1 + if bias: + self.bias = Parameter(torch.Tensor(out_channels)) + else: + self.register_parameter('bias', None) + self.reset_bn_parameters() + + if self.training: + if freeze_bn: + self.freeze_bn_stats() + else: + self.update_bn_stats() + else: + self.freeze_bn_stats() + + def reset_running_stats(self): + self.bn.reset_running_stats() + + def reset_bn_parameters(self): + self.bn.reset_running_stats() + init.uniform_(self.bn.weight) + init.zeros_(self.bn.bias) + # note: below is actully for conv, not BN + if self.bias is not None: + fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight) + bound = 1 / math.sqrt(fan_in) + init.uniform_(self.bias, -bound, bound) + + def reset_parameters(self): + super(_ConvTransposeBnNd, self).reset_parameters() + + def update_bn_stats(self): + self.freeze_bn = False + self.bn.training = True + return self + + def freeze_bn_stats(self): + self.freeze_bn = True + self.bn.training = False + return self + + # def _forward(self, input): + # assert self.bn.running_var is not None + # running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + # scale_factor = self.bn.weight / running_std + # weight_shape = [1] * len(self.weight.shape) + # weight_shape[1] = -1 + # bias_shape = [1] * len(self.weight.shape) + # bias_shape[1] = -1 + # scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + # # using zero bias here since the bias for original conv + # # will be added later + # if self.bias is not None: + # zero_bias = torch.zeros_like(self.bias) + # else: + # zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + # deconv = self._convtransposed_forward(input, scaled_weight, zero_bias) + # deconv_orig = deconv / scale_factor.reshape(bias_shape) + # if self.bias is not None: + # deconv_orig = deconv_orig + self.bias.reshape(bias_shape) + # deconv = self.bn(deconv_orig) + # return deconv + + def bias_fake_quant_proc(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('error! scale has 0, scale:', scale) + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def _forward(self, input): + assert self.bn.running_var is not None + running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + scale_factor = self.bn.weight / running_std + weight_shape = [1] * len(self.weight.shape) + weight_shape[1] = -1 + bias_shape = [1] * len(self.weight.shape) + bias_shape[1] = -1 + scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + if self.weight_fake_quant.fake_quant_enabled[0] == 1: + _, fused_bias = fuse_deconv_bn_weights(self.weight, self.bias, + self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) + in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) + diff_fused_bias = fused_bias - scale_fused_bias + + if self.bias is not None: + zero_bias = torch.zeros_like(self.bias) + else: + zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + conv = self._convtransposed_forward(input, scaled_weight, zero_bias) + conv_orig = conv / scale_factor.reshape(bias_shape) + if self.bias is not None: + conv_orig = conv_orig + self.bias.reshape(bias_shape) + conv = self.bn(conv_orig) + if self.weight_fake_quant.fake_quant_enabled[0] == 1: + conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 + return conv + + def _convtransposed_forward(self, x, w, b): + raise NotImplementedError( + 'The sub-class must implement this function to forward in the needed dim-version!' + ) + + def extra_repr(self): + # TODO(jerryzh): extend + return super(_ConvTransposeBnNd, self).extra_repr() + + def forward(self, input): + return self._forward(input) + + def train(self, mode=True): + """ + Batchnorm's training behavior is using the self.training flag. Prevent + changing it if BN is frozen. This makes sure that calling `model.train()` + on a model with a frozen BN will behave properly. + """ + self.training = mode + if not self.freeze_bn: + for module in self.children(): + module.train(mode) + return self + + # ===== Serialization version history ===== + # + # Version 1/None + # self + # |--- weight : Tensor + # |--- bias : Tensor + # |--- gamma : Tensor + # |--- beta : Tensor + # |--- running_mean : Tensor + # |--- running_var : Tensor + # |--- num_batches_tracked : Tensor + # + # Version 2 + # self + # |--- weight : Tensor + # |--- bias : Tensor + # |--- bn : Module + # |--- weight : Tensor (moved from v1.self.gamma) + # |--- bias : Tensor (moved from v1.self.beta) + # |--- running_mean : Tensor (moved from v1.self.running_mean) + # |--- running_var : Tensor (moved from v1.self.running_var) + # |--- num_batches_tracked : Tensor (moved from v1.self.num_batches_tracked) + def _load_from_state_dict(self, state_dict, prefix, local_metadata, strict, + missing_keys, unexpected_keys, error_msgs): + version = local_metadata.get('version', None) + if version is None or version == 1: + # BN related parameters and buffers were moved into the BN module for v2 + v2_to_v1_names = { + 'bn.weight': 'gamma', + 'bn.bias': 'beta', + 'bn.running_mean': 'running_mean', + 'bn.running_var': 'running_var', + 'bn.num_batches_tracked': 'num_batches_tracked', + } + for v2_name, v1_name in v2_to_v1_names.items(): + if prefix + v1_name in state_dict: + state_dict[prefix + v2_name] = state_dict[prefix + v1_name] + state_dict.pop(prefix + v1_name) + elif prefix + v2_name in state_dict: + # there was a brief period where forward compatibility + # for this module was broken (between + # https://github.com/pytorch/pytorch/pull/38478 + # and https://github.com/pytorch/pytorch/pull/38820) + # and modules emitted the v2 state_dict format while + # specifying that version == 1. This patches the forward + # compatibility issue by allowing the v2 style entries to + # be used. + pass + elif strict: + missing_keys.append(prefix + v2_name) + + super(_ConvTransposeBnNd, + self)._load_from_state_dict(state_dict, prefix, local_metadata, + strict, missing_keys, + unexpected_keys, error_msgs) + + @classmethod + def from_float(cls, mod): + r"""Create a qat module from a float module or qparams_dict + + Args: `mod` a float module, either produced by torch.quantization utilities + or directly from user + """ + assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + \ + cls._FLOAT_MODULE.__name__ + assert hasattr( + mod, 'qconfig'), 'Input float module must have qconfig defined' + assert mod.qconfig, 'Input float module must have a valid qconfig' + qconfig = mod.qconfig + deconv, bn = mod[0], mod[1] + qat_deconvbn = cls(deconv.in_channels, deconv.out_channels, + deconv.kernel_size, deconv.stride, deconv.bias + is not None, deconv.transposed, deconv.padding, + deconv.output_padding, deconv.groups, + deconv.dilation, deconv.padding_mode, bn.eps, + bn.momentum, False, qconfig) + qat_deconvbn.weight = deconv.weight + qat_deconvbn.bias = deconv.bias + qat_deconvbn.bn.weight = bn.weight + qat_deconvbn.bn.bias = bn.bias + qat_deconvbn.bn.running_mean = bn.running_mean + qat_deconvbn.bn.running_var = bn.running_var + # mypy error: Cannot determine type of 'num_batches_tracked' + qat_deconvbn.bn.num_batches_tracked = bn.num_batches_tracked # type: ignore[has-type] + return qat_deconvbn + + +class ConvTransposeBn2d_sophgo(_ConvTransposeBnNd, nn.ConvTranspose2d): + _FLOAT_MODULE = qnni.ConvTransposeBn2d + + def __init__( + self, + # ConvTransposeBnNd args + in_channels, + out_channels, + kernel_size, + stride=1, + bias=None, + transposed=True, + padding=0, + output_padding=0, + groups=1, + dilation=1, + padding_mode='zeros', + # bn args + # BatchNormNd args + # num_features: out_channels + eps=1e-05, + momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None): + kernel_size = _pair(kernel_size) + stride = _pair(stride) + padding = _pair(padding) + dilation = _pair(dilation) + _ConvTransposeBnNd.__init__(self, in_channels, out_channels, + kernel_size, stride, bias, transposed, + padding, output_padding, groups, dilation, + padding_mode, eps, momentum, freeze_bn, + qconfig) + + def _convtransposed_forward(self, x, w, b): + output_padding = self._output_padding(x, None, self.stride, + self.padding, self.kernel_size, + self.dilation) + return F.conv_transpose2d(x, w, b, self.stride, self.padding, + output_padding, self.groups, self.dilation) + + +class ConvTransposeBnReLU2d_sophgo(ConvTransposeBn2d_sophgo): + _FLOAT_MODULE = qnni.ConvTransposeBnReLU2d + + def __init__( + self, + # ConvTransposeBnNd args + in_channels, + out_channels, + kernel_size, + stride=1, + bias=None, + transposed=True, + padding=0, + output_padding=0, + groups=1, + dilation=1, + padding_mode='zeros', + # bn args + # BatchNormNd args + # num_features: out_channels + eps=1e-05, + momentum=0.1, + # affine: True + # track_running_stats: True + # Args for this module + freeze_bn=False, + qconfig=None): + # super(ConvTransposeBnReLU2d, self).__init__(in_channels, out_channels, kernel_size, stride, + # padding, dilation, groups, bias, + # padding_mode, eps, momentum, + # freeze_bn, + # qconfig) + super(ConvTransposeBnReLU2d_sophgo, + self).__init__(in_channels, + out_channels, + kernel_size, + stride=stride, + bias=bias, + transposed=transposed, + padding=padding, + output_padding=output_padding, + groups=groups, + dilation=dilation, + padding_mode=padding_mode, + eps=eps, + momentum=momentum, + freeze_bn=freeze_bn, + qconfig=qconfig) + + def forward(self, input): + return F.relu(ConvTransposeBn2d_sophgo._forward(self, input)) + + @classmethod + def from_float(cls, mod): + return super(ConvTransposeBnReLU2d_sophgo, cls).from_float(mod) + + +class ConvTransposeReLU2d_sophgo(qnnqat.ConvTranspose2d_sophgo): + _FLOAT_MODULE = qnni.ConvTransposeReLU2d + _FLOAT_DECONV_MODULE = nn.ConvTranspose2d + _FLOAT_BN_MODULE = None + _FLOAT_RELU_MODULE = nn.ReLU + + def __init__( + self, + # ConvTransposeBnNd args + in_channels, + out_channels, + kernel_size, + stride=1, + bias=None, + transposed=True, + padding=0, + output_padding=0, + groups=1, + dilation=1, + padding_mode='zeros', + qconfig=None): + + super(ConvTransposeReLU2d_sophgo, + self).__init__(in_channels, + out_channels, + kernel_size, + stride=stride, + bias=bias, + padding=padding, + output_padding=output_padding, + groups=groups, + dilation=dilation, + padding_mode=padding_mode, + qconfig=qconfig) + assert qconfig, 'qconfig must be provided for QAT module' + + def forward(self, input, output_size=None): + return F.relu(qnnqat.ConvTranspose2d_sophgo.forward(input, output_size)) \ No newline at end of file diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py index 94f1d0bb..452b14d9 100644 --- a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -79,8 +79,6 @@ def bias_fake_quant(self, bias, scale_w, in_scale): scale = scale_w*in_scale if torch.nonzero(scale).size()[0] != scale.numel(): print('Linear error! scale has 0, scale:', scale) - scale[torch.abs(scale) < 1e-10] = 1e-10 - print('new scale:', scale) bias_q = bias/scale bias = (bias_q.round()-bias_q).detach() + bias_q @@ -199,20 +197,21 @@ def __init__(self, in_features, out_features, bias=True, self.weight_fake_quant = qconfig.weight(factory_kwargs=factory_kwargs) def bias_fake_quant(self, bias, scale_w, in_scale): - scale = scale_w*in_scale - if torch.nonzero(scale).size()[0] != scale.numel(): - print('Linear error! scale has 0, scale:', scale) - scale[torch.abs(scale) < 1e-10] = 1e-10 - print('new scale:', scale) - - bias_q = bias/scale - bias = (bias_q.round()-bias_q).detach() + bias_q - bias = bias*scale + if bias is not None: + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('Linear error! scale has 0, scale:', scale) + scale[torch.abs(scale) < 1e-10] = 1e-10 + print('new scale:', scale) + + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale return bias def _forward(self, input): - # print('xxx1') - in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale + assert hasattr(self, 'input_fake_quantizer') + in_scale = self.input_fake_quantizer.scale conv = F.linear(input, self.weight_fake_quant(self.weight), self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) return conv diff --git a/mqbench/nn/qat/modules/__init__.py b/mqbench/nn/qat/modules/__init__.py index bb13511c..73cde0e8 100644 --- a/mqbench/nn/qat/modules/__init__.py +++ b/mqbench/nn/qat/modules/__init__.py @@ -1,4 +1,4 @@ from .linear import Linear -from .deconv import ConvTranspose2d +from .deconv import ConvTranspose2d, ConvTranspose2d_sophgo from .conv import Conv2d, Conv2d_sophgo from .embedding import Embedding \ No newline at end of file diff --git a/mqbench/nn/qat/modules/deconv.py b/mqbench/nn/qat/modules/deconv.py index ec7ef825..e5ae9193 100644 --- a/mqbench/nn/qat/modules/deconv.py +++ b/mqbench/nn/qat/modules/deconv.py @@ -30,6 +30,60 @@ def forward(self, x, output_size=None): x, self.weight_fake_quant(self.weight), self.bias, self.stride, self.padding, output_padding, self.groups, self.dilation) + @classmethod + def from_float(cls, mod): + assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + \ + cls._FLOAT_MODULE.__name__ + assert mod.qconfig, 'Input float module must have a valid qconfig' + if type(mod) == ConvTransposeReLU2d: + mod = mod[0] + qconfig = mod.qconfig + qat_deconv = cls(mod.in_channels, mod.out_channels, mod.kernel_size, + stride=mod.stride, padding=mod.padding, output_padding=mod.output_padding, + groups=mod.groups, bias=mod.bias is not None, dilation=mod.dilation, + padding_mode=mod.padding_mode, qconfig=qconfig) + qat_deconv.weight = mod.weight + qat_deconv.bias = mod.bias + return qat_deconv + +class ConvTranspose2d_sophgo(nn.ConvTranspose2d): + _FLOAT_MODULE = nn.ConvTranspose2d + + def __init__(self, in_channels, out_channels, kernel_size, + stride=1, padding=0, output_padding=0, + groups=1, bias=True, dilation=1, + padding_mode='zeros', qconfig=None): + super().__init__(in_channels, out_channels, kernel_size, + stride=stride, padding=padding, output_padding=output_padding, + groups=groups, bias=bias, dilation=dilation, padding_mode=padding_mode) + assert qconfig, 'qconfig must be provided for QAT module' + self.qconfig = qconfig + self.weight_fake_quant = qconfig.weight() + # ConvTranspose do per-channel quantize on output channel. + if self.weight_fake_quant.ch_axis != -1: + self.weight_fake_quant.ch_axis = 1 + self.weight_fake_quant.activation_post_process.ch_axis = 1 + + def bias_fake_quant_proc(self, bias, scale_w, in_scale): + scale = scale_w*in_scale + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale + return bias + + def forward(self, x, output_size=None): + output_padding = self._output_padding( + x, output_size, self.stride, self.padding, self.kernel_size, self.dilation + ) + + bias = self.bias + if self.bias is not None and self.weight_fake_quant.fake_quant_enabled[0] == 1: + in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + bias = self.bias_fake_quant_proc(self.bias, self.weight_fake_quant.scale, in_scale) + return F.conv_transpose2d( + x, self.weight_fake_quant(self.weight), bias, self.stride, self.padding, + output_padding, self.groups, self.dilation) + @classmethod def from_float(cls, mod): assert type(mod) == cls._FLOAT_MODULE, 'qat.' + cls.__name__ + '.from_float only works for ' + \ From e612cfdc2637c6dc23233d8dd5bcbf2f72fd73bf Mon Sep 17 00:00:00 2001 From: wangxc2006 <68397007@qq.com> Date: Tue, 15 Nov 2022 14:30:46 +0800 Subject: [PATCH 05/29] add some verfiy code --- application/imagenet_example/main.py | 127 ++++++++++++++++-- mqbench/convert_deploy.py | 13 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 3 +- mqbench/deploy/common.py | 2 +- mqbench/deploy/deploy_linear.py | 69 +++------- mqbench/fusion_method.py | 10 +- .../qat/modules/linear_fused_sophgo_tpu.py | 2 +- 7 files changed, 153 insertions(+), 73 deletions(-) diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index c5a60b53..3c712a95 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -26,8 +26,8 @@ model_names = sorted(name for name in models.__dict__ if name.islower() and not name.startswith("__") and callable(models.__dict__[name])) -cali_batch_size = 10 +cali_batch_size = 16 parser = argparse.ArgumentParser(description='PyTorch ImageNet Training') parser.add_argument('--train_data', metavar='DIR', help='path to dataset', required=True) @@ -181,7 +181,11 @@ def main_worker(gpu, ngpus_per_node, args): else: print("=> creating model '{}'".format(args.arch)) model = models.__dict__[args.arch]() - print('ori module:', model) + # print('ori module:', model) + + # model = models._utils.IntermediateLayerGetter(model,{'layer1': 'feat1', 'layer3': 'feat2'}) + # print(model(out_put)) + # for internal cluster if args.model_path: state_dict = torch.load(args.model_path) @@ -194,8 +198,9 @@ def main_worker(gpu, ngpus_per_node, args): else: model = model.cpu() if args.pre_eval_and_export: - print('原始onnx模型精度') - validate(val_loader, model.eval(), criterion, args) + print('??onnx????') + validate(val_loader, model.eval(), criterion, args) #?????model.cuda(),??? + kwargs = { 'input_shape_dict': {'data': [cali_batch_size, 3, 224, 224]}, 'output_path': args.output_path, @@ -207,13 +212,15 @@ def main_worker(gpu, ngpus_per_node, args): module_tmp = module_tmp.cpu() convert_onnx(module_tmp.eval(), **kwargs) del module_tmp - model = model.train() + model = model.train() #prepare?????train??!! + # quantize model if args.quant: prepare_custom_config_dict= { + # 'extra_qconfig_dict':{'w_fakequantize':'PACTFakeQuantize'} } model = prepare_by_platform(model, args.backend, prepare_custom_config_dict) - print('prepared module:', model) + # print('prepared module:', model) if not torch.cuda.is_available(): print('using CPU, this will be slow') elif args.distributed: @@ -246,6 +253,7 @@ def main_worker(gpu, ngpus_per_node, args): if args.cpu: model = model.cpu() else: + # model = torch.nn.DataParallel(model).cuda() #???gpu?????????resume??cpu?? model = model.cuda() # define loss function (criterion) and optimizer @@ -262,7 +270,10 @@ def main_worker(gpu, ngpus_per_node, args): if args.quant and not args.cpu: enable_calibration(model) calibrate(cali_loader, model, args) + cudnn.benchmark = True + # cudnn.deterministic = True #???????? + if args.quant: enable_quantization(model) @@ -285,6 +296,8 @@ def main_worker(gpu, ngpus_per_node, args): best_acc1 = best_acc1.to(args.gpu) state_dict = checkpoint['state_dict'] + # if args.cpu: + # model = torch.nn.DataParallel(model).cpu() model_dict = model.state_dict() if 'module.' in list(state_dict.keys())[0] and 'module.' not in list(model_dict.keys())[0]: for k in list(state_dict.keys()): @@ -302,7 +315,7 @@ def main_worker(gpu, ngpus_per_node, args): exit(1) if args.evaluate: - print('resume模型精度') + print('resume????') from mqbench.convert_deploy import convert_merge_bn module_tmp2 = copy.deepcopy(model) convert_merge_bn(module_tmp2.eval()) @@ -338,10 +351,27 @@ def main_worker(gpu, ngpus_per_node, args): 'best_acc1': best_acc1, 'optimizer' : optimizer.state_dict(), }, is_best, filename=filename) + + print('disable_all?????:') + disable_all(model) + validate(val_loader, model, criterion, args) + + enable_quantization(model) gen_test_ref_data(cali_loader, model, args) convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [cali_batch_size, 3, 224, 224]}, model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) + model_path = os.path.join(args.output_path, '{}.pt'.format('{}_mqmoble'.format(args.arch))) + model_pt = torch.load(model_path) + if args.gpu is not None: + model_pt = model_pt.cuda(args.gpu) + else: + model_pt = model_pt.cpu() + print('load fused bn pt?????:') + validate(val_loader, model_pt, criterion, args) + + validate_onnx(criterion, args) + def prepare_dataloader(args): traindir = os.path.join(args.train_data, 'train') valdir = os.path.join(args.val_data, 'val') @@ -383,6 +413,23 @@ def prepare_dataloader(args): return train_loader, train_sampler, val_loader, cali_loader +def prepare_dataloader_batch(args, batch_size): + valdir = os.path.join(args.val_data, 'val') + normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], + std=[0.229, 0.224, 0.225]) + + val_loader = torch.utils.data.DataLoader( + datasets.ImageFolder(valdir, transforms.Compose([ + transforms.Resize(256), + transforms.CenterCrop(224), + transforms.ToTensor(), + normalize, + ])), + batch_size=batch_size, shuffle=False, + num_workers=args.workers, pin_memory=True) + + return val_loader + def get_node_name_by_module_name(qname, model): nodes = list(model.graph.nodes) modules = dict(model.named_modules()) @@ -402,6 +449,7 @@ def calibrate(cali_loader, model, args): print("End calibration.") return def gen_test_ref_data(cali_loader, model, args): + return model.eval() global layer_names hook_handles = [] @@ -411,6 +459,9 @@ def gen_test_ref_data(cali_loader, model, args): for name, child in model.named_modules(): if not any([i in str(type(child)) for i in exclude_module]): print("add hook on", str(type(child)), name) + # if '_dup' in name: + # name = name[:-5] + # layer_names.append(name.replace('.','_')) node_name = get_node_name_by_module_name(name, model) layer_names.append(node_name) hd = child.register_forward_hook(hook=hook) @@ -483,13 +534,14 @@ def train(train_loader, model, criterion, optimizer, epoch, args): if i % args.print_freq == 0: progress.display(i) + # # ???????????? # for param in model.named_parameters(): # sum = torch.isnan(param[1]).sum() # if sum > 0: # print(param[0], 'has Nan', param[1].shape, 'sum:', sum) if args.fast_test: - if i % 100 == 0: + if i % 64 == 0: break def validate(val_loader, model, criterion, args): @@ -548,6 +600,65 @@ def validate(val_loader, model, criterion, args): return top1.avg +def validate_onnx(criterion, args): + import onnxruntime as rt + val_loader = prepare_dataloader_batch(args, cali_batch_size) + model_path = os.path.join(args.output_path, '{}_deploy_model.onnx'.format('{}_mqmoble'.format(args.arch))) + sess = rt.InferenceSession(model_path, providers = ['CUDAExecutionProvider', 'CPUExecutionProvider']) + input_name = sess.get_inputs()[0].name + + batch_time = AverageMeter('Time', ':6.3f') + losses = AverageMeter('Loss', ':.4e') + top1 = AverageMeter('Acc@1', ':6.2f') + top5 = AverageMeter('Acc@5', ':6.2f') + progress = ProgressMeter( + len(val_loader), + [batch_time, losses, top1, top5], + prefix='Test: ') + + + with torch.no_grad(): + end = time.time() + for i, (images, target) in enumerate(val_loader): + if not args.cpu: + if args.gpu is not None: + images = images.cuda(args.gpu, non_blocking=True) + if torch.cuda.is_available(): + target = target.cuda(args.gpu, non_blocking=True) + else: + images = images.cpu() + target = target.cpu() + + # compute output + # output = model(images) + output = sess.run(None, {input_name:images.clone().detach().cpu().numpy()}) + output = torch.from_numpy(output[0]).cuda(args.gpu, non_blocking=True) + loss = criterion(output, target) + + # measure accuracy and record loss + acc1, acc5 = accuracy(output, target, topk=(1, 5)) + losses.update(loss.item(), images.size(0)) + top1.update(acc1[0], images.size(0)) + top5.update(acc5[0], images.size(0)) + + # measure elapsed time + batch_time.update(time.time() - end) + end = time.time() + + if i % args.print_freq == 0: + progress.display(i) + + # if args.fast_test: + # if i % 100 == 0: + # break + + # TODO: this should also be done with the ProgressMeter + print(' * Acc@1 {top1.avg:.3f} Acc@5 {top5.avg:.3f}' + .format(top1=top1, top5=top5)) + + return top1.avg + + def save_checkpoint(state, is_best, filename='checkpoint.pth.tar'): torch.save(state, filename) if is_best: diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index 9fde92e4..936a7015 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -35,8 +35,8 @@ @register_deploy_function(BackendType.OPENVINO) @register_deploy_function(BackendType.Sophgo_TPU) def convert_merge_bn(model: GraphModule, **kwargs): - print('wlog before convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) - print('wlog before convert_merge_bn, model.graph:', model.graph) + # print('wlog before convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) + # print('wlog before convert_merge_bn, model.graph:', model.graph) logger.info("Merge BN for deploy.") nodes = list(model.graph.nodes) modules = dict(model.named_modules()) @@ -44,8 +44,8 @@ def convert_merge_bn(model: GraphModule, **kwargs): if node.op == 'call_module': if node.target in modules and type(modules[node.target]) in FUSED_MODULE_CONVERT_FUNCTION: FUSED_MODULE_CONVERT_FUNCTION[type(modules[node.target])](model, node) - print('wlog after convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) - print('wlog after convert_merge_bn, model.graph:', model.graph) + # print('wlog after convert_merge_bn, model.named_modules:', dict(model.named_modules())['']) + # print('wlog after convert_merge_bn, model.graph:', model.graph) @register_deploy_function(BackendType.Academic_NLP) @register_deploy_function(BackendType.Tensorrt_NLP) @@ -61,6 +61,9 @@ def convert_merge_bn(model: GraphModule, **kwargs): @register_deploy_function(BackendType.OPENVINO) @register_deploy_function(BackendType.Sophgo_TPU) def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_path, **kwargs): + pt_file_name = onnx_model_path.split('.') + pt_file_name[-1] = 'pt' + torch.save(model, '.'.join(pt_file_name)) logger.info("Export to onnx, onnx_model_path:{}".format(onnx_model_path)) model = model.cpu() output_names = kwargs.get('output_names', []) @@ -95,7 +98,6 @@ def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_p custom_opsets={'' : opset_version}, enable_onnx_checker=False) - @register_deploy_function(BackendType.Tensorrt) def convert_onnx_qlinear(model: GraphModule, onnx_model_path, model_name, **kwargs): if kwargs.get('deploy_to_qlinear', False): @@ -119,6 +121,7 @@ def deploy_qparams_openvino(model: GraphModule, onnx_model_path, model_name, **k def deploy_qparams_tensorrt(model: GraphModule, onnx_model_path, model_name, **kwargs): logger.info("Extract qparams for TensorRT.") remove_fakequantize_and_collect_params(onnx_model_path, model_name, backend='tensorrt') + @register_deploy_function(BackendType.Sophgo_TPU) def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, **kwargs): logger.info("Extract qparams for sophgo_tpu.") diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index eb4253ae..3315cd49 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -107,4 +107,5 @@ def _set_fake_quantizer_to_next_weight_layer(self, model: GraphModule): for user in node.users: if (user.op == "call_module" and isinstance(modules[user.target], self._layers_need_scale_form_input_fake_quantizer)): setattr(modules[user.target], "input_fake_quantizer", fake_quantizer) - print('wlog:', user.target,'\'type is:', type(modules[user.target]), "add input_fake_quantizer") \ No newline at end of file + print('wlog:', user.target,'\'type is:', type(modules[user.target]), "add input_fake_quantizer") + return model \ No newline at end of file diff --git a/mqbench/deploy/common.py b/mqbench/deploy/common.py index 54dab486..25af2af2 100644 --- a/mqbench/deploy/common.py +++ b/mqbench/deploy/common.py @@ -190,7 +190,7 @@ def remove_fake_pad_op(self, graph, name2data, inp2node, out2node): return - +#杈撳嚭tensor鍒拌妭鐐癸紝杈撳叆tensor鍒拌妭鐐瑰強璇ヨ緭鍏ュ簭鍙 def update_inp2node_out2node(graph): out2node = {} inp2node = {} diff --git a/mqbench/deploy/deploy_linear.py b/mqbench/deploy/deploy_linear.py index 8129414f..2795ca4c 100644 --- a/mqbench/deploy/deploy_linear.py +++ b/mqbench/deploy/deploy_linear.py @@ -117,33 +117,6 @@ def clip_weight(self, node, name2data, inp2node, named_initializer): named_initializer[tensor_name].raw_data = new_data.raw_data def post_process_clip_ranges(self, clip_ranges, graph, inp2node, out2node): - def find_the_closest_clip_range(node): - input_0 = node.input[0] - tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') - if tensor_name[-1] == '_': - tensor_name = tensor_name[:-1] - if tensor_name in clip_ranges: - return tensor_name - elif node.op_type in ['Flatten', 'Resize'] and node.output[0] in inp2node: - return find_the_closest_clip_range(inp2node[node.output[0]][0][0]) - else: - return None - - for node in graph.node: - if node.op_type in ['Flatten', 'Resize']: - tensor_name = find_the_closest_clip_range(node) - if tensor_name: - old = clip_ranges[tensor_name] - new_name = node.input[0] - new_name = '{}_{}'.format(new_name, out2node[new_name].op_type if new_name in out2node else '') - if new_name[-1] == '_': - new_name = tensor_name[:-1] - clip_ranges[new_name] = copy.deepcopy(old) - clip_ranges[new_name]['ori_name'] = new_name - logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') - return clip_ranges - - def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): def find_the_closest_clip_range(node): input_0 = node.input[0] tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') @@ -211,16 +184,14 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) logger.info("Vitis-DPU does not support per-channel quatization.") raise NotImplementedError("Vitis-DPU does not support per-channel quatization.") elif backend == 'sophgo_tpu': - - if len(next_nodes) == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']:#?????????1???,??1??????conv?? - next_node_output = next_nodes[0][0].output[0] #???????1???????1???tensor - if next_node_output in inp2node: - print(next_node_output, 'not in inp2node') - if len(inp2node[next_node_output]) == 0: - print(next_node_output, ',users not exsit') - if inp2node[next_node_output][0][0].op_type == 'Relu':##???????1???conv????1??????Relu(fake->conv->relu) + #鍗风Н鏉冮噸per-channel閲忓寲鍙傛暟锛宐ias鐨刾er-chan閲忓寲鍙傛暟娌℃湁鍘昏皟浼 + if len(next_nodes) == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']:#褰撳墠浼噺鍖栬妭鐐瑰彧鏈1涓悗缁э紝涓旂1涓悗缁ц妭鐐逛负conv绫诲瀷 + next_node_output = next_nodes[0][0].output[0] + if inp2node[next_node_output][0][0].op_type == 'Relu':##浼噺鍖栬妭鐐圭殑绗1涓悗缁onv鑺傜偣鐨勭1涓悗缁ц妭鐐逛负Relu(fake->conv->relu) + #鑻ユ槸fake->conv->relu,鍥犱负relu浼氳瀺鍚堝埌鍓嶉潰conv锛屾晠鐢╮elu鐨勮緭鍑簍ensor鍚+Relu浣滀负閲忓寲鍙傛暟淇濆瓨tensor鍚 tensor_name = '{}_{}'.format(inp2node[next_node_output][0][0].output[0], 'Relu') else: + #鑻ユ槸fake->conv->not_relu_type,鐩存帴鐢╟onv鐨勮緭鍑簍ensor鍚+conv浣滀负閲忓寲鍙傛暟淇濆瓨tensor鍚 tensor_name = '{}_{}'.format(next_node_output, next_nodes[0][0].op_type) tensor_name += '_{}'.format('weight' if next_nodes[0][1] == 1 else 'bias' ) clip_ranges[tensor_name] = {'step': [float(x) for x in scale], @@ -236,10 +207,10 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) if backend == 'sophgo_tpu': assert next_nodes[0][0].op_type == 'Gemm' tensor_name += '{}_{}_weight'.format(inp2node[node.output[0]][0][0].output[0], inp2node[node.output[0]][0][0].op_type) - clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #??????????? + clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #瀵圭О閲忓寲鏃惰繖涓弬鏁扮敓鏁 'min': float(scale * (qmin - zero_point)), 'max': float(scale * (qmax - zero_point)), - 'ori_name':pre_layer_name} + 'ori_name': 'none'} elif len(next_nodes) == 1 and next_nodes[0][1] == 2 and next_nodes[0][0].op_type in ['Gemm', 'Conv']: # fake quantize for bias assert backend == 'vitis' @@ -251,14 +222,6 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) # fake quantize for activations self.deal_with_activation_fakequant(node, inp2node) tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) - scale_name = node.input[1] - pre_layer_name = 'none' - post_str = '_post_act_fake_quantizer.scale' - if scale_name.endswith(post_str): - pre_layer_name = scale_name[:len(scale_name)-len(post_str)] - else: - print('not _post_act_fake_quantizer') - input_0 = node.input[0] for out in graph.output: if out.name == node.output[0]: out.name = tensor_name @@ -272,14 +235,14 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) 'max': float(scale * (qmax - zero_point))} ] elif backend == 'sophgo_tpu': - input_0 = node.input[0] - tensor_name += '_{}'.format(out2node[input_0].op_type if input_0 in out2node else '') - if tensor_name[-1] == '_': - tensor_name = tensor_name[:-1] - clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #对称量化时这个参数生效 + scale_name = node.input[1] + post_str = '_post_act_fake_quantizer.scale' + if tensor_name in out2node: + tensor_name += '_{}'.format(out2node[tensor_name].op_type) + clip_ranges[tensor_name] = {'threshold':float(scale * max(-qmin, qmax)), #瀵圭О閲忓寲鏃惰繖涓弬鏁扮敓鏁 'min': float(scale * (qmin - zero_point)), 'max': float(scale * (qmax - zero_point)), - 'ori_name':pre_layer_name} + 'ori_name': scale_name[:len(scale_name)-len(post_str)] if scale_name.endswith(post_str) else 'none'} if backend == 'ppl': clip_ranges[tensor_name] = {'step': float(scale), 'zero_point': int(zero_point), @@ -302,7 +265,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) if name in (out2node.keys() | inp2node.keys()): continue graph.initializer.remove(initial_data) - + clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) if backend == 'tensorrt': context = {"tensorrt": {"blob_range": clip_ranges}} @@ -315,7 +278,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) elif backend == 'ppl-cuda': context = {'ppl-cuda': clip_ranges} elif backend == 'sophgo_tpu': - clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node) + #clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) context = {'sophgo_tpu': clip_ranges} context['w_qscheme'] = '' context['a_qscheme'] = '' diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index 3c0ba997..a06216b7 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -15,12 +15,15 @@ def convert_qnniqat_linear(model, fused_node): modules = dict(model.named_modules()) fused_module = modules[fused_node.target] + # Create a Linear from FusedModule. linear = torch.nn.Linear(fused_module.in_features, fused_module.out_features, fused_module.bias is not None) linear.weight = fused_module.weight if fused_module.bias is not None: linear.bias = fused_module.bias + # We need nn.qat.linear here to export weight quantize node. linear.qconfig = fused_module.qconfig linear = torch.nn.qat.Linear.from_float(linear) + # Attach weight fake quantize params. linear.weight_fake_quant = fused_module.weight_fake_quant linear_parent_name, linear_name = _parent_name(fused_node.target) setattr(modules[linear_parent_name], linear_name, linear) @@ -48,7 +51,6 @@ def convert_qnnqat_conv2d(model, fused_node): def convert_qnnqat_deconv2d(model, fused_node): modules = dict(model.named_modules()) fused_module = modules[fused_node.target] - # Create a ConvTranspose2d from FusedModule. deconv = torch.nn.ConvTranspose2d(fused_module.in_channels, fused_module.out_channels, fused_module.kernel_size, stride=fused_module.stride, padding=fused_module.padding, output_padding=fused_module.output_padding, groups=fused_module.groups, bias=fused_module.bias is not None, @@ -58,24 +60,24 @@ def convert_qnnqat_deconv2d(model, fused_node): if fused_module.bias is not None: deconv.bias = fused_module.bias fused_deconv = fuse_deconv_bn_eval(deconv.eval(), fused_module.bn) - # We need nn.qat.conv here to export weight quantize node. fused_deconv.qconfig = fused_module.qconfig fused_deconv = qnnqat.ConvTranspose2d.from_float(fused_deconv) - # Attach weight fake quantize params. fused_deconv.weight_fake_quant = fused_module.weight_fake_quant deconv_parent_name, deconv_name = _parent_name(fused_node.target) setattr(modules[deconv_parent_name], deconv_name, fused_deconv) - @register_convert_function(qnniqat.LinearReLU_sophgo) def linearert_qnniqat_linearrelu(model, fused_node): convert_qnniqat_linear(model, fused_node) modules = dict(model.named_modules()) fused_module = modules[fused_node.target] + # We need to Insert Relu after Merged linear. linear_parent_name, linear_name = _parent_name(fused_node.target) relu_name = 'relu' + # Maybe has another name, but we cannot know for now. if not hasattr(modules[linear_parent_name], relu_name): setattr(modules[linear_parent_name], relu_name, torch.nn.ReLU(inplace=True).train(fused_module.training)) + # Update modules. modules = dict(model.named_modules()) graph = model.graph nodes = list(model.graph.nodes) diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py index 452b14d9..84550d0b 100644 --- a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -211,7 +211,7 @@ def bias_fake_quant(self, bias, scale_w, in_scale): def _forward(self, input): assert hasattr(self, 'input_fake_quantizer') - in_scale = self.input_fake_quantizer.scale + in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale conv = F.linear(input, self.weight_fake_quant(self.weight), self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) return conv From 99775c45426d8cb7067bd00f9f3165b3952a54ea Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Thu, 1 Dec 2022 15:23:03 +0800 Subject: [PATCH 06/29] for qat int8 release --- application/imagenet_example/main.py | 38 +- application/yolov5_example/.dockerignore | 222 ++++ application/yolov5_example/.gitattributes | 2 + application/yolov5_example/.gitignore | 256 ++++ .../yolov5_example/.pre-commit-config.yaml | 64 + application/yolov5_example/CONTRIBUTING.md | 98 ++ application/yolov5_example/LICENSE | 674 ++++++++++ application/yolov5_example/README.md | 363 ++++++ .../yolov5_example/classify/predict.py | 109 ++ application/yolov5_example/classify/train.py | 325 +++++ application/yolov5_example/classify/val.py | 158 +++ .../yolov5_example/data/Argoverse.yaml | 67 + .../yolov5_example/data/GlobalWheat2020.yaml | 54 + application/yolov5_example/data/ImageNet.yaml | 156 +++ .../yolov5_example/data/Objects365.yaml | 114 ++ application/yolov5_example/data/SKU-110K.yaml | 53 + application/yolov5_example/data/VOC.yaml | 81 ++ application/yolov5_example/data/VisDrone.yaml | 61 + application/yolov5_example/data/coco.yaml | 45 + application/yolov5_example/data/coco128.yaml | 30 + .../data/hyps/hyp.Objects365.yaml | 34 + .../yolov5_example/data/hyps/hyp.VOC.yaml | 40 + .../data/hyps/hyp.scratch-high.yaml | 34 + .../data/hyps/hyp.scratch-low.yaml | 34 + .../data/hyps/hyp.scratch-med.yaml | 34 + .../yolov5_example/data/images/bus.jpg | Bin 0 -> 487438 bytes .../yolov5_example/data/images/zidane.jpg | Bin 0 -> 168949 bytes .../data/scripts/download_weights.sh | 21 + .../yolov5_example/data/scripts/get_coco.sh | 56 + .../data/scripts/get_coco128.sh | 17 + .../data/scripts/get_imagenet.sh | 51 + application/yolov5_example/data/xView.yaml | 102 ++ application/yolov5_example/detect.py | 257 ++++ application/yolov5_example/export.py | 616 +++++++++ application/yolov5_example/hubconf.py | 160 +++ application/yolov5_example/models/__init__.py | 0 application/yolov5_example/models/common.py | 771 +++++++++++ .../yolov5_example/models/experimental.py | 107 ++ .../yolov5_example/models/hub/anchors.yaml | 59 + .../yolov5_example/models/hub/yolov3-spp.yaml | 51 + .../models/hub/yolov3-tiny.yaml | 41 + .../yolov5_example/models/hub/yolov3.yaml | 51 + .../models/hub/yolov5-bifpn.yaml | 48 + .../yolov5_example/models/hub/yolov5-fpn.yaml | 42 + .../yolov5_example/models/hub/yolov5-p2.yaml | 54 + .../yolov5_example/models/hub/yolov5-p34.yaml | 41 + .../yolov5_example/models/hub/yolov5-p6.yaml | 56 + .../yolov5_example/models/hub/yolov5-p7.yaml | 67 + .../models/hub/yolov5-panet.yaml | 48 + .../yolov5_example/models/hub/yolov5l6.yaml | 60 + .../yolov5_example/models/hub/yolov5m6.yaml | 60 + .../yolov5_example/models/hub/yolov5n6.yaml | 60 + .../models/hub/yolov5s-ghost.yaml | 48 + .../models/hub/yolov5s-transformer.yaml | 48 + .../yolov5_example/models/hub/yolov5s6.yaml | 60 + .../yolov5_example/models/hub/yolov5x6.yaml | 60 + application/yolov5_example/models/tf.py | 574 ++++++++ application/yolov5_example/models/yolo.py | 360 +++++ .../yolov5_example/models/yolov5l.yaml | 48 + .../yolov5_example/models/yolov5m.yaml | 48 + .../yolov5_example/models/yolov5n.yaml | 48 + .../yolov5_example/models/yolov5s.yaml | 48 + .../yolov5_example/models/yolov5x.yaml | 48 + .../yolov5_example/mqbench.code-workspace | 11 + application/yolov5_example/requirements.txt | 43 + application/yolov5_example/setup.cfg | 59 + application/yolov5_example/train.py | 727 +++++++++++ application/yolov5_example/tutorial.ipynb | 1151 ++++++++++++++++ application/yolov5_example/utils/__init__.py | 36 + .../yolov5_example/utils/activations.py | 103 ++ .../yolov5_example/utils/augmentations.py | 350 +++++ .../yolov5_example/utils/autoanchor.py | 170 +++ application/yolov5_example/utils/autobatch.py | 66 + .../yolov5_example/utils/aws/__init__.py | 0 application/yolov5_example/utils/aws/mime.sh | 26 + .../yolov5_example/utils/aws/resume.py | 40 + .../yolov5_example/utils/aws/userdata.sh | 27 + .../yolov5_example/utils/benchmarks.py | 157 +++ application/yolov5_example/utils/callbacks.py | 71 + .../yolov5_example/utils/dataloaders.py | 1158 +++++++++++++++++ .../yolov5_example/utils/docker/Dockerfile | 68 + .../utils/docker/Dockerfile-arm64 | 42 + .../utils/docker/Dockerfile-cpu | 39 + application/yolov5_example/utils/downloads.py | 180 +++ .../utils/flask_rest_api/README.md | 73 ++ .../utils/flask_rest_api/example_request.py | 19 + .../utils/flask_rest_api/restapi.py | 48 + application/yolov5_example/utils/general.py | 1051 +++++++++++++++ .../utils/google_app_engine/Dockerfile | 25 + .../additional_requirements.txt | 4 + .../utils/google_app_engine/app.yaml | 14 + .../yolov5_example/utils/loggers/__init__.py | 308 +++++ .../utils/loggers/clearml/README.md | 222 ++++ .../utils/loggers/clearml/__init__.py | 0 .../utils/loggers/clearml/clearml_utils.py | 156 +++ .../utils/loggers/clearml/hpo.py | 84 ++ .../utils/loggers/wandb/README.md | 162 +++ .../utils/loggers/wandb/__init__.py | 0 .../utils/loggers/wandb/log_dataset.py | 27 + .../utils/loggers/wandb/sweep.py | 41 + .../utils/loggers/wandb/sweep.yaml | 143 ++ .../utils/loggers/wandb/wandb_utils.py | 584 +++++++++ application/yolov5_example/utils/loss.py | 234 ++++ application/yolov5_example/utils/metrics.py | 364 ++++++ application/yolov5_example/utils/plots.py | 519 ++++++++ .../yolov5_example/utils/torch_utils.py | 433 ++++++ application/yolov5_example/val.py | 396 ++++++ .../custom_quantizer/sophgo_tpu_quantizer.py | 4 + mqbench/deploy/deploy_linear.py | 60 +- .../qat/modules/conv_fused_sophgo_tpu.py | 6 - .../qat/modules/linear_fused_sophgo_tpu.py | 13 +- mqbench/train_all_model.py | 4 +- 112 files changed, 16909 insertions(+), 41 deletions(-) create mode 100644 application/yolov5_example/.dockerignore create mode 100644 application/yolov5_example/.gitattributes create mode 100644 application/yolov5_example/.gitignore create mode 100644 application/yolov5_example/.pre-commit-config.yaml create mode 100644 application/yolov5_example/CONTRIBUTING.md create mode 100644 application/yolov5_example/LICENSE create mode 100644 application/yolov5_example/README.md create mode 100644 application/yolov5_example/classify/predict.py create mode 100644 application/yolov5_example/classify/train.py create mode 100644 application/yolov5_example/classify/val.py create mode 100644 application/yolov5_example/data/Argoverse.yaml create mode 100644 application/yolov5_example/data/GlobalWheat2020.yaml create mode 100644 application/yolov5_example/data/ImageNet.yaml create mode 100644 application/yolov5_example/data/Objects365.yaml create mode 100644 application/yolov5_example/data/SKU-110K.yaml create mode 100644 application/yolov5_example/data/VOC.yaml create mode 100644 application/yolov5_example/data/VisDrone.yaml create mode 100644 application/yolov5_example/data/coco.yaml create mode 100644 application/yolov5_example/data/coco128.yaml create mode 100644 application/yolov5_example/data/hyps/hyp.Objects365.yaml create mode 100644 application/yolov5_example/data/hyps/hyp.VOC.yaml create mode 100644 application/yolov5_example/data/hyps/hyp.scratch-high.yaml create mode 100644 application/yolov5_example/data/hyps/hyp.scratch-low.yaml create mode 100644 application/yolov5_example/data/hyps/hyp.scratch-med.yaml create mode 100644 application/yolov5_example/data/images/bus.jpg create mode 100644 application/yolov5_example/data/images/zidane.jpg create mode 100644 application/yolov5_example/data/scripts/download_weights.sh create mode 100644 application/yolov5_example/data/scripts/get_coco.sh create mode 100644 application/yolov5_example/data/scripts/get_coco128.sh create mode 100644 application/yolov5_example/data/scripts/get_imagenet.sh create mode 100644 application/yolov5_example/data/xView.yaml create mode 100644 application/yolov5_example/detect.py create mode 100644 application/yolov5_example/export.py create mode 100644 application/yolov5_example/hubconf.py create mode 100644 application/yolov5_example/models/__init__.py create mode 100644 application/yolov5_example/models/common.py create mode 100644 application/yolov5_example/models/experimental.py create mode 100644 application/yolov5_example/models/hub/anchors.yaml create mode 100644 application/yolov5_example/models/hub/yolov3-spp.yaml create mode 100644 application/yolov5_example/models/hub/yolov3-tiny.yaml create mode 100644 application/yolov5_example/models/hub/yolov3.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-bifpn.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-fpn.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-p2.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-p34.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-p6.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-p7.yaml create mode 100644 application/yolov5_example/models/hub/yolov5-panet.yaml create mode 100644 application/yolov5_example/models/hub/yolov5l6.yaml create mode 100644 application/yolov5_example/models/hub/yolov5m6.yaml create mode 100644 application/yolov5_example/models/hub/yolov5n6.yaml create mode 100644 application/yolov5_example/models/hub/yolov5s-ghost.yaml create mode 100644 application/yolov5_example/models/hub/yolov5s-transformer.yaml create mode 100644 application/yolov5_example/models/hub/yolov5s6.yaml create mode 100644 application/yolov5_example/models/hub/yolov5x6.yaml create mode 100644 application/yolov5_example/models/tf.py create mode 100644 application/yolov5_example/models/yolo.py create mode 100644 application/yolov5_example/models/yolov5l.yaml create mode 100644 application/yolov5_example/models/yolov5m.yaml create mode 100644 application/yolov5_example/models/yolov5n.yaml create mode 100644 application/yolov5_example/models/yolov5s.yaml create mode 100644 application/yolov5_example/models/yolov5x.yaml create mode 100644 application/yolov5_example/mqbench.code-workspace create mode 100644 application/yolov5_example/requirements.txt create mode 100644 application/yolov5_example/setup.cfg create mode 100644 application/yolov5_example/train.py create mode 100644 application/yolov5_example/tutorial.ipynb create mode 100644 application/yolov5_example/utils/__init__.py create mode 100644 application/yolov5_example/utils/activations.py create mode 100644 application/yolov5_example/utils/augmentations.py create mode 100644 application/yolov5_example/utils/autoanchor.py create mode 100644 application/yolov5_example/utils/autobatch.py create mode 100644 application/yolov5_example/utils/aws/__init__.py create mode 100644 application/yolov5_example/utils/aws/mime.sh create mode 100644 application/yolov5_example/utils/aws/resume.py create mode 100644 application/yolov5_example/utils/aws/userdata.sh create mode 100644 application/yolov5_example/utils/benchmarks.py create mode 100644 application/yolov5_example/utils/callbacks.py create mode 100644 application/yolov5_example/utils/dataloaders.py create mode 100644 application/yolov5_example/utils/docker/Dockerfile create mode 100644 application/yolov5_example/utils/docker/Dockerfile-arm64 create mode 100644 application/yolov5_example/utils/docker/Dockerfile-cpu create mode 100644 application/yolov5_example/utils/downloads.py create mode 100644 application/yolov5_example/utils/flask_rest_api/README.md create mode 100644 application/yolov5_example/utils/flask_rest_api/example_request.py create mode 100644 application/yolov5_example/utils/flask_rest_api/restapi.py create mode 100644 application/yolov5_example/utils/general.py create mode 100644 application/yolov5_example/utils/google_app_engine/Dockerfile create mode 100644 application/yolov5_example/utils/google_app_engine/additional_requirements.txt create mode 100644 application/yolov5_example/utils/google_app_engine/app.yaml create mode 100644 application/yolov5_example/utils/loggers/__init__.py create mode 100644 application/yolov5_example/utils/loggers/clearml/README.md create mode 100644 application/yolov5_example/utils/loggers/clearml/__init__.py create mode 100644 application/yolov5_example/utils/loggers/clearml/clearml_utils.py create mode 100644 application/yolov5_example/utils/loggers/clearml/hpo.py create mode 100644 application/yolov5_example/utils/loggers/wandb/README.md create mode 100644 application/yolov5_example/utils/loggers/wandb/__init__.py create mode 100644 application/yolov5_example/utils/loggers/wandb/log_dataset.py create mode 100644 application/yolov5_example/utils/loggers/wandb/sweep.py create mode 100644 application/yolov5_example/utils/loggers/wandb/sweep.yaml create mode 100644 application/yolov5_example/utils/loggers/wandb/wandb_utils.py create mode 100644 application/yolov5_example/utils/loss.py create mode 100644 application/yolov5_example/utils/metrics.py create mode 100644 application/yolov5_example/utils/plots.py create mode 100644 application/yolov5_example/utils/torch_utils.py create mode 100644 application/yolov5_example/val.py diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index 3c712a95..5bfde3ad 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -91,15 +91,16 @@ parser.add_argument('--fast_test', action='store_true') parser.add_argument('--cpu', action='store_true') parser.add_argument('--pre_eval_and_export', action='store_true') +parser.add_argument('--deploy_batch_size', default=1, type=int, help='deploy_batch_size.') BackendMap = {'tensorrt': BackendType.Tensorrt, + 'sophgo_tpu': BackendType.Sophgo_TPU, 'nnie': BackendType.NNIE, 'tensorrt_nlp': BackendType.Tensorrt_NLP, 'ppl': BackendType.PPLW8A16, 'openvino': BackendType.OPENVINO, 'snpe': BackendType.SNPE, 'vitis': BackendType.Vitis, - 'sophgo_tpu': BackendType.Sophgo_TPU, 'tengine_u8': BackendType.Tengine_u8} best_acc1 = 0 @@ -198,11 +199,11 @@ def main_worker(gpu, ngpus_per_node, args): else: model = model.cpu() if args.pre_eval_and_export: - print('??onnx????') + print('原始onnx模型精度') validate(val_loader, model.eval(), criterion, args) #?????model.cuda(),??? kwargs = { - 'input_shape_dict': {'data': [cali_batch_size, 3, 224, 224]}, + 'input_shape_dict': {'data': [args.deploy_batch_size, 3, 224, 224]}, 'output_path': args.output_path, 'model_name': args.arch, 'dummy_input': None, @@ -271,12 +272,12 @@ def main_worker(gpu, ngpus_per_node, args): enable_calibration(model) calibrate(cali_loader, model, args) - cudnn.benchmark = True - # cudnn.deterministic = True #???????? - if args.quant: enable_quantization(model) + cudnn.benchmark = True + # cudnn.deterministic = True #避免计算结果波动 + # optionally resume from a checkpoint if args.resume: if os.path.isfile(args.resume): @@ -315,18 +316,17 @@ def main_worker(gpu, ngpus_per_node, args): exit(1) if args.evaluate: - print('resume????') + print('resume模型精度') from mqbench.convert_deploy import convert_merge_bn module_tmp2 = copy.deepcopy(model) convert_merge_bn(module_tmp2.eval()) validate(val_loader, module_tmp2, criterion, args) del module_tmp2 gen_test_ref_data(cali_loader, model, args) - convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [cali_batch_size, 3, 224, 224]}, + convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [args.deploy_batch_size, 3, 224, 224]}, model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) exit(0) - - filename= os.path.join(args.output_path, 'checkpoint.pth.tar') + for epoch in range(args.start_epoch, args.epochs): if args.distributed: train_sampler.set_epoch(epoch) @@ -336,6 +336,10 @@ def main_worker(gpu, ngpus_per_node, args): train(train_loader, model, criterion, optimizer, epoch, args) # evaluate on validation set + if epoch == args.epochs - 1: + print('qat训练后的带量化节点的eval精度:') + else: + print(f'epoch{epoch}训练后eval精度:') acc1 = validate(val_loader, model, criterion, args) # remember best acc@1 and save checkpoint @@ -350,15 +354,16 @@ def main_worker(gpu, ngpus_per_node, args): 'state_dict': model.state_dict(), 'best_acc1': best_acc1, 'optimizer' : optimizer.state_dict(), - }, is_best, filename=filename) + }, is_best, filename=os.path.join(args.output_path, 'checkpoint.pth.tar')) - print('disable_all?????:') + print('disable_all后测试精度:') disable_all(model) validate(val_loader, model, criterion, args) enable_quantization(model) gen_test_ref_data(cali_loader, model, args) - convert_deploy(model.eval(), args.backend, input_shape_dict={'data': [cali_batch_size, 3, 224, 224]}, + convert_deploy(model.eval(), args.backend, input_shape_dict= + {'data': [args.deploy_batch_size, 3, 224, 224]}, model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) model_path = os.path.join(args.output_path, '{}.pt'.format('{}_mqmoble'.format(args.arch))) @@ -367,7 +372,7 @@ def main_worker(gpu, ngpus_per_node, args): model_pt = model_pt.cuda(args.gpu) else: model_pt = model_pt.cpu() - print('load fused bn pt?????:') + print('load fused bn pt后测试精度:') validate(val_loader, model_pt, criterion, args) validate_onnx(criterion, args) @@ -534,7 +539,7 @@ def train(train_loader, model, criterion, optimizer, epoch, args): if i % args.print_freq == 0: progress.display(i) - # # ???????????? + # # 检查训练过程参数是否异常 # for param in model.named_parameters(): # sum = torch.isnan(param[1]).sum() # if sum > 0: @@ -602,7 +607,7 @@ def validate(val_loader, model, criterion, args): def validate_onnx(criterion, args): import onnxruntime as rt - val_loader = prepare_dataloader_batch(args, cali_batch_size) + val_loader = prepare_dataloader_batch(args, args.deploy_batch_size) model_path = os.path.join(args.output_path, '{}_deploy_model.onnx'.format('{}_mqmoble'.format(args.arch))) sess = rt.InferenceSession(model_path, providers = ['CUDAExecutionProvider', 'CPUExecutionProvider']) input_name = sess.get_inputs()[0].name @@ -653,6 +658,7 @@ def validate_onnx(criterion, args): # break # TODO: this should also be done with the ProgressMeter + print('deploy_model.onnx完成所有处理后的onnxruntime测试精度:') print(' * Acc@1 {top1.avg:.3f} Acc@5 {top5.avg:.3f}' .format(top1=top1, top5=top5)) diff --git a/application/yolov5_example/.dockerignore b/application/yolov5_example/.dockerignore new file mode 100644 index 00000000..3b669254 --- /dev/null +++ b/application/yolov5_example/.dockerignore @@ -0,0 +1,222 @@ +# Repo-specific DockerIgnore ------------------------------------------------------------------------------------------- +.git +.cache +.idea +runs +output +coco +storage.googleapis.com + +data/samples/* +**/results*.csv +*.jpg + +# Neural Network weights ----------------------------------------------------------------------------------------------- +**/*.pt +**/*.pth +**/*.onnx +**/*.engine +**/*.mlmodel +**/*.torchscript +**/*.torchscript.pt +**/*.tflite +**/*.h5 +**/*.pb +*_saved_model/ +*_web_model/ +*_openvino_model/ + +# Below Copied From .gitignore ----------------------------------------------------------------------------------------- +# Below Copied From .gitignore ----------------------------------------------------------------------------------------- + + +# GitHub Python GitIgnore ---------------------------------------------------------------------------------------------- +# Byte-compiled / optimized / DLL files +__pycache__/ +*.py[cod] +*$py.class + +# C extensions +*.so + +# Distribution / packaging +.Python +env/ +build/ +develop-eggs/ +dist/ +downloads/ +eggs/ +.eggs/ +lib/ +lib64/ +parts/ +sdist/ +var/ +wheels/ +*.egg-info/ +wandb/ +.installed.cfg +*.egg + +# PyInstaller +# Usually these files are written by a python script from a template +# before PyInstaller builds the exe, so as to inject date/other infos into it. +*.manifest +*.spec + +# Installer logs +pip-log.txt +pip-delete-this-directory.txt + +# Unit test / coverage reports +htmlcov/ +.tox/ +.coverage +.coverage.* +.cache +nosetests.xml +coverage.xml +*.cover +.hypothesis/ + +# Translations +*.mo +*.pot + +# Django stuff: +*.log +local_settings.py + +# Flask stuff: +instance/ +.webassets-cache + +# Scrapy stuff: +.scrapy + +# Sphinx documentation +docs/_build/ + +# PyBuilder +target/ + +# Jupyter Notebook +.ipynb_checkpoints + +# pyenv +.python-version + +# celery beat schedule file +celerybeat-schedule + +# SageMath parsed files +*.sage.py + +# dotenv +.env + +# virtualenv +.venv* +venv*/ +ENV*/ + +# Spyder project settings +.spyderproject +.spyproject + +# Rope project settings +.ropeproject + +# mkdocs documentation +/site + +# mypy +.mypy_cache/ + + +# https://github.com/github/gitignore/blob/master/Global/macOS.gitignore ----------------------------------------------- + +# General +.DS_Store +.AppleDouble +.LSOverride + +# Icon must end with two \r +Icon +Icon? + +# Thumbnails +._* + +# Files that might appear in the root of a volume +.DocumentRevisions-V100 +.fseventsd +.Spotlight-V100 +.TemporaryItems +.Trashes +.VolumeIcon.icns +.com.apple.timemachine.donotpresent + +# Directories potentially created on remote AFP share +.AppleDB +.AppleDesktop +Network Trash Folder +Temporary Items +.apdisk + + +# https://github.com/github/gitignore/blob/master/Global/JetBrains.gitignore +# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio and WebStorm +# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839 + +# User-specific stuff: +.idea/* +.idea/**/workspace.xml +.idea/**/tasks.xml +.idea/dictionaries +.html # Bokeh Plots +.pg # TensorFlow Frozen Graphs +.avi # videos + +# Sensitive or high-churn files: +.idea/**/dataSources/ +.idea/**/dataSources.ids +.idea/**/dataSources.local.xml +.idea/**/sqlDataSources.xml +.idea/**/dynamic.xml +.idea/**/uiDesigner.xml + +# Gradle: +.idea/**/gradle.xml +.idea/**/libraries + +# CMake +cmake-build-debug/ +cmake-build-release/ + +# Mongo Explorer plugin: +.idea/**/mongoSettings.xml + +## File-based project format: +*.iws + +## Plugin-specific files: + +# IntelliJ +out/ + +# mpeltonen/sbt-idea plugin +.idea_modules/ + +# JIRA plugin +atlassian-ide-plugin.xml + +# Cursive Clojure plugin +.idea/replstate.xml + +# Crashlytics plugin (for Android Studio and IntelliJ) +com_crashlytics_export_strings.xml +crashlytics.properties +crashlytics-build.properties +fabric.properties diff --git a/application/yolov5_example/.gitattributes b/application/yolov5_example/.gitattributes new file mode 100644 index 00000000..dad4239e --- /dev/null +++ b/application/yolov5_example/.gitattributes @@ -0,0 +1,2 @@ +# this drop notebooks from GitHub language stats +*.ipynb linguist-vendored diff --git a/application/yolov5_example/.gitignore b/application/yolov5_example/.gitignore new file mode 100644 index 00000000..69a00843 --- /dev/null +++ b/application/yolov5_example/.gitignore @@ -0,0 +1,256 @@ +# Repo-specific GitIgnore ---------------------------------------------------------------------------------------------- +*.jpg +*.jpeg +*.png +*.bmp +*.tif +*.tiff +*.heic +*.JPG +*.JPEG +*.PNG +*.BMP +*.TIF +*.TIFF +*.HEIC +*.mp4 +*.mov +*.MOV +*.avi +*.data +*.json +*.cfg +!setup.cfg +!cfg/yolov3*.cfg + +storage.googleapis.com +runs/* +data/* +data/images/* +!data/*.yaml +!data/hyps +!data/scripts +!data/images +!data/images/zidane.jpg +!data/images/bus.jpg +!data/*.sh + +results*.csv + +# Datasets ------------------------------------------------------------------------------------------------------------- +coco/ +coco128/ +VOC/ + +# MATLAB GitIgnore ----------------------------------------------------------------------------------------------------- +*.m~ +*.mat +!targets*.mat + +# Neural Network weights ----------------------------------------------------------------------------------------------- +*.weights +*.pt +*.pb +*.onnx +*.engine +*.mlmodel +*.torchscript +*.tflite +*.h5 +*_saved_model/ +*_web_model/ +*_openvino_model/ +darknet53.conv.74 +yolov3-tiny.conv.15 + +# GitHub Python GitIgnore ---------------------------------------------------------------------------------------------- +# Byte-compiled / optimized / DLL files +__pycache__/ +*.py[cod] +*$py.class + +# C extensions +*.so + +# Distribution / packaging +.Python +env/ +build/ +develop-eggs/ +dist/ +downloads/ +eggs/ +.eggs/ +lib/ +lib64/ +parts/ +sdist/ +var/ +wheels/ +*.egg-info/ +/wandb/ +.installed.cfg +*.egg + + +# PyInstaller +# Usually these files are written by a python script from a template +# before PyInstaller builds the exe, so as to inject date/other infos into it. +*.manifest +*.spec + +# Installer logs +pip-log.txt +pip-delete-this-directory.txt + +# Unit test / coverage reports +htmlcov/ +.tox/ +.coverage +.coverage.* +.cache +nosetests.xml +coverage.xml +*.cover +.hypothesis/ + +# Translations +*.mo +*.pot + +# Django stuff: +*.log +local_settings.py + +# Flask stuff: +instance/ +.webassets-cache + +# Scrapy stuff: +.scrapy + +# Sphinx documentation +docs/_build/ + +# PyBuilder +target/ + +# Jupyter Notebook +.ipynb_checkpoints + +# pyenv +.python-version + +# celery beat schedule file +celerybeat-schedule + +# SageMath parsed files +*.sage.py + +# dotenv +.env + +# virtualenv +.venv* +venv*/ +ENV*/ + +# Spyder project settings +.spyderproject +.spyproject + +# Rope project settings +.ropeproject + +# mkdocs documentation +/site + +# mypy +.mypy_cache/ + + +# https://github.com/github/gitignore/blob/master/Global/macOS.gitignore ----------------------------------------------- + +# General +.DS_Store +.AppleDouble +.LSOverride + +# Icon must end with two \r +Icon +Icon? + +# Thumbnails +._* + +# Files that might appear in the root of a volume +.DocumentRevisions-V100 +.fseventsd +.Spotlight-V100 +.TemporaryItems +.Trashes +.VolumeIcon.icns +.com.apple.timemachine.donotpresent + +# Directories potentially created on remote AFP share +.AppleDB +.AppleDesktop +Network Trash Folder +Temporary Items +.apdisk + + +# https://github.com/github/gitignore/blob/master/Global/JetBrains.gitignore +# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio and WebStorm +# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839 + +# User-specific stuff: +.idea/* +.idea/**/workspace.xml +.idea/**/tasks.xml +.idea/dictionaries +.html # Bokeh Plots +.pg # TensorFlow Frozen Graphs +.avi # videos + +# Sensitive or high-churn files: +.idea/**/dataSources/ +.idea/**/dataSources.ids +.idea/**/dataSources.local.xml +.idea/**/sqlDataSources.xml +.idea/**/dynamic.xml +.idea/**/uiDesigner.xml + +# Gradle: +.idea/**/gradle.xml +.idea/**/libraries + +# CMake +cmake-build-debug/ +cmake-build-release/ + +# Mongo Explorer plugin: +.idea/**/mongoSettings.xml + +## File-based project format: +*.iws + +## Plugin-specific files: + +# IntelliJ +out/ + +# mpeltonen/sbt-idea plugin +.idea_modules/ + +# JIRA plugin +atlassian-ide-plugin.xml + +# Cursive Clojure plugin +.idea/replstate.xml + +# Crashlytics plugin (for Android Studio and IntelliJ) +com_crashlytics_export_strings.xml +crashlytics.properties +crashlytics-build.properties +fabric.properties diff --git a/application/yolov5_example/.pre-commit-config.yaml b/application/yolov5_example/.pre-commit-config.yaml new file mode 100644 index 00000000..43aca019 --- /dev/null +++ b/application/yolov5_example/.pre-commit-config.yaml @@ -0,0 +1,64 @@ +# Define hooks for code formations +# Will be applied on any updated commit files if a user has installed and linked commit hook + +default_language_version: + python: python3.8 + +# Define bot property if installed via https://github.com/marketplace/pre-commit-ci +ci: + autofix_prs: true + autoupdate_commit_msg: '[pre-commit.ci] pre-commit suggestions' + autoupdate_schedule: monthly + # submodules: true + +repos: + - repo: https://github.com/pre-commit/pre-commit-hooks + rev: v4.3.0 + hooks: + # - id: end-of-file-fixer + - id: trailing-whitespace + - id: check-case-conflict + - id: check-yaml + - id: check-toml + - id: pretty-format-json + - id: check-docstring-first + + - repo: https://github.com/asottile/pyupgrade + rev: v2.37.3 + hooks: + - id: pyupgrade + name: Upgrade code + args: [ --py37-plus ] + + - repo: https://github.com/PyCQA/isort + rev: 5.10.1 + hooks: + - id: isort + name: Sort imports + + - repo: https://github.com/pre-commit/mirrors-yapf + rev: v0.32.0 + hooks: + - id: yapf + name: YAPF formatting + + - repo: https://github.com/executablebooks/mdformat + rev: 0.7.14 + hooks: + - id: mdformat + name: MD formatting + additional_dependencies: + - mdformat-gfm + - mdformat-black + exclude: "README.md|README_cn.md" + + - repo: https://github.com/asottile/yesqa + rev: v1.3.0 + hooks: + - id: yesqa + + - repo: https://github.com/PyCQA/flake8 + rev: 5.0.2 + hooks: + - id: flake8 + name: PEP8 diff --git a/application/yolov5_example/CONTRIBUTING.md b/application/yolov5_example/CONTRIBUTING.md new file mode 100644 index 00000000..13b9b73b --- /dev/null +++ b/application/yolov5_example/CONTRIBUTING.md @@ -0,0 +1,98 @@ +## Contributing to YOLOv5 馃殌 + +We love your input! We want to make contributing to YOLOv5 as easy and transparent as possible, whether it's: + +- Reporting a bug +- Discussing the current state of the code +- Submitting a fix +- Proposing a new feature +- Becoming a maintainer + +YOLOv5 works so well due to our combined community effort, and for every small improvement you contribute you will be +helping push the frontiers of what's possible in AI 馃槂! + +## Submitting a Pull Request (PR) 馃洜锔 + +Submitting a PR is easy! This example shows how to submit a PR for updating `requirements.txt` in 4 steps: + +### 1. Select File to Update + +Select `requirements.txt` to update by clicking on it in GitHub. + +

PR_step1

+ +### 2. Click 'Edit this file' + +Button is in top-right corner. + +

PR_step2

+ +### 3. Make Changes + +Change `matplotlib` version from `3.2.2` to `3.3`. + +

PR_step3

+ +### 4. Preview Changes and Submit PR + +Click on the **Preview changes** tab to verify your updates. At the bottom of the screen select 'Create a **new branch** +for this commit', assign your branch a descriptive name such as `fix/matplotlib_version` and click the green **Propose +changes** button. All done, your PR is now submitted to YOLOv5 for review and approval 馃槂! + +

PR_step4

+ +### PR recommendations + +To allow your work to be integrated as seamlessly as possible, we advise you to: + +- 鉁 Verify your PR is **up-to-date with upstream/master.** If your PR is behind upstream/master an + automatic [GitHub Actions](https://github.com/ultralytics/yolov5/blob/master/.github/workflows/rebase.yml) merge may + be attempted by writing /rebase in a new comment, or by running the following code, replacing 'feature' with the name + of your local branch: + +```bash +git remote add upstream https://github.com/ultralytics/yolov5.git +git fetch upstream +# git checkout feature # <--- replace 'feature' with local branch name +git merge upstream/master +git push -u origin -f +``` + +- 鉁 Verify all Continuous Integration (CI) **checks are passing**. +- 鉁 Reduce changes to the absolute **minimum** required for your bug fix or feature addition. _"It is not daily increase + but daily decrease, hack away the unessential. The closer to the source, the less wastage there is."_ 鈥 Bruce Lee + +## Submitting a Bug Report 馃悰 + +If you spot a problem with YOLOv5 please submit a Bug Report! + +For us to start investigating a possible problem we need to be able to reproduce it ourselves first. We've created a few +short guidelines below to help users provide what we need in order to get started. + +When asking a question, people will be better able to provide help if you provide **code** that they can easily +understand and use to **reproduce** the problem. This is referred to by community members as creating +a [minimum reproducible example](https://stackoverflow.com/help/minimal-reproducible-example). Your code that reproduces +the problem should be: + +- 鉁 **Minimal** 鈥 Use as little code as possible that still produces the same problem +- 鉁 **Complete** 鈥 Provide **all** parts someone else needs to reproduce your problem in the question itself +- 鉁 **Reproducible** 鈥 Test the code you're about to provide to make sure it reproduces the problem + +In addition to the above requirements, for [Ultralytics](https://ultralytics.com/) to provide assistance your code +should be: + +- 鉁 **Current** 鈥 Verify that your code is up-to-date with current + GitHub [master](https://github.com/ultralytics/yolov5/tree/master), and if necessary `git pull` or `git clone` a new + copy to ensure your problem has not already been resolved by previous commits. +- 鉁 **Unmodified** 鈥 Your problem must be reproducible without any modifications to the codebase in this + repository. [Ultralytics](https://ultralytics.com/) does not provide support for custom code 鈿狅笍. + +If you believe your problem meets all of the above criteria, please close this issue and raise a new one using the 馃悰 +**Bug Report** [template](https://github.com/ultralytics/yolov5/issues/new/choose) and providing +a [minimum reproducible example](https://stackoverflow.com/help/minimal-reproducible-example) to help us better +understand and diagnose your problem. + +## License + +By contributing, you agree that your contributions will be licensed under +the [GPL-3.0 license](https://choosealicense.com/licenses/gpl-3.0/) diff --git a/application/yolov5_example/LICENSE b/application/yolov5_example/LICENSE new file mode 100644 index 00000000..92b370f0 --- /dev/null +++ b/application/yolov5_example/LICENSE @@ -0,0 +1,674 @@ +GNU GENERAL PUBLIC LICENSE + Version 3, 29 June 2007 + + Copyright (C) 2007 Free Software Foundation, Inc. + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + + Preamble + + The GNU General Public License is a free, copyleft license for +software and other kinds of works. + + The licenses for most software and other practical works are designed +to take away your freedom to share and change the works. By contrast, +the GNU General Public License is intended to guarantee your freedom to +share and change all versions of a program--to make sure it remains free +software for all its users. We, the Free Software Foundation, use the +GNU General Public License for most of our software; it applies also to +any other work released this way by its authors. You can apply it to +your programs, too. + + When we speak of free software, we are referring to freedom, not +price. Our General Public Licenses are designed to make sure that you +have the freedom to distribute copies of free software (and charge for +them if you wish), that you receive source code or can get it if you +want it, that you can change the software or use pieces of it in new +free programs, and that you know you can do these things. + + To protect your rights, we need to prevent others from denying you +these rights or asking you to surrender the rights. Therefore, you have +certain responsibilities if you distribute copies of the software, or if +you modify it: responsibilities to respect the freedom of others. + + For example, if you distribute copies of such a program, whether +gratis or for a fee, you must pass on to the recipients the same +freedoms that you received. You must make sure that they, too, receive +or can get the source code. And you must show them these terms so they +know their rights. + + Developers that use the GNU GPL protect your rights with two steps: +(1) assert copyright on the software, and (2) offer you this License +giving you legal permission to copy, distribute and/or modify it. + + For the developers' and authors' protection, the GPL clearly explains +that there is no warranty for this free software. For both users' and +authors' sake, the GPL requires that modified versions be marked as +changed, so that their problems will not be attributed erroneously to +authors of previous versions. + + Some devices are designed to deny users access to install or run +modified versions of the software inside them, although the manufacturer +can do so. This is fundamentally incompatible with the aim of +protecting users' freedom to change the software. The systematic +pattern of such abuse occurs in the area of products for individuals to +use, which is precisely where it is most unacceptable. Therefore, we +have designed this version of the GPL to prohibit the practice for those +products. If such problems arise substantially in other domains, we +stand ready to extend this provision to those domains in future versions +of the GPL, as needed to protect the freedom of users. + + Finally, every program is threatened constantly by software patents. +States should not allow patents to restrict development and use of +software on general-purpose computers, but in those that do, we wish to +avoid the special danger that patents applied to a free program could +make it effectively proprietary. To prevent this, the GPL assures that +patents cannot be used to render the program non-free. + + The precise terms and conditions for copying, distribution and +modification follow. + + TERMS AND CONDITIONS + + 0. Definitions. + + "This License" refers to version 3 of the GNU General Public License. + + "Copyright" also means copyright-like laws that apply to other kinds of +works, such as semiconductor masks. + + "The Program" refers to any copyrightable work licensed under this +License. Each licensee is addressed as "you". "Licensees" and +"recipients" may be individuals or organizations. + + To "modify" a work means to copy from or adapt all or part of the work +in a fashion requiring copyright permission, other than the making of an +exact copy. The resulting work is called a "modified version" of the +earlier work or a work "based on" the earlier work. + + A "covered work" means either the unmodified Program or a work based +on the Program. + + To "propagate" a work means to do anything with it that, without +permission, would make you directly or secondarily liable for +infringement under applicable copyright law, except executing it on a +computer or modifying a private copy. Propagation includes copying, +distribution (with or without modification), making available to the +public, and in some countries other activities as well. + + To "convey" a work means any kind of propagation that enables other +parties to make or receive copies. Mere interaction with a user through +a computer network, with no transfer of a copy, is not conveying. + + An interactive user interface displays "Appropriate Legal Notices" +to the extent that it includes a convenient and prominently visible +feature that (1) displays an appropriate copyright notice, and (2) +tells the user that there is no warranty for the work (except to the +extent that warranties are provided), that licensees may convey the +work under this License, and how to view a copy of this License. If +the interface presents a list of user commands or options, such as a +menu, a prominent item in the list meets this criterion. + + 1. Source Code. + + The "source code" for a work means the preferred form of the work +for making modifications to it. "Object code" means any non-source +form of a work. + + A "Standard Interface" means an interface that either is an official +standard defined by a recognized standards body, or, in the case of +interfaces specified for a particular programming language, one that +is widely used among developers working in that language. + + The "System Libraries" of an executable work include anything, other +than the work as a whole, that (a) is included in the normal form of +packaging a Major Component, but which is not part of that Major +Component, and (b) serves only to enable use of the work with that +Major Component, or to implement a Standard Interface for which an +implementation is available to the public in source code form. A +"Major Component", in this context, means a major essential component +(kernel, window system, and so on) of the specific operating system +(if any) on which the executable work runs, or a compiler used to +produce the work, or an object code interpreter used to run it. + + The "Corresponding Source" for a work in object code form means all +the source code needed to generate, install, and (for an executable +work) run the object code and to modify the work, including scripts to +control those activities. However, it does not include the work's +System Libraries, or general-purpose tools or generally available free +programs which are used unmodified in performing those activities but +which are not part of the work. For example, Corresponding Source +includes interface definition files associated with source files for +the work, and the source code for shared libraries and dynamically +linked subprograms that the work is specifically designed to require, +such as by intimate data communication or control flow between those +subprograms and other parts of the work. + + The Corresponding Source need not include anything that users +can regenerate automatically from other parts of the Corresponding +Source. + + The Corresponding Source for a work in source code form is that +same work. + + 2. Basic Permissions. + + All rights granted under this License are granted for the term of +copyright on the Program, and are irrevocable provided the stated +conditions are met. This License explicitly affirms your unlimited +permission to run the unmodified Program. The output from running a +covered work is covered by this License only if the output, given its +content, constitutes a covered work. This License acknowledges your +rights of fair use or other equivalent, as provided by copyright law. + + You may make, run and propagate covered works that you do not +convey, without conditions so long as your license otherwise remains +in force. You may convey covered works to others for the sole purpose +of having them make modifications exclusively for you, or provide you +with facilities for running those works, provided that you comply with +the terms of this License in conveying all material for which you do +not control copyright. Those thus making or running the covered works +for you must do so exclusively on your behalf, under your direction +and control, on terms that prohibit them from making any copies of +your copyrighted material outside their relationship with you. + + Conveying under any other circumstances is permitted solely under +the conditions stated below. Sublicensing is not allowed; section 10 +makes it unnecessary. + + 3. Protecting Users' Legal Rights From Anti-Circumvention Law. + + No covered work shall be deemed part of an effective technological +measure under any applicable law fulfilling obligations under article +11 of the WIPO copyright treaty adopted on 20 December 1996, or +similar laws prohibiting or restricting circumvention of such +measures. + + When you convey a covered work, you waive any legal power to forbid +circumvention of technological measures to the extent such circumvention +is effected by exercising rights under this License with respect to +the covered work, and you disclaim any intention to limit operation or +modification of the work as a means of enforcing, against the work's +users, your or third parties' legal rights to forbid circumvention of +technological measures. + + 4. Conveying Verbatim Copies. + + You may convey verbatim copies of the Program's source code as you +receive it, in any medium, provided that you conspicuously and +appropriately publish on each copy an appropriate copyright notice; +keep intact all notices stating that this License and any +non-permissive terms added in accord with section 7 apply to the code; +keep intact all notices of the absence of any warranty; and give all +recipients a copy of this License along with the Program. + + You may charge any price or no price for each copy that you convey, +and you may offer support or warranty protection for a fee. + + 5. Conveying Modified Source Versions. + + You may convey a work based on the Program, or the modifications to +produce it from the Program, in the form of source code under the +terms of section 4, provided that you also meet all of these conditions: + + a) The work must carry prominent notices stating that you modified + it, and giving a relevant date. + + b) The work must carry prominent notices stating that it is + released under this License and any conditions added under section + 7. This requirement modifies the requirement in section 4 to + "keep intact all notices". + + c) You must license the entire work, as a whole, under this + License to anyone who comes into possession of a copy. This + License will therefore apply, along with any applicable section 7 + additional terms, to the whole of the work, and all its parts, + regardless of how they are packaged. This License gives no + permission to license the work in any other way, but it does not + invalidate such permission if you have separately received it. + + d) If the work has interactive user interfaces, each must display + Appropriate Legal Notices; however, if the Program has interactive + interfaces that do not display Appropriate Legal Notices, your + work need not make them do so. + + A compilation of a covered work with other separate and independent +works, which are not by their nature extensions of the covered work, +and which are not combined with it such as to form a larger program, +in or on a volume of a storage or distribution medium, is called an +"aggregate" if the compilation and its resulting copyright are not +used to limit the access or legal rights of the compilation's users +beyond what the individual works permit. Inclusion of a covered work +in an aggregate does not cause this License to apply to the other +parts of the aggregate. + + 6. Conveying Non-Source Forms. + + You may convey a covered work in object code form under the terms +of sections 4 and 5, provided that you also convey the +machine-readable Corresponding Source under the terms of this License, +in one of these ways: + + a) Convey the object code in, or embodied in, a physical product + (including a physical distribution medium), accompanied by the + Corresponding Source fixed on a durable physical medium + customarily used for software interchange. + + b) Convey the object code in, or embodied in, a physical product + (including a physical distribution medium), accompanied by a + written offer, valid for at least three years and valid for as + long as you offer spare parts or customer support for that product + model, to give anyone who possesses the object code either (1) a + copy of the Corresponding Source for all the software in the + product that is covered by this License, on a durable physical + medium customarily used for software interchange, for a price no + more than your reasonable cost of physically performing this + conveying of source, or (2) access to copy the + Corresponding Source from a network server at no charge. + + c) Convey individual copies of the object code with a copy of the + written offer to provide the Corresponding Source. This + alternative is allowed only occasionally and noncommercially, and + only if you received the object code with such an offer, in accord + with subsection 6b. + + d) Convey the object code by offering access from a designated + place (gratis or for a charge), and offer equivalent access to the + Corresponding Source in the same way through the same place at no + further charge. You need not require recipients to copy the + Corresponding Source along with the object code. If the place to + copy the object code is a network server, the Corresponding Source + may be on a different server (operated by you or a third party) + that supports equivalent copying facilities, provided you maintain + clear directions next to the object code saying where to find the + Corresponding Source. Regardless of what server hosts the + Corresponding Source, you remain obligated to ensure that it is + available for as long as needed to satisfy these requirements. + + e) Convey the object code using peer-to-peer transmission, provided + you inform other peers where the object code and Corresponding + Source of the work are being offered to the general public at no + charge under subsection 6d. + + A separable portion of the object code, whose source code is excluded +from the Corresponding Source as a System Library, need not be +included in conveying the object code work. + + A "User Product" is either (1) a "consumer product", which means any +tangible personal property which is normally used for personal, family, +or household purposes, or (2) anything designed or sold for incorporation +into a dwelling. In determining whether a product is a consumer product, +doubtful cases shall be resolved in favor of coverage. For a particular +product received by a particular user, "normally used" refers to a +typical or common use of that class of product, regardless of the status +of the particular user or of the way in which the particular user +actually uses, or expects or is expected to use, the product. A product +is a consumer product regardless of whether the product has substantial +commercial, industrial or non-consumer uses, unless such uses represent +the only significant mode of use of the product. + + "Installation Information" for a User Product means any methods, +procedures, authorization keys, or other information required to install +and execute modified versions of a covered work in that User Product from +a modified version of its Corresponding Source. The information must +suffice to ensure that the continued functioning of the modified object +code is in no case prevented or interfered with solely because +modification has been made. + + If you convey an object code work under this section in, or with, or +specifically for use in, a User Product, and the conveying occurs as +part of a transaction in which the right of possession and use of the +User Product is transferred to the recipient in perpetuity or for a +fixed term (regardless of how the transaction is characterized), the +Corresponding Source conveyed under this section must be accompanied +by the Installation Information. But this requirement does not apply +if neither you nor any third party retains the ability to install +modified object code on the User Product (for example, the work has +been installed in ROM). + + The requirement to provide Installation Information does not include a +requirement to continue to provide support service, warranty, or updates +for a work that has been modified or installed by the recipient, or for +the User Product in which it has been modified or installed. Access to a +network may be denied when the modification itself materially and +adversely affects the operation of the network or violates the rules and +protocols for communication across the network. + + Corresponding Source conveyed, and Installation Information provided, +in accord with this section must be in a format that is publicly +documented (and with an implementation available to the public in +source code form), and must require no special password or key for +unpacking, reading or copying. + + 7. Additional Terms. + + "Additional permissions" are terms that supplement the terms of this +License by making exceptions from one or more of its conditions. +Additional permissions that are applicable to the entire Program shall +be treated as though they were included in this License, to the extent +that they are valid under applicable law. If additional permissions +apply only to part of the Program, that part may be used separately +under those permissions, but the entire Program remains governed by +this License without regard to the additional permissions. + + When you convey a copy of a covered work, you may at your option +remove any additional permissions from that copy, or from any part of +it. (Additional permissions may be written to require their own +removal in certain cases when you modify the work.) You may place +additional permissions on material, added by you to a covered work, +for which you have or can give appropriate copyright permission. + + Notwithstanding any other provision of this License, for material you +add to a covered work, you may (if authorized by the copyright holders of +that material) supplement the terms of this License with terms: + + a) Disclaiming warranty or limiting liability differently from the + terms of sections 15 and 16 of this License; or + + b) Requiring preservation of specified reasonable legal notices or + author attributions in that material or in the Appropriate Legal + Notices displayed by works containing it; or + + c) Prohibiting misrepresentation of the origin of that material, or + requiring that modified versions of such material be marked in + reasonable ways as different from the original version; or + + d) Limiting the use for publicity purposes of names of licensors or + authors of the material; or + + e) Declining to grant rights under trademark law for use of some + trade names, trademarks, or service marks; or + + f) Requiring indemnification of licensors and authors of that + material by anyone who conveys the material (or modified versions of + it) with contractual assumptions of liability to the recipient, for + any liability that these contractual assumptions directly impose on + those licensors and authors. + + All other non-permissive additional terms are considered "further +restrictions" within the meaning of section 10. If the Program as you +received it, or any part of it, contains a notice stating that it is +governed by this License along with a term that is a further +restriction, you may remove that term. If a license document contains +a further restriction but permits relicensing or conveying under this +License, you may add to a covered work material governed by the terms +of that license document, provided that the further restriction does +not survive such relicensing or conveying. + + If you add terms to a covered work in accord with this section, you +must place, in the relevant source files, a statement of the +additional terms that apply to those files, or a notice indicating +where to find the applicable terms. + + Additional terms, permissive or non-permissive, may be stated in the +form of a separately written license, or stated as exceptions; +the above requirements apply either way. + + 8. Termination. + + You may not propagate or modify a covered work except as expressly +provided under this License. Any attempt otherwise to propagate or +modify it is void, and will automatically terminate your rights under +this License (including any patent licenses granted under the third +paragraph of section 11). + + However, if you cease all violation of this License, then your +license from a particular copyright holder is reinstated (a) +provisionally, unless and until the copyright holder explicitly and +finally terminates your license, and (b) permanently, if the copyright +holder fails to notify you of the violation by some reasonable means +prior to 60 days after the cessation. + + Moreover, your license from a particular copyright holder is +reinstated permanently if the copyright holder notifies you of the +violation by some reasonable means, this is the first time you have +received notice of violation of this License (for any work) from that +copyright holder, and you cure the violation prior to 30 days after +your receipt of the notice. + + Termination of your rights under this section does not terminate the +licenses of parties who have received copies or rights from you under +this License. If your rights have been terminated and not permanently +reinstated, you do not qualify to receive new licenses for the same +material under section 10. + + 9. Acceptance Not Required for Having Copies. + + You are not required to accept this License in order to receive or +run a copy of the Program. Ancillary propagation of a covered work +occurring solely as a consequence of using peer-to-peer transmission +to receive a copy likewise does not require acceptance. However, +nothing other than this License grants you permission to propagate or +modify any covered work. These actions infringe copyright if you do +not accept this License. Therefore, by modifying or propagating a +covered work, you indicate your acceptance of this License to do so. + + 10. Automatic Licensing of Downstream Recipients. + + Each time you convey a covered work, the recipient automatically +receives a license from the original licensors, to run, modify and +propagate that work, subject to this License. You are not responsible +for enforcing compliance by third parties with this License. + + An "entity transaction" is a transaction transferring control of an +organization, or substantially all assets of one, or subdividing an +organization, or merging organizations. If propagation of a covered +work results from an entity transaction, each party to that +transaction who receives a copy of the work also receives whatever +licenses to the work the party's predecessor in interest had or could +give under the previous paragraph, plus a right to possession of the +Corresponding Source of the work from the predecessor in interest, if +the predecessor has it or can get it with reasonable efforts. + + You may not impose any further restrictions on the exercise of the +rights granted or affirmed under this License. For example, you may +not impose a license fee, royalty, or other charge for exercise of +rights granted under this License, and you may not initiate litigation +(including a cross-claim or counterclaim in a lawsuit) alleging that +any patent claim is infringed by making, using, selling, offering for +sale, or importing the Program or any portion of it. + + 11. Patents. + + A "contributor" is a copyright holder who authorizes use under this +License of the Program or a work on which the Program is based. The +work thus licensed is called the contributor's "contributor version". + + A contributor's "essential patent claims" are all patent claims +owned or controlled by the contributor, whether already acquired or +hereafter acquired, that would be infringed by some manner, permitted +by this License, of making, using, or selling its contributor version, +but do not include claims that would be infringed only as a +consequence of further modification of the contributor version. For +purposes of this definition, "control" includes the right to grant +patent sublicenses in a manner consistent with the requirements of +this License. + + Each contributor grants you a non-exclusive, worldwide, royalty-free +patent license under the contributor's essential patent claims, to +make, use, sell, offer for sale, import and otherwise run, modify and +propagate the contents of its contributor version. + + In the following three paragraphs, a "patent license" is any express +agreement or commitment, however denominated, not to enforce a patent +(such as an express permission to practice a patent or covenant not to +sue for patent infringement). To "grant" such a patent license to a +party means to make such an agreement or commitment not to enforce a +patent against the party. + + If you convey a covered work, knowingly relying on a patent license, +and the Corresponding Source of the work is not available for anyone +to copy, free of charge and under the terms of this License, through a +publicly available network server or other readily accessible means, +then you must either (1) cause the Corresponding Source to be so +available, or (2) arrange to deprive yourself of the benefit of the +patent license for this particular work, or (3) arrange, in a manner +consistent with the requirements of this License, to extend the patent +license to downstream recipients. "Knowingly relying" means you have +actual knowledge that, but for the patent license, your conveying the +covered work in a country, or your recipient's use of the covered work +in a country, would infringe one or more identifiable patents in that +country that you have reason to believe are valid. + + If, pursuant to or in connection with a single transaction or +arrangement, you convey, or propagate by procuring conveyance of, a +covered work, and grant a patent license to some of the parties +receiving the covered work authorizing them to use, propagate, modify +or convey a specific copy of the covered work, then the patent license +you grant is automatically extended to all recipients of the covered +work and works based on it. + + A patent license is "discriminatory" if it does not include within +the scope of its coverage, prohibits the exercise of, or is +conditioned on the non-exercise of one or more of the rights that are +specifically granted under this License. You may not convey a covered +work if you are a party to an arrangement with a third party that is +in the business of distributing software, under which you make payment +to the third party based on the extent of your activity of conveying +the work, and under which the third party grants, to any of the +parties who would receive the covered work from you, a discriminatory +patent license (a) in connection with copies of the covered work +conveyed by you (or copies made from those copies), or (b) primarily +for and in connection with specific products or compilations that +contain the covered work, unless you entered into that arrangement, +or that patent license was granted, prior to 28 March 2007. + + Nothing in this License shall be construed as excluding or limiting +any implied license or other defenses to infringement that may +otherwise be available to you under applicable patent law. + + 12. No Surrender of Others' Freedom. + + If conditions are imposed on you (whether by court order, agreement or +otherwise) that contradict the conditions of this License, they do not +excuse you from the conditions of this License. If you cannot convey a +covered work so as to satisfy simultaneously your obligations under this +License and any other pertinent obligations, then as a consequence you may +not convey it at all. For example, if you agree to terms that obligate you +to collect a royalty for further conveying from those to whom you convey +the Program, the only way you could satisfy both those terms and this +License would be to refrain entirely from conveying the Program. + + 13. Use with the GNU Affero General Public License. + + Notwithstanding any other provision of this License, you have +permission to link or combine any covered work with a work licensed +under version 3 of the GNU Affero General Public License into a single +combined work, and to convey the resulting work. The terms of this +License will continue to apply to the part which is the covered work, +but the special requirements of the GNU Affero General Public License, +section 13, concerning interaction through a network will apply to the +combination as such. + + 14. Revised Versions of this License. + + The Free Software Foundation may publish revised and/or new versions of +the GNU General Public License from time to time. Such new versions will +be similar in spirit to the present version, but may differ in detail to +address new problems or concerns. + + Each version is given a distinguishing version number. If the +Program specifies that a certain numbered version of the GNU General +Public License "or any later version" applies to it, you have the +option of following the terms and conditions either of that numbered +version or of any later version published by the Free Software +Foundation. If the Program does not specify a version number of the +GNU General Public License, you may choose any version ever published +by the Free Software Foundation. + + If the Program specifies that a proxy can decide which future +versions of the GNU General Public License can be used, that proxy's +public statement of acceptance of a version permanently authorizes you +to choose that version for the Program. + + Later license versions may give you additional or different +permissions. However, no additional obligations are imposed on any +author or copyright holder as a result of your choosing to follow a +later version. + + 15. Disclaimer of Warranty. + + THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY +APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT +HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY +OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, +THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR +PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM +IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF +ALL NECESSARY SERVICING, REPAIR OR CORRECTION. + + 16. Limitation of Liability. + + IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING +WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS +THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY +GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE +USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF +DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD +PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), +EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF +SUCH DAMAGES. + + 17. Interpretation of Sections 15 and 16. + + If the disclaimer of warranty and limitation of liability provided +above cannot be given local legal effect according to their terms, +reviewing courts shall apply local law that most closely approximates +an absolute waiver of all civil liability in connection with the +Program, unless a warranty or assumption of liability accompanies a +copy of the Program in return for a fee. + + END OF TERMS AND CONDITIONS + + How to Apply These Terms to Your New Programs + + If you develop a new program, and you want it to be of the greatest +possible use to the public, the best way to achieve this is to make it +free software which everyone can redistribute and change under these terms. + + To do so, attach the following notices to the program. It is safest +to attach them to the start of each source file to most effectively +state the exclusion of warranty; and each file should have at least +the "copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + +Also add information on how to contact you by electronic and paper mail. + + If the program does terminal interaction, make it output a short +notice like this when it starts in an interactive mode: + + Copyright (C) + This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. + This is free software, and you are welcome to redistribute it + under certain conditions; type `show c' for details. + +The hypothetical commands `show w' and `show c' should show the appropriate +parts of the General Public License. Of course, your program's commands +might be different; for a GUI interface, you would use an "about box". + + You should also get your employer (if you work as a programmer) or school, +if any, to sign a "copyright disclaimer" for the program, if necessary. +For more information on this, and how to apply and follow the GNU GPL, see +. + + The GNU General Public License does not permit incorporating your program +into proprietary programs. If your program is a subroutine library, you +may consider it more useful to permit linking proprietary applications with +the library. If this is what you want to do, use the GNU Lesser General +Public License instead of this License. But first, please read +. diff --git a/application/yolov5_example/README.md b/application/yolov5_example/README.md new file mode 100644 index 00000000..b368d1d6 --- /dev/null +++ b/application/yolov5_example/README.md @@ -0,0 +1,363 @@ +
+

+ + +

+ +English | [绠浣撲腑鏂嘳(.github/README_cn.md) +
+
+ CI CPU testing + YOLOv5 Citation + Docker Pulls +
+ Open In Colab + Open In Kaggle + Join Forum +
+ +
+

+YOLOv5 馃殌 is a family of object detection architectures and models pretrained on the COCO dataset, and represents Ultralytics + open-source research into future vision AI methods, incorporating lessons learned and best practices evolved over thousands of hours of research and development. +

+ +
+ + + + + + + + + + + + + + + + + + + + +
+ + + +
+ +##
Documentation
+ +See the [YOLOv5 Docs](https://docs.ultralytics.com) for full documentation on training, testing and deployment. + +##
Quick Start Examples
+ +
+Install + +Clone repo and install [requirements.txt](https://github.com/ultralytics/yolov5/blob/master/requirements.txt) in a +[**Python>=3.7.0**](https://www.python.org/) environment, including +[**PyTorch>=1.7**](https://pytorch.org/get-started/locally/). + +```bash +git clone https://github.com/ultralytics/yolov5 # clone +cd yolov5 +pip install -r requirements.txt # install +``` + +
+ +
+Inference + +YOLOv5 [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36) inference. [Models](https://github.com/ultralytics/yolov5/tree/master/models) download automatically from the latest +YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). + +```python +import torch + +# Model +model = torch.hub.load('ultralytics/yolov5', 'yolov5s') # or yolov5n - yolov5x6, custom + +# Images +img = 'https://ultralytics.com/images/zidane.jpg' # or file, Path, PIL, OpenCV, numpy, list + +# Inference +results = model(img) + +# Results +results.print() # or .show(), .save(), .crop(), .pandas(), etc. +``` + +
+ +
+Inference with detect.py + +`detect.py` runs inference on a variety of sources, downloading [models](https://github.com/ultralytics/yolov5/tree/master/models) automatically from +the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases) and saving results to `runs/detect`. + +```bash +python detect.py --source 0 # webcam + img.jpg # image + vid.mp4 # video + path/ # directory + 'path/*.jpg' # glob + 'https://youtu.be/Zgi9g1ksQHc' # YouTube + 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream +``` + +
+ +
+Training + +The commands below reproduce YOLOv5 [COCO](https://github.com/ultralytics/yolov5/blob/master/data/scripts/get_coco.sh) +results. [Models](https://github.com/ultralytics/yolov5/tree/master/models) +and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) download automatically from the latest +YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training times for YOLOv5n/s/m/l/x are +1/2/4/6/8 days on a V100 GPU ([Multi-GPU](https://github.com/ultralytics/yolov5/issues/475) times faster). Use the +largest `--batch-size` possible, or pass `--batch-size -1` for +YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092). Batch sizes shown for V100-16GB. + +```bash +python train.py --data coco.yaml --cfg yolov5n.yaml --weights '' --batch-size 128 + yolov5s 64 + yolov5m 40 + yolov5l 24 + yolov5x 16 +``` + + + +
+ +
+Tutorials + +- [Train Custom Data](https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data)聽 馃殌 RECOMMENDED +- [Tips for Best Training Results](https://github.com/ultralytics/yolov5/wiki/Tips-for-Best-Training-Results)聽 鈽橈笍 + RECOMMENDED +- [Multi-GPU Training](https://github.com/ultralytics/yolov5/issues/475) +- [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36) 馃専 NEW +- [TFLite, ONNX, CoreML, TensorRT Export](https://github.com/ultralytics/yolov5/issues/251) 馃殌 +- [Test-Time Augmentation (TTA)](https://github.com/ultralytics/yolov5/issues/303) +- [Model Ensembling](https://github.com/ultralytics/yolov5/issues/318) +- [Model Pruning/Sparsity](https://github.com/ultralytics/yolov5/issues/304) +- [Hyperparameter Evolution](https://github.com/ultralytics/yolov5/issues/607) +- [Transfer Learning with Frozen Layers](https://github.com/ultralytics/yolov5/issues/1314) +- [Architecture Summary](https://github.com/ultralytics/yolov5/issues/6998) 馃専 NEW +- [Weights & Biases Logging](https://github.com/ultralytics/yolov5/issues/1289) +- [Roboflow for Datasets, Labeling, and Active Learning](https://github.com/ultralytics/yolov5/issues/4975)聽 馃専 NEW +- [ClearML Logging](https://github.com/ultralytics/yolov5/tree/master/utils/loggers/clearml) 馃専 NEW +- [Deci Platform](https://github.com/ultralytics/yolov5/wiki/Deci-Platform) 馃専 NEW + +
+ +##
Environments
+ +Get started in seconds with our verified environments. Click each icon below for details. + +
+ + + + + + + + + + + + + + +
+ +##
Integrations
+ +
+ + + + + + + + + + + +
+ +|Deci 猸 NEW|ClearML 猸 NEW|Roboflow|Weights & Biases +|:-:|:-:|:-:|:-:| +|Automatically compile and quantize YOLOv5 for better inference performance in one click at [Deci](https://bit.ly/yolov5-deci-platform)|Automatically track, visualize and even remotely train YOLOv5 using [ClearML](https://cutt.ly/yolov5-readme-clearml) (open-source!)|Label and export your custom datasets directly to YOLOv5 for training with [Roboflow](https://roboflow.com/?ref=ultralytics) |Automatically track and visualize all your YOLOv5 training runs in the cloud with [Weights & Biases](https://wandb.ai/site?utm_campaign=repo_yolo_readme) + + +##
Why YOLOv5
+ +

+
+ YOLOv5-P5 640 Figure (click to expand) + +

+
+
+ Figure Notes (click to expand) + +- **COCO AP val** denotes mAP@0.5:0.95 metric measured on the 5000-image [COCO val2017](http://cocodataset.org) dataset over various inference sizes from 256 to 1536. +- **GPU Speed** measures average inference time per image on [COCO val2017](http://cocodataset.org) dataset using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) V100 instance at batch-size 32. +- **EfficientDet** data from [google/automl](https://github.com/google/automl) at batch size 8. +- **Reproduce** by `python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n6.pt yolov5s6.pt yolov5m6.pt yolov5l6.pt yolov5x6.pt` + +
+ +### Pretrained Checkpoints + +| Model | size
(pixels) | mAPval
0.5:0.95 | mAPval
0.5 | Speed
CPU b1
(ms) | Speed
V100 b1
(ms) | Speed
V100 b32
(ms) | params
(M) | FLOPs
@640 (B) | +|------------------------------------------------------------------------------------------------------|-----------------------|-------------------------|--------------------|------------------------------|-------------------------------|--------------------------------|--------------------|------------------------| +| [YOLOv5n](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5n.pt) | 640 | 28.0 | 45.7 | **45** | **6.3** | **0.6** | **1.9** | **4.5** | +| [YOLOv5s](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5s.pt) | 640 | 37.4 | 56.8 | 98 | 6.4 | 0.9 | 7.2 | 16.5 | +| [YOLOv5m](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5m.pt) | 640 | 45.4 | 64.1 | 224 | 8.2 | 1.7 | 21.2 | 49.0 | +| [YOLOv5l](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5l.pt) | 640 | 49.0 | 67.3 | 430 | 10.1 | 2.7 | 46.5 | 109.1 | +| [YOLOv5x](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5x.pt) | 640 | 50.7 | 68.9 | 766 | 12.1 | 4.8 | 86.7 | 205.7 | +| | | | | | | | | | +| [YOLOv5n6](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5n6.pt) | 1280 | 36.0 | 54.4 | 153 | 8.1 | 2.1 | 3.2 | 4.6 | +| [YOLOv5s6](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5s6.pt) | 1280 | 44.8 | 63.7 | 385 | 8.2 | 3.6 | 12.6 | 16.8 | +| [YOLOv5m6](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5m6.pt) | 1280 | 51.3 | 69.3 | 887 | 11.1 | 6.8 | 35.7 | 50.0 | +| [YOLOv5l6](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5l6.pt) | 1280 | 53.7 | 71.3 | 1784 | 15.8 | 10.5 | 76.8 | 111.4 | +| [YOLOv5x6](https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5x6.pt)
+ [TTA][TTA] | 1280
1536 | 55.0
**55.8** | 72.7
**72.7** | 3136
- | 26.2
- | 19.4
- | 140.7
- | 209.8
- | + +
+ Table Notes (click to expand) + +- All checkpoints are trained to 300 epochs with default settings. Nano and Small models use [hyp.scratch-low.yaml](https://github.com/ultralytics/yolov5/blob/master/data/hyps/hyp.scratch-low.yaml) hyps, all others use [hyp.scratch-high.yaml](https://github.com/ultralytics/yolov5/blob/master/data/hyps/hyp.scratch-high.yaml). +- **mAPval** values are for single-model single-scale on [COCO val2017](http://cocodataset.org) dataset.
Reproduce by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65` +- **Speed** averaged over COCO val images using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) instance. NMS times (~1 ms/img) not included.
Reproduce by `python val.py --data coco.yaml --img 640 --task speed --batch 1` +- **TTA** [Test Time Augmentation](https://github.com/ultralytics/yolov5/issues/303) includes reflection and scale augmentations.
Reproduce by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment` + +
+ +##
Classification 猸 NEW
+ +YOLOv5 [release v6.2](https://github.com/ultralytics/yolov5/releases) brings support for classification model training, validation, prediction and export! We've made training classifier models super simple. Click below to get started. + +
+ Classification Checkpoints (click to expand) + +
+ +We trained YOLOv5-cls classification models on ImageNet for 90 epochs using a 4xA100 instance, and we trained ResNet and EfficientNet models alongside with the same default training settings to compare. We exported all models to ONNX FP32 for CPU speed tests and to TensorRT FP16 for GPU speed tests. We ran all speed tests on Google [Colab Pro](https://colab.research.google.com/signup) for easy reproducibility. + +| Model | size
(pixels) | acc
top1 | acc
top5 | Training
90 epochs
4xA100 (hours) | Speed
ONNX CPU
(ms) | Speed
TensorRT V100
(ms) | params
(M) | FLOPs
@224 (B) | +|----------------------------------------------------------------------------------------------------|-----------------------|------------------|------------------|----------------------------------------------|--------------------------------|-------------------------------------|--------------------|------------------------| +| [YOLOv5n-cls](https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5n-cls.pt) | 224 | 64.6 | 85.4 | 7:59 | **3.3** | **0.5** | **2.5** | **0.5** | +| [YOLOv5s-cls](https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5s-cls.pt) | 224 | 71.5 | 90.2 | 8:09 | 6.6 | 0.6 | 5.4 | 1.4 | +| [YOLOv5m-cls](https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5m-cls.pt) | 224 | 75.9 | 92.9 | 10:06 | 15.5 | 0.9 | 12.9 | 3.9 | +| [YOLOv5l-cls](https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5l-cls.pt) | 224 | 78.0 | 94.0 | 11:56 | 26.9 | 1.4 | 26.5 | 8.5 | +| [YOLOv5x-cls](https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5x-cls.pt) | 224 | **79.0** | **94.4** | 15:04 | 54.3 | 1.8 | 48.1 | 15.9 | +| | +| [ResNet18](https://github.com/ultralytics/yolov5/releases/download/v6.2/resnet18.pt) | 224 | 70.3 | 89.5 | **6:47** | 11.2 | 0.5 | 11.7 | 3.7 | +| [ResNet34](https://github.com/ultralytics/yolov5/releases/download/v6.2/resnet34.pt) | 224 | 73.9 | 91.8 | 8:33 | 20.6 | 0.9 | 21.8 | 7.4 | +| [ResNet50](https://github.com/ultralytics/yolov5/releases/download/v6.2/resnet50.pt) | 224 | 76.8 | 93.4 | 11:10 | 23.4 | 1.0 | 25.6 | 8.5 | +| [ResNet101](https://github.com/ultralytics/yolov5/releases/download/v6.2/resnet101.pt) | 224 | 78.5 | 94.3 | 17:10 | 42.1 | 1.9 | 44.5 | 15.9 | +| | +| [EfficientNet_b0](https://github.com/ultralytics/yolov5/releases/download/v6.2/efficientnet_b0.pt) | 224 | 75.1 | 92.4 | 13:03 | 12.5 | 1.3 | 5.3 | 1.0 | +| [EfficientNet_b1](https://github.com/ultralytics/yolov5/releases/download/v6.2/efficientnet_b1.pt) | 224 | 76.4 | 93.2 | 17:04 | 14.9 | 1.6 | 7.8 | 1.5 | +| [EfficientNet_b2](https://github.com/ultralytics/yolov5/releases/download/v6.2/efficientnet_b2.pt) | 224 | 76.6 | 93.4 | 17:10 | 15.9 | 1.6 | 9.1 | 1.7 | +| [EfficientNet_b3](https://github.com/ultralytics/yolov5/releases/download/v6.2/efficientnet_b3.pt) | 224 | 77.7 | 94.0 | 19:19 | 18.9 | 1.9 | 12.2 | 2.4 | + +
+ Table Notes (click to expand) + +- All checkpoints are trained to 90 epochs with SGD optimizer with lr0=0.001 at image size 224 and all default settings. Runs logged to https://wandb.ai/glenn-jocher/YOLOv5-Classifier-v6-2. +- **Accuracy** values are for single-model single-scale on [ImageNet-1k](https://www.image-net.org/index.php) dataset.
Reproduce by `python classify/val.py --data ../datasets/imagenet --img 224` +- **Speed** averaged over 100 inference images using a Google [Colab Pro](https://colab.research.google.com/signup) V100 High-RAM instance.
Reproduce by `python classify/val.py --data ../datasets/imagenet --img 224 --batch 1` +- **Export** to ONNX at FP32 and TensorRT at FP16 done with `export.py`.
Reproduce by `python export.py --weights yolov5s-cls.pt --include engine onnx --imgsz 224` +
+
+ +
+ Classification Usage Examples (click to expand) + +### Train +YOLOv5 classification training supports auto-download of MNIST, Fashion-MNIST, CIFAR10, CIFAR100, Imagenette, Imagewoof, and ImageNet datasets with the `--data` argument. To start training on MNIST for example use `--data mnist`. + +```bash +# Single-GPU +python classify/train.py --model yolov5s-cls.pt --data cifar100 --epochs 5 --img 224 --batch 128 + +# Multi-GPU DDP +python -m torch.distributed.run --nproc_per_node 4 --master_port 1 classify/train.py --model yolov5s-cls.pt --data imagenet --epochs 5 --img 224 --device 0,1,2,3 +``` + +### Val +Validate accuracy on a pretrained model. To validate YOLOv5s-cls accuracy on ImageNet. +```bash +bash data/scripts/get_imagenet.sh --val # download ImageNet val split (6.3G, 50000 images) +python classify/val.py --weights yolov5s-cls.pt --data ../datasets/imagenet --img 224 +``` + +### Predict +Run a classification prediction on an image. +```bash +python classify/predict.py --weights yolov5s-cls.pt --data data/images/bus.jpg +``` +```python +model = torch.hub.load('ultralytics/yolov5', 'custom', 'yolov5s-cls.pt') # load from PyTorch Hub +``` + +### Export +Export a group of trained YOLOv5-cls, ResNet and EfficientNet models to ONNX and TensorRT. +```bash +python export.py --weights yolov5s-cls.pt resnet50.pt efficientnet_b0.pt --include onnx engine --img 224 +``` +
+ + +##
Contribute
+ +We love your input! We want to make contributing to YOLOv5 as easy and transparent as possible. Please see our [Contributing Guide](CONTRIBUTING.md) to get started, and fill out the [YOLOv5 Survey](https://ultralytics.com/survey?utm_source=github&utm_medium=social&utm_campaign=Survey) to send us feedback on your experiences. Thank you to all our contributors! + + + + +##
Contact
+ +For YOLOv5 bugs and feature requests please visit [GitHub Issues](https://github.com/ultralytics/yolov5/issues). For business inquiries or +professional support requests please visit [https://ultralytics.com/contact](https://ultralytics.com/contact). + +
+
+ + + + + + + + + + + + + + + + + + + + +
+ +[assets]: https://github.com/ultralytics/yolov5/releases +[tta]: https://github.com/ultralytics/yolov5/issues/303 diff --git a/application/yolov5_example/classify/predict.py b/application/yolov5_example/classify/predict.py new file mode 100644 index 00000000..419830d4 --- /dev/null +++ b/application/yolov5_example/classify/predict.py @@ -0,0 +1,109 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Run classification inference on images + +Usage: + $ python classify/predict.py --weights yolov5s-cls.pt --source im.jpg +""" + +import argparse +import os +import sys +from pathlib import Path + +import cv2 +import torch.nn.functional as F + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from classify.train import imshow_cls +from models.common import DetectMultiBackend +from utils.augmentations import classify_transforms +from utils.general import LOGGER, check_requirements, colorstr, increment_path, print_args +from utils.torch_utils import select_device, smart_inference_mode, time_sync + + +@smart_inference_mode() +def run( + weights=ROOT / 'yolov5s-cls.pt', # model.pt path(s) + source=ROOT / 'data/images/bus.jpg', # file/dir/URL/glob, 0 for webcam + imgsz=224, # inference size + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + half=False, # use FP16 half-precision inference + dnn=False, # use OpenCV DNN for ONNX inference + show=True, + project=ROOT / 'runs/predict-cls', # save to project/name + name='exp', # save to project/name + exist_ok=False, # existing project/name ok, do not increment +): + file = str(source) + seen, dt = 1, [0.0, 0.0, 0.0] + device = select_device(device) + + # Directories + save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run + save_dir.mkdir(parents=True, exist_ok=True) # make dir + + # Transforms + transforms = classify_transforms(imgsz) + + # Load model + model = DetectMultiBackend(weights, device=device, dnn=dnn, fp16=half) + model.warmup(imgsz=(1, 3, imgsz, imgsz)) # warmup + + # Image + t1 = time_sync() + im = cv2.cvtColor(cv2.imread(file), cv2.COLOR_BGR2RGB) + im = transforms(im).unsqueeze(0).to(device) + im = im.half() if model.fp16 else im.float() + t2 = time_sync() + dt[0] += t2 - t1 + + # Inference + results = model(im) + t3 = time_sync() + dt[1] += t3 - t2 + + p = F.softmax(results, dim=1) # probabilities + i = p.argsort(1, descending=True)[:, :5].squeeze() # top 5 indices + dt[2] += time_sync() - t3 + LOGGER.info(f"image 1/1 {file}: {imgsz}x{imgsz} {', '.join(f'{model.names[j]} {p[0, j]:.2f}' for j in i)}") + + # Print results + t = tuple(x / seen * 1E3 for x in dt) # speeds per image + shape = (1, 3, imgsz, imgsz) + LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms post-process per image at shape {shape}' % t) + if show: + imshow_cls(im, f=save_dir / Path(file).name, verbose=True) + LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}") + return p + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5s-cls.pt', help='model path(s)') + parser.add_argument('--source', type=str, default=ROOT / 'data/images/bus.jpg', help='file') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=224, help='train, val image size (pixels)') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') + parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference') + parser.add_argument('--project', default=ROOT / 'runs/predict-cls', help='save to project/name') + parser.add_argument('--name', default='exp', help='save to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + opt = parser.parse_args() + print_args(vars(opt)) + return opt + + +def main(opt): + check_requirements(exclude=('tensorboard', 'thop')) + run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/classify/train.py b/application/yolov5_example/classify/train.py new file mode 100644 index 00000000..f2b46556 --- /dev/null +++ b/application/yolov5_example/classify/train.py @@ -0,0 +1,325 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Train a YOLOv5 classifier model on a classification dataset +Datasets: --data mnist, fashion-mnist, cifar10, cifar100, imagenette, imagewoof, imagenet, or 'path/to/custom/dataset' + +Usage: + $ python classify/train.py --model yolov5s-cls.pt --data cifar100 --epochs 5 --img 128 + $ python -m torch.distributed.run --nproc_per_node 4 --master_port 1 classify/train.py --model yolov5s-cls.pt --data imagenet --epochs 5 --img 224 --device 0,1,2,3 +""" + +import argparse +import os +import subprocess +import sys +import time +from copy import deepcopy +from datetime import datetime +from pathlib import Path + +import torch +import torch.distributed as dist +import torch.hub as hub +import torch.optim.lr_scheduler as lr_scheduler +import torchvision +from torch.cuda import amp +from tqdm import tqdm + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from classify import val as validate +from models.experimental import attempt_load +from models.yolo import ClassificationModel, DetectionModel +from utils.dataloaders import create_classification_dataloader +from utils.general import (DATASETS_DIR, LOGGER, WorkingDirectory, check_git_status, check_requirements, colorstr, + download, increment_path, init_seeds, print_args, yaml_save) +from utils.loggers import GenericLogger +from utils.plots import imshow_cls +from utils.torch_utils import (ModelEMA, model_info, reshape_classifier_output, select_device, smart_DDP, + smart_optimizer, smartCrossEntropyLoss, torch_distributed_zero_first) + +LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html +RANK = int(os.getenv('RANK', -1)) +WORLD_SIZE = int(os.getenv('WORLD_SIZE', 1)) + + +def train(opt, device): + init_seeds(opt.seed + 1 + RANK, deterministic=True) + save_dir, data, bs, epochs, nw, imgsz, pretrained = \ + opt.save_dir, Path(opt.data), opt.batch_size, opt.epochs, min(os.cpu_count() - 1, opt.workers), \ + opt.imgsz, str(opt.pretrained).lower() == 'true' + cuda = device.type != 'cpu' + + # Directories + wdir = save_dir / 'weights' + wdir.mkdir(parents=True, exist_ok=True) # make dir + last, best = wdir / 'last.pt', wdir / 'best.pt' + + # Save run settings + yaml_save(save_dir / 'opt.yaml', vars(opt)) + + # Logger + logger = GenericLogger(opt=opt, console_logger=LOGGER) if RANK in {-1, 0} else None + + # Download Dataset + with torch_distributed_zero_first(LOCAL_RANK), WorkingDirectory(ROOT): + data_dir = data if data.is_dir() else (DATASETS_DIR / data) + if not data_dir.is_dir(): + LOGGER.info(f'\nDataset not found 鈿狅笍, missing path {data_dir}, attempting download...') + t = time.time() + if str(data) == 'imagenet': + subprocess.run(f"bash {ROOT / 'data/scripts/get_imagenet.sh'}", shell=True, check=True) + else: + url = f'https://github.com/ultralytics/yolov5/releases/download/v1.0/{data}.zip' + download(url, dir=data_dir.parent) + s = f"Dataset download success 鉁 ({time.time() - t:.1f}s), saved to {colorstr('bold', data_dir)}\n" + LOGGER.info(s) + + # Dataloaders + nc = len([x for x in (data_dir / 'train').glob('*') if x.is_dir()]) # number of classes + trainloader = create_classification_dataloader(path=data_dir / 'train', + imgsz=imgsz, + batch_size=bs // WORLD_SIZE, + augment=True, + cache=opt.cache, + rank=LOCAL_RANK, + workers=nw) + + test_dir = data_dir / 'test' if (data_dir / 'test').exists() else data_dir / 'val' # data/test or data/val + if RANK in {-1, 0}: + testloader = create_classification_dataloader(path=test_dir, + imgsz=imgsz, + batch_size=bs // WORLD_SIZE * 2, + augment=False, + cache=opt.cache, + rank=-1, + workers=nw) + + # Model + with torch_distributed_zero_first(LOCAL_RANK), WorkingDirectory(ROOT): + if Path(opt.model).is_file() or opt.model.endswith('.pt'): + model = attempt_load(opt.model, device='cpu', fuse=False) + elif opt.model in torchvision.models.__dict__: # TorchVision models i.e. resnet50, efficientnet_b0 + model = torchvision.models.__dict__[opt.model](weights='IMAGENET1K_V1' if pretrained else None) + else: + m = hub.list('ultralytics/yolov5') # + hub.list('pytorch/vision') # models + raise ModuleNotFoundError(f'--model {opt.model} not found. Available models are: \n' + '\n'.join(m)) + if isinstance(model, DetectionModel): + LOGGER.warning("WARNING: pass YOLOv5 classifier model with '-cls' suffix, i.e. '--model yolov5s-cls.pt'") + model = ClassificationModel(model=model, nc=nc, cutoff=opt.cutoff or 10) # convert to classification model + reshape_classifier_output(model, nc) # update class count + for p in model.parameters(): + p.requires_grad = True # for training + for m in model.modules(): + if not pretrained and hasattr(m, 'reset_parameters'): + m.reset_parameters() + if isinstance(m, torch.nn.Dropout) and opt.dropout is not None: + m.p = opt.dropout # set dropout + model = model.to(device) + names = trainloader.dataset.classes # class names + model.names = names # attach class names + + # Info + if RANK in {-1, 0}: + model_info(model) + if opt.verbose: + LOGGER.info(model) + images, labels = next(iter(trainloader)) + file = imshow_cls(images[:25], labels[:25], names=names, f=save_dir / 'train_images.jpg') + logger.log_images(file, name='Train Examples') + logger.log_graph(model, imgsz) # log model + + # Optimizer + optimizer = smart_optimizer(model, opt.optimizer, opt.lr0, momentum=0.9, decay=5e-5) + + # Scheduler + lrf = 0.01 # final lr (fraction of lr0) + # lf = lambda x: ((1 + math.cos(x * math.pi / epochs)) / 2) * (1 - lrf) + lrf # cosine + lf = lambda x: (1 - x / epochs) * (1 - lrf) + lrf # linear + scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf) + # scheduler = lr_scheduler.OneCycleLR(optimizer, max_lr=lr0, total_steps=epochs, pct_start=0.1, + # final_div_factor=1 / 25 / lrf) + + # EMA + ema = ModelEMA(model) if RANK in {-1, 0} else None + + # DDP mode + if cuda and RANK != -1: + model = smart_DDP(model) + + # Train + t0 = time.time() + criterion = smartCrossEntropyLoss(label_smoothing=opt.label_smoothing) # loss function + best_fitness = 0.0 + scaler = amp.GradScaler(enabled=cuda) + val = test_dir.stem # 'val' or 'test' + LOGGER.info(f'Image sizes {imgsz} train, {imgsz} test\n' + f'Using {nw * WORLD_SIZE} dataloader workers\n' + f"Logging results to {colorstr('bold', save_dir)}\n" + f'Starting {opt.model} training on {data} dataset with {nc} classes for {epochs} epochs...\n\n' + f"{'Epoch':>10}{'GPU_mem':>10}{'train_loss':>12}{f'{val}_loss':>12}{'top1_acc':>12}{'top5_acc':>12}") + for epoch in range(epochs): # loop over the dataset multiple times + tloss, vloss, fitness = 0.0, 0.0, 0.0 # train loss, val loss, fitness + model.train() + if RANK != -1: + trainloader.sampler.set_epoch(epoch) + pbar = enumerate(trainloader) + if RANK in {-1, 0}: + pbar = tqdm(enumerate(trainloader), total=len(trainloader), bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') + for i, (images, labels) in pbar: # progress bar + images, labels = images.to(device, non_blocking=True), labels.to(device) + + # Forward + with amp.autocast(enabled=cuda): # stability issues when enabled + loss = criterion(model(images), labels) + + # Backward + scaler.scale(loss).backward() + + # Optimize + scaler.unscale_(optimizer) # unscale gradients + torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=10.0) # clip gradients + scaler.step(optimizer) + scaler.update() + optimizer.zero_grad() + if ema: + ema.update(model) + + if RANK in {-1, 0}: + # Print + tloss = (tloss * i + loss.item()) / (i + 1) # update mean losses + mem = '%.3gG' % (torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0) # (GB) + pbar.desc = f"{f'{epoch + 1}/{epochs}':>10}{mem:>10}{tloss:>12.3g}" + ' ' * 36 + + # Test + if i == len(pbar) - 1: # last batch + top1, top5, vloss = validate.run(model=ema.ema, + dataloader=testloader, + criterion=criterion, + pbar=pbar) # test accuracy, loss + fitness = top1 # define fitness as top1 accuracy + + # Scheduler + scheduler.step() + + # Log metrics + if RANK in {-1, 0}: + # Best fitness + if fitness > best_fitness: + best_fitness = fitness + + # Log + metrics = { + "train/loss": tloss, + f"{val}/loss": vloss, + "metrics/accuracy_top1": top1, + "metrics/accuracy_top5": top5, + "lr/0": optimizer.param_groups[0]['lr']} # learning rate + logger.log_metrics(metrics, epoch) + + # Save model + final_epoch = epoch + 1 == epochs + if (not opt.nosave) or final_epoch: + ckpt = { + 'epoch': epoch, + 'best_fitness': best_fitness, + 'model': deepcopy(ema.ema).half(), # deepcopy(de_parallel(model)).half(), + 'ema': None, # deepcopy(ema.ema).half(), + 'updates': ema.updates, + 'optimizer': None, # optimizer.state_dict(), + 'opt': vars(opt), + 'date': datetime.now().isoformat()} + + # Save last, best and delete + torch.save(ckpt, last) + if best_fitness == fitness: + torch.save(ckpt, best) + del ckpt + + # Train complete + if RANK in {-1, 0} and final_epoch: + LOGGER.info(f'\nTraining complete ({(time.time() - t0) / 3600:.3f} hours)' + f"\nResults saved to {colorstr('bold', save_dir)}" + f"\nPredict: python classify/predict.py --weights {best} --source im.jpg" + f"\nValidate: python classify/val.py --weights {best} --data {data_dir}" + f"\nExport: python export.py --weights {best} --include onnx" + f"\nPyTorch Hub: model = torch.hub.load('ultralytics/yolov5', 'custom', '{best}')" + f"\nVisualize: https://netron.app\n") + + # Plot examples + images, labels = (x[:25] for x in next(iter(testloader))) # first 25 images and labels + pred = torch.max(ema.ema((images.half() if cuda else images.float()).to(device)), 1)[1] + file = imshow_cls(images, labels, pred, names, verbose=False, f=save_dir / 'test_images.jpg') + + # Log results + meta = {"epochs": epochs, "top1_acc": best_fitness, "date": datetime.now().isoformat()} + logger.log_images(file, name='Test Examples (true-predicted)', epoch=epoch) + logger.log_model(best, epochs, metadata=meta) + + +def parse_opt(known=False): + parser = argparse.ArgumentParser() + parser.add_argument('--model', type=str, default='yolov5s-cls.pt', help='initial weights path') + parser.add_argument('--data', type=str, default='mnist', help='cifar10, cifar100, mnist, imagenet, etc.') + parser.add_argument('--epochs', type=int, default=10) + parser.add_argument('--batch-size', type=int, default=64, help='total batch size for all GPUs') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=128, help='train, val image size (pixels)') + parser.add_argument('--nosave', action='store_true', help='only save final checkpoint') + parser.add_argument('--cache', type=str, nargs='?', const='ram', help='--cache images in "ram" (default) or "disk"') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--workers', type=int, default=8, help='max dataloader workers (per RANK in DDP mode)') + parser.add_argument('--project', default=ROOT / 'runs/train-cls', help='save to project/name') + parser.add_argument('--name', default='exp', help='save to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + parser.add_argument('--pretrained', nargs='?', const=True, default=True, help='start from i.e. --pretrained False') + parser.add_argument('--optimizer', choices=['SGD', 'Adam', 'AdamW', 'RMSProp'], default='Adam', help='optimizer') + parser.add_argument('--lr0', type=float, default=0.001, help='initial learning rate') + parser.add_argument('--label-smoothing', type=float, default=0.1, help='Label smoothing epsilon') + parser.add_argument('--cutoff', type=int, default=None, help='Model layer cutoff index for Classify() head') + parser.add_argument('--dropout', type=float, default=None, help='Dropout (fraction)') + parser.add_argument('--verbose', action='store_true', help='Verbose mode') + parser.add_argument('--seed', type=int, default=0, help='Global training seed') + parser.add_argument('--local_rank', type=int, default=-1, help='Automatic DDP Multi-GPU argument, do not modify') + return parser.parse_known_args()[0] if known else parser.parse_args() + + +def main(opt): + # Checks + if RANK in {-1, 0}: + print_args(vars(opt)) + check_git_status() + check_requirements() + + # DDP mode + device = select_device(opt.device, batch_size=opt.batch_size) + if LOCAL_RANK != -1: + assert opt.batch_size != -1, 'AutoBatch is coming soon for classification, please pass a valid --batch-size' + assert opt.batch_size % WORLD_SIZE == 0, f'--batch-size {opt.batch_size} must be multiple of WORLD_SIZE' + assert torch.cuda.device_count() > LOCAL_RANK, 'insufficient CUDA devices for DDP command' + torch.cuda.set_device(LOCAL_RANK) + device = torch.device('cuda', LOCAL_RANK) + dist.init_process_group(backend="nccl" if dist.is_nccl_available() else "gloo") + + # Parameters + opt.save_dir = increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok) # increment run + + # Train + train(opt, device) + + +def run(**kwargs): + # Usage: from yolov5 import classify; classify.train.run(data=mnist, imgsz=320, model='yolov5m') + opt = parse_opt(True) + for k, v in kwargs.items(): + setattr(opt, k, v) + main(opt) + return opt + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/classify/val.py b/application/yolov5_example/classify/val.py new file mode 100644 index 00000000..0930ba8c --- /dev/null +++ b/application/yolov5_example/classify/val.py @@ -0,0 +1,158 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Validate a classification model on a dataset + +Usage: + $ python classify/val.py --weights yolov5s-cls.pt --data ../datasets/imagenet +""" + +import argparse +import os +import sys +from pathlib import Path + +import torch +from tqdm import tqdm + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from models.common import DetectMultiBackend +from utils.dataloaders import create_classification_dataloader +from utils.general import LOGGER, check_img_size, check_requirements, colorstr, increment_path, print_args +from utils.torch_utils import select_device, smart_inference_mode, time_sync + + +@smart_inference_mode() +def run( + data=ROOT / '../datasets/mnist', # dataset dir + weights=ROOT / 'yolov5s-cls.pt', # model.pt path(s) + batch_size=128, # batch size + imgsz=224, # inference size (pixels) + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + workers=8, # max dataloader workers (per RANK in DDP mode) + verbose=False, # verbose output + project=ROOT / 'runs/val-cls', # save to project/name + name='exp', # save to project/name + exist_ok=False, # existing project/name ok, do not increment + half=True, # use FP16 half-precision inference + dnn=False, # use OpenCV DNN for ONNX inference + model=None, + dataloader=None, + criterion=None, + pbar=None, +): + # Initialize/load model and set device + training = model is not None + if training: # called by train.py + device, pt, jit, engine = next(model.parameters()).device, True, False, False # get model device, PyTorch model + half &= device.type != 'cpu' # half precision only supported on CUDA + model.half() if half else model.float() + else: # called directly + device = select_device(device, batch_size=batch_size) + + # Directories + save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run + save_dir.mkdir(parents=True, exist_ok=True) # make dir + + # Load model + model = DetectMultiBackend(weights, device=device, dnn=dnn, fp16=half) + stride, pt, jit, engine = model.stride, model.pt, model.jit, model.engine + imgsz = check_img_size(imgsz, s=stride) # check image size + half = model.fp16 # FP16 supported on limited backends with CUDA + if engine: + batch_size = model.batch_size + else: + device = model.device + if not (pt or jit): + batch_size = 1 # export.py models default to batch-size 1 + LOGGER.info(f'Forcing --batch-size 1 square inference (1,3,{imgsz},{imgsz}) for non-PyTorch models') + + # Dataloader + data = Path(data) + test_dir = data / 'test' if (data / 'test').exists() else data / 'val' # data/test or data/val + dataloader = create_classification_dataloader(path=test_dir, + imgsz=imgsz, + batch_size=batch_size, + augment=False, + rank=-1, + workers=workers) + + model.eval() + pred, targets, loss, dt = [], [], 0, [0.0, 0.0, 0.0] + n = len(dataloader) # number of batches + action = 'validating' if dataloader.dataset.root.stem == 'val' else 'testing' + desc = f"{pbar.desc[:-36]}{action:>36}" if pbar else f"{action}" + bar = tqdm(dataloader, desc, n, not training, bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}', position=0) + with torch.cuda.amp.autocast(enabled=device.type != 'cpu'): + for images, labels in bar: + t1 = time_sync() + images, labels = images.to(device, non_blocking=True), labels.to(device) + t2 = time_sync() + dt[0] += t2 - t1 + + y = model(images) + t3 = time_sync() + dt[1] += t3 - t2 + + pred.append(y.argsort(1, descending=True)[:, :5]) + targets.append(labels) + if criterion: + loss += criterion(y, labels) + dt[2] += time_sync() - t3 + + loss /= n + pred, targets = torch.cat(pred), torch.cat(targets) + correct = (targets[:, None] == pred).float() + acc = torch.stack((correct[:, 0], correct.max(1).values), dim=1) # (top1, top5) accuracy + top1, top5 = acc.mean(0).tolist() + + if pbar: + pbar.desc = f"{pbar.desc[:-36]}{loss:>12.3g}{top1:>12.3g}{top5:>12.3g}" + if verbose: # all classes + LOGGER.info(f"{'Class':>24}{'Images':>12}{'top1_acc':>12}{'top5_acc':>12}") + LOGGER.info(f"{'all':>24}{targets.shape[0]:>12}{top1:>12.3g}{top5:>12.3g}") + for i, c in enumerate(model.names): + aci = acc[targets == i] + top1i, top5i = aci.mean(0).tolist() + LOGGER.info(f"{c:>24}{aci.shape[0]:>12}{top1i:>12.3g}{top5i:>12.3g}") + + # Print results + t = tuple(x / len(dataloader.dataset.samples) * 1E3 for x in dt) # speeds per image + shape = (1, 3, imgsz, imgsz) + LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms post-process per image at shape {shape}' % t) + LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}") + + return top1, top5, loss + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--data', type=str, default=ROOT / '../datasets/mnist', help='dataset path') + parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5s-cls.pt', help='model.pt path(s)') + parser.add_argument('--batch-size', type=int, default=128, help='batch size') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=224, help='inference size (pixels)') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--workers', type=int, default=8, help='max dataloader workers (per RANK in DDP mode)') + parser.add_argument('--verbose', nargs='?', const=True, default=True, help='verbose output') + parser.add_argument('--project', default=ROOT / 'runs/val-cls', help='save to project/name') + parser.add_argument('--name', default='exp', help='save to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') + parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference') + opt = parser.parse_args() + print_args(vars(opt)) + return opt + + +def main(opt): + check_requirements(exclude=('tensorboard', 'thop')) + run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/data/Argoverse.yaml b/application/yolov5_example/data/Argoverse.yaml new file mode 100644 index 00000000..9d21296e --- /dev/null +++ b/application/yolov5_example/data/Argoverse.yaml @@ -0,0 +1,67 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Argoverse-HD dataset (ring-front-center camera) http://www.cs.cmu.edu/~mengtial/proj/streaming/ by Argo AI +# Example usage: python train.py --data Argoverse.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 Argoverse 鈫 downloads here (31.3 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/Argoverse # dataset root dir +train: Argoverse-1.1/images/train/ # train images (relative to 'path') 39384 images +val: Argoverse-1.1/images/val/ # val images (relative to 'path') 15062 images +test: Argoverse-1.1/images/test/ # test images (optional) https://eval.ai/web/challenges/challenge-page/800/overview + +# Classes +nc: 8 # number of classes +names: ['person', 'bicycle', 'car', 'motorcycle', 'bus', 'truck', 'traffic_light', 'stop_sign'] # class names + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + import json + + from tqdm import tqdm + from utils.general import download, Path + + + def argoverse2yolo(set): + labels = {} + a = json.load(open(set, "rb")) + for annot in tqdm(a['annotations'], desc=f"Converting {set} to YOLOv5 format..."): + img_id = annot['image_id'] + img_name = a['images'][img_id]['name'] + img_label_name = f'{img_name[:-3]}txt' + + cls = annot['category_id'] # instance class id + x_center, y_center, width, height = annot['bbox'] + x_center = (x_center + width / 2) / 1920.0 # offset and scale + y_center = (y_center + height / 2) / 1200.0 # offset and scale + width /= 1920.0 # scale + height /= 1200.0 # scale + + img_dir = set.parents[2] / 'Argoverse-1.1' / 'labels' / a['seq_dirs'][a['images'][annot['image_id']]['sid']] + if not img_dir.exists(): + img_dir.mkdir(parents=True, exist_ok=True) + + k = str(img_dir / img_label_name) + if k not in labels: + labels[k] = [] + labels[k].append(f"{cls} {x_center} {y_center} {width} {height}\n") + + for k in labels: + with open(k, "w") as f: + f.writelines(labels[k]) + + + # Download + dir = Path('../datasets/Argoverse') # dataset root dir + urls = ['https://argoverse-hd.s3.us-east-2.amazonaws.com/Argoverse-HD-Full.zip'] + download(urls, dir=dir, delete=False) + + # Convert + annotations_dir = 'Argoverse-HD/annotations/' + (dir / 'Argoverse-1.1' / 'tracking').rename(dir / 'Argoverse-1.1' / 'images') # rename 'tracking' to 'images' + for d in "train.json", "val.json": + argoverse2yolo(dir / annotations_dir / d) # convert VisDrone annotations to YOLO labels diff --git a/application/yolov5_example/data/GlobalWheat2020.yaml b/application/yolov5_example/data/GlobalWheat2020.yaml new file mode 100644 index 00000000..4c43693f --- /dev/null +++ b/application/yolov5_example/data/GlobalWheat2020.yaml @@ -0,0 +1,54 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Global Wheat 2020 dataset http://www.global-wheat.com/ by University of Saskatchewan +# Example usage: python train.py --data GlobalWheat2020.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 GlobalWheat2020 鈫 downloads here (7.0 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/GlobalWheat2020 # dataset root dir +train: # train images (relative to 'path') 3422 images + - images/arvalis_1 + - images/arvalis_2 + - images/arvalis_3 + - images/ethz_1 + - images/rres_1 + - images/inrae_1 + - images/usask_1 +val: # val images (relative to 'path') 748 images (WARNING: train set contains ethz_1) + - images/ethz_1 +test: # test images (optional) 1276 images + - images/utokyo_1 + - images/utokyo_2 + - images/nau_1 + - images/uq_1 + +# Classes +nc: 1 # number of classes +names: ['wheat_head'] # class names + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + from utils.general import download, Path + + + # Download + dir = Path(yaml['path']) # dataset root dir + urls = ['https://zenodo.org/record/4298502/files/global-wheat-codalab-official.zip', + 'https://github.com/ultralytics/yolov5/releases/download/v1.0/GlobalWheat2020_labels.zip'] + download(urls, dir=dir) + + # Make Directories + for p in 'annotations', 'images', 'labels': + (dir / p).mkdir(parents=True, exist_ok=True) + + # Move + for p in 'arvalis_1', 'arvalis_2', 'arvalis_3', 'ethz_1', 'rres_1', 'inrae_1', 'usask_1', \ + 'utokyo_1', 'utokyo_2', 'nau_1', 'uq_1': + (dir / p).rename(dir / 'images' / p) # move to /images + f = (dir / p).with_suffix('.json') # json file + if f.exists(): + f.rename((dir / 'annotations' / p).with_suffix('.json')) # move to /annotations diff --git a/application/yolov5_example/data/ImageNet.yaml b/application/yolov5_example/data/ImageNet.yaml new file mode 100644 index 00000000..9f89b426 --- /dev/null +++ b/application/yolov5_example/data/ImageNet.yaml @@ -0,0 +1,156 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# ImageNet-1k dataset https://www.image-net.org/index.php by Stanford University +# Simplified class names from https://github.com/anishathalye/imagenet-simple-labels +# Example usage: python classify/train.py --data imagenet +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 imagenet 鈫 downloads here (144 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/imagenet # dataset root dir +train: train # train images (relative to 'path') 1281167 images +val: val # val images (relative to 'path') 50000 images +test: # test images (optional) + +# Classes +nc: 1000 # number of classes +names: ['tench', 'goldfish', 'great white shark', 'tiger shark', 'hammerhead shark', 'electric ray', 'stingray', 'cock', + 'hen', 'ostrich', 'brambling', 'goldfinch', 'house finch', 'junco', 'indigo bunting', 'American robin', + 'bulbul', 'jay', 'magpie', 'chickadee', 'American dipper', 'kite', 'bald eagle', 'vulture', 'great grey owl', + 'fire salamander', 'smooth newt', 'newt', 'spotted salamander', 'axolotl', 'American bullfrog', 'tree frog', + 'tailed frog', 'loggerhead sea turtle', 'leatherback sea turtle', 'mud turtle', 'terrapin', 'box turtle', + 'banded gecko', 'green iguana', 'Carolina anole', 'desert grassland whiptail lizard', 'agama', + 'frilled-necked lizard', 'alligator lizard', 'Gila monster', 'European green lizard', 'chameleon', + 'Komodo dragon', 'Nile crocodile', 'American alligator', 'triceratops', 'worm snake', 'ring-necked snake', + 'eastern hog-nosed snake', 'smooth green snake', 'kingsnake', 'garter snake', 'water snake', 'vine snake', + 'night snake', 'boa constrictor', 'African rock python', 'Indian cobra', 'green mamba', 'sea snake', + 'Saharan horned viper', 'eastern diamondback rattlesnake', 'sidewinder', 'trilobite', 'harvestman', 'scorpion', + 'yellow garden spider', 'barn spider', 'European garden spider', 'southern black widow', 'tarantula', + 'wolf spider', 'tick', 'centipede', 'black grouse', 'ptarmigan', 'ruffed grouse', 'prairie grouse', 'peacock', + 'quail', 'partridge', 'grey parrot', 'macaw', 'sulphur-crested cockatoo', 'lorikeet', 'coucal', 'bee eater', + 'hornbill', 'hummingbird', 'jacamar', 'toucan', 'duck', 'red-breasted merganser', 'goose', 'black swan', + 'tusker', 'echidna', 'platypus', 'wallaby', 'koala', 'wombat', 'jellyfish', 'sea anemone', 'brain coral', + 'flatworm', 'nematode', 'conch', 'snail', 'slug', 'sea slug', 'chiton', 'chambered nautilus', 'Dungeness crab', + 'rock crab', 'fiddler crab', 'red king crab', 'American lobster', 'spiny lobster', 'crayfish', 'hermit crab', + 'isopod', 'white stork', 'black stork', 'spoonbill', 'flamingo', 'little blue heron', 'great egret', 'bittern', + 'crane (bird)', 'limpkin', 'common gallinule', 'American coot', 'bustard', 'ruddy turnstone', 'dunlin', + 'common redshank', 'dowitcher', 'oystercatcher', 'pelican', 'king penguin', 'albatross', 'grey whale', + 'killer whale', 'dugong', 'sea lion', 'Chihuahua', 'Japanese Chin', 'Maltese', 'Pekingese', 'Shih Tzu', + 'King Charles Spaniel', 'Papillon', 'toy terrier', 'Rhodesian Ridgeback', 'Afghan Hound', 'Basset Hound', + 'Beagle', 'Bloodhound', 'Bluetick Coonhound', 'Black and Tan Coonhound', 'Treeing Walker Coonhound', + 'English foxhound', 'Redbone Coonhound', 'borzoi', 'Irish Wolfhound', 'Italian Greyhound', 'Whippet', + 'Ibizan Hound', 'Norwegian Elkhound', 'Otterhound', 'Saluki', 'Scottish Deerhound', 'Weimaraner', + 'Staffordshire Bull Terrier', 'American Staffordshire Terrier', 'Bedlington Terrier', 'Border Terrier', + 'Kerry Blue Terrier', 'Irish Terrier', 'Norfolk Terrier', 'Norwich Terrier', 'Yorkshire Terrier', + 'Wire Fox Terrier', 'Lakeland Terrier', 'Sealyham Terrier', 'Airedale Terrier', 'Cairn Terrier', + 'Australian Terrier', 'Dandie Dinmont Terrier', 'Boston Terrier', 'Miniature Schnauzer', 'Giant Schnauzer', + 'Standard Schnauzer', 'Scottish Terrier', 'Tibetan Terrier', 'Australian Silky Terrier', + 'Soft-coated Wheaten Terrier', 'West Highland White Terrier', 'Lhasa Apso', 'Flat-Coated Retriever', + 'Curly-coated Retriever', 'Golden Retriever', 'Labrador Retriever', 'Chesapeake Bay Retriever', + 'German Shorthaired Pointer', 'Vizsla', 'English Setter', 'Irish Setter', 'Gordon Setter', 'Brittany', + 'Clumber Spaniel', 'English Springer Spaniel', 'Welsh Springer Spaniel', 'Cocker Spaniels', 'Sussex Spaniel', + 'Irish Water Spaniel', 'Kuvasz', 'Schipperke', 'Groenendael', 'Malinois', 'Briard', 'Australian Kelpie', + 'Komondor', 'Old English Sheepdog', 'Shetland Sheepdog', 'collie', 'Border Collie', 'Bouvier des Flandres', + 'Rottweiler', 'German Shepherd Dog', 'Dobermann', 'Miniature Pinscher', 'Greater Swiss Mountain Dog', + 'Bernese Mountain Dog', 'Appenzeller Sennenhund', 'Entlebucher Sennenhund', 'Boxer', 'Bullmastiff', + 'Tibetan Mastiff', 'French Bulldog', 'Great Dane', 'St. Bernard', 'husky', 'Alaskan Malamute', 'Siberian Husky', + 'Dalmatian', 'Affenpinscher', 'Basenji', 'pug', 'Leonberger', 'Newfoundland', 'Pyrenean Mountain Dog', + 'Samoyed', 'Pomeranian', 'Chow Chow', 'Keeshond', 'Griffon Bruxellois', 'Pembroke Welsh Corgi', + 'Cardigan Welsh Corgi', 'Toy Poodle', 'Miniature Poodle', 'Standard Poodle', 'Mexican hairless dog', + 'grey wolf', 'Alaskan tundra wolf', 'red wolf', 'coyote', 'dingo', 'dhole', 'African wild dog', 'hyena', + 'red fox', 'kit fox', 'Arctic fox', 'grey fox', 'tabby cat', 'tiger cat', 'Persian cat', 'Siamese cat', + 'Egyptian Mau', 'cougar', 'lynx', 'leopard', 'snow leopard', 'jaguar', 'lion', 'tiger', 'cheetah', 'brown bear', + 'American black bear', 'polar bear', 'sloth bear', 'mongoose', 'meerkat', 'tiger beetle', 'ladybug', + 'ground beetle', 'longhorn beetle', 'leaf beetle', 'dung beetle', 'rhinoceros beetle', 'weevil', 'fly', 'bee', + 'ant', 'grasshopper', 'cricket', 'stick insect', 'cockroach', 'mantis', 'cicada', 'leafhopper', 'lacewing', + 'dragonfly', 'damselfly', 'red admiral', 'ringlet', 'monarch butterfly', 'small white', 'sulphur butterfly', + 'gossamer-winged butterfly', 'starfish', 'sea urchin', 'sea cucumber', 'cottontail rabbit', 'hare', + 'Angora rabbit', 'hamster', 'porcupine', 'fox squirrel', 'marmot', 'beaver', 'guinea pig', 'common sorrel', + 'zebra', 'pig', 'wild boar', 'warthog', 'hippopotamus', 'ox', 'water buffalo', 'bison', 'ram', 'bighorn sheep', + 'Alpine ibex', 'hartebeest', 'impala', 'gazelle', 'dromedary', 'llama', 'weasel', 'mink', 'European polecat', + 'black-footed ferret', 'otter', 'skunk', 'badger', 'armadillo', 'three-toed sloth', 'orangutan', 'gorilla', + 'chimpanzee', 'gibbon', 'siamang', 'guenon', 'patas monkey', 'baboon', 'macaque', 'langur', + 'black-and-white colobus', 'proboscis monkey', 'marmoset', 'white-headed capuchin', 'howler monkey', 'titi', + "Geoffroy's spider monkey", 'common squirrel monkey', 'ring-tailed lemur', 'indri', 'Asian elephant', + 'African bush elephant', 'red panda', 'giant panda', 'snoek', 'eel', 'coho salmon', 'rock beauty', 'clownfish', + 'sturgeon', 'garfish', 'lionfish', 'pufferfish', 'abacus', 'abaya', 'academic gown', 'accordion', + 'acoustic guitar', 'aircraft carrier', 'airliner', 'airship', 'altar', 'ambulance', 'amphibious vehicle', + 'analog clock', 'apiary', 'apron', 'waste container', 'assault rifle', 'backpack', 'bakery', 'balance beam', + 'balloon', 'ballpoint pen', 'Band-Aid', 'banjo', 'baluster', 'barbell', 'barber chair', 'barbershop', 'barn', + 'barometer', 'barrel', 'wheelbarrow', 'baseball', 'basketball', 'bassinet', 'bassoon', 'swimming cap', + 'bath towel', 'bathtub', 'station wagon', 'lighthouse', 'beaker', 'military cap', 'beer bottle', 'beer glass', + 'bell-cot', 'bib', 'tandem bicycle', 'bikini', 'ring binder', 'binoculars', 'birdhouse', 'boathouse', + 'bobsleigh', 'bolo tie', 'poke bonnet', 'bookcase', 'bookstore', 'bottle cap', 'bow', 'bow tie', 'brass', 'bra', + 'breakwater', 'breastplate', 'broom', 'bucket', 'buckle', 'bulletproof vest', 'high-speed train', + 'butcher shop', 'taxicab', 'cauldron', 'candle', 'cannon', 'canoe', 'can opener', 'cardigan', 'car mirror', + 'carousel', 'tool kit', 'carton', 'car wheel', 'automated teller machine', 'cassette', 'cassette player', + 'castle', 'catamaran', 'CD player', 'cello', 'mobile phone', 'chain', 'chain-link fence', 'chain mail', + 'chainsaw', 'chest', 'chiffonier', 'chime', 'china cabinet', 'Christmas stocking', 'church', 'movie theater', + 'cleaver', 'cliff dwelling', 'cloak', 'clogs', 'cocktail shaker', 'coffee mug', 'coffeemaker', 'coil', + 'combination lock', 'computer keyboard', 'confectionery store', 'container ship', 'convertible', 'corkscrew', + 'cornet', 'cowboy boot', 'cowboy hat', 'cradle', 'crane (machine)', 'crash helmet', 'crate', 'infant bed', + 'Crock Pot', 'croquet ball', 'crutch', 'cuirass', 'dam', 'desk', 'desktop computer', 'rotary dial telephone', + 'diaper', 'digital clock', 'digital watch', 'dining table', 'dishcloth', 'dishwasher', 'disc brake', 'dock', + 'dog sled', 'dome', 'doormat', 'drilling rig', 'drum', 'drumstick', 'dumbbell', 'Dutch oven', 'electric fan', + 'electric guitar', 'electric locomotive', 'entertainment center', 'envelope', 'espresso machine', 'face powder', + 'feather boa', 'filing cabinet', 'fireboat', 'fire engine', 'fire screen sheet', 'flagpole', 'flute', + 'folding chair', 'football helmet', 'forklift', 'fountain', 'fountain pen', 'four-poster bed', 'freight car', + 'French horn', 'frying pan', 'fur coat', 'garbage truck', 'gas mask', 'gas pump', 'goblet', 'go-kart', + 'golf ball', 'golf cart', 'gondola', 'gong', 'gown', 'grand piano', 'greenhouse', 'grille', 'grocery store', + 'guillotine', 'barrette', 'hair spray', 'half-track', 'hammer', 'hamper', 'hair dryer', 'hand-held computer', + 'handkerchief', 'hard disk drive', 'harmonica', 'harp', 'harvester', 'hatchet', 'holster', 'home theater', + 'honeycomb', 'hook', 'hoop skirt', 'horizontal bar', 'horse-drawn vehicle', 'hourglass', 'iPod', 'clothes iron', + "jack-o'-lantern", 'jeans', 'jeep', 'T-shirt', 'jigsaw puzzle', 'pulled rickshaw', 'joystick', 'kimono', + 'knee pad', 'knot', 'lab coat', 'ladle', 'lampshade', 'laptop computer', 'lawn mower', 'lens cap', + 'paper knife', 'library', 'lifeboat', 'lighter', 'limousine', 'ocean liner', 'lipstick', 'slip-on shoe', + 'lotion', 'speaker', 'loupe', 'sawmill', 'magnetic compass', 'mail bag', 'mailbox', 'tights', 'tank suit', + 'manhole cover', 'maraca', 'marimba', 'mask', 'match', 'maypole', 'maze', 'measuring cup', 'medicine chest', + 'megalith', 'microphone', 'microwave oven', 'military uniform', 'milk can', 'minibus', 'miniskirt', 'minivan', + 'missile', 'mitten', 'mixing bowl', 'mobile home', 'Model T', 'modem', 'monastery', 'monitor', 'moped', + 'mortar', 'square academic cap', 'mosque', 'mosquito net', 'scooter', 'mountain bike', 'tent', 'computer mouse', + 'mousetrap', 'moving van', 'muzzle', 'nail', 'neck brace', 'necklace', 'nipple', 'notebook computer', 'obelisk', + 'oboe', 'ocarina', 'odometer', 'oil filter', 'organ', 'oscilloscope', 'overskirt', 'bullock cart', + 'oxygen mask', 'packet', 'paddle', 'paddle wheel', 'padlock', 'paintbrush', 'pajamas', 'palace', 'pan flute', + 'paper towel', 'parachute', 'parallel bars', 'park bench', 'parking meter', 'passenger car', 'patio', + 'payphone', 'pedestal', 'pencil case', 'pencil sharpener', 'perfume', 'Petri dish', 'photocopier', 'plectrum', + 'Pickelhaube', 'picket fence', 'pickup truck', 'pier', 'piggy bank', 'pill bottle', 'pillow', 'ping-pong ball', + 'pinwheel', 'pirate ship', 'pitcher', 'hand plane', 'planetarium', 'plastic bag', 'plate rack', 'plow', + 'plunger', 'Polaroid camera', 'pole', 'police van', 'poncho', 'billiard table', 'soda bottle', 'pot', + "potter's wheel", 'power drill', 'prayer rug', 'printer', 'prison', 'projectile', 'projector', 'hockey puck', + 'punching bag', 'purse', 'quill', 'quilt', 'race car', 'racket', 'radiator', 'radio', 'radio telescope', + 'rain barrel', 'recreational vehicle', 'reel', 'reflex camera', 'refrigerator', 'remote control', 'restaurant', + 'revolver', 'rifle', 'rocking chair', 'rotisserie', 'eraser', 'rugby ball', 'ruler', 'running shoe', 'safe', + 'safety pin', 'salt shaker', 'sandal', 'sarong', 'saxophone', 'scabbard', 'weighing scale', 'school bus', + 'schooner', 'scoreboard', 'CRT screen', 'screw', 'screwdriver', 'seat belt', 'sewing machine', 'shield', + 'shoe store', 'shoji', 'shopping basket', 'shopping cart', 'shovel', 'shower cap', 'shower curtain', 'ski', + 'ski mask', 'sleeping bag', 'slide rule', 'sliding door', 'slot machine', 'snorkel', 'snowmobile', 'snowplow', + 'soap dispenser', 'soccer ball', 'sock', 'solar thermal collector', 'sombrero', 'soup bowl', 'space bar', + 'space heater', 'space shuttle', 'spatula', 'motorboat', 'spider web', 'spindle', 'sports car', 'spotlight', + 'stage', 'steam locomotive', 'through arch bridge', 'steel drum', 'stethoscope', 'scarf', 'stone wall', + 'stopwatch', 'stove', 'strainer', 'tram', 'stretcher', 'couch', 'stupa', 'submarine', 'suit', 'sundial', + 'sunglass', 'sunglasses', 'sunscreen', 'suspension bridge', 'mop', 'sweatshirt', 'swimsuit', 'swing', 'switch', + 'syringe', 'table lamp', 'tank', 'tape player', 'teapot', 'teddy bear', 'television', 'tennis ball', + 'thatched roof', 'front curtain', 'thimble', 'threshing machine', 'throne', 'tile roof', 'toaster', + 'tobacco shop', 'toilet seat', 'torch', 'totem pole', 'tow truck', 'toy store', 'tractor', 'semi-trailer truck', + 'tray', 'trench coat', 'tricycle', 'trimaran', 'tripod', 'triumphal arch', 'trolleybus', 'trombone', 'tub', + 'turnstile', 'typewriter keyboard', 'umbrella', 'unicycle', 'upright piano', 'vacuum cleaner', 'vase', 'vault', + 'velvet', 'vending machine', 'vestment', 'viaduct', 'violin', 'volleyball', 'waffle iron', 'wall clock', + 'wallet', 'wardrobe', 'military aircraft', 'sink', 'washing machine', 'water bottle', 'water jug', + 'water tower', 'whiskey jug', 'whistle', 'wig', 'window screen', 'window shade', 'Windsor tie', 'wine bottle', + 'wing', 'wok', 'wooden spoon', 'wool', 'split-rail fence', 'shipwreck', 'yawl', 'yurt', 'website', 'comic book', + 'crossword', 'traffic sign', 'traffic light', 'dust jacket', 'menu', 'plate', 'guacamole', 'consomme', + 'hot pot', 'trifle', 'ice cream', 'ice pop', 'baguette', 'bagel', 'pretzel', 'cheeseburger', 'hot dog', + 'mashed potato', 'cabbage', 'broccoli', 'cauliflower', 'zucchini', 'spaghetti squash', 'acorn squash', + 'butternut squash', 'cucumber', 'artichoke', 'bell pepper', 'cardoon', 'mushroom', 'Granny Smith', 'strawberry', + 'orange', 'lemon', 'fig', 'pineapple', 'banana', 'jackfruit', 'custard apple', 'pomegranate', 'hay', + 'carbonara', 'chocolate syrup', 'dough', 'meatloaf', 'pizza', 'pot pie', 'burrito', 'red wine', 'espresso', + 'cup', 'eggnog', 'alp', 'bubble', 'cliff', 'coral reef', 'geyser', 'lakeshore', 'promontory', 'shoal', + 'seashore', 'valley', 'volcano', 'baseball player', 'bridegroom', 'scuba diver', 'rapeseed', 'daisy', + "yellow lady's slipper", 'corn', 'acorn', 'rose hip', 'horse chestnut seed', 'coral fungus', 'agaric', + 'gyromitra', 'stinkhorn mushroom', 'earth star', 'hen-of-the-woods', 'bolete', 'ear', + 'toilet paper'] # class names + +# Download script/URL (optional) +download: data/scripts/get_imagenet.sh diff --git a/application/yolov5_example/data/Objects365.yaml b/application/yolov5_example/data/Objects365.yaml new file mode 100644 index 00000000..4cc94753 --- /dev/null +++ b/application/yolov5_example/data/Objects365.yaml @@ -0,0 +1,114 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Objects365 dataset https://www.objects365.org/ by Megvii +# Example usage: python train.py --data Objects365.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 Objects365 鈫 downloads here (712 GB = 367G data + 345G zips) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/Objects365 # dataset root dir +train: images/train # train images (relative to 'path') 1742289 images +val: images/val # val images (relative to 'path') 80000 images +test: # test images (optional) + +# Classes +nc: 365 # number of classes +names: ['Person', 'Sneakers', 'Chair', 'Other Shoes', 'Hat', 'Car', 'Lamp', 'Glasses', 'Bottle', 'Desk', 'Cup', + 'Street Lights', 'Cabinet/shelf', 'Handbag/Satchel', 'Bracelet', 'Plate', 'Picture/Frame', 'Helmet', 'Book', + 'Gloves', 'Storage box', 'Boat', 'Leather Shoes', 'Flower', 'Bench', 'Potted Plant', 'Bowl/Basin', 'Flag', + 'Pillow', 'Boots', 'Vase', 'Microphone', 'Necklace', 'Ring', 'SUV', 'Wine Glass', 'Belt', 'Monitor/TV', + 'Backpack', 'Umbrella', 'Traffic Light', 'Speaker', 'Watch', 'Tie', 'Trash bin Can', 'Slippers', 'Bicycle', + 'Stool', 'Barrel/bucket', 'Van', 'Couch', 'Sandals', 'Basket', 'Drum', 'Pen/Pencil', 'Bus', 'Wild Bird', + 'High Heels', 'Motorcycle', 'Guitar', 'Carpet', 'Cell Phone', 'Bread', 'Camera', 'Canned', 'Truck', + 'Traffic cone', 'Cymbal', 'Lifesaver', 'Towel', 'Stuffed Toy', 'Candle', 'Sailboat', 'Laptop', 'Awning', + 'Bed', 'Faucet', 'Tent', 'Horse', 'Mirror', 'Power outlet', 'Sink', 'Apple', 'Air Conditioner', 'Knife', + 'Hockey Stick', 'Paddle', 'Pickup Truck', 'Fork', 'Traffic Sign', 'Balloon', 'Tripod', 'Dog', 'Spoon', 'Clock', + 'Pot', 'Cow', 'Cake', 'Dinning Table', 'Sheep', 'Hanger', 'Blackboard/Whiteboard', 'Napkin', 'Other Fish', + 'Orange/Tangerine', 'Toiletry', 'Keyboard', 'Tomato', 'Lantern', 'Machinery Vehicle', 'Fan', + 'Green Vegetables', 'Banana', 'Baseball Glove', 'Airplane', 'Mouse', 'Train', 'Pumpkin', 'Soccer', 'Skiboard', + 'Luggage', 'Nightstand', 'Tea pot', 'Telephone', 'Trolley', 'Head Phone', 'Sports Car', 'Stop Sign', + 'Dessert', 'Scooter', 'Stroller', 'Crane', 'Remote', 'Refrigerator', 'Oven', 'Lemon', 'Duck', 'Baseball Bat', + 'Surveillance Camera', 'Cat', 'Jug', 'Broccoli', 'Piano', 'Pizza', 'Elephant', 'Skateboard', 'Surfboard', + 'Gun', 'Skating and Skiing shoes', 'Gas stove', 'Donut', 'Bow Tie', 'Carrot', 'Toilet', 'Kite', 'Strawberry', + 'Other Balls', 'Shovel', 'Pepper', 'Computer Box', 'Toilet Paper', 'Cleaning Products', 'Chopsticks', + 'Microwave', 'Pigeon', 'Baseball', 'Cutting/chopping Board', 'Coffee Table', 'Side Table', 'Scissors', + 'Marker', 'Pie', 'Ladder', 'Snowboard', 'Cookies', 'Radiator', 'Fire Hydrant', 'Basketball', 'Zebra', 'Grape', + 'Giraffe', 'Potato', 'Sausage', 'Tricycle', 'Violin', 'Egg', 'Fire Extinguisher', 'Candy', 'Fire Truck', + 'Billiards', 'Converter', 'Bathtub', 'Wheelchair', 'Golf Club', 'Briefcase', 'Cucumber', 'Cigar/Cigarette', + 'Paint Brush', 'Pear', 'Heavy Truck', 'Hamburger', 'Extractor', 'Extension Cord', 'Tong', 'Tennis Racket', + 'Folder', 'American Football', 'earphone', 'Mask', 'Kettle', 'Tennis', 'Ship', 'Swing', 'Coffee Machine', + 'Slide', 'Carriage', 'Onion', 'Green beans', 'Projector', 'Frisbee', 'Washing Machine/Drying Machine', + 'Chicken', 'Printer', 'Watermelon', 'Saxophone', 'Tissue', 'Toothbrush', 'Ice cream', 'Hot-air balloon', + 'Cello', 'French Fries', 'Scale', 'Trophy', 'Cabbage', 'Hot dog', 'Blender', 'Peach', 'Rice', 'Wallet/Purse', + 'Volleyball', 'Deer', 'Goose', 'Tape', 'Tablet', 'Cosmetics', 'Trumpet', 'Pineapple', 'Golf Ball', + 'Ambulance', 'Parking meter', 'Mango', 'Key', 'Hurdle', 'Fishing Rod', 'Medal', 'Flute', 'Brush', 'Penguin', + 'Megaphone', 'Corn', 'Lettuce', 'Garlic', 'Swan', 'Helicopter', 'Green Onion', 'Sandwich', 'Nuts', + 'Speed Limit Sign', 'Induction Cooker', 'Broom', 'Trombone', 'Plum', 'Rickshaw', 'Goldfish', 'Kiwi fruit', + 'Router/modem', 'Poker Card', 'Toaster', 'Shrimp', 'Sushi', 'Cheese', 'Notepaper', 'Cherry', 'Pliers', 'CD', + 'Pasta', 'Hammer', 'Cue', 'Avocado', 'Hamimelon', 'Flask', 'Mushroom', 'Screwdriver', 'Soap', 'Recorder', + 'Bear', 'Eggplant', 'Board Eraser', 'Coconut', 'Tape Measure/Ruler', 'Pig', 'Showerhead', 'Globe', 'Chips', + 'Steak', 'Crosswalk Sign', 'Stapler', 'Camel', 'Formula 1', 'Pomegranate', 'Dishwasher', 'Crab', + 'Hoverboard', 'Meat ball', 'Rice Cooker', 'Tuba', 'Calculator', 'Papaya', 'Antelope', 'Parrot', 'Seal', + 'Butterfly', 'Dumbbell', 'Donkey', 'Lion', 'Urinal', 'Dolphin', 'Electric Drill', 'Hair Dryer', 'Egg tart', + 'Jellyfish', 'Treadmill', 'Lighter', 'Grapefruit', 'Game board', 'Mop', 'Radish', 'Baozi', 'Target', 'French', + 'Spring Rolls', 'Monkey', 'Rabbit', 'Pencil Case', 'Yak', 'Red Cabbage', 'Binoculars', 'Asparagus', 'Barbell', + 'Scallop', 'Noddles', 'Comb', 'Dumpling', 'Oyster', 'Table Tennis paddle', 'Cosmetics Brush/Eyeliner Pencil', + 'Chainsaw', 'Eraser', 'Lobster', 'Durian', 'Okra', 'Lipstick', 'Cosmetics Mirror', 'Curling', 'Table Tennis'] + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + from tqdm import tqdm + + from utils.general import Path, check_requirements, download, np, xyxy2xywhn + + check_requirements(('pycocotools>=2.0',)) + from pycocotools.coco import COCO + + # Make Directories + dir = Path(yaml['path']) # dataset root dir + for p in 'images', 'labels': + (dir / p).mkdir(parents=True, exist_ok=True) + for q in 'train', 'val': + (dir / p / q).mkdir(parents=True, exist_ok=True) + + # Train, Val Splits + for split, patches in [('train', 50 + 1), ('val', 43 + 1)]: + print(f"Processing {split} in {patches} patches ...") + images, labels = dir / 'images' / split, dir / 'labels' / split + + # Download + url = f"https://dorc.ks3-cn-beijing.ksyun.com/data-set/2020Objects365%E6%95%B0%E6%8D%AE%E9%9B%86/{split}/" + if split == 'train': + download([f'{url}zhiyuan_objv2_{split}.tar.gz'], dir=dir, delete=False) # annotations json + download([f'{url}patch{i}.tar.gz' for i in range(patches)], dir=images, curl=True, delete=False, threads=8) + elif split == 'val': + download([f'{url}zhiyuan_objv2_{split}.json'], dir=dir, delete=False) # annotations json + download([f'{url}images/v1/patch{i}.tar.gz' for i in range(15 + 1)], dir=images, curl=True, delete=False, threads=8) + download([f'{url}images/v2/patch{i}.tar.gz' for i in range(16, patches)], dir=images, curl=True, delete=False, threads=8) + + # Move + for f in tqdm(images.rglob('*.jpg'), desc=f'Moving {split} images'): + f.rename(images / f.name) # move to /images/{split} + + # Labels + coco = COCO(dir / f'zhiyuan_objv2_{split}.json') + names = [x["name"] for x in coco.loadCats(coco.getCatIds())] + for cid, cat in enumerate(names): + catIds = coco.getCatIds(catNms=[cat]) + imgIds = coco.getImgIds(catIds=catIds) + for im in tqdm(coco.loadImgs(imgIds), desc=f'Class {cid + 1}/{len(names)} {cat}'): + width, height = im["width"], im["height"] + path = Path(im["file_name"]) # image filename + try: + with open(labels / path.with_suffix('.txt').name, 'a') as file: + annIds = coco.getAnnIds(imgIds=im["id"], catIds=catIds, iscrowd=None) + for a in coco.loadAnns(annIds): + x, y, w, h = a['bbox'] # bounding box in xywh (xy top-left corner) + xyxy = np.array([x, y, x + w, y + h])[None] # pixels(1,4) + x, y, w, h = xyxy2xywhn(xyxy, w=width, h=height, clip=True)[0] # normalized and clipped + file.write(f"{cid} {x:.5f} {y:.5f} {w:.5f} {h:.5f}\n") + except Exception as e: + print(e) diff --git a/application/yolov5_example/data/SKU-110K.yaml b/application/yolov5_example/data/SKU-110K.yaml new file mode 100644 index 00000000..2acf34d1 --- /dev/null +++ b/application/yolov5_example/data/SKU-110K.yaml @@ -0,0 +1,53 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# SKU-110K retail items dataset https://github.com/eg4000/SKU110K_CVPR19 by Trax Retail +# Example usage: python train.py --data SKU-110K.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 SKU-110K 鈫 downloads here (13.6 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/SKU-110K # dataset root dir +train: train.txt # train images (relative to 'path') 8219 images +val: val.txt # val images (relative to 'path') 588 images +test: test.txt # test images (optional) 2936 images + +# Classes +nc: 1 # number of classes +names: ['object'] # class names + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + import shutil + from tqdm import tqdm + from utils.general import np, pd, Path, download, xyxy2xywh + + + # Download + dir = Path(yaml['path']) # dataset root dir + parent = Path(dir.parent) # download dir + urls = ['http://trax-geometry.s3.amazonaws.com/cvpr_challenge/SKU110K_fixed.tar.gz'] + download(urls, dir=parent, delete=False) + + # Rename directories + if dir.exists(): + shutil.rmtree(dir) + (parent / 'SKU110K_fixed').rename(dir) # rename dir + (dir / 'labels').mkdir(parents=True, exist_ok=True) # create labels dir + + # Convert labels + names = 'image', 'x1', 'y1', 'x2', 'y2', 'class', 'image_width', 'image_height' # column names + for d in 'annotations_train.csv', 'annotations_val.csv', 'annotations_test.csv': + x = pd.read_csv(dir / 'annotations' / d, names=names).values # annotations + images, unique_images = x[:, 0], np.unique(x[:, 0]) + with open((dir / d).with_suffix('.txt').__str__().replace('annotations_', ''), 'w') as f: + f.writelines(f'./images/{s}\n' for s in unique_images) + for im in tqdm(unique_images, desc=f'Converting {dir / d}'): + cls = 0 # single-class dataset + with open((dir / 'labels' / im).with_suffix('.txt'), 'a') as f: + for r in x[images == im]: + w, h = r[6], r[7] # image width, height + xywh = xyxy2xywh(np.array([[r[1] / w, r[2] / h, r[3] / w, r[4] / h]]))[0] # instance + f.write(f"{cls} {xywh[0]:.5f} {xywh[1]:.5f} {xywh[2]:.5f} {xywh[3]:.5f}\n") # write label diff --git a/application/yolov5_example/data/VOC.yaml b/application/yolov5_example/data/VOC.yaml new file mode 100644 index 00000000..636ddc42 --- /dev/null +++ b/application/yolov5_example/data/VOC.yaml @@ -0,0 +1,81 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# PASCAL VOC dataset http://host.robots.ox.ac.uk/pascal/VOC by University of Oxford +# Example usage: python train.py --data VOC.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 VOC 鈫 downloads here (2.8 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/VOC +train: # train images (relative to 'path') 16551 images + - images/train2012 + - images/train2007 + - images/val2012 + - images/val2007 +val: # val images (relative to 'path') 4952 images + - images/test2007 +test: # test images (optional) + - images/test2007 + +# Classes +nc: 20 # number of classes +names: ['aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', + 'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor'] # class names + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + import xml.etree.ElementTree as ET + + from tqdm import tqdm + from utils.general import download, Path + + + def convert_label(path, lb_path, year, image_id): + def convert_box(size, box): + dw, dh = 1. / size[0], 1. / size[1] + x, y, w, h = (box[0] + box[1]) / 2.0 - 1, (box[2] + box[3]) / 2.0 - 1, box[1] - box[0], box[3] - box[2] + return x * dw, y * dh, w * dw, h * dh + + in_file = open(path / f'VOC{year}/Annotations/{image_id}.xml') + out_file = open(lb_path, 'w') + tree = ET.parse(in_file) + root = tree.getroot() + size = root.find('size') + w = int(size.find('width').text) + h = int(size.find('height').text) + + for obj in root.iter('object'): + cls = obj.find('name').text + if cls in yaml['names'] and not int(obj.find('difficult').text) == 1: + xmlbox = obj.find('bndbox') + bb = convert_box((w, h), [float(xmlbox.find(x).text) for x in ('xmin', 'xmax', 'ymin', 'ymax')]) + cls_id = yaml['names'].index(cls) # class id + out_file.write(" ".join([str(a) for a in (cls_id, *bb)]) + '\n') + + + # Download + dir = Path(yaml['path']) # dataset root dir + url = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/' + urls = [f'{url}VOCtrainval_06-Nov-2007.zip', # 446MB, 5012 images + f'{url}VOCtest_06-Nov-2007.zip', # 438MB, 4953 images + f'{url}VOCtrainval_11-May-2012.zip'] # 1.95GB, 17126 images + download(urls, dir=dir / 'images', delete=False, curl=True, threads=3) + + # Convert + path = dir / 'images/VOCdevkit' + for year, image_set in ('2012', 'train'), ('2012', 'val'), ('2007', 'train'), ('2007', 'val'), ('2007', 'test'): + imgs_path = dir / 'images' / f'{image_set}{year}' + lbs_path = dir / 'labels' / f'{image_set}{year}' + imgs_path.mkdir(exist_ok=True, parents=True) + lbs_path.mkdir(exist_ok=True, parents=True) + + with open(path / f'VOC{year}/ImageSets/Main/{image_set}.txt') as f: + image_ids = f.read().strip().split() + for id in tqdm(image_ids, desc=f'{image_set}{year}'): + f = path / f'VOC{year}/JPEGImages/{id}.jpg' # old img path + lb_path = (lbs_path / f.name).with_suffix('.txt') # new label path + f.rename(imgs_path / f.name) # move image + convert_label(path, lb_path, year, id) # convert labels to YOLO format diff --git a/application/yolov5_example/data/VisDrone.yaml b/application/yolov5_example/data/VisDrone.yaml new file mode 100644 index 00000000..10337b46 --- /dev/null +++ b/application/yolov5_example/data/VisDrone.yaml @@ -0,0 +1,61 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# VisDrone2019-DET dataset https://github.com/VisDrone/VisDrone-Dataset by Tianjin University +# Example usage: python train.py --data VisDrone.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 VisDrone 鈫 downloads here (2.3 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/VisDrone # dataset root dir +train: VisDrone2019-DET-train/images # train images (relative to 'path') 6471 images +val: VisDrone2019-DET-val/images # val images (relative to 'path') 548 images +test: VisDrone2019-DET-test-dev/images # test images (optional) 1610 images + +# Classes +nc: 10 # number of classes +names: ['pedestrian', 'people', 'bicycle', 'car', 'van', 'truck', 'tricycle', 'awning-tricycle', 'bus', 'motor'] + + +# Download script/URL (optional) --------------------------------------------------------------------------------------- +download: | + from utils.general import download, os, Path + + def visdrone2yolo(dir): + from PIL import Image + from tqdm import tqdm + + def convert_box(size, box): + # Convert VisDrone box to YOLO xywh box + dw = 1. / size[0] + dh = 1. / size[1] + return (box[0] + box[2] / 2) * dw, (box[1] + box[3] / 2) * dh, box[2] * dw, box[3] * dh + + (dir / 'labels').mkdir(parents=True, exist_ok=True) # make labels directory + pbar = tqdm((dir / 'annotations').glob('*.txt'), desc=f'Converting {dir}') + for f in pbar: + img_size = Image.open((dir / 'images' / f.name).with_suffix('.jpg')).size + lines = [] + with open(f, 'r') as file: # read annotation.txt + for row in [x.split(',') for x in file.read().strip().splitlines()]: + if row[4] == '0': # VisDrone 'ignored regions' class 0 + continue + cls = int(row[5]) - 1 + box = convert_box(img_size, tuple(map(int, row[:4]))) + lines.append(f"{cls} {' '.join(f'{x:.6f}' for x in box)}\n") + with open(str(f).replace(os.sep + 'annotations' + os.sep, os.sep + 'labels' + os.sep), 'w') as fl: + fl.writelines(lines) # write label.txt + + + # Download + dir = Path(yaml['path']) # dataset root dir + urls = ['https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-train.zip', + 'https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-val.zip', + 'https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-test-dev.zip', + 'https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-test-challenge.zip'] + download(urls, dir=dir, curl=True, threads=4) + + # Convert + for d in 'VisDrone2019-DET-train', 'VisDrone2019-DET-val', 'VisDrone2019-DET-test-dev': + visdrone2yolo(dir / d) # convert VisDrone annotations to YOLO labels diff --git a/application/yolov5_example/data/coco.yaml b/application/yolov5_example/data/coco.yaml new file mode 100644 index 00000000..0c0c4ada --- /dev/null +++ b/application/yolov5_example/data/coco.yaml @@ -0,0 +1,45 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# COCO 2017 dataset http://cocodataset.org by Microsoft +# Example usage: python train.py --data coco.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 coco 鈫 downloads here (20.1 GB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/coco # dataset root dir +train: train2017.txt # train images (relative to 'path') 118287 images +val: val2017.txt # val images (relative to 'path') 5000 images +test: test-dev2017.txt # 20288 of 40670 images, submit to https://competitions.codalab.org/competitions/20794 + +# Classes +nc: 80 # number of classes +names: ['person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light', + 'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow', + 'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie', 'suitcase', 'frisbee', + 'skis', 'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard', + 'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple', + 'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', + 'potted plant', 'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone', + 'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear', + 'hair drier', 'toothbrush'] # class names + + +# Download script/URL (optional) +download: | + from utils.general import download, Path + + + # Download labels + segments = False # segment or box labels + dir = Path(yaml['path']) # dataset root dir + url = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/' + urls = [url + ('coco2017labels-segments.zip' if segments else 'coco2017labels.zip')] # labels + download(urls, dir=dir.parent) + + # Download data + urls = ['http://images.cocodataset.org/zips/train2017.zip', # 19G, 118k images + 'http://images.cocodataset.org/zips/val2017.zip', # 1G, 5k images + 'http://images.cocodataset.org/zips/test2017.zip'] # 7G, 41k images (optional) + download(urls, dir=dir / 'images', threads=3) diff --git a/application/yolov5_example/data/coco128.yaml b/application/yolov5_example/data/coco128.yaml new file mode 100644 index 00000000..2517d207 --- /dev/null +++ b/application/yolov5_example/data/coco128.yaml @@ -0,0 +1,30 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# COCO128 dataset https://www.kaggle.com/ultralytics/coco128 (first 128 images from COCO train2017) by Ultralytics +# Example usage: python train.py --data coco128.yaml +# parent +# 鈹溾攢鈹 yolov5 +# 鈹斺攢鈹 datasets +# 鈹斺攢鈹 coco128 鈫 downloads here (7 MB) + + +# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..] +path: ../datasets/coco128 # dataset root dir +train: images/train2017 # train images (relative to 'path') 128 images +val: images/train2017 # val images (relative to 'path') 128 images +test: # test images (optional) + +# Classes +nc: 80 # number of classes +names: ['person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light', + 'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow', + 'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie', 'suitcase', 'frisbee', + 'skis', 'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard', + 'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple', + 'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', + 'potted plant', 'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone', + 'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear', + 'hair drier', 'toothbrush'] # class names + + +# Download script/URL (optional) +download: https://ultralytics.com/assets/coco128.zip diff --git a/application/yolov5_example/data/hyps/hyp.Objects365.yaml b/application/yolov5_example/data/hyps/hyp.Objects365.yaml new file mode 100644 index 00000000..74971740 --- /dev/null +++ b/application/yolov5_example/data/hyps/hyp.Objects365.yaml @@ -0,0 +1,34 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Hyperparameters for Objects365 training +# python train.py --weights yolov5m.pt --data Objects365.yaml --evolve +# See Hyperparameter Evolution tutorial for details https://github.com/ultralytics/yolov5#tutorials + +lr0: 0.00258 +lrf: 0.17 +momentum: 0.779 +weight_decay: 0.00058 +warmup_epochs: 1.33 +warmup_momentum: 0.86 +warmup_bias_lr: 0.0711 +box: 0.0539 +cls: 0.299 +cls_pw: 0.825 +obj: 0.632 +obj_pw: 1.0 +iou_t: 0.2 +anchor_t: 3.44 +anchors: 3.2 +fl_gamma: 0.0 +hsv_h: 0.0188 +hsv_s: 0.704 +hsv_v: 0.36 +degrees: 0.0 +translate: 0.0902 +scale: 0.491 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +mosaic: 1.0 +mixup: 0.0 +copy_paste: 0.0 diff --git a/application/yolov5_example/data/hyps/hyp.VOC.yaml b/application/yolov5_example/data/hyps/hyp.VOC.yaml new file mode 100644 index 00000000..0aa4e7d9 --- /dev/null +++ b/application/yolov5_example/data/hyps/hyp.VOC.yaml @@ -0,0 +1,40 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Hyperparameters for VOC training +# python train.py --batch 128 --weights yolov5m6.pt --data VOC.yaml --epochs 50 --img 512 --hyp hyp.scratch-med.yaml --evolve +# See Hyperparameter Evolution tutorial for details https://github.com/ultralytics/yolov5#tutorials + +# YOLOv5 Hyperparameter Evolution Results +# Best generation: 467 +# Last generation: 996 +# metrics/precision, metrics/recall, metrics/mAP_0.5, metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss +# 0.87729, 0.85125, 0.91286, 0.72664, 0.0076739, 0.0042529, 0.0013865 + +lr0: 0.00334 +lrf: 0.15135 +momentum: 0.74832 +weight_decay: 0.00025 +warmup_epochs: 3.3835 +warmup_momentum: 0.59462 +warmup_bias_lr: 0.18657 +box: 0.02 +cls: 0.21638 +cls_pw: 0.5 +obj: 0.51728 +obj_pw: 0.67198 +iou_t: 0.2 +anchor_t: 3.3744 +fl_gamma: 0.0 +hsv_h: 0.01041 +hsv_s: 0.54703 +hsv_v: 0.27739 +degrees: 0.0 +translate: 0.04591 +scale: 0.75544 +shear: 0.0 +perspective: 0.0 +flipud: 0.0 +fliplr: 0.5 +mosaic: 0.85834 +mixup: 0.04266 +copy_paste: 0.0 +anchors: 3.412 diff --git a/application/yolov5_example/data/hyps/hyp.scratch-high.yaml b/application/yolov5_example/data/hyps/hyp.scratch-high.yaml new file mode 100644 index 00000000..123cc840 --- /dev/null +++ b/application/yolov5_example/data/hyps/hyp.scratch-high.yaml @@ -0,0 +1,34 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Hyperparameters for high-augmentation COCO training from scratch +# python train.py --batch 32 --cfg yolov5m6.yaml --weights '' --data coco.yaml --img 1280 --epochs 300 +# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials + +lr0: 0.01 # initial learning rate (SGD=1E-2, Adam=1E-3) +lrf: 0.1 # final OneCycleLR learning rate (lr0 * lrf) +momentum: 0.937 # SGD momentum/Adam beta1 +weight_decay: 0.0005 # optimizer weight decay 5e-4 +warmup_epochs: 3.0 # warmup epochs (fractions ok) +warmup_momentum: 0.8 # warmup initial momentum +warmup_bias_lr: 0.1 # warmup initial bias lr +box: 0.05 # box loss gain +cls: 0.3 # cls loss gain +cls_pw: 1.0 # cls BCELoss positive_weight +obj: 0.7 # obj loss gain (scale with pixels) +obj_pw: 1.0 # obj BCELoss positive_weight +iou_t: 0.20 # IoU training threshold +anchor_t: 4.0 # anchor-multiple threshold +# anchors: 3 # anchors per output layer (0 to ignore) +fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5) +hsv_h: 0.015 # image HSV-Hue augmentation (fraction) +hsv_s: 0.7 # image HSV-Saturation augmentation (fraction) +hsv_v: 0.4 # image HSV-Value augmentation (fraction) +degrees: 0.0 # image rotation (+/- deg) +translate: 0.1 # image translation (+/- fraction) +scale: 0.9 # image scale (+/- gain) +shear: 0.0 # image shear (+/- deg) +perspective: 0.0 # image perspective (+/- fraction), range 0-0.001 +flipud: 0.0 # image flip up-down (probability) +fliplr: 0.5 # image flip left-right (probability) +mosaic: 1.0 # image mosaic (probability) +mixup: 0.1 # image mixup (probability) +copy_paste: 0.1 # segment copy-paste (probability) diff --git a/application/yolov5_example/data/hyps/hyp.scratch-low.yaml b/application/yolov5_example/data/hyps/hyp.scratch-low.yaml new file mode 100644 index 00000000..b77bf312 --- /dev/null +++ b/application/yolov5_example/data/hyps/hyp.scratch-low.yaml @@ -0,0 +1,34 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Hyperparameters for low-augmentation COCO training from scratch +# python train.py --batch 64 --cfg yolov5n6.yaml --weights '' --data coco.yaml --img 640 --epochs 300 --linear +# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials + +lr0: 0.001 # initial learning rate (SGD=1E-2, Adam=1E-3) +lrf: 0.01 # final OneCycleLR learning rate (lr0 * lrf) +momentum: 0.937 # SGD momentum/Adam beta1 +weight_decay: 0.0005 # optimizer weight decay 5e-4 +warmup_epochs: 3.0 # warmup epochs (fractions ok) +warmup_momentum: 0.8 # warmup initial momentum +warmup_bias_lr: 0.1 # warmup initial bias lr +box: 0.05 # box loss gain +cls: 0.5 # cls loss gain +cls_pw: 1.0 # cls BCELoss positive_weight +obj: 1.0 # obj loss gain (scale with pixels) +obj_pw: 1.0 # obj BCELoss positive_weight +iou_t: 0.20 # IoU training threshold +anchor_t: 4.0 # anchor-multiple threshold +# anchors: 3 # anchors per output layer (0 to ignore) +fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5) +hsv_h: 0.015 # image HSV-Hue augmentation (fraction) +hsv_s: 0.7 # image HSV-Saturation augmentation (fraction) +hsv_v: 0.4 # image HSV-Value augmentation (fraction) +degrees: 0.0 # image rotation (+/- deg) +translate: 0.1 # image translation (+/- fraction) +scale: 0.5 # image scale (+/- gain) +shear: 0.0 # image shear (+/- deg) +perspective: 0.0 # image perspective (+/- fraction), range 0-0.001 +flipud: 0.0 # image flip up-down (probability) +fliplr: 0.5 # image flip left-right (probability) +mosaic: 1.0 # image mosaic (probability) +mixup: 0.0 # image mixup (probability) +copy_paste: 0.0 # segment copy-paste (probability) diff --git a/application/yolov5_example/data/hyps/hyp.scratch-med.yaml b/application/yolov5_example/data/hyps/hyp.scratch-med.yaml new file mode 100644 index 00000000..d6867d75 --- /dev/null +++ b/application/yolov5_example/data/hyps/hyp.scratch-med.yaml @@ -0,0 +1,34 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Hyperparameters for medium-augmentation COCO training from scratch +# python train.py --batch 32 --cfg yolov5m6.yaml --weights '' --data coco.yaml --img 1280 --epochs 300 +# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials + +lr0: 0.01 # initial learning rate (SGD=1E-2, Adam=1E-3) +lrf: 0.1 # final OneCycleLR learning rate (lr0 * lrf) +momentum: 0.937 # SGD momentum/Adam beta1 +weight_decay: 0.0005 # optimizer weight decay 5e-4 +warmup_epochs: 3.0 # warmup epochs (fractions ok) +warmup_momentum: 0.8 # warmup initial momentum +warmup_bias_lr: 0.1 # warmup initial bias lr +box: 0.05 # box loss gain +cls: 0.3 # cls loss gain +cls_pw: 1.0 # cls BCELoss positive_weight +obj: 0.7 # obj loss gain (scale with pixels) +obj_pw: 1.0 # obj BCELoss positive_weight +iou_t: 0.20 # IoU training threshold +anchor_t: 4.0 # anchor-multiple threshold +# anchors: 3 # anchors per output layer (0 to ignore) +fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5) +hsv_h: 0.015 # image HSV-Hue augmentation (fraction) +hsv_s: 0.7 # image HSV-Saturation augmentation (fraction) +hsv_v: 0.4 # image HSV-Value augmentation (fraction) +degrees: 0.0 # image rotation (+/- deg) +translate: 0.1 # image translation (+/- fraction) +scale: 0.9 # image scale (+/- gain) +shear: 0.0 # image shear (+/- deg) +perspective: 0.0 # image perspective (+/- fraction), range 0-0.001 +flipud: 0.0 # image flip up-down (probability) +fliplr: 0.5 # image flip left-right (probability) +mosaic: 1.0 # image mosaic (probability) +mixup: 0.1 # image mixup (probability) +copy_paste: 0.0 # segment copy-paste (probability) diff --git a/application/yolov5_example/data/images/bus.jpg b/application/yolov5_example/data/images/bus.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b43e311165c785f000eb7493ff8fb662d06a3f83 GIT binary patch literal 487438 zcmeFYbyODL*Ec*DA)V4KCDNT2AR-_jqKI^Ncc+4McY{cYlr%_p2}pN$ch@`USHHjY ztovT;S?_xPdw=KR%zS3gKKtywPs~1NhP&~*c>q&NTv8l>Kmgzc_yg`15Z%O_j12%l zMh2h<000p{g)1o8qRR#s;EfbnZ1OACDp zrMLGaQ2TpLexKRcnc0{*0f3#AjgOy|gP)a+f(<-!^K)I_N2wvPif4y5P7$TpJnjO9wb!b`Bsi z@?ZSvkIrnYte;^1cn_BT6YL)h$NL%opSr--$@s1ReNO-~Vg8}tW7z+|@c&@YvzZV6 zgUj%++XQl<@??Vd0>j10{8);zj_CSu`JjxAYeZx0Bc3MzYk_)P$STH zFwEd3CuqTayBYxitPbdvKRlpje{mTo`47Dm=uaPlBLk>@ z(Lo*HXaSl(JPj*>x&%g#S0XBdw$Q(o);tsq5Im`h= zpa8N0V+1gOifV#296_y87w1gHgu0Am0N!VJRz zNd)hE0#NX`4S5YC4ak7{c>oE37-$bOj23v`9Iysn1JVFL%o`YSKp5Z!`Po2zKL|BU z6l4kT1wMmbNCw`+JcXeHEJ4jEKpi=O9DoXD3vvnA0dO*AULEnA_ zGC;3>0y2RKz~X+^|CL~V^$ZaHIsYaDFsOfe%P=_I!NEU3KzwIlrfq2Z&OzJEUSHYH z*j(RMSKCbgoxYW&t`R2WAvmw8%rzYoVhbd8gChYE3!n5c6AQ24Z-KuioqKpcy8bf> z-`$-|0&x08cXv-H0ALjM?rxb8fVc^ROypn&^@If6pV@fWL^wFOxS04jIK`MaIoQRS zp7B0sW#Z!$eawYKDCYUMkaYhLMW+xz)0?AllzMq|b%_iXa zULSBy{ykg$AI}^L9c^7x3O##mGYVa8bA20a3T|_A3Ii7Q`{xdT0s0IaivY|k@BpHF zDd1dza1I`S(G~CMaDV8Cm>|0M@^4Q+_(caXA{{vA{8JW@71ZNbS?T+-Xn%QLfFAut zm%pcD{iVMK=aYZw1b^vD0O>C}3MgL-^y9t!`?}2Ti-U;{%=dcS&)MJ^^WGzRVE+BI z2Y$r`ARSrt7acil95jRt#P@UheYrp0d-VI7)fE7$Spl>|us%|}Kj(kvd5{Y(+ZO;H z8s9A-;~Cl6S@E;5SlBXa>sjjPGwWKKvp8v6v9K|-vH(zFCo9nF`gRmL`i909f|PqT z4U`nddV-XSTr#XOR-*bw#xGoK^xwL?lGk-H)#cNp6c(a@I`KQ1Tbb+IX;V0vn_1ZM zI|)+$63!3edol|p#V-~+Q$b24nb#DemNxnnT+E!ztdwAB8$APlx#!~llm)K@DgRN@ z(b19Fk%QUN#*l@LkB^Uqm7Rs1oeAV%vURqw({^IAu%-GV;W;=WY>ch!j4drF?j>sL zSlZhOQi3)AR>|Dzuh{?5_CJhPMwWJ#wnmo!3;W;OzZ`a7$ckUoMqk@b|2dciDev9R z#LB_M$|nCOAYukXhu@4cmU_kp&i^ape@L^mk+-xo6QukdXHio8t(O13^zRD6R{7s* zV*%aD@{eEd;ok-lqWEw6uLu6?f&Y5ozaIFn2mb4U|NnX5-w;pV0*oLW!2k}p+X0-w zt)Aa;4+RG^E5Q3qQU>lmE(HK#5{L=_OH3m{0>-*901OF?Fz|qlgO!_ug9JthK)jDZ zSUFhPNMLyW^8)&Z>9;JJU$QXK-j+vFJh&RsMwth^gcarQ7GXFgMMX7V%gKsMz7+el zQ3gY;4embOS8i@$XCo)^j6z9Sg#uv#jMNdps2mTV(AKrJ5_$dl<-N>5=U?*w9!^Jp zMasYs)4i;2q8Kk4PfQu!goh&oo48(o+5TsMtOv%D;H_&g>+4$C*nwg9eSA*kXlHeg z{XvXl11bgLx002=-D`y*HLnAwi-#a>z7P`!g6kvn$0KmP^f7g5T z=9e!act8H7dCviWV8h+rUCqBVsRjTj21_DW|4XCe1s6=w(O4~;Jm%RH`4vc{T z037Cki!5buUGnpOtz&Dz!T#$$1oauXD4e~!JE8^Gb5Q_r6MlDh_4)4ZCIeg|O#?uq z#a$zSg$TzBOAia71Yocruvn10*82r31RnA${Z<^j{%7@!gp2|*l!MDS2rMiN94tKi zedmU_-)|K+^#9vK}QpO{=&Tv}dPU0dJS zJUBc$J~=%*zqq`&3j)CYHVZudw(S3~3*7F4z`()5!Xez-1%YwAHyjHN{t?>)Y>_tz z+SU&#**_rSJc~##ZbG8sklV-Au^B+dqvl+oIk-3Nmu3HVhI#$Jvh1&6|JpSLh=Qa2 zKEZ&0urM%SQ^A4>4m1Hw@DJdBr3e2cgkK5qJ|X>0cc7485@2B9z(1r1@DKjp`@aTv zGvGD>$K51=3JY#2V8LR6TTVAP`&kQR)8Q3ubM3sHf~F(dHL6v}Y8|*gG-ncUJ7ld# zYk6cI(K^1J^FFy$%r#!?32|^}>QK3r{}HoddzQ!`T0Y2lTt-~5z+JV(JqQ)Oy?9%B zoTq5@K#x1FM2{?r{)I3rZi7mezT;|Boubi9Wnd&_FSk^b@7_-vf8EC5b`9ye7Ld;+RKVL$i%u0L zGFFFGA!Iip;*_HaoNIxb!cyQPdDPCzN=|rVs-XHfk^@tHi7ffyT&nRQE%EjkuhYRm zH2d1RQ-0C8#4%l3Z)G`N{}wFgDfRJeWG|=SNU{ahe z@m-vHUumAK3K`)jM|}z!M%-^~Po(ule}dX@tGtp$V@ux7{aHIhg8AcUsz+K#o&7}` zC*|;0t=*~NhCE+evtcd10#@>Rj<5uUl6F@(6<9f!?F7_hkh3DWQM(Mml6)2~g76n7 zxQ*FKoZ5-=dUlG*Y@xP?$-Zr5DfaIl${=lW!TpReOe^#?)j!_ zXmyZf)ime{x(ab-=zhF-}W0Atv$} z`pNm>MZL(+gho!*gpb4NiS^mY5(w4f^ECHfE1djXNY#`Rpqk~>M44JY9T18Th))#r zQF_TNUt{xdjiuWE+O&oY_h&h;PyJUE=X_l6uXO$O_@Sm9WnungPn>+hf}q5FKC+Vr zvKSgFvmy_DC^{n^n%l9Z<`G(&jk#vHNe`6s<=go)+aqSh$IG7jL`Picd9NfSx<{I8 zAu^+AE%9z?>Uv$)iP;TFL?}EA8d2I{@xWLY!mn2qM%{8GQXlzY$rE8rMG#0k$67{x z@m&ACKwR=zlqor~#_X!y4 zzs3IXmifI_P0XACt_HN}4VTy`zAL7keE-r|=9hsoMH7$h*YVV6wU_PstXMqtFpy13 zBuwuBB&Kn{43)CRdUUI=Ty)CA**4sLEtOI?p$B@O1&K~KyILLb)(Lo#+nhToSg9h~ zv^P^Bx5NH?u@i%0!~q}udbF0cPqB6a54ZFORhVAI>ynl^2*>YD3g04a9Wd2C%r*Kx zTY8g|DM~-n=59eGFlGeHP9?gcrpiX|;1--7@7qYCxH1#At$mPJ25QxbgPbdB)c7EQZ6o+sI%mYQf}FLtb2wKG&_ zBWr|1Sg#f5)saR@I>_9Uu?(?d@vQXcj+?#FhT zYeBJIWPw*Qov)yX8LsdHSWpdl2iOE7cY`4y->aNMyOD*lZ&?|1Xh9-C#D91yv zraSem6YdyyyJ^2ZXYaq5UF@QO+V@U5HG^RpP)Q$<07!Kq-#NrW?}vuF^GrJT-$p0ZO8{@wm0& zoCdwpG`$zY_gZFA2Ay=kU!Ex~@OT5=x_ok2dGm)uNKc8T#y=m^4tTZH&>As12tBsSf%BKEY*S;gy zo@9PLwDGuz<=hXWZEs?A+o+y%2&%fjJ2(@|+wZTN!g=^3Jz9c7d0Pn=%4ky|@h03_ z9V)q`>F;9ZXsDoxwbA?>4L4@_iNylq%t{ALeao%{$+BZnc23H$OAV(pE-R+RWJYG% zf`gyktVJYsdBlsu`tEc2tz`P^>SJnbw$Nm9k->Vh^R$+VCp0+|x)x z%~9T*W%jxQME3Rjf`lg~@vD6=a{L&3VwF~Oy66ZYUXc%+;l>>Idut_}m#UbyY0a64 z9w4$r&px$yYdl5S{qB@>+T31CT!fE7W5aR2KPb^R)u3UJqJH+nk61;vGH=o!GG+TN z^Os|B%Ck;_tLFT>w(!@e(1j|zU~liLU18R<#mZ28G^jw!;FpSu!qxp=YphnrSCsgp zw;MZ`Ypt?7GGu!#9&CPBRByZxLN~efsSGgo3Z|_NSPA6-dJlznT2IgW#-uAF%*#?> z^s7e}g1ZeMo*t1AYvuxBdofRW-RxnZ%dWHy3k5u?RbH3G7lo;hzXy`mUA`wR}gb`{hG{=`cw-e2GlRxP9Z;vxaFLnOXo0A!?I<2mu zE`B7VD<9_d~^{tef~he11e)Fo~|05sKZh^fxKPW zATsisw7pwk>RRfYIa0}#RhFslaHf~Pu52KK)z?*7%|FFp(qoU}AY7!VU_paD=R^Pg z@Mg!fV@<`Rw7cC@Wmf;^`99K{;RB1$XZ)I;qWR|tS8s0n>PgR|_B|IqIwtyE)9@R% zF626W&;$De6Vse%?Vwq!?NU!ds`NVHAiFvbX(RmFWx6SA?x}Hnp53~2#GYo`+_psS z#%0b<@4jiXw|Mc;?s1C4!3j>qhXrNY>dM4z#(~~?S?)=t>G!*>352&I<@i_31D48K zIKht8}?gfaIHmELs@0pS{X zcLTIs(AaaKx)y=tWd!-T1XQU;Pb>2x(%8x41pxuy(go3#XPe88|6rlH+*2Bc_n&sm zPL`+!8!#G-?*Lpe>#L_umaU1xmqIzZ3p)mM7N6;9Lj?*%WkQ;(E~3k{8=DZ|mU4?~ z<`d|7#}y;b<S!9@-eApp`w=bRoa>?TwfJ1K*AG&cDc@EtuHeXgO-MT1qE>>rbpT?8#>I~GVBBNTh| zx3i;UXbUD!FU)?R)IHmhN^5D*llkamGaSmZZ)7kg5IV#ol;n+z16Ye>PQClPR z#;Kv;d!s%`2i%gNaC2=-N8E=o(Eu8(??$H|9O>XAYTF5SWufSE{jQ7cRR<`-NYuny zm4n+ADHl+Ym`(s)6gG9XZfh$~{%xS7iHBoohwF=1EHa^C@@h^#mDDG#+g_X-`|B_~d2K@nE%zV)~O-#Yledh-lF;>3-nK(fn|?QCvGvCt^ivzR6|&)Si+ zJZ57*)B4BGq0Z8qV!-08*QsOY^JH3p5P>nPoAA&s%SzROV8-~f_yGO4(3fyBRImdm zxhUUANEU16&gl$4GG1$6;bS28TA5?dz1g44YTrN#yWvK3)F; zYi-5AqbX)U%GxrkKS_w;Ty+(T*Plm5XO<+w--Ao@ zP2O6EC~#ZG)Acc&WGtQDyggmJNuVjI2zAacsui`R9^!buVZxcl6s?8bVVmJ>2t-7A z!MjmiIXitnJX&nn@S&L#1GkH_q=+s z09GE1IpeRf_CxYFxOOUh7I$PK2wl{a;O=OwuI!aDe0hchNRzIS8%;Rm<4_I6ZhABlq0J1X#sN`{HZW-6vm zuv9Z~4c|8KP&20g)29=bkL)=l^MOTHM1<_q?-TBTEx9(E9_uoUn9ljj;V~Z@+o`~d zGh$Msj?~3Py5V*tG~dRvB9im+1K6fpjKOU;(WL^N{+3Y%6`v$Y$FFLlrl>}SPnEhj z=$A-k9A@d9R|m7N81o~UpVC`d2V0*jHs0dh#NGjhvOcp^mN?VrCLMCayB13UV18r^iR^Bs={c9IKM<;f*Uybp~`RhqcRPx4BMCKk-w zUQMh*59EpSphcOg{pPx{*Srvdr$42MrEf~vMRQ(I^{u{m94qrFMl|r2&W+f* zNXf*5pu(%~Ayd77F)U-l7M1g}YGMeiEj&Hax^2JI?XYFh!an&mmdpOS6-`!b#J-0u ze3>#<>x0A=6ZA%dhG2Cv+@LlgaatB!H3Z2~e@N(z(ufAf8c23Va zyTKUW`#49+rRkMcD%@Ph(~$}{lJ@d|Ho?(8+r))2!lZe=uC1-Tz4SfJvW>IJ&bQ~w zKD>{!T;o0p8(DgL;@T5(Ho%*X9&X|l(@z^$s^*^{68b*tF|;Y}C#3qM2L~>+>*1aJ z?f>({p*z78aDtI?JgSPTkhw_YV{kQzD<+0R)pY!eQvr|=BWARdA~&s$(( zD$Z5SY9C}_sVFI3&NGL5K%hN7=1!zyQ{REcAD+TRTaY8GYIX8;+T)9Xu(a> z!V^$YW!QS36iSL!&fmXlJ6XLx&|EPKec_mJ29*^8sF zaNR?$Bg1BAy{L}FsizXx?y6n_hGvn|%HRCm&QR!y+%DH`A7Y3vHUl=_yl03;yyb8qfrEDc zf_m;%4!?q4>|{uCsN8cp36kUMHm9|nhuV(r`=T1&FGOit5T%%Hs{$Y$vO@%d3}eXQ zz2V5A58O@LgpGr)LLQ%b9^hpdbH)y~noD(|;Mm_>t{H3ChVpLhscNg71}++1Sf0My zpDp+_t0wlPv6DI^z&uG2R;Q3V(i?h^c&WGu^VK*4fso83t(E8^?o|6cV(|{ZHyJd# zyctq|9=ko??W6bflbM`U@MUsx;H7IvG275*^>H*xFPEbTZ%xG>BaZ&ma=h%Filyvl zH{Hu&1<&;e?*QQi6_|-n-vUYCFJz_q53{!)2*to@T&CUqyg~VC*FZbp9WSqg)X&Ge zv!M}Xwpr|+@>J*!km?x3!_+Z+twYD|UZYbZt+cbm4n@VS+UsMupu$(L%)Ppyclhb< zt9vtLPOr*W&^e>Z_(`|mjW0>z@T1aRr4JfUcrcNZ*SOXyZE|FYtO>49{2U0syN<*+ zDnpj;fH9??0?_W=8i%M`7X^PqY@jySc1Kl0N9AYaDF>&ac86%hj#MLV_t71&nVFJ2 z^}XGy?G?vlis7oH1x++sBkIx1E@>IDkMzk)sWBc*d!flA??RMFr_pO23RsvZCJ(Sh znV)hJNVV7#Tk;PX>EfX~qQ`U?#~M?;ESyUELABMm`rTc5*$%lHxn)0phJOlQoWx$R zu4eSy*mZp(XFEZ3I~Rt{$uE{T*?KB8rbQ-Sh|4g z{bLy!Guoq1_L!ZKnf~3I&si-4_jHy!r#L;nmVXOMENNoc&DBC z<-pX37K#M9^Kr)9iaIklLHD(Nh-cr%O0+8l9{wZEf6Ppe!>_t%5Z6FK7r+ zrl?q)P>g*QB(nWjPpCHk68mWl1{~zAl(Y&j)C*BK3^&78-yTWWC?xy@2pg>4*d=?B zIEnf&JV(%3kb$5}a@}ZgbWLD^lo{66-m2v+)htJ7>J%IxJItK%7TSuPa6D=*V^8L% z0`QDuF6075c47@^>+tYHow$AG$re+Tw!Uo4wkr=2iTES(lDA#om@0{I)xMCrZnX_f z$c^BsZ*2~*yF?!`Pbm0|W+RF_pJoyxLv+OL9CDjx@4SS8&k+=q|51RtY%!z0q$UODLAAdM#V3e!SrAC=$#|%w+T<#J>Za1*-!OEz895 zA+)cTh6uE580cTrx4TX8PR^uJBpK`bh28=5#jjj!$_o!ZbapJXy|XR$s&2BZNwX+# z{@Rzc>RoI?ulo7pqa7@ zWU5*P1pV;yk-pk;krrG^2BsH>B z2Z`J*>4pBuQnhczzT^S*#KUGqTK!4(`lf;-!4VygULSN}OCGWVj&|1!5*=NJj)m$^ zBb>##k?N4-8X;q(&iyJ!S2uL|#;Acm-6Wwnx z75}_$UNb-iK0oo@VQ1x&$U=wC%!h}awd~bfUhvL$fIlZOLrr>a51Eu(W+*_)5!$#` zVDD3^e(PhaBgwN&e};>+_?pXbpdp++lQ6uag8#6=G{d4$x0J^GO7k#DY36Z$Pp zHuA4&m}-z1gXJQHjK&Wk(ESO`mBgbE%jf4|y6MDd0&S}TdURNY$2W%^!SwOFOe+ZO za%tG8{+LgDt7)WkuG%t(Ro+v;A~*P``Vfa8)>z|-e#j8*i&it-A*lUQkp_dH5bk>7 zgAB(k6=SxI0gv_PKB7`J^4dWw@ypBJH97UB!!4`0l+~ZM`}h{9T&WIN+Q|3CW&Owq zi=&jz6_-%Xbt`Q#!=8`kb450rLB97Zy?>F8ykwbHQ!i^7;PJ$>iQm&G40V0!N8NsQ zMHGct_2_ZW9e{2wQsYV(uA$7K2MOcU%_V5}%&T!nCw7RR+Zc_zwR(Lh4QC)e)!Fl& zH@26XiiH(t?Xwf9wqKX=3=Z` zU-hmW6C$(Nq35}w%8x_wUh2UhS!5JR-OhnK;%g#fC*!w|rq&bfN7g1LXAdWqVtaPj z1Q0G#hgmtOP&55j)xwG4#`@A8-oo7W3*(sU4V7cSZ>hxe^)WHWqKS8`@;;%O$seRQ zjDtk)TrYOD{NqLh9 zEyu5nR1(61;m4)AhX%Fm=%$$A%3n@fq^qgpTU(U%_uPMx7BtD#VxofJUv9j{ShCid zD7Cpk{}>{(8vWKJ)Jg?4OzPwtiG)vibh2krLW?G2-f2v~7Z(P;lLks#kZcvvU<_QB zF3FT)`m$6|L_fT;A41~ZuJ-W4PyAMn4WeMYD7whmQ&*(h-fxiv7j39cRKl$syYy#) z;OntT3^BHwjHXlmsqx7a#Y0!WYwYJb2D)bIs*ibjQsYQ(Tu#Z>yZeok> zu1()!uK3%Ek_z~rh^`US4pY9Md4_X*#Di|bissHt%x@D`&3Plsovrk}y`yxzO=~;x zG&l&ib6l(y?!hFI_7RKjUdM`(2g@*d8`%?e(YDR8(aPc=cPX6kDH&H(lV=aA{4j{y zYV0gBpyp7!WjBFEw1+L&3$f*4DIRIrL*Vqp#H8Ov7Z5gA}_<5>kJhR9yehS zsE8WwyNVz9w0|MUtvJ-F3kemHlV+6b*cGuS?SHv`Kp?5`&UCOWZc=j}eUf0hIfPJq zebg;1eIspOriO{7Ij!>Xha#w>D^8*tn+fKGx|EX6;VKnriR3B=ha_WUC(pz8ZEwWU zJT+YGn^mcQk|C;jP!QcJdFqC}_arat4@rG^p0X&5F}3c(9QCEm*8V2lDS)Fjy>zAd z>#mxf7I#xSN1w!Tjrn2^%D8?_iFxPyl-Vl6q8{kW)`iW*bSSjCZc4z;(Qv@30xzGg zYB(gN@9@bYBgqnPyC?EOtz=%-SB_Tk?RjSKad7euvRHlyop@{1F(&L|bxZv{aQEFX zBj4O%$~f$U2F{Pn{dp@D`3JJW8Vof9*GxiR2r@rd*P=t*Fp{=3KimOjEFH3N8)*cb zuC7opwj?b=f))YfoKD=$=YTci0DwwD%%`cH4 zo0WS*q0mwktL9_XV%16fp6eoM-)-}IPNVSobEtr>c}~+!Ax*ni=C115w7ER)mdxDu>%bs6*A%Ht+TPb-8i-wqF??gQvZ6~7h+SnvY}GDm zcs0VZBe5V1zb=e^&9^{~(Nf71^`a-~qYbQeRuw`Yn(Uan_Wz6mEvNq7Ap<&1$m06+m2;h-II1v|>a8!VxIH)glyB1S}An zKZ7uIkyzEfeTKIXP_4XzFWdqC}f1vCu$eHXgfwrN$<;lxoXt5Kl4c2il+|nvdKou(w z)Qk_boE^I}!*>10KhV0jxauaV_TxF3wvUeDC5}eWZ^n92nLDDei?=;!Zl|?wrs(QJ z0wN+AX4AzpJrL|yJ)A?j*5B1cmYSPqfve5Uj`(Z&{ked(1B2p46jLPH^0Lrgy?8D9 zVlQgBxmFRV9_$5j&%w?QMPXcf>X0DuIHRCQB{;FC8IewEc11bmVYufE@`x^wIA|%zXWr@9(lTTM`!`d4Vp%SKx8*b_-BDUIs)32JJh(zEv-z;m@ z8h$U;Jt|D*QYnefkmIiC@Jm}}&4Fj+uvvBVkiaTblNN4AN#G8dn~I3U!G(`paBi9j zX|)haSZqk&!`f81)ZF%4(2JdQ)=rWc)OrGE&atZa(Hyy&`Y4oN@#t(}2ICeRBS0lI z&s8{j^!udbx0b3Ze8Shx>YwIeUIiw z5X*;0z!b+;9rE1#L#(wAG@PNkk3(rd>WSJ}6~dF~icj<|i z_z@euZ}xud>~v8->eX4cn9`3&AD*UCtR5Z zv?ReCeVH+9bgb49nz*+Em5H@!1ofn*vWBR|CaLVPo2*$rJ;V$%0(vhnb~__EC@6 z53P+ZpYQOgeixiukoBG)^0b%>Ov4&9G(N#Va1#HSA^8MZ!TrtL3bSq#g!{d>@w5c< z#8v93E}f887s;~qg<@A4N;Vj%Oe!wgeYZR45ll^42vS}`M-f**>8dF4juEayh?rI! z-k=QWp5|saO750TcPcW$sTc}`ym)zI7PQ< zB_Q}?N^?-`RDwir=*ZG5<~23tXBRr;Tj(Qa8y91KZqlaaguY)YLfk07SEiWjKKX7C zhF1JC7?ZCZ-x`*;i_~<&Lf?EPI^6SeWTgQUW?D@dhKJQ&cP}`#WwH($F6XSRO1=*C9Ja}1h6;Crz3u*c=ZeZC8N4W6@u@D>TeA)2G&l|9KW|kRSL^ceb1$#<5vM!$ zJ?E(%bRGdE!?fYN?wi@8>NSY-H{cgEJAF268d+ptiw^F9cDxlm#ftsxuvk59@?PsN z-yjEN$!XvD2zs5<8k2&QZV9hphje1xz|R7PXJv`|FG)lnxul(>NVpvLNE-_;SUU|~ zg^P!LPH$c#FkR#rs=>%|d986;7si9MJMiZ4Mz*Pca!)?hGDzr=Ym&5Sl0D*}ZB zc6E`O^t%^JgpF7bw#XRI6KLv^V}FX-eld9{%UOuWt6hP{d4`<~P=>tr)p3c@$Y)sh z3_#j{{Iq1hFysKEg@2liak!Rap;u|kA?~2=OW;8Q8AAy7(`vSg&NS`2kV562 z1T%2~mSmC?qbY-;>=2{drK1HVv8>T@m>$%{NtBOA1uI8d(bVbE?Mp^ZGf}Y2)!aecu(a$AFV<*>k?mqs!W>BrT z9m~YhVnJ%rgKA?ZY!)1qiVxO$Iw*`Xjtpf3CEcD5`4QQp?KU!s-*4hQpx6AB-^;%b zvRE+NnI=hZSk&@A$UbPXA(&Ecu{Vo)9)UzDEWlo3KHX%UxZPcH;@>JB6i09A(a}5D zIF_Nn@n%(0uugVk3EWM+1Fmr}#A**3f}vDP=2RX027ZN`3JwmpEJH8fZWbHo(YNL( z4hBn_(+lZDq_)5Q!3)c^-E{}RKhlK1W*DgFx*bVF;|lY;MDQ=G<6kyOG+*)a-_}Zc zJ8|l2{HQqT>gCI}Mut<$Py3o&=PTfX;Nm8@mhIqd-XhELAg%S5zG^r<>0RK_OK zU<{8}>4=ZSbwx!r{X_3v*W2oEBF(Wyiu-%l*rlgW_x-ku7c7J3Fpj@yjk(F*0f?D# zZ#Bipa~_ms(#o`p zBTb-r>cT|{mtp(;r_DFvLi9?Fw@BwP3o}yAhLt5(?3xnri!*9fz62VBHc?bvk2?b! z^{citHE6@E&3yFbLLv@3k$DzUG3HG#63@>-qmFM?Lx&?$~qZo-qbW=M_^~P68{uxh2}9e zRU?>|WJ?}JrR#!Ctg9O{UCTYx{9t}nSsQ=1X#ysF0I!8_ zz17WE1K3ly1NIl`zE=eMtzY=6pYmydU%bwAO5)%+DLwsu#2&>o zn#%u$;7ZPAUnnO`Wv!)XR?Im=yv(q*p*S@I=MKQy<*Z&z6iCmJVN;G7dr~crsi&i~ zmJ7f(71coP1@99=4ZboN+qFJ)(C5H{1jI!x}%S)SB480F3*bFJ;r6;tO5LW4{ZC*V(ko@ce+& zP8?mJUYgLy;yEhR6-%AuqYX3|Z*XE;$Fc4wC}0w@tej+=RTg>t)FmPka*FWfv}hSB zHInx8&Gg3cja`hXZ@i?6B z5%D#XW)Q-mPY|w_@7TK3>rcPw%zwZ`8EL&MxaFltFP+D|SmyXj-pl#x+76cFdq0M8 zmv=$Lyy5u%J@~rCeZo``mkQy&Zfr-6J92Zpq`oeN3}SMfYjW!b2-Gp*=DfRXO?jBo zM809dbZW3$uY661EOCxhBEm-M2(Aj#<8CkIn!j>I zA<6BUv{Jn${v>!?&Nb!pJiaW4j!4JmHIFErdIZMTSV$Od-i;7`I8NPNv?&DgJ)|N(Y?2Z30>*rFC zf4gjjcuQJeL*I;@I*;@)_Rdn^;QDnBUWxU<&Ga=a`l}p`)h8@Ez6*scOhYS6! zf-VFZO{&g&;yDmf2ZC~%lxGKF;pwC}p$w_2o1xn*WQSWS+{a_b5@$`?2d9i5@{L~= z%oMD^dA!2E7*p_Ib`u!eK~E~E(6vByv?|T^_{2*7MzaIICGDK)@s999jO@9J2GRW% zX9hN~hvTTiSc5^%kX|1!PJW8>aGhbt6W+7QqQP*-#FC}`7V&`gBd#>`v_L<%ITNc< z$w1o6b*Za!Pi`^QZ(zY`2!Y4LRk-RlH0=E?ic=1Twz<)As!U4DbKBjoV1;PTedSlI z=PvdR`@4Be$+gi}@+j;_<0Z~!64h{HD`y{s=QGp3`L!C=4;UX1%l^!w&lFqaT^XR& z=`EvYy0*qQJC(C;S{Es*azi?x?{DMA+SEj?RW#32;e{t^zGR5z+c?ZshIA4m;Ff!6s%u!N-!@d(9ZOtT&d2$hA-mx`Bhx8P zJnHFzgBMHw3V`#_ef0L&tQ(niTcp=1eikFga9eNaXzcP@))b?a<&_>7BqzW8#EkAj zULgFsL7`l3&6=7;3vo|VD+u~tuPp+kQ#(wR6Z zWGGx#xRyF>2siUb(HY`WfeoIESZ}8Esq}>k;=X?-}_jnI#Mm7B6 zRoy86Hp=%et@CqFzSol9J; z5UynG-D-lLHm0z^7C(#+JJUl^8`&s{9+>10Nuy}vm7w|@NBCyz^`yEw{pu>-xD^j1 zBCUwBa-sUY$1>*9(S`<>-KcM|JBo{V%BkxI{$&nKVdTWVJ`QIc`CxB!*(F7Dg7QwP z-PE+c!u2;C>QnC<8k3y5HWH7h@&`s-sr<({FABMOHdK+Jr?5gT2tT9tP^-Wq$A_7LZhP;#r)$AaDr8;8Qvw?>Ma2d=_ zf(M-)eE8~ zIa>R?yQxSbfmy^OSg)-2q<*|#i6BOdk4kj?Dw&X$uJWJWu&I?Bnc5kBP*3qbo zimoF%n)U}6B%?7?FLa5m9M%Yd(PQhF^`e=s&nB;PgafFO2Cv!XQP8+og&(S$;B#M0 zeS3Z1x0sb;*B~sHjI*Fdw+6_<(i$sSYU9X=c1Gz22z zF!{%D#a31I;!JyHy{cSGN$EJWa8$~ryj{;su{P4e9b1M21RA5Mcxuw-Lvc34As7c3 zamV$n8-EdLI(k~f_AhT7t|3*1UhUio!=3;qi7TwOo5R*9(s*`duz9l6E$y?&Bys#r zGz{1$)T#Vy(1vS=3cZ5g}ALi|ecq@2oa19mGxbe`WCZ0E|GB! z+beRhFf5q$UKoBNy*ESfu9}juju8C(#re;!Nv}?yGj>NKs|#Ak;XjQ(;FaIAj-ld& z@lK93n;9p#id&?YNfvI=LdB#L8%9Vz;N1^Vv2l3<6(kQXLNu| z!^sd(E@jg$WYdW&mc{HS@bo<7jzJj6uh>@j0pn{|O-dgWEY{-UCMx#v?~*hp3hc_m zc{urkf-%Kuc)!LzA^3mcX{FcYRX_++C?m{#;Z!s%1{tx=+*Wlm=(YRAcSA6<(O=U~ z>+U}-yc_#U{6F!PuKxfG>lRvrt~|twRAxeYDO`sK7&t0A)Y_-Ss3BHpF*s6LlpGQD z8O?iti@pnd67ipftu>F0{sFbR*PN)lyR)^93y(Hn1njRG+g!9`k_j&3Xl41U$Mp}` z58+j&b$^5YG4Te4h}6Srs9d(MWJv%6JU6PBED6aWE>VI0A=12wWfbo0&c%&`TR-di z?0!am-o7Bxd`a*>Q@FR0UhZ3qWR7mTBc5=xDC{=^7d?46AoE|A-X8d!acAHs@LYOi z!gh5pGfXYpayA1gRF@!woH*;+zkmM#XkXZK#@3z^)#KMBw7k;hh|6(zrG2g`h(ov- z?Jfn%#?O-;(ripwykNZT}wT-;OQK#xF+7&+MxY#E8hGs za(}X|HJKoa_H$(%FubXe5F>G9C{U~R$s}W_HR3vMtETAQE^7-}v`ubEN;%UmW^}q3 z!6I9H83O#k5b>sRE7-h6;=NZ;v=*}XGf&WGgyh_5;^N$e+BSKLszjj=K;&Ro+*7TG zqZLjoMtt@o>osPo%37m=_=)h(Lh+5Q#m(-iuIRV-QKGr9YgsJjbO&jgJ(1jdf39qC zfn05$hV|_yLuS)qx`Dn`w^-be7=7{Op(OPAiwykTPj99C3!}@g+ej~baiN#N+Zm2S zDED*)N`OWQBOUT9o$)@Sr)f6>NtWuzRdP@7beWCJYDXAiQFb}(yq|jX@L7#7;lf&Z zo^C%4TBGKgSMGextzI?$H2YZ=Nw~momP3!c4{g~Q4ZvZ?TIcj_HYxPww9}87e|GE@ z?ieIdCL7LBc9vtXKHRo5#@^@aSA)Pex?;h{L9ew74$cW zeh>K5K(s@l$7y{ugXB9zXb*fOU?FBb&m8ox5%6w}@x#NnT9wC$A-mHrWYh^sB6!OA zKmeNp5}_Cmo8$~Sahlz<;?>M*(rIjTZ+EUotfsGNVs7PW7U5-kR7oT%LyhGMGXUMm zIme}X=YYI@{i&H(*zbio&VI$FFmn2bfNsOGjh&BTVsh-Y3_*J>vfWiWr7@!$wrAwlj=<;qdfn;w#3b3Yu@P z$omX7XUSGPy7fO&e{4N)d<##2UM|+&Tf2K(i@jFp`4e5tV|$B2Tgr`=M`*)t9EN?o ze8&~<{{RL)H253hr|rX~{7dlWr#;obhdvu>5wcFnWi0Z~s31gW=8;AvMBNg~%0VN3 z8BRZ+JAaBAv*!xrhT%koFnVXMY*QiHl8jDk;dih-sj3oz;` z>~?d6V;d`wFm|3t)K+E3%3E;TkJ7$wa7TKaUgtY|HqgwyHwy3{+FCCYOYo~(xA0V} zZ>z~;a+Y$kG=U{Bz8h*w0~~uwhy9^m?R?1jiBEBw@z0I7aOvI-y}Y`8k-=qkDPnjV zVSio+9<|+;E?V^E*{{r1W$w$+`0wLShxMy}_#qaVp!lx(;yp`6@CS#lCzpJY$X>~9 zrSjzoEWTz}3h2@@NPhRqa6gwMy+_tX+;R=DS;SK~BYLESEG~5 z9aBM-)(K_seu-d| z*P}0a+UF$L#GeWH zFX8r^;BOX31^4_c$uB3??m#%Rv9f97(xm&_R$1@Km=F#l3J*b^v*KOcn$N*Mfc_XGiaLA z%&RAvV;Nz_GV@3E&Q3VbIIp7r0A@XQ<3rYbHR0_eQq?ret#`zKVuduzhqaB-;ZrA` zBVpzusBN*3L1W0k$ocEQ`i$2;6!AUF4>6j;6*w6L&pM3bj9`#K9S^mAMf);%Bg6hA z@a4CS{8Onxq}q6ESTo0QHP6{gT--*f97a%~(LP=7Rf}!la@em&F2qWWMN$gwcUL|K zEyK#CXu-F$>u#s?Z~Fl0ve|yl-Zay+k^P?>x`&G=mgzPbW7~6lLq@VAD?8&GijcTD z1mIWJ{{Rs!_01FZR`3>&;ajMM{4;nG-YcC(Hkf>qS}>8Lgv1$vT;Oe73j8+xj(=*6 zbK&QKzAWoDR@2}3Ueiat)Z=#CkUTe%T*V=T&WO=8LS&iMyB=~$uh*SF(6n((tXD8h4oZ|AxDjuJPiA5?tB z@Kg3`y73q7bKwR3rmyArnRTdmV_mzR=H!bVI^5gc>G4Bv1%y&=@*QV=@w>V+9KP%> zJ#+pEiJ^EK;b(-uYJU%WKGUqUeLqY1i{dEl?5vhJWVXV_@JmagJ;G?=sFS<7m{KX*d($40D?t)Dbw8G{{X@p;Iuj8{dD?%N3Zc?+P_Zj!zY~(&oenA)N@~$x;}$p zrhmaNJ{tH-PnPFSw}(jZ6}mOMFPaD2w89Xyir|MQ4sntR!3P9>nO?|N?&2>=ANkkA z=QKU}HP&0J!*s;`ztas2Bf+LCe@ek1(n|eNAg}Xia

L*Ns zBO<5GD-3K?)li)0C$0eKImSkO_C-mybrv@h3v(o^%W|v!&ushFipUx{G5yBR;-;y6 zsu|$J51ixhsqKQsW@2-_Ul{BP)^ga*ZZ$^8;#SVzB|i)RDe?mviII*6T3F6vjeR3u z{RCB&3S%nV_9C$~XtNw*QUl=g{&}mCK@0?OjGiz^=BTdSsxi=tx&-se9$aiu$35!f zElipy%5Pw}SmL{O50Vl@Mj0KmfI$3BMPqR{+oYM1OEBjgXY>7P?fh{{%Gtpk=(beP z%Dm&PZ-$@ml4Vp&lH|83_QhpnVY?iTM;$tI`c^I8otOy{g<;vw-hDD_sB+O-8PI2r zohzNxW|hfOS9W*=anil4ZMkl6VcXRlJ;sf#J;Tc+1>xgwmO~H62EJtdzN4Q{{{Vtg ze$p_qmHz;WcYY2=L6Jci^fVTq%lG zvr#bIOo{l6@CRY(gV@)oU0mt=k{o0x3KRe_v;o)JyllS<8jbTG6T>-uEL5hdE7;^@ z@Ep=3Ew0BSI5xL?z#T`*NFDlOxwG)zT_06%5m?z9JMwbZky{w9x#aD+gAxL9)G#?6 zM-|*$*;(D@NMcoNv>dVJhtf&HyX?IXU+}_0v}XqOTTxMEF*U zccj{8hk!q8-`hXoY*1Nv2jYgm4W+&Sw7R>GQ$jleHCV1+Z?4%C_Rj+!W_S<&3p4v% z_%cO<;%A5aKjPR-hzkg=G-<+c4jvY}hDPg*#~3FAfnGu3&xalu)+C-U5n07|l^a>t z<{ildbLrnXuay2Ncy7zXUL~7I)4z3Yw@48S7TS0u0^PC5=DdpVlwI37@leNL5~~`w zU)OW{4*in9Xc+$hYM+AIU&gB~F5|=TYI9w{Y-g4^T^iDMfsdIV%2mqlPK~sUz#M_~ zpNur?OK*eTCxQ`!Hkoo6k0QcMx5U5CVp;sPg9xgW3_|lP*~)*Knbr>jk3G5 z!RTw}{{R4n`|Lc2j#HF8a!yWur_#QHbPAhd;Cpwkh{VMEicRdz>sZ8_Hhu2(T2y)N zCbyds#>`NwjIIZ$ITYPCZH^Ut9M#1G&vw8+IqB|e&YtH~QfPjeWd)tblW75jza-l` zL!5uIr*BVc?d+Q2=L;;-2Rwsu40A92^&ghw@y&A)tH&>!3HgEGpL*QU?`~tdBIg8j z2iT0)jf-7P;Ccm-)CcdD-V30ohJs>^>GvdDf8+If#_}CHXwKPTbVXb=WQ>8>uHjj5 zt9gE6eCvZVeekzPHux;Iidh8cKYDcjy$N+U8|fh4d33mV30*1TX@VN@CL?hy??q%1MmX6)HT)WXU*8X zt;cM(+H0X0jzAY3fm2>tDr}XRafc@v=DlibOQ@_RwYdU9qmA*Y;P(utf(Jv2<7Kha zZcv0dWe14eNk2f?{&ncrR+5pODf_JRnC+d|j_OMNeZ^BtON6<&Ert%uAOp5Q1b$+= z3)?$wIX-D#HeT|h0gk!cxfRb_YjCykirkBXBxn6o1{p>jeXvezt`zx|(VLZxxKV_>DIR7h%sTesOl-kBxDQsdsa@GW9K+MXHPQEr^pB0x)1I0 zMIo(=b@PCc#~JJ1wdA!gacZh~N13vHe7&F#`|6u<8$li;QJnrAd8VLk7MYsbY!_|0 zz+#7u!#~fT*1DTp{{R(OS_@4#QJy<;Fk>?#-k075NSfKs5a7$-c}yMxZL zbl1gyT~9X=fuk?8_a21*01|v>r^n^V9+7nuKgv?>O*VD_@&q?=yg9+o`s{YD^TU2R zGF#kgI_wQ?6JkWli40=_N=q%fO9t$WjjjPw4S5COi8N?jDE;HUea9RT(z)c*w96}p zxYOW{-boJTcn6Z?;~6apLVp_ij29c#rvCtW1pPGs05i?0lT&NmXW5bJ+HRrY=hQ52 zA-NI8WCDm_---pb8U<5o*~WDS>%!>h6$G6=`8=Du9lrq!-( z-bJ{OjlVQ^2^8qTMmB>Yd5RBxk}3D^SUx1a@m{TMXD*ql>RueW@`NIN7WznI3$(c3 zZ+e%IwmHG$u;2k-Nkfh>bu^l5)qj!aJc_JeG+ocBJYiusi9W++bLHOzypB;Ekq7xu zNxU|ABM1B`d|TjO3N4JQr&z6{Z6R9Onk*LJaKA5JdRNDu8~BUyhvDtSL&d%@gT#7S zkr)0Zx?90>&Ucr)(b!URNotxk>^6&Gb*@|9sybX~f-AOwyJKTp znG)gg$IOv;$;8ZrFpMiOXk>4ZG<-}5fwHm4Z0>x zv)GuVB4}SpV*T@doq;R6l|JK> zj@(y;h@HL@vYamOBz!@wdTNx&su@W9Z{}Pt`rsoZ8HuJ zAaP!MpTkXR(rdqmlF6seW7@yidcD@3v8g!$G~Ga7>7~cf21YmskI{Ww;r^W0 z-)yyW6!JDhyAVSVGnF8(PMEH~I~gX@&9%mIR|Ld)`twT-o7SfN*)&aD6*((d{$YMF zf59z3Xg>;G$*AahkM^uhhVr~mrMnpzXZMY$=(1f~M!osAKrzz3F!5*X=lex?ZrW`_ z{t)YpSuKz&>K+u-S6ht+c#K1BviVz#{Pw^v_le@avURl7G~#Y8U4ry|#iAsCyqVy4 z^sYO{>EN5F(^B#Fi)nUXQ!kk%ymyZ3N`~^JjboHQ-Q;o&eI5^tYeVBno=t3)65B)g znb$mjf8om`;lCMcm)a%l=V-9j*5gZ+#(rfnT}2GCbqlx?&wSTw;SUo0F|@Tdx~GSA z3yB!QU14&B!Jg54PK zdMtZE>0h2+F#iC8O8(B8^2j^~r|A~AClW^51;(S{iHJDM$rY@Z7EZv91KR=HBbxdQ zCmP}8tyMJpqtBj6Rx90iKM5o8zOQwzN2hpWP&U?4NV$t%p5J;iH~>UZ6^bSaIV|J! zuVe8Ch?l{7V&8ac0L6<7nXbMb z_@7;#DAch{XHK-%VHp|uUrLtNeoqCS2Rvl=uYkNo`yl*%@Ns)x8$r>$LnK8;-wf+8 z3x+(gFK4JFoPm#0s$@CHIqP3Vh0o|zx3a7E{{X=}df8;-cQ_qNUi={O*Md>4E)_K` zGDZa^1r}>F^KJnAp{<)CfIOoH{u5j^$HLD9+^}65X(LFBd148{+Kzs7Q_0#+GClag zt}6ck$KM$K6B~;Uh+1XuhGBN*U2jFYg57`s5$BkrwvljqmIs`1T#ldPZ7Wo~{{V!7 zd&ksmWpEX=HiRb7bHXYjgX&i-J?q|~ijq&6T^>bTzMP%bkI?scZG1zbTHQ6=hfuke zD~Kh3y2U1tFe}qx9WjEwg1pbeHhwCy*Y12p;tM-VEw>?I20le(sxXP%` z2N^Z9seDY*H0wxh{3!Z}xRHcqH8B#vJLAo7-e=I@4!qZ;X{W`ywt)A#lr}RXBRMH| z2bxzsHWCTr(379kl}b`q*f986QF3v!ZCLq};tz}UOB=}M@eHs z2F7kMIQypovJHFYgW#)6uYh{j?ICBWTCni`x1?zBHN(5yWdR;Jrn@$$fEcIf?m>XI3$|7|7kw^MFC(shzEfcOcjEw}@)lyNbyxvaa{;|S!Rek8sYFV)-;u1QEz0sSfzKGPPw~#F ziK4XLVQEJV{pSAY_pi!%C--xwH#mo3_SBl=;%^XrqHO;HR6fuby z7I_`80_2{6SHF>9rCSjQ#a~tVoS3SrFlzk{)ArrB(JjB=mzu7eAreDvpy&}oCMHrO zS#1Kg=?gP3C+`!x-oKfz0BAzrjhW^ z;r6+GrCV4}Jh!c4jcmD3vaRgJ3WYI7tX3f^_FcH#3Fj|$s|^cCzPIq*!|A$@pKmfr zEM!I&B6ouYB#2@k7OSHdoq~w{NIk9aW^f)W)4C@*B(+ z(3t+yEU_lkX>w4mB8-AT)|b2QpEHBNZ%$8J9~gXM)@*g{RzI_g3wsG+idYAj7FLE? zB=U0(S)N5gHw^3!Nf_<=A^St>+GFTG8Srhq(WaxN__qH3O90Acg3nUWQgbV+oCIs7 zDZB2OK2gpq9Z==7|=3Nfj7-3&BTRlG2C%AQ1kVH&jf~)Z^86c6oM+VeJN9NE1s|#CmsXFeCKeNviw+Nsgnm~m@ zsTeHa6T?^6`WJw_A9?Zr0K>lyyhWkOp!jo4)kdFbqg*ndv`q!ReECY25<9~S%EXey zHr70lFu!Wgg_i#SvG;{MG4R8}dQ0hkD$)Ekrt2D~h|7U@ajI&UahWYtC@865iy9Rm z?r8?$^5wka_HOYPiF`@%yWvgminR>@@$Uig`SExX8)HFjVIb00yB1>tZwJv64; zZEg6TeiI1{Je4-RtD{ z@JnaGejWI#UkQH6KM{Oes=@Yyr)hVW;tOM!XyjYin6H>r7ue+B;2d#Zpg*)Oo1&c` z!nzK>9Cp%b`p%VTsW4ecbh4gAPBN^>$`7VSdBu2``Q1wUf51OMz(K1?9p{3gzVT=6 zz3}_u?v;Gcp?|`@)KWPA0C%_jJ5sY`AbDlU{n*J5@{xcrPJT%Jlss~pFYGz{U28i0 zZEL4#-YocsG*?4uBsiSO@~&Uyewo2J7#_bse#ScI_UDZ}KjB&BQGMa>3tZSc23Bq9 zs(>6RIoqCj1A*yZ6#Nm<@8A9i&HG1bJ|~^$x?7)!Qq4Z*1?IMvR%iJfXLHBM+*I-o z;lca@SanO&~Q#0jsf=KypQAjT0O_YE6WXLXM#Dd zt=2b;9m_CgXwLz-;PnJolR`fFI-{ZSckE^?SH=2)u}~N65)~|0dv1|3#48X1!5nds z#%t~$7y$Y@u7BmcKl&K2oIhtD1ls7@?e~am^+dlC+j%BEs2v-2T(;e<*+2z8hQ5;V za)G5p%06Rr2>$@b>**Y>bICq*Y2FR753UVlg&*FXJT~wy2;GdHM-0dKn4Bo->s$xRQF7 zX6i}mN&f)r4l9xH%<<|zBGz?Fm}QR8Ngb`Q2qZt~!Gx?zvH4m@V7MQ`bIo*`oced! zm>a0jsn5*Y4;-Et59e4^XH;Ukn#puym&*W>K_dY49ep5Rkpb}AV+1&{{8^{hfsX3Xy_IAT6tGDT`6YTIPV2h+AQ z*0EnwTQ8mB^A{i=Z+hvVDJyJt7(GAF6;VblX>u#IQxcWxdQ{p}ThxZ*1Fz#$uT|rN zZb2Eyt$i#HE1KU658ftCNA5FG;z4N94qF5R?_4GBl#-wA1#BE~&l#@n@UlQAhz<$d zGupiF36k#CSAkhpGmI0Q_u{>*cfxT{TC*Lt85%CRP#AqiKh7)WANVLmsIdP4f=2$) zYaT)m4zdsX>2f*x*VH$65Kkg3VEllRd)Ld~@KKAEPxvH9?GF)c#k2zCo)66=YB=`A zeSd||{uU|Uw7>M0c-&cksLR!UX#OkgwM)BV>nKLr#FE7489o01&MT>dT#%xNjc~jI z4lsRwm5FaPw6Yl11ahH|%so$joxjd%SG$23t!^z!FdxH{*Xldh^pk_!{AQ*3wK~f! zQY)k~IuI8m?IFk1;O8HOb~-)Gb4hlHL1edJcJw=jI{t>dm+hK(BHlG@xX$+IpmjZe zrFFgo)NL+xTl+gan4z<9j@&ejxNbi|j{SRAE;b9;$@v~mb0-`u+U#juL}G>#GQ`Uk z21ZET&;J0bwY5zxMa$fyZY)nAVD#ucsympzTUB7c>mnU-fKN_3=DC}#UTa-maAIwhLECVObHUG+ z4^K~8@Vk51WdPjqxNv%}zE5iUm*8X;lW3Zz6nwWy_>U+MYt+S3wKmVusa3l^miT`H z-b*YXZ3!9Aq38P7#-FrQOK;iX$S z9PLr4NTGiWSL|2AYj|%j?jAwA$&4HxLEd(dLE!OUjsF1fNKF3#!wu+atPb68!~=tz z5A7*`r{!O=9}3@W_vG~vr~ZU%Jf?(nk3;OTs;*cnQvT}vkI(-Aj~){6_kzAZU1&ZT zwiecMM-|7EpluB*Ze?HgLFwt!wNqFa+l=6Y&%Jh^wB%xMj~CaZ$i79jtb2rI*O-Ah zE6B!sisE$3+c}{S!*d*x5^%Wyf&S^iug){nqll=YtdnQx_y*l5x7iX|M}g-9$sH>~ z-A|lKp1o>y(dJBC1Ph*{k&u5%V2#p7UYR-lYv(FdUC%}e#=J00_!#+HJdU;3Xqt3! z{L?n#d!K5s}rvuS0luIvAccO@MS9esyvwS^ofc zj04d3sy1J_Eb0pvQ@8wT?VcID#_o9b?OZIhyP9_xm#-LJO;eK)pkbW#01i!A^2C9G z$r;5~ynXJGAlfiF?lIc5r7JRJ9S=cp1O*si6wVJL?&RjYR%kT#SCocW`T1Aho-Mv_f_TP1g`cxGi2QLKyq97--4O&xS2>$fh5!L_2Ui5=wR_nG z2-m3!Pph%bkF4pz$sa&?Z}yA$5#o`HFB4nYv+!foBa+vX+jNVt{{Y`r@6h~1@K=YU zuxqR2x|2{>@&%Km>SK=^n<20O$Q^6>n)rSDOL$vfQxA!BnRJVlByH3h-ZFk|ge%1( zBM^G+WA9&F{6F}MY2bP8o5J^3dMjTAE@4>aid9Zm{cLif#(3I6>&<<>W56yq_EKs2 zpF4-*NWYDJ$M5Ea;{N~$>T<@`_vo+2R$RuzKFkJx0&Aht^wDvpTFVm&6vVuzAQAM< zeoJ^i_MY)gm`}DgkFVQ>7>4TJOF2m#b2Ym5kEMIYllwvVTT+EDB=F9wq+IRD zlS{j{x`3SaQr7c-Ck?FUQ{HvI{w`+OJqY@aA z&u%#tw6{*W9ifw!%N(9hPg>{huMCm|9S%)q(Mn3?Th%fmGd4lU#YY5LhBGk8jWT|s zCgz^t1O-kQ=jbbL&INrwBaO~?1}E2H7_NDHr%WfI;{GqvZLf53a(+gQmA(D{07{hU zYkA>kiUPi9ggS@haNk4Px}9z<))QGyfTs)f1pfd!r>8^Y{{W+o6(TA)-~usEBx-r;trICdL9ey$;8*a=!3Ynw)Z-oSn6H^*vsaM1xlDiv9mRVptmE*=;c-pf zjXo>*hi4~~;SF+I?V*2htH*C}-!^cfSk}@)M?APcje9?ab-h}W^Y)0!^~h$8vN7qEV_ob}$1)wvS3N?W2>dF( zxmrH+)fu`FmWx9F0EqRE6vKQq8REau?W5klRnng=*E|s-hJ@#@GH^5RUQyv|KN9Jd z5$IM}zOtI`H&l6%k@-9hPQb(xNB|D}SFc!I>RLo!@g=x;7>p4cvTpwXduagvAk|y# zZ|w68S5ngLFRvz&IG*Ihzi23U{{UK^T)!`GpFZ^IX3$Rc6GY1b-po8pZSb>Rg3j{q zQn=IgX%5KWOf$8@N)PV@W(6!$AcMPfWOG?I9u$X0d%N9VUbxmY##qIoUEEIQ7aO;S zgtUCKa6UjW!OwWCt|QTPDD@2*_f*o(7%+->?-jvdMpFJcT|mc0oL5z=YPzS2?Q{L7 zrAsV1bdOTAdt-5qku>OMjx6v)qb?6Sal!WKW|SwSW%uZC)WOa>qEX_VM#|gI(#t^& z?5dH4UBM6lIBap;el_PG7?Z_*Gw_G}A>InJv6|Qv^90tATr`9*03>e8r0wAEE9+2c zo+9|K;rClzN5Xo%x@;kkKAbIet77}O#J18Kl~BFf9zAQgweghtq|l3v8faycD7Mo) zDKix~&y^!2ybKe7!~!_Ybi#Ea?u?~ME$VzCx>tiVPZH1JUlP8fqP)RM*x56|@(e7D zu)E1_6druFM&SK3;##llk^3)vVevaf;AQomOs=Kf7FEag^Vg?aC^`7POx_g~3l zz~jAsn)u_v1L7K3n@k|r188?_%V9!k0h$S;ZaLUPESV(Z zI0K$YuUznV!%LqJm5WW))us@g=5poT+Xg~hmOnlLC$Do}{sW3JaE+V4uk~|+wo^sx zvp~HjTLRYQ{KDnMOZdX_6_5D@W9a z0&=8u1Xrr~ixDWd7s~v<;M~GFeFXJaK5G4vz5@Iq{h0LqM#sR~mZ^JrtfjcrX4Pel zo<_Ibu%vNC6iN5nI~Sh$BEJ6c{GSe5QlM;X9X37>uid?sm@&k;9PQoNsY=x~^dP=_MXc0TO5)}b*N_TiXg zAoe}$ox0TRA%UJ*6@0#1C$>1RoO~za?~OXfg>kHS!{JAYd^jxE&6Hl-MRZJ9$dOXU zS*F@S=wl8$pgdQb+y2yl2t0DJ#o^6=Pm<~1{yoLaWzN%r3{Fx&eTvtB@_NvI@;yaf zoR_-Jui|#Od#>qM4Hq(7@_`9bRYAzl7&y-+znV|kPWikS`$=ftF7WNWhN-NLerrn| zF}_KNxtNGj=I&)^tnK58Lp+Xm959U}nUzoUC#d-2Mr}^+-%wq<;dWNWcPT^2_xAq) z8vb*B5qxR!2f-f}T6o&fMXKphY7<^b6zWRc$Y;*<4xF$6DJ`9u#?i%orQzK-A8MbM z>U>`kJ!5|}`YZ82fG>2#)jltLVbsrw{9Lv-enSLh({zXpynw-^TDfm1Fh9Mza@RK* z`JG}B!yocj@RAKJJSPPD+#XbEYi|dZBsq-vQ57KL=D{0!VCOaZWAQKk3RUBu7TV9H z{5bH9zl9>YjpCN(9YRS9EAZ>~FCbd{%1AB_@Oe@VetLKk&Kuk5Z0CU<+0r=FZv<~s zfz; zzi5s)!ij>ZP$b$F$nS#NN#u(1hWI;iZEJ6FsH8VCSvFvFW`*Je<%u{W1atn^df7gQ z#={L&R+O4B{3D_G)5ZQUhSycq;4#^uyJ+vfE18zm~$>3K7{iET( zg*H0hgmrjEu^zpoNN;W}VmT6k-(=I{x!Q{)4CF~W9FSLO%C{aBr6!xITi9Pq1XoKG zyCF@)%F_9643fXjQJ0}q2u9*CI@f(GNjvn|^kBV|eeS2#{{XeW#f^94et|T<5w*BR zwXMrOx2L`A)`G5f>#M-=CjMBjDZGO)Aof1SwQRL2&lk&tjum`n+_ObZc;x7+) z1I4~1o5Nb83Ex`O?XB*<$*fo^+uT^dbvqYnR?#GqM)Kth=XaK}#~$`6;o~WLT2Gbl zf5`i6Elxy}KR|zBW`^7Lnn8Rouy{i(0o!{BAb$`w``g{p>gfEbA| zN$p!}{{UoEl7ANJH&Yfb{hhSu z&0;@%nHiK1{B#9B!;m=o zAIxan!Wg=)!YkSSCHQ~h&&3ZB=^6~0Vz-O*gqu{+F72-&vb(o{qT9iWlzrV`#_Wj%wwCT^Zd`r3F_YZVU(>#2TaLEsDME>lBO- z7XvvYfsQL)3&Ue_Sd4}k$gg2S`W-jY#LGSP$Ii0GSp2(XjHwv+Mj86zx{DcZc1Ih) z3ji`>Y>%h}3g;c*mE)A41wp|BrD$6SuDrOVRv@Vt!5uS| ztGPf%*HP`99-j5ctatWy>kKdEK6|AD_z426+uvIO^1BAkK4F^d!_oJX zGpRi+j*jYR&AY4M3=@-Fbgp5tDnf!f5#G12E`nQ_<|^HIIT;wPV(INJqVi;hRmi|G zGN1f(;PYOFE*#M@!|u_{-3isi$~O*xduKSWn}6V=@U74M5>NJrW&~%$H!~)abG_E-pi)k{{X=;{xn!f(cHr|rKC~?A(X6_PzF}| zgUB`Yei%Dv6z|$!`b#`cEO)@m)<61EKZwl*&)e84FbZ2chaR6!_1fxJ6F-pvI0W?X z$La^IeNWz97XK*(L zj&)>o{oj}aD}Y8<1RQb*JqXYAl?+rWE1!_!C^aL*ZPrVzUUZG}h;t+3sbY5H@c#ho z*Q)$EjTchAxm<|S)<_WZk{N+!KZ=1*@W;csCh$JB;_Iu0MAN2d21nj8V{sVc?=Ttf zO!#+mzV}cp%-awI+)RFbLGjhyr;T-Jbut(xp>(Mr7&}&0 zRaVDr4z=ZEbmb>DkCEVdytAmEk@R_r^GpUnM6mA3$KJ=f^slY|0A}lzmqOK=6VmQ6 zhU5NOgZlf|$W|JuyRli|DpkLTumO4uoM%4WGhc4}0q}mQqG-2~T-r}&FD@c6$#9IT zF6i^UTNohp7~_iW#z}Ho=jqgGv+Mr=0Pby8L;wIIKD}|%C$Q^Z2>#O&L;nB?*pF^Z zSC=Kp{{XJ5w;x|>`?KNN$sVP56~+=s<3B4LF~=Nw`t`5R{{Y%T>K6E+aTbN>+OOuz|r3ZiuhG2d#ylc zk0}TKK5Oy|{s~e_e#0IZXTU9}YfZ=fgk~Sozgd0~+%iw76W7X2eK#80E{&W|t;yqT z9XR%i{EvwL0BK!j_Rsc;)u9p*50!2Tjs`rH!Q-jTP_ep-?jJ5vu0Y{jZvJ%-?K5); z{h{@vIQy{MFXaLp)*gW=HiA|Arx-Q)hY-j8qNl6;&(X5k{_ZQu9sEjjwUxH&D4Ye(rA)J&($_J>s-aO$Zk`21ozG>O>bl~gplJW88z|sVPn}& z=Tf&4P4dYdyyJj;xtbRr@EH`0{{X050JX$75aS$_^9Mfw9zo&0?L=SsS1TE` zc#Kb$a(|Utw<@k4E-}XhgO9?zn$WvCVFhgt=F?Jw=5ZC3(lH0_?h48- zI6QxoTx3&=sM-t1^Aw|CJ`X@odX~^Z0_?O#h@vK07>tsDo->2_R|J)w=8{Klx00f; zJa?)x+~pYL1J^ZLM!OIrH04Pj^3g;804yKYnCQ$ECK10QCNjuz?hfJl*3|bQqtX0L zJddM9#~B7N*zd<`;ABB8!s8&7UAY}_3F}=JqdQ({!rD;Dg8=UI4&+8G|o!L6mHCT00ZBO{gl_@m`xi;3pySN1CPvC;2-=J7s48+ zi+(EnIMZxmEduG+5X#J(;#ZpDHBIMmE5Tw84@_~!e#3#)@wD6N?|s^yMt5(k%Kdcv z&*q2wKJLOJ9p$(PtQ?m+fNyQyy}mq&v4euEHj8hlr5G4zzVohPEX1^*8!>cN5e8mBxz<+@-#|h zj5KACkW?^HxNt~2xdR~Mn*NH0YZ+-cpNm%E+Kc#AKd-!@;eXm&<7b2!UejF~v`PRF z>TxB^VONeCKPq5x(3uauYu@}n`+odSu+#_IbX!-pWW=%CTuy|r=*lJ%fsgLw{cGl{ zZ-#nSyLzx_Z)-6M%%&*_Q@11*8DI~-Yt8Rt(ClZD8=0lCb&#Vxt0Jc0afKulIl;jl zb6U9J>geKgO9PluR)oJF^FLN}5BMn-k`$j7X_vM*d=D)|=tt2Uk@VU4*Rp6IwlD1I ze|R;kZxBbK-E?;)?B>rO-4V#Z9yfaTuf(k{OZ$9&Tso8&I0Tl54t`VoN*PZb2_5@Y zW?TIojM7CXnHv(jMmk`tVB?_~!Q-`KJeGvCvC)Ojaj;*^o!3 z52b%CFnn6^WtOOQD=l*4O}O68z8Cv3V#gx{Mx``XBo|ax3$U(dSW!$(Dja{yCnSI0Es^?HTVZn~kfhK|jyXGIOlKn;lp`ONc(|yfj+$hb>W0CD zqbIM>)^)TJM{e>1fXG4j6|+9+ZO@r+k^cZp+%Wdwg*6(8)d~_oKBN5q06NDE{o<-h z>c-HuNNhw3cvzKBzR}Hmy>V-CV=F>q$OPes3Nz2sj+ON^7ZFV=`D_>faxys_^{)}v zwDSe7^(O@F+x6*>TKXPjq>eP1;+`r*OM5$uQRTE(3zl9EA1Z65wZCG(DfIkD*0OcI zEZk|=5XZx+?p@tCA>2AsZ5JqvqXdD)deWktEO5@FQaa?lLd6DDf_n6*HTi9hkLSz= z7HCLe!OxhZuh5*<3rvDWj#kDA2B_|Q^F+mHdvQq zBT}?&Hre$ZILMMjrT2sLE(Ul#e_FOYW#TPs!;xtk>y1{%NWaux#+NDcE*l_jCbfitq`R+zGyoQ)lFVV&(NT63+eD^x*e>x zmQVr}V!F5uj=5$tK2N4orykW(;e26lc$$xayiY!faKsy%eM;3eOF5VhTkpXvWPk_C zIO~p0b+AQmATp=kD~_EgXjOMKfMdUrWUs~|&x^|@!GP01ta~WX3f*GSI zNcJR*_OEvbnn^UL6N0`DEojc);oroM59{`@CWqonYuH(qEG@!IA(vrdj=+|{^k5r3 z>*+m8&hlLr=I2h;JjfNqA|?O@1auN9Bsjs{o}Dl{*XMj62EH5WwvbzRlSj0_du;i0 z&uUP-p#lauE|dY92T?Yk=2 z02ADO4IJAJw4KrB)WJ&kiS^Eb;p@#~#UI;R^0b0D@dT zYEKn>Irv}VZF1X8)lx|s#S-L+id{5Pvyl-3WM>#TAch#~Fh*;@d^6+Scf;DO8jp;F z>Q@kn6YWjv|E8QE8krrBzE~(C>Z>c){(i$A$b?E@g^HJh3c@`y>Rk zd6Ui=ImjS_4>jXbfuO8Sp<;vYSiscCMNXlemaMuG>wd%4JY^Lv*nRa5)T&eL=-~O@@)7Po_bp=n}2q zf)HbWJ*H(wIbL_MBd!V0r%ox6oR)mg=R$K_G|jeldw^hq@RFl(Ac^E!xG^^2>Y zBVgE2LL|Wq0o)oRgU)O7tK*ma6Sm_~iM%uLFIwB-O)h^n)ZC!EkX>97*J30xTD+07 z1q=fntfv_x75xPKRkqhOLmbd)_j;X-jEE8l#P8-DhY~61n{rEI7|waGA@SG4odO%{ zJ6k(ge4?N%aA8yv(`t-yk4nZlg?u}`IcjNED!R2zPvtAe{{XT70LRY-YA>g*m1U@y zmnX`*lFH&n136oJ*yDucb{8xG$30DXUx<83`#}6N(i2DcgQsi$9@E%#xzh=@Rbqbq z)Dg*Q2mQ3~)K~SGz6;fK8+(|nUPnQl(9ek@Ax>}-P|7-TI#-Ho{utABC|M%VtnDvd zw;6QtErD(AlEorsh0k1)nEvoRE7ijBwMS^yifj7$eg~UVhv}vJw`2Ka@PCi2Z;|dT zM7pSziAn8m@f!R2cx6ob0)Ca#S$t5|?X?*qyLc__n+gTq`3^m$NW%XB-#~iT@0Y~? z0NBU&U-*p#x}U^f4`}JBvc5&`lMT(p(Sh>$H?}`wxJFhZmLx87o^xL@>R<3jzXod- z(Q97?bvf-BK!3AaY7lBDT#xx{ZL1_VTyz01F_F_5&(Yzy%L}S^kK%Y3yo$4T(Vv-3 z@l!#TMX{E~JDD3e7L4vlQ_c$~{{ULI<+kt_h?gE9vyJ@N_JzE+AP@*39l#8O)4prg zz9oObIsX7^Z-G)(c5sp{?9BQO&knXwh?oNPtj@f(Fzh*OOg0!~XyQXnr5@EH?fqj@}Izv&}U2 z$t-hu{!_+0rhs{ov+h&8IM~sKYndA2A~*s|c^tA8 z=tB{MkVb2=(LN?m8XJETYB8*s!lZVN0~I`+qeuxI{_hp*e`rlRNjn_$vYJu5r@KDY zwDEV0{6Q4o416W3Y4^IJFx~0;wcY3TV@nJxE?F-VA7#`nT%)rt&ps{QSa04&Grj`r zuxdI+x#Da8014Pyj}*fli_aabR+jT>dW6v}{k5{8W}4pZqJh|47m@}O;kHPqzd7i> zBf0TZl3QsyE#9F5uJ*XZYM36XdwN66rkq@FR_wWo-uC1h0SR9mCid_dn1{v!Cx;rHzE;H%6303F+SGJR$%jV{LC z{{T(8f#ijyn%>_0D#3cOT}K@8!L~**qej8_E9_qbd_AxHLj8jD-`Vp@(L5XCd)w>l z`LyfOKeaXQ?Hh~pb8%~BrLDS2Vm^5xlIH621+`b-CI)NZy+=pWt~E*ZO{UiRH@>2e4}!k|JQt<-li}>wx);SAGVbo~ z^HJ1cHx^p_7grB)CYPvPKqZ(!cmCqyNWl!|e4aSwD>N-ppS1oB} zHS^DLEGy;9Ylbnxs;}iltf=YzPRFh^vK5{oA5u1C0r(pJN`GJtJ5|@eZ_fyPQ`RPo zTx&PpD)9uGd>GkfB#SoGkSWA&=I%vIjn0vIoga_tNNTr|=4mhr{D&TZbg#;MM2u-- zY12#W-RZl#zvO+N49V1^Cf{H3JLqp=Y4=KY=Qt#Pg?X3l7va1A02%xw8b*x1Y>%kg zIT;+RWy!`>hJK&}>MNkK*L?ZHTrR*k1oX{$&+RX(Nn!Ao-b+PeA=Rx&2_wcx*)B+E z&h9$#_}9QfvT{C+!KRtU{>)w();vGseOvoR#M%$+=+n#$j2Zmw?fF@bR1WwafMcJd zJ|Gr|>A;-syNsy(MSREmE_m0)9wYI-uVb#r(!iQ}NgM8GP2y0bBRYm4401`&y?r_3 zd3^Jxss0Ox-a&N{90-JPYxj2!dSnye)~z#ok=;!l~e)3@hY=0?x|0A9MB z(Zs1-|JD7~{iiiUdHXvzhkhk^j%^3V8moDF71WNG5MNzAxOKfqwD{IZVuCpgml4}d zm^L;Gg$BMY*8U6lw))#n@eZwPsA)Euv|fGP{mgOP-rfwsDVO_DGkvDs=HNEOmomnL z#xR6|>pl_q39o!Dv%m3Q#A{h*(zS@L^uG>h`n{TJ_WBL9kVf~rE$y9cEvA^Fo2Xhq zvIPPb9J76gYvT_9!}~DB;eQZEY2j;)QpFnDz*tEVLem95=(fZe{Dwv==f?7RFM*w^ zEBJQ4Y3wAZVPnj$zaN+5q5Dl0Yd!V;t^G6e+r)k&@Ylni3u=BO)$T4Xt#sw{ZdO?f zdGcf|Fzoqbc_opRA&(di!)OMZ@XtcjVDRmxv*I*EtlS5CrG<;ak~tg1OhWG*1riW9 z1pr}s`L~Q^kH-En(|#F7$#0`*jTQc(6DVbp@l;C`M3C8Nk%XtXa~pLP^wz84Zye~~ z4Pfz&wug4wc4Cg+Xrn$;?qY}wjQMH^#^>A9jFL0;bF?X0?)5znY42)s`m5Nm)P@W~%9HEO~}k$z~_mdT~k-w5&Jqwr%T;YyFcm$UrgR?yh5M&1V$W zq~W{h2LdgTRv+ z&`ah^KUD-}r6A`%%U>Y=*q2c+$K3|jIG4?q31yYQ%Vb26p>jz$3NSr}I#=6&v$dKJ z0FZ!_=DMlR`)&NqetG`Y_n&M1tYXsMQdY}Mj45TgRRr^)~Pufaer{`Fh z-ZnfM`%1a6{h0-rc8uyejE+j+7E^G?ub~J5uNC-F6T^9^JfAdX00Q;ro-xO#y?yiI zPYC=me}no)o2qN9sBI#W1&Cq%z?;qn4?x76jycY2pjV1Pg*jx*aIj}__57ko70e3ukKEG+#C5p!u^H)k&q2OM+!tKmNpWxQ3aW4-;z z@0^*MzT`>R=~fMp321Te!LN#SnEc-pYLkvX((VYr=N@Z*d}Es5gU{B=DS z>TmcP##f#gyEfOXC9snuDC(oBBeQeP59eRApRxzUty@RyEtF=0EKR zIg{f4uF6TQ3Nw<8Rlke77Uz)bRMr{nalte@{Q~Om^A(4)`$_BBNUAP#f zB+sbW7FFeFYPgH9mnk> zs!J&#I!3%><|zzd{$7;2TsGc7Sk5w|jAPorCE_ZPc#4+(XZfF_WzqK|=#HA&W?7>M zI*<-MD@t)WB}Pc=&{j+l6=f%&tuHmPgU)N?>PqrGm~3cR!yuOk=O50xcwRW;!}11y zD&YLanCwQ|n&>p*h^iZ|Mt%KjgA$IaM{{Fjk0;t35w%V}gjU>fn2Q|p%D4xx3I%f( zdepN{ZOZM+x8;iHq0~{Xglsy6=eOMzIgEszfq(sU zOMH>8&@<^&C_`n}PfcJ1SbZ+X*Fjm6+>fi92)#E{{Vu*-Rcl{ z@4#`bwD%Uzrd&IuJ4g;4033lt+oh#xLI>asv z3#ZQzXFP^?M#cv?Aa$-wTJYwgEu_3ntptUbfWsg$Bmxgycdfm@HAg4aBPqeC*&NmP zfczEW2o~yFiy0f8qIi6V1Re-=+J8#&-xhdG=p~m$yp}a{zxNj=c(OU^CU6EiSJB@H zJ_vZr<2Q`7{{RB`E&%e~INu?f2x5X2KRnVk#L%ynnHd3#bHF^;ar;&N$~yl5?CB1Y z`p=0py?OOe>QhdCI9o}#a7L9LNaa=lkntV37&tfwvX)~?HB}__XEVV~!O6!%^SSQz zSSE@+NiFXpoH{X*5|5Afl;hW&4z=wcvfh-}9~?Xyq6Y~kvVq<`+eSAL*RBWv_s3fD z?NT)tmJtDrthxHFeX;u`S-k!{_(ihc%JaNae9}wHZwqg5paM9++s=FBb>_W1#$UOT z@_&;)dRTa@Oj_MP)cvdY84cu`Ec%SbOPQ8g7?KMy0PIB_&p@g^mG&Qk?i$ltx6ySk z^h?b~-tl9%o8)JdO$?0~Q_q+YSnWLEA46Vc`yyGUo#Cs!dUe5BgmE#)1}AArACBM1 zSI{0G(csm558+LI?2omth%O!FJhKBk3}>j|bE9LIc)l2> zjW>Smf)sJoH*ER?&3+a9Rrrr5g}-R;9>d{nbXi$eLs!Eo`)tf5o~5xSn3;ei;( z!(*emIhbF#+&_8H@f--D|#(xudLS_Up zd}@}z_smV^^w01fI@hm>$gxzOkh=0b`Wc-WYUN|`9?;6O6u+P0#laj z&o$vzUlDX!6K{sRZJ=vW#)XV`h}IK@1LX=O*h?Om2Q}%p7y90l9CB;^AGz|RUBYcT zRJ)HJgswn6@J)Q~D-SgN=(KuJsTTd&TO!oJxn$^URM1e;c`Gz`z4Rq3n zyo~2pOBniX^FE&xg!0+`b$Kr4LxG$gm6K-_g2p)`Q|0411A+DQ>5A62XxwhwmFPdl zwZusyDrvI4?5wfG>T*aUo@s-_cDmk~Zu*o_OKy%?QdC(NAx}7PM?Xr_mKdgk%k<7Y zkLz7NpJ=B+SQDb|{Z|#{RmIt8bT9QKht%gZcoi4;=BBDrP8@AQZgNcQ$3cI5Om zm7si6gIkF&^{HTCZn3P1tuX1!VG0g9^u+XMdq*IqBF zSz1_Xch@$@4xI~sBS;BV$>Ld^5qNQ2EiZR z7&!LLZ+~ypn(vM+ zJ--8);hc9qJNTFI6ZU5K#bF$NF4C=a61+y5v=;iB8CxL=J<2V{;tmEU&N>?TC&s_< zOTU3)DK5SrPj{frB#9J~=~|7&lxSEkRV?*Hy|XGY^A%(sjn5yh`d@(jAtbF8f<<<_ zR!=mvR}2)A2G&px(0XV2*3{kzw$n7z29X?lNd%-r%VJL%BcbDt_1%Zh>(TdNHoj*q zTm?D&NgvG*#EEQiZ<57dcseWv~c|5vwkZU&Jpd{~yY{{X}7Q^kf)_)AvbS6!_WE&MTD2%EOflLmjnV;t7s;JcTn?Ww8R=gA z;Xe*|hvF`hr{Q0P{AJ=@LSMSgp)%PvKJv$O-b^E&nVEqc=D(-gui2}>dVEppdWVB; zwKD(`eVjSCiA#(CBDgLWj<^7RwaV+?2EG{H{fJrke^~HjmKkQa@V(F2nadBkBpFy{ z+IdpZupErw;=P5&j#gCTZ|l(cXUAU!{8;h-0D=#O^eq!u zTbty*)NFj`gHDm!`ypz>QSr8~BF3;|PhH5SRs%B(eA2Nb5C$=T2T@;K z_{rV14J0p5wkAOI1XssDvMq()o8ygfCs7{zKmvo1F(}SDkU1O^*Raiff8xe?^&K3r z`C9b{=bGw*NgVb5Jr2(8iiY$Y);XSB@PAMKy=Yy{8IL2ltX<#d{EDLk|JME0zxW+= z{{XYuUxwP|r><&VJ4+2x;J&E@_LusdwaB<;jvJ?rSq!nMlgmjR;|L*=HdWfcGydCu z7(N;3UMJ+GX9GuN9@+T-vIOiK4Oui*Oqe%oKqcba%r!71VyzAGB|S z{5}1Wnl6zQjGh(KG_7Y<7UJS1vb+~IvuRfU0BN2kjvI-mk+w{*NxfW$d^twPz5SMS zPmkIyzsK(i*}-U@B=H2=uCcD_JFb^dNbSthOy6jl6|u2`Le|n-8Aka-u?qY@@itFZ zijuF0e6fdZoz?Zz&gbkDD@r;^_x`^R>!*qQA+XVWBjL?QSDG_C*N`z6ZxNVJCBrIt ziiN=qwB{p&ljWe!E7m?1Yl)$2k?R*Ndp*Xh1;ke?V=Qs4q!=n=yJrms#bO#&kz0Eo zmkbwff5QDI;OEEh9e7(?_OTq{d@1maZm%_c7Dy-5w5ihG*jOl$E^XyEQX;~(vu$NFh|4Pi09HGP zuaCu2#M+YkM`FaqE>*egelxsnTfi!HEh1Ph^vk&I;MDF!FwJvgDzwIF64-eY`IDC0 zFcG*14-%3MM{Dpy;wyY5pW`65jA-%gnZc9y-a|1@v;@XB1_U3QaK?7xzH0IJ{1i{Y zx>kXw{5trBHjm+-9-TrhQ(A8c3w>4z9`;z4J4yb~uNA?#6Wf{Ph)U`mq}dv>n)imXnS)_swuxafadnEZ{k_3f`Z=Wo| ziQ%h8cpN0=)znqiw*32c(B`L)sFIDk`sw|3JV#HvNp7APm5M|_mA0L!o|$Z5XX{-Q zQL;0Xa5nz6?w&32r;Pp)d_cFl_#LGkTf)9E)vj&su5YYvAkuAML%-~DTgvc+oxxxv zc-<8gs{Zpa*LCru{uG~rb)OITTjEXbhv2Jgk#zPkHN3(*ksM0P4aT6d#>{1B9$N)F z&k93f8-oqT6jft`%Y<} z3;Z?Wn0!0o-3s%=m+@Jx!!oFAD~~m@c7ofKE)FGJu6|ZgfnLT@#58ej#wn(fdiB3! zmNhQaQr`W#A3oc>XrvY67&TYz(OSrxLFA5}^;#W9$IUhka~GOZf0kWC?anuCCpgc1 zVyV1aqzcQOrz0GWE8XjJqWhz))FemKWgu?jgWMjO{41K7$goJ<MZ%+$K6^y+%2aRo#JQ7$AEZ@h{rWcOS4v#w$;~ohfj0jPANj1L?(mUx$j>wLeM! z0O=AuJ}7s=%ho^oQa>(#XP5r~iV_(C`|RXrrhLK=r?q}se$iKx+Dc`<{HyZc<2Y;2+h4^>gLq4@suXlqw6$67W}xpwm1fWVUhKX_Ct~EX9wWpl)|1v9~)$PAm3DU$BwQ9PZg|bSp{(@8Q`O<* zdTl?3dF@{oYXwJ$H545_*FH1vD}I&sjfh3C(*gmIVb3@@+J8UG-o7-|20TN2!~y=7 zIripOAFX?wF8%5JkIZ;Qd(|JO>LRXC5s(P$o~@prTKxe1idkXs?zpdwji`;FcmDuK zLO&dnpKAC7;6fN2?dk8{zW)7+_|bKCA24h67{Ke~FHgh0ds(c%a?$!81zuJ^P5vLT zT~Mhx1~S9G0rVOA*XG~HyS@Ja7;mnWfKBX~A639S^yy!vJ`_}r`cW9#3Uwb(&c87J zI7B-@;x9V%m@~fumh|u9Ke1AqkfeTHi?9q;KR4C z+9Hx3JBdk@kI45?@)tE_<|VpKqHP{c6>Xq){K-NpWrf=^W3u z{`qpt`d6J+tm)ZB@JT`X(MFh9uGRccg9GQ%J*@)gJ1>IONyw!RL_)J+QK{`yV7 z%C{oAifPv2BS{&5i6p_}*@yM6cFiwyte#X@e8d>XYNRGYz=Ag(N8v+UAdk#=ri_2N zzyAP5mZ(x)KH^0%&lqu@80V<`REVCnvpNkOd zw@_qEE?_Ow%!E1|pYKH3s&{vqMeLvHD`_jQ7(csl?emO}&b>Rru$i>vk{zUc!PN7{ zNWmSr^{zh3)9sVVW>h+{+NmKsd-mg0+d-&qdd5E+#@%;q&8Z5wbxHgrh0QAdGdcg4axg!~XymAhEZA z+C?stGx=ct>5rI$a2+=ehmQTbz@OpeO%ENp%9*)^>k6n&y_+`_ISnl?PAT6v(BuEx35W zWMT4{*_da)UMuW>*%HN{;|GC82O6BskaPb4EawgRezou7)AmmNpX7XwI;(h^N)PQT z{pkLczi00%)8WL*yGQz5xCG!Fs2~dV-3i{u;O2uC@7}Je?4?HGm${RYPalWjUo-q3 za|gpzY@Lg8&C~&o7ySMe^p1(B{iosYgRM?NUF%YeAG&N{*;;N$W&PRmKt+F7L`7-Wz* z<2`Glw$d%(D;!6LZ<2$I*jyKjIqRHMn2u@OIcZ4 zWVMH;N-^?WOBu&TZR7G_f1PyJ*E0zfaU8qyOJi@{9>kB*yo&lWwhXrnrBh%WgI8h6 zQ}4$;#X$ElAk6m6IQz)KhJ8C4d{@?Z`HsMUya?y(j8-sjn&_$QXxUFM_L>(Q4B&nh zm3WM@nTQ8#5`LMjyLB^MB=D~ELUPIT41wH_O6M>47>kab#dLFtQ5W2MpuONZe21jD7vSLWs z5WwMvJ6HCD@ox6|*Wo{m@4Pv1Z1p&_4N2{;1DO1|MYH*DynW5BgS)p{{y%jO*&E{C zo24bxdU({9RS&)!sP0}&^z1@bZ5@ABzTLqESa^Ex$5Do9JLmo~bEw$#+L_9XIIaUfqpI`TU4 z(y^`E+Fen$0%sD(@^U2M8#&N=Q zBL`EJSo(~sise?aqOO}iaWwDS7S~AAptJFAh}UlfYL7j=!xQs=%1&nCWu z_$Tqp!k-sK>-bjM{{T)pd5dwpNSryFa<0J~hm5JoZKu#z=Q8+G-^FdQATvg>+|fxFQpi{Q&v9n3y74o zmgngGyn6DgLdM?jxn(U2^Vq8fUO(QgZA9KZqMO*Igah}m9CSU@w-v|fT9w4dFUbUC zWAv`OQ@du?jHH=RA;xGJFo%N{{VFV07{N>gtLX` z)_jqLPbu7ip4k0yUTZdm9FW|~F|}BM!6c{y73%sh#;b)GEfcOjqwuarOONdl$q_qF z;xK(MILEDN&f=AfCa&y!{i;~b(GCVxl$OVChmWONo)_B~0ALTwy#rFvC6XD^GI9Z7 z#|OSEkGM%z4ZsW@=Yd|NB^fKDlbxdtGRG^_e_G~kyvOlG!7vBe)t5XT802%;pXXZU zd#Pps0)x~ZGyXL7v1@&BUz3xtt{0rIAIhS1(a{>Osc(JMd@GD#{IAOM>~cR!=xjAR ztp>_-Xt_zi6Gns%V8O_a>FT>Yi5X?b;AY+)4h6qz1FO%AWc^5Ysay}$8B(+oO+y^+lq3mPBH<>;<Wz58qUW{ z)GQJ+mwD&0x@N{cSm#MC_j`#zKGo`86}^jB)BfFerQGSxfR>JN9EDkVKQs*@ZO0() z=cWfr;%B@JBD7u%@}gk(Qb}+8bxl#cf*Ey7OK_W7c>pY#N92w-E*r1CdU#ysqWI%; zj;;$8^!DBy0^TnK6z()?^L=6{@}|@%D;8 zw5;_zn8Zn+F52QmL)STm=w8LndMHeK`hAWAJj2@4Ry@2;Uprnb^$kDacf;QiTh?E~dkCj}V%!FGX@(DkfMH}-7PMUx){c$V)=j_fkr!@FuSC-4xb zVFWe~=*h@fVK# zQ{Y`oO`BMAaI+G5!{5{9yQ4mS;t}@g1#=`3_Rf{hG$- zfwzcmgl>9v2dzQz2T=$1Yw_jeQ^!A*px-HsGQTF?LC_P&Am_bwtTmP6E`+>H({bCT zhrxfbc9-yvL%!B6eidp{>W;Q{M*4e+9!C2^ZZo0=3561O!6*C705$d}i!K?p4I(Bv z0`(7G_3$t3F>3~s;(cb=NEw?=flfDY>nSB$EJi|sf)sJw*WErXnZL9&au3RH6Z0VA zzDBgKHqUi@TAlUN5!Zo=&GK`|&-wf+^vo_HQV8UBtm6!S=M@)SjCDW%*Zmv#!>D*$ z_Pz1`zP}3oC1|(45YcpI2`$~tsaX80R1h{OMy4v^;E}zQ*~KwS0~L00jR4 z()5oW{3q3XPw_j$@pxZYu+jBBbHlpKlSM9}rQ1OOYej}xBV?29cN0%}G_+#N- zxvv;>saE<)d`o$MtY7ND#>&$}n^^HS zv81J@oj#X+^X!Uy_lxZlEz$_vQWmbV+a!xJEx-+5e|#YLkK>=(GvVF-q48r{(QfqX zo83m!LA|%%JW^Rl2_sRwn&KEHmtDU!&7pXYUhw=M3huP&HIK2p z_Yz(RWW0tJ8S`ClE!4sUgZGXi!tcpA2QBIz2Dtb&;ZF$IXlvlx2^lXf}=6dTz=kh%FN7D4a4(W>y zlWv5`kj%_wk&d0qc%xC${5Lw@>9AeSmX{a< z&4nC?j^smt=tLo0hC2N~f>LS`5!I-Jz8P8ta z)@|0f<~Y|Ww4Sb{f&Tz}kyW(`zR6|gFU+9)-|qFuuO!qwRU~%lYpTZ9=;wEz_hMB! zE4j(S!+VCrW!k-vgB{)qUCL`Bv z!V~mN;*Yjc;?UeTGSnBNt7lMLJuSZ?~I9P%Qy2S1iZEA;2!-TkJk@b=0rF6JAC`#gNZB1&@0yPzbI{VVd*#`jW5 z`)+u$%z%Vyh^0qR3fm&_?djhFzgWUXUjG20@ild_KVv*Z^0&j6wT0F-{Kglut9Y zF7~=!8|EzwDeKU2_;X(#YZf|z)vvG7H$wMU7}yIDiCfd_Ut;N7WB$)GVBkh@dK0(w zuac(LC4&82!l9PpUb6M4feZXd8(T$R`ac{B!zcg!x&^zKw!*(K36&2xaX*^ zvwvlF(=7i0wj$H5NoX}6DV7OI1cU$(vV+Gt%C|YMTOCo#4lMl-4-}-2(@zS>t9#CP zi;u^pepq~ByM+8#)VqU!a7f4ZS%4?s+P_V_8GRka$w+rA7}R{rj{dnd`CIW?U_LGC zO5Bg_3$u?=z5f8EeB9NQ&j*Yr`-t1RKdUd;0$sncXTw14yp0kvTc$~o$v=&Ll6)SR zHO8Px{{WUqG4*=6?Ql7t_UJt}RJ zLZwbOe=$iE^6%UaBRMsNZFbQ!^&09;3kg;6*srL~WCD!Z8x@zUd zRy`>$B}G`3h!`C`s#x!CzS3>yCPZL(a2S5IjSTF=o}!#2mF{;&z;J&-)~YnS7tW5- zT|NcTUh)XjJaC|=L+ZGy&mzWfC6od=A1sPL@1TF3L~qT%h9aGFwoDv!$z$nU@7vnu zH-~VK%mf~J=Yj=v)}v;Osn2s#S#9oN6FiWW2P@_y#78*$%EW);weSb+bZ(o*nq|eh z>=4=EnH-UVc*g_Z^RKffxr!K?bexN5(D1_k(EQ@>^(eU0q2Y{m~L`5l~0C{G*1*=sRM)j_=@Sg1lp=?D%8jy(d!hBMNP(m_f>eNd zZs2^G_Nag0p&E2Ce$qMx>?#8v4_(Cg;O$$OT|nvFlla%gUMleN!=Xv6s~zbGkSPI% zasJkOoSawaam7mMuE_bhVdFR}pHEzV$KMt7*vh_}YbvSX;JQ`a$55}Z{+08G?NRV{ ze}&%+{5Pn0hzPZPTT;0#515Qn-ULNW;WD`{#eoN=0rahphMyI_Cuv$pmqypGtrjN{ zgpOe>e<&ZpD0Lr93g$m)-xzqa;$MX{y-UTn{{U@$Do0pb$ciR!v}_N&#uqr+3h`Z( zF*mCizUC9cwFfkQVBBwu#z?lqVyrTegMpp=dh^f^PHXfB_BM&6__5&LNq4@wXC(D; z1<5C#c_W(q{JUufj3mh;5f>eZC5IXB#~rFW@o9DlBc&V{JPjWXOZW1Zf-bj5e#VY4cJ zTx4VPuNBZ4?j|Bu+MsfH&N;5`#%L{-mvK8sRyil!pXXngVxxC+=;JGBcOPm007U>C zuq2%QDI&g8DB~(l2d#3gda^{VpOyIo+NZO;fo?zS}k;NC4ctWY<%s zNZVrL3Kmb-73Z*DTZKE1e_!ccv<^0+Mt_B|^cb%$w!5A1eNCIY){TB%)mFy#b_a0d zZfiPgg%OhciXpg#V;DK$RxoR!Qg=Tt{AUlv3$KgbGx7I@?yaNK^iPP|CWANGQfMwx z4NBs{Br{w?B4x*(Sr$cPF$T^;?LVRa0JUGm?}~p2{w4T&-{HT*>wg$%H}>~$tGrUr zY;9UO6>a>ZZ5yjIp=Myf5)=`f;QF`Qe`ntV{5SEAy{ve4=1be%I$QhG2?cI0Y@l~X z+vX9toy0!C!*)6f`BTB)7-zrNt+j6(d_VC`thZCzdDCc9jY`p8<^^cft((Pi`$kuY zTjvZ3;Ai&U8F(txB@9YV*64Z}<~FW9(v}vMl+#wxU)_q?G~JqQJM~vSjs2(mJD}-5 z1pG;<*#}5&v~5Kuibvi`EE#>0F2cDCM>*%xzn*V}*B%1+Kk)AA>%%%tv^G}uT6Sz= zB@yQ_uwbgpLT=7mXy+N`ziNMJ{{V_!4e?LxyW`dH8r;FCX#N=0{QE1bON&d3fAn%9 z43{W?v1D#RZf4u86@}juX8EDQ}(r^ zh0+&VEw)GH8Op6sk~oCO)+6BZ<<0IG3-8Lf-}YmJq2(70K&@B?<^k8V})Fd z1xbE@U_U>HUrNE1W70ecrTA9w%SmQca@>FdZ@|d|yX5^V^a=^{E1w@Ka*B37b@4WV z;q6NDX>2~uVlDF9-F(c6{ogUyErHyUPpx?K_;<>GFGgp!h-4LlR|)|f;dxQ`4uci+ zz3+x>%&yjqT%3X5Ae;_IU$1)dy>~^rw7zLDHYolc3t;p5SJBa^z7)>}v|lykk2Q+Y z{7)P@g#L3fgP8*7JF)kaE^F;i*r!0&ZN4hOb*C&%0Ml+IoU>&8mSU^&gVp@doCCnm zHS$HgGEc6@1bnf8{d2gIM?X$0^k?>8vy$hX;O5+SjIU|#f zyw{JN&HY|T-!0`>xXAr%(X1rW{0VV;ZvncVJ9aA?XD-`t0*}Y}*P(nSv%b@H%a0Cf zR?}KtYPxl{r)w)le5n!*$ufuOy?3rb>&_1bxxWWXsA_sj+F7WJY6Z#Uecp5USI}A= zhuC#(6D&4cXlxD?j%>3AmGtb3+}Ria50VYnyjT6cGfZQ3Op zi8<@&2{nhJ+*^oJH2cg6Ao0&?(Z02oYf~Z>`LlozTKJlh_q9ERvC#N>c;wR|mK^c} zAI7`uClW?jADbLkllX1fV|6Eyfq~k+8&8@fh-93cy!Ow`_81C5g;KH ze8ETqI6um-8(DtW=LBe-v%DU0jk)5#AvCYr7U#rTm96KE=AXou7Ru1u-V4bruAoxe zn_{9YZgbv6$>S%1Urcyk_MGs*jyro@f5cjIM0T(Cgf`lYVLsu80Q@J&uY=0+T2-5CO7R;z!Tha*1loTe`%=o8Cu%jZ4+HSS&%Gp zHlKFHe-D#3)4^-5C#o^6<3E@ca zt16hJcTKUnm2ftJA=)x>Ps@OKu9DMNjwwn?rFSk)n{{Sgm``1~g z>hRmVmy!pOVpejIqdT_1P8C-qo;@lfN0&#q?peRU((FPgVYWa|{0oYUbf039c$4K= z@^jPB)9qPqbenk;e69-sdW@6BIhor?mewvyBQF)DgQ?NzP{SNwY=SVZ(IM-|`LSH~ zwcy=1R`B!_YLG*B=knfkVfYS4K45To;~1{9M1@A5hv1?{PvXbh)Yh2Gcu!AKdGo^K z+_}YYx_r%@@RgC_J~8-pV=dCyM)LihB?ytT`SC8`H)0vYkpe*jJ1gd&6?_o*fAE_` z@n*T8jlAdYt#mENiK2m&?M#oUUTfxm z+Qtxm4{OsX*nGPMKBWDoECbVmJvvvnfyY*alvJIYvCUeHWUkJ8_FwRKg*2^WQ1M5H zbvW+fiM0^PK9*zhq>5Q% z3lWkA<|5;JBNWf^5976(R-w&Kl{-fMt>3P{*K2kCtNA@J+?-se)(Zza>=eR_Q^;!A5L5hdl-+Nzjr-J+64g#>EK8|7l)x95k0e`;Tg zI_JbK8{)T*HLYvJ{spquUf05Q-aL*LvAHc_a~<}vrCdfJloyU?i$|6jOluP}yhV!9 zmb$pBRpxw&U9ZV}?0DFUmR3uD_yHfpo6n9v9Q-k&{?FbD*Yr<`lV4tGnvSKeY8rKp zmwF4yv+cH8Lzlc%wntHH#@ZPl%nlZ1h0pCqoACbtz}6oN^nVq2iswP`t=^lfKAUN& z>bjHZvs*r(ie|AoWBj^pmDGz7k{E=Za;$PrNy}IB>7;9)0b$p)C;gmnH9Nbze-UXq zE&a1ex5l5^D;$ztYC3({x`Rx-i&DEWBf@^p(9FQIL{ut&N1w9i_CAvwvHsA$5!8G; ztT?jKt*$&x;s(7s!g-Bt2A_Dkq;a&8UA*!~aJGvw2@36+b-@P{Q^Yybofpd(-$i@B zUbg#_TAcW5i95gR`uULf=fr*w(|>2EJ{kOSZBtd(^y`TGwCbC84#r zYoH~WV)Ki`o_dd>RXGsV0)QvHtI6S zOEuM;aNDfXEF)@pjCL^UI{kaV-VXR9;mvpUt@wfD_!c{>%Ny&xCtB3TpP*RHBSUU{ zySe4cK_sfPTe$hsZ{0~A$5pSQJ_}ra&EE}vAKrMk;l0(Lias6q>dRBrZgnjRNNy5I zH8~o2C)90h%flg)F<^^wOpNM-A1!>HToqhoB(T~myZgFd)vey2RJw#eXhjl-W6)&nXHa<=9dU~KH}-b;wBH##3jY8VJ}2oKzl8K#-|X16 zNW4d8EH!BF5m6#8wv1;@=4B z3O;1MK42ZeQ~vTQ=2nRgJ94Hj^7|njq{?!_iSbxDizAsqDJi9x84aA;G3`Vi3h;0e?a1>xJ zF^((pGsRaHmcQ^vFN}{n1XsNA>M17xvv`v9C>iQW;8*F-?OlJ5*aPE2L_rPVa^UBT zA1?~we5*c)hNAwoo`Rbyt} z(ev{Sqyg9p{RjPu;E;SS@k}xj7C#v3>7JZ1jfnko_2Rw?@a603AMjEM9$lqVS-jeR zi{%QVc6~4yeQWP1)RU=Kne305sdr|6fqv6EcaMG<{>z&8#NP;bmhVuTN``$zY1-6Q z*H)|{Z{CgZ#Rb59iosjWAQQ@({Mq=0ZQ=-Z+k>yIjM{)iM>gJNx&!4YD!aV(!0D1e zIL{g5^)L44ceVY6d|r_f8skR1CmfxH9h)QE^Iw(!02RDKZ+Di#E5Sg*$a?zwD!K_N=On70U?7a}Yl_SlM?gsri|6jQwN#J$N5O z)ULH{IvrnBMomIUU7uSprAe_x79QqIE`F&seHMhhqf8oxR;vW$B!s>lW?l)&5 z-qJ`IM=6O|qr$E?ZQMYVV>lSEhdg<$>Ng(~wXG{vvxZ45ujNs546h1H=5At4Y_jb+ z&QueQ#=I)nN!g{)@HnCz)K{tfV*dcaJ+7mZ_9*yeB4Z7rBYeF%mCtW_{T}#FC>o{e zXRHAA{{U!L;GgWP`&VjT2R;P+2lzjE@b6jI8^_i&+UOTrE#-{fV_AltVRSCq&O2xF z_sO%&U{x5Y$i;q^XF-*at~u% z)`vaTnF((tu0iNybt6Aq^ImiDkM@=LQQ{AYx0l`p@vfP14yx%q2-k7M96V@Ltalz; z9pjbb1adeztv2|F2B0mVI(zYWeX)m|c|nsYzQ{YwWm?BlOQ zd!3EgM-nnRh8S!P2iM-VV~JYiDzmaeSrW6*?65f$u+(m?9u90AiA^{cxjT*gOH*0G4zZBF}P zZFPf#7Pg##83;U9eWZ^94?Xu47#Z(fb)$P2#n??kTQ-7Q zN!uILp!E5;>t7-M)VfR;vS`LP+4d`TWd(DUD5H-C;cRV6E z&N4yDxZBsz*UjIwhM}+N`VNnEqQqWU^=;CUh4$h(7?KB0M_V;9GWzc%?-=j`DzXw3vJpyVaVsySKVK?laKgbS`*+( zsp089T#=vOTM`KzjP&c%^R6f2CyYO4jc3C*M_Bkzq=^;{$u))ctEbGM0C<)wmTdLG z!64V@Sbj^TBjqPHqwb#__)oM@vM)D-ZAjjwczom zX_p2`q?H&08>qlya!DOMx|;S~M^5+y;_W8d=fJ)mywfB`^6ntKzn4&66-t=pmKbIj zIR|q-;m&i{#y%%_OT(HUg?`(m+K42EXw^t|VTU*%hHgRq4R=wEKWv|wlc^_3-1tIG zt>besB|w0yEUvRz7n`BJW|04 zC1T~e4q17xvwvi1WnYYb6_F3!p5=%f91B~Fp4c_@7}fikdLIFet!#A|+jf6NzX{_h z;X7+9L-Xw}SZ9{N&;I~kzUR>G6@CWzDk#V@CZRN&N%={VX^vYQjEuXKGRk~Wu6 z@co^%YWI++MG8E}*nXIN8k!wPN^7q&Dd9Vpw2RmCs`r|_(|CT`YnBWo+m(JsGhED9 zT1s3R7XozKt%Tyo0l)R*qleui2Mb1De+OD8HVcos24Hn02(T z>96E}LoKH1TI?r6*43WI_#0{P#SfD>^5nyR3E0P;-D~Hc2!78W4(&BzNaS@*KK}B^~L*0cz@!anfo?;L%H~i ztR}B{b>R!mHpI*Yw2oKKIK{++mok8$%;$o44C23@PapovpS2gke}H};{i%Fo;LEFj zhdwKo7YpO#esQPd3QZSeUF|#Xtz~|)zaC6h2 zQ~06qf8jsu_n~Q8*NQcrU&NY@qkAR17FtKyG`l%SLgD8##dmcKc>zmA2`=3iAP+F` z$L&cMgMK3Db~F4m@HVO8y$JoDT^q!97hWcg>e+ye#mx72Fv?jD9XyqUq>8P%hC%cr zgy)Yw9bId_hn4ltODgWn;AHWi!{=6#{{T^&No4!~0NNJvz^m+ClVRr_7}iFru<9^9 z-h->!UPH!wi6WR_@%$u_$rbhI?FHfg0EK@Lyk!T(4}zW*hroUu*Yu4(?kyiop6**a zdxwbm=GiY=M~XK22av2|jDxf98vO6^PlPl!)^BXDpo&Q?WMuQ2G!2;gVU#H!hd8d9 z7dw(-i@mU0a$OgfN zJ?@+N;C@y4v*Ey+PN-nCk-`WvBQm>2GOPhm#{l*?>tCjyuryMAtH#o8KfBU&m@+zi z*E17?><7JJn!!2}{<@w8n$2l?KgplAUJ+<5G?LJP$ZX@e+sEl%%LR$k?{qzCMj%O| zET~b~Hry`+bjDAoKhNfE2U0e=YTL%dI>dPPAoTaIrmkX{Q$>Ufq*@Y$8Og@jPyWqc zl5q@IvQNzYgDZO2^nDKfCz2v%Y#s=7R@_c{k+lA`>5~5dXMKgGaRJz*;Te$o@WH>G zc>e&0)EjX8lbnH!e!c5!RJd~n6!$8AgjeU-on>Utej09i4XwSbWX&47hE;5+3@}DV z-6Z~%F}{IMl@!U5+c^Cg173Zj-At-iCjnRV;C_|VTgy3O%-fXqBQ@q?Y3r%cPN!qJ zwsvqBE`FSk&aO1vb>qEq@w`4-pDYeZ`EtC2?TXm8okU=XWWmo4clvj1p7r6*DC~@u z#g>&^oSt~=NgEx-M^8$61S^w+kHVxD@UxjY2iS^GK=(Q=1LeAtwm<{(&2xw{vqla; zcLW}T<~=JCX1`{)hT=GqWPr>fl0dB3EX<_|8&0{TNsoR(?p;<=g z{K)R%TY~{C_B#Q&jpC@UQZ`!ODV5=Z+#%~n-Hu#U> zba^#x0f<+?COIO(91Y>Cm}JHj=?YGq4A0l0dJ$z8UzDz9M`@)IJw{ zD%Gx+SJLcLSb6P3tZ8cC2$AB8=blL=8Hr#82zr@{MZ}>@kJ+J7p-C574+{cI2Lx-Bqd8RQaod|i5Okppy z?T|Ohe5M~6c+*e#jimSh>6`6E!duU$T?K-90=a2QKK-QTG8inRA2SL`*blw_!5{Ea zZ-AZ}{eqzIPsg8$wmu*6W~Z#`s3F$x;e9^J+BrPvZiVKd6xPN`Hmlmj6k9;exQ&A3 z{W0;M<3EAFWX}%8@KfTihZDya7dDsL{kE@b7+JLY`-xM@p3>uEr1nz7BX3ke_$r7$sy6$YlpD zf)07-C!TAZvp`&or_k#FnryOvdZD`J9kf zBn+;4rCytskA06xRf}m~bM0*zl3gNV%rYX6oc9iM`PPlP?$R{GW67H%>aG5I*CFA{ zI3m*Si?&@ORxGS}Ju{Jxq-MHHRx#68e!I!de9ik?DTCnGh$V1US*&tTLyMS{j)%83^xf*+rnZ-I1bLMK1Z8*`9Xa=} z0{x}*ZC}OT2=zS!PKY(-$+eO(DGG!|IyuWIY<#M5$K}?#aPHMr`=LLC`S129m(0HL zDzPmR+gd6(BLsPpWB{Oj-Uf0(&j*8FSZgxf+Uojr7X%fO?L546Cu?K>0Iy#!e#o8^ z@gIh@zY-m0DFR(v*~0|0NQGR9q}r>s*ea-QOE*q<9X`bI$AkP?qKz-a8jYg;j@w(4 zPivTEig>=&6fzQmNS)N2at}f}5^LPWuij{I#%kwUTVjQX_ z95KYf9!c&Q-dG`4A)0dr&@sjcHR*ak>{;Xqg)8 zV#6R(G3Ou=wm3LBub;db@uT9Gi?l0Cn=KFfH^G{v(9^BFO>$$gOXUVRZ%j}Q>6(qQX_{57-MzeemBh2O?R6E(+s7zb z;faB@NcOr$8Df=#BmV#nLl;?8lxjQnwZF^q@;_l%nQ3N^gFkC;*>~dXzqB5a`#*eG z@T{reYj|!ozY}KqOA8+nTEL%XhfiDBL3CMJ{L6{@g{aWrzrq}Ogo(m~%^t;HTx3_%~$~KG}nOAG?kPy2OoA)37 z3Kjb`U;f*_2rTb(>+ccRTj{#h!+7^kyVU|qb)v&=?=6M9Z+NY(rn*Re)FP4tENLQy zA;v!~J|t-x2ZcOYq5L@T-+{HQ6H@Y|(=>TC+i9ZGQquGz-v0n+N2sgY?~OdQ{m9-6 z4C9P_UNbeIsdC}n-pc(G>ify<*)Kh~s>#)YlheP;^y+-C@hjm!?CEMeN${(|zA>}8 z(lre(@nO|8IBoSA^)C=u7*-1_)YD^vIH8VdCW;B}2yKyN$C?Q}&+Oa#IQXCTk@)-J zYdu$9y|&Y=E%f_e4|swrNHp846$Hm`{g{^)=F(eRTbppcQaZ#O$nY7aUn%@E;|8^= zcpt^qng!kVnWXsFPVnSUD4y3$zkP z{{WAFX?=UaUMbUl9r%uWEh5uXwX|(!#$7v3w~of&eZ{oWi`IE#nqo!4k|vDrX#_`e zZ9X0))r|*UH8p)2+wlIiIH^j`$Lk-)uND6Q!9@Q6XPp+~;z!5-02+-#+rv`F;v4uE zMZUOe5pg0~A+e73NiB7GUS(^KE(nm@7g}k zb3?ZRIo|~m2xo#}xP1f-Sl&Yv%GwVJI{i(ld%`Q(5{5tWjng0L@ z#og_je`afXUaXI&YI=3e$+VImv_{ffNp`b&ZZ8tlJSd<@R&U-n>lM7ZPs1Az8+=F9 zv>V3wnSW@WBJnM>7VwjCs@UnDU{sPtlgK-4lGsaczF5Rhzd2@c^Gk^*N~9Jd7E^q; zy57pyS}m{U{EnE@QGCnr(BM7>=)MluJ_Yzs_HU2GdajYG>AD`BZ>e8+k~4VPt6DOj zK1rHXWV=~ei%j3^)$q;Od7h)-8@~>GNdEwZdE-m#eKH94OC{)vX;om6Lq#J+60@%0 zS8!|?&UbJ!ex25TY9EQ($HA=@+UDOv@bASb#kQZRY4>uxpJNkS3!}bMaCbukw#HeR zhWri8cs_XkFS7ABhqO{0-x#1U+T3tp>t;92-;gwx; z@5XBf#a6OIYo}ZYn3(S(ImaPf>|^u*^ZHfYN6enpqn;ITGP@*DNIk&}I3u_`)+ODw zq}qH?h(ZLOAOvKl0AbT<7_YsRMay%UM(q4%*8C;l{Wtyy?eUw#Hr6poWo@eX+GwMn zGX$2;Uzu&BPc*Sth(?82au{y=#G3s+{j%+CG>`Zu_r|>^K!G8K)4?wlyb&(dMTrzf zh!?g101EuY@l3N?{{X=zeld|GyJWxd)IMH8QqSTWpfk7=ocbKs>BsGnC{Ne}QPC)z#}_%_h~(YkCJRFha3_1_f4jcF@G z<)0$e*KXWnf8e{W9<>j%WX579NNgw^ApN!M^St7Srs~2S&S@Pu>{D zcsqw)nI69N^WKrI>iTA*d#l^KF^WQ}SZ)|HHhIXw`W*fh?7k10SJ7-5Cd)0NB z>*{k}6@T{oO;+)>t21pQvj>F6*2>Bp{N7l=&N14)lPt2GMcnxA9gXW%`F^LN-uS0R z)aTGd+G|?R8`>70IWpU0lx5>=E9YqF#jtr(n){#jR+CJ%)Vx7^;f+Oa3BQr%X?Dt7 zln0R@1&IV6MnNR?ug(iwk0$F(o+(-xP^uz~<&+Y{2EhZbBad!t?7!Kb((gvM)#s2y zA(rK4jiPj#GrBy3x-xJTxL|O@7{INubeAGgKFWn%edFxE*<}&F7-;N(s7qE{oab{~ zN!OhA$E|*P`08;t#xD^&a2rx}J%5LPx6|^kyZ-=WyZhPx4e1Sh5hXi;F@_1WOD5bF z_sOr%Ul!^RABz4X7+^oOB*`QYRSS*7zdbN(hDzM)`rPsOrT+l4AJFIQ=P@4+zhh;| z{nG1hGJ1c+n(=e>!+urv-|e<6um1pGkB-)Nkw-Fr!VjrR71S^f5JruFJZ?q-l=TE; zSLb)^?c*&r*TTQCwv9A%nXYa1uP0UJ zu*b(D!6fUx*(Dd<*+N{4V{H{3hC8hirzWarWbJcMj#$nFZAAm7XR^ zU63eL1&+`^KiWUq9(!i^xqEGMHuWsM6&@dxb9;ah!5+9_PE z%&I0(`(N#o{KFvN^Iwlwt12}sCf$_(09hZRQinWg_vQZpBjBHdzaM@%cyGjBE7yDh zYvRpkM$|k}qIi2rveo=Wd1Tr=+PWgf(!yiAmUNa`H+;+jCIA7lwffiap#Iq&1pTWm zz6*R-_>HFc+rsjAX7gV&{{UpgXM8RqzFTCq1v8jXNVt`cnBX129=`_M#W#ljBKXw? z0Z)s64+3$8@}!E+Z{V?BNA4$62|2Dnzql^7q$jAFIoJJfDJe;>-Yq_sPs?v1z-6_08ujYl85 zcORi7Q_^cRLo-O=0twuqhu5I~b(Ddv;gvkXcahJa9@Nh=+}gdh)19aA$n^Z{dIoi5 zbdMl0`SIY4b_7;@zHG?T4hI7Y-Sg?iBE7xL0wo_Su0T`M9V^H@b>oc(!dg?!;ypqg zHYMlJxtcgv<-t6=EN2F^*FP!i zUq_K;6{**C9A4d-!HmU4#cFDJd%)k?7Eo$hb<)Q3pna?e+DSZaQ~Kh*d;O&Cwf_M4 zR&F$RXyk(WJ>_Hz(Pdm@{{Xe?Ub*n=<7dJxR?5#y(R^K`U0uZ*#PF@;Tf?CtfmR^o z9tiAsu2bUFKW_M0t=vMQL=s!LVaopiuIrJ^dP6Xs9bKH};AGwCqu z4$?@z9mqKc3nSaG=zm)I+u|0fZKrA0-XE7uSR}f&DLjFtQg~3UvB3c0yI>x*>_4^* z>~s83(XLUy)_6|Y!*Vhk%jA9|^sC<-t{Pblm?Z4e!hzqXQD36q?aM2!$HrEZN-24t z9!0Ok;VA{2FbRx~r#@$vK{y_VIs7@S-;Z81v%S+}(JZXvw-N-Dy1sA~v(p^z>7Lc= zKMtaf_g{|Pzjz+-WFB~pbAWNs;@*|UlD3D zo1YiSY|Oa`7mf}~0y)9s0N|g>zd-*0V=F0bel+-I>4x^3Oud+>U~s0|Koy5)85vW^ z&*k(OjVU|!J|8fm<&L8z(Vx*D!UUZ>6JcifMULGak6)W$1Ft8iB=h}#h0;7p8vG&f zu8}(BrjLA51CA2$_bxHdP)|JvO8JlBro7iW1+X}ZF$$fz$-@8uCyM$m(crfD1>t9S z%e*e1XJrJHW}Z;su*d+f9Q8QwUzhO(K5Q(}`kq@WPB%G=9}(VN%5AO4PFUlBd-bo7 ze{3JyD@py7bS(p0)%02Qr`K*SZ4k$C9kejKF$q6$S*ZUKlkw2N9-}5pyIwb z@vr<8{{Z&O@qNMB<6Dh3CIwm@HXC~=@J}FZib+7nAM3IYAlK*BwdQWzV|MJY^cXn< z@D&Ba<<1lrJpkvgap_;u_%0F4=u6zAADX}CbK@`=+O_F()ti4iepWp@#lIK+B>YD* zpB-zLy2>%$tEgRE-Xjuw$9Fuj=QzO%IrOeqO}@Xflr@x}WHR#0Y&?vdWMo7Xeqy;p zb0WSt7-7Kd2VefZY1-Uez^qpcGoPD`u5vM7MNWh%=@nz9l_)K2bsy`H@Et}O_X6#l z@JT=9{xzX)=&BHmu0C9J`VX(^SP;uR4(TTWhDisN&$;hbqI@C(Ku|tWjE;XE^^DfK z9Y*|jNhdIt3;-BDxc>kO(2naX$;Yp$$NBo#I}OXMjpQ**5I_WR+w1t%zd2SwWFY|P zPY0h<&T3jZVO@Qpm(S*1$iwck?ZH2o_wR$pYWnNouf<(c!5<5r!6!EQedWwsK_la zSarCP&ec5G#FsKh6736|g6uMP42)+dfz5Q+rwLL@dmotQc&UXRop`sInM*HHR6^lWp}GWy2aF1k;wR#>T<$3RQ>rR zD6YX&46zEw9N^ax;jf5#ori>Pycw(8i>n*!Q4A(gx?4tPNd%D=9J1i;QaH{51B%$t z?2n8jyt6FuTi$N6hzuDA+7Ff-k&s(!bQ}ZR^`%1&sY8-)spjIV`+BNTKWlX_*-PRv z@HdCNN#GlCdmYY~XDs#&Ad2D{%t0I~$iZ9@f(}Dt?+%suCF4sA_&i~)Sn3F2xr+AY zIZ=ouB#GS}PBJnXNI2u(zoO3r{?y(E)h10h#qckOFX3g0V{;_7ZX)>+yyF@eizre= zg9AJ<9M|)~<15K8yjid7Z-XV?qo`b*9m|<+Ws~JS)Nc7taf4qek>YF0RUuhg+a9hu zjWm;kJc-#M)!Dc1^5Bn#pVq%rKVdjAapM`7^@~f4vvnXzaKqbYYPR{>{|%o2b|w{vT8k^MSYQ<&H?kNjT$- z0bev}ax8uiym-}1q*4^)<>V94divMX{xP|lE87cBH{0m&s*LgrtR<9=naIt4Q^a~< zoAWQXWAVKb>%X8oRrVi(OkZK|sOp^flVt z>M`6x(UX9A=jmUR*KJhL`b4fa(#@kSw0yW9%C(|%235cM*{4|rXs{S?7mRyUY)0k- zv92oB+if#wt0P4&302Qw!Tf7l37M^O$DFrL+>BQ`T!ahgD%>|1v}Zo2kIJ~G6IDLv zOBCgh8l5HBt}=N!u5K%!sk@--S}8KRd5U`U=CU+Q>oj|f(#Qr$+n>Z&J00+m=8{EL zVD0l`t~-O)wDg8;(z5>mbaeVxAZL*+_5qB(bM-ae{{V!Ml%A&_@sm@wUxVH;lJv4d zo*&fN*s&@UTR!ZLPftqzc>Wo9U%(zI*6jZP;SbfPw4VwLx@70i5yQ3e5J?+t$QjE4 zoMaPUynnQP^l*O8UlJr{KfCaruFd}dfUncuznj0=UR2TiU9NbCP%_CLli|zT)?LXf zHZhuEEkKiEe`2O%K_}?_en4 z-oZJE5*VeHH(xAo`4>$1wAydPq2v4g2IK7?WBVL%+Dwe8E8$VYIgI2CoE+qV)0+D; z;$MU9yjP%G{9n*5b<2BXiuLZJzqPVDY>!~86vEm?a_mx1n8!26GfL;nQi=t0=QD(5 z1y@CF{QD7yhb$X!U(oso#yrP#ulBy3robk#yn@O`S9Oh|c;>4MhAMY@x1oR*0TrIa6IkdoaoWS%u^X8SnTAAIWH|^yShl- z?8F5^%z`&*%OLt&OR|IESByM6@o!GhZ}fc~tzpxyY=5+M388}W_T_G7w)<4paN0C$ z7T1ZUiB+5wB^a?k0emd|k^Un5F|_el?K$C{2Ts*|dGPnfvDoUGM~JVZ({41$Z!KC| zyUQrxmf{UfU~AbCB3Vd|Sjz&>76;fr1i#>?-wiw)@ZZB<3qNRmYew-twD$IzcCjv@ z9kNfS*vch`P8)PJxVXI3?9wRy-F*~K6i;&mipvy}1o({Wgz)vBHX5Gp@=tc2nsrHh zmt=H%s;X6WN6_u!?*m=xx85i54UVKDww~pz;M49crMF_k%Mee7iJnH>RUHccOOv5Lz@yuCv`i(JCTPPw~w2^jg7R#sS>1R?vNlg}O<`#tzuSnpvveg>ahwfKAEX{_}No0)e;&{|u?Bo?}@oN4Ewy@}(w6Gtjrz+Iz2{#Y;d$BVoP z;n{8M{wC`;w()7Ygu07a=~`Mvdp4b@X^SK>!0RNj&Z>%hyMhl!AVtjd8cqbVVWQ@n;TBreh1Tf+YUWfFrdO}}!w80)6^S0{*N_-CNsd_%sz)Oy2CC>7>lv2s{w%D(NjqVxINr&H9z zW4W4zwLQAxMm{dLO#RpiBh($=)K|*RE|2} z=N`56vd!jr!I=(9S}Nnyaj%kpYyE24{{Z0>x`m{xBu4v6RbYNh8-fAI?Vd^WuX_xa zGXDS)JFWQ~m+Z-^Y7_iExz;>Ut^g=O*S>#fO)bpPCZBGm>f+wo(g$a`n`zvsET6lH zBpcu6WZkr_dp?V4;0eW|cuvFn3H&<^%zBm7?AI{gA`~#(!26B5MI17yl5r%5dXU># z5I$%44WRg@+feaGiY#T8#$7oqV@#FAx9HN#`?6IT$_R@_%65SHW5@^5#9t5mH+N&< zFB;xlXtp}^)~{pY8<`!Y(ykh3xLe4ru291=O|_saA^=N=T#!R{<``E}rR;f?sJ+uY z55yiW)U-`qygtxeBf$i=4RWrR5ZOFWBRtYbM1D*~vmxBB7w(61jQvre{?MPXmY1Sg z>Ao!Zd*b`;R^ABWyt%cywlGa_iWXTS&C;JO+!jozJ4oPH$KMhDHvC!m(9&o=3cI(m z@n(^zMR3~X)y!9S8ia5_$#l}#m`pG|)rokbhA220aM3p-#iQ`Qg0;xkTF|^xV`pP{ z1ormh!ehT!)*L8+1!o z`M2{wPK4E!o~QIr;m;Pq;O~ho7sT6b89ZUGO>N=Zsb5#CMTSpZpWs;J55OH-WFrx^A!HZy4yaY3IbZ8m6~q*BU+C(i5jx++JBr z{hMgAtmYY37m?++RF2`KbyPo~e+&K-=pP6?C*#|Hg`W_uRV|Ir*z~Clw!5brU9n#n z^$6D12f&@Cw}p~j#ubYWHkild`G#3Wo+4F~vToMa_e;*pU+TuN#NXXtmY;`X^1s91 zu;!2B-xk<>K=`}jTdfPlo++9ed#x>eE2!H?bPLJEmzu1z-Ai#al79aAV=>4;I|7y4 z^{2sa_#`Zz5cox@_%p@&z2E#NFMAfJ9k-Of*zjC5Zzzh`;x@RtUGC;>vV|iYE&#XV zAKDL4);tB{ABSEDzPHpYwI8z09oLPOax##X3kMf_WLDR7rqfj1+DKYp3`( z@yo=%AhbGWkK!FK{t`VD$tIDg>Ts&;nh52@t>;Wxp?E@#rZt&KW-+-cqYPix=+9Nm zq_yA9_SfdR9XOh6?CEb$>&QQ34+3c432%H!@KeSbUX!e7Gh8kAhp%r|`$*MVE9Ph~ z9zhkeNA|Uv0C%jRq>*C)DF{!ee`ffr!rv0@Zu~HOMdLpUc&SCj!p&=WHSwNz`x5F4 zs$0sfhJ#|q9B}0pHd*6F#&LW*@x$QHhxNZ0-+XrQu9czQ-d;tn>T7Pc#>Qx&jun`#f{8kle)@ z$|6}V8*&LSn8=VMfosX3LcLi^6%yo&Z=!egeOF8OJoi_YUZ!7*d^hk<#2!Dsiu&H` z!&fV*#cGzb!h`!pHH;*S6^pXE%8mn06;dZn#R@k4zct(bz+V@B5`0Lt@kRHFb&nP5 z6Er5v#8Fw;E~BKi^T?Cww$^b>V$>azbkd2*7WUzGv2sIjci#~8PmG=x)E~#65A`1j zc-O+ZkkkIqBSfCo`VJY_@Ok5?Pubw zO|97?T&0E0mZM~ir)qk0Lv0*ZPkz`%o;d~bM=>tY=5^f~GLHN;EJW+5S65A~?wb5R zuf*BnC(C$C$o&BD{{V>oD*QX}>U=!Z$(zKNi#^@-#1`T0B72r;tu1BSC}s|2^5T%Q z02Ppy2m$LgW$`7b@aCmwq}%@hV)##2OZyA!a^m6bF5`8QJB>XgjiPrfT(m07GAVp; z!2WMq>%Xmdc zMOVfZzf9A|kXggxFNz*0zO>UdiwkeCNo`{=p7vFOJ@+OLHaN_aMIKCN$OmgOoV)Sz z8Kn$H3lk-Wk2U+PZLQaBG}_mCCU#Jt+Gob!wO_;CSHhk$pT>8(nAMKAs9r~S+XPZu zUq!M;w~;bIQ3R2+PCoQ_EETKd-DfYQV&7?)Nx3(Zvvv;M#eQsOt~vDPze&C=d>GQc zA=*!Wp!k`uZhTLw4Q(!^+|evCNVc-=LO@G_Zz`vjNE3)bM#lr;FN%K)d~5KQ#@6S= zk8N>hd#MXMd#gE2k-Gur#z-oxo@_wKokjx<+m(R*XNLHuhANVzBvBwx^c4N^%B-ifm#Qy-<3&wx6zx)$l z;?sNwyPHz+97>^>hKPOc_O zG3TkS$$N=>kCVq?r#vi_b#{;DeW?0@kCRFZ2qS#)s~tddNK-)g%l906aP679ma0~`$Vn*5B^ zto2PZ;Fp8nQMPE->Gey9W(*`*VS9!#vbg1e4U$Gw9Dq*-y>s@^@n((i?^f|=!Ec0D zUM8#-lA|_0*DGb6-);v%xSS7R$KzjK=pVGN z#3ucx{vUir*YB=pxA5Y;D{RKrM1tDknPx~XB+Q7Vog^Xpl-j#igkB@w3quBUr(#uKl2ZuFX6JD~s9%E`S&u)nmWVe=~S{ZW6R1kW6%1$tT zn*E>bbSR|NZoD6<$9ojBiCKw_2$RTARAIN}amZy2Uuk~de-yke@tgL3wAXwgs#;#( zX!`D*FP3B9aAT5XfNcoTNf=oqR&jzGXNwApZq4i1lF}b68IARFCc}cjqRH( zWLs>4sz5t~0l{KBXNueVQa;KqO4o1nKGosZGg14`nLlQ~8l+#N%!j zNtb{JQgC|X8L!O`8BCJ=P}e{s{y(2RI1bO${yF~uJpPt`&eL7pe#d_gyje6C3U971 zU{{tg9CDZ)@X=x2wvbMDZR8vba(^uRMSBF_5jEC=Sn|4*+tf%4U>53_3^2zT$sH@I z*K@z0=6tslxqAyE`V;+=wJ$Hgf3n0V=K?Pg=Nx~_ocQ{TSK8mUf?8_7u+PSuJL`Dv zmQM)Yd1)$cQ7CwfP6zjq@XALafFi#)zh^%fX}TVh`zd&GbdjgLx7VasK!s2}tjIjU zxH5)g!P~T-lpVw={SE!9HQg%D_D1-g+RX9WK-w3Vx@HKOlIGn&B;&4qYN^BBN7UsH z_jJ?pJrnkrTWv4nG?tOaw)R^YV|wisbD} z#H_N6U?1g7YssRMa5x-v`hSgilYLIgb7@%bqLqpSL%4Sbs2!=dvLeF4j~#jUu5KMl zGmsw)M;)q-#nkNKuN-`%9-pmTCS^9w+g(EPD`Dp@4&H;P9_PJ$WAT^a&w@N({w^ul{`5PI=ab8v8`)DV!5UPSA{o;l_ z!8!V5cCTv*iEx$TbIOu&XMq05ejn1jJ@L!qzNKLta9&+_&r62lScjHnL7?fg0Kjp9 zB(gtGyyW+<1NdPSOZzlfq-Y{!@o$Y=_hV-*uiXvbE;+y-m3yb`(XHt>KO4R}NoR8u zj~|Tm<07N06jkuc9^s3@#QV`v!f9WU7<0?X&6z9>u)cR-OU&3z! z>z*a}qjB~{X=L$f^NET%RG;j-84g5>;0^LKagJE$n)$0=@L$4BKjII>*d){;)9!9; zG-<4?#luGgR|HMx8(3jwRgzCBOn_q~bl3;?jyAmm`EohcMivZIqE(AYxD{^HE2pxl=?5i z-{g2TadfBfqtX5-_}<$4;;edCh4s7nEp+V$#>V0+q%wb|r_UjY?!>Rgav+H zGguxf_^bOsT=&Y2#R0GRJO?A>+Ypi4}Gbobe&!Bd-T@lFTyjA&!MOa-2Vbj#)``1E@9cV?zdzoFTQg_(-b5^srZxh^2Z+UAj=+yZ>XpUe5 zZet{JglqwiUOg-HqxM-Ed><8j6Mt^eIA0B28>wS;U?+H;b}ZkR70Y@=iL1Il&{augQ;&9vFwizq9jrli}pI9t+gQr=!}%ccI)* zZbDw_kXcHt7n$U^G0Z^RhiStB#e3)N1#jc;hkvltJ~a4c;%oD$Xv-Xy*4nX#_fKFh z+BXsgfKCfXxPr3`HZzfse>cu>9N1Y|@2BFA(em6+?-rXL>0z#TYS2$DiqC3}!QXm^ z1G&eRP6y#%6o0`-?X<5P{>HvBj>6_SCeqf|R7D|`kJ#>Gb^{=FLc@yWelvKR;-A8Q zhML#Jdkd=%5VnbGZf3ET)=M^&DNz1Y(azyO$tny)oMR(7uO|5C<3AbSd<*fNw~Tb? z*H8Y(xJ$__FMiU{iDUVgTiince=OjTB*_5abgz@h^0y34D)D!;m*COs<`vYbN}F8| z=S@+T>L}EMwEX=P=PvIw=*Y$DmkM?{0l>9oj zm!2B%CXIO{N{t?&s9GkWa>17(Lvc1)j?=?>U{~o+gtfgZNw$v1LWU~{f^Jc52)951 z#xRSF^x`V>kFR`X?mQV zRlV+`Z*dG#zRmk$GNfV3#tShVU>0K8^E zcc|Pxb>orGHT@5N!BnBuv@F^{3vQBgo;`u@{{ZXe zw=|tv(A&BN_4};*5Lt^si-rb?V z!ac2;w0^xVanqAucK93q2(j>M!!}ZC{{R#GU9D@k7O=dA+fI%hHr7T}Bz(kAb_ROl zU~3$IA7wNnZhmh~f8c&64K5$dD?7>C)t}E4ej)KhkYF-Ihk&ZQF5Zf`=brfLD(;cs z&k^{$P>NXWueD>#Mu`-%svq`Qw<=?|Omwg5C-#f|jz45CgP*cjjyy;3*TFUxTE?Ga z65DGR#$7`G-Q>85OUHF|(n_th$oVnBz`)|a8N4^IUFcImCZDd&1-x&#!v~xb9OM)@ zPQKN7HyP8%Qk7Z|vQFCiFY`Ri*Cn17Y7>>)S3Hz_8Sz(!^t+9J#1;a|J5((GXD@@m zA$$E!Kb?F2i{R*dF{8zK_LN4MVwF%LD*1#fugxK5EEo(B2hdj?9kLnjWMW*R`T|XUUuklo)pd7u1gu!g1C=30*zI94suj+l&kzmGs}l zxZttZY~cANnoTJsXDNYz@ggQkJBSDG3ixt=6l)h66~q?kRtV!}eY{{WJ$dKfkHWs# z_>(jj-Y}0$Mg{FOTSa0UKYB&uZy|o^&#!v?zlv_fIe->{I`E9(MitIUItcU)frfRTp!Z7h;R1r$FHw$)oCXn9gVxEBE0If z^)AOxCH1UtnWXZR{{Vd%dHU_g>sPJrZLVw^?9#_JT>ZvY>y9`z!@+rSmjFj80B0lD z6|o)G?oJr?CmF0|PeA25XuQCHKT*@@D$H_}gAIUvYbqtBz)9Q}~&wVxJ4qY;%L7&ZFk@uNyzW8lY(bSUm*fv@yyyFntZS~D_v zd)KEcoQnLZ@Mplk9{6Kz$A|o5b#HYG2C|eZpo#Z#To-(LEfs|3;6Th7&DhBz%Rh45#YZC+&%Uo4QiI*8=?L~@yF{>@(qyk+sX zP4T|5quF?J_52@UJ@wtrtf{AK5?jd;UoT0%xO7Izh?%Vl77WPy&^qqiI_@q=p>JzGa9qUzpJ(P zY15#;=ZSRL=8JPk z>2(~)Z=jLwHZ}xDnHkx*YXs*Lj zk)eiZS(Z_<-6hJ-+tD$#*?wi0z_&jdEOal5?|Y)@8inqiXQkM|Wo+s+>sw2yAw;y( z-r`qygl^(E!w`$L5N|u{>yL-p--Es$d^GVF#7zgpI;5IjscmuL>%BT#OO2Y{;zWY> z)>kcPVtauhUn=TXv_p`05rO%>Jy44)&H)r_h`afMRd6-!=^;@3f{{RH!_;2vO z+ep_w8MUpQs!8FkS3uD`MWMC*&x&jhiKUwU)_o?+O@>&V!I~)L+R?SDDoDj!a}PiG zY5O(!Ewqgn!Tuc4p8Li=4}wi+#;GQidpk#}JcZ_kwFH$O8-3BV`{uSGxeQCon(;4$ zciQK`%`$%!SzFp^%cqO!G#x7WUQJwI$nn_QHKnzt`}uD%6GWF0BgXtFYNR2BSrDwg6(`Y;%^md_EBkj^*^>gtYf>2Pt@;LS=|=eTU%#0a9NY~W}RJH zLnQCD-Dkp7*1JMQIoqot54QHXs_AV;tz_okJ)QkfqXUb2HMQ) z8b^p_^XAfYcp7V_7LnOp?UPh%p>Y+MxQq9e&lGntNb2uja$Wd0_HOvK9*N?sPYrzBnTO%Xb`MBatKi%@8VvSCToT zI5|>)Z2%vjb=kZFs(5!opTkma z9@Fi-ts_lunRQJ{L2qXvGB|zmxkW@xgs_uiGcux+kzYN7sY!0YWw&f+_ZQLJfpt#cb= z%XcO~T0Hr_SY66w+pz6!hXTE8Quu-3%?JA;@@-z;S+)5MoGUuXX0cp{bFqx7Sp&oo z*Um_avxwYp9Q-f)MErHLljF6&f%MCPHoxILBG|#A+}p;DBr70uslg;6VwcH7%(pB_ zaI2Dp8ulA;nzGjS9wQ$Z$3vv}bTn^Sl4cPN#E zf#sdhLno9zMk?PeeH-yV$KMJ56Y4e={{R_Y?^d|i?6mt|DcoEw>)R`{KX~@;(?rmzd{)=9MQ?oa!4&d9*HS1FTS}{!VAYeN{?;D@EOcv2D?K*X$%YuwKjM?n8|761 zb0(}D@!f!9+yh;XjarNMbljJw$DN49sJ&1B)8al4{3iHa`&0Pm#JWGnUx=E=hdd87 zMjajzVJbGM_E{R{Nux+@UeeJZ5xvxr+(OEm*qJuQT>QoHWA<3^N9@h}UTM*MOt#ZL zCHPZN(ypY@6Is&aSJW&nS&|!t33og&TwCo6X{lT05?g#xDo8;4XZF{n@vcJQX6eLR()LsyzXYUxANkK-ZReD0jv49 z4OX>EKGkU_eLCs$zubMjO4{3@^k?j!FM)q&-`Xh~5}M0(xsf@Ycqn^Kcb(_yn_6UK=>iZN;MWLdUubgz=l@;aExtxtBXrI$}s@BJsK z*Gi@1eb3N;jFWsD_}lQ>_rzZZwCyic{@MF2?Tysew~;N}hUL8Ev+*9CBU{a73rNjr z1eWYNt9foo9yJxFrQuHlX`dLrIp1npC&Q`O=o4GtYE!0*Zuax9mbzx6YcgM76^G4e zjGkdP3S(&SR~`-GZ`!}$d;ZMuXkWG6$njK==wE5l^mr_dy_TJ#CZz9S1DLabrPnTegT>mOx_%SsU$jS$eje$*4$!Qg?D%&2(mSJTr}|w#`p%`QUFe<{)FINmKW8it6n8fg#A6~GYl~^s=De{i_e0E* zD}Hvb>ZibU{{V(Y_+jHek3Y0jaW8*eA!Decp+sU3^W{JS(YK>IeIJ=4&W| znY>jcs#*zJ?&)6Q8^(KySw7a1C?#TmWG;S9{e-`0ZyRbaKCw25tWDwX9BcQmcwt;3}{yRSiVsPk9`KAEtEG`a5{ab z5XOr-A-59wkVy|aTW}*TenC~lQ^i6!rnSAAR$RR_=?L~&n3P7r>kD1rqTm(s4c`3VJ^d=c-8K0);~IVm5ahBMTwb#TKfmbdQXBp zLGbq8>*72Uzl9rF)5eQZAh=Pv(ZB@59$=YZDHIxuDp`^htwG(T+r03Ucq;;)CbpNHD7fo*kbEo;I0&btppJN5XZ_;cO#zJJK+1U^jK8&2lzK1hx$$>e@ZX5y@dlcF26YP=qLR;1 zvb=$$W}Go8QM0wh>@z{K|;aMof$@B4UM9a;G_66kSg)k;_2#iG z7dDo+N#rM!r`k2OgIPlm3%yy1eUcY5kzQXG@P;b_(~4=ur+e>fTg$flSss30n9{Fz z7PUSO{{Vu6{{X=ed~e`S3+w(n{g(Vss(eMCLeb*XJWr!)zi7Gf1hPmBmpYw{!5%$M z-f+U;Cgr%=lkGCdo=EYIt7j!=60w1-_$o;VJcZjw7$zf@8h1is*?hJc$D%Zjw0_({0pB<_0VBkL`u~ z7JtD$KWyt29wyRlz8d^(@d1wd%IigwPm@^Gq>9k3)Eb4HShdtwO@Tby%WFv%@T3w; z1Ywm4_xLZxoJ}rF24@E*-^7;MbzYr+1LnAwfM{Z)?Pxw%`JclEi10;WZ;Ys0xtI<{ za=+nF-rikH;wV;V8I`SNP>Z$lH>p2>J!>k)RbR2XJU1~#aWIbJ-^^I<0aLbk(on9c zD&f&tLj@`r2FR&`MM5*y;DT^!|@}-!bv0*qkbAcPyiW);VNkJQ4^h zN2sqW*Y7;7G7)eckWT?(fH43#Ambn$k&X>my3`e;Cgdw8Am<0=`WjYTkzQv!aTO`W za!I4xejof<_^+h=3%JsB{{Rq67K`B_VD=hUlW+EMCZ&4NO3St(C?FPd`_8_CzI40R zZ#7L`d;7-!0FRl3&Y)ofvg2?Zk<%E?Dy^*M-$1pp^2{D`M$#|L1At2*AP^e_<#U{M z7_5|UD)a?l2I2yec{n)d+w`wPnch}A@mYmhjlylaAGluv{2`-wg4^~|@lK@;tn%rR zYf*tcv~sx-!7T9Ykl~cZ>A6AN6rI2Z8}8d&s*t$f+P(xe4qc%x=A zW6dXSPVzUNxXG{0FW4VU)Vw|WA^2v~T3FIXmg~qwwie|R?JNqihH%VTSeD2KI0FMG z^tbku)M1laa@<&1%}or`$|G z&>+b_dtg@2UaA4FFSFIk!lF}xjetP}8tm5R%K3L^oQ!+)ugt4blUkoy2X%A5i!e#H zBXI%$0Agn-NBgW;RIpv$$#M2kblo~j5Q(V5Et6e>!te6g?1QH1!m25|-pR#ZO zkS;y5+PtaLX`5M{ToTv-Sp4#SOB6UwZZvm*=~HMoxJ>M9vz>q*pz~U{ z8iPv`$WL6?J$rt$={HeK>wy^~BO{MrQC;+-p)z(pW&N6Gwuj?q#>p+CKqm2ym${Gt zNpBD7ha`?q%t`u^dz$=o{i5{?d4FbK2t#ZHGAD@qT&|#FmDZ+KCm7Cg`QpD-{{Uv+ z5a>tacgE-QBiRRwG}e*6>c%6`Y!S&MV*tk4!Q&j);-ALP8tJ|!_y^#vS3sA~vAgjn zjHfILkXGYZmV0xREC6Vv3lYpx%!{2aNZQqY&Bg|SN_m>NP*$ls=xqBl5la*@xjeU z;9rRn=ns8sHHeg4U5N(QD&yr?_TiLd-~*1l)=!4~Nd}Fo>T=!3A@iOszFrs&s(3jg zk)CVysZyP!k@66fbiT(maTH1MTT$@+w$mTl;fu^{nHPL<3}-ka=E)fSYqZwC;!wez3anBnW!M7)yF79STJs$@R<^wOy{PK>D+#}XMlFCw#EXNBe)|5>8iPmsGg&|nGV6XC=N*?^vnTbhNd2Pqs3-k}ZM2cU&Azt`(jD6u z&AlK7#~C|;uL1p|*;n>&vaxBULi+EAKh?I@Io2&n=Leje@z8*4)W2y{J@4!-X=`J0 z_Dtr|Qn=b5xWaFcyk=8w~`6Kb+u&o%w7NVxr;d~B+$9O-v( zjCCsPs-OYXo^kyv$o~LnABEl+)ch&pZyDcPG`F_84ASODEfN@_Z@C%yLxYY+2tDi7 zziJlL;{BccMJ|?!b#Nnv2cIXF92=OGw-r|BJ$_IRwN(9}Cz3CKem1=^mT5Fwz*Zxk zN)=)W+DKA!kT(!GQgAEaGJI6&#+{z?fBR|l_~>2@`5q>~y%2D5vl= z_mae`!^?1nSCNJ=PV=8%O8r05{5xR_!7Zw+PRgnR@%I!020CLL4%PX=X{+hhKk!X9 zc`+b|z{_hRAZObrk|QcY0)AHEy*bIR)h`WO=~|-4J+|ajZ#NI}xZVNA(g7JAn~~nU z?j*$2iTIQZ-P)UWZV~$XGBpBpnaNWw4$nGmkPqMcVp?+iem5}-o zU93@9%#j(QM-dFiy;X|y)D{GT?hSm5vpefFdY@HmVa`tH>@Y0(%S9PwQ_kWtM<1nrUwlrt`0?@M zOYu*`*8a@Vba-xHz5f7CxVXBO_8Vqfu?USJjUbL!MU8-#Oej94zK5??b zjn~H4&dbpJ^NMP@*NeA9%&vT4;rsnFQ`42h$m=8uq#jv%u;YM-xyL@Wv++i2DEw_> zb*x)Cfh;2lIAEb(9qh@-U`sCnNXJ$g!LFC%XYA+k3*mo-tUOKQEh@qtK1<&qUFrA6 z=3@H??q-Tu5xlj*CHEeB@@va}F1%7-rQW3=vXt8@4ap?g0a)%P1CTICUI{0)ecW*C z+Cu$qeA7}`dB#gyqwB8)>H3$%PYYRip32#+ZEs>ri9JuzLc$3GExZuj7yi1dvv`r7L5OC|pRM~X!=5Yj0B09W_W(C{=j7TzJ$8ulrXn&RCT%tnuvyoZsy8$MR}w#$9j%?6x^^6Z=O0ql@b|#uxcY;G ziu|@vvGs3Fj_%)6dxS9al_ccz-n1gr?jul$N8?;|y9k|{R`1mG#cPLG!K^0BjIX)a zTVJy<1Fj8jMRZ6F)1`5Cj1c6WK)|ioE(0b>1L;{f+-g=j%X@;#9kMbr&uXwQBi=zg zfE(+Y=i;0kdv~W9e858tHvk`CX(%CF?ycriW^s%Ve_GC-SmKSG{n5@vGTP_;5tELb zkL6tcv#BMHopX5L%CM&5ft5}dB!FCn9CsuRE26Y76p=7?IA4nzMyujqg&rsH_K$eh z6W?jFMHRr@Of2lXWC2RHR~?A@*XO>O;6K`D!?7x8-XpfvRiEWcms6gnZYPU({zb2J zZ;6_}jI<9Dcz?qe5ZmA0M?BW+dwT_=CAgM0`&l!U%u%KaqJmr;E_ttA`X48sLC&;!dmaaI z@jK$>zl^+X;GI56uNTDHD`~pkoDA!3(2bHzZf$LyQtBAG7$*`(wU}d?=sq2O(ASne z1=TdG%|}qv;=YydQ$^G4qmsi>HZmeHk)Vc2(p!8nRxKPVqYg^~OZ<8JJ^1s$J{i+L zX&;DQAHTcM{84oC%pg0Vx`ODHGAk+h;#mjplOt$9FdPx*m-m_%#SaJ1uV^;9oLXOp zq=oKnwCk%Dw9~Ei8C*2*!E*6$`&^M2PnLzU`<^m6KF$*nhpeY4JK5R#o)na#`CC1U z;eU-U^?h3R;%2p|=~g}&@Q#xHJ85o}*HD8`*)5IdkrXPebeDmcK^s4na*E$K%23nz z-zUM}+f(7^g!~iW4;JZd@qbU$fbgb+v9){e5L#VZMGNUu>y~!578-u33q;dCn|BaT zGs_-!&9hg|UMhpYQCfJLR`B?_Tg_6+X)U1EwB&0@V)>pa?P6&OV;)Eil|Tk!IV@}J zUxPmp^!x9Cn(yrS@P|m#^!uNQ7g{#A4xe**EZ1XIMO!wK*HO8a(V*E3DhZ0+gsZn_ zmht(A&98~4g_N<5=`Xt0*ZvD4qfRf{?0mi9pV_O%mwyDMJ}}X5JT0n02uq3e%S(+W zSmj(QGhiNvMg5uVym9+PSbog2UyTF9mY-shO-fmA?r!z_ zk!>V`^2#fN_Dgvk(ba%Ok>Ef{JR)@;5B>*!(q9w&ZuZ_k)I3$=9cl!)&@}0_KN7>K zE}d~Kk=))!$5qrNF_|r(ks$@X&*w9l72Q=U+O0ol{{R*E&rZ4Uw6}ui#(oyp7HuIU zyS6ve>4GU=`#rAg1FD4q?c8H5M>34@Zmg$x%i*%oUfp{C0Kq)y(u0yq zW8wAiM~U=56lmH!n!UcO;VT_N(c52xPjP&gPYRp&D8@PEp4LPb&lr(o7~Z*7t9~8$ zm+-0=+GNzbt}PeRNugmTf-)&74eN1%rTEU&D!CI z$s*SeuIpYM@mqLn_JRGQwC@^fHroEJb)@RocD6PazuH=TjkBlNbg?uRHjuTX(Zz3c zjK$$!G!3(=lkH+DDaCt5b+Su;_y*XRIuDwY*8NZEcVF>e?3r=^yiMS1 z9Z1Em*y$522Eyv#D|wc(%?hJiM;P+154PUyon47wl>9-^Ecq=Ckn<>rvINY_#iL3&eBj@mYPA9W6Tdz#;?9SLGa$w_RH{3!LNiLwHJ)MEv0zR#@-#4@B3#_xwVf` zpH#GViYd$=XNbW*qi)wwJdE+PykS-)N7%l02Oy0MN7gA+=H}8-nB8;x@hEh$PW`ReZW7<(gF$d#4RvSytZZ4YW~t z1{38;r>S9 z%dZfO+r_5GclIEr=GrBe<|c+>%%(+_Mt__4XyI$(;08a3s zySBH`cHLf0VH8TRyf)FolEQ{~82PAWR+?zCfxdD)ALGZyop0eG)3twyz5vzjd|iEg zYkT2$y0&|Zn+uyLq#GiKK0{njZ!ecEu4I*?B*r(XboGB1d|CL7@c#f@(vwl}<@bzr ze~FsQ>pHfXZFzNVqB~lw7U^ueWtuf>G<6#!FvW2pKf=V1zonH_ohVdIZQp+!%f&H$24*W*(Ro0QFcs|!&ZBNNt-A4Z2_QO`Ncs$sQl0$7Qkk0BHFkd0JDzqSC zUCB3ud>!zsTJWZqtm+h<_dnZj7sj<(s`lQZwW{{sQdPSsYM0n6Vvitbsl6#`x3zbv zy|>ohDt4^ei4j5gK6(Cu{P20@KIgvAxvuxcM%vw;e=(;@hr4t!)Xmt^qnYkHb7c*4sopg6NsKkg zv?N|f{2~?-{6J+F3F5tXI=gVknP|}x1?mm?x)B8={_-z|aizGb>g+w>B1lh$LK$c6 zLg1wS;D}>OG(9=Ptfp{iC|0E3Ouy^u9;XGE!>n7xZPACpC`--v?}57{t`t+Ta~Z_UaMA8ew}v`~_{y1XNtSh)X?s3OqY>WO88RN@#Q3A= zdvNO|-A!#~lh6b<=BW-GUN`Je7l}3U_}3)G#2*6*D7SQ3B!WClY8*wIT4ijF2H$Qf zbH3s^3q_|}TB;k&+~K7CO`TG|LfVdk;bCqw{*FVkq5Zma7SQFM8~-8ZBw}4$oC`g; z(67`M>|ZUJdBXx z_5GVz%FSh}lt1c32yK-y^6aTFiH-8EC0RFeJ*gYgsbz_SDlMC% z=nCB(J-3o*oidnwv|W7LxtC}os)h`4PPCRhD&DE0I{8-T8a^j<^VpjD)vF&vis% zHP;1h{Tmy*x4i3Vp69LYxW^YTgxXI&g1QfOJ;(G{wT!i7BCoRUzF4H+KnlO6TOi)eBjh z%ZGqS?ic75n8r{Tjm%fUObdIa z;R6{=9rW)soP*KgZEbBr6jW;ib-fS8&2vOlH(XWsy0ywCh+4R z#=Bzu@P<2$V@;)WKV=Z&TV^(le^>l1%T-_?t^^=4GK7-8J2?#^`k32!L!Q2F;S=Z+ zEV-;J=W|~1NApUt95aY49qcB@?Ng=qyD6}Y70y6XqZ!1B0JU**9eAfL!zttsS7*Jp zK-D6G;G>F@AeA$5KhY|?_atuRr5%-s<=s5}oDsqn-p|Iw#yTeeQ`p2{m4hp}qpDX| z#y$7cN|dPAiz^1jry?VHvi$iy2Dhmxu(z0~O`)eD699VhSQx%@F=nrqDSl9or#oXy zcdqpOSxbb3lNm_SDAGTbJ5j?#F5y_%(H>|D)e};b$0!?PjyD7Vy?U5K6>~6%8sn%t z?d}uNs1>*DS-Spjdf$6oR@;w7MPlrfCS_&lm?HBDUm+ko;RT;dO|#p&*U{~INLwP) zAAnCtfAy^3JU+JoCORSh{mP>*SXx^B9r zJ>aTd5%NWOqZ3f;!veO$vV?<4cq>lLY!dST?O=QjXMPF3XbWP{?yXVN&zVzYRTFW& zRdYsrS3tJt@L>|$KO7Me&<6&=pPmjG*CJNE`;7fVSY}{)vQk(Mrr~6icqPKXqpVC_ zSVgwsti?by>W|)a8d69NvyK49-vjb|D}@p*DHjXhcMBg#$$lxmbe9bn&z%VnfQYA> zat#<$bZ|S8jg1UxVi>pf7i;w!WZa3>Mh@W4C98^o?RgE$adkJ-LAY=$;|O^!WPdM+ zsRMAAQwpuWkf^CeFo@qu7i!6Qvo!5X+#yB@>iG2r{g9k?A6~X36#vEC5)XvZc9Fse zrY_>%BNa2kDB$de zuIu=?!DB6XFCtC@_NI-5v|4BJ#-Nvll%HO>`p@laGi_#x>Vc|AEpX~U)-&cwgw!!1 zjs^VYF793)8h%fkkyK*(&#ZHNwX-{Br=vSfXQ~ROEOJX*lE?3*B+&RLgI8~jFrVAe zD0B2|%ldroCxffKyS?6_ePJMwYw2F{uyvVN(Hw_mt34QAlWiUliU&3fhxrsFJzSXC zz3LBP7v9g-8q1l1&xEe&Rz%kbY;>s$&Qsf}PAQ4~-rYo3&UZ)V3*BYa3J7&E-Otd& ztM>*Hybh2q*?yh2w8afnxh5650sWIp z##=lwKIGBxVDx7=X<{ouBZV?VVO#2_hG&J6>+S*+;KKyt-lc0xFN*q;D)Y2fhl4v7 zWK^#6q3-tGf|mcU#+&`%U{WK&r?S6(12?`EuvRIu!elT}h$IFKSZ(TFeBVr^sYD!1 zs`DS-d*oG&yhzPZ+Z)*d84+Bra;m6?y!i~}j;NbJYKZ8pq7;l9x7rdjL0;y6hXlAI zgh*FR)=dTG-s${9e`1kS)zF|EVva9A1~rG0&~Ho*0+tDf*6BMQlyrUH0+C@QW?_nJ zEc5XW=0H*<#qad+ZDhOolEA zwYRw*3$g$C3-#IT1KN06tKf(pS)o)RsOWM?Oqtj`2HVZo;NodH8W$i~o$3dBrmg=@v`G~VXPuFXjQ5$w`5x8UA43FFaD#rdlx-JHpf>2bRM z2*9slEDaECp0f!Gmk+9RGDf$J2|Kq%#X_E&kjEhEW1WwVjzgyRT3Q zTT!^|sRW2~wO}DZ>GyHEIIBhaSyLqzdbp3a%`v}7KkLe($FSh#15S}H-2C0}t`+Fc zDxoM03&L>SN;!LtT0TI~km?mcJHz7lt~7f4 zr(fvkWFn9V20_XpPNVN1Vh#?q^RJZCtyA~Z3yw3q>6^ngIh$Iv!_|4T#m?!T%V_bdyUMx#Tjy52HOv(>mJcdeP7(zIa_A z@YT~nGLO9c4_vEuEy`I5q5drTN%9BLp;)}9m28q_3@o-k@%#|AtD4=*lxkt!|7OSt zyogJ2jHuSe&M;nq>MSnfX!)XuN32^l%2vsNDZ|gp%rK>W^=r)KjZX@lHu;#<$G*o{ z=_u+;vtE&P9$u1{{b`y;`g+~ee<`L}n#Y7$`OK|H4j?_C6D*$B-bpUae4J&ht z)C!DArH4emK)U~V-#DrJj9L*q9nSR~(m2E(G=fb?WO-P7=<8@KOeT{ zTG8uuhfvGEtoiYdp~KFNhp)wPN1F4a7n!2o-PdYY0Y##J#>TR7<|5{QnPe!c+(MRw zpPGE%AXA-9S?lH;=PlN64;|}9Z(X>uESLTtUvS1zY32fKP>bPbWo6QhWn;8CSDg+k zRAzoHXU&q4B=& zzZ=ME@z2bt5p`?Y+|}6B)!2Y3-VER{l^ml|`Zj=Hvf=%kB}v!i){U^G>8{>qb$=8p zdQr=h+?L*@yBo@~C}>qUf(y5J!@wZHb9nE#c%vIN!)N+vy!%n@PrOjGg_hY9E+1Tv zIMo{>I-dDuo&EkJ)p+KFF{*ms&5g7zI4hPy*bAr!$}ucEEx)YO9B6|SWz4FdzqlZg z&0dJR(<4r4$=CQts#n+EB4+*bDLF28w${x4ry()vp5oGtF@Jtr&akb+jE27S*q(=H z?~Py4twray=ZDt=Rpp(!0H2F}(~`Gpi;rr(u@#bn-yZY~Ngs*8>-Cgu_}ar+aJG8o zl*8d_hp`-=a+BO&+%D@3;*w$y2xvBuu(Nw2?yjHBzl9~eX~|9qWTkqv78 z{_)W+io1pJy&91lOEvgf)%K(2c> z8@q$kclve%{sBol#0pJw=8_B-PHSwbp7g(N8yh|q!QUmU37C*&Sv*|fk?l6gVUHKNRIC+vVvTW4ZT?B&ZSDOA?`Z|^4Xwk3C^hZ!fp`?X>J;o&JN9s;R%7sS81 zru(c@eF$!Y?>ZsURT?{#Up#@V^9{#ad-t+mo?hxk;>MTN_neqtsxueFajpE931p@- zc=fa@bq=d%&11r9#0AB{yU)A0X;}*Ag~mfN6@y==nOy+6s<1$kbp~9Lpa>1Y85ca| zH^Z+VUNt7F!emd6SM$}nJ-s&$ahwDL6L?0DXHn1(G*ZK@dS)%c`e^+J*Z#(g8 zufzxpH+IO{?53_?qWx?kK2x#z&NK3$Lh1G|M|gqDC3%GGXGs8!k9l0Zrs{0kr<5`N z2aLm;mhPw=i-!3QrR{x)IG80>{9|Oi^^}Kd0LxYNe$!!5Q{AMV%B;Vlt(Jc`N&(Yo zF@|F8_%v@mpV{4|8fZKVBs-~^wV0vXAG5REe?Y@)HQ++G+}pDe5R}7{C=A}!vWpF+ z09}7)gk@TBzFVbiVI(XORgm57ib7_5Dn2|@A}A%Hc1?Q=611bLWULj~w8O)#dRZz1 z!_?SQEIaZLX^L{9#>4vmS82L{zbVp~C?{8RJwu*d;xqdO*#zt_UmRL~N8K3f{*6<6 zWp3R%J68jRw1uS%f4G-8^FLNAJ-ET~0A-1_%B=y*aOtO!?7*NP7QTUU_Q#CVK>uRhSvPQn|u95a%o4lWjE}8GHxV?FpGUF))8^{ z!q5bQFIVX|6jtEk`Mxvj`^hpXBkt?L53RlyCO;hihauvSrDq27@oJ-gd-Z9rLi-**g;Qt%e}b__+IM z8OhO8fY-q+6{Se45^M{uoM=lQ@YlbqF>1|Vn9m#UBQSdUjkaf-O-w5uDNZY;Pu=dF zkuh^)0s0*v`5aQ#vSf10F+x$ANr$gFA|vm)cu_MlB)vFFB1#jIh3G(!H$KcM$>14g zZ+NWB_78K%*BZNP{j@4JzQ@Ydi$JXRIwAyQB&nFYWD`4ZH7uc$XM6qry?}?7_z*7{ zLadn-n;5EDjhl+KFU8xo+FK{h{P>dX_uQ|JttiHy>N_W%ce01)Wo1ecisF_rt3b=1W3B`?ilf_h7ys{*PTTGeGIy)^{rR`eu zeo0yON&%UsBUl{LCpnn4R41bpIoG*xtTXEVHW>S(IeCRm@B=2>^!6ZWSfc{L@vYKv zF6zVjAZe)b4AAkEuP?^K=hmGP>ZiZR_w;u?Y2XxjKAqKcOVUT3&G2PBHB^#HT0c(q z0V0IH4vL90HOI)VjI<;w$1RIR{C)M zkM`AyM`@)Ca!@rum{QHZKOu(5jE7wIepNGzuTZns2YtV;rO;4qFt;Fa$Gpfd_6$XK z_bqs7?Ti9q1nrz?3!eOfPyWn}cXP%6s6PCGjqhA{%d)ui!%_j4z1H@prS+Y&HL53X zLn*C-p~ITD{pbmlQ2~Pdb3UnW)59GJI`hcU5cHeomUpkV9zR%0+{*`VkjFCR2J)x5 zP>`(T5XSKk2-}f-_3-?t%;r7BcXACeBc8i%)fl0QJ5xMGvScIx`)@s;k<)%Y=YPdg zL9CwK65<)Q={dJo!&h)uda){rW**l+;Y>XvC+px9BdL!|c>U-we>B&3SYif>gTKnV zftiKe#gZOzm<-91Wk@O7rB))N0Nu z0t3WpwC@Xo`LP1v=7i@>l8_pC5M(A6UlJ{<}HOvD`PP`sdxa zk;S7sMt=wPS~sqHX6uO4HG7c+D}FHE>;4!*g5^Lu9&6!}1dYS6K>I?A-c92G(sfKH#wbIh+r@b>s0f^z7xB?Syq%heyb85&Fzeg2>Gb61I~xV)aC> z>3D&wUPbsvN`Pvop=q))@{o8S3DbE_K#{3r+F8nQnXQF(M?EJo%eT6xqM<-!t<+ME zWR2fD^>{^t#lJ~=W|Ung8^#=u`^z7A5q@Wx|5tktNtjTG3Uo<86HG6y3<&zf<%o!9 z3_0b=|E=5Q-r)$eqipeIoRqkZyH&cM0#e+wVyqE!FeR|PU#gmG8IqwednKV}TrTVR znty42S=C5K;y1u_@jOqYkf;QSF%L^(fhP9wt||loQ^G`i+h4=O082X;Dn-Y+3u{22 z#>TeDm!qzAp9HsjP6nQ<>QRa1vKHM5>b)i_GSp5IW0c{%CePov`=Lm9K;I_R*UhF- z5OaKCr=zRQh-)x^{#>a&dZ{QpIS!!t5!M|m4l7H2eDnzl5KfUpARyF*k$Fxj5pHIYg;E<|<5aq+w zDm={WW@I65A_99~k9>T)ujn2l9_P+^p)`47(^t`D9o+4ez_E~3iD6i+K4oFL!8YP=_ zzBaW;L2(6;=1+)$ep^b&I@)*uxm;$E(L8o--^*>XKs!WP|MKS&m%M`W`_CZp=EZyR zZO;y4@z$j2P<$B6T^10WK#VAe31QI*_G zwo4Re4?prUZ=>-ah%FGu9QGN3ei`>o;$aO?mU$bebZeOYK1xP@taHh@FEH}5lk9z4 znr>6%(r4pgJq=At{{20E0z=9i7$6XXf2)fr98i>*>Co5`qCW6vV;^jaDsf_R&gPtH zh;w?A025!8eNx3UO9$tze%v`ynzO5En}!@h*L>2=P3`}3!@05f6U2i+n0j_o1@?8TzOiWr8+b=T(i9> zYCq!?c&bLE#MR$kvD3wk9PV|;eJRE+T+sP{cr-3X@QD2CN>k@OP-oN?(&X=&V9wol z8-Wa7j>Z_b85t7Fon8~u>1L5T5NdC5{S%-O(3cmqQPZ@j$TZW%<`P)(0h%_@mYbP& z;3uhrT2e+kai#h=v^EFD*d_{xG1ATJtB5|vByv?Bxy+WU1`5It;Ck1GXL147`hdCqk}}tJN-pS;9On$66#3 zG5_^5zv}6d?qjSRrWAL26J*iVD6!#CLN{LvuZc;MX`Ai_Q~8$}H{%xtOLa@8W|NP1 zy1s`#ojA{F&(gDlb7ZI(6C-z-uvKjtq&sWDn+Y5iR+jqQnxeN0DeECRlFjn1w@ zM(C4)k-8$!4{xT#8dA-*xEex?N7;ogPLKAG>zkoR3HC*p<}HEz#|OXpB$Yd_?bvz!{9oJ>JZ+(G!_C zJdn?gT&t#yH*-SqFQAdD;!G>w_|~32H=T!-e{uJd=41CtWkG*6bURq3B}mLCEhkb#J%a*O71e-AyLGx> z2%*7%5lIp41{z56sNGD1S&kqL0YS&QmcJS3?BKyRZSN9YdBFs6q(!CL%j=ZnBW+mW z#yD>grl4|FX_FMFJH*hu zWY7N)%yE~dFM*>)o{ufPV-m`#nS%xSR;mPlLu`$<&5A=|TBSIEa9+(1H}bE4%FGV3 z+qnx~t~@!B>UC!x5M+FkHKi!7NQ0Zr;R?7MtJ{`Z*sRXN5hqjn$&a)BOooWJ$^31> zrOBzvoa5-PyKQ_PiauY(?aEQdp<8d)e@*uA6+{DQIuvU0ZF4{*`+n}$Xu|A?jcen~ zDT~ellxYkT=of=+(!{y)SI~2%!MJ`cdiu|0HEVHur76WAr_58Iw|Bg4Q>8d^N;UKQ z`G#Y$ep&yB!IWvpUiOWEc)w82cE`Jwjz17_l%~FF^EIc{vY*=ukZAkYU3CX-B{Y8& zcPQg88Y~Rnr{YrkbQ5I+*LF{Q8k(EKb@}*H204iy_KVlGP$oobf0Mbj9RN}(3gD*a zs*1E2^R2~W?UKL$?g=r@o0v;hvA%5W2G7R3@eV%8Z0(+>P01Q=5xBt+kK=orIF`qt zP(!2*`GZx=^K*0=Mj+t(0ULMhiGbODf^}wQ$fcs(4lO^rv3InJ2#o-+$crFA%OI2P&`7vZinVVIE_`FqM@XM@}g$sFL_xDqpvlRf7?$^+<=alabnLv`{q~}2hx8K zo{?7mCez5+6ZA$Ov1U(BQ`z-Ta`82T8{Wp57|24W>VR);^7lmmd-de^|K_nNNP=nv zIXTo|DASS`Xab;-uLX{yy%>3`B5Cz{MKxrLybvAjDm9A3L@2_|61F ziLxDDCsCv|bH0)KbmVjT5ZVU6)m`uj)-uUH_z!Q0dTAhKrN2~7{F&XHeG64*OWG)# z!)~h4QfCJTj=*1g0i<>U5)1iGJ~15?nRBl?fjNKYPTOpEuK5X4nYB;8OFXz56D^Vw zfutI#q#$g=LZG;1O0O=Z29Dh!n|Yj9is_T9HXCQw$zFp{jhamR7ONKsd6rYi=Y)N7 zIQx<6oU3Z>Yn^cq&eyoCNSN19E4eAQ>n1N)NkLy6wx}szNs|Blg?7T^X8<0BaARdq zQQ?=>)S3<2HVbGl@vXAJ$RJB?2vYne3q*-{Rh4L>FcECkgksd}<_l z+;acDFw(Z`yZ2@Rbw6uDS7Yg*!}RNHw^C0xoW$^O%Ig6m+j|58>Avef;W3Zf@Pat2w#VFHuYU+JAH4Vtp#5n^<&ktWa z0|4aVQc?H)eh>*pt5uQgvl$tx)ma@4ZYOGerzshJ4%d;QJJMTioOWhCy4xX?%t_=R zrULvQ9vNc3k?7(MIMC)-`l5{w;%{HP>>TpuHC+4Shm(&S^+PA@8Tz+C85jt~9g2(8 zm>p!JM9~hT7c#+~o8W4?-4p*LRpeb>a+Hr>rj(YyySzcg0OXa^l&fRHss&-!{ckpo zyP`&N7H1H3L)@yI(Jb~su}d3RsaB?OW_#w6p(H4ptCg<)EVF*4h`ph62Buz zTnTS}R9&XuDnh$AGm>~B6)AD;jw@rfUp8+gkoB%R2ik1*sl&J{`{ZD{Q1}nE+1Q&W zER#$nk{Nl6S_nYAyv!{>%p6k)LB8zMQqUb(6T*dhvQCs*)wDTOrP1I}0CP22KN4qb z*qjg%Z|!$l8mV4SbFa_HWnm>y)ctErm^x&Rn#LtKPZP)ho^cU#MptMpaeazCK5~04 zf!M!pZ1RHD4+$T){F2(cz6TN!mFkeb_G_-)YaThQJDBy?Ro(?MLPs|)_o89ch5y>n z-ND7su8Ca*({D|!O}IcAs2Lw$+;Bsxw9OZBHtm%-?$NKdF%M4wzy&tNyW$X9tOru4 zUpGp2?;}hO!D>55Kk84L8S^9TR5_N9*hg$VYeNs(tUM0=zRZ}CEUNj{*n0avyoTn; z5K^4<7jI5c82hxnqhadUb!}l?102m$pMk4gMRn`@S4PP(&;RngF@Ix9@|`xb1bf{T z7aWR{lpJTImn}PYk)!ZkX8+y1hexLKw+T$OG{(IPhl~*`^aW>Vmr&B|jD36Y$6<2B zTG%E1h3$U(!R)nraDG8Xmr9>#pcbx;y1Orx_Jsawj_sgF6hwxQv>g%Dp(vwLg~8#d zlcS5@h^e)iU^iBqg#q~#xcRFY#nLt&0N7OIp?yS)6h%aBO}DFiOj~pBS0CL^z($^r zSaK9!uWUjgCTzH$`A6e|z@?1od}9!}@&zKu$?+QP8vc0u$&ez&^?EF>wOTNvHKU=K zed{!2aU1R=hq6)wofCNG|QrHtX?suLJ(JaZoi5gd*CACD*3{GbU<}cI}W55+G~nCK%}~xi^DOl&=tZ3fl0?T9%tDkD(*LDaKKa#3ZJWYfiN;9YBOp8_t(Sa*g*Q z$1iToR;}J9J4o-o6mFlx*~?Tp6n=dIE#?wsS|9Gvl1mfJrAijRAy zMVP7{tJ#_X9zw%4TY{POXHQ*An?8&6%0E~#j3RE5ba1_R_ic}M(FynDROEvxB8`3_ zYy0z`&HF&`P1@?)=PsHu%o&PXX8~isQo8iy#a@juva-pDG>AC$y(r&wj}HMN-~akk zr0^>)VxEn1|smGY8{*q`l z_dK8P*hoY2SdY8gV9Uv>d>S&9_V!Pl{!D{1cMex^@fW`rHLRM!xmBMx!k+H!P8^jD zdnRi0EzhNy;On)-|Jl*_96%#UCQ|3!!;?rLW?70H5hBKD4K$hE7lQ?h(3@EgoCZ?b zb5CNHpzM2$Jj^+SgtUi#Y@%})aktgL9zPMa%|lGziu0tc!&v>p2if#a#SgGQmCq`l z_4eg9dfWW;`dn*k>ap!fmz~k6n;@_3Sngs*6T{!-_T$@^W~G)e+rML9A8Q`FsRrUG z1!=odzxg+s5-y2@2<+KDIKeKX+$h3E^Su-_hDrF2fD|y^skzZo%xk_sN6TE=BjhrM zpN#a2KI-&;I?o$ZRcCd(XJ@CsWew=zX)c5Ysm}>or&oW^ zXB5=jv9)E0uZSRJ#&2;OI(aqx);1fRpp4c=fg;2op#(<$uA6t-loV)5~s zG?ezaR!8%UO)( z#TXp$d(>FXAq9%`dVjC_TUls#$aS1-VX=apFpDwXK$qQ&Cv zv&HnFm@RvGan{NkHViIE;1K~%x$2B%Ok6u3zEQ120SzNXlW;HUe%9QNsed7oJ?3!P z_zgzGWjy;B##*8+cv^ky_{YrvUT-5mtnOCfrhDYQ3>ZXq&kU`^1&+kH!8lmzoS7sm zj;$5$))r!qm16*7DsPM4?lae|nY>KaU4id?{twT2GO2(>2q*X3QI> zLH1ld&jBQ0SF-DDE2UdTj#J#h{}|{E`j-Sjj23Z>wt5g2HfydoqHwN#a4R8}w0o$_ zs7Z^6yS@EBB*d|f1x?wwzrKeXCjN&fl_2yQ)i^w0VLK3b6(v#0PzwwAc1S7tz1Y$$ z&C$`ODOG*Vhp2&L$h4jSpZ65>Y6)7~dmi#vR-xjG&Cg8DvYWtC@sctNN zQ-1GTxVusmefFv+>2-C|obg3m@a>b3q2>cWsF}H;hZ5b@mAWwQ5!ZiF^ZGx$d!pgn z7i%(LiG?bB|3r);`HVV_0fF(1)722_zJIXDCe)`e3lhTcoR79-0!1c2Py`46(pTso zB_De)DYjyKa(9JG7TQqQ%ACO()ZoB!88dqS;r*CfB(1=`l3vL3c`}FEVqayE={%mu zt)A@86ul2?1>tS|KBo3T4~ZC-dN`Yl*uw4m=wN-4pTNzE)-?1y~}F_gLK9TB8*ILZ95Z zWD?n0cL1M-x@~e*lyw8nfO+m=W`5!JsL3tbepSO7tf`HngM$&iTD>2$NwSn!v(kiT zE*@dejcRK!vU{-LDNJgER8A*s(r`|Kfjee8DwE}AET!%<$IGy?B9d*A7lWZXOgvc1 zH1q&6Gyw)efWg|-@1h0*^O}A#F2({kBDgdC9*(D2fe>)iZ{^ZiVk3USW${;L7@2F^9f+O4=>CWI zm><(^i&C~kHST>D$DKJ1#87iq75~Y2^6YrLmQx>*886B5Fi5&?|`( z`Pu-Zhj}K;m8n%^P56Zu*2>xC_0po4!fUYcd4GgM#!xCXA?gCoRmv+xKzDYwDsAsq&|N zswq%-k3VbD+pFi@I^B-AATIEhZiG)#T%b@qR-t-7lZ|I))5)Q0)}3AGx%k9;@u`xn z41A~CZ=XJjfVdAr39&Y7FM?m6YMs5w2vcaDYI`PX^|SY#zo0NV>$$l>u9eKjc8&%> z3?|Xg42osa9D9V!a(MWjorjW2${nueChVntlgNiyiz-n@5abVW#%jR*Af?Pd)!rfl z?=33-ty1@xl}2FAX(!VeMT#nDEu`q|;h(*)w->)++WVOn>H@FmKbd`K1KSp^w&ky+glp8bTRtSx))lw^2POa#o1WoXy(Ls)d#P=o~Sb z#6t}cH%r6B?@w%qwqaQla%|9AQ-pqG>F}2D_}0E|-Hq<;tFdb2qV6Qa;sy83bU`XRoC_d--5f z2q7&O9%vUx!MCI+muG&@==dgbu1KX3!Y$>fjRC+t%KB1ePE&)7e>?n&`%cqd5c;v? z?uRzT{j9bOM#~>UaCz2P{j7P5akUbD+Iu*Q&F;Q6HXkNDD&5*4Ze3REy)s$v&UXU<%jK6CB z~3sRkb>JG{vAlj`t?!Vj)<@4%O$T6BUc718l_B_e-W$ft~V1k(_CitA{)DabM8 z#4&PVv7heKM)Vlt!p*YZe3HON<3}m+TWL73HbjcL!>Ps4i!4~}vQ{qISz|&!bP_mq zFuFs_X1GKJXXHqf&hn2xzG6JJApX!d%lq<)-lt322*5~BWY6%Sex(TeHWistZAztz zg07M1KOdFU*ExaV?cvJpcBwi65?}068NT3YFwokUGJEOnnY_*xPpjv1q3cr}`z*{O z|Irg9eLvGy-5sS!jY%62J<_xPaUd#I2glQLCi)}IdiXD{6taPzyEq)$tWYJjUY|7Z zJ_jfiB|-Q>jCUnjjJ!pOJy}$SM4CNJ>FTVk)aa?XWVhI#At{IH^Tp%O?9)J73rXT3 zf+w{w&RDGi{tAm_kzTxX2IaK<-?hI<)zArZUA$OZWW&#YQ%>2INfip09d&;MMx{T? z+y93*Upk@}gU2f~ERTzw&B5eIya}9nRL=DY?NU`&J`9bu z+IW$Xn_=G-^#6xPA|RpgjGv%s>PMS}u%z4?N9rk4y=+Vz?FKfS5s$_8hP=Ni1H|1n zfzyaM(v|bSh5WtIMc1PC`rWDjnejwniK@(87AX$gJw0j5*U`|ouEoXd)(q_r8aCT< zJ-{0U@dN_8W^?@3>CeWiQ#(>+-s?{+$8bnn6gkFT;B2tYmZZ}{?CnT}g;GxluX?tC z4WGBTUCPL)+iS|nl8pe)3pV4eeQ~f{p39YSGd-?3I_Po)^>fD3*n+Pi$sw-CR%!}? z#;a5JJW}QWAVJ#|P)D5a159OF@X1nh|JlcI8GZYY6#ia{8zcK zQuZ|&5gu`SOx>qR?q~7-wQr}~-R&`T&nIH7!j+qzYNcCQ8r`#g<=_j$-}^KObwB3i z&-{hS()@X5tn`VYyq8G4KCNAPUubl_26~v!CY4eA?p*O5a#Y&15oM)S@8Kwf=5sj;Dmal4GsaNzOdNaSQzeU0YHJ<|>`hq^sVoEfs_^wPYd66`sc zdBzp#_bGjn@qq7A49!2)0HlD@@jhzb{W)K=V5os_)v7`5RqjfWFtw52QZ}{&m1ENZ zX?OH_9d-Y{;oI0P^(_gRoMiFW*Yfl&2WM7_msNRtmRY$EV#y18@HE~HXD2rKH@tRx<8YF!CK#voKI-G!QN$W!`v{c87{2f z7bSv9-fgNY6;Cq4*whW!S^U|J4NzXvZuZFe+^K>!+!Fr-ohqO$y?ElmB%5WRUsWh$ zNxIZABXUEVDkq{2MYGSx&8TP7sC&z3e#QQUYy6dEx!ei|**RA^^lkh7Mh}6vk{(3kLImZVPEdbn z*cgx_?vEg==mgVP9L}qSC1~+4T11;szd-TyBp0b>b{)UZC7x-2%`x9Q2g}Z`p*3VK z*B1?CYpFgA0Q$5XyxIer?XnTgcL$}jP>C$m{(*`g(>5~V%PJFV8q#~`#g2cJtHa0h zmq2j&+g9fYuhHdP2%d}>r`S51Nd!;~ncAw_o|Ak^$-dfxju6s)K9D&N-pH}%+`x9M9R7nARRn%FKBeZM=L*aY*batXsc(3CBhBU_prL!{xx znIV~|+xQ`p|KYReqV(JHw@K5wv9DWbEJ7KLHQ;{3QF0UxEF|z!0E;If7ezGE$G~i6WONID{0H z---}W?)P>QJ}5dcqOyGP$*MkUEJY`U_i5$x@X?5#e~*3t!+Y96iIVHjC)bTvWS9{= z6@a*nEuKHI36inzE&$gUsyJCc3AAs=e!i z7?{j+buE;1OQ!Dgaa2iWL%NrYU8fltOH!?D2uaDml=aj~ZQ$NZpLr%xTuqq}N8jS$ z)D@>x{wrROBhb;=az#$?kLg(-L4O9ut&}%k`7FgNM~#=QhaZ1G0M%zhZ9+*8@5wOB z!QjoN>4xRe?njt9y}e+9`+u6hmKxs4i@Nwp#V^sF@oA9ssgX0wt2em$0x0ii?@5u9 zN4VNganVhLp&w`){~{SRaBmpTq#Nx9VKe`?W@1|Osip*Ha`@2^DrBYV%!tj$X%ZHP zOc1m3nzkKvvsew5VHF(aqS|%)eG63G^#QYL8ZI|$Xa^wpzUjd&M=wr}<>MY{$=SM~ zNOE{J^&$G3QERGnO;Ky==A&lG1V`6Yyo2Airi{2`3wB?jrVdw#EM+Z=if)W6wwacn zy6n_;HcSS1>dN?|M?ZJi zaha#ucRqMyfi5mzGT)bF$*5DUbua#hd?~v+zVlr5Ovxk8$tG(`+#oCe^&zNoO$-;L zO+O^No3eGvGA5w_81Oe12wAr7^8ebQ7Y5m)_-)6WPWgoRg1t7+KC-)Ih>k}jMrTvwLbZBrD@uSQIwfH z)ZenRaatVJAg4riZ#GAD$ie`dr%|#NgKObi+Qd32O)#E@cg1RN;8|40#cRtKdfY~- zT+VvPQm@MGg4%8uP>k&3y`%Mw0x}GhfUQQH&I}wJEpGsErjKPKb+Ds1k}@#ie7CaP zbLJLi`dkBDAgJmv5%O{x=8OPaXsNX$C!?!;a)cg=`?dLfaxYhNgGY?*s&#iYH^tC^ z_-82=?dXJssVJnf_>vxf+6~_}YP>%$Wqi`kgS(R>>uO+JNL@{g0Rns{r#mOGSp-e?2pzi_3@4^(MhYgMlZbhyM@nnHkF)#jzjLmFwnd&IsPiff8a(k7abE&s=ev{ z6Ph8t9X%|M{C(5kGus#LsSnwM*=aE~uvC{Q;9?EnZ8GDkjLa zjYbx1KEj!LfA}_4PP%_(vPpHOC-VuLI&6wlZ^!y)>i2+!#J$jkeN7^ze<5!$nOlTO z?)*RI6xm&_GAS(Mar93lz6vO+@`T(&al{9e1Q;#Zn9Dp3j0>JNfR>8Ep7pqur zKm|jmeaCg*2Ad6`s<*Os`bQ3* z)QURB&y=~|>x@`CMT!0YQFPW}O}%{>M@0eYZctL`2B`^3gD6PDCIZ5kbdB16(%m5- zOiDnyyGuYqx?@P!Mh+PB-TOD!buKuZ^PK1T-uL~v)8yBp@@5Ltgt?Rq)lu7#A8LWG zq*r;SMHFMD3rr`~{Jqn3N=VdVBoat7@q+XsExxjbl1>Y$zu38fh)Et;1Vwq?E_Kpi z47CCc_3O~UgtE!KO9FG^zbP6n+!|s@S{8oW^bsRZDwXlGF3}ioAn)83nn%nOd0u{dxUf{>;XQx z+!SIbwY+#{#Z?Jb$4UE!IEb2J59uRNT2es3CW&{qR60W{;`0efepu#`fCC%KU01fgUD zLb_*(e*vf>xjZDbs*8U9h)*urS*09Z_sk$EmH0~UV$V6pDIjq!-MYOaUWbzrjAy_~ zmv*w_<@Jr+Df}k_D>l_Jn!hHK?!KyP!qNs|0a>*@8TSx45?|!eqpuzBc>aVWG`J@m5g`*lRf;u9t1ZH*CNUZeSCl;E6WO6j6(;~M7{%cj& zmb5K-615wW+Wu?p$I<7;g=0u$-z`JF=o#SOET-u4E1n1h+7M*~)8d92bs+02HWCMZ zZ=%W}d8Srr{${pSLHo>HGL7$E^JWa!4cC0VS`PtwAQ7?3(%n8qxG^{(_FWNh|0b}o zerwu~z0Ulcv611=!&=-@Tl%lG*;C?$kpJ{(zT48|2mJ@#PA|byr}r-MHpW~8vS5L~ zZbD`=8S0&gYo`nEmV*|%3v(tFe6kG!mb+tO@!i+5UukjF5IBcUakMB~Wh7Qj_#EO@ zMCxlZ^C;VeP}w%B=DNey;dPMUe@A5V;<-IxoKbdd)9w8}&XpD`IR3?LQ!+K8Ta4UI zDKmgf5#)lF)Skl8L(jU>8*qYC^~6ux;0@)oMGu!7IsHiZpcCB@!-LMJ5jMO36`NlQ ziqfuN3^$D%Z&jhJg94d5VfKwvQgyQ*m)!i)_i6&)eKa**UHJVqA|i>WvB^Y`VI_$J zvU$}jN{)YLK{3X98=W)`e%0w(Ig7ZZcTJ6<+hkc?ZQ^B(+Oj;hR8=|J+Klw4;X!Ry zWgydk1U-V_VoDMy2-w6TlaCbq!@X$XyJ|W4%C(n-%r-@$`ANUCZC0v=VS}J(z)6Pl zKY~)2`g^AWJ7E76}UXOUCM; z)$03kv2g!K$=~-ntVZh*+5WtD*trfcyyn&Rur2p*au04zBmI>5J8KL-sywaSIDvq!?|mUmgga*HhY{r=uK zl~m2dFMdusV7Q5H$V*t}Z8kV?t||-DVx)drjuZ`9RX~MDi&RYMPRxkpL+eHkNQ3=_ zZd?;0W+6kF&x_vXgYBKhG=4{aGciQX;U{oR9puoCIV4}Q<5eSrZkUeiDQCR~bbbGy;5vA0&@wxz^TDR zF>99@m~ekE<-narU(U1WH+9MSjfwnCpJYa>>RWXbr)XXH`fXUU%U*sg_!577uyJyz zc54rfHg#cnWo@(W(qK07;mFtH^3g5(v>@Rh;Q|0S(b%->>;a*-Vseg#PtpHn8Mu8Q zm)sA`_~!f6+y`Rdv;_5^kCfjjttZB_*^dy;=5TJ&OZ*n-H4aK`07ELSTyCmggfbeB zNi-(nJknXxV>nfmDW^WpmPr{Fv%k7k)<}ccKy6O()woM_Z$2uUC9^KPnH3_q4)o0d8 zF!O5Wn?`;jQi!dkZ#W3XRo?H&o={fC7;bc=wNQ59ZRJqsN%01$s=9`$&9_8ss$Z%x zb9*V((mpXlz9-K@>XUIDXBZX?ekVux6@TesEVKsi66wXClybq`p%(3qJ;F3gY}m2I z)c%57dF!;5D&<{g$(vTx8jigD?q_A54xahnA%kM`Ul6Z1_=?|u7^!5xFqQ_!t3~qBzdqhQ z4%SJU$06=KdzX=18jHXFG*PyZ7ZI{78o9-Kc$oPTy1oZ@hQf8mZ>F(G4EsNVM{}?) z7_sJsel!e#2E@fI*-|YiHN@M0+j-7g_UQn8^$k5(4oio=!jY)p|H@z#3_+2_*qzO4 za!NG9SIaw4B-+Z@oOr75G205dZ{?V{Tm7n7-KlU$nB>Xz=Qn2VBzWd}aMXoAv#E?l zA~34N%g6ubtG{2OcTDE_)Tv@6jy(f^UHCL;qy38~#?PP!Zt2Wuh<(^_Ykhp0#54nh#WV`&C!QR;gQAD}y$@}CB`|ie0 z)T>GiPiMH)(hMeaeU0A1hxL_N0SEuAk5#F?tLAn~VAh)lY1giS(DKfO;7i53IJiyV z$AkZWcdkNVL*@D%auT3GHI8c~;{Hd#W?ZQ^DZTc1SlkV{BEto5OSdQ5e1A z&o|B+U%`I_xm>dnni;*#N8UiL)$Y`eyw+#Zog$RtSq~2Sec#}@SL&8a5P8r1${`;{ zK-S|diz#VLIhmjTpnT##0-+!H{na^xG(uqhZ;FTky0K4Ib5)M@%iwGqTx;&+!b(LQ zx-#~BZj}-(Qt7poWroff*54Ow{DZw0n5WKS%0u2gWX!HyKlOP3@hov~;W#9{-v`|g zUjce!d&cD?e&uDz@Tu{|F3i~TV$bRxI>qUX8}|Zd^w&H_))1u)tUVj{st<``zwaq$x9=Tnj!+djb+7Ejc-+kOiz1y=Uhp9Lrpb}FZ5z=r&*fjH z(oW-zc;i|1`PWmxxML?}AJpS^5D1XRDBx&G$8yIvv{x0U-@6Y!n5Nu`qU-(u@;HA) z87Us#og8n22OF0cEAxHsfeOT{Px~KxmMZwg8Z&vgt z3o$8Irvu?2{HlYgg3+czJ}wa1`dIftFWwU;4t{OMOQU3`ds# z2JMgdvrT!?m^eZzU+ADM*8F>=k z!8Gl3{3k!dZn7NaWgR3dBnEJVYqg%w18J%mmxX#n-Cf^X6Z*dKS8~U%yPVZaUV3}^PjVcFnPYf_#;IR z@oYFk6o&Iw)p3nnkso!UJ8eCcGd|KjqnjB!(ez%Yyj*C^N61}F93_)Dr~<0sq{6zU z16iXkHF!OvLz`@cG-b|GgLagv9`JptM#^{u42v+f;%e5q_PhD#Fq-4z99B_k23&_oGjd=#>J~@8r;bPteXd z+jM{FNEct&asd3ZI>tEIiPAoq zUqZeKTVxE!2|H#w>bSYMWsL}e-`s*HH|`ClcY6|SHtKRfQZe}d6rpo9rU=z&hVj^- zjN#z$oY@o)Z>!Cg*;Y&V%X5_Ay;)pobxNsglp2}hM2Kv=NVc9%N%5X>!P&nWG-({b zf#Hd#%7ysryuSNh1|!--kOnYUKut{2^{s?;(Q8AWYwm)!NwS05_#e)4(b4#F^D|2y z$d6-bqc8CdFJLDuUt_S_i<-9jln+Vnk^;2XCM`=kfed6cZLs}*c&g_tOcp(x5r24R zyf5UZ`;TBmdMI+Eg&rm<+X+vW$<0JQt<~Sc>0;M0;V4;TXZLNB5bpV|Nha;ea|hTQ zM_&SEd(BI2arUl{8m^K+52yPw>y}haFMa`LJHN*-^de89F7!G0L)#z$H*wiMPJm+( z?@^`8OydxMct zSSO0&q7OH=jG7z5#bd{}=z1jX{Te#wmuCB_#uM{l6N?5-tF0Zj-aduKHZDV1)nePN zL%QInw8C!Y=~&v^ONChz((fY%_8- z|2X2rCYe)Th5_fb-boKwj|35R2w?c%p_ERdq<@uNFa>GGlYeUyd_-vdc89CJ*ENnW z+qBtysG`GmJVLBS-{cD3ix^rvy4#3eqk)Zwcekjh_!L7HUo#-hWCZ1hEy*srmXiO| zWSiS!$hv(J;FJaYpNy11Um1;4oW<{2-)70IQLf9gQ|>eS?CIyjN-P9TLvc zX*<5Cd|=M;2+>n(wZilfYBLoh%ei}5zBSHG=W6je>Dh~W9o7RcI9Jyq@h^PVS`EGj zNUH$a*$ZNo4iq@h(23ZHLiQ{=*jf~-!KGz>xYB(nF;LR;Fvh>7a57MUO88&{g56Ug zDNVxE*I#P8F+HD5jS+j-*Z&$4vup6T%Z6mv9+FYV$CtFwqW6tezW>y#i*rA3JakLBdKOHWmjn?C^Q>UtAQ?^eN;r-1r|sutldMobm`x0gdlv znTU03((|&L@=*XXYru{S95_+0{}C*uZvaA^>WBlxAwP#4Ypb0U-qojjXdZ4lByRB7 zUMwo7X6quebq4U5gV3qs{UQt^A~P<7BlMO)qX0ZfEdc#!kt-k-;qx`;trpq%9^z{l z8+CBhP5xyX%5DTU9=}dbgM>aTMqPeOQYfnt(}~usv28K)NwQG*#=t}2x1+yVdDx7G ztcQ19;x!8$)s~6+Qg(7-p8EEeF&b&y!wLO7>VI9-?XJe@_dy(<pm3LazYP!jlHW zf057lod$<(A3S9(7SLiE5c9xll&v-%eOYH$9^gAIuM>7R(;_2?^A(<6KYUvC=ZV}0 zMs7F4?=F^@>W$lLMW^-Gwq43xzN}kHwEFEIJTADJM{QnZ031#z9kpoM8afzfs5&Mn zc-w0Q$AJ;^H<{O;c}ly3rCq)_VU6X!C#oQ0bfV$boo&G`Q-uM0sM9es7%z14Zz+LDUDdKH@zexP zv2g=kUMB!E?+bd1TassL(P}3oz%yfo;5Irkt9~!{|XpA~+v!GuuW)>g@9~lT#M+lpobN zcxQZO)fh8~jE>-`%?^H~Rqpg#j}lEb#B%cL8=J(%!9~`J22bl7A|+Xvl^akC2`8|` z^y)WF<|ZFbg=slC%Lza16JlY0sT@bZ-IAm*=eEA03oY%X+D}K5`p(lsL zoqV{yer3rKVgeJt)l@I=bHG`!owEej$`iJTFX)hHjGzRY|8ZmLC?E)?IYuri?$HG838Zw(~mR|l_IBGu^Etjtg&RP{Y=l1 zygj8-`asd}enf|7Tr6Lg&EgKBH&KtdRTrG$AV{#>@V%;z3Jv$m9#v*)>UZ4Z+#kuK z%4Bl%{6fROc*fHQaN;>`;8;DE6Kj(ZfkQ?~NdjVKid^V`37bbEupd>sjtFzcN;2dl zmDD#zb0zd_Y=72BJTf}n%=ut*#`^RdfwnpcYZfW!hhar#jbAxs#?MSUQDOus#-wW^ z8+asg0^FgQFYp*h?0rns*eYvEY-jfJ5zot)1jqgUTGJQou>BE>KS$-~Ntj{>W5-Oyrwa`4uD6TmKyFsa;^bQ(as+z1ztRO@*oDCyazIX z>aQp5xeEU#{rjH>Ysgz28}j63+G);8`q79|rv0%pUVga+PV({fGh~*6sr0CUbzp=4 zRtt0|O+C}>(TCH%Fs3mAQZ6CuU%xXHAp4Y|2+9B#(a1L0^|eO}oK*T{d*JegQQpxu zMxT!|o=suHl9JbxeG^FJsaT-`%us6{3lo2?tJdECRj5CjL_9`9M8Andd7|ICrCuC*%=WljS?=Rl zvDTP-9;m@V7QW{1J6C2}X}_R(L4U-X1CSAnu4|7TZVm0L{%uOj zeP6S46Y)h0$;r8l9}5GldCD(8c=?CrC|vW^etO^8reXcr`LgmvG3D;KyI$o%j-cfP zmKpV+PkuZWN42mF&IAM>@;ZM9`gZ<^tWGvB2q+Rri)Ls=72#+elw(|Q>d}$BT8A!WCnDeGyTL9L3&MZv zRL!?v5Jn)Lt0bP{GwC39O#Gm6rb!P;U9`)Qu;jbeSx>y64wRZ}6%r|%Wt^-yUl51z zR%WzZvi^tu)p$>RGWRR47i)k%iLX?Aq6n&l26Q7`s1BzCwCsEwt&?@)-<`@_OZv*3 za9J#E7;EJ#^mn>sEltK+0L zw}PPf((|=jQuF)emzT1Z1b>)ro!y%j`h!dinBQKh&(;*Gw*cY$OUD-^H{sE3+N)Gf zqJt)6r$ZaJ{(>()d-}d3N-R=)IMmK4SLcvu7xbJ!^eN42rqFBF_2bZWLey*te^>(8 z#P9Q5OP;yuic(fxT9uf^cHEHMqB zwZ{%xm0%)^X=D?kk5QGx+K??t0cfR2T7&#>!Z~Kst5p~Jcg)dfCuL@3wZvQ6)O4!o z7;gTJcjvGW9~e@8fcI2fZ7L>?DBkohi;*}R85Nk)YwGyR5>rRkNSM;f#t z#{UsS0m!gmRB}HHfA1Q(!`!3sm-RE8uNvkUUd*DU!SNYxN>M|#kY(=JA95V3xtAgH zkQ4ZVJG)Jnxiq4eC{Eaapcc4)X(H=qcP_KIQrQ1Z%ZF?cahdTxT`!8(H7+3(Y?cSj z%kJ`+#(um@X>&rSBkR!Yp8{c%gZd`zP1B!Pq_KP?8FsaG^{4o3QTlIQq`R(N{%0}z zk@S!DA$^@?@2}*A;gp8MeN!>J9UA51Bz@Pg_FmAYa&aLvkAzWC#=UOA}^}85?p$nu3w|0~*nI1r*ZDU~}ZBSJ5im ze)l&k>0yqsn1(!YRootT33S75?RF+%E*kq5Be)(&bom+G8F`fdr@sYbywk_97#Ci& z?`@NBL~$e2W!qtr>t8wj7AJ|1~e^zL0b ze*+Ho5i5(<1<|F}twlEYB!lHsTcu<6c&PQHU~m2gqEtIRMB7HiS-6oi1(um2-cI8b z&7sLXEhiDeFrlId=_V1Uhd&U$8*_CH4O6qLnv0~_=6{}q(9`H_OAHCt<_(^yXO$}d zBQQWQq;Il|OafT<mymw89wa$|d069L~^o zN~FN*kLK|EEguF;+5weK&aqa|J$zF{BdW+O{w>P5HLze+jcnGbW%NN1(1t?j-K(@KD_zk99glSXem{-&QYr0pIMM#sHS7fHl zBfXf&AX|q)e!*UxNaM^C9jkZm?A)0@y*iaSR+kUmWa{p|%#Xyy`&L+Ko3_=|ROFVv zo7BOr%>EP2XYu#W%K5E(-q2j@iRjFBIq1Jn1@OKx~K@ZMS21fssMzZHW# zKIp?Sp$b>Zu5y?fI+`r9zW&Zg+jOcEm2z=q!t0*uJ)>K`E@_noQt{y&jOiQob3JQp z2Y%kc(v=a%QduV?Rzl+Ch+a>wHL7lGzkNPU2S~N? z@Fjz(+_FdMNdaw5s9T446ecYhEAe(Sdr;6KD6*d2W>X_H(TkIG#W8)6BGtz&W23FE zO0=l|(vwDy*^GtvD?A;H&M!Udxl=9?KH>3GwRzAjApTd!5M>EUy5sYKRu? zgHs)J+(U)d&xLzMD+pV3qgGcG%|iemTSJEE1`aL4e04_H1k69>aK+t%!j6@RFs%KI zpgKIDS+eMWWF5L)n=|3y2bE3NA&c$UXL7lGxm)7eqww{FI)nE#MbV$a4CxYe5{Pn~ zx(t}9wCe=&#u{uNvl?(gzS<=jft~3yx&BBM=#vVRqEMt@!)=WMVneOQbh~pk4|giM>{O^ga2lman;Jg4SyXNX8^!KIC23`SHi} zS?01eQ+m>~x5sH=GlG-1x?Qsu*^g%x;vZ5{^ZmS0&J|+#S@XJKgJP>C(Z*v(O(==S zKkxRZQH?Cww?~cc646cml&DI!=>W;iK<88ccGVrNnZoA8?Zo5H!asbHj71Iun!k)l z(!5bX30wui9%F+0b3`YxPZ>fBm&jE&xO~eWu4JikcnE#hNF9FiiJ*$`KS|IZU%yZ? zMR0w2{Zu8W?|ScTN)fbn9}x;7ap`!vT7O(N$Iv3&nwn|qeO=Rp&+5JN&3FrMJ0u3F zLM#8Q5kb2G`2WDGl*t`rI9NivXvU(PMfo>be>4XuRG%crzkK-qkD_QU)I_BYH4sc4 zHZZ^&M0~~Ylf%TjS{`J)pvm# zH=G*(eEL%(cRvp+-D-Ld*0}fUIQrZ?@Q>5Gx9%p}xZn%v@L3B^<**-LzL(O)(uOdN zmmI_8FGz)nRqv4o0R#>MWH;2}#Im5Aaga&jHVKbVV7nynFaM9C=!onsZ(m%NodA`Q z9Q~5n%PzappAD$Fv`6S0qa}<~`vdLf*^5Yq)yXlwOfYL+9$E6mA(2PFvZsfVw;}>5 z+UKOO5gwLh+NvV`cOhki=PJIlbAT=lojB%v+Vr&C?j$< z*Y38&^2@Go#px9CVDwbx9{-oP`ZJw*IR$vOj9k%|Fa;CQ*?Q3hkjDup5mq$`j}}cR z;%1V&IhVee3x|n1rg=va$|kwZH{%NCwunM6qO*+I5~D7U0R5ALG$$@^^sLQm6G!Dq z6gUeK0xWX&8Kd87R?w(DiR6xLd@$!Tl6*a5K`+@vK71MA zcYyzD%m&%!C0;=GI4LS#SR8!)g*zu+8|wem8PfLf#ocOk72;$`_S^;}iii-8Z+|LH zjR|Qt(C+Y7R-3r7#VVrO%5SsXr-LWz+$(XzdiKi!A~iO+*q`$qfwrDFFbK~spjj?! z!a6cwV!2#-J_NSBV_w*|7jS_fQ~=!vk)6{ZQ}J^~!Mm3~T~IETocjg!E^Z!_7uG-m zi>K7wfeaq81k{pMW&s#JEITUgG1qDCb+%+V zF`C^sNrLTJy*^Po@@$mvKqg?I6sjVj#^`1#LqtilRO_v0AlwdjbC&>S&y-G6*K~6$o%%*3gimV5D9gcP{pO{ zoo7tgme61RUF__E z-&AgHMNW5{BaE8dw&EVVvBGjcv=dB{yZ)0-bXMt~h14zHvR!8KV*cq3e_S_$Z$nyPqKW&ME)FmX-iP_%rW#hcQFBN1m6#kfbKZ;#!l3@^Y4R&^!XxFM z?lJN~FVWfvFtZI^cw z{`s!;h#(IvUQ$23f`a#nQe#Z|GGQei#mjeTtY7u|V34CEs)KEAvW?UPDwXx9JL5)l zdM`X24{c!Vsp{U|ZCNH>?4WTQZ6xp2S&!+hBE*Xz97=to&vgi{h9C%bWbncN+UHc!M@>6hQCU zcWT<@}*F@bUJn4wx*F$%`~~eoH&;UV(LXL0mSu?`1h&b&9?VtGB{`UUjFv zN(FQ8HPDb27vAT(9Y$(e#P&+>{=p2$FiS+WQpN;{7z33gC#!wc9+f3d)Uv(ztKBZZ zPy8idW5$^|xusCZU}kLlKp~foDB@vTbj!04t)7Un`^6__i{V)6yxEta-Zd)W0;2Dk zzhdsR$Ka`fc8qbB)byAl$Oh)d_UU$F4a4~%_mfU?kceymSZ09ZssHYmS|bW zeN{BkU>qR-%$atn5`S45iz?(mWVAtpI!%jC0c@i1^$d7DLmCzuU?~dcI$o$Dnwg$U z$4?2n@dtNsCj~TV=Y1o}JJ0FjA*FjFm!=|Spg-mFryoHVs2!9k) z8NQ3!{B}bJ#PylvK;awqefptSq(ytHH0|I9%!eN4`=yEc#>%h=8cgm+g4MTGb17E#b^ZTlVhK^)+9cqR{Jj(`h=t4 zd}pp6Zl!@l1lK>^0-5H=Z3I>A>>0>!ed0RL(@ug2CP zCOeI>@&Qksh!QuNj*b%W59G6I*nf}rn<_}`>IG!>$5s(2XBs5&)Rk-9B+#!GEsLBx zjIYGVzVSU4cY`o>Wy-03^BE(xNuKSLfjtR$5h&Io5c!gj080Li`qzM%-x+Au?k+T^ z(4IH;MkrzUaZS~FFbDy_|CS#_DKL{a{wI_NAnR2vQ@ci*pUF$^Er`i@YxCrKB@-mw znSeBMCXn+7BCHr^7vQw6$Jw13uxKCi`M59rG_fd}bOe;<`7(c;a@+Y!W9W-##5}n| z6pD@7cR!UC0tCN7K^qrw(QJ#cQ))lifqg6O!>V6sX7@t=6eJ4RB|m;#&AX6->@!uhx7c`^(gtFo;3*KZAF6_*g zwjW6T{Wh@v_pJMpg(9^jl6+&H$k}3quwtdWHUB|kfT8{b%04#9+n+-ewb^Z67#`$; z2ktMC*j~xD+$=8F$)+#reo+ZyzOT2{V-}gDnYY5M@O1++*YN}bAX0o_y)hoeK{k2p zgD&2H#izc|zx2%sS%Q>t@}KcWxu+x+=fO7w$wUlDfET z-Z_t9mDAznS=mp^%DJO7x4rA4Y__9A(|={o{`D1UYKiRSadcJGK(nsLM}8I-QhwI8 zSn?-7Ngtze;;sCfhT3Sp?9&VDr?x?=OEq!tzg1u1N|?3y8ID>^X1rk_>3L*UQn0hd zmEy={D84Z1@D@EWvQZM3u6PsJP?u16-V)Pr5=i<8)G$itPJ;YhtxdebDl+jq#6%BO z`yQ1Zf)m*7WPyoCcT)Ewm1xGzEgIX}8q&s)TSG=I78;SF9K0c)m~BnzL8G2219aHi zO?W&|2$QjyrB90q>KQveQMHy1;hM3F6MkVw(@*|KC}2%A^o;oZlk8zxmMc!qeHKKI z|Jq}$$s|^^aFvo3&iOp0LH0oU=x|PE zKoeVNtoipkj&v*J0}aePQ&3oRGbd|Z*riVrf?r?S@I@uBqu)ztIY|ODlB(lOqqxX>=(R2q*uk-jW9^g zoVh6MMS?zA*{D!riGu~*2<2k_rv88+fC_6iE&3xJ86M(u!15Ax857H99w4zaR66eG zt;b~{mdM^`(8r+}q#o?|<#a~QgIW(>Q;?c-ySjSdZQE5ge+>dI#~4NRaJcftXF=k zG4vcx5#Wj5iP);eq}xGTtOg?U^^r>Qi>C^qjm^mxwn5;$+NAdp5t}l{!?M3_i8!=w zr2|Zxat4vi(_R*L5BgSlYNjhiq<^tz>D>3_^%a?YNH(!%p(lPC0X0|pN6_Nh;ZT{| zhBuZy_>4A|Gm7Yi3I+JzT8gJXtGOfkKPoN^lxls zWb99)6vN{w%0F+XHzj4orj(^MHk7n3I3JxVwFy5K$t;QRc7ZfrQ=p4_%U@lZpdRHG zjH6|HVO;2eA|!LYLLfxb--^}r&zrE1HEEeK;ThdSfBojB@l;H^4FnNzWZfkshjda=X}yyg zW8e zT~u68T$G3uKMNJ=O8mB-KprS!k*->%)Nby0dTIjHU;v1hlHAJayU5}+{!L)lP()ZZ z{ac$|Wscb+T^POX4syYvIfY&``c!NBye%=1)2}xIQM5Bd^W=7q|h4j`s+2V{!^8v zpWL;{L8`$~IF1_)S6vCqQ|2&wbNc$mH+Ja2EB%?;JCd6hZidKJ(mvaT;O&xED+Ye8 z=BU8qOvoSZtLSy2L6^t|yy<6cgD5OOwh;3l{=R1Qb0M~>=1F}WOhD3sEkRUga)hD)V!ZLkL0eMESzvT zTqNcW+P>$^_wiMIpY0Je<89LD=~2KlOfw$FBYzyb4hqISpZ4s$k5z7IX`S=3u&^jD zn3i(jf;uG|qc%_CENVaP#yX1RC+V1R4(a&PFmR63jO@QFSYdxEv*y17nY*}M#Y>9M zpN$r^0CNBD$YQ`gL`(MIBxma?36{ZzKppvtO7`})@82X?G?NycWci0R7g z?j>#X(nO<`mh`D?TvtCn>m~YoKTu`-`oM}a6b@bsrwY){x(q{T6F)RZ5&umM`gs@+&tQuNX$08oe}OTh&NQnXaJ22H;2#<6 zAwEv-BIgHQ-~vbu&P=wtmrJuyglDafzD&ClJ?M12ox9Y-0DL#{W$|sK^A>>>OGDK$ z$CNdyvco2FZ@a9aWGXxtJJT>y+}IkWqLtmJkvuVj4{hJ9$j?ZAJ1i-pe)GgZsGBlo<${UQR!p6(Pcv**u3p{u00}^EHwXKpK~xyQ zUOYVDxKMs2U$4XxZ0^7|I83uNW)Wc#QpgitMj(O0Z!>n$Q9|Ie74@qc?>{0HVAuo3#}eBq zFWU9Z3zZ$*%N57Roha9f2K`5U-p|%U*!3-|n;ZQ-z(_0bZ&?p5-ukWj$-YsEuC&fQ zDzQ@Z7ya=TRA-Q%syf-Y#Zs^AvItq% z3n%sWeZhG)JZuSbrZU-iEg!hGdF{PxaSabuDRE$A7@l;dFidKCg73IIj^j!SsIe#4 zrBTMV-Bsi0%)~6%C5_sM6tCa*$^9dMS>jU9ijJsnbt$6&l(r~%@4<^Y|6U2b?S1WZ zhRM&jr9$4PgU^isVvje47!@C1XKa3K^DvX_I&@d_n+6-&vz53X(3fncKe+ra&3G^Z zD@_GhnQ&qKN1!?r`pRhoz8DQm7|g{Ww6bfv8c}UE6oF&K@NB=tne);6J4czB1(8fb zMq!8d#@w*Az^Pk7+%)2$vQU8FO#K>%sb1ke)xK;)rtTugcWGAAU&6}*dfV6H-#;ht zsuA*h*(6&cR)x5@cK&HEJI%J{=gkWKZpMFmUKucG=s=j7p zrk~n6AanaGW?vS9pxEm~y#X!BUv4-{b1pLWimAG_(4!Q7E#NT%d6Y#XzziZRZPDgY zo8}05Q!nY=RWwwBn_Bsc`j0ZAnbJduD4Y4RI?D@U6gTvQyd#3O8($JDi z`pxh})H$LGexw5?C70b44dKw=fOHaCQDe%?PyNriDhF@bU{f-Eq-&f|nrrc~HFC#> z1+03x41gQt3W&%sd5j+-e{Yz`iaBd;3LOQ5M-EoHL~-okW=bEXJzzCUTl9E2P*R zColhaCpO(1gJhOVWhV{^Icc`P!-U|;gn%PTynAhxmoJ?c|F}co zq~`g&CqodYFKCN%<#0Nhq9(|@R?uYw) zjGOG$T7AE^+IdZj@tlOOmmvtf>WwX)OQN`V}q_-)6>BbyWocs88g&rfM9v6u*0O!MXe2EzJUF0 z_UlTCX~6m#Cx+MStk&%RJNszQTfX+<0ip@!udt5@DbezwON{Y(=L-G`Snr0rI)Hs^ z5_Q|#gdAL8`7)?1A%3?02;exruw(4h&`E_=B7G3sXr9_$!aqThEde>-KERQW;74qp zgRCN7iZM$%D@g$eD!u1K2NKO#NuU{$5+j4SG?^}x*agtTp6oR%Kx^4J{imaD!oZjyUErqud&b6%q~!uqgXYhF+qONj`}=7oddp z!01noGX8W)c5k5+Zz8@WCHAZ79iP@@2E2WVfrkCVd9lpQpK$nGgvMP8w3(o1aYw;D zg#w60Si{ypf5v|}+=x5_EQtpSa4IX_U!%dWOKpXSM-ubp5Bbv!$_k?9^4EZc&`^MQ zFUyqG!Gomf5iYZc9pl|94Qxrg{1MMkZ5?Vp1mKV0bGaU7&cDxyD?z2|AVj(&>sb+5 z_gwEB2qCYD;yI0|yw9ugH3#s#?sFN?@(~iCBoX#kB8e5a^a08t@B1wyZct!oDjIFVH>7~(rHl!LQ$Syu| z|D)6Sc5n}XMVcq#sInfck|J?G*G$bWmSzMJ=F_@#283_>5Vd&GO7!?4Qn5sK#TD?` zV0zzD0-Qw(K56wlQkus($B10Y!_M0QMPu#PJ4e5|E`xug+%|a^bpca_8vxRfG$jM;m*T_WF0s>$QdjrYU9ZGA$2ia*ouEpRx8I@;v z$3Av5)^oXskmiZOJZdhT{P4zt>)A3vR}T>>!LK=;0A|- znpKh9BZH<%LGjX9YW7s7UQ4o8SnmbUB?_Y%03IXZZVwGZ*CgM4>)!P5shQx0GZdg# z>?U%8*E{{>tKg(HGw*I;Er0pZC2lvh)w}dA;91Cm>wXV1V zZl}hIQ(2)LqD@__@Q-xw zf69vG^&Jg8yVXjK3jh@sU5Ct;bho#twjQv#!?9mVOQx{=2h<*P>(ifRe7usgT#6Ko zBy`RnpP-npZkN4|mhoKHnU|4wI`~mF$aBeISrX@x-FF+LLkAZ86zwMwY> zav^4N@TF!yS;T2eK_9i`Kb%sJWo1mU=xwdlmKs{=K{RWpR})skTQe{B(TSed@`zUS zjgD{52H%jd@28<9L31@g_{l?*OigUtce-g_(?ZhsVh<8-1iQ%?gc&ap?Lp`p+o^lVcREN!)Cw6Ys2%BmZ!| zwZgI*&Z6p;t5yH~+4T5Fo$phCR(z6iI68p+FC08F83o3VLq?XEqT zCSrP1wCrw$EsCj!57`mxTArG}bL!GCN0&>)5irYP3i86~H}`RpqO0Egq25i1kP+`9P! z2LA2Fxh;)q@1+{!Rrdq-1~SaF2v>u{iOoE6rlkW&a!ssOyGDm#IIm##iJM;G6PO?e zW9u(oD|WOK_vh5GviLw6_b|_mdy&1GnH2Cn-dkM-&=!Y-QId_uY6tvZ^;GR%e;p=RlSL!Nf>g!F( zNcNVSYT7_$@T$I!`Hwb8&&V9JehE>48d2h3MWPLXaG{4|UgiA$!KZlGcid)v-&~mt z*8R5SINOG;!Y-@F+-Tpz16b^i%-#g!p)3T5DwhYa_vPR_qUppw*dxj)8HLZ@LMHFv zcLa_ZF)4HJw}I__BTKFTNw`V%nD(=*Q&TOfr^19wuhj;mgBAUK0D2dM3S`hT%{3Xvd4y6#MJPBVqa}TBOBeAH$LuiLq9=^c!8G zSbz{*4?nRgQZL4+Pw9CP6xtWfxMaQ#q7DA`uIMIqS`bJmb~jCf_X7Tv(T2(njYaGx zH??*ix#TxTmD7RvNDdmJYt6E=O#sInsw_mm6!2q)vVFfTWW5D$%Fr*+p8vYUu%rHg zt89o{z>5J-^;ddZ?thj}KGx@$$}ZJJo@vSW99+g+gN@(yqWw1?9ol05GqNP|jDF}= zK>F;|^egQL(sh3zvGWp9hZu@tLrfp6iq%D&k6yEq40LR;tG;j#O(`6U$QXP=S!8j|3+=NZ3eSVC|$l0ET@h})B>Kypv^ zqqU%H2kPS=D&Y^?`*wbWzucuB(Yc6k;iTHRyXp5kTTfJKQRNg@w3`#ALXiw)Ni^hon)*LNX-&SY)nlU(Y3mKGV> z*T&`z2QBB1SsySgc%7)Na~@Zh9zdUddQ@~I$ChR-Zmf~azN}dWwfh-E&U9SG+7d9$ zYHC{fqj;CyTJVS1S!$yuA;x5;gP{yVkg8&0rZdnM-?^Lm!->S5sACrS{adKR_6sgH zA)a@ZBSGi~oK&X}I=Br3G~8Tr>MJVrsDHeC$ao)wQ70=B9|BJu?|_br&nn&7pn}iEsTNRv0mKJZH1v zjledSEjp@0w`?M&jb>b*oppw(kB`0v@@*AqTN?;>iuAKW1($uv79l{ESdH9E6IAw@ zZm8OkJ>Ti4u#G6TDVK(duQ@4v;a>~4S{quSs9SxCWg*tu_vIBgiM;{~&970=4J|Ju zjZ$QPcFD4TaG8}&rJ7GkL|b!Smh*NtUN#+{v5Jm|svD~{(y*~i*2Os-Gb4X(sQ)Ccbj5%{;fw7U|5hnpHCLI|P&)b1>WI_5 zU(V)=GIjBl*&QffOpf4h7q3IGG^!H;F^)irMafzjd{V49 zn(uTaurY@rXIS5h#e(e3d6D=7PCP{SIvS1btxbxP*L@;L-V`O7h4goB zorI}HY{iF0pZ=T|y&SS%{x^x$|Foj~ZnmLSlu45ABsV;tfVTdXX&h@V$-d@Y^Lpy% zkA3-0`lx^kdt+-k;h8}Zr`U|phY8g@5}N-#H-CZ)Wfse|s&@A$EiCs9zW3t$FH!oj zn9!*zd5d*2G@8~6l!$B4i8ARo2icR%h(YaVH1=JqPKy%k_5XgYq|Ho0M@yK7?I0G| zEaBfS-W7nl4O?yf-kH6_T|0Yu6`6gxkJa9vXd6Z4a3fy}B&g@AYX4fV)+k+_M#JWQ z+z$Pa0X7DHF9RbJ>H|`l2F*09?yW@g?R@os`-s>|RE#XA>g3mva-pZ0bc=nQe)z0} z8oLwbdFp=4UHIaQmqIApmk{-IYg*}Irw_jshJ?3dlBlem-DK{MW9}(YBmx!DBn+rbecE11IF61r0_*lKE+#|G{BfyoMdvohtke{f@dbL)# z7D<<`$oreFL|px)lGO37ZjQs1eu~fRUhP5KUIf@C9@5{mM(|>)>o5EG_^(ywuH2R7 zAwR2YON7ahdw;yJIZ+C-KWfU2*IeiyPQZQvF~$$2gMMd>iaGoZtvk3|Td-)WC9H)U zh>LMZ4`yHgst|g!tAF8o-3fn^qWUb13RxBkxch%ZPR?B=K8pnt*mueN6%awI+nKh~ zo28{!Sol*P?v2((r#KCNCtrMPg_aqfv^M1#NWb?$?Ue-b&d5-$yU_=ab-nl_y0NJa zK9*Ab!~ZL;>cZP*O1~^as$@M@*z{UBPzLM`!eZzJq5x0lsc~A549etJRg& zIc!(6QI(_cT8F(fmf}^>4j;9-5@&5%D?pHbRq5OIabYE8i{o0N_B-T5P@zm3cz_dB zDy(1K5R0F=G?-$lBHg8jbA{u5mpyMd-<?s&PBbht0gYJ*j=~m8}yDvhfB`@XDCf>Gpm7G^HYLvM+1<4JMR#o|a_>bWa2 z$|9JqibvMA$*Orw@dGU$;gvWLCjB1`1pDZH=-3@gsL%ef0ndVy<*EgVvA9deoCQf zF;Z1oDV@gl%sF9x?;PexdsQYZVW*^O?|H7tOVahifJxA}iCLnn?g77HTVl1X>A_7@ zmz5IGVAt;cY|G$dw5i<}HcRU7sh$I>1T*IVjtnIC+DH8i`^yQ@Ln*h!t(5;N1y5$? z257{5-V!{~oNhF3nDru&dHi9X%BAP{7Zc|sC}0t_vV7|xVD4l-F(~|+e zB(C#?kdP$)oxx6U@@6@Ea$xjU;GI;#wCE+YZ;jftYXACJ!{a|+nSdZC!iLoOy$Uro zP`RLQ#5?M^*)S*P89$YAMrnA^GNS(^O`ZZWfAVX8 z($XsBUnIVf7&Z+&&#{g02f3Bx}R#d=m9lDqZU<}M`84R)fWUW{| z79?wGZp@1mB8ca_;x*px-m@$CJAe|;mDr#A;pg6rZHd%=V7dj$UGdlIUIob;QmVH1 zqLShtD1pCh$Ick^iTv(Q+w64S@*iqh!$Wb<>b*nc`z$OH7aem!hEmA|2#;I(I{g~T ztBgL+o?IFUm@Cj-ZwKgymtfMzb_aFsFxN0X>7 zxvm!XjvTW!WW=gN=r?mYWl_)R9=BYL8_{p`4X%R$mm3lK`J+=4^$yU1chWELQvCmi zAyi|B%y=lPs)ubu5uYl*_2G}#HU)!e$3>jD$T&+hL?trG{3SD)&0xV};5tYQt03US z8X4{uvVFU#m}B=3=Q$w)Mh?J&{P6XxV7`oTB!0Ks!{yd(;r*ti5Lp$Lhh$S-!|Dnv zzaLceBgLPGYge$>iqz?n{uB7{=f-*OKM4WEo$*xMZboG=0Zlr9ONe zK8&>%0D7)9*y4-c{bgV9b$F=gex+PmMa98jV7!?D3@w@?j(K{9C4Bs$=khG!j&HM^ zL3gOygfY?6=cD8uXb@hmWx4G-yjwz@!i7tIRg9!qh_}_#ZYuBq+(d89I)UF)5x7pH zONn`3=q6c2qc%RXYlS=6G_!XW7%QW>r3>Czg&>TXD3%p05DDLP5nK9=Ut3P{?rdr? zLV`z-ztK&(pTN~DkvdvwwnO#w&&^jJ*isxOV{$antiU_lTuLmj1WHbNIWv|ww$uj8 z!l$cIcX9VT*)1q_0F!k1(>G9wdDz;?vKQ2D{1pEi`QbTeTJG)X7x0=kB4k6Qv2RL+XZ zng%-YZu0po$L8wdfxQaAV~0ZTYv9 zi6MW61gW*&V#+^He?7q{LRD$Z8Bqb{J6>)i4hD~E*c!s&mu#Yaj*v%iI}#_K>$$!H zQqzd_jA<8DR%AkjZid#qN7$;s**~1m!5{|+V;9|_g^31|Kf^sAZ*17_|JcNz<3PI- zbB-gPF~82ue2%+MmQI&w?fYoBMdDMY#i3qJ65uGe)A$U+eByKWSwm=mP@Fa2R;@66 z(=KbJM-VCMM@P>pb$+8sMfXKr43+~$9e^M{5O1bz_XlWl^?b)`{*NJ<=$EiEPGIdB z1$XOMFQD}Ax|;)nH11XIg71HDLD$nEb)sDiN;5KAWohVAKFhsC^qH4{gWV4ioxVmS zxD}dpvD4KmjYDT7e-<;HXv(_7FMk7b1=`;afh-A{!f%i_v3$d_^w*O3SXr8@>Pj*c zr^NIUyf|?Zh-V0ooVComGds@}nQqPa2#)HAzfs@x` z>%zMfWQ|H*(k(rNYe}$9zzM0C9iB>Z%NJ+!P4ggT2RKn|mg-8X8o7wPZZ}wh<4upq zi+r8fS{|BO`XK*r4st z@j9FgQNn0NqnE5b;yPQYJEpJJ(&^itO&i!ZYpmgBtipf6#rt98Km0r(8cKqRf-Uz> zk2{<^cm2kN-DI{?v#r4-7zgU~$bM!b$MWb2@+eH4)Hr1S_; zb%4?6Y3?_IuHX=!+lx%lKpaXJ0YChP0LdGn<%My|2QR$nw4ZFHNYRf0zAp`+gJ@p6 z`$ik=2ZKO>6zmyMO7VVbXp&xFAjq}%T*N7w5T?ji&=&?V_EjFD_laq!_*AXO+TPq2 zU6@bsRZ5*P%xUn@*T$lU>Q>FIs_Mh-g~>oPUFz=pjgNSAv7Wis#O|p;=9C~Ryl6?C zLap35a*Nv*@a)>xn^T-h`Lyeo*&*HBkg$(6gs=Z7_&J0TA{#PO?qkNU50PQmvQana zyNf%D=6%te+%;*>vEQYmLVl5gY_NMA!g^sf!2ECuFvAf0**WJ{C2$oxmn$zCMo`+6 zExuUc9pVvNPcb;&YA(b`exrN?v;dC@?z0 zzIdbx6Hl#Wf=-Y0WcVba$RBz!n9)UM9)S(vdKf6ekknV)@HQc{BrjhE%-%VaY$V4 zGM?1EUkZuPmX|)IK}dWM}kH4nMH)wF29EkY3;H_y(O9glx>gx+!7fPseB#&}I|8Pzu)mQJqh~_V9ssZ3X5Xl0TLv#Kz%j zKMuxY9EuooXB({*s%R2y$hRb^0r_OYUsevN>*m^!7EGH*KIW}^d#csC2c-~E7dJ)- zd~P%uf28~KkJqOF7rY78mJ3(D#wRFM%ckVSG?!141kX*pM+3vB1DN1;0JKVTynt-q zFo+CfLfZ976?_hq-#iHp66zZ|S9kwmI9_gly}KJgVAEPJ=zvgXR9cmH6Mo0$$*Qo# zET6G15DlWikZgiU1K*-zH+b5)K86PVVwag?2gu z_LNUN7Yn~#5tF8fw$r>T*AXK~z@g7*-2RAKAyU7-7(+4JLF2G7T#_-?*p_x#md`Li zdb}&SU2C5%$T;ztVv~YVOgxv`cpABOB8bT7E2uviY;^yj>hitFh<|)l>VYg0@*cM>3 zr|MGKxL+8VmxQG^{KJV85+C^g^w^vwe@+>dX2G0QuFfBASFA4eYO+T(t+-Axo`Z}7 zmtw^2nSqedLk?9Rl{FnD3SYwkKG5awz;~A0XRX?9j0gFkKUZ$gFuF*l_*`ST#B#Oo zi4=>X11o)PGTDutap~vM#)sl5zfxF|6!}qc@h=rWFqC1(@5Gud@ane`7vKZ*5sg0d zZxcCtXoiY4W@*58ntvG~`HOOy{llAjd98sFB4JVt8FqNzn z%?r0h`Z0~qEzUe9AEC3~R%Y+E6kJd(!}!c>yQnG1v7`^KXSpmpX2;+~-zre?i}CA* zg|B~N)W7`rU+nY<6ymqt%s=#R3agy)x@+p>S3ITg-s9S%L7b5qcsJw;mcw_+B+i%n zSHtA)t6jw?7V~4nI5N{jXZm9|(G{V)e>me)m*rJk{6nB8?)2F*_Z3Pvd9gf#i`VY8Fi}V^+vvivblF#ElEM*kRZ-`fDOTPp#CO0r=%BDbt-9$@e*%K2g4b^}N`z-kIG>uJh)AhHBfT)49tyP^8=g zWJV~4$YFCWPRcuVW90*h_r=C;yWI<@jbhb-hl~z?eSnM9n6t=8^7ELbOcl2#o;vCaul$}3pnr2tZ*&@$eqCXQn~duavlYjUuznYq zf#BRU=pWZV>jqJ=qfP^wLme-TIB7jZ)yIQ= zCAlcpnRJS5_6)^o*8_THj8Q^`pyg2*d%En!-3d91n&=1X-` zlMT&Tka!^*vVx6J%GEv#XHxNB^UFb0!51JKT(p4&UV)ELmOkgalVa$(Hqqmn zY?8P7wq8s{o`>)UfqICt{u7TQdBmbwrBQbV@kQTGRm>D^lzeLr_x!L8<9|be0gDbu z2lt?|_gvGLYL;YU*qPkV68oJ&PPa?erjG0a?5!p@s|g{s|=> zHAZFm3bQ{uv=mwrL>$=Ek857_Zr|O+g4Y|Esp`{r!N36v)I~oU=+$St&K;*5+7y_C z#%HB-;QS_8uHJ8ssUp+hGLJ=lYn&GIV`nnkU4v)L)UUIQjkdK_(EZ57S2*26jn>PK zb&5}7)_5f!aoRuX!iRg#!t)A%2Drxx>O_RR=lk)pOd z18>Gt<;PKqJS-9`?d5}tfqG~_Pt^l7SnJ;*WLOqdSP%W#0FTQ{8S%2EHM8a~m$7QJ zo?eurEc%@>=N|jzWgor%;TTHPyrs31(}g z;+0RPCf?pOWiLKB3Xgkq0O9d1+AOWDSW&T^ZU~XaOb8n)@)mEQid!zST820}=Mjx+ zhkj|kq4^zc1^eTh0w^xdW4`sHXg-W064o<2EuVx`tn88b5hV7?bZ@bPrXPYZ{J&^1 zA-?lCFvz{8!HVU8UaTze%N{O)_OfE~=_WkXV(W~Fy&#XjR_c4`J9nyRg^=9G@w)yA zeT!XRW-lusn-jnY^^qgRuM_`lULBHSqK@YG8Q$AFQG6r6c^nXz4sW?N4el5LREFQj zUHPgAi0$DoJBEI5#dagk;u>L@@tdPl1HM=2pk;p9E>pBYyb~Ppk+WyNIzLEw!!D{c z1n*t_p|P%tWDk2$x_sP~qy%gy@p^B@1n?Eam^{&HaH2?_xoU2D#CEZS*=$V{&!<3n z5{oybUu0c+XHjg(%=fNDyySZS>f))OHRp5Fhw|O9>S2^v1P4Wd@D$(uP|_lisc|j!&$k4&=hbcsLK3iH*?;PQZfGLV zt!s8~s!9%jp>e1FKfr}RQKD(Gjmo`jKicwSM*7M7a^bd#1vQJ>~4b&-aq^RIiN^F~oio_wY)_KrR51yee zenTT@uc9BF9v+{GYL1Nch~Q#>0owM5EY8EG!l*>4M1iXS+P3rh6uX1Rkr?Xlvf2w* zL=$1gb4hi)9BM8~1T+D5L>bYfTM3yHSA#iNq9BCBDg7#XFkG zMd(^;lKRGQ{UTuR%3?Yci6p^BdJ$MhrI79==nrGrwWSRX3XWsfPsDru%R&O((JCnB zAq_yw7u9Gz#piXjB0|ueFG>!?Nl?eIPA+F>W~WHdF@JU?G5>hr@lk0E0kpHDp%kmO=$c5 z;htZRXe$Wfau^RGgZ5|!I#u?N%q(6SJB_svxhOa(rKQorvU&&3TN!blFvgG1c4`b^ z37L@eK~>o`SrS9V??|C5{#Z6Fm!mM|0}6x)+rsaMkonP<7Fty=jJZPzdv7CXxC4vSTM@hYRAk{8L13ac9Yjw+us1_hL3Na~xR5s~O|q)h=;sX4kB;=09)YBk@q2CUfB_kyfKi*c0Ek z94GZ^=`61%I7ix}usCyN_5}@W{`Lz&>t|%YC~46m**nRXN6c087;b7a9!So`^DV^9 z5(3;I@ejuh_KZ73EY{I|#as(>VT{1887B=bib!ap>w!J-xDmtMgVx%Q{%DAip-3S3 z&v$E}$-I+7x`E{wVZ$<~7t2xq z<$UTjlI(rUor_5)Z-Uf^MomAo^8`D3eB+EfP<4t*FUzuVwjp5sSn^54(OZ1ZT3#CT zFYc}it;!21#9i}nZ4M)O|MxZ8(*6zhLbdHt*Mb#z!^FFwNg$MrhFTm6FfOf}gEb+& ztk9lUSk{9GLssO(fIUgehgC;kIcKV!5LLb+=dEun6uz+XU`FgRq>-tZzAO@>x>*@y ze4#kAR~r!h>B#G1qgT^~mDh>ZAi&RxpJskE z=Zsc^bl$3Ach}r}g9i>>%@<~g$JnLoMwEjj5nDzw<^4WzF6^_-E<+n`dJi&UZ)xuL ztwCr$=<`L$$pP{lzkf(M(~6}rjJP1E`C+>>2*m{%q(xqz2qovWi#(y}l7Ho&XW@1Dsc>iwTYu05x7+GxsjWT|h%YIcmVZjT<=s`LB z!6$p`D~?CUZFW&uCI}q2x-8Dw+M>Yj&47P?0s5_@aouL`SYLBwjf;$C@yO7hVW4OC zn}kAqgWeXX3g0Cq4*o|iXV&kAJ±R>MFOp7JM4pD#r*Z9T}X!IUIE8D;^=LFcP@ zZk>W}tyGO}0FtfP@UbpX0!d-T)*x;i_!+v^zKs1EAXPB>x_hWN@!$?j^*jFJVZddj z)xgxChY(pQxCAN)2>2T9V!ugqAPnO9hX3K@+t(#x!ax2kg*VwnHB4k*C?a>h0d?B2 z*!4T;THQ=5H zcY}A8FX{7TKG3k8;#ce|8f!AQ+wtAbtq){ip2-D7rEf0v`^zBVO8o&E5{HZi>Q?Ar z(_ahfuN8s$Tir4?MdxdVrCgz0JQTuu!-=KTBdbt++RYqLzZ=`rV@cR( z6x8YH7P?FJ2s$TSbg$p|{J!d0;$=1Y5V{$dUWim`6nHR`e%}uUaq5J;{{0>Y;OktU1d&;jpsigIIW2TCsn#*|E?12KQsyGzrH2I{^DvBuV>$|Z3|XD zBE#@9aj|@$)9~v7jYf5`&J+9Pk;xSpV}2%Hmm^*|3z^qI53&ok|5_ehx1&4(_i}Sb z&sFvguq3lQTWYu7ZPx`Oj&!!`(}j9Ti$ia)-t;C%Jm&}t<0U68R?i>AgfiCaj|pkB zdX$X~#=HdXnGYy&n~glkAOs{*2_v;%eD7!SewtowFoqo`jJ=5emQ-U|VIX?Hw5z-R zz5+{eD3)|c?+p`QI4x5B?oe$Lp^(Dm#mDgQeUhKH2lq1zzW9>R>c>8mSMq^c8+5)z zbaI(=m9NbZPEAD0xz_Fj!R98|sRCl_TWHfW%4xy>Z$BpQv40?>IS)jKmHZ!dK#v=w zUh)lp`oXt3F>TX{_!_q3>sOH`Bg{2Q>yEtv?BC#GE?5~9kf7!?8wR?Z2K~}M2GL{K zM(Jf8r?LNV1W+?*(0|(*h6Ue=O7}h7X z)a+H+K1uU&fjvJElooC1W zguCF#Y_4#gQMp;ip?%iadmty;+@WXkby3qxlIpMG55sptDTI-D z4P1Y|m#bFHD`M;PsC2Lte6);oGuk^JCL2Hd|Mi>l08fsFo$`o5$?dNXjrPp(e5(lL zSnLH;=wChA!p#rP)p_q`F~4tk#J%=nQ2F*wr<(mQeNc1MkIeNN|9cOoWDvzCxefS3 zlDBYjjNFPE*=U0Dyrf3DNxV5{ceqiIyy)A2quW&Lbm{v0;{w3<6uA~zA|2REDkLO@ z4szHUVrFuEGjv{JOD%cwTC=fp^GvzX6T1wklYZ$FAOzObLzL!6iE_GcYKn9$64k6f?4)$`M6;KUJHX=lPaE+ zj9~MHF);7i;Qgbi`VZ$dz_umO2ZSTCL)#8-#rNNuW$k+U@+iVJFUKhc0|6ejngqk` zkZDsr8XU-seC9$EB|}hX&a>r{Q@WJmKD^B$Lw^H7y={0s1p(-bMJkV^CwAT+tQ8hJ ziZ&#M1_l8=-~N3Z#;?%Pm~XDHQ&=HKFLpHjAC5|h^J%0utx6bfsOoLV8DtE60tcS^ z8`TFqUSvR!YklH%f8`jR7S(4{q1EQfGAZ&_pB|3exl^_FMt^-&TV%hN%*rsC=7f%P zqjLB{)#GMmy4#2!-l0=s<_k4$iZ}(TmWa*S4Qdse_(3`*+NYz+@10UgjWmo0#q9KC ztTb=3VMkV&u`c33eUzx{qFtXuhoTE!^^YT;hf{1ukJMy7XDU4X_UNVhH@h>j$<(=v z_I?NrP(2rQ-|fwx2*~Jc#&*61%;hfY)QBXii-pu^+x?oBO_vE)3X^VKnjXhLs~jAK z)@Y)*Z!i?oz;)Fz6~h}jk4HAYhN5k?7F5ON~p`ORTqvIAUc_BNe=^Y3Oe`ZcddmQ{qU zQG7@A_V3DC1gUbRga6t8DEkDXj3i(1l|Ye3UIf1^XS@;e&#(t+`3QjYH@bXn{N zq6=Z67DZ>z9M++)!eT5p5rmxoS_Vi$!S%nl06M0poa9-KXHxmMBha*BU zPR*0mp8$sEQFeUpACu*wy#eowRNnS_xe zsg1x>pbQAI!M2+DIDDl<2PHCCOP$3_^%inv|(1;Xv@S(nI?EF>ikkP)? zl5%*rI!nOO(c(1cu$6?Vd!eKrT<$A=y><`mC?D2Bf}f#ekqiUbrgG2!TrPRdTtW#J zhHlMJTT`$YDSPn%UChhB$k>u zX$>(r61X^dBO7igM%dgtEUf%EfhA;2#Pv9r5UUAwzIWj7FbE2+YcIyKLTk*SC8SaI|2Fb#BlQxne^zaI4fF}3wR zN%~tmlD~+m%Fcn+!{i|JYJxQ_-7s-02`nVX3GHzMuNN{)eO=t|dRdPs=IygYxxHX4Cf{gOJ(erPP^4a*0TrI88OZ>JbfgUHVqO&>3c z1}Em-3BLT#SKS!EtYVMst3}bD!6qlI{ls^uaI|6YbOBXy`Wb+}3s5?-&+VyRRc<`X z@%ZSnna$%?^Bm{$MaOIT=BC&o`o9C;m&o1R-Eh)ZpA_;A;-_2EJ(V0G%KN3^2Qb|h z!M)P4Iai4~ZI1YNBDa3L+r&hFRGNEE%l!N_w!cOW!2_Zi{amzEEkus z$4KkBMdn1Ax#t`?%J&5wTB0*aBYV&NuDGt#l0rI7`*$FZKGJo=rV>$Xdl53=jkN{P zrnLPX_kG&wXx12)u3?p4%rm}SVH}k9w;Byn!HID==|3D>NL-gNim4jK&=XAjSD-@~ zd_1sm!l2m{GixDpb2DX5_wC*1l0vA!FQDSK!Fq@jV$Ov7mGp8hAPJH6N{t(~fk9py zCu;AJ2FK6W!zin!96*kJ0xUgNZqI42sgOP=4S*wyW~?0R(V|!V4^+kgrO_aSP^MGd zCYmnz$xG{gX8hj(t7j%M8b}3Lmxdv0H%orV#}yC-z2i(^L-6%wbl8o zSh5Z8v(#`?_nB-Iw6_4E3{O~Vw9QywaaH$uGbM83ra9v!&pUW{Y9XW3pMhuSV&UGo zs?c`>=lXqY{b=l~?aYR;aE9XT2>e{En%|hB{v6_SYl>Ar4kq>;-3XoVva=8I4q@vn zWV#MZ)M%yBxL9v&prKT)aGFg~5&J&i>QkQ^_gS@jYY|=}5DP^p@Ck2#LpmH_M_@`m zFB0Xxe9Bhj!qKV07XaJ%iDu6rVVFccG5f?EX5^$*n2CfSp~ z4>|hzQ)^1_&iaU6x7zZSkniLfDH5(A$6QPn;q7fSE!Byl;FWZ zd>^j&^#s+G-LX7cB2Wy~idOPf+we(nG>i+kve={C61 z0`)=WUvI(&$(~_<%h94BPDQ}ke61REU&-`EOO;_^IMKD~M*fWdVI)C|csiX>aZ~V?sNEKwfky0JK+#4zO>M!avd4l6LY8SNH1;n;7Lgna?fou1bvAq3peL z(D~d&F_{!H?P-j+?T^{w6g_|-bW(w

wcVd9TWT)t}<5p825`(7@t>InCF68vSmf zMEG70(&Lpc`X#zXZ?sD{Ul80D*2audKcX9>e1Y&C^xK3xtluT40TPDXL}cwb<7)y> zh3u7)VEO_xXbm41YB;Hkk;E;%nWf;$D+2p*)Pt_hiH&-`n}L8WmOYn^q1OeHPnf`MWsSk2 zxgtofMQjsTX$&4l3EHylWDfa<^MpB8Zl-zW)b_I=mYlBxMh8%oC!u+PX=tm7`!qxo zn56D=lGAlFxxgz5CzH@o&3E|3${Xf$Xfva+fDN(-D?K;Wt|zu6(|D4sBym@xDjKpp zgJ8ExRThMKynoC4o7hLln=kHa&0zIN@ELta%Iae3#l6a}XJ6U(gB&)>qEL2y)Ky2j zXHnr>>=d*;L*;$!zRH6n!C;gMf3@Z4Z7_R%b5a_9mG>N%@En7!#abB^AY9SO-|NlC zX=&2Jrn$Hto1~N9XUpYb&nqNPW;@_pCn*eh=1zgPO<0a8c6QH+JgNpGxej+tBC<6= zj0f}G&2c0FY;c)rDi=5k5KU)VP0H!yo?;aqKm3)M%j-UnhHVcL+#$1XITdWwH79ug zV@t}vyj0|z;nlM>QjW28#GQWp?@)s$v zrH27Om6IdD2yTo;%b<=Qp3KyI_iS-BFg1QoyAzUue>&I;FaL)VvR0qbbGES7tDjGd z@+>hqFb*o8NP8+>9O!Z0ZnKk=uhz<|$u^l-Rm1_Sm<$iA19O5c;V4$OkgjJ$0!$ct z!w$_&34pfkQS)g@^8MbYr^CX$G5%@-?q5~z3>E;?N;6E31_?xzA|JhKZW8b*b|`~1 z5DJ~~XUc3bAAhXt+kMmR*%(C&`v$jPTfA9^9Z#;;Z?47M%^2&|+wiTww{B|-o?J31 zHCMXoUc^&nA;hY|T@q0|i34kk9-CW0k<&Mj)7$m5jf0MFJ(KTbGkxezZ9c6T2pvzA3iov0oD^A0}3z9Udg{C8mHq_*nc&?&_;OmwobtepT@qtG!fKTGH2L zaXZFOzoil!eSlCd`LwfqPu=I}yu-#UC&8C+)ekp71wH8XjZ(tE|v#~(cQ@sV?y>ENN{4%0`i+CQOBeoN0j zt{LC>A4zB7*W~-YeG~;j5J9>nrKM{yQE6$(0aF_39yz5u1*8R}W7O#GMnXDANy$bI z82dc?{+@qeuY0fiy07cJ&ht3l$1iSY2F^F!NkfmmVWGs9Tvo=f2}FYfW#}ihk7-(F z)T`Fbzzj21zQr2J6M%4g#?+PO`tvTPk*UkOr(2r&Zl&fE<*TB;nOCVlj~sInR}ZF_ zn{RK@G?M06ssar8YYQryq(3n0E2&=ZC+vei1-xvnlaUPb%4e7En!Y5Jr2X6a3E^d6?7v{c&Lrcg%%FR+VT@~!3`Qu*dmd1C_X~zLy|s7|$^iOV9sk7h zE{?OrAy4p(hw43vRI+G~m-B$uI+O5BbA>BNgrehPFu z0{`{@bz%1(9>5-KuPxdJ{KkYz(YyTo*y+!kKq9-t~U&9c?Gb-4{ID;6{a)Kwz~$v-ISsXmd5K-( z7Xxk6es`5L4p*|@A(y{C4Gilvr~HOqR_Jm(iKa<9TESdbQPlD}IZkNMY*A;G+D##x zfIUgz#UB1d0P=v(gLsV6!k9uin(K)!4MmN`>pbAvqeyXZD7Bw{!*ymvi0&T&S1w;q z7fv=VFspCu{Fe63M(a)F{a@jG&R%-@SOVb3;F3xsOaS;}yl zzwxdYIHSvlLq!WHNFbbf*M}umEEqySm-q(#0GZM7y`=lpY7M)0Ln13HUy@X~64f3E zi+zh%M%^Dj*YG_}&(kkE98A$k$Vu2USqDw9jVpX+Y$P8JE3m@{RXjLLdz+=OyvkG( z{N#|2ZDmkiG$gY9T~*J2c&6|`CE?i_*)aSfKdsmgLz@f1a6VP4_@pvOiE<)^kMF|y z;?eWs^=xIaU;ZBE>_m<(Srv@g(1`gC03*Fdu9D!dxhjvDlOW!Jj|~?5?e>ULb<39k z#mZORx8`KMZ6`EW@%O*PeiLNj;?<)G6$HB~G=`e8JTK3S5awf+;+>gT{jmYIUd;1k zk;IZm9^pMWyUkhxId+1jt{4ZWz4C8HUWfjS=mS4@_S`Nj%3fY{7;`dwp9y)oIub)W zwW4DzKLyOz{NqKeA{u%1I_OrSS7z4y2nej~a_m~6m|q$@ zPC;(iwXP#4)Eygkva4RH7U_IGAmg+V)sRPXGy`!k zS*b8v%mC@PG-auF#H*+Z!*f#FMo2zWZ||McQW12CciAxs^|9bwn&_jxAo7|0-wg?* zMSrEhqlGjtMt$*mLIMaB10{y?`Y$A#cB8rH@!v-;@em9-YMRB zW0yr_`12@SxLNnFm$kxo2=fpdg}FfbV-U_%aQBhlJaDh<=ND%JyiVkN42L6h*@%CW z8_Cw0jaK|WeZO=k7urxA9he{BISWsSY;EUd59HHtNTNO{T>l*0t2jO!!Tu~@p{K_j z@4{t;@aoO)cKtixIxg7AdP_sJZykts-4miZ@S`e8dRFLI%QxcYLSigpBaog}Lb4|$ zvepH8wr-~sn#ZpR=bcUvwe6k4b&v7TcVZQ8KKU4#mITHjkIgY}y|Zg_vv)mm@|{an zOzKys=)BJ!#9hHb)RRpHuBN!_`IDS+$hlr3Ajg6~pnH2Z8vPa&DiIt|X3%3m$g8Za zz@)$!%nPMbIT8p<)F$oBvS+X?wTT^8xOI8hYWX#^p|(d)q`WoH-#)J?FY+^72!@eH z%Js*}8}mC|#^4{F-T1ZVjn7;Ehj&uzR(T9YE1Vp!aQv%5DMwZ=;>I9ef*jkOqEl*^ zTTs+X9U4w$g;^NjBXWI=ar&53<#X(d+|G^pvfUc3kMHYw6UuH?x4Kq<{kb%AY|S{$ zp8J1zhV@L1nl60As@Zs)-Wwx6*QsjDCz`r5+yZp#zWeSH4)@t(5*E^+h(Uut)=nL#J|+r z>^dbwc{eL>x$<6a$672H4iY+AMj!8>B*SpSHu3}Plh*cNQbX-5IpozZ$jHwXECRs z*^A`GBy$u3NPCb=ioD;5ZZ|9|&M96Tv}tZ_PW+{#^yG0)R5v~#b9x3~Y*}TxVRS$r zyrxk!DF9=Semn7KQ2e_xE0$WgRf@2@M(#;vQR2In8FaWoYe~uF4y2!@0UVRxn@1Y} z$Y{l01!`{mho@v>7+8YwLWvcT;DUf{CEW6a`UU9rXD6E8-5UB^B3_B}oC-LWh>Vm> zT{(p9@1&2ZN@T~fB+a3cCT0HM3Ww$uwwF`aNm}PrDO;B_;0O#wVZYn5HU_#$b@3lw z{B7zF!jmAfX5hsELfse0Y{?SlCm)A8?N28uSfA4P4#OZ9920iaO}6rzW|@?8Y#o;_ z!$mRkaHk6x$z{riI@V`ZE;Q$S#!w4)$A5P9nN9&LxOygB{@d?PAmnhB06W(F2D%?# z;=qa#7WbP>y;BKzC%ngJreRp};{Eo;_k|%hJ9A=poC|_NS?3CZ#^52U@<`DFQQ?Z&whw-bDEVH$ym9m&Qi?QMNy zdsUUGEUTnRshsj{WV7LIWoUB$- zQojJGvky%3R2DCsCtTII)4=H28fB7#_jU+pxqx~997J4jU#e-zn11UznX2dB|DT>B zp5{G4TSFqvd;74y8)iz%vyO#SlUkc=z;PS++~tbp{a>#RCDGGsTgrEw)mZ8-3hRu| zGDAq^zpVGaf?jn~6^BzWCIW^n-X%>*dkxh}O??^bysr~qBZhJtC-+hxoO3BNAH2j> zrexe9g1cXTqHJTeEm^pu#N`QHIz)r0%mCa7vF?zeb%_r+AK=To5*$nG+6L|`!PhF9 zW*u^ZX8Ae4(Ik+UdKD9OF~#qev#X?~bz?uD!GfV==~m@Vbt|{aZ}sq|-0w1g1%vvS zrBapCQGA1LRGTa(Pi(eluji+XnsANuPA>DI9ic%M`rgoeJUjPGTGi-j0`*tM_)a##m7wwDZ^{O-2)#xuLP}h61P~e=k_H0oYT&m!v`l0 zZ8r5Qk;I@92CQBWa?D_O1m(_dU%D(%<@w*qn+@Qb9d3}j>}pctvcxd_+x)JgaU3{e zqHf(P$lDDQ1!;eFfQ9Y^t=Qwn?TulOa^yS8uh;Qs68FMp{45RYF~Z$-DLxI;O=!c- zV{n~whWp@0lw4{LX_^QG)K^Clh-ouI-5Qh!jlJkQTZW{30KX03D~3r8LcOQYPeHhb zkZp8Ze!S#{eo1Tj#fN~$buHYBF!aKvY|zBIoZ`oscH5$r z8tluh;koTu;q@ZRLtwq+3IDtZETWUguJOFC>AMZ`WW9Uw#>bbwRYG*Gnyg6q%FB5W zYSLrAx{+~D0m1~%JUvMpJ$o&aOoMUwfbl{eL;d*j|HH#$!6_hKbdh(;VOR1V1o0f^ z;-eNKSKiOJ10NSxkj|&bZ7ezSryWwJ+0C5o?a}4uuB(RdC;p2|{8sy0-2YKtlgkOz z9G}31k}^7@>AghT|;_+0K=Kp6ZhCcz-eMTIzT?-^ z3!Omk5In=WruLJP?_CW{>iU0?cAK3W&)adxsV!g+ENX*%KYaHhP7@b@p;~n^FHO&+ zu0_+!zMJra?l|%4V4=$2RDILgA!Vzp!Hps$*1WxH10;}bXwaIgOahtuZu2ATYadUE z(4M*{Y`SwC>nyhBJ=NR!LOn-R&0)<__ZeFY-Fk$)N$F~37@5Ej_Z&HBe)&44s;aQ5 zsWbQ0wsv{&X24I92iJrO10b}U1Jh`aB|=(|ga)D_=)}yKyjuTSf#03%(yB0|Rvk|P z$TNHu1DYaLaAJHe|BIZG_qb;E;U)~)yLY|?*o_qH)T z2J6DJz%5)<92e4Pxe6DsVL-p5LDbPoZW{SMdK}%H;(Pav2A_9yua%7YU6nmE#~ciI z?vg>mZMs_O=l61Z(W*~ZjZAnzM%z}OYSk>s0d^89H#Wnby%hjTXAB zGb^I9v4~H%f14vGr1!v}{TrA6Y*yHMJ$4Y9kCL1aF;2nLk))H?-bbakA#@ z%C#|aT2+B@Cwp*pABsH3&u0E3VCts1oR@%=or7JtaTQSz0pbMhEjepMnFhA~n15_9 z%o(Bjh+h2sOnY54iU^oN!W(wRd7b=3k^i4nG3 z#ZPxCgbb}DZkEDhPw%iB)e#aRyc42lC5q^GDBkV9bSQbHCqrbrpz~Hu-p8l(D??=r zI>5w1RSJy>=U|OqQD(|ASG38Uat|&6#U)y@3-`|ZU|;hnT?8&v+kK0cj(EhZ*Qx}B z*fO=bRnwLL=&-gZc4M`(HY>-?Ah=yuz3FWi=Ee5bMqi-8qVeJFb%JLgNGD#m_J@Fw z3s?Q8GrG4wseccPY$*(l8X-EKUPYpqzu_#ej}7m$0fAfMQJx)dm;H_C*$PA-$Nky; z%KIDBrji~`6w9peh#cvG@Ph{)$~xf0>HlOtKc<{(T0nxb6GDKS$>3VUUJaA$!3}zV zwHLzWp=Rb=d!Mvf1`qL*`?um#O2=ZE7qCxVtf@|WVPuB_tQr5r zCoDDHT8vU#pPRmtGFr{^VIsn>f}l;8+fh1tW&RiEN7wXf>Fo`tH(Xu7?mJuTlk3kV z3U_Q8V)Q1)_iY0SZnPg&#zqAU8N6wWZItjiUc`FDr=H%joTzVgMPkKVA~JW^q66N- zYg?nLbXWqs!~0Ep?_d;G*5`9U$CE*h6kjiKQr@M!fo1OF`a3eT1vG{&P=X32)gG-B zwiE6E%7`l%jc`obc4J7*P%E{ye$rNvkK{hiLB{k#;f~OTg}-ja3lH`}9ESb9lZTO;#AU*M7KA1xAsMbRYwq(;;ZtFm zNv&-t=WkmAribYdXJq}gg!F2OMzx4U3TKO`}&|2AbcZqvq>LHj@GZ1GK>|bep1n1 zngG+`W_Dl|l58RASx5ss!S76G+qd_z9>-wUs1Be{gyvKbZ>h0KZc($=*!Q+lCJAr)@zb1;=pb7S3Jn3a#x}h6r8V zP4jOqku9}@Ilo#{IFk|-aRAqn4D)q*7|DZ09B$Llws)@sRFHWXO7$UV1JkWh0tq_d zg9LTHZIgczdlBwV_$S|!0paTBEvx0SD>x&H+{Y5LSA1xDaSVR$VoF6GjEfyeVI@o` zm4ggfPC)f)0zRWtafR^Xl!(pT3IwS}fUw8bU)blF-CQ@T}_WIdr7}r=DVowd?~~PV^D#Ejk133+XXWpR1+CA6mK6?oc#q zzd1>nDM>BLz-f-~)>A<*sWnzMJ0 z@+?q-#a71pR*~C!-w|&ll1mB+1`XL`GtZz==Cps>g$wQGTTem`QW}=^r<-2W5+=Qj z*SGt95j)sjqEpY93%_UYp40zOUYU2slIteOIXgHLjkUdDv4Qr;1`uw$p!Ln!@l}4L zz|_XRc2pLa6=qY2F6M@Jh;DI*1ekYGI&;ML+B%LC$681I*zXh_62Ha2!#lV#U?J{5 z%r*7FzMNy=3_7cdj(3+K!S$?<^3e=i*C_~rxs%|UsHR*I5B{&rxdmM$L)$PDI3{7* zRUGPpe-?9;JX5 zxj#5NadIhqpkE_U^UoDBHM}%c4m0__vx!KAqLWG4XHWYo!O7H5b|`PdHs(T@~! zQ_OO(XZtkv`jup(YsPYyy!Snq8U2q~wKyst*qX;cS#)&RFUa*V?2bG@9bLP%WsiEy zh3w1|KVMb+v2gbmvMkM(lphtt|LXW-_EmE%l- zR=$B3qJYYZuav{9^_Fxt-0Q6_E9}^(hRB#JL*q^sh>gS_FVKg#^M0;Rnc7~28t>oK z9Iw2I!yik#YjgOwR~Q}=E6?RRpjO}Yu5k|j1H(shb`y%W_=TFWbTzebcoqbYfT) zG%qjG(tckapy;wiMmoXTrbTFfwi;=D{y-MaHxS4CSCsnmb0XT!*A}8U)7&DxPY+J; zwylp=hk;g?77r?{nX`&h)$MYDQl}2y#2>?6wZthsdusKGOa^kbKO&x=0@D=yZM#3z zPZpg=-et!UX^%*V&RiRWl^yC1kSwHspzD<_CC-0apcJ}msrZq}J6KiT8X+KlA!mO6 z-e=W`JB?(UAy!w!mif*@-?O4PL_7Q>;KZ+*i3~k^|-(8uAVfL_vRC zip&=KuROZa$gNr$!Gh8cbnyuHJ`u7SW?)3nF8*a2E{Ggm{^fi!iq}(_qY#cym=?x| z@{WF9x9Qq5GXW zC-CuLh1`X=cN)RwO`OwbJUINVzv2^H@)Fm!1b5X554PI_k*x1Z1pj1i4VLs@7QSU5 zB>uRcxtB?sk-+|F9|NWT7b?RX=qsD`+FaAhs#zx~)-sj_iG>%WTNUFiuW5Zg7=%`N zL_ODj_}pEJ6^YY75_i};aSoahY{lkce5Pn`XTLkzo>;H0gXZ;DiRgxVv2andm2V>) zk~arNhjn~EX21G9xR=$-dOYr|OvsW@M6O=eP~2$y&Nje;LgcyOSC829zrY+#7Yfg9 zZ!J?lXqd;{_@@fXm3T;1>Q~m~fg~iYtMX_dG{5MT7n4M*uvN3wgQvNRbW^j@w=0`z zg^w5dm}DHn2%KaG{Ni&S;O`m1Ntsu)KWUkYr8)aQd~@iqli}#X&g`l7PbThXBcVt4 z@SyY+tPn>ziP_2P$>mcoIVZ@ffavLb^;*7el5BlHg&uN^_C2_S)g#a8E(1=9O#Z`j z<~#Mi*UH!qs6R565Dfk^nq75d>(YNmEa6nh#Jl6(>;^^!6C1}&uDD1`*jQx;) z7_@K^IOP9~vxwj#+nNQF+BLy`J}5nJ6dLJX(!7^!Xss7t+-Fn0to#X_Sf&eXhxATrd(vmjcV*2#qm4#W zBZXoC>bS~ffZiv*oz9;ttHR!g$j91;=%EqUTh~J)@^1bj72e_TSmIj3Qy@;-(W4~i(EeSahgA${mdZ@<@X@1phg$GVfrDpF$C{wj>#%t>L)5k+ePKZ^k&v~cw03z=`rQG-^&jCbe1qi;{QpvS7X znkjnYiKkFhCw)tr0(7&=>l9xji6;8uF1@LaSDI~pXXurKt1CT$c6aSme`^r@3QFvN zWvFY?s)vLNh)5eH)4wP(uR10GxW#Q1vZ87ypL`N++MG%NGIrA1CqZnxIo8Vg53rVbSx6ISv$48y)NoisnQe z6*GVpooR<6NDWp_Zb=7D)+)UlsngO5PIaO-JKJ!QKg)Q8e`$8*%s68c@?>8Ed5Hrq z58n z7dg4Zie-zPLI|GHTHqlt;wT0cwAP&x`o&f0pJyf89WTr}{&+>5z?+X3OmYv$r-}~K z(!-wACL^9^6eRNuF77Is7Sz$WLNDE;;AkFGx~A{16Tqta1Oio00X-Gyk~0 zV`j9NACB=^@9XQll$y}MK|?lT<@pY&>0!~9Gbhmcg68mi9s^>q;qvHpQrtk zWQ3!tYQFKAkureppo{ z00{8lI{sBFv-VKQkXPvmd%T0*l?!TNAR(5eDLn?ud4wx90S^k6(6XAdr(#v!95jY_ zYnoRtCbiq8bzM}*+aAXRt6ID`qN<*+lPAOtW9Z58CeH8ggA2zY%yr66M&ScDn62l=%0#wVcszm4Y(7pfyaAM zHy)8TLuG{OjN_8tC*IiTnl>7*w{V81_yuWAFqp zewfR$+s>9wp{Bn=vYQr`aIlBb!a^4W)Jm7C%X zim`(G5zW*Cm6!Dm!M{QV(Y(e>0m`2pKaP29Rhd~B?0fzmOm|DDcA9WE=SR+U%Flk9 zT}Pcb&MU~4lQe_{opOBo#4K|@^`$pke?el%vdr4!YU4`bzDjIkMbo}vY`Aa+0J4c* z@CKjtoD4lg{KjQ1T4ua_H1Z<(nRa2*p;BIf@d35lLF3a%1k-J;oN+?g5qK0BR#}Om zBF%)nmkTkf8zZS19ufIK0NCP9M11%Uk5+o0iSJT&6NaJm7&H;jmH`e50I?KCdb@XC zW$_!@#LalaagFO93j#lkZeN%QF)%xW>^wg9dhqpN5;}ukEvc6T*r4?hpNNO<%S!D< zY7QE22i6|kjy%EhTVO(+56a6Tm0Bub5tp)UcAt>Wh1Di~N&&xH;w1C zYedT*t?*VCJhnEL{jUtruwv!n4TpJc-(;L7+H!&2L6V}sr_lQR>)XGZA~HCGH|xUh zItbs^M*OyVOXbCh6KUm@+UAfuDSRG)_TiJT%)%enbwxAOf73IG$qcA$rOmTJ>@)nz z#DJ8i@)$3Lm=+x}b3;urF?DB*0p|o8m zH@fBB;n_w^me7d{A-Hc(;Q>zxBv>pM!6|RV%D=nB81R?KYybkQ7nEoEYq+fiI^pD` zH#n^&z&*P18u_>OcE)N8OJ-=Z%3p^k0hDM5ri82aLClTutDvQ&$W*hBpo2-P`bfOK zohEmaml(n#s@G>J9b@tGIp3|Bx~4D<@*o##j9QET6p5Lc0_6HRy0_1I;+GI@+w`6) ze+<{}N_yPSFuX*w@|2tS^asA{!*X|_yt8#GFUK|r!0SrvN)&yt{#HvF;k$4i9Y)Jw zf|)1mET8hi<%wm|ug2`71+jlT{P!ZhOD+qsoe6T@q1X7q?B48ft7QxcTHo5m!QP>8 zh|$2+e!iB~S0Yj@DihJm_LOPNNt7m*IigybD*Nx`2)Owjozg7~+LhhiSUCYhZD&QaU$=)t_S3iwH4yZ<>QWl&Q%6LGN-+WIV~#GYQP zX<;EPlVhf%(ojU1)rhmnU+;^`n-6R}v?Mt2JVf)45Zs9MY@7F4!B5k-;8RM03`%5AZT7bi z=D~t`m)lt^k#4_UeBJ^WO~4Y`$#LMeWY?$K)1x_#<)8<6_%24-A zl7W5D{mVuemd_5$KU*#S6eRl{(%>KHEk(ez*U<99ktmi-3_uc_^8>~~ZwB0gUJHorO=?bJ=biBwo3lp$FK|-?okoo5npzQU(-D zZHlV;dVDI{v7iR5g2X<9_PA7 z;pd-wE8Z{NWU)KXgnevyKU(y3`4-v(jbB$4F=xqhTs3KzcPrXScG;HiGRK{JG zK&C4zW`L|;2IB;Kl3v=H69evsRQo}X&}}r~09k?rl#>5lhXO{u(r&}?PF9tk!{qEB zEhwNgp9bgv*C*T)-qXs^S=2}OAw`W*`w;=+NaX73deLb(yo=zlb!?ym(KL5~wvbFN zPdXQy|NAy1?pWk)utPBlh8u&hP|3GYCn0{-HT;3gc4*2P3osjfPJ=i;5=ng_cyscP zKVUBXE48n-;ISOf-TUQe`YZAhh7NNacomKgM1gwq$S;W+ItUnsZNglh2^JXj<;ZGh zskiC()s9PQi)4SLGiFjXnDA?chUea%kpmi z_qp6j$L!x`eJ~uEag8O|nijgITpm|%48rnS4YPi?`&{ZMWEQriuQ9rg-RCsc^MK36 zGxTNpPFr->X(_WG742S9nAwt7tIKNg4`|a z!S*~{F$<;>a%%2(D_G&=@GQczxJNvvy{pZ(Gfm>7*#Ob*(i9+A`R`*N6YGh_Z(CzI z%Ke>6DJ2ITV2CVhmvPnWB2GbUwN;x>>5Q_of$m<5r~Qmo

v4oLeD|uyN41evsX-b8Td)J^9v_hclwJ# zzkzVCy#6@SC4kV+Dcj<|N3WNstf@xf10%YU>}H~}Cx9CV>NlT80ZrW!Cj|$|5P|c0 zg^rGy#Ebj2xbZX;3X9;uINXx`J^2BvvT(hpkGZSv`WVD=5jh52bcNpzWdL(B?<#$? zr9oe5(9Hw>Kn&%qmJ1c`1}oVxMY(buzpXacE+_H zYL|{|Qfa#8!r5bQ7J4gtbP$Yk|KVl+Uq@pvcWeun!Ub~Zif}umk53Vp>2r`M7dUa| zCNCEim>i;xwEomXfE=SU`wveZWo}s=;E0A5IT&V|-hTWo*KqcJx_mA5ZS2QV9&%lR z*TkrUfj|$M<28hFHN1nn|CXcmd!m@Tdhr*@G6xPN#IkHD=bel?{V4sq(V7UYV@k0* zVd~J&)pez&q^D5^+Ke)4-;5+#zpAh@|73Q+)Q;y!YhEIX4xxS)JbK?(YXAFl&rc6b z`ttP)RxGYC{X=>nEW+g)+TSP=&bZXk%P$%%8OBh5ZdJPmHc&;4b%*a5SDLdniljd6 z>T5-?u8Yt>7rh^fh9tlCTbL}t-Q}mR_!i+OLft|YcFR&7b0nS24XQ6S43uRw=1`Fo zD%&fR(FYBLURXwtqQU@j+T$RfD08EQ4`X(V1=LI~VqMk7_C7jYuG@ zu_`PK^=Fs5J~CV>^55FgyU6c6x##__mqXu}t|@pqek403I+?|@2l9YNi1J~v+zs_kG%ia1V`Mp4!QA<#Bw55!+vN3_~ac@`ARi+qlbmMdrrLS zz{_K+ix3iGKwqacKl5bglh-T7N-f^VxyipoBTu}YVIyl~SmKv=N&)Zhi!L+>x2_7y z0|39_S%J3db#Ux@=fbbxD>IR^ea$)UJpJ?3qUgPKN+Sx+rt!*HuE!c)Gd-F*dn!?X zVtwS(@9SA~kR^RBo>S!P_6xU-ImMJUEqqM#E|+A0!D(B!GqHRTosjnSde5et@<5D*BsB0^v1}-2%#An1OB;z_bFl^ufZoi z;9*ijRaF2fs^Q(We`uU*t!e{Fg_Uo>HV27YG!k9-_{r8TU0W4!d|*L0x#c$O?jv^E zoanv=`vT*$5I{z=j1etdMw2L#T-6PJOEGMw`PBn3I3~>DtKN%BdsCp8^yhm*DwYqO zd$+aG4PF8vgud6RPHSA?ED;(bZMxsQWBXhDln&zG*Hyc4e6k&AafJ^I2ROk- zI25+db#d4Xmu(j=o4CQPXXF`;&V%!KRzX(C+)2&XaevmZM=Pf*!;u~H?e`th#;!by1DUTv%_8+~~&-#UeRc|C8saH*XXNNk7I zx)|;%XV@=&d7?Z(sNqnhe!Q(4O_lZ}d1k?*qQPG)#ob)AWjyrLLM9{yOMHJvwtlco zkb&zL&+4HDBg&7c-}NP$85=7zWrCJ_oER5yd1s<^YUPUpj3O$}zrL{;Gq<6692e)k zrO~jl>wkf*eC5u^xrS9go4Za}p6%5`4_(Qi4UzGO2mEAF1r|*Y=vB7j?{u|zuy56$ zGgy^VlDr5XRJ{T(8FsI8pbZf&(dMKmMt6V#kfMRBzRlk|o*}Yx!pW&~RaWW4=ej3R ztN4TNZr~hX)BXw3HZBP5^c$8Ft*3|`4N;j@97xRx#Z84E8ta9WX{Do$HNV_>2D`K7 zQb$kl@XyoxyU@I+7)rivvcXzp#GBa$K3OuHf@NdaUltU_-O?QSDqPyCb&e}qd$v|t z`KyvC;d1^1nT3&Zgp+l_0&;F%ifJTvz^Zk?7N z4Y+=xQA_MASyrY@>94q=Y4rsRzKLQem(BM{*p2!Fr32cNt;Ar7S#WrwEsjn~bSeV9 zpxDwm_c?b+3tkgEdbq9GeioeZvNqn3I)#-ORYx&)PA&cPrXC@uca`4KK3iQK04iHq zJxJ5GpVwPi*CLrulNi9n7;#0-7=6+>t&>_J#}_Gm5Gg;8AI2-=H#%|YkEM6|j1ZO5 zE6q_3YIWM}vh}#zMhSJ3+V+B*B*Sr2Jkxlcc|Ne}CU4I0 zN$FQ>lPMFG^8dQFld@ZPS`?R zu2%1}6ouT(8Fg9&8whbUa%s_w5l4?cPtBW{ephX9{=in(nGUd;;6k zT-l`R7)$%(QvEA|`zwv~6q7>F^(i?2KfKz!o;IOp*xx}^sGR6A*ED9@h0vOY*qO)^ z$L%*g1b=j&?mQ$)r1%*3L)LKXR``CW$h&JZMOiZRelZ9rB*eZ6MDwm;OWaU}W`FWD znx{`L*KDp$;7x7GFN7P@F2yZ$rsdrx$Ka@vp1( z#_X%iP)5J1Bln?Rj>$_^!aIH($m4YJd24B@)0-XN z!@8MH{K^1+CkMe-tZQnS5%UN67kkC3Qz}@|=|8fSzCEAsve{bo0&p3dN1K%X3JS~_ zJEFZ6F=LRgxeo=1Iu!;ynsIWpcz$1@Grvd6@pX)m+*Ng|NiML}umC&Z!V)RYaQXuD z^>f#3VVkCC!|<6j^Ww{hEfBbNAspruL}VA2>J<(quYbNMB{^M?>z3`fkKw~D#b7Vf z#xi7yuHP1~_#=j}pS`1_)Nb_syfq$~V7h*9$1en)Xz+HD7f)18N*Okny@oAIN&CMl zGnrVu8TZuBY>c?lx#PyJ;e#U|X`nA(%! zZ66-!h;{LBT0Y|l01o3K3S5@;`Znre{VQf%M}5=Aihl98@h}p$y{q@%)KfYYT0Sy9 znGvD1r8@=JOMsEaT@mz>Q9HBF^3=-J5D(pWb&a=R%8hh;fg)zpZzqn0$s%D-1FGDH z2%DJKMTzHDPjNO|mCqQ1R>lpA`lsk=}-3M&TSCH%0hxpU;+Ve&sMh*MXLk7qRS5 z!1pPou!2JBvWBLn%bVQFjrIh^S&T!82@BVPaVtJ8ZB!Ep`E2YJQy1^w0(1sHXAgu6BnW7(I{W`dN*2z{GoNMN!wi z7C!8es3xMW!y}u0wCR6UJ~VXo>6pIV>3(&&?tC%fd%&-t4u{`Up*Zc}`m)Ih5%>Bc zk!jBA72sZpCqfNH7pvgpOxcs2L0o$=MlfdCjNhf8!>A~mUEg2S4qkUH9BTX(Iay`X z=1zS7qjV^rJ84QY_Y#H!k`VNqus1n$h%+{#+;&AWXcz)Y9^C}6WVteY>z*X_k=(#} zYCI6M4O$c$E`?*FtSd6U(cvnXev9AXW_1srZRO2ZDUf73q!Py(P zYI|u$4|u`_1_c&C$M>7?zfk1uyhpuWJKqy|t#*l5fkzfVzPM0~wp2m(XL_gHJ4!o%cznN=v=IZMGe-<*q1)nb8^h zz($4tFyfzw+!(U3eQz-d!ZGE7UHO+&{!X4PAjH?xy_oY09a=dnrx^#4Q@eGyN$_vF z%5)dixZ-3Z76UmcUxDafRH@B47*$^;S#?@fSQo9o3tD)bnl<)ZhCeKFn-dz{sR|8G zLZ{*`x1ruF7lSrdQ4o!gvPa0Do+F#C(&ASo$I6=^KB4Hu0Eq<08OWF->RFhN5Ifts zQhU-{AN+9~A4J4=>dS?_emuQS6-~X?KU*_N^%okmhVy^#IecG0-n{<#iR4+1tWFs4 zHB~=E(%6L@VyV6ml0k7by1r#}$!3%~7&!UQp*ao|&v&m@s4q3BUJe{RHmBFFkW&4U z(}{S`3RSB@i9G%Tv>}UbH7(_kSf-?J_#E!KY>m)p_OCvkyQ3vOY7pSYQn)-tPxUOY zc$nZku2ycV?GJEftv9e_6g%5X@;@a2J7P>fgST;Hp)_bZT#BP_JYcQU3NYJY$PORenp*FiVFkb2Y0^%yfg=<}Gx zt!3~z>^!=NZ%y=`Cv@$f*WI_#JGY6CnaYox@czA&{gft!&++xa@ZlRK)rUnC&c{kY z2bF8@Yoj?i60JX=V&kFJd@)HEGS;%re(!T>kl($X*-_>2(=}X<0}j-p=(QEy_A^4wFE8Y~Ty}@+gh?lR3Blm8^*YRO z;hV1>i+;I(4s-_vLCx7-_F^X<0|EeRgt{_|wP-*4W^1U66@D&9E_JtQ&pznG{V&So3X+EKEuFyxc=s)D@OB6O`S=u zfXlX-vn%M1>n>RSvW+BI2Dw-rJT%~!rCG4AJon^X$nN0JeU_7E7FZDc(7=1DdAs^e z&q1E459Z6Q!rj!%=fn4@lqtDrovDRrHEoA2vCy3A$+tq!!BbAo0u3wk?nWaF@5^?- zZU~KA_5IUni**9>TvlH#_d;n7uWw_UGBv}KHD_4o;j#{-qsJ=v;}&FuOn#SDIilU` zynp7@txJaW>+oL_KLduom`hMdF_B6C3J-jt?tgEw$7d^?uA(0!%iO)W`QjisutE^R zUgjb7*Hs1^Q;B}3>%`J;P`vELY(dW9p~$as%^7ZIqEZxh2XO4G?R!{2Dx}TjM(KM~ zL$=2w?*GtJgzYh#D#>&yt@c^c-g8-+L$^QST7Wro`q|w~J)X%ImlJ$oUCK)TlAbom zsXRYAXmIm5_-YR%9jLYr(sGKu{rBB43?e(wDX8_#Nba0>H474YI4qT7ak?G*xqha@ zr>FQx1(g<7spOqyX3)FUk<3?`b94Ph$swy1>2;D93%_SR0B(;N&z%e^P4D&vin&2| z#KAC=qrT0S*}b!SW!>t?_?5vRo)pdu;T6wMSY=?b-$Fjxh9%mhI^U_pZr;fk>W7b; z`z^O$Q}8@!dQ$l8&Q%#eX6ov|eJ9vE@lg8jgQFXLzEg5DuB!XE60;9U3{t7Vwc4Y` zKgzUy@xFa^mtwS~R3|Q+)zvqa2vnyD#>yTTSe{@BP!#FIPRq;%%FOwn*zp^+a5zw5B5s<3oVbdkPwE6!ah;7bdi7UPLZ)Djd(C4`GCMdiu@ezj7|rfbrjAD}1)gI{ zZolOQT=GtB`+erZ;6v}`Gr@;)?J2T@+l3E{YKZWX5-jK2mzzTylI1=C%&z4`Ki*ft z1BnV-kV7&1O--rzPbLffUkt82u{NG%ypW=+hN<4Sw=2y{%u4nfZU*>jzaXuQ_FA<{ZPuZeM z+?C4ql5h7?_Ph6$^q~7=h!DfmIQ6e$a2s0ta`k=+Kj#l_p)(ZcA4Ohg*Ns@wRuxIRJo3jDUIc{HtR~ z{{VuFe$vy(7{{TLugIjig@KPU-(?u))0K#I+dvIQ^wRXoP|-zBbijP^EB}3fr(w3o9dY z+d0Mn=z8J!qvFrSKZ!Bg=zkDAMR}{*q@c!^A_nrAjyIuLk%N6(2cgK~p}LFhKfHE_ z@~^ocbuauBi}r^W+ z5M{)?e8hG*ugxj-CY581!xB#lc~>p>RpnPJy8{7R*mtIBo-a)gQnb~x&4t7_CGrQ$ z1cBS1P!}HJvi7xa%go4hVD)Kz$LJb=!BIXu+(W#klN)yu>4SrwN6B;4au2;aPxvZF z$J=67D}4^(RDrl@t~2eAwW-a1dVE6DPJ!_s#QGa3@~ri}MF3I;)O&`OJdBgfZo5yR zu0RC3k|M+fED#O==fBdrooOzw?-#R=W%nPYZT|oTZur$|!W)}C0c33LU$D3*Z(IX@ zGux5HXs`Sg*W+v|vVUY~WFF07Hh&ynaq0dw`N#10#`n;8I`6~UXV@T{QfVWRFKle) zhIRAHtg_t88_pUyRH8<#s5dvtSgEf!)jTudWW1Jpt34|GOv5Q_2}o3bbY)bJz#Mwk zZy8oCrzR&#FLNu}_{ ziRC*ghc?$6M(^(*wVk8f@$|3E{{V+R0Qd>0Sv8-Cbo~m+jlr4jH3oHdw~=^QQ|9uQZIc!R*+1btTO+efmuic49(Q)EnVO%!f$%%Olgmj`JdF!UpgKec4vDw7Xm zPx(>%9WVSAgW}D*7~8{o6Eczr`!&uw{{VPh*T1-MB|`qyGR( zzpZ{k=pGIDN2F`%ta!J>{{U}mxDxW&16?J%rU{lGE^VS*#AmpNvjE^gQFTvV} zh?i5l(CuTjEXA!Y-dOFnvXB5paCqSI+@964#!24NdJ*iY+RiBb?X~{^g3EkRTiC3x zd@-O}qv!YXt=Ez;G6{9LI0vBj>slKB0Q?p=;|75Qa^CnuLv@&Zj-uypK-tK-IRqSn zGBIC+UlV*K;e8jyvuWCPoVG~O!#TG}?j%U0njrEfKd2~Il_S?Q%S~3?42~KbsNC}0;1Ca9yw-0MQKdCEVtth>+m#=+7k}_ue~Y39 z8h?ZIIOQXUbh~h*^!W|K=jpn>8~FbK{{RI9_|yAG#TJk8Ust+qA_rLpk*8ZFmYiJ) zD=R>fyUxQS9(=L@Lb9r`&3x~5@hef8WQya#UkjdPa0jN@9eZ$n>qcm`W*X#6h>H%9(DuLWouZkZl7 zo?9_W=G2e$kU2$-5M&kI4BniU z>Hb>&hw_V}{8_j0h0c|({6f~^y}Y>6ts3>O-8aEBP(ZK%Ta}5R!mfB56fQZh8`ZQu zQcY6kOATvK5Hzl1@~pK6k6~lCVMW?k9SZPAPAm4a;P?CzaAaQ*ZZ-mdhk<^P|S`x9Gf|@NNV)@-@Dl@wZ+} zbR}Y6?0r1p2MRJUI39oxYX1Nyv3Z>&{hZa>KhtRN>#O_4eNX?^`wzw+75@Ne?*x1s z*ZwYems!4+-%Fa>1-3VKer?2Z&lEeuPo&%es?U%ga5&=?_>22r{{X>j{yAuxt9&Q; z1Fv0KrNx_R+QzeUZKb{1$omhIqutrs*?B%ntdTs=6qzJ>A~r(x_1Ep;@pAtF;RozJ z;@fK*=_Y>>-f6lr#pf|AE!Es*T~^ub zAW}@17a`OnG4u?~4%3l^4Y|swBz?4>@JkPjf3*k3y(DNa7lm|fZg}l%ZMAr0ONi8| zV1m_ME@Y9gPCTg3Kzba1K|ix+>_y-Y+55uQ(|iH&qFfuBba&N5K|ZH(8=z%*A9-|K zF+NNxsDzLfyYX3;wt_jKLd)Ji7`_C4!9G3xk-Sx)YQ8gNJ<+U(4d!Hzt$%9y^^Gs$$AWa7M#ka=8V8B4r1RmDBCZ=whBYEZTy6~b z&j$w`4SdVtX%Tg+n@F$~&Yq~ljBK1a{P_CUhh84hd)ppU8)(cR7P(BJ7dF@>B92E@gK^#<3b7DBZ=m9 zrtY1Om_9Z97t_8D{06twbd~XI0Vv0B4Vez8C(`xq+@U%?LXoQRIm(VpQ4(C4**BxD$jpJP>>n;_rhz6ZY>L2aW(kRTpxlJ4^3j2z&3hF>i}9G7F-zDNH6g1UTMwzl|rci}w_Upq_q zWY(5y@!%!apl{=wQ=dHK2|$i8c^*avas_3P(u~|=Ew6S?VO_<>D|5r+wby0%CE%?; z!~$f7!%5j_Cu1YD4>QQp?q*`fX$vu9^EXq?dDfTU-78nWmsQr8VX?DAad#^pl_7|3 zZQZaG9R5A)-aJjHCa0zNXTUMX<=pst!oDB3yI<9sHPI)A;$WCXC4tBA@#qKB{{Yusg8DNwfitHsAEb<^DVX|k^vBWMF2pGwAaw&xR`w>)8dH1eFPeEqTxYwf?;W5UyEelN7wORzS+ z74b#LP^KWe4}>K6PveHwHtz7%`L~XBjaZF7=zu)YjN;aB zlrR_&$@3hmlAod4-h~O0*3qL@^R7@1hX-;K`Hm}))pRH?^}%m(3eOz=R5HfO8aIu? ztE&USQVs|;+iKQL@Y}~)4~2C`d9+PZ_$HTgBW^Iuy+6>VC%73fG5hP4KQiR~R$pEH zPO6uY@yLbdSbh;*;>!+-v_6+5ry4tB!CWiy1u$){79#K7S-m1 z!~Q(HlMKtPUPl@2h?!6^_5kG9o@3 zwuTfwo@23YzM%z4uUUk$c*ZgJao3D~6>H(V$p?q8%)c`{(DnZSWg8!sY5Mc;wvZ4{ z$bTLy=rErOJiJ^|n@H$BDO&Ei)DVIsxbVF0bDWimXh_G?J*(hfiTahsh2yK&vVuAM z`x{w`$O{mBgkFG*=N$VCSLk=e8*likZDYpoKf$`z2T_GjMRi^wCg;Q2bm|Er+9b&Jjpaf= zTIRg4C))Tq+IaUF$F+SO;=K<~vG{M`U1rK9j!T<8O($SK);+$dBB4H|W06PD0bY_( z=H-1)Mx%C8N09tBdx(5rV<%~?}a*in#YG5Ow(-( zUq_aWWq8%5=)r*m{MgPw!RT>bY4C!^U3cP-hxC2N{U-6D+@yy2t>#n6yNhiH$d~RVGBsO~$RU{Gh>BV@>>8Z(5cUm2dw7wRx@r9M`&YHFskupe@ z@=RL?%E~*-B9am2D>Q>?0U41&7_9w4v_-VJw36*^red*xh{}wKy_!U3360MDCNs5i za4-&)qIfev)?vD{vyV}=iGO$}i|pQDDYftzz8^e^0M7560D59-`o*-?%^sO=3$Y&x-45#@;sw1#3x z1b_!M_b0)tsINQ^ZzC{R<8Lewa9e5T0QJZD#e4YdY@-B{JL_SSrlT#70)N@l#W-7u zCAG3xvPdW#w@`DEI`P#0b#F}g8zSArFGB5*5Z+PyjkN8;fN(N0aB>D~?5QsFX19`5 zibV^YU<02_`uo&2*Gn{E-2#Ki+y+j2{*~l^XjOV#;&=*a-5dvn`~jkPXG4=ugaahl zGEB+=W6*kGM{msYTtDoqs4s;-YA=Sm%-fl5{8gsfI)Dck(|=|*LFxdyVtt3_?H95U za1;0fs~*GD*UJ9@+RxxGi5C7Tn21+E*R1t7C!)t;e-+s$8U6^fT45?me97*5*lKZ$ zRzB(YW2BD__&&qJ7ZQ1r=!n`=x#Kcv(lwjs-*hKFhP^^K0UT8?6=+g;bKWO z`zMRF_CPuWxzwk-h=1R?l-7l;i-FG-{-yAaBiiM7da3USzxgz(0nfXO^kXy%abLDHj)rF zlJGH!xaSs?$nqSql$EZIMR1Nql)@heJ?!(SLYYvJIbO|OY{g;gB_8~NQw;H(GX zT-WZ|enH4Tt$&x5B`CWy{Ya8c-E}M@$f@%Z2N%wuR*-JmbF^{D@0|3hBSKe! zgYEu*r9I=ykU+@?f%?`;N<9b4+1Xo7Y?i3;qhY|t>33xR0N*36Z9%E%@QBHlhot2U{Ty#YQ7u2ji9)@k~5<;}PAA;)Y9k+cOm^FkrygcGKDe$%M`_r}I~I0r?8Nd(&QGp; z*4~wW9CC;+A`?aq(qh>JqSHPH2E*7oTq1F(xTNq8|pUir|J{kLmQ4HlEOJ0 z5<3{=Zim(R$fWqoVXbE$jy0~wt`7sXwb&4#g$L+ zf^q=%9ji{l(lLP>90QJhd-ScT;rW%W(lMNs&s5O#AA`C(TuF1P-D)n%petB9gd-%8 zxQ?fT?N&7_e}nffvMkrg6cMu8(P9MSzV{3Ky*=y9WrY>@7CaiOGblSj{HvxodurP= z=JGu*(^U9dr(1b3=$997lHP61sF^?RsbB}-O=wB+UKW8vMb1<_QQJt2<6b`tvw>tBfEJ7fNkIwRcz#SbXBYYzi_Hq$&0aSY8g*E(rZ^Cy(aO%31ou?g0#z()cMc@w;NozE=agZaH z77x1)tT+q^KBKRG{cF^v2MAc=oEy;^dM=%(*he(i4Y(vyG@Rq*3Vkz*%J|c-#Sg}B z6-8-8-QTJdI?WvyxUR(fm^+aldu z1{Uy_2jzDR&6!wr^4BbRBCz~FrT9ll@o=@(B|44Q*~7&mAy`8rx|rmOZe2-b!vGj{ z+C^<$`0r5g?Yy23@jjMb&80WGbc*6PQM4xY+!#o>z&@VYuMF`NaLcLb`mUB8H@wOe z;kNLyE)QPDxMz=)(~Zwsju|b_tu+seUk~K9j>}N^bE0VyCO}OKON&j4VsVKiTXsHl zyaDECIO7AF@Shp$QLfvFwCU_^-NUL$8zL@7G6@*(QEC1^@fMXCX(LOJqX4;d<0Ot# zuU>elwC^8W*|TShNY zBYaAax;I7HDsWVF+Ii{@MS8pZNQzcAuvEjK3O$5wo)G<6;>D;~Bl^uA9^vxw4W(5=fH~&_gU3@*-DTuNFe8rT9-~m9nQ1jn{9JdzLF@J5=ojw^PH(fkSm|v-~-7d z5_{J@e{l@fa={z1c@g6qa0m{iaxx!C?!@}$xJcnb!KRs22;G(B@J3GFyc+2~ zBx%TF@F$3^2wQuf25H)ICzxbA^n1(eNe1EpatD?P$E9)eD1z60+LYh4p&d!;I&=s1 zG)_wB9l9BQ8y5*}HIf4&!z|mk9#EvLqaMc~06DIf8(k12^Iq?YWg;l!XDa2+a!UDZ zHph%GE~k-<^ImpD?!Rkotx1KHV<&;*o-Wu!Z0zN@ zjy-*&nIR4(ZzL*hXvimP%F3+S9TYYVN22(mQ+9&?08Te3EhM!;!R89 z2Cn`p)MFOL=xt&$QJydrm}e)hGNXWc3|Bv<+g|9m>~x(%A{iPJHM6pV<_s`WS3UNV z!Ov`ST<`X++>*7*f~<^LQ}{vP6O4O)KJ|X;^HH~UxxSg^lZIz1smKG6r_-lT&b3sz zVxlQcJy_e_PgaqNyz(qy>;rJ>J09fvoRWR(kMSMV^Xv9kSF=b^D2?|?xP0iq93FmR zFiv>D6{o1`Uun9Q?^2me&gMrfLgcqO;~4xaj@R|A6k9EXmryOk&l19hRP!I_0PS3o ze(~UQ*NW=HQ+LrE?X#qU;l9hcTlZnHv|~BOM;XsS>Ds>3@D8(U2ZBBt>RNJ2f;c=w zr`!qKb3E4ijCVklOB-2jWd#^WNCSbK=dVohj->r->purcV15tj zQP{MSOJm~m9PJ?56kq8Zh(J|N()LO+OJwAOj0*i9!t7VF&v5PaZGG`;^*%eqOh2{B zaS>e={^ptYJ&n>ThE@enPr|;|{gf^~)#D4z8X{C`T1Dhp9YZSReR%{NSLf!Da$ek= zBwrjnKWKPMsan*I-@J_T{{W8kukB6c+aPH$@d5M~EirBt;BzSPJ|gi~b8G@Rrgwwby@T ztx_xKfJ=mn##c=m%Dgy6@=jDXKt08Ok=$99(8K1KiVfLDa!YQKeb4ouhqEVFFQHZM zC`JA0AOF|f)7ol)`$GHz_VbJp*mG+i7AA4yB zf)C>7#b1irpTUn9*lJ%3bvdoIXuV^aD1s-pwg0!@0@Dl&E5`WEVP;i8(<_AM9o$o;0gJ>q*`6!>$;8lIPbWis4b z+*rC>%M_`|bY)V(f$yGc3rGE}ziO=#?I7^)#IF|Gpu$HKI@S8G-T7BO2<=(fEqq?pPG9+7*S|4XRZ$y-!T3TpvJAde^JQ`4&-3GdEHKv|KwsuU% z129DX6&S7m01o_C_+M?}&1TwNe$>6Iz|0Z+#zpz|{Hx9fARhk!TKHS`Oo9AA@Ymu^ z#lFEB=r+ufLUM=glU+%)9mhOi9&w(S0=(0}Wk2B}@eY1f3vUvE#}WC!ab7IxMe`-H zb~r1+PIhPN`}}_RL2VzKF0u&j-b}xqLjM3Nttagj@NVQ!r^l~EWtJCs5XCaH1M-jt z%#a;`&rWe)nyuuB%&3_NqnzV8;C(4HEfV8j@xO+l(+1nijeAJ4P=4!uvi2BuvBAJ4 zz$5}jO9gCT#Aroo#SAoCvHLmue|XQp`p3hs5qKj@)!{mt_=Y<{Zkw{_PqI~(FiMUP z43pCrMdhX4 z!Los_u39**BbcH>hjWt4w+I>WovO`%es?rZF;_dG6&po9r@8+C!AO29cvHZiwKs~s zF~m|<_-?u`k!d9EtqS;yRaLjsfk$wI>@Z6$?BrpB08asH@RRnw@eYOJ&)M(9o)ywH zYb|Q`Q)@_JSZ*VlN0w{oqDcJ~`AL6IRAB=t=hHJ|!eSXVD z(QJq+w>LU{g|(Cr!ginCw8X2p$|lxk!B@vu;NJ&o8V-y`zjC*_we*d7sNAmLCcB%N zSyhV_j1jgZ&Js0L906Y(_?!0G_*vo)iI3v{02OIEgIHPL>KB?OkzgXqUFsJOX1ivZ zWLFItCKlkvCXsMU5Hht-_$hyZE$-gOL+~!CJgUSWw%^*qxgNexo<}`#oK`e&wVYIK z&{(I>IWzQb%SE?WSpXS0&N}hwU4^xj7%WFl0sjC#_4y~M{{X>BynL3hOQZN7N3$0( zmsgth>hEq2ItZW#^8&iB2Y%7t@K7I$nk3ry!;J$>coF{qbS(6RV;7bMkn4d zFsE=e)_gtqx$$D_;3w@G{w(okwSRGAW#OGEeB0^6MJ?v7G)`4iqejf3GB&9yMtH?@ z*1~(KsV!~!o{k?cs-)9Re2?2Jtu5_5S)0PJD7cApDOkY9(0O7y8l(F~d`$RJWgH$E z*L+W-$1U#tkUj0Uobp8~Iy6AOe8}BHh8(Up0tm0=uH#Cd#NQqu{?-2gO^)YW^5ZuQ zW+9D8ovZF?WRY2P>N#~s z%g(b{$Wu$X{Wkv3_BZ-R?C);!TrQ(=3V3BwTX!KRmJ2n!Vyt-!BbNH-J?gK(PabJH zr^5Y4;_pg{Enx8-@<_@@`by7yt(h~~Lv15(J9*A)xcF0Mx=+F=beI$<7rJ5&yc>Nc z5J(=uv5&^SS^b|i>lyw9Ug;Jme68ZE>uZ9%?Pt_P%2kF#28ELZcgX4uejST#!fkbz z`Jbj?*YA?@I?sxKwMB=Ae``Gxz;-?ux0RCKH#0P$cH%3kM+*_-<^J`2vHt)CXZY!< zc;nzsh2dMBF5ceXQrB%RJm}?PBK@8TrN%t5!!%n@BLMo>BjZhG%fuhIjl9sGHEm#y z(oaD!vYIyU*`{7A%Rk_yS44a@@KV1dYm-8A#tKx|YdlaY0X<12tq>swZjjb0PRLk#xq*#19_ ze!S0;J^SJD-KX6~aCS6qdEn$Q&1m?7;s&*uRB))-Zv72#9vRkQ({85o%K~}_iQn-HNw$n-f8iX~?T@zTY~%`0 z)DkQ6Bjd;?#2P&E43^OBuumMRoOADAr2Z0h0n z9Be%6&kju~`CmnY2alPflHhan{{RXwa{cnZ^||5FmcQ3?%V}|rJN)LnUstqvWPw}&3}9q)oORE=dhVSA{{X_6wV6|k ze-K{J!9f;!_5vAy}a=1N~0T#Bn-pUjDmAq zoPV_E!w7+b*TRNA{{Seq9sNaje-nNW=sG8cEzFV`lHr{Jke%ZS8w8Gruf2SPJ{|ab zPVlwe-Qn|ZL{dnAF#)po$#c8wVzIe3C`uv5ShjLO;XvpB{7E(Ubog%#{J)DnZnA=lasIy(=sk1Cx(|%?`L&M^ z*e;I*R`J5{DN-5Lv%lmDFhR#m``6VU2L3DAX#NvACH0xS+cOe(g~m^*1ob_u;13CH zYrDHEhGyRuLm&1*#(jDYde_sQ1QypYI?EQ~oGSn_ILA(y_pU5%RZUOWwVRDFUf~rQz7!|{eXFEqA^IizH zP&9y;l$>=cGmlK-yKN%g*K40CK^Q-DaqEsN&Bo<4X;W=2OsHa%npBFT{ zLGcp|fl_OGR$wv6!9)K5px29mQ0Hm&JzL5y@jpj?BWp72pB#K`uZ`YTuXUyBap!t5aaYH3h6?F%-sGG+CT4`@1(ZiU`UQ*PZl zpY&{ZCcbvI(Uw1+DP0`kec(spG5J^djl>ugOC4I1`!bUIn@9SAk>mZo28D0tgkRqj z&bBzmNXX;3>6}%URshI=ouIcI9y)M%tQc)I2-D@c$j?GC$@+f3(z+`hGg_UN5h@m7 zLV~26o}Rp)Ok%ueNwd~fdmjG);r{@HdOo$GuZZn@VQRK7e`lt7vS}L8D|>z$?AGt( z+02Uam0-AbMk4{vMtWUK;J1aeJ1e{GZ^u?PvD#|$XcIH&`ud3VOL&a(*}in{B}m5d zr=pIgzC^t6Hm#~ds0r_85$X*HnFD~092|&9Bbdn~k=&DA){EiW{WikqQn|jG@@u)x z%ae{xc_0>FkHl7L1 zSYHSDJya}SG}7hcf%Xkte!pP<0QFZ3rdiwF?vu^gx3S0L-mVQfhbloU?@d!n`u_lc zWa`EAI>|ga;mcV#@jjgDG52kCFns71IP6KsHC(HqlWU?n$ow~;OND^;wP*0> z!PiVSc<)Lbat(&HkUR9S#dkVy!tVjXpXq-Pw54Kr15MW(uhUQC>s&h7L2Jq?{c%vv z>keEVM?dHP0Iywd3m2}rJU_CR`*MU!Pi`;MzYT^7%3|gYLn{wx4 z;79{~@CZ8yIr&dU801yLgd+9*Kgfkwx<1p?{s-D=S623I;^-xo>rvEn$RlkNUX06m zq^Xkf01bf)BnX9=6M_fIO?;iJKA>c2nXz@f;0tD?5FH1L7HJ{D<&@_jd;0^^nnhGR zf-G-oB~4=b`%x|fO#XPYsG#i_@2Pr z!48e#Uko+6#;I&?tZ%M^62r_ylwTsB?`M}yN2RP%Nm^G4q#$47mR(f3d z=06$h_PRv3-|&KXO5zI{79n+QrQi85`HEc;ifTk|Dj499cJqJ%tPM-X{u;W}oBJog zeh`8;MPx~|+sRk~pW_;eZUl7sMmZeU?0@W~`yuOpwEqCY{{WAF4Xmu%<502Ed@-%) zM4u`-wOcEj`NSzAVU_M-FKsIWg07<+5nq)4EAXDX2iWynI|=nm0lB=JgnzW!5`JWI zF)I2TfcdK3u_)+738hDQ9z#bg2mQ6<-vjsol&*049r64*);jj-QA6=Y zNm1^k_~+dEnVRLush^ zkHY#UlF%!Nj)!F}jgq?MgFU-TcFiefCFJumc;RqKq47t<15DK8)in78*0UgNkl~fN zz#G&t3IWOK*l}F$w{3r>cxy_xy0MzkkuHgLQl?2KaCVKuB!yglHTEdhG(BF@-%PZ+ zv>@5Rb8!sO$m;`mG|f`t!pd~Gg-_oMB*Eh&a)1v{ z$FF+yDWxr0g-LSkbm`)F^l2uukHWqZ)`YGIlf=3*>enfS+C*~;m{K^-0SE!-ob!R2 z{ymO%+b@8=8%Klx04)3>nfB-HhxM*rS)ye|h_>=Ra85@A;~(Qu=WtYzu_NWiK<}PC z{rVZ0%3=bug2U48Ta0M~q95h3#P{{RRt!;Rba z>>dxa-M1OsS)_IL>zwn9S1PJqNKjiG4y5~Z>rweqZYWiXqVbXUK<$%RwHY^bzT*X6 zPjk7l@y+eU%g1Hl4~LR$LgyY3)0ikx^Dzf@037EPJ+H*6VvXki0E9!~^}9p_h=;+N zS$QzqSVqPcRVqGzpBY{hlTY}428A}Y{h2nE{k5p-LQ89FDdZ5LEfHY;V@VPtuF}fR z=NxTP_^xe}U-7;6blWwR?JWRU;Y5l2n0Ve$kX_GLBbFyPsi!!pS}Mi4O2=5AA3RxP zkv^5-?+IE;H*fO%FQRRHpOl3*e=Zk;lG!|hI#u0A_KEnHG-~2M4S}&7o({MY2zewT=`JA-q{V*bQnJW z0O4A+o~)NvOGbNz-^cHbx_yPM{mbe0_WF!ob)>_?o*z|_1_3d%>1+hZ48iZ?%0MhCgwi2-9o>FuN zwg*99J3}~?N}b{eQ{*;d&J)`Zgb*Qy_#!K)%{{Zl{=#ttM$+0Cfx+JuW8ovpHOnt49<5qslq<~a(g3o%s$kMxfNI%=C$9hIH$+Q<2? zn$fB4`Jk@TR!hClvA^J9aIKee>Pa; z@>KC2^hVo_%BLWHTYtf4zieOGZ&2~ChqO-!Ur*xi82CCH>npU=8*Z5l*WY9_!8PzD zWiZ@YI?659Og;wThOg%`>-LEFpFPZzMI=!xv0#d_5TKlZ36PQpPI<5MGmf%~c+A?p zdRB36>8`g)zsBeKdBD|b(aEsXaWQIYHfgIhtGD4}|Iq$>zu>O_00g`-`#XQZKzvW| z_u=$5R{kGNCquf^zq9Vr^4e&pv(K3nzGJ@EWNa*f7-N-HDo^7Vhzd5hbnE*g28yk!^gkyP#+Qat4uabljzzQcGISg<|1RZc+Orh7jOUu1317H_|4*&Wd7W{ z+B3FNMhCFx{A=hk=``D9W(ip|taEpH0hPx%CjeKv_*i&HTD7>K9jXPb4RKT^MEO&-F>_FD0_v1croHpcTw(W9NAFqtCz zFqHo7i)}k(;BF)ka%<*~4(n-Wt!j6#G-|WzFrGeVgE0HPHV6 zV<=&SDod;Q$>e56&SJ8(ipWL=IYP+J4mSgyalribpTk=8_LsJrc=Ke`H4!WrE1jm- zBhv+W`rw{8ubyor=VR|_%0K(g`B=k^DdsmKFU4L!pmb%h1I$gw)4UA9TIQkxP zex8-={{XWcjm?k8{U5|d{>&^i{Z~;BfX8Gyb;>J%0U)=U$^QTX&$zi?*}I(fw7FTr z{@0rAhm5`<{4AOqi6ok0_q)Q4qwN>TB9MBp20k)5!RStF_S3_@J@6Kt;Xf4ki{jsh zKjAU(zLRQ_K9PjqG+t5?6iDczNM=zg<-)LzFX6X_{w(;aRQ$|`0C2hFZG^?9^$iwts>ywEb9#};LX}dep|=Kd z%y52FoDSd$$IyHculTdXQt93uwl_1g%`3AUENach_5k2>&;n0jabI@-0KrCn5d1Fq zE&F8XUj}{{eWJ@s(k*mri~lS1q{YszL%Hga;oc3D3me8a^!e+rcP2 zG2qQj^s8?*=9Fpst=eXlo6c!WTU!oJalqvKpx0gwwN*uP+nL|ZbhgmHW&OQfAaxk-GC8Z|wJOH*k>7-?hr>QvYS%w${9pT7{78cO?(4_j z6E%MhY5GO=)sOaEdW84H%yJ`XlHA7;$v~@dI*thn>z5V#eftA^aQ%h9Z9jp{@K5%k z(DZAI%c&xS{a$vNSK0fNNWBf{Cne_d*VK?KBcbBB)8X+ zJ;b*XIVMvf-o?oqvRjf#JC9tP8ulLz{A2Mq!EFynx9~;eVt3g)31eA7X zdj2)%<{3kAQKanA=-}~il%?&KhxK87{{RFa_;aoNVbm{v6nt2R!nay3tlEXPiQ+&n zE+C5MbW&PLlWlnHrSlZ-&y-=m%win=a(>#su$RVf**{;K#U48rmtGsaTb&leR<)US zOQ>4e-OkYni7y#woHB+aFwQfIt?^_23hU#~+V{hr8rOVX;oUPp@i&CyTX^*iC`crj z&Z)jh4%ESr#tWca21Q_N(!MSJ(H<@RsXhuXfgxAblS`XQzPM0X-ZA#XfX^y7c*Kpl z*CZjq$i;Ke;vB+?`kddm-nZY@;>Pf%2CROy6=@`W=kZ75KBM7phj)Gt@bh_h*IM=E zshvUH1TblEqpFdRCEN<+jui3pV!S{05b(~a;r{^HyFl=EryeeBE;MU-kxv2AOM@vK zXCMx9&&oIzgYk<}w6eRiMwj>6rmy17MN|)*rJT0b%tsg@O8{~B^P2bX*?uc%be$el z8)DQoUCo~7>`dc6pEGCbYsAc;p@sKW@;n?r@wM!H9pOE0PY{0CejwLEEM+e>Q~?QL z{(~~x2FV>;W>59&NBk82);Rn#;dmm+5o*^r1E3jrBv5|^kPpVG`~$H%kL@4gCAi$P ztez#D=dM>(0hjOx6`}tC1wV57H-)5(U`S>@yoE^5^y$+z!^2L%f>(pMm68*roCpW6+&_<6>%JaceYaJ&gYJfrI31fKjyUc`eGVRn*2jtVIQI0uT++cL|$&Gqg{1G$_;<%p9Vu-fF0Ub0>7EGN$^7fkydQln_rX0%-c7_%6a^i-3D(Z~pYZ&|ohW&(}Cf-^`EZ7e956npBG%50sH2u}o&2TV5HB{v%%T(A*FPu&o(?Dlmy90j-urZTr3Q zO*`?VcQ49fTA+=PV>`#N+8q|8cAMoUfY72>DgWgyrLV5%^7j40v;}O$E!^+11b|mS z?{v|gu}9Ej#%mg7C^4W4OMx-8uw#GgqoEg5nMqzUWvSlGYc`NCF$4XHqqRftuH>G^ zryI17yG(oi?Ak%Vq}pzKWDV}e%n9=RM;8Hy=kY@Q9-f|Nw>eQ!4i6ddvaI{ zzGc`S+}km#sGzz~f4`I)Hu^~(#aGne>{g#E8!y-|n=lJ=ocL{15JzE<>V$~E{lR>b zMbkg72I^B~xvfn^Wnv1VMP?lXVaH5MwOdJ@;gtMdy-1$Yi^b0qTTbt6Fq@Y>I&V?Y2?^wLj!xAS>=r8=!5DrzRdtN^a z#H*$tPG8dPH~Yl{x`_2%s2oGG3-0JNTpM&-ZZ+4wI)6OlHG>DJQmk}B zBO1b1rFM+BevuusL)Un7pm1rf`CDQ0D)Jz-@U8_A$K~IOG6oUdUt){MUgwj6e!fiD zHBNVusY!wVMM3OkcyB4t0@RzGKdsI^JT0z-!G$=1^_Kf@)+CcSF0vW0CTFj?Z{!iK z4BOk~MS@vKgjuv&I$uan@fncezoa8V;c#;tWQfI4lx0)Tw9u=|YLR;n-V{_yo{+4D zd^{O$Zd4Nzgc{%cz8D(IX+c9fHNb-`gl@9IJ9bez<%yVwz{Dr71BxJ962%t34MM0a z@RqQtE!(8j5%~I&fuwrd{g|*YHwvdLzA=q_Acl*#@T}cYcef(D7uCm!8EmqLv|5&}Ot_n$L7o>ufu5Vwzb=|jqS!7u zg>3EPuf8n27Ha8IPIeWVt0^+)wb}B%$hY!M&!YE8puQ@1A-{8B%I$CFr)eV}pz>sw zy{n+&3Z!?8x$V=qEM-=2pxVlWNan{Jl0OH)!m>0OL53vL`+JgK&xTzY-+mtW`(aC@ ze_WT7)1$o%saYAj`I<^91Qu=-I=(DeJE$w0^!kS!BQYa;(wsWgDnPJ1^E{c1h^|0oOdqrwuA2v?qH4r8z10U54K zJF3j0+fNVMj)IGy`_p9mKO!O(-M$VVtp06mC(d?bAPR~VPUXzF_Tg?lDdU+UfXc^Nst_5y|;4*w=h!~RjOt)DQymvO#7tgDqmSH>R%ln&`$37^rLbX`f?^M;*mu+GE=DSK?rnpv{7I9Wo?N`T3{^-~;8 zX+52N*8hmuqhc2~jUrFsl9=$=9dyr4(NK<_fIF^krrQyg7y|m*n|IwXq zGE!lYQV`nHYy^MSIHB3Fr~&YmeyTKkL`E2d)JMCN zSSL-->oz_h8_=YZrlsH%-V-eA*gNbU?vwVu5bg2y$}UzxF7~Z+$aqSzPrZ!(h!zgR z(?_}%nRVrIN4+7&N<2*u0qW?n15{3KaSe59fQAnSNo@7Y>OaNM#! z%}~s04H0jq>RFLKbH;bMWJ^X?)^(U>$j`LiXjf_A-xcS2+TXlJWjb;jzo_eaTa*C?LZZDeRK_sf8mbrQG`Gy+{Mlbnhiz(^KK9DQ zX#fq==U8S!g3rw1aMy=4qHk-Udusyv=uLeL2rh0N4Cy8||E+ z@}D*^Z{@48Y#3VjLK$nZTFUq+kd0dfT>~XMc(O6Su+Ck!{%XIUd%7tM|M2^b>;irq zP{9BR60>%)pOF@&#bztceSOtDE|-c@dV0v9zwA>5J5o^ur=&BjEd>cgiB=gSFB#Jk z@+Nj!=Be+q_q;MMs>( zE^GA@6n&&d>gMTltbQRs&>9eD3dYoW(3-x|o5-&r6`8nKP;02^`$I@eo{AIJIo|Le z?d*23&T>y=r04_pkbAgy<3A%b1hmnyxL`R7ui(J^0k2G1RRhNJGD^n9_kU>Lh(F!% zIv)Xkmv1kU^8mGfIob#^Je)4Vz^f{lEvgzU<5%UP5p60{2{J=MB7C9m+O^|d5AnD=l z2Nrh)rFlndEx|{3DXH8%;v&;$W(ip3+5_+!V1c7MjesIL!0BdEKou&hTjKoBOq#6^ zBi0b_5Xc9(XAO}$9twZKKLkzW}Dv7IuqXFF#wh0xd*o7Ifg}KnYW6SyED1gXB+Aih3iJ_KlvN;#@A} zwH*z^e#JVsUWH7G5+BM#u(ku z5?0br-eT!M(B#EiLtmV#I?`KJkF7<2j0QgEDIg`f)9qE;iq<=3HCNz7uHN(;E1y0{ zT)W>oNOYe`cSx<2SASFf{@Lg2ZEzv!e00ZZ)K`p2rXAqDeynE%lZ{B_*RILxFKK>%Wzap4dPfp7( zW@Fwin%wgg1zz8eKaz2FVocRdC?>T-u;dSC%a&m|g2kfNsw*d~WnCv7teI|lK8W86 z1%Kg2SmVb4)!R{L?vI>WAGNSAwxnbYvZjS(d9mwm8_BD_9dG!uoA1WtZ1dFg&hH5C z7d)FGM5a!7N_0dD==3%rm}i-+j8ocGCHutA^mOphh8R90!Wev9`om!9GfBQtYk zaThT2K7@O1z$-2GF17r9UOb{nSYjk1BJ%jc5DxHBdQ48l8!VDzL*8OX_KzqNdn?#G zUVHxNBO*rNm8UB3tum2yjiwi@qkh-T)`&@WZuagwsYuy#Eb$PxjLt1JS``?m_iqeG zxCMlwb|sTi4x1fiE+hGyUSBSgk8wZo;~Es|6~%FlN_%8{%NxB*ZW&h(Zzm*AzOUFf z_>hxq&v}pii)A8gq_qf&6+cA*!kVNoQmZrFD`ZY$iTTme97PITnYj+re#wVxvhvW*p%`VA```xgJR{K#mG*`8?tshM!g9O04sx|1A-yb(wMpIZ0Qd9#0&4u zp%fFt)gwA_!Szh>bKrP0Py=%r#k|+65y9fZTS940@8!+N_}GjUkA3YnkR4aQu?iJe z$xm2gv{q=&F)Na_Qk5nmlGtW!BhYR&q}oHCm$!zQCUFFsIMS{;K1-0edu1Q^>fs!J zI2cq!**-Zl`RPA@t|5@;SoTG07juv-%WZ-zUDv^X`^hK!k%=cP)pe;AD|f!M!NYKX zQWKeo46bTk`kwUKd#3|Tv~;w6q_~uF_r@bYd^j#NAuOCiH>o8 zu6Ii0G2gH$Q8lxVM#^r)e2^uX~RC^B;c*jcPy1o%jUaZKwIY1u4AC!s~>O zZn$A+B>StTx(b*a2qmlcy5hcGY^62iE?FnEQ}isQ?fXI_lm+w^rJeG`oDxIKlbs~M z@A}E`1pF=avq51S{Cg=!13IqWxv*lzWqUAb6qaO2?SHQHaH05EYp&yB!#RXkmG5uz z{INz&mC=jOoN>;Bv-BP7oFx$=$W{(GrPpm2yg9C95Aw-OjYUQ^$lp0|jerr<#s)qt z8_*q{{JzzS7gWl9=jk1JR#03n<4iQ(|sntI@jV<&c9)kxJY;r*LA zWFgT(SBgjTqYCK+t$_0t?6V`Nm+OrV?tKInoC|&MJF^273d{*NC;QK*3MYm8~xm4H~kX+&1X_W(LV9N7*)7cKW!mZ1>4A|%0_*U6rOdv z6eISXDh@9+y#9-E+jK_2huU!cQDDHyvyFd59^LO6ETOuvcgqz&T~f4%kf*j@yNUZ@ zS2H;?oK{E!QhmBo@iHiSM|V_O)VT~~J|TyZL3F*B{Iwaye@->SQkfh4E3B6^=Jpx( z72z*zXiWC$&h+)>d1t5VXR#7EBAvj1xcfzTViR{O4MV|0vs1d}l_R{PHYWM`Ip0;B zwkTQmD1mqg`38w?@Ev4Z$i2xRz$%DAH}ri1fmhR7NJqVqm5!8&P*?De5 z?MxEA>^(SbO&qjQ5mH`pouhdA8`U-Fyubx`2BAEjHSNav>e-yM4vaH_Y$Khm`Lz1_ ztA%OE;7XNPC!Ap~EhQFb0-v*(br{VnC84%h7^pP{@gP0ils4;M<|IA*{Midh+%{PE z{Ykc;-m&a9po73B$vOdY;#mlj6y_GuPISWY8a?d0Nb#e?x7`Sn;TL=J@6d)yB@4JS zUmXsmM=rem;){|g1YqSD5yS69fSd}0z&e1v`VzD77{r% zP~?TZ=oH;dYUq4*)Bn2XY8Ifemo0vF4j%=F8Rr$RY0bgprj3bh3zl~0(V*X?13%z% z_X913Fy_ghm|oMrxULDYmlp&@n;=@@P1wr|)6Cv30CQQBM(&VX>9Omvo4(#ZB0lTp zMtwU1eU1Ss1AaJ6IG5jdFJ(Dl9W5*?h^)wuNrB*kIF{^xM1P~!r&BJTjM8=l-pUS> zYdbRgBvtch9u{*%`?4m0dhT$)XePbLDTj8guu0SBd=43NEg`24Ihc0&bG=SFf%uew zAon)j`Hf@RvGd#ty#mDOgW`4N1uwM#8ICp6BtHx|(!Fr|q;je~v37|YeO3^{rODY6 zNxU*%3vrR_W5fI-nubs1)}Kt9w%Fgpv?w9Qy$3`6)uTyUho8K|d31XNNkut-TTJ|2 zlrJ?>mf0IAISKQ`+;G$wz$*xv#p7?fd{C(jJ0{ zALPs6Gxm4N_pt9lN$WT3djgcpMn=v$Tvd@Y*n};MpBqrgBcr?Vy<9J-_N&AC%Cs&Z zh>nG*i-l8O^18x-$t%CHa}D$ak$hsE>dCnJlO8^TPc`c%)Z9S zec_2cwDa2GOC_;CsJnswD$7!Mw|^VqFI|~oXf(e~Z_<$+Bg=r%3b&f@mMc*3+NP?G zn!FQ7ZnMp_r-rBB0}}JQabn#u9LnFHMZ|w9fSmDuDmE6zs|!B=o6;efgzPXgFq-LY30SMA~tb-vz8C=Kj-!qX_hplq7OScS}D62)M1prc*2qL(9ef7 ze>4t1g-nxE^isE;Ld^w^q-*J7gT=a}mcWOZM2>L&8Jeqfw)QA{vjr{lUSF@#FpCf) zvyf3{zwgyxS}a_%2~k@~(@(rxE3{zRFD$R-G>E}kBT`Mscu(Epy%V$h}gMyl5zv^pe!j^fQsH2LQhJeY>Zy@m62p9Uq!C zFe2s(b|*+e^1K!^G?oZMHH_xC1Pi-S5&XXkf1ydB_9HKgK>XNfug{Ix-h^6KYn8Ne z)>7{eP+&EIYZm8Edk1^k9rVcIP$*lVwsq!g*P;vm29+WEwA8Papa}Ns+_tZ7aeYMX zQF#A`VSZw9JwLB~=Af6V>5w@83(@HNsfEF<)=ETdmwtz4wz*?D#)T7iJdH=<3siQ-3apy_i7bNy@uFH0WEghRmsxpJ&X78h=DlrkgZK2)u4%D3`naK5y`0_1ig(X#h^pf{6*P~k(u5&|t}6#PExW7^wR1vyq`R_H5GZ1hbA zYYstNn@7RMEQ23)mgYx%Kb#?lLi1>+gdK{rksIiw;|R6a5f487Oi?|I^8NwWW+4id z3X7>?joaYx?w2UjLJ344Zcgx(h;G`IPkxFfV3^1W*58lFOFyqenDN0;Dd)sdUPYY* zh|^&~xZz#?qp`x~0E7^!K0jJg~(;l9EN-;zwOom&;*8jrg)P59sNv zfyh`jQ^@qY$|u|&y)4ZErxx-dLjdB$i$v_8En0S>Tc4e7+6Vn?jC<;EGp*y|)~|Ke zKJ7l0J?BpA&grkb*qr;^AML(_kx%k3QBb*++WgV&@iLP|=j)*20<7jI=i-IF6lR~1 z&z4PJztto{!iI!%JSnMoN6+BEH{p)j>7P$Fe`^jG=Ul&SDBHGUS5JNA64Ge_ZYL$E z<~jN&{TF6{_7rQ*i0@qojnMK$4TZBldW`a+;LV+oDPR>aG7owYjpo{ptakkn{MDz9 z{m~&`UW3+4?gc)xuJYmnZboOkQ*)&wO3G5FE(`KQFa_ zHuxP3X5mD!A2DJ02Re@a;%Gv1NdbbIVgtFlVotf@{yyAm&~1F;$T{x56=66x>Q>YE zEd;ZkRJ>1xeF5ve0z10Trl9YxAMus0F)|>>}!skpb=+SeAB_X_f znXIEeWCmZB+0jdX_n5!kyHBk6xhC3`2k^7!fbq=$Y(c_}ZaD1>R}r3|u=Ds8@CU2G zsq@V<-1~~#973)jY>v3c{<0ivK^TY7HdS^Wna(<`f8aYm=~%s{u_y-;{F(Fj7cR_h zA+pbK){Dgja=GgH4w$!KaIsU^fj%_^9N;KtJ@(Nknuu1WOa>LCHvE~1g0a z!>b0o{_yFNwkZGp2J&M7=&+ zC6E{fH^sA`-LOfnD37DYO4S(;v*MP8R!&Da4QFJ8d<*(ZrF0ONwG!vjA@ukby)h#` zDoFZoJJFF0xN?JHnuz)1umn$K?o3_@@0)efB#)<7_X84V;4c~i>&fF$4fZnPC4s4_ zRu?k8yT%=2fb=_lN{y4A9E?iC9C5)MQG~e0?U<;YBs6t%Ex`eQR(vCVVd*<^)BO&XR2~RII za`}_kYcIa}>Qdo8h|CTUEnN1HF*3n{^f4D+V{EfQ96tp(u#)J=b}9fUUk(!)b!NCt z+|KNiT?s39q@j;>X)z9< z#V;oO6^plJt?X6JOOWCS%gY$UsRUVJrPc+$U0>u`Xf=b$hSyvJ6~=s~5%rN;lX2ez zMJSUm83cQ`h95qqE$mD-V!Z!wNLsG!@MpY~!1C;WKk|V~gl1;GH5ZKKNr8ks#x7hC z^bHEHk{otUY~cEmL8mrIe;%|tsiz-iGHpF60Dr1I{u($(*Y4-Ydl1`=G)Qphz%#X@ z7re*KjrUWSZlsK%2t~TuA1eX019eYXPYl3WBfM2q(VAlM+iP!dl?}2^LN66$hI*~g zo+?B)U>yyIaqYLKvw+gTUL0?#h=i}>rR0aLCv>l?OxzkEaZ22YoAYFxywA4NV^9dH zNPBoLT(Hs=(}_yY?N3e+K*v_1QeL@sS~r5!}z=Ds_z${Ie!n|=lhcY_SN&O z5x?n;&y>`~3U@SYe@6_OQFUDSF@GDE3^^jd`(HWf7U#=8aq?c`r)b`a!1ACT)tW=P zSrKIDWj0DaNS}E_&~%V^Svjw#@rb=mtz6xs-sav%_01XgM?HS528kM;4Z36^(W+9O z!?ku|Z#X|y?anN>)HGm)Fl@?wh<;u(L-ZB@r}TAP91KJ ze#@KxZr4DtdHePj8Z@#ySV|M?k&_!9ejHiNvQx@u-K8PA*~mnBhbvQGdqgV1q)>DN zg<&1{aM^*&^}Nb%JEP$FM`SX!Y64>Z#8F_WdKE3vqq+%2$KjLqX8_dv%H3KdMFUWK z)M21}#PHKECW@?2t}}}3AB7%_3stF|2urZUQWX#zU8HpJNHzz>qqX8rZvRt zW;Q?kn-+c68ZEkS_Uar-ngdDbU!&0RJw*Mfm@Knub5jZ$ZUXzUBTieRy;Dn5HS(27 zsJ(i(dpVX{@(F=O8pV4?D^5;3AI_0zp}rPuJTa6NRdCESk+;4(&}x#I65eXyZ)W`8$8@Jj-ScgN!kKpN zs2TL5kguyGz9*}qIUQFd#_H3*O&qGl8VS3DU5a)wf}Dndo&WCx8j~f*0>rw8)m6~; zU%d+vt&c&;DDTp@RU5Z*Vf**!+Z4$V+5Fg#V_0-vefuPrH}B`(=Z%8 z;fjZiyx^|EMkBVsRM>3IG}UB@Hx(N8rBS<7Hm4AZXRgDDEKow2l$Iw&D0N~czyL+O*&OSoh%&TQsK98Kv{NL=YO z^%3oC`W6S@5eFD!o{2XHGF!eBd(re=mU}AsTB7vZ^EU$${GN|21>1;xWtaB%k7qkg z!t|!L@S_wpQn621LW@+vyxNR5Bct7u5U(#?Gd%&J!c1wzHCEA8+$7W`o~mq=6oFa? zE56de^+H5r#4ke=EteN*kFDN?C$CFyCnY++WW8j{)iV^&jQcdO0`mz`Vf9 z>KR>7nC^@yt@O>{g31w_Cw**pmrx6psZ@f$%&qhF_uJn-jM9RrV4!|$s#R#ugyCdf ze4Tbh`*XMLJd*V7S%X0j$(fYC69>_eRO}~AWS0`2ifvz5@Fo6jjdpL@V65(4Wi3VC zg|E}i_V@>>>nIfXLC6%C-I~079&4hVKtJ-?%6=g(V@}z8MGgCXy{w)K+D=8gILy)O z5x7>YTMUJHgVJk+$l8e~nBaN;h(49lp6U=zKoIT|dIa_?X1lv)WfmhDJN&y@dC0rGy1PU=K@@?eZlXY(sNGSTYaza{BROFlWag#ng3p3_pOVVfg5_j6= zw$b`iZC>=o%RA*%>lW}@+Mr&~)~qLoWS0yFGR;NnwiglLmUg4}Q}{rg$I8g2F~ zA7I=uHP)mn#r(~USwt#V_J^CAu7fW7mHTuV9=Eu7eRw48_PHy8@iXc6b~nN8D~4)GT)PQ`QyNN| zIne=l<{#v#gt1n~S?LlH*_8L`h9C%@jyCsshPGCc^FSkMrlQ528Clv02_V9KqqYVS zs$vlM{*$A=Z?rG2tGnXsm1?PA>Nkw`H&btu`AZzqT43-M>T1%~p7$sz@gI@E5T>&_RSW~&q@)R_K~ zb>a*e1-mQlkC{ILP=nyznLvzh*qWJQ^-CWP&H#gmwNy8ox7R*ho12;ADx=`PMCHl| zrnlvTV0FU#aBq1`n4Ho10Hc=gj*655y5IC$N!$8SG&)WU`iQ*x%t!9TXfEMPlJGhbqVJGUr$E?4p)ki zYg7yN@6d-^>uiXTFfwg_{pixA9x3V+l8y7>kd56Z0ZD|?T<k(NfSM^Tg(h+D;=LV9zELTPRDh^Y-;C#1bD#hhi z%nf(*W@claI}6ZZ1IFcjIKOVZ4&Tn07?=1Kg6>|zk^bWQEI_Ig|UJShuS$tBkb}{CvvE#qeBL@h_ zKYhQcE3qA_#ASjv%FRo!;8y{7eEa4Nc*}dhD{u$o*Cn|Xy>=Y+CYn#=YIwN0Dce(O zh+}rVaPM*1N0t3k_R!|mHDEi{-C3N8nU`KU;%6@bB?h5g9ge=61Aj%z2lz>(&A>SF z=AP&Bb7sw|wZLF<i1fr;xCz*p2gX*^(x|=*PwGqEjpaR#DSZNH6eB45zXXPn%UlJ z;;}ho{6a&$5cB}yp4#6@QJt0=^A@BvMs^qz=l>Q_t0c3r`Ss(LUVVKOu-|B;9L_9H1d>u5*1(rvODD=T3UeQRxRMgYdH7Gw{U2D|Ui728<%V1UtP$EB1v}`|(`wYJMdCDNZ7Jje>vQV4|nh-JcDyJd??-ZuF_nA!Lft zq~`t0DzZh1#nN_ocWVrcta`zM&L4*Uv2f4tiq;m?W4ir!gQAmo3jh_pnxR1J8$o|o zu#f9%v?Sw1amWQPbKc?dpX@+ZxTP|wJK&>@l+)*R>`__Kh zF3J1q_>nJn3+j+*eO+M)O-djXIcJT3FqMlRgOMOwqiRo^m`iI)rj7447lfA{aUo*O zI_UtvhCO6mXGmP1Qj0NPD_nToM(d$c4jHm>!lO<9i2_;77QjyX4VkGFOX-;uT)wkUKi(Cudoj`*H1N z_KEDP{}-GfC$!d~GN!Yum8D5abU3#wZ5zgyoNIp}m8_n{s?x^9vNgxt6m{nn>{sDh zi^=F!OD)o@;)brxLKA!IT!~#k^;=_oNMjMR#u^4)#^3h=&W+KrbCs#{EDkx1NpnZ^ zUuY}Z?zLH392;<>3fj|?;}pE$}bCCO))SSegN-h*toRL(dgNFkU- zgFsaOgL=UiWIS4$LV~+y^*`MsQ?gFHE%+3z=bHdRk0Rdri%VQHAB9NI-BzvQ#+8Up zxFa;Byb>o`l($V&!rxK;2bUnp#LeX3K2(_y-c3N4%_*xK8r<|?;ri4fQ;sVt>{K>J zs7WhHmZnG*1qJNts8;${X8=AjFIG~2VN_NO3k!`DHMmB%%xJb zxKZ0NcjGhJ>8lz)uoF0;iZ5E_N~}~@RjrPabE=|s?GhD|Sss~z?h!($DKQFy;CMQ>4u3oPSSMnHd{cCyy82MS?CmW2t%pDXs z*k9Y1|D9SU@&GOI)MRfo+5L~Q1UyWHLg-1Y(XUTYBZ7q4lU9p>VXr_t^vN@(o=h&{ zWvO3`U>@AuM6O}hBZv=Rn;sXtp37MUnyf)XSLbi34lM2#2^=V^2B!{Ubijqh%wI=@ zM+m;T=Li@1_f)^+BaSqI&rHu!1{vvSl19Y$z^v06Tb18aIW51$`LTcT>EDgPnGI%L z<5G8-Bj7Hqio|Ec1b=ZFe|vz4+3l#2v^kE$@tNk@@B=S%`hH!Z6mKZ z!T7iwN<^PYHN$1?C}Q!+$|s|^UBd$3sKb|$tW~habQ=Tdy#?I@U9I=AH>E1pPUYGj znSZ89@XwM;r(Y6=mO_tDk##bdu|E(S`$&QU3Bc}~bdVqm+1#hZc<6HOFELzFRcVruV zHrC7`dGRkK6DN{)fi3@HEhLq9Xo|+N?0=e+nJh3S ztzZ7UQivCEU4GlgWZf=J%VV3ZUZ2ma|6yo+3vqK-7OyFXqZG&SbS}xU(B`mb-18zt zo>V8QF0eqdlVs%BdgMxt-s|>^L~@Ya^%OowIt@3Q4J|U|q*zS{25yXSq>1%bmdnWS zm=wsHAsLCj9La`x;EK3umgZ_neMgtCE^{X<_!bQo+MR#t%D-EP*rcjmR?uAkCIO+) z3fN=mK$*#{TH1I_%oEzHZ_JYcq&*f7Ef#NkX1x0-gZay{Yuwvj_sULn?}-Ur?WAz~ z4Tj6y7tUPN4>E2WJ#XZ5{iRHq;pZ4MGTZi~s#jJM1lU8$G)#HsXTObzWeqZChAN4f zv)7Wt<_yG`0}}lI5y_{ieUU z4=udz)UWp;NjCJW3UOVkJ$%{QVM3Xv)MPl>Bx$h+Jg(4V@(osYK|HAR1{P^SUD{8o z<_BC3Uv~)(`2_7v>xC*}=u1};`7=Hzz{&HtIiHZM6R~G=23!=R2)xtNvg$6dZYu0Y z=GUH=QpqG;z7Md(n2uTSTZOAt3J#pSQ90xE(o%R%yU<5kum|K%*Y}Ja{#jaoTMur3 z*aRl+LSnfQt(_(d9VxO8rhPQ-EVGi<)YN$tINsbRy3gWV(D7MGk4buF2i^vlr&Gc< z7h5Wv^wct}#mbV&Z*>@wXu0ygZ7rr~nJyT$`a2ZnxyHX^SWcrZl zYx_3Xe8kNvIPNH6rjtMswydqL&6MYeVG9_Fw{AM42!0H8u8(3_50y1)gD=CdFX0Ea znD1+VsA}MX)CFI2q8t{r%3@!wRVDIy5bkg#B%8{iev`_n)E~k7!!m`BjxoK{gY?Q| zojY>Y1;}iLX1oe~X=PEJtApdwohk%-45kT`1kX9a!3`Z&^@U3I6Rouj9q=f^`Uh$M zI^YYEt+g8v$)M2P@+ebrG5^2v7{!Ef#lEHcS&p33Ui?ZMDWQyaf%M?j#MT;kht2I@ z+2=TBG>2>|${pgH0;Xva%rz6Fb69?wHoQ3_Lb|-z=;GWc$9xFVL4Fj~Lb>Buy_{G5 zl?mblI70Fp`p;)}OVZTUg?DV@%xPVQxqj ziaRv5Y7i3y>|zWtROQzE>ygk9G-w~_WvnyiE}Adc9W#{8^hyo~tpTrol@*^VYjs)) z2T<_m$mQ>b+{c*A3Dec77L}z2d(_XaMkGjZvb}i};1Nsk<`WEPq2Y!hvtUM7+^baz zc`mHks&nmk#q@@ffR^m%j4!4X1Ur=mknUFZg&+7UVI*B}Ygi48Ul-tBlYtl^A?(C# zFne!a_Uy$P*CkLN*(q9;w~EO$UD?;Z@JL274}~?C z_x?qmFXg;7PrXu#HMllIu03_S1J#1QZ-p>n2^}4w-ad%L(S{NFTish=x1MbF>Y>xg z6i%cD{oQUtWU%hwXwx}14~$5*OMA+wuz4@`hlEgE>&4}56wnSl3trXg9A@ra>tYwj z&aFv(n>=p@XuSPLWLY!n=8`e)@JHVt^}e3^uZDEl<-HXbM@Dvob07bk`$3dgzZ+is zD4?RY(@xUaym){~ApcN#DpM}tCAI_R2Wb+4aPS2c_t9c2Ad_H$?NOQZktytq7m8pUq7X_gal zGxbaTw?S4;`DbS&p79@%WGp?bD{U(IYt|s~%3d0TS*A#&O=&)MvXETAN0FTVvM=u) zv4~^JtFOo%i^Is9``F`)ySLegvUh_VC#w&A7k50cKS7}#BS%uCRTP{C?4FH-8;_j7 z(WspOe@geUc%~)`-g<@1Ss+q6;8?Deb}(fS3mVww>es!@fJTNdm`(7D#0qku@zC3E zm|q){WeE~?2p>ysu}6>N9xThTMBNfzVtAN)3C?p&_AxSl0qIliO%LjvZkN-Jtk0o#dq|2HFW_y!SX~2Hb|1z`5T534y0%x1qB4f^d2x>wsv) z1U#C2fq8FU`zZ3BVfnEzv{=&j-%JU-rkiE(|7y*&?)6fyXY%7CUqhtbu8dzmh!>3Y2(&#ykL+iJ5%VghT6W~ z1i}PVW*6vSo4Kx7UCZt=rp)`d%9whjqOTL+y-vR_`*5at}?F4w~eBtf`A~M z0)kS~4HJ=;?woXtMmi?lEg%ij(%p>iP8r?O8#!R?|L*;?kNe%*b3fO0opYUYocn;J zsQ3QB=Gr<)H$1hw!lm7EQZm23JwdtAfJ9OA17)O}rH!ol_7smS7}K?d;^4Z$e;Qw& zR0?cI_@(4DUBFNyK{cxJ%#o}7UtcV>N?x8DY7|iKe0ubVgJeTop)KG6^EZyl+m77{ z)rjT;XfHRnGdi!ASvUDu2bT`ce?MPR}n zl~aZ`sdwW)K3j(#={|hdrAX?@{IT${I+^FT`F5#e|Cgg@lFXOa>%qcH=wEFrCEwo* zmM{eQdwwZ{%U52+d7`eH*T$YJY3ADbf=I5K>@$U4ndsK7wa~F4; z;;#G-T~CG0p$L6?nZ{r8*FcCZXG zmv%(^#$S!Z__2Cb)-j8iHRdB~wz{kKyuCE%$kv~m6h&Zdg~z@}qa9w5(rw`1VI-}` z_Q@s@8mbg^6o^wg?0>MT@J8>P=k|>b9IAJaD7su`$)&0M)+Uiy7AN^4H2xl2#tFE4 zcx#`&wZD6u$u^h;AT}Cg9%MQRq-zh}0XZNmXxn8^&GE~gy<&-bwb-@7&=H#w1pF6I zK{_qnb}gmb^o}|vz5LkFIjZSQN`o~IEP<^WXpoJWr%p(;!@g8(Q;3*5=8_q_{rzC; znBSl%O!-&ZzW8mdLr>p_o)Pq_-UZU;Q5=B=!xKtRsJTmjd-wNmysLMV`LcL@RN5G# zULS8AR2Ju5)RKHDkm%#(Jmaaiy(%OgsL%#P5l%?d4y*@AI2EHrJt zMlft+)oiuBRU^)p$CpqP?+3mA7^B0H!Zh;CwW-Y^Q4ZVom}ylDX0_B+vOef)MxtB@roC8T3|Z6UfG3}lSq>dJ-cBB8{aBYlaPsB|Q?X<{>$D4Wv)~_BJg}rbIZbO-`Ns@<1%!mXK`)&g=-Oi zSXyfTKrh~>GLCO>#;&f6S~sqN%j;oSQAz!8PI=+3L9iUoCuU~miU1(2B`{U0L&IpFk>XW4YV3Xaz~A{-b>ZDKzQ_!kbTn7VZ$ZtD?IC4HHmg*yQfDK!vc;n6=2$nE&W_ri$Rl{aHWwMxDhWw5TlK|IemhCdvS8cSdo`JJbE% z7|949%Gn`$wP9C7+W*Hk79jmb{JZQ&)D2meFdAMrwOYDrT<44adeP%1}H`a9zsi)r*)e^SrAYX%r_{Uy*2R?a2< zB5-}Uwy>?laO*tVTn5he<<8OS!9}Wr-(mFQ>MapW7S;A>j9}B~P$*5C)@*Ih-Pcz* z;3XUXT*0^(a+&?@Sz~T%%}9P9+pCs@{L%j~PL@xnkt7U#u>kux(_{W$*Q<7PK)ZgP zcI-7Y;a&5+?yrRBIjbui`{r|0blN@EjlL4a1Ip;`P}TpuUl zaOms8K61)$!Zxnc<4wMr>Y7m})=C`ffpUkTjegG0y(3hGxL!9eA!UkXvA%6Zb}ssV zWELlM!G#4425$MAz)28lw`vQO$H~jc2wD1C9?46*dsLg z#%>i}_aqWH1jgv3D)i@=MTsJYRSb|CI|v#*A7`_-jsMPBlD4Apim|>NCu3a3Qi;Jj zw0(rnVs=clZheSBRRwEk3hZfs7bsXgFQKV~xU!7@!uxBUF&&GzBi_6B27+K?7&tT9 zft)I$J%jbE$Qn`fVMvKNlDXmU?OFjjPM?zASl#X=WXQNs-qg4u_P6w<7!h{lm;AzM z^(=K;X;gA}t7L6Z)w4EHI?yHOIq0=cA4D#2y-y@8eNY+yH|PR@(+P&ZmXL*)%w4EW zu!3kb|0s|=kL@Wmizl3t);+9Jdld8!nDu{|Io!WV)?fs+^DW*Am$zWzbW=WfzacU_ zAuY+DiAcDuos$4Re(+a|zI*`1B7MOA7TZW=ZjmHp!VGGNL1hZ*CHk?@(Mq9m`W`~&2JhDLM5%5)?}le?PMu`~2; z2h|1V*!?a=0Xsm;Fz_xS9M8@iOCZ`rEJNVDybw?vDL_unD0 z)byogrsIwt*lQI0_J#qg;h|MS1nh@wK}l&)U;n?%3t)?BI)BmS^BcTQ8T7Uvsm%Wg zxK$3!XKbUPhc64AFo0QX5W&;H=TAidTeYCPrkwQtniai1D0u-O=A5Z98n3ZJd4+BD zgY`xVM?Phnfg`G*Qv{g?F2k~bYdNd0{3$N zhZfQVZ%C#Ow4*{in2tC9mf!v*Gv`^w7cKrqQ1|`{UtBTx6quKV1Id=8zSBiKJ~|`S z7gE%-CNeyL?Ds5j8>|{;lpmO0#no%B>Jti`iN1|OMT6~$BIl7p^MVYKhRY!iNUq&Y zQd^i^%goBEfx)8i^s#b;InOod;p0<_8^z=u}4x zLz+I`hIgRR0<4ZD108PWWfD;fOl%u;W1M0(#aFe@|6xkp1@yoPryJ1y52JQ7jtv&b zxD4!VZa%PG$XJ+mpvG2zCc)2ZQdpzzU31WX8wt1inb)*A0H8*{Ms?sgB^v6M_i43) zW$J!(vn?|_klzfExkA8D^jG@Z)C@BXK?BoELaq1lWoh-}A_4|`=n*x)cQRGAYi9#U zo&*5=qs8NzY7+=?)Qj9Zv9ubmfZH(KCnnAF!IjSOX2*Vs?ANh2v9}l3E%DcoroaVL z#;*8ndO=vxs*dys$tZ1q+vx;4HV$18v2N#oFI=0bX@*NS3}P+wd~nXOdyWs3^QvK8v#_+lrVyW;petk&g0_a}vA4s;w#s)ck zWo;7$ww{ha_uqGC*gFt>f8}|z1&92Z@BCP!9INX_#$$6u(*%EBeEwYGp6Ytp+H(lU zRiEt^K=LnqL@Q8N%p8D7)py8CX?-=u7ub9T_FN}yi(?K8|#As>lcFH4J* z$$(8P?VoK&E%ouaW%sJgU4s zi6m4R(T=p(yTunP7q@Y9R*0*@Rqx+@AYQNPPj!k;SjP#R2gIe=K!T~RH(M!F9Hn!e zR;AvM9jR=nzm!#2Q-{1G?Ma^C^!eQzQE^3oWRDAuYvy4$HG-knl5NuBE^9=@mIl zP&)E|w&RSUnwb~gO?<>N=uy?#UR_+iamF`QuVj72$IBe*(?H9}Uah#xvRsDbZka`@ z`Q{R}FU*NtRR3=DOW?AyPBS**!f7nfNyrp zCaRo?mN~;OhSsIjXc3@+%(Ev*V!Lx_lL47m%Hv>LJ#Tgx<}=|v`tF=zw0(brLk!U{ zgz$Irmu0tsfEOU;P{0F4ao~L5%qHfU6sPGg` z+dqRiIW1jQjBFM5JT4dZzCH0?0GB9`B&5T+w2Kg%8lr#njEl=^i%4Np|sD2orS z#aWjv#F2Ye4l;~$H6wyp`4u6QVbGuOiS;u!%?fzN8_KGi{cZw4Rl|Fd_ zDyrOjrCuY_x5TF-y7}v4v}c+?&JAi#_b)gJo@r7ku*(xv7LhR$OV^Cnpr=fjGSN!K z*M-Up0m~(;G&?uhjVgBmB?<=bmJ7Tiq-E{_C0xeb7=Fa2Jvu3}12%gE|j z8JQ(?U;mtaH))TW(NxCnYV;IyT&nYk9k#RnmpnTC)=*AI(|a!yr4-IglWOs2t9;RZ z&TeHzT-+G7(x=;REh45rDlO$*T9erZp9+^E^k(R}-Nfl6^7%M{2+q-Vc0OTHRU<_&y(0~OY06Q~~+5sA!`*Zj;An>X@=oXyoCX_n)SQtb- z&fHsmFKbPTW#f?;CCq(b8VhEY;sl9h%zaxt_h-tKtB!gb4c({f;}F8RkBKZ|V-UgX zmw5S$zJX$T|3bI_8Y*MM0*^1mUFz<7YLt?y8NP^? z`iPh=SEOo31XZq|)D8Wp;QD!)!8!0FQD?9=nDy0^_SQGO5o+roib#PDk#n3m$8p?a zQT#>*bN68l;Ie_xGqZZAz@IAf- z177+|le$4Hu5bd4iML#@*E3-Nf-joeZ|KEBmXRp3>!nDz3VQu;Nv>DS|1fEuQsVLM zwr27~R?aT(OR3d&f4fJ3-DEk^a0bc6huZ=tYpyGrBKn-}`wlTAxu=cOJ%(`gZm$Xc zZlaDH3Xobd*}UkBRXYJPp{5Mu($e9pJuO#cz>lW$XT8owZL+z5tWA@d`V7J# z*oo{+(0>?-j=%LUeQCb-NIy9A-jO}|U3*rH1IbiwVu5S>C|i`Yz(iN%Q?ZR~lr&s@<^Da4^Ts;6Z{$OZ&XlzLO*OHoGs~?OYiyoC!jR6s;Kb zLECJO45le04VQBkbdZKjf_c1eAa;D>M3;M&|M z1U&I+2rwmDowwaM2^f&%7LQBb-+Z?0vco13A5r208kH-2>lT-$+D|F9`Q&>I2=j_X z0h%lw_=@85Ey~n80|~%lDBpqpJBKH9H0>mSBFbnGx*7GoS>Zp7dO`evKV?GCr!;pCfsFc!-s9BA?Z)Xe`=7=sGe1yKMZ9hHG5E)i&}%U9=IHhG)oqe0 zVNq6*PV+0_Nnc_F~+I5a(z`yotL>v1FKdK&&{r%e~JWAFYHB+)8 zB3@hBLst}=O?!Qr3bt({5p((bYv|S}*Y3BSW4U+9s@$yLd%+hhQTs-%1VO?q+^2jx zMt~1fIHHtN$^WG2W<3lf*uZ}jL*cJZXN_s;W{dcW9JbrUvqeRR>5iT9^{KYEr55 zn*4^1o;&=ME#8HhcAdNo5zNgH=FtRG$IHD4kd=bW?`Hm-j**eaVaD&36%{6w8=FFr z=igUY6QMrwMs#=t>#Nq!Bz>*hh2Rmnk8fM%tL_x8FcJ%%cAxHO?))c2*kILq1DRb< zQ#n{h>0=R6nw7TE(SO&&)C+5;!M=lB4(}4;51wjr7=jBFSualymhv;-GGkPUL=Lm^ zFQh_ShbQ{!+||8@rRTpVkROwax%mje$8-g>dPb%wBym)ESM#eQ`ntse=<&ZSIGK^x zh2FEQH1mlE`||Bwa*QK^l}7eb9~YXqq_kUGGSz209aF!(Az1o(q;yIy#C5DJSro&o z;WA7wCn3L!FPp3|6 z%Dud~Ib>YAi%p1F?eUdsv1f_~1$@3zYPH?vI1s2C_H%B%wtjI?Mw?hjwnuGU$ygg8 z#G+}bS=^lcC&7Q0?XSy+*34bCg~mPXUjr?XxNr%qa~9uwhaLts!>Knwq2C1}>vD3j zS5T8iCLsbBX$N|76yser^N>)oi8n~AAr?8eVmu`Ft>Mcy_*rh@e;8GI4mEEL=qDSx zd??vyAABE|)k^<0XAV@onusm?i%aA4DiMsne)H++?hC1uvc|VW@w)S-fpdW$7r88T zC^F;1pl))qQyFFeT4G@ewqwx0&6oCb>qqT(Vfp-Z3)i5uNRA)m2jh-=1gpPX2)>H! z*#tqm%@L{9hVpQ6#YJaFw5pQ`58-8vumHI1>Op~FmGEg(GilfsJ z%db%j-MV}DCOJN=8=htxb2wK)rDHx?+?pA6dct?OTUOemX_vXgVbXuQq5mgL=%VtXeFB)KnfE)C}I|X6N z0}M?bNRVpkdEz*eV4#qs_@FcGx2K z5z3GH{^E5AE$q3o$2W$Y?#8BlzusvL=rQe{C6b1tS3|s&U@XJ$!Hl^}ZeD*Hc5^Y% z?ki8%FbMMX3F>9xqw=~P47DI7CdB}IdV+46)I|F$mm$&~rc|oGMit61Lpe8jKh>EI z{GkzL?;tuMihyk8LSUcKkdA4bMOoldE<0L|C?U;wm3+8kTIwtie>4Jee94nmn@amU zt$1YU&|DEPw?Pe8m8a!&>#VN&pqhDCsHCS{ytwLlf!K-=b9OqNBEob7uX&`+dmcJ|< z?N3EwDpl<89fD`yFj>03##9BB?3yfcSu}rRg@=ESq=_Yy!YgU<3=GPnkd8 zUljiC$^aTfYtc_nybzY5E^=8MJx7^?YhEeqc8DT0ME8=8ihRRQ1-o-Edcg>}n&#bQ zng+B5Z(v;Xz}WAGD1BoAJ4q_o^bi=qC3tL?N74BXO-&XK=f0lz*V$HR1Q3~++NK=w z%a--it+#b8ERX0H23h2iI&o$C{b1_P$8xyx`ws&lI{Mu?*m)~2a+BaXW;!}VDL^Vf z=|Zo*VYbwcnYU|TsR3uf{7O5+a8hTNk1jW`&yt}klMP8LjUL}emJBRf;px$p!o)@E z2tDJ!{7(n*8|t{_lGwgKfiQhF1dCTJ@YHv6NDZfOOOi zeLx#T7TCFBatj#RUc4r)iyvb%uWN}nm_F`ZqH9vANTbkukX+f}jvG#K~-X~ho;KMJ@ za?sFA#uFrv6=ay+GrZiGKwyLb48ZJpj{|AvGv_th+v+_a^~pX?+}vibhc1Vi2j8%# zh++JP(PoWCZOqX&7rjD|tt=_V=BMe<#wK}9>4s&e-zS@QUO7*oeL&0=m=S&mZg)#C2Ub>MO1|P1-UW#g*A&=qEA)S(k=^P*(zyR)b zL?<@`$*%7mJ!+_r#3X~#m0NlKwF?M+`ykh5qHb`iF2^NTF$jO_Q z5Pl3qE&qo>)uhJo&y*j&zb6DC@t`3h^-kq{8$4*UAkf?#yi#Qu_GPJ#MYp-`S78dM z2(Kn@g9=avszC_vqF#h=#W)HN>Wt`;^eZfms7}$FL8%?0Z&5tB_p`q^dLZF>C8%bwk7UmjnUJArw+d7l0Riu-FV+<;#nONlOU zmn@m5$Ye{N)+$OR;8X52w|e}w3fet+6Fa8p@I#R;2{xU88?<0&RSOM?%a*LL;P0$-FFlUB{n>K-(U`YM`^dOxI)8+xtR{cwdHt9qz zbB7`&0qt2_-J0$K?3yvXz9b#Q;aHN#^~z2Tnis^-I8_Bp(d(S)WQO&t!`Y$)uiW4= zHWemD$0@_sWOa0*LXNI31^YS^3gkRVc`59?ci>aB5)~`$sXUO)pA8Ne+>ELP_hrGl zvV&!I7#Ix3i|tPA-h3H#+1Ua4WpkuG2QWJ8tkq+iC3wP@k+gAeqds3{Z;M+t1jm+I z$SrYa;;{H9LGi{YVRoX1RoPM_S+>uCKC>>`xKqc-E(BTvg-wYz{#P9(ta=1`Lk0lE zf__H39?#bVn1--1d_dUQpYd%CD4h(jq`t8|nPI%V+a^$}zeRJkY@DX#IyE2kpM(&p z+u$(rC3L(k;86lms0P;y4`i@=XcTI1I$JOdEdFzx%ea73T(zF{0tsm_$oSn7KSX!I z)E8E4?KQeX(}RX%UU&o)tyL0uHb#c)kL5x-BirdEPp+0_p9%wMl*Po2?1f2K3Yrz2 zUp$vA))l5=KWR)o|D9KQ;_K$6OjL7Ss0y=62y<}D2z_7wMYthMmzHd%wNw8|h5zx#t6mop>-nWd?*f@pyt`s(6ymyP6@+jofUq9rsL`0!ts(qi*a(Sf)+_wzWNIpEl9q_Ic zGP9sr9b4Vm)wfz7M@z}jmdFM7spH)=gDQb<= zN1EO%k+Zg3torf9)wzKgAFH-z`C7_iL4s~4PHeE|hED2$l6-?Egt?cgG{qfNqB8jT zGzDeQ?FoUA*Q}!#ubw)WP=fH5=o@ka`i-^wFbexCx{hx|CDJ#8fl?<7O>DzYell=C zM5p{t)|u+^;AkmpS$c9n;gkxiZswr0%U6)Y)Q0AO2AO{K8ub$zQ3+O3dT{Q|i`rnz zptuv-P!e))S7iR_A(2x+@z34l*PyWKm5D}Zrn1Z)5tJG~;k9@Bu40QjZT-5;OOaki zJd1H+->qO3*JU`TZtz+hO8v!r^&%$oNi0h6ABuIMU{UI=$%v!*tJ^j!qgYZVAK7{P zY#60Qtoxvg;e7qt8|DKmuH06Z+mj<>eLAvu-`eKGiFBysVo6Kcm;6Cz^aXpA_{OXK z)hA1O_g(Z4A}S~EgD&|#!`qZUNYQ1d$}_mDS;1FYB}Ms1hqR@C&(n4Zldb$XToiU% zFQUXe8rzLbx18nT1-D2ahkEzm*?~F-fzK8^R6aC0V+!PI`|iCiObO{2ap$zhJH(E# zSjFl3)0k51o-AlR-vLk$q%s`#fRpw{`SBHxByZU6m}4aiXq5i?oEs>Mm2I;Dk%u@d z#`u$l3w<-#pEr(ojiTLc@puAVp7hT31J1%=?NAabMo_sh&WNR-0Ze|I0XPOx20LZw zs)f1^et?1-1v)%%?kv^m^;Ojz_W!fu6R5=Y-$ltUOm>Az>;2ZFpw+ypj6{Zyd-c?j`t@?f=n#h`Nak; zQijk<(I023wh^8I)Z@&Pc1`L-xv9eSUB(mAao@{=-$KX$)(JHJtnk#~zQvzagRz$; zcJ)eM%nzVijj5KD<1UoE_eSxd)Tk}%&Tu2YDh`8qmbxUu{hxcmy95)@j`D^D@ADqD z11CUmrMV-A0`0Drl*xu%Yy&E5D{J-h&q??$lppqXMV@!3uur@j*5|W=jG(Av(|iWSA&SOAVp5orfony^){(&~;h_o!fSgp z`g@Gxj+ut(c9qeI(SdwNZOV>hcV$~Mm|rrrHjJ%})3E++U>e#tK;yGAK$r>ystb69 z@Kc@LwcskgMw407tqw|;Ydo0-5?hu3>WR`pyisp3S~2u4fM;(XIgwcC_gqBaGz9=p zN=Vwdyk}%y{}`V{a4~G4E>$_pa~Bk9p8V`CvHlRJqt!W*Orcc~wCIJ-`8`71`+L#3 zT=p)1KL+789CxCeJysnxPzvp+F0&m#0{!~{DXaFDWq%b9SCBio3AK2Y&hmuZKb^uP zOD&}G%<{F#qc+s! zPeB@MkfiAYfV$zcm|}lCcwV>wMC62=R39CyDTAeVtL8<&+N$S2$6l+8Z7|MJen`o7 zbg??$PYps%`Ie-*T9XL^->h_Ssx;c%0UIRfGq~7DNPmWu8ijk?98^xJFRTknS_c#F_sKi-z6J!7jOf z44n39o{t>@=Djf%t^MhkUEmLWP}$CCl<2MMQj*eeDodlngcl(aep7P{-bS@Ack?0E z1AE>#FYZH!AxR|zs}VB|L_Nz_AI;1_lAVM&Lqaaxs_0tEdwn)~SR9c4Tqdna_fZX0%EL_m0ZLa-YtH5NW(lb zt$Ad!g)71-)EBE+G@cT+&yh{}7inNL_{)b`snB6<5k<(stB?Wph{$WU0 zVhlj#-O+nhl}*5uJS7;F>;|-p7yL?0NdOP;9dbAA-6Dck>=1>YGA&dKj2yHjyVHf% zk~is3*cT(Saj;XoCbZAh``=1q-Rniqf1UT>(lS{f_BnGE?`xah@O<`*UUa7+eW!Fl zjTm9Nx_u1|o3}mlp_{3*!C5b9{+u+yJcRWp!Ys*Us#TAh?2}J8PCPfSd$7$d-*6fJdA6b@?Co6bJ(wEek**H{RM>pr#UfGgD&(>M_n(%bne+u|8FG z9zee)Yx|dCkJSPvAz|r63wn?7y> z3soAbg&l2B5P7<8Ajcw?xyEDG+jGZXMk6L7YjFY!<^4?Bx`X%XiQWk095iZ zLcg{td~d^tYr`wfoEvbK`9s{}rp7wt=T1>hf|ThvRAfq6577J(!L~U-^vvJzDQ^ev z)nCOPK%v~)`HNFn`1A3$&CB5ZVvl!ZvH8fTFbqN*&`|=3qK8!()%o6~wOT66zoL*t z+=7Qpls07RCqqkW-f2Ia6&L{|IS-Y>gc~1QXP^!;{5^l0kOc0L5GC)H8MEZQC!t0) z-*l95-kGco)%P-;x%@S2YUhy%CMw&3B>Kya!bQGr6rI?YoaccNlGA6-gjfDjA3qdR zQ@e2puA-j7yoNzWtM@pKE#Jn7^GTB7FdsIy_|EL_e}K*1VY>_~YG_kHhPdw;2Vx1s zT!x!{4~xj8aZ&>W`1Hzn+>sFeQ~5BA6DkutE_^;>9^73HOU;YW+EI(}sM4HUFfu$| zsux1^%1QGb`sSt9O3DVHa{#ct7CK4c{X2TWV2h8G$6Pi={tB?tXcjwWop-Dx`Z6}0 zmLX9Ej&p;TXa&wK*=DcKLS5sx8{=`zurM^kRUYf9m^mM)buxcLIbw;z>sdC1ti0c8ukWo z((J94pLJ~vN7+?sTeDAbtOOr79wuk#Pk$c4U9@kYDa^rf`+{T@UwqH}(#WUo>r6O+ zz2oy`zB%DpK@>*p796JnLi?BA zO>piQUP%n%z!-x+8CVNSJP#J- zR?<8clRe~x1iWQI2kd;%iH80L`dgvoW!x)xUn&QZeFEAP9JFS|DcRPJ$+;u6-Y#mR z*SSuXAT`~8z%-R^U{XopemzF-bhR}E&i0$UyDTTkhaCu!Ba@g!vK`B_Tir|Ftssi= zSdXn>j$5Gg(pwX!K53+|?p~f5`|>Vs99#78j$RZeT?1x$@9j^&EL@;KOW?%^#p67U*vAc$4 zzz$?i-6HKdETGj(eCm>ya>2wPR@m|+Tktm0XlYr#zUy?Ws|=Ztl=@6lPPz3TMz-D+ zIV;xyDtrIVAt4`W11B0ny`6R9;%?M3vyUw}z%HViUI)C#EZxpC1r2;;)+)-dq@yGj z^BMxpBc0(g%8QNV=)lyLcP=q38GY7 zy*P}0*t)w=X>q;|lOQIdzrTC%n)FpBDqfjBOZKcgQQoB-nvu-^q+DhR|#b{Pn5D)tk{zKPC}s2tv;GL-lSZ-y?FVn zO22UqM=_KUNYCaA#}yA$RokYN)`pHlZZX>MvyF`wi>m^L$cR}k?=(FswP(Y=<<>W} z(1(ThaoAgA*!X65quoyPXKGTmeaTc7SBsOpYtY5hD|EW)OFd%A3j&XYpR@B3-I%o0?) zI67e-=UI+Bexe>>u|4%~62HEXQ+!jPDKaJF3v)GfxDfd9Ufs(g4^?|}WO7ixweIA| zYpI)XH=D-hy&G%8cr*P;bc)>yKOO5%;Wl0l4fecB=8dcYj2+By07<~W{BCtUn7OkBw=*oNw5VWCIwA#RiZ zL~`7Kg%4K9un1c$SgDhXWqt5$PUO;F)ble;C$)`Z*CB5}s=Xn|We7gXibc6L zIP#dk7m~EvRxN0vpo8m)JK=gV<42wRsI(xVSge?DJEQ1YW^cPhUduSH=EE`%8~a^1KoXJ3IoxT$yQq4iyO;%R>*>*1inQ@b8bG% zoo{>r12lOJmtA5tCSwWkD=2L*C|R87uh%WtEWpYHS053!YaL>W=!fHYE&99ed^6N* zgS1<4HhXF+NsKhE>ROO>$~bAr@Qm)B05B1%(>KJzV1q+_HmS`2=nl(;R@RZd6K(2D=ePejcGle9;VJ(Q9cAkf(l?q-#&2DMrzebH|eBdjUu$=I>ne$R0P%Z*PCOGC#DA`gzsfF9cp$! zZqfB)mdQPN;^R$vrtLk!9sRnh!nFVry97xO$0-&?jP9%}_j@i)gyCf5D_5CK{rNR zk#5E=)l1q0)nC)wOKtOYS(t~IB5WBzHF%4)!K&{VFLR(h%$43)9+)P8G= ziu7lDe2yBOK{Kk>wvZP2o->t_^a&ZaIm3|^O6ZsNpAr8g`FN+$v=e%13(FEVoFMTg z5ka#n+yi-iTzhZCmS+^d^g+XNtvAMKEftIO$1}`QUzD3e&7R-Ck2O@)BrlTXqvbd& z%X)tis}g?d9ob`WH%)oUXItK%<@()}*00fO8&)~Pa*HyaZe%a~NU7U*#!NZrlj=4) zYzk^v>v!es$}G8EMbSQ~ccX`MEYuuF`fL+JPGkC!PHtGos@lby3cg$axQ*0P3~Ya35G&%aoyQdj~W7fyJfXv zrnObUuu2EE=<|R4^`Vbbx%zApKSV_e@k6QtsOcOWEU;Oxr52X}Te%F45xrf8n}h#h zkW0N@vlFBL<^R|*oFdyAjvY%*I5c;=9Uab?{oycAOT*i_4}vG?Of2ub(?y{Ia;N=M z<6-A?L|YWj{X!;$QKcxi$9H)MYTC5bC3G(c#zQD8&9j9Q`^dH>D*QN4W#Q?YB^SAW zL)?Y=+I030Vqo<|7b+CV+}3Itr5t*Nxh_`06Q>T#%Ca4u#$~M^Gy^F<9$FJg_RFQj zW_6t0)4Rf*dp3z2I$gapS;I=Ct&;ZT>9SuID~d4R1!mt=ZoV4a&JkM?=win^wF(%L z!BWJ-a!=O&n2)*9)xZ0OS}p9X@k|fRSZ3+8HG{=pi>kn%#g$nO@%=ql#F|HOdE4}u zzJQ~{z+g}HV`moF3*&s1l>}kEjNgb`)p03lFV$2hBuJul7p_1DGpZx-E}$v-En4?45-7NQBhQj$$UIn+ zZM)#bb6!;#kK5kgrh|o193?3#RZ@QB>2*gTPO@^@@2R7IyPL>ApJepV5;^%(k)yLX zIQDmk6Q`=UC*8Et{&&c0`;IFJu`c`9De;l_E6O^azT9E>TUY7U^%t|Ql(!`W~hrd?X5KP`VP~#@Erh1l_k}TST zD_Uc#79z!JXA^hhNM|T8;ozjqSq+YOE1xEIZca!9%WVohB;(e-krWel{d*r{QvWC% z)j2XeF7eHy*OZ6Pb%e!LuMi#!uri`^yYc_m?BGLa@FlQynvzoVFeqAzO}YYMqtz1r zugn`c_|AZeMPY=c91AABD{_*pH{M>|;a6kdBG!=~%l#5Fmi?wzl88>l_m@9)1A@k=LXT@+ZX&?sFI6dD$ntESsXSfoXOLGK^H)?^I*j?>=4Pp@M^-s z?*Q~9n}yhpHkpwLO1f(0%gwl4t`of4>`9;<+_tS|g|GL{pB5w)hqfnAT;hdrYLs#B zMfMXs7k`-DU6U*F6= z28=15{E1tG&=en32fBxELlM>QbNLz_m0qhz#d|9Hu9!R&`TJQsL0E!8-}Z#5hR5Ur z0=rW?1@p)UB#KAyhY5IxcTMNLtX}mAHV|OALogW=dV={^qj#ab1~CfEYigpZ?iZ7R ze$=o1MxIm~8XA)U|1CZ`-})>$djLsZAQgVwaS6nk%M(UD{)Wo#n2Fp=V`6dhPl83f z@?grqki15n@N{_L0P5|MLlc*?@_eSBvPYA`?Ahtgk1+mzj)WT@@xPBQvVmK6^zWCE zZ0ML~BY0qf&%2y!L;Bz0RxeleS0<|Z!?9AWY2!nmKs}xK-2TkJn<67ui(|i*I_q5a zNGEiP47n$Z`7Rg4#r+u^T>aD|KI3`?U^_>9Yyul@IuBMlwRT>IVFr$qYb6qDN4_}> z>1Eh8je_NE)r3Np#CejQ?fvsB3p~RsD5Ad>;XHhFupVMNu;&Q zz-*laZ<9(FAarpX0G>8Pt|L+^8oN0WNgf^Fy?i4Z&F1&RmA!o*1-@e3ofzM+Pp$~b zeR!v`h7ebJ%0)V&*SJ7!PL5nq4VDgroWHo9jL5=l9E>YNip_fYSm$VprOIz2j5{y# z*h*dyv@)RnJoNXr2agA^7QeUBj8BP}{8-yO-4Nk=ERjb4bc zPI%?5bR=N~f5vr_W4_Yb z@THp5y*``Bp5AB2eTMBA>%S>SMLxg)E^ESY;X>mO*Ez5IV>1!7H7<$y_4#d(} zFg}qR!F!a|uRfjcM+FZ<@`-L*afXoBqu!7~8>J}jbF_J!M%yUkpXci%#|pH5%Tr=U zt#A^FS{0_TXC5q=EmTwPy*)!N0K=OEdZ>ZDo|@1l)2HL=AU-`$jM33&?w8kxg4>yC z34TR=%gy~~#1WsFMoBQlcuWeY+=pFMeRH#O7bk)4M+Iu>+Gch@V#|gFOEiANEVQf< zJV)%_&4;rj-4^5WOlk$gCc10)jBwWg?Zh{Cz%vzZW%JS;z$1dKCUWmqwB7JmPxz0a zZ62|}-TVdB;CdPBizS|t7m)-5wA@Ct7D=nI(Nllg)d6mEoVV+`y&O+S`#Nb4&xwuaCJQ7+t3%dJg;w{Pz8!{0FM| zd&C-l#V-TNEw$i%{(MYZX2Z0QLhZt~0Xf^W;EZ!$Rs0nA<0p=^TWcQ}!}bkYILvI= zA7{GR_n8RK&D^g&_LJ7Wc8)LD%axv_==(_CJx|gtclI>C@xQ^v_?_@KRP%q|YuUIpc&#uyG55zq7O`!D|0{v7zN`*M6x_&?#+xzs#2eJQfhuC62x zEUBi(EiP`FN05z^$_qynY80VT0Kq5veEo*LE?)QpURX7&`-aj8TX7VrBgG3yK?=O8 z0(!13%XM z8~Ef>kX>zw*-m$aewbXA1Evl;X1|&5fIkL2H+B0Nl?Q;kyBh zkSHn{VkG_M9Fj0ie@ovA{ujfhX||ewg}h&;c#bA4rGQE^tU391s){-`2k_Kj5lgu-ApZY>(O3SMg4a z(CPj)@U^^}mAtuv7&QYNXLY^Ij#CF=AtYxT4l|X=KcVFrk0vTIZ$$q9ue+z>F;IW?PA(j408!mdIJ(lIrQyb75h4~YrZem{6~EV2Ip1N ziHke7H{WIa2t42%SJA(*-j8!>;QdP51)diX$Xg(1%VK);`_0=fP0Q>@GIg!hrh7(ht~ACe3)l)CXHJw8-_V5I)MFg$Ope7 zz1|z9={nPc>X+P(9P1tMLHVC7>;4<@K8DjvIEq;2BpHDva7S_2pMF0o_OID%Oj}JO z;!lb%QAM_!;k34W*=7F#O~1DT-~Lz;$EmIp;{=~385%3gf4arZ*=l7A2Wy?STt z{jF)bf9(GN@osC^%gbe>#-MUp)_ds&G0xq{xjly`fnQyToTY|N+qAzE%ECoPtL&3L zI@T{W8>?2i)PpaWs>I}yNj&E~0&~!Q6}RCFRDDL(VquiEc?c|WcfQeqjtIx(b66fE zh2d4VhiOg2jDyJ;>JB>oRo!T4E%jMsMLD;Q8)e2v2$^o4rN27*Oh+n(;wmz=IX>s> z&+Mt<3qK2Z7fkSep9I#g<1Z6x5xW*vlIFuo@TQ<6SwQZWdW?<~5ECOR4=3@D!&-zo zUEYzXN=kWiSkC8+6-J2n_FVoI`dRP`TCvlx?_$pk3m^41Il(5*jS8XJQ4U;y@#jCq0NY^89St7Tj39s7~lcB z2eCNhbjNHSyo&wv{e=Gj;G}wniSY8@_I&uxZ#5Wn==NzBv8S1EmWhG(c|l;(d3_@% z=G%qjjGx0D7eUwWwFR`&V@V_AlP}2{1sk`!^cepD^$Pli`%w5@@Z-aN62I^ZT}k5f z)NUr!uVk8EB1=%1g=4pGmK7{;#DgPkCnOwK%JHuc(#)`~l-1tp_Gjw85phg%?AnXO zKixKzmzDKDs9FC2X+MJcpNH+V`8Azd%(AN%T@vnAMj+sD${6RN2N}mT?%%ba#GPlx zx{tved8~Np8!AL zqJIYLyd&{bOt-tguz4QZUoB+0vbsP~C%N9ybcFmS2+AYhYC50LlhA%AU2>&!Xu*F4V5P1thmINvX#3s#spj zrrpRPWhJcdVP!GR9syCEpTi(Av*Y?bgW*TQzuJpa@UM@)7qph%1=FmPG0Zb?mzKb! z_IEJOU6NFAM1b+#Nm48NW;m%+()u&}bj9G~Jz77EU)m@3B>4C6OH0!)?({o-UqiC7 zNUn9S4-1>MIA(EkePYG#tz=-~AIpwNLZEhaC9Ct&ZC-201optq6Aj2y=Te7|*&Jbp zFni=z_6hq=TzJFwZ1|sZy z13%!Z9|E+Gg8u*&tv(|BKGNF9!yZ4k^B>BLEvy>Vxj)(wDPNW|aD{FXA%M;2FzDFy zuvorjsq=XJv+cnA(;Ah-WtJvFGxxfI@9kBg@dl$6t=*eTh;PIKWh#xIv&rU0Rhdgh zLovxlUFu4@)*00mD2zDV{J%Zsg9^yxK1%GTLbXD*q` zat6)fPC(#=+B=-rvkAAcJfkr>sx>@FmhsH)8D0j()gQIIz8PxLg*o0l#_g2;Ov^7It1EH-%LB*dUz495pc-$- zzZYr5soSjIzX0@>>S+My{gwp%YxO!ODx|6B2l-d#AMI$#XZuii#UR3@*QqEZWZ~^? z+o-H+mAGBpDjV{&mHNJ+$+qC zrI2pjfyY6dkEtK2u5?J)%MMRm9x70im-&09nI!`rj7d;b8ne9fG8 z+IIBn0sjCVD*STq@U+s%HjvpI^VDNC$^tiUL%4oDy7BK;y!i)&+FOsAfd?CMKK2g- zo;!Qo4;wwER;15R@o<2fXbao^L{x#hVB0y1zg zc941#*FFCLr6XJ%U@;AY-y`}Qf1i5n_I=J5d!1Z&t1ZclU;=(ojAxE|^*`iSX4=KQ z&w%apshlmvz3ib-wpgnnJHYF=Jbi1&{MC_w)?o&oEQ zxhK>0tm;us7Nuj?$B49%%x3_4^#1@o zs-?Z!%Nj^<0l+yNan$ksD=kdc`o882Vvpnl*?1j$cfiGYQHMr*4fQ=0E*EGUv7Y>P z(ndv{Z^Xh>UT$0meG|bKf=MvEKy;VakFt-~9go_3O~QHLUKHw9ykt@@1D| zZfWOhwpP__G%HgmiEPtQP+Kg-PjMWfKmg~TE$ps*iwG31UH)U@p%PnOmZd8xp~?G4=5Oayk;(g8UUq>?<# z76Ph1?})AS>~DnR9OEi@_2-T$^XoS*-#o9m#xeBZW9eKuwo^|Vhn^O-`DU7SZ96^n zTCb7OLmSC1WRL5Q!{6|7{k84X>*9Zcaz+H7C)m73lFI6O78e#UBDp=GuHQhIMdRC|6U1P+$<@bvGO{m;zTCT4<-(KHcv6kZM=5xAPrjYW| zIqG--VN@!GQ3$>_5iD*&Ja)(B&3W~%wF}uoWN7yl0Dyh5$n9F%J)XDXtJbja%nf+s z<{=7&MH%ItimLQD$pyP+zk66X#x~g_#hxLzc1;q9?=BgN{{X6xka3PjO#c9p>s^fETzuH+qJ3}`B`wVX5xW`jpJn7p10F|hOb#Ex10mmfc7!~z*!dY}r z8EHNt@C2ny+SQuNY8aK>EE~Q>1mkxhvD_*2uOk;h!&x18d8s`QNBI4w+cm5{CV?Gp z5mq^skE(P(q{^F{+-xPG18b@0$CYz&i9^H~4tE zezr{S9DiZINeTI+PyzP^h8wTSbI*GDhvWBy^uG`4+BLS6UMqQ$;y|oRY%3xH19id3 z;O4Nw$|>?a&fHDq9SiwSH@UIXOwt{u zMTzHRP@zFXy+9zGjAZd%ZSZfxW5mA^G|MpwxU;sAZ!XB;L{Ng`aq06(u|WPT@H*qu zej4h~>t7OlU*cCgjIzfNZoGMt1GhfWjE?x?vc*(#a!AgcU9Qf1;U{10&A zQF8Xtks6i4OCf0wR{sE&YDhW99@X}-zrVH&ZEois0c^w@131YMWZ-)7UmzrdQvI#8 zWU*vp02{#Ju^p7eft+OJyJwEo_3VudM|J}f$j5wlt~^?_VHeowryJd$Jbu)^E`v+> zQzoBtcQ4vCE6Zr2jf%gRk+_;+_Y^J>NzO;0JlE$P-l-;@!gyrcEAAj3pmUb`*W16g zO~`E{!#Z-NUQ4T~TL&3ciC6i(dqsgGXlDXt&Q)+gJUh`OF zw%EhWY!W~}d4c|Bx;;AcZl2c@9jZt?V;?93Iriu2kH)zK(^_Vgyst6$gago?{B!MI z!{HwTUHG?M(!3oSbALL;F$7^_AH0~5f9aaGJ@H8F ze+u@MSQcH1yJL@qQgRdy03A9Gp4Gwl{{X^Po+r`nA)jIbj@e@iuhR9 z=B@22cRrp}Wg2#ljUSc%GrQFFeHQasyFY1@Yc!6>E}_2jdY(>u;=CzuCu?}FtQ^NA z8#jOkHz~@o>z;$C_OGfuN8#9J(rzrFV5U)ynH+PI`2M}C=8Z2xy}s41bn9>2x&h|N z)e%AYnb>i+(0>u&SLj$ev6LG}i$b2Zd!B{hn^`rvX7M(Rcge#$V;mgx9RRGzVz;o~eV*A{%Q7w>FY`#pJqsTA z>Bj(7!M;QNq`(GTD`(JW_56C&%59##I%+D&gI<+1y&mUKjYOt3+{#J$jxaOy>&1B5 zc%4}TJ;?-&oWiG{uX^@fbS|Oc>xklPg-|eg=O>H;I_IT)&?nSTYyjYK(0_$?VD94X zis!Na*6Cz}SGD*X;yGlQ9d7h(LRnKIb04uv0pNE7u4~JDEvSgRBc^HafG3vP+$-SY zX^AkqNI2WhNdOVgabBei_S1YW@ivzMd18-5)W%5KNLv}sJM`^eHuw)wl0OaT0_EOY zn|nxu2)}h4iqL_`AAIA0eR(4|uIzm6QD3onTyII+x$Ab&A|yucNV2nth`}I+#&QV8 z)6@Ca=9i4TH9fb6wR`z)?d1Dq?0Ty-dz<%nx6?}24BHj!IuE?ee1NMrK%>0g@{_ZNC+?ET>OzQRkZYf{}$oE^y&P_{iV3%NMy&MSs< z(8Dx^lDplO?1V7ra$MVuk+i!IjxrFFjF0!Lcfq&v-slkB5?xEMxt0R;XM^&&CMA9HM%%}q8+U6c~3j+*o_g2;iQ^Hjv!8c=w`_SoOG{Y{{VtlTzKQeKer4J_?udi z`VBk7+LofxFj%7A?-UV)h+zDooQ!f5^VC;edJ_7TTvB#XPv>MTzN}@ArH8fon*7h{ zBK}CPu3F+kQq_Lyu5$kX-2)iF#d-dzs_FLkEF+0`%^NE)4I}N(M*bWF+r4#OCTZ0+ zzc}J9#d0~@?ibV6ydPfDmP@}j`+wgOG;#5}D#}4Q$pGVMBms<%f9JVo(oWGor)l8R zTAA9d)}N^Pjc}>k#{7=)aCvEU*_AUvIWJ1m|H=?-!fOhR{ z?#@9N$TjRvyd=3_Q;l6p=)v%ov!Q%R@XA_hw)=eG)>rYB^2mL-bz$ai9$bK~K>z{H zax?QDk7*~v{{R);co$!dB3p7}jj&tHg^1X4Bp6UW_TEV+HS2yA)GvG^@g7KS8_9=W z^I(-Hj3A2s4XjnN0Ul&PZpazh2sP6F&|VOIOF~Z+SzFw|bgj}bkqa>lM&%N>R6S!+ z^8x@P74bQq6M}ja>tDM+8$1#4r~V2V@M3=n++D4u)r`?zXt%17$u^&Oh*gFZ=V?+_L?jZ8o}gFq zqx&j&u{Di;TPG@BdwDWYf^!iEoZyUuf;!jfcfs$BULp8T@oP`kZD$7FKbA{xG4RPC zHbDqoK?Zy6=;(ap5PTc$M6LN+QxB&h1P{oT5#1KaW9+knBW$J>B9qo1U zySL$E^iDhC@9lr~`&rI0ZFEwyQvU$Su71J%H2A0DX?!#X_M+=c3_Bh${46>7l zjn{x!1}by7aadoq=Y+g>@#FS9@xQ}Qg1T*`{j(0@m7bOYD z%m+2(ntjZgHh}&lj@sJRXnB$C=2umXmw7v4!zzM0Ge%b<8$cNFUxMGX?z!Q=6zW=5 zsj1C1wDQRc-6xqYLbmM2yJlcUcD7ishWtsUUMYo+rK3T>^}LNA2fhoO{VVbF!q+j|_^!g{ zLD7}__49rG>-3Hpr)brE9KS?}=zKrp4}iZKG|LYX_~%r!`(~NptMR7Z-9|>( z*_X_T_VfFDo9aH76RZwmOI!C$k+tKhE^ z+1%;Z?XSYK+TKAmwZf>fu#z&-PO-?LmQc!BKyX1^f(?CbJXL8WPPMJft!?bP{LcJW z)a61IaJN$B_wUkI>EzM!>P=yFBzAXEZ)xRFt};w`>V3{SbjMz6*gPxYokLrw{lse#zgkcgKH^TD1Dd!v6qZ6wCXvh7nUM$iV9y&Ps|2Z&T6pRShl?ig&QFE3b^QFAwC9eJAXwUYH|zKHl= z#a=P~+=j!$_SUGDI!>8yrKF4l$vklGCESsLxtx%SF^;5@#|dNMO=rYfwYG)f=)Ti; ze>^s@Mlczsxs@S}7bGrPMgSHdWQ>qGuCQp2sp^TQ>8m7{w;wDvkVf*&C$E&r!kRBiD(%{&Pjx;lnTpS%Tk)WyP_ zIbhO#E<7#*RT@%@MMA> zOQf<7-Z{0jMT+*+47TVO&OkU>_X58Sm3V_spX#{y@>E;)erw44eD}i1!arNf<;_X$ z-FZLBdFX#TJ{0iEc$>o7rl$m?-RLokwSbM@b-Or8Spdl|4&dE6LV6G@l)07C=&qu1 zG;C3W=X&iJ!TJy3?c7&)@bATvd?@&hr1;Q{(Avwk9 z_{L3dz*gyQ_>vR@ukj}tAoS!{?VrTo7e%ao#@3!B3+0Hln-|Rq z_S)Wf_eu8nQsNaWz)50>kSe~=P=Xi^pU*~(@hV>wc;*ibX;L%Y>JzP`Y=uS@_?3vr zAS!Nbj+n1(_~oH#{ww%3@v8p-O~j)~@ivL6S#yuvK?F84$fdD`5$?`=SMHL-EB^6l zf0LDKJG76`U)u}*3Qh2n_Sf<6g#0t`eCiEp;XQ5yv$wg8trlxipKE4Vn1)gIkf_2< zrz#i%xF@;&s4e_a@zeGg@m`3d`)gFt-rHWVw6kxP(mOeo?xc-E=2rVZow;~vhV6h3 zYx$A*8{!WcS@_Xx{0}3+dpfac;kAtuj zSJmTdsbP)7sM-d?8-_d;Va5nNdvj1}w`(+;LXRp{&Zj#e zP%|qWbU;ArNEP(7r58n1*)(uTS+vX69vj!S`Tqd1Z6J{z=PFxiD;l>AAt2xc8Eg_V za0e#8Q~m<#HrE<{z2Z0#p|RDqMr$N&77$t6!(^Gq9bq{3uO`qw5Zrh!^}IQ8ByHjZ zA@4u}RMV1AmLdnF%UL8$mxWZkene=%rCRqY@e2u#fVUIqGn6?b5d3w7Hha?~HPJ5tUb%DKV*I zg$%%85_)GD?V6qs4?_wo>ND^N&z8M9ahL1)RrIm2vA1~RxtT5J+Fl_b3m{%pvW3CH z4p8Ha^~vMY>syxA{v3_`(BdW? z3(KAkau4Ox9qEp`d-=CE_J~PRo^fU;j;95-=ieEvBU1eXO9jrisJ@)KOwyFLjrXir z95*K*ft+NLIsTQ5*D}bhB)zf8>OubiKmB^sOIW^Gm7{JrC9(kTp2D#b_R>csfMy4k zU{6k`9mlchS}5OQ70V`8X3BMJbmxz!u&3V^Xqr~&yq4*VezmI(p`=?zV_SbQm=Th_ z{YOky;o>-on36rguyKL3@q_EeI6T&T@>se0rOm+vR#GFQh1Z(atSyx#sTZw*ENN2Y?Ep4rV<`V8Keg| zCy($a9<|VTWoK&*TS7o_b7a%Tq+m3*-#lX&{vb&DX0>$GS27!I=+1s!MoVu6&3k7i z$ByPT1zdsyE^)^>!0%e03tszG@s*AI^1H`sUglEenBpWZG7q>J&p}!8S}eZ~{6V+o zXfJFe-RQHd=n9@Yra1PkPla=vW2?^DSN_SlIb8g_s5$SQr1bZ#p`|%8WbaWkoR?6$ zPc_Ek(~YcfKvrxP&usd0TH2n0Wv6&*(^;1P0L#0(8T%xPtmQL};Ks)aMzmWjZzYtitcag~B~J&Oo(Eo;J!z$VP)-S}5Uf%} zj#r6eAo0hb!2JIJ&TBxz@UO`1K~Pu|(T7v?#aI^VV2P$tlFUiYcpL%8T#x5k$!l!4 zEYN^mx(suSboC%qN~%u57_E+LPBNyw4607-5rfx(!5xQgJJ+a8vt3%svXQ<>Tye(Y zM;r>^_eT}n7bfcW*vLBIju5NCX*Zy zK+{hU!(n&_uMNOI;nux-;V+B_g7r(E?MOmf>J0)Gk8V(_5xK)05EZg|o;Vfdnmyxb zR)1leb|NgLfXacM=i5DSE1}b5ZGQGGCq`wG;&N9GLZ~VW9FlM_qd)z6%N1stdks>% zRyzCWbp2oBzPY67vWX$_4ZfqNhF%6P?6SmQ9!!z&J^8PiwEqAq?%8io$R$-}QNdh{ zf%WI}&3$37co_UV_=(|dS4Z}`l=NTTkJ?q5pJR7dsUtL;S z>GSzeT+IY-A#l4AM^!4!yGS4mV4Q=&#chSeLfSry^%f$8tdlqgAH2$fL|ObZ>Gh}* z4=H~BLE62~<}20g{tkRu)ty&I@V2LRBsf-r+DU=*KPGSwbI^CidPT47kNZX3L?&+z znB#YpeN}F@wMk zt!|O8J&dEjTC<`26*dka)cjATLnMDJkOZ1mQ;vUm@<#o>>ktnauP+mfr!&x@gPa}O zJtM=KotB;PZpQmd#9D8SY}(n55-OSA+Zbf+J-L(}{%`1)FdG0aK9Y4hzFNgmC z5IjAjXjMd zWPLlq9xT>v@eRT(ds{!j95TqCFa2}}nydRg+oO0(RlIEZ*B46`@8a7OYJXD4C%tt( z7|}dA;SDoVveLEbwJ3EMqJl`@N)Qi_vEHEHIZTHjf*9k{xvvgs_8tV&#*LugUNWqv zZSNhsHs)0g^AP-ey*hRJ1u0*f&co#DdGCN9`&0I{)HR0+QX6aJjh82BiYQ^p^e5&& z)K}YSs7SJfl$jTh00TVx;=X9mJU8Ia4(rY0-7Rlo)Fl%6vx%V$!gdWNKXgFJ9P!6& zRjqf$2|h`eP%51BeCjdZ?gjq<_0z=5+R12C6{BN={jK~&1OJed5c%Jk1rP4JO^5Y|kw6 z@O!(-yYS8{nwR1?iL9gzrZg7L4nwW95)py?B4m(p=t1?bR|QeR3Nc5Mi^E28a%}Rg zGfKXMh$2HJ#F+qOZYMneBp=LXzJUFnJQd>^ekR&$`i_mLM`vRc@+`60&u}D>)5=92 zN^7I$g!1fM11ylmC|KegZkk0{S0KBn=BpJ_T$1N$F8=`4io(VUYB9O` zV{4#zv44LqfvLwC0k_F%EK9WaK61uJKR;^Yyj`MryTn>e)|ug}FEU4uWN#$)w#W;N zHs-jvix>w#PMJ0Nzj*^{akN_S`&;f|AeGkR=Undvvm&+*c?E$3Adz3Ezp$_T5+ma$ z?VoKQ#f@s(&*9&~%S3CfL&iGv3lcT7Nxn&RxMflHuoshZ%$A9r-elNlkBL(Y8g_De z`fhtMsOIeUJ~y`gnSLi~TC83fi&6088lB*8v)f6o*co5S8{9`7ytBI$OqbfHXyo8m z#J85P-D;9p*hdsLJHagv3o{`utuFP zh5RKYqUpC@DW41CuNF-s2!7LjeibFPvTen!>@9>5gyu!s-a`Hp-9(V8!)~EaF9kw? z5q|Fg@CQtGuh4OM4lgg1BaBH(OPiOfPiy(aDf1UFncpO52QTK4-yk>JA5PN8mfwD$Xao+ghUi{oHWcNit3(icmpR zV8DUKc<;@8{{S`UX`Q2T6tTee&(ghK^_tSu$6^5kAcLG8-8Xg}I@gXptlN(VuYSLc zb>Ur;)P|Ivr~lWpqv}&f@ax33g}_;~jZp2|gT;jZ05^W!vtKayTKj&1qkm_}rgLj) zAV*`LEs%0fILhZ8IIp7Y>@_t8MWE7VPohIH7j zyucx|wGrI721bm+xEn$Mtfz*-ZaN;rt25SdensUg_fkv0=z0f=^=WjU2x}MW3N@{T z%!I%{DdE^!phaJErUn{AKv(;{?>T4F};*+AX4hmU*ph{P|=%0Kom*?Trh9NEt20MhHJS zx9th>+QqG=y4Lk%k~w9SW4MeWVHoTU5RC(d0B%;t1mgz1N?1oIu2y`@!Oq;w{Yv=F z;mvEro)n)-(Bb<`u*q+G38A zipwy@NoBQg+`6>?02Qy~j=YyHojiRi^6tOk&*~>qk{FseFhefpC$MAGl6v;%*1Ye; z7R?o`vjtM|p_FbTX^yBlFgBDxO@ z*lGSQoBk1t5g(H<3b8bbA=w|_StpYV7C<2cWA9e>yQ9H#adgvL7{1dCP_Sjd*&91~ znN=VLB>nE4NG7rL4}t#x5qwpsNiV~_KIZmILmDz8pz@kbaviX8!>2_!0AoD+cxlSl zhNcxFo{vNB>C*k%fKRdNSW{%xKMCYgSq}=h64vR`bqIm;}#sZKyAm=sc9v^!@8+Zy0TTC;&X$(^Cfl=lW zvJgve&Hx{HZOaVfoK?LR%fvn}@Q$_czs6$gO`mtz%MU0xiPLYK%IPF!823^;fs9v> ze#w?Hp8?%`Kh-2%N%TAWjVwrnkbcD>Ovc|MC@YKsxbe##`PanVb3U7!(xa`r9w+eo zSiSf=@uuHSy>$@VUHN<56h?x60}+Fcqd6JifyuAcy=!c7CaBUfn|2c}#6ZJtHo4vN z5Ho_KoF0|=qx({LM*7O?A00&Ol6gz^f)tk>Hp)j;A-aqK_}A!H!mkpaiC+vfe;7v- zIlO^olH3IW^1B@woMR)Ck&GJgF$&L_9?VjbO#LAJqco^KDt^cR01dt++N6(l;SGDp zmPU(t7Wa_6OrlUnml0xwh71cRUJea;55p}&;=@~uPejUJx~!3IbdAOXA!dk&03E;{ zp^v3|-|X@IpYb#HT=B=l85&t;j>Kvfs?1eY>>JEQECzWC^8zs6sqhbnwz@pND1=`_ zFO~pMJk(vFC@NY_jqY>;;3HbY~f{Id9Y;v4&~6HP6vBvQ2Q@=2CeMsVS< zQIud1N`)iS1RPiEM~XCitKWoo_qNu;YuHtctgRt;2o*$tv2fV$l?*!(k)MfwW$%ZY zbNJ6y@Q$|-y1z@OLh;Dp$FOH|1}dL<$j=Nvhu`X|r7I)nFw;?tq;7w~K>Ta3_#gHI zm|EPJE<9Nsy@I+-+k^|Z?Fdze-EcT46_Y)BSK$8u!)vWR+u?qpq*9#tQts3!# zW#1Viwm~3$EBi?JmGB$kH^#4lAMm8UC0N0CV|8!kqoS(GZzvd9plp1SDdZl6@Nx}* zExOO_WvBcS)qHW_FC6M~>oMBgNn#@~i&mQ32&iQ!cV&>|h{y*gp2ObCsVO%~FIQ*z zY5AWGpH=r&S!k8l=c)No@gKuJJ@}XVFlk;F_-~@!>(|<&TTS+RBo5+BbTg9WnPX4h zAW~#>P~G!hd+;m%2^sN!S=24{uZ|ksuZQfCCT6kH&)ctCBjw#Ys@udz@hC~T=-hMt z0Qg74{u=P7!dp!j!xk`UI&GkxlG@TWjxB((Br4-MCnTJl_2#@2S^c7X0paa8SH03N zxz6m3&dxBEt`CidJ_F5hvC5;L|mRp4<{f8eoS2Ywv>+JCUFuj5}0*;)Sp z!gb)MPZH_gU`4-{Jvm6ev65aH)@wr|#vGIq2;5+(Bj)c4c*Dm(5jB|gj}hHLrP#tV z6hU_)gT_H(1Hb!681%2F^?!@s2tF8iQ%kV;g{*AT^{dgQ&355SsJd?^N8PoNb}1*E zj@7@msOFjFM-Liv=Ubj9{{RNcd{R#me$0M7_+=H-8+`}FlSvVZW(38m+QEJ#j|3DZ zSti=Uk&NfC!v6q*=xN#qz(3g9KZSGa7cl966TTs7OL$f|tJIn*xiS_sve=H+0u z&f|w>058l%d^7(51X%c!@K5$D`1RwThdQ5+p}4TrG}vrWu0mTw=V&)`+N{edl*cF9 zVw@FVqbPy791n5-0D_8mzu|ZM6?^u)@CWS2@SDc=zwnIs6G7E{MX%}>4-D>Zt|hm& zir&qfm|EmQWZcDFVn-#%%%H5sqNP%GXLTL_01N(Sy~Fr9;p;%mAYMOzJeA3)qNfK^7099~_ovNy<$JgHie{1i6{{Xc< zbZr~MekagwXMv}l4SPwvw}LRF21mJ)CbN)&*a))Z^Ry5N7+(u~VgCRG;Qf^+kKniL z=K^aUGP-!}@9exY1lsnyJd4rwE2xCiBxqaj+&pnc5Ad!l>8}X%BUkXqebxX)kS1jl?D{7W~+Q*Kpb8Qts6DQf;-!5#%c5FCs0j%fo*UJSNMry2Hy%I(D~kQ7 z{hT#V*&kcIvefO(ESnsQtu@Wx$D~Wd11pU!m$E;jW`Aa@#>z?XgWPVH-Cv?u|htsbL@)P)F)K9pVj4JmZIJ znKqTXU$>F@7ZUIkt4E%3_m|sGMwZ;md-rR8&E3EA^Zl9ip%;kt9R~5X+v+y42hU7H zFh5-FV~l>a_LuGb@s2$^M)23d4JuKy!g_m~l~*kr>l);VeRv211ln6IylPKC?4XWC zeuH?Y{t0F9GsJq_zZCxfWSE<;w z(OcJV$k!9%bm{OMIA)W)rBS(gMP5mK85H@FWD@nc;%hd zoNt-YBLrt|6mG(e$2sI5z5f7aUm9BY&*3k`-vww5d8k}!9vjts>&xau+}~z>k;iKQ za2IgMX(VO@3@GjS80jQxC6!4|c9F`GI5<3m_*dQk0I|=Fue>ASFNjyx5-q!0cxwJq z=OmVBjB2FkN-lTM{(h9D8(84J33xuw#Tsx;2f}_Q@ty6+KWewJ zoN4-VDH$mD2^^9+9m6t$bAjH!Q-5T?3HWE>_rxt1SkZnB>3S}t*GLxjc{MF@>|7D^ z7MB+BO1R*zaCiqN2jh>~cS^p~t>BN~tomN7t2oBZYgVjzWghe$_NKB~_tf191P z`&IC(Mfm>!ZFv>X#4jJ~8hk_rmMbf9Z*92r3i2$0l%6mc{Y`oGs;N!3wL0m-+cWaV z;z#V`;r&bFe~Nr*@w-~G(>0A}R@5~$(d-}^wz~dw?D6c`X>*X|R#%cv@dY$9%^-%Ui?Kr1-uUt@fF@NJck z#SKeZy16%FQWp0fe1I&|m4zb@F;F+K5$)V^NCU7H^X=vQF|$RcU<~ygKMz{|)bPu| zEUn=v=9!gGY~+-bc7nQ1Cc4|tPb2tS#ebw$YO&SwoUWg;P;M?;$=S7a(RptC%-ugk z&^$R~9hR9CR?w#HvBUsSSe9%k?!=Nwd}AS z8lG|g0Fi6)4<92Jk;YQ#IQ}WG)cx=1^Hr?!P+eg!!p!{gK!b-RbL-!q(ypQ_D9H5$ zA8LvvKtXldGBNmftC6W?AvwU#JMekOzp(9J#*#<&bBlWqF0!CSGCnd- zZaZh3cA_*d)KR3-cg6aH>UVHj+K^*Nl|qbyTPG*33GG~bQ^^g`l2*>v!N*aKIIfOA z4(m}&v0Pfnh4K(f@;5$#%Ji-4KZSlPzaC|^tedh$1>fx3#udujAdc5=rVe4De4Kfj+!fsrYZ;m&B)>!=&kj z{N#A}Sy}SHWQSP#U}1Bf->oWl(dcBNRaSta?i=M6H^Ldo`}Wjo6i;C#Q2Ir{ThmR74B?(|&z+1;L> zhtj^)v;B}hC0;^iySvjY@}+{!eQb}wAHn9c9Gw3EPPMIN`xWZZg_7&UdR!xrz+F2? z0qRPntAc)vYu2H{S+xEPH2(nb8Gn)F)#9wOo%!)J*ZsdA^FA)Tw2*@&XXIdW>x0&m zTQbMyVBj2MA9ve|{R_4Ijr<=Qqv{?kjzt(Oscm-L4%t}X{PHWWv;CMp5b5!VZZ4&{ zL!8MNyb^!FM3=iU+nV+1cx#_jvW^yCJ1@KKkW!86Vl?C;a56%XkzcC1CxX5OXl=Jk z_<5$vyzLr>p{m{m&r^oFk0D1@mw6Oy&Ooh}V(1HAkLb_Ud1pkHnkZ1H=$Gyt%cK5%*hq82()6tA!ZP zJRf?qq$SRs;!QhHVH2v(g+QL(LnLJ3_8^yW!WQ5RJub98q%qop@u3^w+>`@R1F+}-<) zbIzyAD>FI`1C&y^)4EI?nhb1h639WSZpBg^`#{bRQ712{p0OB_wE2%MpgZ?o{*$;P zmW(y34Z8;JFzTggpfm}qkS%Z~bdvp9S3S$bkBk9G=2akbhg*y(G^$01E52aW-0{+g z&AleGyih^eLo!^QiDk=SW+t4+<*5~QZdwFVG;)oe{X_Z+UaI6R=SX2s%b@ispT1C# ztxh$FVo`g2O2H340jI|f_2%-QKAEd6e|@pp5ay-0Jjt1Bzj6gT-py$hmq-XR|1jAn z0V&`f$ko2MKZd84I(&bv^uaUjMZ77*14ASWDGVai76l%ZT)ZeJtknHSQm>O$coR#I z1AiW6ZSh<0uc@&YncTO0G$CxQr@eW4|4k(j#nvp^5nuR4(-Ez_zLUu^L|KIPsq*gF zt6OpX3lX@K@y_9%-kBX-qzCS1(>?x=q|d-8xCo2u4Uqadb>4jTkTeXCMhc&zSJ~v6 z>509TvpI4yo*%gCfDgVq>b&9kyI@dj&igH>0Jd&-1(WVT0wS7n5(|J2+PJ|I1`Rb0 zqFx?(IS~D(+-EG$lk`8?(>uSsTguHFYQJl9X7~+41+?vVvvUmL+qb6ox+9%SJG&*e zwFQ;KY%HfK#y?Pz719R%yTS)ucjbfox^fxO58k;D=1@DSUj3U-QWp95`qnDqHK!r= zrQ^0}`wl3Tl9|vd*AEj^O=76(@U;&`ehyXM9_-pu%a&7 zf_sPP8)d5~H$;Vi@bJ7-!$Q;?Mm$PYY#FaMU6&kPU5mOY5D8Al5H@tjbiUb)&(@gg z9Im*0W>-ed#lg46hNX@)H4-=g$9uCv^7Q)bG09 zN!^gQ69RiGymAD2za;DrosXIB3j?ib7cP6^PJ^D`6x3)k0G>H+5`7s$*u9cN0s2v9 z&9~L}_n5|kUyOE5`l4Hu!<*qzIn8kYU^HTNZ2YuBuD>ci<9=+!X)A2grca600a$Gb zXHzY44Bv3y`~m-B*_ zX2~J>km9lCP=V=h)=ke!9_WukAFI1S7dN8t2^rbKY(*LJ^k8ivNF%*I!^X7gn+prO z51n_XIEC7ic*R$Db`WX=;x*I}2z212k@1E?xyUDo(!0dJ%m2=X3eBJIMpifC?8*|r zxT!bAxy&4RE_sPf4&Lm%$o$wig8JaM@|{0kDuk35H#6nFnb;FQoNAB-dONc*qc4Dv zPrli*0IIK2MdbBY=SuD(0FfRqEcs?Gp1lkBa`}Qcc+N-53w@Xr!FbjH131{7q$;)>b@?}VDHAtE z{M-!j5U?LK?Nx^jbz`pW))#zSI6}Q?ekQf51h64o$5s56M8*~6ZXH>*SSk@0$1S{a zWlLHcOOVOmh8V!g5;j^TojqdgafMJZUEpouP);?eQ=K>jG7*M)52X}3n|EMij{Wtr zS1kYFYt7`*zdJ_b3uL?j105F!R_foto1NeT0U0d+qE^>`B$~R<)9?QI7vi^IQChj$ z*ufhT1W7f(tK&9pDJX`6u7kVQ))+663W-*ND{ZVk!64*Z+&Xo-ZydX357!T^wCO|e z!70q?RFsi03aM$!^mhRTqlS9(@;pbGyG&b32Op6CE9cbN#TqTIH^7A8c#d~Z_GAzP z)i0t-T%dg6C}mPsfa}EcC2)7K@ye%AImTn@1R{4S48&O7{Kmfv8OaY4MSp|$GU63W zI>y|x0v+aH<$ji2#P{L?>3D_L;|>IVVY|s1#%b=;@71aOymV8#Fn@Z2`}^`xhmoa1 zaPT+Z!A42%$}Zzo<#ny*QW{J^v#kA*c(skgQJ-RTt5;U*+7IgL+aHX4Jr1w|a|iWd z2xRnsBz3S@5#5%+-&oa-S)AtxU!KRG(PAFbL(D9ysoAT=;N9SlPc8@qww<)-}0|mS;(RnsXb0hWEii;aPJ`;AtQtFIc!tJN&$*%`r3W`xMj)bAvW2>W`Gej|xYxtImqhcSez_EcP0AbIZ#Y+Z4x%56&BjlO7k# zF2Ubm^KqYo)UXG|eK;UtG=#DS3ql5K@5tn#bi!Q<{iWn%8HPSgSA2RESZ=PC&t)FJ z{h;yG%>KI~DMp955<)_zVf8w(cF{YTw762i8!80%1vCg{MINxT7FAisP|$+|t6;+931fX|O@Rs(*ZE2tG98ro@QTQH)O(%sEOX;2MR4wq> z_C*aGV9%zeF5qL<<$2fB=}CREUV%&>c!bn=v$>utdMxjTL&`s=%Oc@7r;ttEhOP+s+9Z<3z%X<${g zr>9LEeHXna#f5X4s@{8ctBCu)&^e34@n|%HTd3h3UV;zVEPu)DBmQ}Dz_+uoq_3uZ zSnAu^`6}N^pvu}%&>v&kO)|73YiSkdw@$#hdk*QPGU zFh8ds|5#7@QSi4~)u(*%$UO8`s5+2`bw4NAoeTm5GWv+tfeAYX^Q*>h`&sXNTlS=N zH-`Sc$QOxoI{erBBwy(ZYOdFf91GEqgT>*W6q~nB)7_SAbo1taKa8(7(Xjc3yiNM% z%t6Z4ekJpf0eILbrTo(*_JR~KdMQ- zFD@QDy3GL{0t@GNU((+?B_@T~()Xo^%b#Q5jyK;W9WJaAEEy+v-hhcow7<+gO+Qdx z>-zfY@YU{YgV((m4NLZ}T+LqQC4fEPhRm#%p2@9#YaPz#0T8*#;H?r z=LqIO;fo1gaJz3}=t+I3*c68mk3VSV^MgIYMM?k8@-Rrcb} z%qF&79plMmc~h6OQ**WarK{MT#I*R+QXf5tiFcM)yW90!fM(+*@R{t0kfLJ`u56-H z+%N^`k0-@_4iZKOF&kc@7k#fDsSe`X#vJn~`h6^KBAuycuCl*Knd!^dx4R?f0}|-0=J<(g z7%YlO#Xc(wVw|p^Uj_ZcFM*2UkASv~VG(cTN*<*ov>>ClkXv&b83_>NcR&j z%~1O2)FnE3oW_47kgvqYIhPJ*;5V$>X=3q$BZI?+f=p)$sBxu(c|Ge|pUZaZpr0g_Z&O;%vXQ&xLFfCj`o0T*H@{74##Vvzn%+tZGeN@Dmwp%{$_aT!L&M0} zJ0i*@KtBrpD?A)z@5b1zKdgT#Q1?&Q<13GWqQR}Qc!9cN93W(ZZBnNk3+X})a8938AI-UUi6zKO+U2`0#oSpmD zQ6tT~sv)uNb6tWA>F?RSO}H)=(B}i}jnk?V2m2n^EXtEnfc)(Xe8MrWR z?m|K4X{+}z?9?V91^5RMZWgNfjn!n~42>^VM85^GxF1tD8-FDzYW01~!#I z4ZzW0^)Y)lnW}M2Gj6p|>gXouCoa2@B$_j>si@~lOWzx0Z8yt|Y4_?%F zs$Uq{cz8*6H?2M>yll3M8ZJ>l8jZ|-d@4kts0`S%%-dX$yaBG$Mz|pFO|B^ig0B=r zGlISDewD=yoGdBKO^hbW%oJ^MiY^W0`cTPsTH0p>R|pSqjCW7`9D4Ra@TB)U!Vdh96aR_k7D&ijF+2<}k8*9w(EgHhyK1WF+>Flhh%wvo~&H9}=XlgHD z9l;R7U17A5;Z%z~I$9HT(y2qEHYAxv-6v8CH}7N-GwkYO%uJSu{?;7oX|4Ahk`0+B zI!!jh3Ee;{DK}!CDvB?;ROipW7UZU4&TD?P(jkqBHJ9bdSsv8wc+S&>!j*Pa$q}bU zd@KtfXl#-8Z3F7tVk;s_GIL;!Jd$B?h}-oJZ$`qA{JnchW4u6F7I_fr&?s6a(t>NtoUFStx-rHk z#(TkA$(y?G3CXcnSf7UpQ{etkoK|EY&*BI`q$oC_+6@bU;z@6qBq51b)RH+8hXJGc zAWU(ph&T6-uld9P`=F2S14IVGk_eJPHt3?{pG#n(<9xm8N>R=&!f4#i<`MTD-qPC= zjaH zE^d(U@vH{|@C^||(_{pQI=kj@YJ>mzkBaSCcAy}bG;A^|qBp}*bo|7Y@5Wa1_rPHP z6B04OadhuguTt9e`RUN%SpU+A$zm2?4i$@P}B-KyB5y2;+y zdHFJ3ZA0k|mPDTS&*1Vo{EPShqN?KTA#6b$L~q^JCq98bb#jq)B_O}z*z$acoOJkJ z$PDS*=K||zNpb7`eIr&iz@8m+bZ5@(m6=cJ2f8VHHWmn8U-WnEVgo=T!Bd>#ozX0$ z3eiJL9i*nk-DNoas|p)MF7GMO zj)9YkeKgjLKy^Um05vJ@#wZD6*?V4?y5v3Xi(b=cdLy9nF3FXjb%5n&9vP7f+Ji?! zXk6Mj-`Z`^h9THm)RDgRjZ84xSst^T=N7jq>oUtlxa;fuI__qC<zF*$ki{%aXYQ!%SAE?q6<3g*k{NZM>@dt0_CPB+x-RX`mq@8eCjeuM=7!2d zn+VwEZL}78>oep&7OM4*+Qk|(ps8u&u8p*9s3i|%&9tRNOTJxtmJOOtP5ritn0tM0 z2vmhQoNzk99w@P5b-FDd)zdl6{?)PhUHVdZQ8xLOMLwxF!^a`I0o{H>DfE2vH+%uK z^DluC{}FY8srpSAiyHYkj!(|8>EZh7wl63z8i5NYoK3g>WB zDAl$%5!+(EX+rd)s#-wVMt-GnKZwXdA;A{&c}?q%BL-!*3tM^1_8-X;_;286w@?TZ zkoXKSo9p#mxSA-3@`Bc;rnamPb#--#Y1D1)hr`3`Hg8!!k3G7rI=Cw<18;zQ1T4C) zVJvTc3bmMQ+@^^3kr~Y>=eTWCoio!mN%haAkXkUHo0o;(RR-K7lP&$hvgm_Wpm9$_ zsSlyG10{uXBCkIT+3lzTVDvqYu)YI_lVttjO?^?F*8e z(}nLeVBpqMKxE7s`DUzYtb#v$1t`mu9^lgiz+F8&n!EXPQxuk5iT>MPaJ!;_(T?X) z3FoUj4+7)r6Xvci;?Yxa0<9j>=LtstAir32?lf4(wYZmDHv&gsgS1DrL?+29+U+ne z9~v7%H@()Kdp`wyRM3#Z8+F=g5@Vdc$8)pPCzz8w!lK*qNA9*VV7N{px7DBVW*fZL zp~Q$yMV6bbpCk&u-GNq^YT5--5oA8In{g#5o#y6+@)SZCh!HjR9sfU&WVslNSQT24 z>P`0eE13Ps^~jd}sKNkwn=dAOo`s%5_Y!nn+-}8KYX8kXeE3NKYW7;h;c`l|IRC=Z z{uDAv+v#MYsJ0{P?V=JOwP>~VABhUMfr7&;$%!7xVIQIAZ$>auus{BfB(}KBNv6GT z&uT*T^enPokm-&)q)`3V>GDVPp1!}049{6feofZ98nJ5VM=cr)R#8YApRXg5!8OR~PTv$X~$+xyznlxFFGT9-y4+6{zZy zm$vD1@6+pGFkXHxN4-xs>lm*uG&RJa!<>KS@UPuU__4b2%?yL^wmtVI7Y8%^#fx18af-G8A!n8&YV9vyiP*!3mmObh~XSdtN z{LvRg1mx{z;uAGmyNyycct?UqmAww%r)0dIWZ#?=Dj<3MZa`(P#AKVG`~d3vg|nT^ z*bE;*$LAFIa^Nh;lGN!RFtE|}$H7lbiSeeTLByIboP@BndZ*bQb%6EilSYdi zEn1vsS!^bc3n%4qb>#j>vKL+4M$t}GPCUfsuq2dEooaGeP7mM@;-GB7nq?Pb- zXcQ1-0%hGpQ!0sondH0Kl`X)osHOhCM%pCy!jN#ivjvAEJ5LBIp7oCvY5 zM6?P^oVgxRe8d2lg}r|>PB~A2yI#MIr}+Is=Uye?cLw9@DFRg$@x;m{T0js9474q` z)`8W_TPLVQoFg(Fna1Wb4CdzyRR}afn(ME-TIkkiZC4@u_0{a|WFc8j5Y#I-EdTWAsj;J__*8wkk`9BDm#9+s z(GEa)w&fUteI$DLmUk4rPmoS<=C#qU|5%V_skz871?|hiuhxetFv$wITI;FE@Ug3_U2e2_pjGkA)tiNf_uA;NAJ%iE z&5zDRYF%FA;cgAWF3L7j?i59DLk1z;Kv?CQG2%80M6g-X+@dglZqEg?j}iyoRy-J& z)WJRm;u&73q{vpZkp4M1j{bfDc>rRrY=aNcB;PI z)SCc?RlBXQf|o}J@Of4rQT9Dc%Gw~Z)VGr^&MdxyyH2!0?tvep&ZW==7(@-?sIXM1 z(hNJa`DFU9oe9sh)Q~d&%9WbYRQdeJaMf?j{AqXQE{Bkic~J}~FOVbtbocd%Ni(}+Bh1Dl*n(;kQj&D=mJK2bjq|5CATlYQ^jzdPQP%lsBaqBws z7my@R3%n&!qR3;muT~d8KafzjBC-O7aUC02j(F6Bs`5x(T!sN1;fti0uu7=qO$LE> zUl;4^^&M}7rrA>v&Tq1{J37jAm6Fe;(Px_IMFLi7BrrPBK;ow72$_J<+7m;;*U=%Y zOz+>#a?yt_?s+7fCfP#dHpj2vHUE*g8?9_$ZJ0Ig`hU6N36u{U3HZG_HmEJM(Y5d)OVA@5==jek6DJr7&4#q-S>Fr zXeRWqH^Bxlq1sJMR0D>E%#H_1Vq<^yqbvH$r0zd&Z=p?7UR;)5&LyZ{8|}dRL>H(x z>ko$>W)&wXklksrRlW=;`qp`-^EDgUlJ}vUaBMun0qVaMyLa<=y_nXWaQ_(+rWkTI zNuc5=FjTtN030y&z8GKSr-L_~!5_9j2ee*vTpt(R39$H9o&Aw=Jf&|u zVDr49f;KEyyvf?fPD@VsM{L8Rz|C7wnM=m=aPZ}Z&#=mr@_70{?EYgUIp-^9uVt>> z-YPz<+a{1fyIx5YaT*b{M78RFXzxVAhs#(#Pt@Z?S%w8vUS40SK^47%~WPg2SsDLxtO)umgVD%B{u2OLCn~JyrVQ`pb ztPwI^G$JbsaeKh$H`Z;iCCc@Oa{t#+4jO4^m(}4oRc-`zEq@jhor+%?u~+SzL}bgH z()-tH{Po+ZwoI@82hqcRG6c*||Lc_A@{GSA(n$HDx+QSkq_U%O3Aq1Aa-G-89{_s}rW_q8-6_;h->Y$^m z|M}8K^?NOUDk-Y2Vif{5IcA?Wzb6q8S5*Cv#InYXT6(}L+-t2!?&7ixISC2;kA(9- zk^{ru2O53ip0(FU0nm!qp$bcto#x+oSYHL8qnbs|YVZ*9sFxQzf}V zNO;TPXlt$VN~8NewiAv;TTMcHX~Rs3?S9^$wIHc_wy2Uo$1+XMY_;J_TeHD;?u>7j zao>+DVB$%%9RHDsSy8+3$FE~DKmzI@slspX`~TW;a>;zrqk8}6uSP|oh4JmOyOnq7 zGyxXcy7e}b9iKR#4&I#OGP^=1vm$p&?A8w)(pl1HuQxtIpT4JEn5WGQwJdum)=u<_ zp21d^<0Y_A=EO)$dETnOAG3S}uZrD5Ge?b*^)5$lFwtZYvzxU;dmn2VtYj2r(ApAP z5SwWQGCLMbjtDoBxquG~0%9y>#y{0-iRx#OWd|X>&XUCaqP7xjlWG?t67s3LyBOr# z^y=fwd=5ZMQPbtFE=ONR-w!eTO68p!}z`2+#FEz z)c@>^60+Wm+;404)@>40!;$<)Vn<45nG<@l5Bg=bn6c44%01)C-DK{v#H;%i!mI+E zT5Ou}KXXcYOX;JaP5sSyk(EPSt=Y)yerSRZp4Ro^TAz5Uw%_~+vPa`E^(f!8b@^wm zl8F1Dx-3kU{JM4B->DV86`}9#;u^nXrdX4!!y>I^-9mw^O5#|cVdYIZA1ZzsYt!jB z$28Z}!bB(a$K2d_#3{YHzluzke~C`}_(umoX_1fzLB_!DIV*97^fKql&a}ufNBO|% zonU(^YJByL>Aj0jlG^(}CyAq*j??+rBThuLH$c1K3>l2Ew zlSO{V)6yNCcIT}q+4jrZErcm!GmBUG)*5P_%09i|9^dDdj@O09OY$*0Pj9EirX^X) z^7?H4+_N;+91`Np+>(MWV(JFLojX^cc*2zX8$IRLdoXtgdc|)|qX*B~{Aj zq~)5~VJj^A+W&**i6_88X0Kr@T#d(_gG&zixrwOyp3s0TWQFw}9=6Nut-;Sq*?QAR z+-ht{T-x`}?tR>q<%f3F+EJDFp>Tf{eE-$9h{*|xE}U8qrHPb9IEnSLq)ykryw(57 zV(bfnADpy!EcmR}P`dlOiJj*V$K%l8FNwvzDgjmRW$!SUuhu1(G1CAX+z1={^c?dt8VpBvTwBQ*D$;-8xDw3#2yFP7ryrMX^aJ zMWNRAn4FTOmUZ9U?>aId3%%J`>fBCi=XZU*NfgRn@WgC6=`L@ShJASaRH^t}(F3Ot zGGZwnZU$ut0!}jPn()1#{3=jlXWc=t$h#zMu;pqvii-r=Ll=z9&|gfKSMN?3>DJdp z&G6xPos1v337as}{r!_dzRhct9VCLy@!A01S_NkqM3!&0Rr^^~@OClH)4yVM3{$x> zpf5;@P`FPU*7gGO;9zc!ZdMJsZKf%a~{fPExPjtMW}l;nJSqz68_-cfr+a zt{Z-=tVJrH2z>o@qru9JPHCHYoi25TvwVNxk(eEEK1%V>WO*oWIYxvxna8Ky=n9j& z_5yutOwq#<Zt?^`ZH#6T`Lb1UKAoQ==Mk+g{^mHFTPc9HSG zB==WWIu$2{u2odM2P#R-k&d-Z5dZ1@{E-9;%E)1pwc6<_iWwrW8vYoYx!y{>VVNo# z7G!a^wXLA}NQC+6#AxVxzQSQLt?aR)Km4_Y)6SAkjMC@C-Zv%nBwQp47`OzQb9b>O zc2#yWPVI@wM(Rf!qRmJLaJB2>J*00R%k;+7RenAh)obP#O7Mer(YyhZ4(mY+HghA?I8=uL@CQ{yP*#rm$dHf4FAAwYzgw@cJfipVj zgZRdMr^;&CV}%dPxu`)UJoTqD3xzUYxKp<`dBR2@FTo%5*%Ql9lHA*clW}fy z6TU!>&FzS;^|spB6sSmsi(o~@nOT4_egdY0%8&Gd^M2uQiJSba)T*wVwSd-%4Wipn zWLKT6vtsz=c<`x-_wAXO@38Vm{I9NFi9v5hw06$NYR5TWucp1yny6`?*Ol ziJUX*&e(*zm}S_npob{SM7mfJx6cZ8x_Nc)B2rWDfF9#*;rl{}jvyT4;I>XGUP~}| z1kp|i=*4z@9oLrJZzR49;79?29YEi+-UzRT@IeIy|A@F&d@*LZOJzU;?ac|KcU&}B zbpAa3kSxJ)O0tJyG-{K6y^fy{$Af$47{3{qN(qs@m+!86IZtGXP6vWZah?4bPA_N{EFQ`fSN^X_j( zjb|3L3Ssl70v-xB=`1%OCz+TbqkIXBu!Do;zFg$_bF-DWwRfcI{&R z$iMzp=r1W?44Y`z+9LmZbbZ&~rl>T`RdF1Z0k0oZy@|iDE#UdxA)YtdN_wO=0)mH% z`kmuLM9>T67UlEIwM7)2s%Bxd4iL{ccg z_Z;ql+U*q%)IyQL!ncIRejU6AWxcx+#+Zy-+fi!Y)p#xYVd<=5!@q(<+jLQb?fi3G zMHbA7FtRw>h^86pTzx#<)mDmJW3ZsIThvV%`cRc+vZ~)MjQMWnN3xaT&ky_sUSWN$O*VUt z&v9)NF%<`G6>*9`xvVI~d~1AcHln|zNBe{3`<=k*mVdWaZX@1*u>H&e{3f;>`+A7=UK!~#&byWUJo5=`gdYSp zEUtSB(g9Wg6|Anj!_1{i?D26QZU4{=7S(A9yKC<{zvFKKUN7`iIc5N=969l(}7|4M-!({t3T;_kK^0>0sf_ zT z0aC*5@Z}|QvPD#}%PyE%J>$;v*6C;}4SAz{!G|9+P9MWk=|NMdV^>TOW?!)!ajR$uN&~h8*dVJ3h+K<4!oP6E3M}4inDh_rKZ+sjZ9M7xqKMr@>os{y~P*YG7MvC(UA?8R-4llv?OrOI2O8X-ndeGUWKS{#1!*1(-92EY+P9JrCC2 zRG$&vu5GIE;6UlH_ehG)IajGMy-^CA1~|F2Q1a3yT<;cd$P=V)3J8;czeXLji?Fvi z=I$e*`+6*cCI^OLHq$O%Q_cYTXZ>sP>G_l}*3T9TtCL_(Dwg0X!vW;dvk>TURdEcHZl)?q}-9Log`HaZG2cT}Y zWnpu_Ri;a&Z=TEqU0TUILv+Cm8+W{ds`u0E0&2X2mji(5V_U1c#ti`B5N$|@F|i^0 z-bTy^hHgej3bVOA!1D3GP=7F-VX7>>7eyg73&ovtOD6|LMBlj-w!eBRM4{GA9noA;#bDCGWX$3b)Unit46}Pt%CGHO%@KO(#ndb zR-k5gc?5V(<8?wD9-CUkkgyQ%U^nqpI?4Un zlox5xt1W9;-Yjel6CYwaJ6=8?+A-C;UI^CWLsT5tFTl2q9=hK+S92PX#UvI|xM7u` z-VU!G>Qz$UL3L9rJntt^ssS5ih&eJv`8f|qCUUW?f4w7zqd^r^2aF|?W||GRmeHn| zQ|E*~xwNJ`Ja<0Q%ml0=pAJ<$@`gVvg}tR(^2$Ivim&$G-gG(*21VvP^f)aseebr-p{I<)CVo`fqBaUUZ}l zH|IXDV$Q@yC9Vx?5Kzfx(dpQpGGU-e$QR>H#gaynGpkP_gNSPBALv;vNZ%UW1sb^) z%|%%scha486>&uSAP3)UCiqkqr;-Q0GUo2GiL!I)hY~jsAcQ)kH>Q1`-teVElZn8W|!fdP}rAJgc&4>W=oA)(&ijR648|qW;YVi}4=u zfzPCVx#_f8&KX^o0Z29vK8Ca&_egpF(9qNLdw8jN?h6EZuPC{^-g7ycZmvY$x<)Uy z0yVIp@_jna^xeC1U)_TtmVy^%>!R%^VMOO$XxRCEtIe_S@;2_ESLnzua-IK73y13# z4I>K8cMq+c;vS57{a$@4e)#F{suDSSW=Jgx-UW|niav5d8Dn24<7guZyr%zF`Pew8 z;=5}!BO6mTMl%0C31wGVXqGixx_rrq@<+rykNAlHcHE{ylHiGYjbVk ztd|xiK^Vz>(u8VDL@q=Y6_z-v$Q=9RcnjP#U2o`BSaAQ4M!i%`k7pn^BzukDb`|N{ zShl$f%2KcOJsK=Ok>{IPGBjroM%HMX9xg%2>5X|0-Lu^ z9c=R!LJ7Q`m3Q{Y>USNx*=F1>q^8bI=w1S)8E8X zmV*vUGt&jKx_|uss?N5WE?2D=(ped$9944MSncnjEuPZiSWY_?%#2BE8ovvm#*P_G zbp}t^m#lK5$0@s;#g3qz^*&JSQ`$aZaxH0`4)#ZdnQ83mv*{J8pe^Y$nU$S_=-AtI zHAw~mUm;wsc-9RyM9}#0Q_tsJeskF${1p^81osYWu6XQxzaQP}N~wIO1JPLTb!Rf3 zliXwtD7e1nYydD+YZFZkPpX5VvqNjxXNW##B@=9B+0lgw4|UR&T|EmSTQ>T}uU?re|oj~=}e%l4_N zhcDhc*BfUKJvLCAwp+so%AV*&U?y%9zxOCr0&Bq012j#zkgOUzSjyV!AsrbKRQ)8BO$0y8h>8X{Wmluf;GW29?g`ElyMvJ30$zIw+k(Rn$(f^)1G= zZs3F7*phuZQpt38HP5Wzyt~F?y#8VClLHCs24vg#S0&(xKDh&MemB9uy6>{EVwqi7 zQk2}p?D3PPyY9o<-MCoT55Jrv2R(#%$XJ1|=UJesl7-<7^s_hm!>@0)X*#tcLw}F) z*WSm3)tid{NI+w8mZYeP=uLI0Qo$Vk9;R|1I~Vs6uRp5J29lmnA|BP%#MYDWHl{6e z-Qj$R7srkeQALfLV($$@t_m45tmk9%H;hGYU6_V!Nhg}lg^-syA951 z?brykTQ}QS3@_BIRZAV2oYHG2pBHGC{KbWp_thU(AVVg|`-#i>jI8#_+my_wuzpM4TS1?W3qM{|G7J7(}4-?Oh zxOGp(jLRua&s8tz1k|ivQGQ-K_6`l}aieo$OK20GhTCa2#_$CpvbLLOXfPqf!~nA3 z$2cWOhaSWYw(#d#5Ef}EJHs?UGdCsmBe`VnQa3qDy{y9UJr60?I-TjL!D!bdZL~B9 zvi@Xx!?=RgR9^6XPdZ80lN{)3G9ni7p{I#Oa!kR?Dr4)o# zl=hrpMNPo2{tL*WDkh%zh27XBFZHyu=&L}%3L1M45)uT5tFqN)Sw8bJ>NMpnwE6n~ znh^@>hY_ngNRz4zP+Oj;>@USu+}I){RyBqa9zMDmM!LQyr;*>SRY^CJz7AZei50&% zdq~3Zyt968woRWgHpd_5RB;+ueNt&GXB?mrN#;IHF6ShnBHpmj%K7m zteO9gDnIMkONyUkS0|*JypX4)9JgTGi`7HFs=ikkc%=CMMU4+rfzbS zXux0p8LrV0L>$<*nmFPgE z0Ck}y)_rDbSEJDy9X{tc?(Y+>8_yiQ**Lp9cix(?n!G=8q@k4|Y_&Z+So{n}3X@;H zo9Qi~H(mPK3K13YCGivb_($yDdn8D^?MDYyG&m3MjdqgazxcfjPW1?F!^Fr*YPJa%M6%xdp} z(0G@4mzpKMaWpMR^n9d}{#MzAJ9<9v`)})8k2vgeGiT|CYJhtU6`I0g;VLIQ&VxP4jhacomSk!KSSPsNiowR zUog7$DYGn3mi1zh1~dsj5AK$40)JV{HoY;~{DcR0zHpl6ZDoUO`NIyCH-Qme_{V5S z$J!#mFj{3?>!NTx_KDi@NyuV}oq#gJ>F`U>y*b)uv97jP;d#*)S~&6`Rbnaxga-fi zKawwQmVf7jU?!TSz)GvPo~15#tb01!DN}egV|eLglpjt|?x?yw-@NL?dZ9o_*Tzj+ z!R}o8sh!HC)twD*HB*I{e3_2KITBg@J*&?lU0Fo^;R5rfN(7o+ebVLpU+q=?L@k5! z=%2-4l#W!UNnBO<2X*Ba0(q4$c>yEf&3`9ZEl;fM^~4=5cvsfUgkSh7;lAL(Lzdjx z_Ec|QfNE;yf-%#CFU6=7sqw7&!wh#KhMYWgk9NcVz+!M&wY1DMZEDn}_NJ;RerT=O)ZViQLUh=xcI~RwTB$vo z*t??k-kXpZ5hVFPd0ysCK6mci_jR4;bsop}*nZwlLk2QTITe-b1=9{kyxL%OJ8RjJ z%QcBSb>`o7O*|JTwGEFE2Z~c~8RZ%yI!ynum_fN}Ka4=M+bX+hUG;;|Ksoxx2 zt#Mjs^M8`Ul`idgzc-dDsUT^zohaSED8OUIC|U9REzy@}ynf6&Hsk6>1{QsYjf@m2 z0{hUr#1m+LJ`-9pw$FNAWkqSoLs~@hb30+j`Hz)noXuP5H2$SbZ$4A)iK9Mc*(dh) zxPRR;2kNUCqRna%cyd$=OvV^(GjPmaxIUpPf z2T-VyfqkuyzUqeL&b_g4R9Zk5EKxS1BZWtr}^ zlkBa8UjLfjn?VV|vya=a20w09hu?jyw&luZaD1H3Y^K41X6DKs6LhGeUL6igX*PvE zcMtxFH1wm4bTy`=`Y?1;&u1g!ZC16F&S2)XDGbXuAD) zXPATj=*u#S8ib@Hnw8JKC0F%Rex6M6+~iD#(Q5b6kV@M=nT7;W8E}fN?d^>3DLw9| zU2D4L_W7 zi$F(&u#0r44HVO82ZAXiS0xXxDFPn{48HhFpG%gP*da?j!nLmqaIC}s1yex9F^Vcr z`&zH(Jl)JZ<+AcYV%x!(zQS?~pm&>b5^h6hu#lzdsG!c>wwAo(JL zS_rP8Phxy|N>@qcBSnCV9|rGuU3$DI2}EdXA|T{~FsAsFKspEW^o{F?-pNp~qxbOt znO4~KU5s_dXg5ZFcu+;58*=t}rp7SBru5!ulNHC=G^+)DR?05(p2|b(l+W-r{ zPwr%B|@qvw|3B0&kVH`YX5^Zz8w`g2Ebg*#ob(?9ZFsy#@ZmkP| zK0;M~a>W>yV2rz=jHw2(ckjw`-gUI4@4sym_U5jV`pEWpAp3%a$zjRDBu9RKx~Ws( zd7*}L=Us$^okHXiOM>QR1XqTms+@B4K`%U+3?Gncb#&+{|p`cXC|n^w6uV6>pzZ*+sOC~2jSVPIeast zv0z*f_SwXW7dT8ki|zWSwq;g>mc%=1nC?#1$%j6NMG|}FDtOPvY*BSyoqw_JT19`K ze--f1$eSb~Dd?|(25nn#o#hLo|9&aaI+F3XmWR>wrP8bdkB3+|KVsL#?=10iJ)AyU zF?&IAFWAv%6T+XC2`C@;$Zq_T78C$sfjpm!NaC4)pYKlGh<4j)H_~o=EphfYXezZa zv2dKfHmz|(d_f6af`M#6*g*Oi+v&68uMWcDh>JT# zEH|#O7DRi;4tfBK!HY-ZS!TSime2bl@SU~eIU&9RFa3o=1>??1^@Ii`l4u$UaFGf- zx)TJ}|4x0|hK)VsE(vRE|z<Fn*P zU2mHRub+9SDQtFI#uH>aC8~RS*clPdnP;-K0;d(R6?zh}G@S>DT6{xynrfGOuUjz$A1GgwkYAH}9%N>5fpo?P}255fTg< zi3ET1_E{df-K!7Z;(Tll=R;hLU*|}Q{8x3cqB}jl!SS&ym}l~_@FBb7CO(ge*HBtW z;cZ=?JKXETb)Cm{Ur1hnZv(UtwjQvrX8boC%iJdqD$X-G()Ip1pm03Kw4NBP$Ogspff zUVYRPn>q}xzc~^x4uhr!9@8uoqjgbs2n*DEWno|GyvW${nh?v05+>GKL~fk{mVLe$ussox~A8d0Ct-O)>|qoJ|s8i|nQ^8fu5d6i};bQ0tLFW(RMb4$DM_K>FU0!7%aO;{F zSXR6Kjf@lV{M-*crCz9gH2OA+E7CP#m!P2Kz)*tXu`d77U$qY{{;HNt$g( z_LtP0X62WC!l@o(WXaXB)hshk+X9^xt@5Ya}uE<@~^@eW}*i|i& zkRC|G#wC|$EU1_H>i6pMsGQYmFAcUaSJIDsVRL&#KtHkv+`|g?pszRz0XM`Zxeh=p zE{<&3LzG45D08wM`SDgiuCKKt~B1WF>x_#dQGwIe48}`rzKN>VuVeI)R(= zsps#}(Y~IcvMYq4^y?@`e@M448PE>@xHe#{AIjd9-e)o`=Frj9sd3SxOwA$qlzld8 z_@GZnognR%WmW4%6;p<@nPtg>R||sE&%aU9QjLrijU z6}+$Ov}GjTC9#~VEe>sKCElrHxu>A~?(ezV+KhdFE(_iG%S_=m-v?|iQVwx*CnAo}zf)bqfYnHa=2 zIj7e4z~4Mmwb}*p6c56R2?vnZJ)9VVbdJFRle$?gqfar|peHx{oex@8I^Leh<~YIJ zmgI;JyM2VuzEuS#8&mG4n}#WkK0QO{z=JMM!ScYf%@$6U(gJO5c0Ql#(i`vV587L@ zg-G~7hHhQ<&b5GsA6i<^$A>>lGx_?^NRb{f z;I#%E6gG41DiHdd+VzlQ1&{@t-?B{(bh~vJ*{iObdq{{@P*K1IC`QWs`w}BmHU*94 zaE$9V`rH9fwyDrD*ojy?CA+MmFG%80s0&ky+YWH#P3b@PzuKx1J(jY^XIpWM+oW6! z4B45|Qavz`|4{I_nkLxEn1bA3UKYLQ*S(vZxBy8EIE2xF9bsRJsg)H-WhH=f z`g%iod9OY)dJRkCMC6CYd0HsPiaWY#CjsNd%Qa^CqE7V zyc>{Aq8)qXM6yQkBY5BaaW%mes=0`Bx`!Czazl|DZV-`x$DiN+gIFrHS!8biwb{wK zTQHOFmPf5GPQRXqk^-wYIT7XCi?uaf!u%;G4$T*PYVKW4(>ppcqOX{c8VgIpp#dwt z%lq$%c=jMMqrVwPh0F*meyE^n!_i*3lDMg7m^jbn$q=htfpK5B+G@Ej!t48tabya? z0Ck&xXF^b9)u-g%y#yX_r4Dh%|Ju4sTy@!fkBt5Jt!sp#NaRQ3pSkD4zOK7N_l89L zK1#|vYl%?<*B8mq>9N~G*eBwICKw_r>Se6i3~g(g_Tq7BnEvpWUb~aA#m}ebyoCC} z7KxIS0!61ArWl=yi&gUwE?#~*K=AK%ac=yvvTmE{()hXN4q$8vWM%DCF*vna_Kb!J zqTg(uS8#ZCMQNEfP!svu;&QtvHrAuon_vTPpv=en1KE-HFxp#-K35-->E^c0^_0xT z{-%uQG^T`FGVuoPsSfcrAY5*4b5-vtC`AjM)>;^OC|OFlYYJ=fLMreVEvFOf>TE+C z7f_F;EFs!zu!2me&+2OPvS!ea_A!FYLj^C`hw?`DV{x`W26`XvFxJ{k`;4_i_jF<8+-mnhnl=0%wZw zl`J4;)Q`1)6*)a&1JQ@E0?M;3H~g*EzlnES-D#yhW|LyuA+Rn_BrBdARf#p4RdZXl z+xCgw@{O4fOIo^ueQa7N?6LsnNXpEkMF$#x>zP~>^huxps9JttMB6BE_uoVc>;2r0 zS;1EFX0IhD#U@DOGv`>(L6>CYyd}TJKH#{bp$EfcorHbb!IaFzwn8Jrzo43IJ zjyk4oJq5<9?0EMQpwP~JCn`8-gU0$1(#)u=<}+m_Dd)Sn@BSaJu*Ylo;XctHpdmC`3>~O4fS0w? zJ$dz&rh;0fe&7Yj;cN2Lif&ZUvnN+Lun;m^K+!ncjIDxZ#wGP6k39ju*#Bn-UUpfW zu*7yvnUfR3#G^bErEq(VIJ8Z6tZ~0>zmrL@UXfF8b&)!}?kN5*QU>*=<2*l5o9%O( z0_RVz*@NcJZ!UC;-&?teNDaCFM|84?^Gu}8P(Ob9YdscTH4r{}j8dKS^nNq^(`)CZ z?S7F=&oP{Ts><5Eqx2|Pk_ifLvG5CdFmld&+`0q>#BRAJi&%L#PjlJI%8QWU&^kDe zx^k4&P)L9!*$ibkpQ}+ql@_<3mj5u__d5@8H<9=&Lx6z_A%kXmy?(w&Seya#u}ix;?#CoZgfvDkomA3B zC^ux>5Zm|>fS~929}(gA(=ufGq58CdNe1;XEvL6U(bx9s^|2}-i#DCl{AxXKM##mf0q?iec4l)jY!iHcFqf?Fe?*jZ-V)&m ztEGCNSYZiItLn?U>fVT0b3KYxAodH;jT&mhRyQHf`Xea%<-iUo>E4M@fPL=J>1wJA z&+eq=3aty6ysb^S|JN?XTInwc47YYN_sMTP&g4_F)uL($xq z9@(remy~3nvwE7e*%L}00gd(5SSXmU)~Eb>B-TJxQyVkfa|6cNR3{==KeoVTyjHv) zTLQN6D z7mCw{h9^W^ZHd)}6*)?2v#1$vm=3W8goLM}QcmAAiz<%l#Ym`2kQbk3TnN()!74mT zW%IlkWKcYRLF&Hy-TZI+wI|t}l6Fv+IM1xsMQS5{RSoR0L0C$ac5@J*m7w$Ii$0GU z-qpBg2y*(cRWEh|`j%Ah==nzRg_8fEmJ)RqqUWxo$VRVSO73lI(L!iC@u$F|cD3mQ zxO>*O&5u2o=wr*Uq%)Z^+d!vE63CWkk-OiTJr~;tPJ?C}G2Tx4Hx|@$%;fV-r2F4F zw=LG}FMSDKx>_@nUIDP63bNMRWGEmxM^40LJLbNGrz)&7CLy$7}&%v4i@yed4v4K3kCKdw-Z9Wpu;gXIrxci3G62sCz9a~&cG8O_(Lu1`@L~QLWqh+Tv zEzg%VuOZX*SQgEy7X&ZIf6MKo>a|H!HtI4FWd7cWUfvFJzdBc33)=CC^nt25h7u&z2rO;TY#?_QimhH53fS`U%n)S-2b|r^<`Mm zmEqa!Bt&wYSot<~z1+;#TvQ`*$B_&zTNeZX9?3adUv;umB2}0!r!9I5|KQ;NUVCgMqSU3MZWWB)7**_ zUdhj60?*Xp1l_(&OCt=WTCXxYR-ZJyAW%h35w)hK?J{Hii~9b`~TjmL3f~(_RGauvlLO&(Nb4ryYWL3q^)trirR( z1hXzS-8!CuC7(fordA%mlVqLjzCEUX$U`K@balpYEK4BaNM>+zeYkU**d7b8$h`L^ z&M}HQEJ_nZwynYp3ab5;rPt-?cyC1tOd(9U!G>Fj3cUs78_0$gBKTR0!EX3x$l{bH z^nBN|2I=2a!>Pd#0<160A1swNOyRkZL$-4J>dSKb6meg_gr1pJ<>nZMS@l{;`~ zbJ2mr+0rEGqYM!H1*jq=KMdGyNn5$n+i*UyT2D3`ERiXDSgJBi6qq)z6o2~*T}N*| z`KWndH-%CA;$d#LAieZ6PQSOy)4NPsWT8HOn%)=#fiCyWV=$9@nMmuq-+64YX}n2T z)b_81{gj;-`+40nK3#?tKJBfUdOl8@tV}DF3U5_0W*kvJxA(e-Oqs&xBz^%j=z4oD z7tLe?5_ zAnpP+H))<6Vwa?zr%n{}({53+lLj(7X&#_~= zvWwOpz*9|&NrS2RwwB=^Zq}v5tj5xE?$f=LGzkDyQ+~AF-z-i6H(n2@edz#yQP$a5 zRqV5vPS6zj-8(pXv>L!+)h4=Q!E)fU(#ZXZk#EfDM5SqWyk;?@ET?N0J`Y|5qmPh? z;+R@IBc@ZMIcCuLoDppjo)&q89Scq)^PU=N4KX35lkffsRf!x4Ba0Jp#l+t|^W3D{ zTJjuj`o`|cElH3JVmrH2-7=NXnVvan|Xs-S%O|!V)s;_$f8GQSvh#0AE*ynfuX`q2GwV<$JaSbsv`+34 zbm3)lrGVf3ja(_>@5Dsky#-!ukxIz7%mg8{+^z-nt9-19n^XLw9;!T4W^;gjr&-Ft zKPp9Dpsg{Q5D+UyBGkJUbBkAK8gc`NQm62~V}tt($*Tf2?tcFJ7JHNbh@5jwta2;a z*Q}{o?wZcJOXB8!d0Lr6>1n!kx`8_Hemmq+>L6ot@O@NJV}TPJ!|PJkgQeBZ;H6bU zbob_Z*)`jS|C3+4OcsHlwO{l>1=c1-!h$>2RL# zP(uWYCcha(b@v)Lv2B7X$EW^Uu76{1C9o@IH6Ae{rn}+Y?U>{$&vA7vF|sp10c(Q9 zG$lgr>1=&u(z^@wF1dIRfNJ_q`o4g@Y*%N5oy-!OI#>XqO&qQ`7#-!n<^-{>f?99P zae$tcA*#=CLX|1DfBL*b!GiyAdOtF@+X#|>LEV=`x#0*quNi4+@IFqO>qm}}yA?l+ zQJ>HvAA%fO&^fi}kY%J@Lblv3mpGRvS0(~K7F|6E-K*=^4G=2$!Z_gx?_W8N8OT%< zQ{vS90#W$&7+vvuU*;eBov?qwhq~UrS9h*4J~9OoZcfWX|2gJu5B-ExM$#Tm9z{}g zrr1lMztga*za)?lPqrm*3uUaz|J`cGSPN1fOJqN=c`w_QXDPL};&fqG0Y(yzs%iA< zBIj?ZwzT)2=lDj1^IKE>W_o+;Dn)yihuxiNTD(MPMaTwTSPlQ9TZ!le6pzA6d@+26 zJxZyudCE|^c~MJIfVI#rlY_Y^7wKRU`DVF;)BBlwReqWgHem@&kv?}3L;J>kqcS!M zq7GMA+6a}_zSM=~Aw?541_HrcD_qpzNW;Z@Yu2?^xzu9*^9sxe{&ncgeXAYkd zF0qzWv=eA~Y%b(yU03^pA9u57>$%Aq&aCXY0hr!uKX?Aute{bU|6_{Z_=6LO!rAol zAERa@dks}az{DRKiUfQUayjl9Aojqh(?B>!>sAVrGoVUaP0ihJlDV9em(|qJVfPCs z8QI-IgQe#mf(o^AP#_B@vI69jY{vY!`RiT>{I|_~JgDt&_npADhMff72@wWPLCW;U z9j>2!*@OE+guh2+{Bz>V+$d7T$YA||tN;{EAHqPzt~c+?+{K{Y1EWY36FLk^|L$e;!nvN%_? zfLFgrJv;;ueR1@!*wOCII{u#>G1x;!Uk%+csg(+eHh+(T#B85Kc#fVTvODc2;&n#I zCSC@tSrVt+?b#_7#G77sO*P&%!$#JJ@t#9l;IJi2wc@L5&H_Fjshl7Re#aj8Bf78I-5SrOWW}}b$5TpjaZiEN^(UlGRVi{DuY}=u_uVE$jR1>#EKBT8 z2r(0f+7U@T6SVw9^(~o~#JOpykM4%mFEZ>BxvADjZCGCU|0Ym55JPL&fiOn@>+h#9 z$k8BM(@~#Y*Ifv~691_OKR>1JS?Z~nol2U{spfs9`xGzRc5=6dYtJoBJcR%aAA-SZPp4CQ$TK5JOB z;vm?^wQ0KcM_Rkc@4=dg^Cz5hZL+An;!icZaof8bSw@De%Qsqt5E_zM|LGe*rJ;6rbu%y>Tmo!=aBapnma?R6(J>UXEu`6gH>d)r(?>8LbfnC zcf{fu8Lg7j7BDK!fPC%2Ab~Et6%$P#Y7Bn_BYB_qVo~*TFj)5#tFW1xfR}S3XqxzY zIk;x?{_~RU(7!wgJmh|9j1s*8FnVTg9=-kGUvHfnBy5FJv96ZuAFTTIC1B68qE)RP zytuu69R+h2kCBDZb=|X+c9or^}n85K|iQYY+AWP<{GYsc%t{7hb(|tnCd`F+|kSj2~|Y90j4?`sc68csKninqQgj zoRh&A-|XN?;D_g{q^DM!1R4Au(RYh@UtDm4pb@i^yfa$`x}%hqch zUw0_nFmo^jn9vl5;o<38B!2uVnqX$TR-I`FTmw`{D~^w9O9`F@HUF0<(9lhiZQuWu z`l-UO&w3M%l^-DitdUwA(Az778KeG3RK|X=ap2(Ck&!x5O_)zruG%5>;mZ^+G1{mG zPGU^L*%)#^#W`WnHInD^J9v(3V8%HjSj{CXCK4OENd-1AlG4)kUNRw|nzi5A?f;Kx zccgl)ex}?rr+~pvXGPtnjWv|b@8O^l;0i&wuLrnaytr~g$U^n|EzkPEEjT&TWufUF8)uxraCF?1cNR_J>%8C!`bV5^>*I7N7l@p9n<>*Jc#HW|j*)0J zIt@yv9(N{(w)-jmc1QfNxd_T-q0(>vqTA=H8!ct7Yn4+k=m$jdX|r6K18y6G4GpEQ z3{Qm(rOy>Epe{aTS87?#OdrvId6%+wzBInLwRiYI-&l}`W1$|#W>@6|*nbBr=2~GB zYNot5Sl$WBS0~jukFIS)R^l8Nd*m4ZP(S=T?s!>U%g;x6rF94Y*qc#_NfV>~0xK>k5pUFk;= zglH|7l^zAF{6)ju#6Jq&$G>ew~Im%q#G<4m1nYjx=+hqiQcL2-M=$TCmY*k9uJSy?^LzC zNuhEznP$NRV9=-KIvX zin=dRuBrW#p^2GaV>V=l^$QD!F6$(|l4N25^MsG^jahqtTANa^)XU}7|3b`V;O3gk zNF7kBGbow0GvnB4n=oN(M}Lre$F08;PGBf9-7dPRMmoK;eG;YBV_mQ!Ku!h(^nop{9$hT7< z<4xg0%Wd2F{9(~wtcy(jKM($!h!y6kDS55WQl_56ONg`5ypzEaAvt^1?t#K_Ib9Yt z!0MjJ7qFQo_hq|=M8g@;AQgdMc~X?`UXuoBSf*=A(7Dcit9(7hL+60eZvbNM}W&2&zIOa~F>+iEW(>vZc!rBJFE!*Gs zlE2d)C~kgw0!KOyhOM8hZWMpAeS)&ov*EH%*otMO)NfASuibijjJ-xZ9#q7K-G+ExYJTNB>r8$El{lIp?K?)kYCLqKLpl zJ%%D>O{N+HhI4}tH^%-U2D(%3wiAlFYk6yE3qvwx4oOdnB z&^i^A1*C^r+%^Z1AanX|Q{%7hCNKD6a(u4onWoQ%dmb!o8NHNzGs46xX#eH`WPmdk zs*K^;>Ri7EvYiyi%GDm$-?TDXCW(@w22C33jlYie=m1^~FDRLBY@eQ(*NZU?!WUeZFjs1OZPXl^?s69>;o7*X>Ck`R3}{i!i3}~2KE}1K^q_b8oTjY z>ag+2Xx}Mf!o^049jiTVP9*~#ooOPqayOIJ3}(C3x2V8xQ`WVoYGurg!lnEJ6ybg; zq-Mj)gZAB7BtfC62#Gsc%r>Ec{S%1%JkYRoG>yxHznwm7iuKI*)*X6hG#A47jdguO zXt;i(^;+;Y`jQO4+B|xyrOd3^o zakcf|+`7tjUiUyWcQtXEN!HM}$ViMPvT?Jr4@?H^+&)9dl&U3Mo{1W-MX(OM&!gAxvB`C!r*+LjJE07n;=!{Yc?C7>wknuHdVh@4dOsChSsJMY7g20@ZETydlj{;wT<#8Z|20&6IqbXm-5{xZkY`zz+7fo!dQyf4?jY zf&^6!T(XMJ3a51h1C~aQPUm-hbQ&8JTmX7c4Za+{V#tweOboH)8=E+DS?)YswxP*| zxuvp#NHA&D2WtWtnhhO1`-o`4aqS_+FWK@6(Io>_xLCf(rVpJU*GQhA=_|dDOv3H! zdp0lSmJp6aoL4o0C*KH;`07KuC08^UWp^@!$}z_%Z5>zZYuP9b=Pkee?J(Cm2S^h^ zxBB^vi3DnOBOG*pJFoO;w2m%aLN)u1zVFHND_?Gp?cjs1bL0)&|ZJr&0F0+TQ1+?QTeQ-5}y< z_ByJ0kt?L(<>5n-;NN`RY`f|cDnt*UovIut;RY*ry~@KX)f3CYOWCO`d7>L3UO%aM zpz(ClTqI=EkPZ|4pdF=|y7=8NPcP9u-ObpY)YoZg^x=*am#xBkD?rS91CF^5d8wsX zeE|n#;LnmrBQR4)W#E1)875hJ1qdFbQ-1pbH+~W#!n^%w6&xITI9jwYp)}co&cnKJ zn3=fCk!0~%_~{&eQh=B1Z55Mi)L3;PiT4L3I|=hrd9IY)Yq&f_33>XH@F4QTjIY1F zT79AOtketbIQ_3dk>%V`#z^^}%ED}+;GJUU2iG8o)wgEH(7`^|-gSXzlLb+sC!vRM zhJN*cGqR;y!AHJ*3cG3@1~?Qd-Wq&^17;XhesH_d-kctx#Cq;nEU8K*_weIV2y&o$u)!|%f$xL6F6G+>;(c11l zk$M*aeb?Jbq_!kucfUa~;VaiA!-!CpRTov}yZ_DLYhbSHu}f-w#W5`WAG-bk7yoc4r`%u7NW9KNLx4}t8ZN17lmegA#E0n(RA6Eha&(^+Qv`Vix3#wjx10V} z^}7oZv|X9s-9Ld5tc^*8w@MZhqe2g8ZkrNwnmPSoD2YB=-AIt{kx(dF`ELGv`Hw~S zdxEH&-#fAcQ)ReFs3hrAO);>LivJPW>v)U)Et`M(u0Mjs++5ZolP0rde{v5d*kfF+ zP5;zL#c;MBOa|Ek=f@)ak!MC2E0ke>$NkmlpG^@)o13{j*)yYvXKBUY+`u-P7oRL1 zE8N5~$y}&p*44{pR9ah0Iqt;(dEb)B_a~D?zJk?p65O#^dZkUWRORqGPYxE8%rdaL zpM|mbX>_)H?*g;P>R#T~CbxHvXppCg9ooC5Iz!8}WBOnCwB`cvvKFLBlcEcD6a#A|;qYPl}Kv8+c z(%^_#;xB!B8F_CPuuOj^LUI88T7&qDliwK12-y1W#lxKr;Br#o4kaE0II+-jsh=_9 zzi*+-QoVRuSZq2APQz4y`{2Y@3yuP#lK&3*8WE?k@z~ch$}-g+w98bs!_T~H#sX=4 z)LiGa%Mlypvi%D#?gV^f%5dB|yUn(N>6wa3i1Jeyv6a)W)$SZ+)*YC`m z5N`%TJ((8jZyvuHQh6Hqd^-a!q93tL=)mtLv2vV=dxdWm9O5U zm`fV0XLy16pLG`YKOp&jwmblR3iOsKrdw9-DUH|pI@tX|^IAl;O#!Yz3Mc`Egw|m# zf+$C$-?wW0*Rs+`XYdAps%m}xH`tC8tAZ;mS@V>8knZ1s-pRKTO*;B%HpMH6QD27V{}y}iVmgvy4Z6C8HFzfZgxRmVE$~WaoD_S#3efi z*`qd;+u45YvvGEIZ|B{ep$kzWY70+UF1>|GSK||v-xrz;?N)g&Y)0tD*mma2*J+@G zk$gFQgspzgmrmukH^A)6%EVf@&fe>tuTo4(T5~;*XcQFw1{qb3=`S9+Dp?K@6gW>^ zP^~e_b`jM;YV`H%U@D(8U+zz`-&VJu$vM(-7ro>q3Q*)A>#=K+L#RQpZwWO2?Ln4k zT=eXZ>gsRY&lVR=N@5uJgRS* zKfuzicc^7b{OT?|_|s7c_dURUy2k?qB~5pHyno8W?w$G4*q@nR*9xA-A(Q*c^vL`z zs#b((CEw-({Pu)31;c(){Iy;0xt6f%EppO;YiGZ#JdJ)^VxE%gh{!wj zDfP@RNw%=D;hfZO9q^4_0ra%1z-x~h79WH^0;#niS7$gF+GqLZ|3RLp zOYe*+& zXr_R|J_*AO-yOlB+@p=>2bp<7NU;J^K*QeTZwngF@RMD(TE~geYMn-k+SF5uQmwkU z|L^P8z&-LH3FJi|o*h@;Me*KUIb?3~051|l&JV~jTKcqfRc2t4X-pX%%lRCTqLw#= z(e(eRjZS^En*3Vd%f~)Zh+m6Y_4P0{h2(?tHL2#sH_r^m><;YyHTF(0WIi^ldyM0`P-p$NJTXF{HWbEXR|cXNx=OwnX=sRL(aR*-mhY?z#DJ)zR^$AZsDd+I4wt|DdZDG~n`qJE<6uv29j=$t~K2MKr zQX$28X>5k2!!$HDx~BX{W2zwtM`D1IO_PnU`VQpp*>QUCirETjFKnecdmlHttImE?QS?itt(KAkqqJ+?3rn)Y04+ z#hb93g^pN3z9bE!79QUj46{!)#>{K}@&vG;OTks+A*{Lo^-E01r2~(`FDxYM&;LgR z40w5XqE-&h58^S9N-syQE>D->lQ&H@4TM?;oWZsgdiDubeF_M(9o~C1hN!A?13%$} zCZDj!Y_SP}B=a&X1%U?gK89iR*Iu&CEH%Vr3|HDc5w)ASTx9nvgGUZ0DV+dgymWqw zfO)n1bTX5O&-eaIS5WvLQ8+ULN*RQRYRyyo4o#LWz+}lx(P5`>hV@YL$nh|>#K>ny zU>0Qnud%zL&5pwoT)`c}xWr7Tu^y%Ea@qHhiQSMA!Q_s8PyR?TKW{61v(V6Kf4*Qu zOz?L47ZHeAV`%q0jg3@RcOO+Z{%P?iB*PYt*{gA!=>Eui&^}zg>9q&JlJW9mTn}g0 z*E%p8_5X-QI|8h&kL}$je4TPuK!AiB%N~C**)OKGgoG2Mc=#3PrXlko^sz7FI`Gq> z$cF_H+taPxp2bJYYTxXlT{s1KqWReZmwWhdP?)o?+6f%>AlSFF%hJx9?EB3ZBFFU# z;f78tmGD7ZE{tPXF3WEDRetQjm_KD=ihR#DM>BTB7AQ54?No-AdXRMA@4E!0su`&u zzfz-JonH>;R=^Ssg|1-jl_B;Y_IjY}{^!3E6!dWg8O=@Zk?zIQl`$|6*vB=QMfXj3 z9D$IZBwGfM6rmhCl=iPzjVj7Ip#`SD*|y4^hPUwGygSK_7owyvo2k>&H{@S!EXTc1 zNy_2~Ff1dQ$#dZ~C>{Qv8ZDGzZE%UPQu^&Z$xp9}x4wic{zeRXUj3IbU}-AK=B*Su z*>Z=`@|ciw<;AAMHF2pm9cP<_uP#2^UA}Mbum5eIE{XkN%FO0VO=NjZz}n9ZZJ+pqv~fwctc8zGGNF@K zR!yxPlRj_YX}rm>zb$^?fNg)KC5jSORe>2V#*K_tI_lNb?+6Sa6e+%4oCaMsV@aj-G$oSc3e(g_?a9| z_X}k)Y|n~{uPra$EYIyP;3)}fr80gpVDWDSCe;#&PW2ULG}Srv9Tg6;sqcBnq%brQ z*?dX!1-21~=Kxdo`EEnv<$TflFGrv1OmoME6nb+rh=zy`k&y0i!OJ7-v5Xkl20R5! z-LxJLCCbV-n0c2ont=ShWp*5ITCJ012F4kj;2;hW=)Vf0YMI#0(8^ z2=;Bceu2DGA2`Z%?6`1;_2d6-s!_FhKskz}sPqbT+?K;tSUe6xRcJ;bgbVNCQgcdX zAVQI@@yr)EKKIr;Nm%$Fk;|yO{|!O3=7}fYy`#ezNXN`?7NteB7T$i#dH1#0Xg)xv zco4d@RGH^bLaMOpS;<0H8=`0}ba*R0$vD8d64c0mJbu3xUicB6!nw9l`p=A$>nUcVEVkVSNJ z6oQ}GEfVZ+1mIb8ui5y~5GGmRR_i)4`S4sVO*L&1p{(y=;jADYXc6`&XTx3U|IUev zL4aLlj_1g|*YBQh$S*NyALrrY2{VRd7@3s;QjTDR2p|4WcuUY&bK2yhW7!~WY480Z z^l;o3ufy}1;Jw~1m@4lxk26bdM=~=QDgd2qhQ}bW!%L$715iP)zVykj^TBDM{(w)p_*?$~1<2L(=J-AE{{X^L z%8Pk%uHEV?jrY8ZYY8pc$G4et8lSCyHPF~vH~P!StAz!!AXBg{5MxA;oLSFzLRU2uH4Pa-z(Yq-2VVEd~V`MX1qmOtp#|-(O!1a z{{T1WQnxzyv4ay48ySB87Mn$?wNJVAxa_YpMRI%HQqmn=)JP%P6 zLP8a7txT_sHto%lNx=%camG8IMsv;w>T05~`1_+P^1a55GU{Gp8TJF3jchK174qA! zp{|vp@vM@UwzGKHoPi7hKqICA8T@nj*GcwIi)^6?m4O}hmilD&ABHQP;!^6q^QIU{ zHM9GVF1@s!Be#xKWZ!pa0mo2o6Q9d9yBkSHBSk6%H}vmys}$^ zd1CZUy>t19l!dQxyeIxpw zobXj6hNk5kE6e`?Z9hnSGva-B_EY_ft$rwYYgCg-@yEuCy+X#%SB<=joh!mRG>LC| zZEgw75^9&lVl&24L2L%bS5V?V9O_;+_)-4=1k2TQPm9+VseADg#x`Cd@SdBh%QM|B zoub*l_KvF@H!RabZE+(lvNqKdvEBh!-rq!i(;u|ze#$-=_$BcBz+M6Ht^WXu{5SBk z!#6fwEZ6NN)vl~pQjX$Tmcq_$wvb8Y#$;=TF7C_&Aqchj?c=W({7?O){vlm!{{Ry_ zL3iSR6Znzhu$KE!m(7}Aw79p4aG@4Bl*)ET5=Rnb>_y4X<@p8zzB(%~#6dzF?xU5{ zl$2!o61LGvM)!Am`>UC=`V&3O@VRW3T|#n{`4l-4vRuh?H{G{(yzQ=s>OcGy6U6#w z!`%nrAMBUmBjueF!JZYp*Q87<&o#A^L8Mzq%5u@#Bi-DrYzSHI7DdA-HNyVa-xKEW z{{ZX<`!swB)AEh2_)0$&YOV~!%CbBKT1vWv2djX3L~SLx>M zfAccFHtM$i8~*@;UHlr;SC`a0Tk#8C0eJy5_FX&w077rhy?gfK_=T%}&fm9Bfi*t^ zBwug%N#TzHT3bUe`mZ-hvKOWoz$I2ixhTW}${er+fn2BULExP~#6JRmWN(DJYdIRP zf~UK53dqdfFxR3X_WH0x*O6VX?RoGE;!o{&{{RIz_}$>2g5uXv)@*MtF06Fv?o!@s z(kJo@vq@W?05TAcmv`G?KAr#>RPvjJX7JTOP?2bo4{K7tkEr$ zw=CL)=8b-vyx8KFQzJlK!Ye9lfVv9uuZTJ&t+)IW3*oPWUBqATn*34mJ?z^~8KJV% z(?VubocxAd1MOXQlkhv@KkQHNH{w5tKWJSg{4L_YiJl*mz}_0rZ>5cO87wsmm}S=W zYm0c=LvL#Zyb(ce(z3L-4$H7E$8R3}rTjy?4P~vs#135UlLQYP3wId zZ5U!J(Zf@zkHpiBEKI3;>uu(>w@FD|=yX5uRV(|8@AxRc#eF1WZk?-HX_oPfe308{ z+HypL*aMI~{p;F1EuzeR4F1QTv!t+RQt{`+2l#_?6h%u7I`>(%(=@3Z1CTCs{spK?Mm9_(?HXuHuiS58g0C` z+HRR`Bcu6nTB{)quorP7>{BMy%MC*lh|6eUBSKM@9vbs5RVl`sf>CbCrrVb?(@5*9 zwO7Yw*bK_9VJgvoS(I&UZ40SMyDcwkTi11|^OxoE(QKo#f_4Owe$esDI^Y5$l>-G}MSQ)bd^xv&i#jKR zb!(;G=q@g{45~u130M={?SrpO9CxpO*S<4&CrMU<-d4YqsTjBAcXkEOA6!?2=$hE^ z2a7yq@fL0B);c^^a4tY0uGm7Hk3qbjZ^pT+(X9+6ECn^o8^$ZWlUKj0PUkf&bt<`y zT*{=mrD?a>UM(wnHoBhOt$5=5!(XzWkA5@FGA^s*-ADTx&7>G5A3KndkO!2ky@1_{ z{3G$C*Kz7HSs9#vXKsKU`?*#-_P0_U$vOA@zg&N9%l`oQLOvCIF!)7igvn>4$*)~T z@s>uG@tNZ;qp%Su=NvXWoc=ve2ix21%bkqsaypPOPSN?-^QVegs;!y(DEu+~cK-m$ zR8Q!y4Dy#lEcTB_KhysJwsJbJh8hGnlgV*@rZ_E=%$b;H0FZFJ0$7d*e`@XiCV1Cc z(Cv_TQp$UuEO}QKR!V=+M1$qGjl#Csih^a741@c$U_%<+@bXWl_;}n|g${*U0LnmV zpX7r)@&OGY_FsDK^e7?Gbkwki5!yh7?j>-ZV-^{h`G5n5B$1qvo~N4iyg%SailHpR zZ{N`WYIcL);0wSpP2HwImtX%>QLc)Ea=8nHJ^2-;I!9$d!L(P zc*h#*R7yVLdrI>4SH1MLmAy|Q@o$HdNWPjuZf#aJRJga2CxxJp{Gj=60gEm2yk`bC z=r+3J=Fc7Ii*0)Pj9}(h!(#vg$Ib6wp`Q@`A$Vuw7lh!tjyXTI^kUJpvICiPy}Kn> zJ;WSK6ESHOP^NZ-$HJZ^i+oaB-!2+0(Xl-@As~_I$j{K%^RFCm>&r6)=_^8K?>rmC zKVQrFYkYCjUH)e)ZK~ebOJQ|sa`x!>4;s5{Vc+*elq6$0#&gv1Uqs8`&jnn30gK}K zx4Qy6hPS)7wpf}Kl1qaWMlHl_SH^a;XFqfo3E&%E>g}n(Tc2;%lD=YLhL_pM5zhM(;iXNF#xQ80YzVn(}+yM?{a!ir(H+ zCP?|t&<{^?FnPsy);e{ysI@XIk~Pj*cMhW&-H%MyBXy%C&-#_X1e1bsp1pXlW|g9> zn>L%X_l_OqmR8&VQ`3z909ta|yp4{v)x)DHC65j7+x-6kpJ^qv0f8XldUV13`&UOg z?k9*{4rlC(fOFRz*JlTVT1XN`lWAoHgA3$jbTrQo>8q&gHnxaz819hzD*WGvO6=~n zB{wn{+4k-@;PlO5QxM|pj;uXCWp>a1)%$zkr^bKyBuB;(Z)xMJ4J%0TBv3^y-OSM~ zq^dAjJG>0=`O=wBnC#5wX*f_g_#a>KFN1EpZQ`W(Fwb$}2=(Y<(=BeNd#j6UJ6Q17 ze{E*Zm2~J@1D4tfjIMA$EB^p#FA{i{$G;zZU#;*vP!WKps{5&`jh_v1i1Kl7wsqUpTOthzNcZUX;$#WFJ`!qHOjV}2{yC1 zL`0fpk~8*kle8RUX1q#NAwr9&rF}Z;eu*mc=H<4XPw6Y-$H3nT>0bkW5BSSpj%yq5 z30pjE5N)bb{?zYlf`0dwpq!Jz>F9r%Pm6lb#Qy*Rd>--F!CiXk?W{FlgZ}^+Z?9~M z#=>XQJV|sk*_ntg3R~UFv?~BapD-j0e__A3FN3vji+`}E#P0?89{KIH3%?EDrk!^R znI7I`fy8nlKX|WTM-}{+d|dsjei?j3@Fbo%_}k$E_Bz+YJI@c@4J^m@+jiGsVSO^p z#Xr%exKbn-%19$)l6d>y3Mp5_({3^6N=r_u*|l$0+x1_6z;T{BjvJP>XUo5)==tOR z3P!Lu!;g+KPB3l$Hu!8?`h4C9wvUf*D<6g{^V{|^*R3=!h!N>m@S|GlsP>N|1!A(X z4;D9nyN2}ydRNzw5pAJr|6OBc8?pe zXx(MEM3eVeLANc57=!tLZGM_IohsQU*{*HwBAOXclF1}&2xF7fGJwDkG4Eg59v|g3 zMwfMvPGU$I1T#qsvxE6Ate(qV7?VV3ywo;?@ueZP8le(Jf zUKjACk;=A@rD}RV&5guq^La?e-J{yyLO8GIUMa>`!OJtk$-By%d-hzf^FMastR)Op zECwzyO-6Rvb$s8U`!T9qIUp`nHUJ;izH{-jTCf;c{;{d!lL{B+f3dnxp*c*3xWN#dq3Mb4(D=l92pEo;TMmOl>sF$qa#2#FZk3&|9rOEU*1K*4uAjHq%x zToK2&{6e~)N$~P%dIywYiq|ZnbwMW5qztJuZbGG)fsyZ;;=CoKc#Bl{&3)sWxY{i$ z%F(UF&V(5*BnYp!Ic6X!4q33g0qIpfA$XEesC|!3Qx&=cZ*E=Luu~&?0fh`VVB;kB zIIW#D;*pm=wmffGg8u+hRJfX9wlH5i9E>Ooi~FaM0U5#Imf-L+Ua_v}38Hu+&q|!f z9G2=PEi^ltUok@L+8A)FyK!X$AY@l3;NJ~Rd*T~i9b+z67e)Mn1FU3uYEI={%;yS0 zJaM-=S*vTQO+9V$sWnLni+KynvIq=cwe>LY3r^Qhw?1j;-KJdG+Pi zVn>tr`AJzrY(?q8!yI$b2N>ry_UDH@Eom)}*)Gu+aZEcAAVx(vZNDnvyO6^e6pfL8#h%lzD64umm2n}PK;?s^u5b$@pjfo!q8)XX7NiZbgS?hY~nD+T+z zn;H3=0=ryWmWL(NO&5svU6GQOFh2Zc`n`CUDvri}Ff1AugRoJdU%P=P&FdU5W zUOhJ%$xbUr`J3WuN}OdE@qUk~jq!Wo-jVTt;8%-26#PN8g6B%s^!Z@8j_To|P(mH7 zClSbD*g#vv*~(8^QHibAMfHz~Oofw^$S zMsPpXpH2HkwxJplNOO*OAmh`gKKMSyzs-aG3Wncm{ki@sYBGmJm&Dh5QPjpzsziMNjf8>s2hsXkEg+OUk8)fqYRK4f`a?w@;L`!3>SKDglh4S4s*9}q8w{uOF-hS@#h zt)w_OZPJ`E&uk`6YwHOZO%o)@vJ7Jk0seLQ_4{e-_OZ#Lcs|1kYxLe8_SC) zxY{KAYva63&t4BAuTeWnOO<-dPr+UMkI?)sz{@hvF!*+~Pnq_#b^KkI%=k}EGH;a= zy5}Ca>Hh%MsV?1^g@s55>-}n^E#%Cuc^&c32Bep9a-C1De+JT>(f)!;I^3k&hX7;i zP@MWzsRrN2nyc23mkO7e{YUgtz&g3*Nv@G!i@)f8 zhJMYr&1)}=EtOthJ4CUNq+qijChle6E>9k2dRN!qv$fP#dPcE7IG94Jn?NcH5=S`T zk>8Qcd}sS4c&k$IkH&pMJvL3Nap9ZsyK3)7hT_%HfjtWEY#!OKw0sG6Vz$)ggn-ku zPo6nDn6n>pNy$H6Yo~(e{{RV;)ARh>K4a+}hf(7kV_kV4{z((brSUbaf3=pM;p^sD z#NJxN!90VwDIksxNc^ku58}R?Z=!r$@lKA>xEA{Ms}S;?mkO7&00%r{(;V08SB{qM zf7&9?#c(?`uulssD-cKr0F+)2<&j^VKebmmD~g)A{Qm%rjp4E5S0859E!NKIg~IOK7U_3msKUD7jP}U@cdt9vybUaO8PG$` z!#9${_k(Hv9DV-)rZ;;U?R3pL@$Oy=qT%|5Cj^Z4J$iBVt9IA&Tgr6yEwH+-Rj^0R zl`2O<2Mb+ub2J; zd{*%$z3}ruw(yprad)5}lIKTReDAPoaRw4xqAY6PXLOLdZZ67>;g15o!m!h|4S!9$ z)htA4ntiL@&eA%fh>)s)nfvbS>an7N&QBZyNA(U>pHjuqSc-C1^*@tyo*l1^@* z?!T`+kE}mnAKH+9&p#E{!oDuL5cogEQ%7&97tYH|k@8yV3ywc@PnSPU(6J=qFc0+q z0B>_8vbCMU?0+*ZGmse-*Mr=A+#cfxy?+zUr(SCMgXwx*t6i~4Br&S*#8IwgXTZRX zN8e_`fIufCXZN4>T>ZYh585~D;-&54 zlHI0|s4}|;~;tf^T?#Xd2m6T2R8KsndDVq6L z$Di<9Z;Met>)_7;XnJuxB$v0hckDCJv}qlxegXdg8vHje@K1>GO(z(e;*z*{{Vu!{@h+9fW@Ww zj>AC7&UB3?+7<)QJ<+>4_pg$Cd;4Pk)?X2~FN{7Vc%JPT5K%Qr?GS_5rHQseJ7BGS z?p^d?&MG_NYeFl^?fyxAC&c4^k{KjD+Ax&9NiWR)?sZLnM)2$(XxFc8wD|kA-0@qF zu4HAe&2QRk{tCtWHGB=x?=vD--$mRz9H&g6TT;E zt*L8Tw4vs4`?N$+h@gfbNQwxQ?TR2y(UMC$JGTMlPbt{ti9B=GzM0az6E2rxEPu2e zt=(itw?=KdUmLd{xE062?co;|_UIz#40#Mj2>_l+!OkoC7Y~_X@buLh zFtk>QEfPrl12M>Pc*@ScCT+@EyZ$GR{22Hll}CGXYS{3~z0-TA8(($tpA8u;_}{P@o& z!5`VTR+B-xHk$tciB+^sE=BwCNpl-{_bdMZUhr>lPw|u;O?{Y+BREtUB;XHo(!ZKd z_$Z&pzxZ6Av?b?zu?jb-^<`CX?!c0WDe$#w5z^19pMhJ~lFY3-Y2N)Z%clq!rU45O3}s0N+omZX&@Le)!*RKP zy5m2tX?<$?+|F5ild_2ba;J81MLA>zPaQ{Eu=ZClj#qcmsL1i#HYOoC11-i$rO(Z| zo21s}M!EA%q}wnJ=IVC%&rBlAA@=CCvEiP;`9Pb@)FHaYnR_Ja85@PosCCHQCH zuY}r7`}{KS7l~~AMRz5{Fh{Fvx6s_c_d1QvcC0c5Lw=#g=f~DOUHVDMMR9+m<}ny{waayIs32k4G`$ zYM#MLoFfUVURHN}wzcls*2?xu?D^C7nD~&Nv=_#2h#wT*DQzP0uCJ?KBvO=Vpo37B zLlkbvN;Ag`Dw3qG<=~NnT^H>^`$G7G_MQ0Qt9)_s1%V6~f6y7e@?xnfbF0Q1%n^0#f zboWxRcc0Botg^GQV(LN3IK@*-IBj`Tf#^;KewC|lX>oAjB8{It0;xIttMl6PqfYWx ziuo-&dLN)+sA6i_Nx3~QySwk|aT@1^?0iLiF0Ze7m)Dcv%X4)!liWyDZC&$9%_CrN z12F(_Fe{jo;fICg3es8wx1hgyL5M^E+5+&PdI8XW74!wa!F$`Z^6u^yHuVdhKR@TibXtGFttJc$ z`<9dsPn7rUGmp>kuaV4rZpdk^Yq>up-{zA1PolthH=RyO)RY&SOY+z7K5s)Tcq+gH z$2@%i@AwL{4Y--Sv(dA07>{u;B7Iy*zHWtdNl6ReK;fm?ZtIgo)_@$ zogb5BX%r(Q{E7w-dibn=(jP63n0iu|t(WTm03+?O&*O_wdFRd!;dj zJF?)8LF>=byi@j2(yfzB&~*u0DRrvdYBH0_*&WQc1bzt?E9YzY`@~fx8S}+7^!Hb% zTlYPB9tdO+r6j4nRiC`|S5M2a`APo(1tRe+(*D(cA+^%o{GA#L4LTN91r=ib*OXVO z3*?vo0D+#B`KUW}4KpEGf3}@d1x$lblH};M3dGVLVFB$lz`BF!HYs!Uz`NIoj zX*YH&zE>c6lV3e({#4@e)F~d>wbsW;qVmj@lqzeTPBQ zq$A5oFyzkVh-7v7nSkVBjc@#MOYakF%RSw;-F7(Rezzov3r zhZ#o)C{nMasihe0%K1{h{qF9Pdb9cCI?wR07Y9;>+$uRab&_{!e6F{BG(6YBI`*e; z;V86avVtpJPfv|5eCdijATuhULZ3PE-{vg9WM=^IJq!C1#w#zf&RCtMHDn6J24lkT za0teF;|FhCS3BYT6I{}@DE|QAB4ae{k33shnmDcG7zv%KHz{62ZuUIap=rWvi08U^ z5^(`;}u;dI5=RbYIcq~jXN;u}6{{XMh{L76qd_1wPSS8C1{a@b5 zd@&uKm*R=+br|E#^DUf_uJWeFMKK-;BOn$ekC<=|0Au0rih6a&hx}uxcym(uu5D)7 zE&&70LFGtKpf4L9I+N2C_OFQa-x7HRv};25j)j%{u_oaYt9e0k7_^xy@-T1Qc%c=T)_e{_Rk0Sgs~%HzJIZ6yPvZmr=tLn*6)Q3><#13B@mVUz$Hy z@M|-a>rGAkaohYh{O)}2Jjc^7BUshQRsme#Hgo>~>-=l5w$t?;U%=WHnR5PJtPmB2 zm=n4enn~m*Cn25W=Z;1=u6s(;Wz?;vvVufLbGOW8;1URDJ-%K^&Ooni@YbiGYc}6! zx0FezL<~c6jznajRT(~+-RMPsJ>ntK$tc{r2>Az3n;!Yd?_Mb) z>NeNs?GU4>IV~G-19EUaY!T01D@qkywL0p>C1XDRB!bZS#|22h;F5jMQ}pj#bpHT9 z?#U!)Bn)>wKEM5H+VLi^t1yn`t#rOO)uJLh zdtw5{2*;oU3&0=b`d5;jjCVU6o=*p_;Azm4)>7&^j$aHOi2v64yWwa27Ng)7z^Mm= z{5#=C@e1hn*I#Rx+j(*s(-9b;d!#H`n~2f{5U3kL`Lchajbrvy_<8$Hcq2pjm*MRq zUjTTAQPMu%NVR81dG_Kvs0v2|Y%RUzgcj@4zlhKHBG2rl;{O1Nel*a2E$h%}dY^{$ zjZj@$*iR{0=D*Y5%!p047ijIHkO_AU%3Tp~8;bt`Kre;fG|~Pd>YDGu?NUiJyPF#b z-4fr-jzxy^b4#i(0JBas7>|U1`@6 zMvAhOSr$mhm=el{LdeU&=dTs|_l23wOs5|VOZ#mvEqeZ+GxP5ea%kpwDbQQ4z3hK3 z-wXc$WZxEiLb<=yyj^`gj)A6mh-HkQHs*N9EgVYUFP5hv||&ok%ExVbjgNfP(sHFkhlc4Nf_0Ith6mT8p0nU-rWL05C9FTHm?Bmz~kGV_5E)? zF~m!jaF3C8+8U!VCOlm7s-KCPp8&q&uk2l%W!npcRe&CDnc z9VUwfnl{1NBH?61kG!YXpVwdbCuhSAAK^cVWAKKPcXNJ`!Kht5#n+Z3l3S1&$yk*U zRaI3$QSNe0en$Am;pLaZPm7m2-1m3-b*!^B#e|k6QQyu!V=6p~V*)d|SqA1H1HkM0 ztN#E5&ha;kJa_R^#v1pCH59zE(X^-{wTd{BJ+FB)hi7F~F8*9XbI;Fs-Irl! zmSS^?R@R-H>#dc)uQS*1dmV{}I*wOSyD8q@m)GuJk^0=tn~5%D`>_H~rxoJ50kzaM z?NV|5RwecS6YrQ7@ zk|b$SSSZ`LtGfg&cM@2y2cFy#IIno|H1gbAxe};cZowqvj(}s^>yFs#UpV-(z-kTl z2@)xXCP0NgRV$J6J~`dhKqnsjkzbh>;;pI1X;USjTA;QYx(5q?Tlf21o`galYh~JUSTTQhG72 zw!0kOh2Mo5U9r;cd_^RGy1UyHPr69tbiph~?=Vn5D4=9*>w+sU#oiCTzmc!*r~k&M@<-pDRs-jL6=aHu7S#~JPWpxab|)ZxJ+v2H7uBkePk zSW)CG`DjO%B!u$AD#W^}UN?D)GC=@zQA|!(jf|}uMs-Ak$(6wyBOwVVKiv!q5sor; zob=8?ud6?2{{RB&J|6LwJXJNcQz|dn(51{#!tOFW#%Xu2n4}Lelwgp*#53Xk42r>I zHx{A~w8IPUl2Y?T&RKk{+gjwbpWQ0!!28M$E8RXD>e?oS;>XounZ%LJ7)Wec;DMvq zJojl`B{^Tc<*@~FISjjptwNIIyEtVgosa1I;4~f^*E}C<ulyC;!}gy6K0e**QR&iYwpw?J;dr5yR2ZPV9XMuPo8->mc7S?u zivG^N1zd@AJ4?7hvo4-wku}0?RO*hQRzSouo%@^REO1ETzm0$REuOOv#jhUt?@ocz zb8~uNhDfB0hnC(l0}X*!91Jq3t)ou$5rzg$N=+ZFfAC7*iZ+w@-$;hmD5swJQr-y} zfQ|_Zh9nRM<=P682t6@h(k7{GXW`!*N2tG;S5splk7`a4z{BC>`ND=!KvVMPIL2%I z_kZA&U$wQ}wYP_@HOt4hzqXnH#h8#Lh>khhq~UY4a8Dd{ukV}SuDu7w?}V{qA20CcbP zFz|+@Z5(#7##jhblwl_1IU@rk_sBUsj`jSxf5FZl0IZ`v1pTBmn8Ms@Nj8P37Hn@8 z==V0oRODqxmOxaQW`P6>pjR+2htjEM`Go5n961UoHOt z2HE_3gx?@sW08|rbj4V;cqbhTF#N$4jAg*Yef!q6rP3Mgu1GnI<;QLa=D*a2v_H+M zDI?H<=Od@RK{B7bD&b2L(n6(k(`kN&+?x|TQ)Nc}6BGoH4@F(&yHROFL^pVGe` zziH1B4-$N3*6txcySKRe9C$e!El@;od*(pr+co;raensxbsptF<6 zoC0u8IL;}`_bC)v9Wd@bQ_%LL?snJgpNM~De}w)w{f94nL-5Z|SiDo=9BPx?5T@$- z0maNxGKMS>GbCy-0bC9%@wI_WrgM&%?kmlfAvv|x9S#j8a>benZs9@Q)N%E%(a(l* z3pjVbb;doup4?aGpN64Hwd+XaATOGyC)=OtU!Xo7xsKON{q7Y&&UhHl{{UXUGw~I< zC_~i#()uG)g(+fjGj`Q4xi|WsPX5j}cNZTYuPrWHD114nk5B_OwhjkgK;!z?=q8ga zT0euYubGw1k}PQ;JxcS19-y3H{Y8FUe#v^H-Tu%zl&;&1dIjJZ(taevkOe zP>)XVd{8T5IN9WEFAT&k033mX_}9>Ig$E425or8Z$82iE@kT1NcDno$JgZN&v((5VmsN+)5na2v$>%~rxlizt19AFfcCRq;W{7lsPwh6N5zI(s>J|EsJ7IwYdsn63#*sF5Zrj3Rr8_blt)cHRm>hY9^YRM)w(z7r6`R9TXgUq=_TkdZzhk>%Vx3@- zT02F$gd-|Q2+4G8NWUlwNeleccs9~aLPbDINRC;sNM%#BZph~$xZ|AT>tE1U?CoXa z&x-mkmp+FGxtM9QY7WpUtn-UyNKK$vlu0a!BQrx8Rt3Ib%EagN&H&;Cg~Un|smm8@ z-Rau@0Mq=B%=|*gC5odOkfx<2ewzOPU*vsz<6ns$3er9v>c0>E5a{x0QFww#7g)Kv zCR>KKv{=k`X(sml+|#SgH$@?YU=flzpMep#jdk5;;EaJJZysqPYz|p%zmm+_9+pqDiBV$#UGryVPNrk^K{v;HW}`snB=Ti8kK$vip(z_m9M4UM3Z4E>x|uvP)N`yX*624Y|{`3wZSlRQ=7uEU}gx?2$t* zN6^=l{7tj4TS(?WsNj$n7$-l6KhnQRJbn8n{B+g6F?g>__+#MRTUgSQTD_9e<`@K4 z#@6O$xwjEbaM47_kxGa*u>)ydszo^e0Kq8#0Bj!}-a~b9X`*<2Fj1dT)YaWly5b9H zE(Y&$AU%N{Xono-I7zCQoJ?cZ&Px9PcK-l@(~aO<(-}fk@Ys1u{drx!{l63OerWyr z)_jp;oLBTG;UD-RcB?JIOXL3liB{0Y2h9$XrO56+lU>|w{{YBU^>2ay0N{(ivOkB9 zo8qlU#J(kyApZLLI|TX#)8d7(^?du+pPBt3xML2Rg;ziD#s0Qmh_aufw-MF1wTxiA z}4cyvk52;^7P|Q_;An*V@obWT-+P&9H(C&4~L{<+JZ0H?UN5N$|+G9`) z$-ij}u~-r@#eY&B2K|CRV=n@T07)=S5f=DxGgd_SOgdL*>) zzMX5QfC3`gK^$BTiZ&=X^cDF2bM%kHPNSX{gq{BYnWgwHWA?0rz)G~KDtMa7TeaWk zWB9i54y$3{dr@lfTu-T^1(GY63=0$B9iBvk%q`A$Nwi=B5UJ-^;k>>v_^;wiT@y&P zYn@|PypGDz3ZLHGTu38`9ixQYL1G3K0XzX;XMe#({v`N2;=k=*tm=Lc@~&;PYfEhs zd!)k)tGgDM&mdut!Zwr040$Fd>=wSF{{Vtbd=LKsgs0*TrSbDyfc=BQs%&)a%ksXZ zbrxihWO8lop8$i6(~x;yKT^$j#u!c?d9`?Q&y_nj_L|Z?I@_kl=h=S=)WqkX+4QPj z3QcIbE#KU`pSGU}J{M@81wI$(UkW@n!b@EW(QVAQ0M9Y`r=79Yrj)y%t$J`yGmdH$ z8R#)qz&Pv8f1j9)MC(>!4L19mEJ8W*&)O%7hwX1REKdv4YOUosV{AJhL(6l5$BT@@w>q z{t9RD?S3NscK-l{U8Be((fkmm*~o8}O-f)I?fPT)DY6wg$Ub1jes@_;mr{9Fp`E=H z5`U4e?T-hzYx^%KRch%vzVr0^EAH3Tv-umvZ1!1}Xz0<_H2u}<_MhLU$q>h6=-JI^ zM;i~l(zR_gjbiv_-qusj)0HdRIr+Bn?_2i%2JsZiOoAl(u|LRH-D9}REW>{v5e5GM zZ!4bun<`4=T4CI zcw!Pa&=q$L&x`11%3Yjm3+-_7&vN|$BDm`V<-68 zFU0k~;u+`jvZq0Mxg@^}zrf^mpNhT!(V3u{^h+va6^AkqM{I$N{x#-0&&Qn{KbNKH zhTtBpaKQokFDD;b`o~HAkh}$E(Va&VZ0ZSTQ_g>g1EPX0CV4z+h_Hj`|T z%(4>_K&<7MLon^h;P&FX548T#8pIndY-f!WbySW#`~1z@*l}N{8t;d^KNpLnYCP%_sU;{MFIFWIN}O8_4ciy6utHF^;_KQb*%n^WhJG*IpRJ z7S_t^Z*QBaU1LGj+=TU@yvfz^i|-Fc;OHCqFYX+e>>thu4pZevju+S;C_l{azU0jU>How6mlmg6ImFsu13 zAUWTNUlzOx@Iywn@#n;i7gg~NjUDV3*H#)O&9$Zci*QQ*=4-IfDN+MA{AED_fx)jp z_<8>T1sneWf?oK9c{=C7e~A_wK>q+qpIoxL%Mu3>+v;(G6l9(TK;UH8$se{X{syx6 z2dhou+l05cM~U>f)_t!%`k&oLv;tT%mo71dP*{<-HTg}Vc)ek@MzexBQYRsCIa8JT z1K;qkOASIYh1qoHXQ}-&P5Wv80KpD*8RNF`)t`_2Hf6|>mq^vFQ5AAG2{h?$)MS6W zz;(f|Pxx`+FZd!?#C;0K;m_=^@rPfq)uyzT!%VWfx4XQUHN(#s2_;VE8Nch4`KOL+V~7{g?bJH;T2t2I+S? zox&Ie#iTD3W>_ucJCpq)>4qVQNUbN{`>UGLbYm5Bwv?3HvG~X1i)m!m{7xs3&yN&p zP>e!G-Ip^UC4nFUr~qDh9Zh*wm?l>-OmcUIIQ!THWMkWe&tdJ)It zUu{BCO3zc|>qYankG(zz`1`^i4!lESs`%de%S}y3!xoXo_s)7xB5k|t#!7XXYO zZq@F88T9vzL*gwX3n`F8xS~9MTg2poxF}XZjyfE0Yx7sf-T-fj{{Rm(U+lZ6t!^|+ z>zh|;Lo&6!_*s(6zbVfs_~f;aK20ayAj3Yw;`NKDT46d}i?; zxu+mDe{PwfKg1&}N)mhI5P9jGbBg^%p5Ffe;I*Ca#czueOp;z&Po>$|M>Cw-iqf@UmfU^Xhk2dUm4|TTf-=>K4Z8H2YHdOK}Q^k7djyw_D0d z^55nt_{Ch)rMR%v?=)wyiWn{TNgURV^6h6odL{!7G8mD;=e=q8T05@`cw*jd@)$KM zyL6d`HWW?MpeKWYAO(Al54B?@$J#Duyb^Li`>n{scRcgo)9|mwT)E`mL-%?R<$_4( z{8^}8T-`{avkJJ3GV-bj<8a1QAM$W2jpv0mYfl2%d_$V**w|`U652!M$%#maKKXXz z8$m)o_C{2Zjz$i@sM%TFN>$nJT<-!bsl#w`NAu}kSFY*$-`S#$ZIQ=vvP?z};GcPk z$slKp?E@#1>N_aH?Ii3)N~vgbM^KAZ)h5(Du%cNDFHD643On|%T(bj9(nRtG%-kc8 zISRak&j+dR?Os*kZ4!N6>L{W+06UoX+t)u)j-I%#!^EB;d&}mv(^R}L#ucy#!90A~ z$v*hUKT7nH=8B2W87Q{d$6nmas$G(q`T7ChKF81s*d7v*fMuCLJo3yB;yYE{D@ub+ zjc315(y$vwasgqUMrf98Ov=R?MFgAxxd3}}S-mz0%d0>C*Z5!d4fw&M{1&v+?LHs) z>q*kAY%Za-x3ev8G;I@)DrqhEZ7(QI#D;cNKRXV``n~;`H9JcUD)4C@Mb(a?eCs`1 zQ4K7zzmggqoAz7Cr;spIXN|b#zs|Se{{Z|G{{Z3?3w@<%zBskfwd;6{c2WI`=|0sY zt^3G+)eA9<0s-B%lfM`>`yZqK0Kq{&Z7{StJu%+l$GpWVW?N zw^x}Nx5i1sd692oKo5Xa{&!y?_SJc?wJ-P|rqj%%q03r4r|);hpZFw0U|VYuK(YUe@1~h;0OPE#AKQ z{{Vu3ctb(>VezW+Uk7TVS+Mafyc(sA>axgWiggMJBlv(`ea=r>_%lh-F0VC9a0l** zX36ME44-}u2fcq_aQ-?q@UX=H05VMeTjE?SsN$!GT^aEo?M>s$Z}>zT;mSf?e%}%q zvz$B;nIs)@7_$W!{uOMKk&o10_$9mA>i+-&H6IhjZ5^bqJ-vkK6l)r^){&6%4ZODE zbCciRzApW#w9kZo0Ptj=6TT~3Pc5mE-rLEt$DcLBHptxLosX^#v3>o{S^NI zf_7@21=qi4Sv8M_U-AUFz8_`<8a6~oRZ2QH%p^w)t{a`%@Lh}C&DWf$wW{5kaRi!SC#mPk(3h%Q?=1mxgj>(6TQUxy78wp#7=!Wi>@ z>m+y$u%hiaDtaHi(<43l*P?ih(jsPlJhHAvag2J7y|bUfxgP*P@!IRG?xjwkWiK0U z7<2=!I`PQo2aM*wfadkOKd@<}pkwOg($uKIj!fipyN{cyB%pKFiMk3RK^%`uA!K=RGp-i9S(YA4^duKrJHMb3<)6$<+ov!g2RLI zfCwWTj)ZzwbSd9Nb6s0Q)$JjDI@ipXWvjGgu`T7WZu^~sbJTFY*zZ_VX{S=Wl1RX4 z0^ldiS($$4?{_H*4tUReh83-)%(syw(6LDq_iE^%I7I_7?I4neg23c=1P~2Jb0z#X zw=X5Rc398Px!h!vgBVDL9Wj!!IO+f_Gfior+Os^m=I#sKOkK)`NS|WJgba~_q%%p* z&@cxh7$YYqKU91W@m`Z<<;f-r4e58fY}-zu$Cs4_LJ*++!g}F`IUgx#nq}Ou+z3&i z#TO5@Vz0c^2;U~oX1Hr!ubsrGjY6))^jRosCWqZWf5nTew^Cy`RBMv1X zf#RlMHA@S*BV={J0c6jaBfMGLg|1$=pKw=D*9M!M_nd;ZS^6{?L*_R^IMdE|pII z0Nwed79~arQa22B#~o|?xBmcwY5a1&@WSZ6HPG%Ob(RGx%_qwwQy^Ey2?a{$0F&79 zTzpGHbZbKqYbJVlc+#sG;iDw<{C~{;#toZkdaK#lG;Iv4=0g;CQMhvCbx;@(epSgi zBc3bs1O5wr@X{}gpYTnu7Hc-}quh8lYi({e3e4VJyM?$_z|LAHVUzOuk<@;wcsBmS z#XbYlyi2CURx<*H68`|KO}uh@bmu*5N?n`h#D*Q&daBg33Vnu$se%V^s(|!rPiO&5)QMH5)Fa%qsIsX8@Kgh4h7HoXu zu1^NPk34|OMehWbKw>6&Dy)k zYjkI}aradvTuO(Yg_-d-Oljwf;pwP_@}LHSMg4`@mi_p{U1$my2=&Hsn*1A!d`QN}89Xi(sKzeW z>VH7+Uxv6kxL9KGNy*7x*Rt~4{Lkb!_V4|i{{Y~jpYTeJLrc)LIJK_{_*=yGL}~XV zSji3JM;I3>vO=>QsyeZW$~vhe0;PZlVksT;L<^AOdsi)8Ep*7XJW(ZCM*QymxX2N+y{RBa9u_40GGlJ^r2k!T6P^ zH1-T-E7t&w@Cd;p)B4xPW2tgilwWhyrB|8?UC%f8D|ZFQj3T&?ECE>wQ_ybXp84m$ zt$joLdHf&nU&Sjg1nV9rvc0;L>9>|kHJp(vAc5nQ1dV_Aachp^n!k<`;wnz{qTnG7dtM zjPyTW;q6Lr&Zw%fQj2X_`KG^9PyXww2(9AXDi9b9r2F*SI6HH z{0E|F8kAa=l8pz4{71URbW&K`EyFItrMmeSCOe26fL14E11rz$UM2W-LKGe+2)ymv z^|AQ3haMy1E7khGJL3Lj_Pn1<{(pV`S=1v-n3g_)Nar4(tyI!PZ*rE5j2VGa-^4%} z;C@_ot1%_a626_OFOzd_?+nUM+?qBiB9BlVSbK`eR*%e)5pl44?HK9X@b|CgzH27C zAJ8t+e8yZcHj(06#7ObxG9Aj<8+x+?#zx`E>7Jgo`-%Gr{8PK|PM=||S}U{`7q>RE zO$4SJg@WQJ?njXS04~W_YXyHmSH9u+qj|m?_aFe%2*R8b&j*voxgx()e__8ACx`DZ zZYR@X)$Q$V?Qd>vnU3+~OmGGc?#L*r4JJYB*0*)zFA!=P_P-jN z&1P98xDlukDU%E3qLd18@kPgt{{Uw119+bCyvv)N z62#p6{$?jg7AZ#ljq}EGM+D~;^g0r&v>b}~yY>T?Nk3(60tsLi;@a<2w2mN4Y>M9c zVGtzpNJbp@>+C+(j4Lk#HT-qPb*W1<#7c~>7bj-5PgDB8g`qWu)NV<^ z1Ddl5bAY0t-MHaLHLGHnAc5D9t$eRCXV66&P{hd}C#`%L`*{A-m;V3;d;>4T--j0j zc-O=b6^dBO?$a#g%C*@CAWGmAwgexWG7wcTdk@-w_Qdd?z^f_zDdAryS@Dyu)wIs| zEJ+y>-jyQ+4psm|zZeE!`;cNUA~`y=*iASd0)qmiq_u* zz7lIc5oT3;n@EIKF)_!RdokH=pFrO`9?GQGu165>pL1W2U+_?miP8A4;tij|eLeSC zcx8O+t*ygq7YWL?I+$P~N4$b4_GaNx!SeJW9Ps7U3G*wK0p^Z}i%-xc_+;p=F?2sFK7fA|oR!XSSq zBDc2GF?2IeMUg_IW#>0 z>t?;Q1bLDu_Q=g+YI@_@SlnFck*&?bLnFf$;5^3)b|!mnYz+3UW;r}HZ0zQE*n!C3 zl0N~-{Hpi%rLU9)ta7|ye+*|In_`n?!*lFk*~?Mz$HUJFN#Y@=S!$Z)+yos$`WPKTA5U5XC8K1A4d|&Y&gLTgo!*k#}3u_=DBK55Tq!P|sJCP#-H7YV3 z$HNji$T|7KOT?CQD6&Gwt1&?5r&0q}j5>|`a2h{Py3BF)%8ol%;+z-ZRx2#dDrWe+ z#;tbwl8l>CN>8f0PTee(yC1W0uZkE<&n$*pgv;sCw=LwgZp)*xvVC=DW-cX8cm($U z063{0;$6GR>FN2`6yokjUDpz4JcSCMmH{L4r#uia9&5CUJ4q`s_Q)jv04n{b+9UC= zW^^}tgDMnzgpb_d5&HiCO350DMjl*(=#%#VVt+hi_5CWidn-70A1x%?k&%Lb5Gt}Q z)y5};7TR%~WO6gu5lftv2Ij?~HPy2oREmiWg9V1lK11MFRQ3b3=Xu-&7 z3-Q!^p1C0P@m*K`3Voni>vqpRkEGtiYTKu>pH#Vp=TY*oGuuR30zuB`<96eWWSZ=T zPm7wmiO)P;2=33MzilY22ZFR+J{V#}ePB6Yik~R7pP!U+0VBUAzC`eTouUO*xwM8| zsk8=(2;cqV^5C3)rn|q2fACNrg@3ib!~H+u*MPnf={^O&@aCYJmDaaquj=}}#mq9n zG%IXk)NP_gLhK}GM=T6sc7RQB9v_Y5w|0c?V8PhqfLLc4{4-vJXw;=mLZ*zUILcn` z#utrzG2vvg6KVEV_U)G3BaCIZ@6WG)!oNyC;E^A-AMGpQKaM&l?AxLE=T+0Z0OIcP zygzI9YimP1k|d8KNe#MP$n35j=XyItGDi(k*=3^KN@^xE}t*O&l<&XBtYD2dfKZ>S&F~* zoy#uK-H&fRVHEHd3No#*j!Rc#=__o7`Q`o_f8cK25Ftp@`hu#gN`htl-5w3U?GqK6uGt8vu6K>f9g3^>X|)@G@%K z#l4mHT|HL6GxHuU;A%OIXyTfBTbV8Hx~=+X^m{8V{{W}H2l&t48mELV;qsOhyg$YZ(8Gzi@pbZ0@Y{M@3o6b^&hgy5wNp#y1lqURI555 z5yH)}L6$~H#yT9==gas?=3K_V5G9hqheH&$QfxiExb2yM!8zF6_O3zfd@JFvGSkA+ z3GPsr`$XUf(c6(C#_aDbtNbrE6?h^tMSabFHNw%P+o+$WxAa$b`LpCOTmx36Z-!Rs zqe(9|)NkmbZhp`|89Z_EZ^SLD_^H0mWQa!&nwj#n>pZNHzzZm5ns&t34!{@0VU<8t zPZZJg=hXEZT`tj1U*>Wce5wfC$2@V;s%uji=X=Y^JlEaJ5=mflmM5upCz5lV*RuZ4 zpA0@Bd{gn(k?`l??V{XxpT@WFXxc1t6im-_;Srgia9F_`+~DC6n`>if{98HAKCx0* znd$m@AGPHeKV{2?g4pPFm9V$8)3tkB>uc4L*5>Ez@cEzUuCq&T9C9-%DVGtme(bnV zLX*2HJc7mUV$D?}0bmd~SwXu0GuELT?#(Iu3mLt?-fnHki>#ari24F4YKF--VKAT1>z41Jb z8XYP{`@~5m<~z2KK<&#AJ6JE8mGwJ#OZIPK8X*oQq@Su*wj-AGOSFcHXzNZa& zUD+BDJYF5}ay_XL%X*tvpe?u!&Di}bJ44e_%S%b8;#WJl>9_p#sHF3}QLec!`-r|o z%AJY_A1LD|Kh*kF?QZ@kBuzRJHiatK0|aDq>GiJtm5e1Tl{Jk?V_zx~B>wXpFyk4) z_Rmq?xWg=pN}`+&2h;PdoBNA-WLs6ncwON1&sN4c0CwV!>}!t9qnuJ^Q<^BZ4G%ky+U7QWGtNJ`<#V$zQXNj=V{% z_t3+e^f~OI)P{>S%r-Vo<(7il9-=1k0I<-LR&gDWxZLtDm-bLmoQ83NBO`XxlE}1m(i}_(Dcnb=@$0Z>h|vh_U=(9NLE5nH}R}(*PI+z?@#;_AN~s8@D>jh z!Qrh+$)=wALve4ac^_pE&lLO&yX&F#_AG=tvGjk7{u+El`0wGZTf{aR3TW592ZQZ$X)wTHc(m(@$eGdx9%&Ghk{jj#eJkTV zQuJEuX3tJyqT6fPZqJh%{^B-loNZXX(IC$mQH+ycuOIMA`BLNdV%09RwJ~Xb z@R8ix>9KsHWwo?XLO26CW(@7SCvzUy75U@)M0^-wMj1b7rs}c)Yq-7^4=$dcsHs9Ql^8Om-7`SlYWlipHQ@# zSl@iC1G5%g#AlqIa(+?i+ov_i>ZtL?vD`^1l12nEP@t*kHo9&j*8mK0jz)ck^k5l$ z!U)gt705j}QO{1eV=BCWrPJ<=?ge=#Jb-hYazL)CP}Ir0o;$5M%yG!R zQH_;BVC|1Fik-@VKwg`&a1`~ zYuKd=Eex0=n1Q8D>uRaoUKv3#(EvHtS9#0TI{{{UW|q~_+%JF}U*r`^LMxev!1 zwsDcq$^pl0RL=~raE(W?(&#a@p^8Zx%92%GvZG2$@;0_N1Z;z}{pH7AMSY|DC2Jae zr^LNGMYNU7bE+tN_M- ze2jI#86&4sdkr7NT8D*iW1mNwSVU|#!vQg_;5!J}SQ4a+j!5}WY-LTgAey>p{bm0E zf}8lIuD%|4Z$Z>7podfrWlgiiaPh_!(=8M(UKenn63Aj;za)jQ5A&Vz9@IRco-zvK zaVK*z$ON4C1M%tBzWDf=@iz0~Bo?iGZvmFhAsa^=O#~6E5Ls3r%eCBs(mJjrU>gC0 zSH=D`flB?PqOoakTN{_~Bxl;brweBro@9OBaUaPa#!DZt{9W-yxJ{~>t=TKJ?jkArr471h|9<3zcd+C^dmEFkWMNbmcyF`n4Rb6?HJ?K`4> zV1CdZFVh|OfvL+NAOXa?Wdi^Xf$)FEzeN84;Ef*^m%?8k@4g*gOjbV;+J=NP0ET-~ zNgU+vXxojWKPlvOuesv>@zDqt)u-|zhPe=4;y?Ng39y#ZLu+x z>~M;552(%$Cz|#Tjs6$0)_x3l(?OCs7++|5ghg_4cU!*zDPD8vN7lb9Kj4`k6y~+p zkA!boL=rRf2H~CFqYsp zefkX7fANP(k3#XK*M%)k~d9|!*4dWpC2&xUkO!#6iKS0zVWM;gQr z=O-2Ue)~Wb`WOEI1q0QybUMI4cOwWT0CO%CX3m-wA zPvu{osU;SElNfu*{^7sikoRhz@J?Tcc7Ju`pIelJ)%R+b7;k)mn)}yR)U?K$+IbiV z=TnC5jN|jq75SI@HTcy(XW!Wi-~`%?>kXcvqgg{MHU=cs*fe7sPDp7~eNBA@;yp&? zej2k)E2+>mITAt%8guaCcGU0=f9HMoxD^lMjxQGlGWw3SS4AQ-`@;;GVx1JT)i9e-B?f2GV{Z z>6cm?jo_qrBr#NcWMDd#=cYz${N=xFe}tbC{s#WiUNrb;@rO;h()>rGYd6qe+uX*{ zy2&NXvqvTylN|9n%?pCbByJg5hE0B>;ifWjuO&CHExzaeO&^T-tCiEDoat>QrT42p z#Xr3H&q-h3a5K-9+drJOD{Zan_f@|smeGA9-QjEHPX5D8zvhJEQ>A*PY z>t9Xy#^~R(*?GHMXCQKT9G_3~k9zns{t4&e7O?oItN3=@nBOMuGEIeTeikqGI? z1-hW4nBrxHZLZ2og1@_!OXoQRZO>lS`BP;&>NfsN zWIGYlj)S4c*A@D0;aOkDpR>n_^tkf!-)MJnZ2)9EadpRH4>jG8*Y-Ju3Xe-lKbkzu z_YLjv6{&ZNUxE46;_VXu09^6J9vJZptaji@b7IgTC=+7<3X;ssjH3gx?%RMdSw9iJ z6?s1buk;8l$epcZm6*!1>~j-t+kidE!Onjncg4EY_AlYD6*CceXB<9pAKq430o|Nq z8`nSW9<}u@{V!G3f3xJ0DdU*5CNjZ^-H-z>To7yfUcC=*hn*^#rkeZ@;(9o_P^(U) z9jw#)nfQ0(i@|^5RPhSn4QkHIRw|_7UgiZ^Mtwxl$WI;X0!08BiEOR_0Fb_f;C14& zwQGXD9Pq11*<+{61Du@gx4oTHp4>456}JLKbrfkPau;s^XFk7=dj25HY0497d%xFH z`%eud;F8&C_i6eZuZX8v+R_FkNin(Ojmk>@03*$Qt^UGVm94Lh4V>j6P|<98Lye*G`^;LIYPa5kJ@C?Zag^B>AG#a2x7gug4PQ(aGqRl zMpQ`d;O!tWY!XQ$zAfKd+AuLgY#QkBQcS2;91w7FHu`>bH^ZrPNcBi`nRhjt&i2=G z#u1gAZ4tR=c+`brNmpfFhoX^Pea4^RtBcjrEJd_%?`YckcXE;{F#=M%hWjYw_+ z@m~iGux@cXq z+@;h@a;W(L%YQLMgb|#JbTW4jUl&2(tqaT&Pi+*E003qMn33+p{dqO;pY0Rj>kEI4 zS6AYA<2HI#l0yqOn38#~A~=mk6L1Z=$Oj`F9C4)?LBD%rKW#bbvHPv?`~C_0<3AX< zlf+&t)2$cHVR8MF4BE_#CVp_z2eylOC+1@#9P)FT`e(xb0PsmU^~gNc_=OFzT!PnH zOq&KtQI=c2IRp}W@y&lM8ZNV}=+>e1onKMYf&fBUUR>J%K)@j&krWZw& z&(wd6e-}JE@I$~Co;&diDR{w-#@L^p?(Bo*!h^p*G>4*~EOB4Xey6EwI@YIit?Lst z&F$>cTwF@u^3qvYyp8y?5(i<&YPaKWiGDozp{tJ&c>7ei*6z_tq_ehGHSV#X3mj|H zArxfw&2o!2qxa*89Py0%eifIsIBV5&(Hg=+I`dFVs6_$&;{jA8k5O`b@)!M_;GB96 zNySYLk0hCC8Z^#x^6okOc>F0joZif>FNy32?%@vY>IN6S;lX4xFBId$GfI^Z1ksCAOiYLS$)+h-Sd3jH#n zhI#e`n!KrOz%`Y-y84L7q*oare_fH%uo%y!ZAO(^j?sxSO47?q5Nj^sNxU&zFH0^WuJ}YK}_s}XTFI}J@9sN0|p}Ukw?G$Ewo_Hkx03%<`7mL5( zsQ&=9R=){NH&)WLUi-A|J~1b_4RqHy&q8tAk6QV&#b34O?Fr*6lP8KicdlG0A1X(A zExK|BK}%Bt9(d$_Rp?XU3`^ElL_B;tE1%ZZvErR8Mv3oyM|Eqb#kY3*nWKYf#t0?Z zdV1H3>i+x zu8>P9InGRuS8!EF?jhkq<-`I zYw;)i6ldVhn`#O{cF2Pqlati$CC{+Qa~j3qaCzv?#@A z)UTRT{`Z!P2yQzE=ca4<<+||(yQmUxE~S$^V{yRa94O#cAV3CiKPvVp@Rk-f(&Bi@ zv-W$$pYT%Oh}ZErPZitSu>^$C^rcJ#+kMEmAc5a-@5Ow#;?LWQ;tz?7jbmNb^+Zy9 zwVLhVjQSRi83Je6NBUR6ptOn=j()kRHk=YV*L^&Oj9au(i^R`n&v3u^<>EQ+<-W1C zvx?$l^GwnClCUJO%P|?jJ^T7sHK=@9_=l;&BDmG=QCWUvM%X>Ex1Qd=rFobD$uERJQMy+dXU3YT@G1mqh{6St$Tl;wp>eci^y4Bq*6zJn?L8;tH*sTXJiLp!5GFe z4^jOqI1X?+A45vmA1Bip$8TT4x?`c7qMnTVxAro$rNH*}ug{BnO9`p^Cee?0U4 z)%tbLHw1)&-i75rKdvj|FBIItcNmvTb!gvqP!PpV00-NT&c3Dp0D^aT)52d0zi3#z zckwRz9T!2ew3cb_E^@HOk+61Vd>x=D`f*-uNBgJT?T^II$3KX&K))9}TVo`FUOjtP zU`m2`)NT(#a6@BnILBJiw${GgrAK4tEZ=3Bq_+g1jaE)%MN#uJI}EmR1_vI#74c)n z+6Rb!EBKSe9vYc$EN``qLh9OWtcBs8`g!7Qt}+87DcVLxc>vaChvPG2cpaRebC%uC z-1T5M;Pae*mGtzmE>xn{_B>jXT+H=tejOM_mpUnAX!6G(3}9f44wyfu6_=;$;&RYk zC=a|SJ#sOepH6yp{3`I&tu?!7uk`t2N$}*w9}9)g-Oh3Xj!8c;Bx1PBsp5v?d*D0d z6+9gJetcHc+qI;O94)cw(CahAh>+av2oIKQTaM_X9JL>$OGkE44!(P!1~wqMgIT>_s#nf{0#V=sQ%Lb03ALu zctgWpC}>WB;olGHdbAc-cFA!veX{;#xY>zj^5Txs5!kFD-19Eg{$T8NJy*mQLR}I9 z;PIHof+uR)?JaQEbl5QEQ67jzyqN* z$(rGQu?|No9+m;}Nw`A(XS#mT-XpW|hsJ-2-X+uSW4ujU#oGR%4a5z;Io5q@_E_ad z95Tp`tgXl@z>W=k&bP2x+C0R0Mn9c~0m&S6@9SM%^~}~0L1iY=WsQ`k;g%qV8-2hB zy?Kv`FB!;q2kzZ|VUFPAra!H1m%`CdM{XhOdmP!g`)%W0%a8&Fy7i5h?Wm^SAV(S#BCnHqIDrA|EIuvpR#*ImS(G7j3s> zNV_z48eZ8SYsvSw4o@Iw8T21c^^Dr17A0XA&lwp%k2L$6TesR|x`-<00Z9#leLIur zNv3(0!-md#@+lC1|Iz+R{{UnE0NIEB6RV~8zwt3*p29hqt?Vz9vR+Hq%$&0+5QZf3 zes%*VZbAN=J`;Y|J}>=`eggQ@aF$(EtMl^4v(mHb!>< zv0!kae=NQiYflb`5WDhj6&;%8w=&BKJOWbzHtoj)ft-PpUR&|YTlnweFNcfpzu{E& z^F@4a9(GxhAsk4!l4g;%l_d@Ww*ZpEoLA=)#Y+{5jH+#a!2L>|3K)DpBvt8+f5>WR2u&!2Wc0(Wdy?N0@-QZWpr(^FI;(#y_+Ng*0FING`N^?rnr` zD>bdm6BSfYc}?oX;N&WXC#ExBr62H1kJ>->RsEE{BwKjP#8w*Kwc;H{-YY|46b#bZ z-iCawtZ*^7m0Kh^3A?g|8LzyJS`_f|mK}3G$oVWEwsq34r>=+k8Gpeq{vGH)0{;MJ zy*A_fPQ>3_!uOsb@>x#F&RxD>?m{DOL6OjL$u;@E{{RJF_|f|__;2E0hJG$ekfaIL7+`%U!=`VtoAF@}+pV-bng_e3(mE*lD&XQEqOpNl} zS`zm%BLd(ov4Uhw72U=$CerU% zsZhx{l$`VOfXp-Niu(8Vb+8NjKIz(Sn|B@m0FirozH+uf-#L@MNFX;k{5Y?iJ|TX{ z{{Zk$AB)Q*T(QUOYerav=$hC`Gc&?|OB7#WbNd>g+B9~3TatjVQuckj`@7cS< zAF}@dgnU8prZkWb>zMs;rYjdT92)692adO^F zC6613BxfhL=Uhg=s7GNcSc1|RfL!1CR=q?84~2!XbYF~br;BP0%Ska5#B-0CROTf1LK;h2SvA%b1| z5gLu&r#b1=;=JcncrKJcsg3FLWk@Z%f({4=C)^w!)KX6OCQ>}RU%rl686Dkb3RIN^ zfPSD6w=1|0yx7ki*9&!HHe-@#+0QuIKvl;jhFIr0$iU=x_4Hp6++3k|Nj5IhrO8l8 zQ-P90ect2KCzIENW|i(8C2-6#+_Lk*=jI9t_3ANR%@pl(jlE2#7YMrnby5ckyp%b{ zNCW_VfIh%;T+X|5c?v~xpyWE^doRqxfC3f;ev6&JXWF`JsG+%5j@3adcV}pRaq}Ix zZW#2zPb*Q`ux6~iQPu}6MasLOfwq^2!x{|gRm~+ zxE%12{MbD)j(zc3sSU%~%l7ba*vM`;JyiYQTm#3~y>gm$#goSzYSJU92>x-xZX=9H z#1r@&4x=3hQw_wir=EoQWNrkK+yj$_UBr?y1_AtQ2+_09#mimDyia3m8$tF)mvqcy z5rO8UK#)cd?Ftw#%*f2y!6v--#TsM`gfRKbsmaOR>73_3=RTF!-_H%j%wh|PCknqZ zMy$-~v^NpqwvrD)*E#1rSk#dug`j3tWZ)JVU=9!6&gD4c4TE0Z5!weGSMMLnNBk5M z!;j(r0NSs_Y7w^Gt*y{32o2^ypt9r)f>ex;PPqoWgZ6dt^~b_*jh+_xgk51xBU87Q zRwM4h&Lf!q5xW}u5B>_ZZeabU?crbH&}=?zU^Zm9IRImU)9|m&OITV~DV*+BP(9cX zEB8F71@)Y=F>3dXzeDqEb#G;hl-EXorzh-V;+PVSx+@4!B@QHT>_D z(ViOzeax&@dwemrYAD3 zhj&m3e(Cq!!XTo_YyXvv2Q9y&yVzz{B z{0?j5Tzc2#IhJ!geOY5FD>=6&mrJ{!xn%h~Fc@lhI&HNCmEQUaNbAy`k_3=38;5b& zf;k!M%|^K7Qk}m1SDQMD%=D4{BYYY7^1rsn>~rC-gnDkZ3_lb+0QREZ!c0Q@l4MqnB#Ps2%&rS*1@Xz*+_)YPpzS1Qt75$|4yP>!e zz)s?*`&7ugkep$Nuj1=K@fN$`ZB`9)!5(9Z>`x**9&m}0KF{IWJ)9tpD|?&M1haboRhl*eqEXH3@YEa6* zZsTJb+iIV_A%;MaOzAe|R$ds4jhgKTAN5<_G!LGpO4w-JD^&oyT3(_3Gu^_Z?F!R8oFNVN5} zp7y=Zt5aQeyTp175nJh(7xyb+6f3qgw2n_A-B1ny z5XP9t72x_)NN(551)H$PTK>DsoRta@slOt>t3R5t$}X*EN$GcQk@Sb`&*CrmLB2Zp zUs{y0id(^E_+q(B#Aq;a*fXj2z%}|O@iFI?=6hJj{Clxk2_$7fJv*Q0#ePo*zR~Vs z)ik%o#gnbVmKnps!bLd8KQfB_P56VTLE?+#@ivD#i)~IAZqZa=GF(O6>7t>)7%B->|sDT@p+ zF>xLbUw$#`Uza`-(XO>e7Pj$^EzVfJeq4RvIvg7Nv*1>?bW7pkeNHlNrHn~x>Zb-+ ziDg21=NyXgE7faQ?t0v}i$6Ox8($vYPkZ2~OWs^rU3n8Ly6y=CtqQOl46sarfuE_b zp}rJpy6%y#%c^R3cQN@fB-4KFR9&e*Ma)i@~(F` z?IkC!y*!WUd;~qjnUOu6)xt$;jgb=>RR#zJij0f{%WWKV;=e?{WQZS7_Z7|Hp^r?Ibxbl(|xU*Ybe(|E5>H*;LZ1WM5?ixVpnT(6lK$pi%;o}ZuC zU$M``9|ZW5BNr(f{HYH^i3nsa22?*gAE4t0)%2TdJ13HOZC)tbIP)YfaKw-| z6O+doz&^ay{{VqnM};-d4@Kj@6>8U-d~w8JMs>upUMoC`M5?Z#lx#aTZaaoR1D{s# zKZg7@d!tR_?-kw~okuMphTt+o6aC^!Df4cmB#)GNa17<&j(<0fDJ!2%JUia!IpQxi zKZZUT(`I;pYwc?2kgTW={7ax8FC^_8XPlhiVz>w`MYJY+i54-?Zo-4>jPu9Rx<83` z_J>f?w9P%H3uztWI?6_Kbu27}f+`}CE%;Q(UgVRS=irLgFv4^@w&gZ}InPs`NbB$T zS5zzLaMh!Cp&Yj`K>kcEk};N0PCb9fug<^Q3s&%kkK*ki>||Egjk!0G6)UpSBwI6J zkQld^6lXhkWDYa>A9pE^JV>E2fyW$YoQ|fyHos~Xjs7z$frUCiq40A>T9O!UQe z;OgF1k`yV~pDdpl>QJkTX`zWSTbvPt$9(Zz-lO8XtBJ%i+(S2fFCZNB;E!KZUW2Jv zLm@W`kqmQ!pp*I>X1M)M+SL@p3{iu$j#M6c^c){fYuKa8-i6NR?4SG+yZ#B=`(J+1 zo+P&Tp{K{G{5$w|#60@^sU@7+i9^JCd~yY6h=KddvxEzQ36QyH{*w5W;7<_zJ@}KZ zco)Uj7WaCdnRfA9&uur_W0pgQ^A>g_xOHIJJc4k?AlK%1{1d;&`u3;swKSg=Yr4Js z6G5`x*85SsyS0zY1c>8;=6NHC0ryrURR95i0LOX%00kWVyl%b${?@)O_+g=VNp%a0 zL8sh(rt19L+p=9)!y=VPqLxk=6~ka1q=Q^@%PUr>;|F*5A~6-C82iWD_YAV>yBzJ2 zP7ec+KAr2Bj_U*>oyth*$s@lWwfy>cv;GQS`%Y^SzxFxNG?jMI@x^r>o`n4@(%j?NoF7A7^!Qs6^piY!W|UL1 zKd23BShCY&d(B>WtYBRH_j1Pw0CXhC$^rNNYv&&ne%W8L=Y{Rw4Qt2xG}1!E7%b;E z)0bX0h^F6nG|C&Z(>F5$hL40Fd09jHj4O?8N?xJV*Zk1uggora6zontzBq zK`6=n#e&&J20HJ)k`vd_N%pS<@pt?cZ^d?Dzu^t=EtFYL-OH+K6G+3^qPBr!&O3R> z>tBXbd|>e#hDf4>D}T>GWIa6z0DV9e&ENR{0L9vnWV*ayj0o6>h%iPEMfCh@*rUR@ zXzs7`GkB`+Z4c4!AAiA1e`{@e*u|#nHaZMB8-|sq#S=GtZ@X)z{D?g{10udP@u%${ z@w>-T%|DBLX{=v)$C%~5onyzeP_jfn1h1atmT6mbtY6)UC-tNR=V{>AuS+Y#!$`px zw>PY$Y8%L7<|3uR$YlUyL7cvgLuc1<8^1G}f^AOjb=vG$4_|7eNb61Sxb-~Os|^*b z9EtC#Gfz7lgYA(`XPLhH6ZJe%l0XNwJVxYk{{YvkP9kk00D_(qHDE?c`eX9^=~mj| zn1x=4=~Z~4I)qye02Df&20C@C>t#AFTw}kn6{{AM^2Cf3g2#?}j@bJ9R8G1^PO9j0 z+c%7kJ+V&_k(_=N(?_Gqe{}xp!R;11)O(}%)$7cq%)SLNI@=y~VU@u>RbI*Zyq z$0hb+Fzb(OfA#4_vpx<#4!>ITX#6uJso4}#FoFk51&&!%w-qCiE`fO} zf{`9OpU{14>w8awmz&wxf}1|+1h(Vvk<$l`m7;zgcymeABGT{gt)ad-WQC?hk%lv$ zDbC}-;{*9t63x1!T6hYw*2h)*CipJZJRz&=$rwxj0Jj@fn6J&`q%p`5dtkc~ao--5 z^)>dBs6%Du$|GWM3hX`c)wu&5O=bKv@NLh6yg4Pc{iWTH+O8gTTc{?0ND3l{kIqEj z9~j=5QgSo68uVR4$)>kgjeh4O1B{II$NA#EM<0pym7K176fo^8K8KIqXxg+)HZ((a zc>%GC{dNBUf<;678d&^7)b#INrFr1c3hlf_uZM zUhA5#!y9iA-D>wUYIhB;-QH?w^IbomBS&X*Bfyzy9fG85)bZNB=J;FtCwv&bO=jYqs9&pQKp+{?*721tfcop%unH8{1bon zv-qQSvHUaemx;9}Ai5YdJMj~3BktVXMFpwnkUZQTxxh90De&L^2oL`N1t9UIlFQ;x z4cq8VA?xA|Lr=11&VFKTCDdO4@=uqX;1SOs=*MmFtKuZ_$8Yvjg4K5e9o5jkR6i>WgX@AjR~>#Y#Le9#Z%-w|t@S^N zFNc5dYySYn3!`oS01G4HR)b=vZu=s?pndwmMX9c?_{hZEV*{GmWfnGq>Nb^vkCBOX0Xt4BjBW zymfauvW2=Zz}!(>kUcZ{*P{4)!F~<3v+}gfHtzW0xVc#uIqCA9iZXkiM{4sb@dhq` zYwkOKp+Y(+{s-bG!jJeQXZ#a`;asVzX}<~dy>4ZZW`a);>Ni^J#X14B#+r<-v7Nwu z)#Q!_Yv$khEyu+V0DjNE@K1eH;3t8tZ1ino;;qygUA>;4XAP>pooPHb8oQN-A%$o{ z+X47@D3M2b6TYk^} z9PuuV;S1X>X4_YXQM}V7zX&bkkQflhac?9Zdorw!g@KQ9GB9IY_)Kh-Wjc=LvY}4N zC1d=!JWn7OR}1!ax{^lQ=n*rMh8%|riUmh$;bv8yOjtN|;=*982fPbVkyuhsILTq-VTGv)ErB(>;yJL)6t471MBMg}~% zglzk>$0ry*q>iVhXFj8;NvOqfr=PON2r3zwWMEsX5z}$}agHl$^45EaT6Z2|WMPIg zh3|qtGD)nfyKPoGzq8vRljJ7|fM-1Q9P{{B(ZgSWvPDfsM4m#7k*jMHA0LN@}r%iXG>6%&l97;*r0Z>ji zbtLuZJBs9vw|a+Uaj{XrI}cCFqA--TOshT5|I+?5yb=3G_|xF#nQxZ`EzU4ao}@Dh?mUidME#5XkhG7ECJi?8 z;`XdHpY1C&Yg-AD2?dln`6j!R#17XSGAow~pc2G{9)~`+`#S#49x&AP{{Rd=fu(q^ zJ2JO&TidUnZxhIO$rx!J_shZa6eB1g99QE10K?yo{yy6QnZ5Z*u1{L>(ZB z$Y`)*k&ZWvW18`$j#qd*Dz)#v=esH#$yAy0XYKLuhgkiYe`ufhNc=tF+g*3VS9cfs z&D@j9kjtmpT{XO>dsv<~^JTSbVm!4dLUI@os(62enqHlt{95>d9g&gW#7(DbQ?!yZ zgEQLwkN10y6$)9d)lrP(^(MZD{jUBZYoE6V!OL$M&8N@cNcO7ODX4ljM0|B zXyw`%o)1ySGbm5lxXo&C&GOAON#iJPZmzAZY}9%1NRme7Nnm;C4^U19ed~*`yRq@r z%IY>NDm2nGe`lErIFd2*s;)j@4t{OK9C2K)#+?txy0?b6Xt#HIGtw zKqHWqVj3`~1P@x@hWIYNX*og?wY!s=@fAEQ6KD3}uW2_nF1mE120VhUpx}K*F~`3- zH2w^zdyv9OEBS31>&nIWw8 z*e&%ar#(I~!pVErNAL83-rqLgAL}E*DNwdK$6MjjE>pQ zDmcY?b+}_Np`__O4w`xH7H>%(QOk7%Vo3xnKV~q#5&4JdB?94;joBN5DXG4Bu zeUSE0Ufh`^c|nM6MUk*v9)!!aqvvo1n0&zFu5;DL;fSvxX~e3q;07$JV%$bgW4jr~ z6o3!qUx_~#f8ewlEP4i^rThc%we6;(a=uNzp{vgeEDA{7C)s233o%iEjUZ8+oZ}Vw zXX3BhhxVZHZ^Q)njd$X^%S-6hRy4D3Cfx#z0=DOTTd|LvsbT=h;=Z3N;S5z(?<*co zD?g(brl-{3@K=30D_`1f%I?->ni;e!SioQTX7EiqNQ4fDBpi0leqw3C((V~p3S%EjHu^=3IZ1Msa+4OzuK*_!J_V-)2oozLoL{t4ClLSB41{hqWfD^1hm)GjTo zEE>D6k5J za6=C)n9vff85@H)LHQ-%Z`os3_yzku{4BBYGqtw4uV~jA#-kC#G!tB1nHJoTc=Drk zQ|NPD=GOx*=0`lUlgTHkHTZ86Wy*$QTAPiZGyHX5gFcHS&1X{y2`OEFlO^#kr>=Of zSh(>IiLYnWH4C>gO>uK0OEiw(d7aqhnD=AU);uIyD9$Uo(ZPo zdtG)vm9N^;Q77#wvF1*sqGh=t)6+<$mJ$dlOQfj9pX%6_Q0>tj}?Byf3!D;E}1Pf zgw>!`3@6j>VVMs&BWtu$WOYBg>G;>(w$`_V}^-KDwW=8aUu)Tv2YEw-0r z-rpnSB0%K#t2T*gHQGgaA%-S6cU9h0dhN*1(;~em;?KbU0E?dj_4M$^#t#i@o*KWB zKQ%4&OF8W1X!jM~f@DD=XQReh4`w;9D#!;Tn)bP(&aqgR$9P&IXWC8|B)Of$YEjs>0JQ7S{4PpEw7S zI^wi7#e69qCK{S_o%-MWk6YGa7TTxUEtoa5 z#a&r>T#z>boE`|TyZ#EvXX38{=+~OGc}oZ<(qy<-+Of5x@3vVK;d0Q)ki?AO@m~$; zZE34$@IfIvlwpVfU>=wQ1L{X!^pV(VR;>DLa=3!@Cz`7xta6C+x6D9xw$66s9A}EG z&nx=OBN)3heD)78jVVeK>G~f~>i+<=4wvC=VB1?rv(0HR3bv|WDg0rZAZ{npzO?)#m;c-pW-0CET0@AUh}KXMubL2EF3@ z4-dvx0ex{CCT>c^9m1C&bqk(=^~P)Z*Zqk+1K|Gvg?|ooZw~lE+ACOYQ6#%@k22!n z^Bj`Qyv`IOce2FW7F;_`5vc5xypTK~EWcM4DEb zTSH^^QxdAAQJyVCds(7=&Sc7d^|O%Jz^|0PIC#UrmR>Ts(EK)$&2QmLXeDMY*q2gC z8DofJ?--1jEJkv!4gfj&@Ak_5ptT=_{xa463-~V0)5Xxt`hDEd1_IfJ3aoBG6Hd{x zuyOK4tO&`k$&GU6$4{}3!nO*G%OINa=}A6(nB)RRN-l7jP?{EOPuR zO4i$DZimV6FAm~rVqRyDZML+(p7MV!75z?j$t1X*INi%Lovg<>BpjALvdSf`kj(uYpC^Qhan6&cR}bDaK_>s}j)H2pXW@0ZI{aWLovYtS+G z8De`^^WHY2acAvVYd3UxC&mp>-T0CnG%+H=+9)|7^ZZ@#uu z^Bo>}RFA!w=7Z%P%n8p2rF;{r>USEpqc){;_vdoNdS{^g2Ney5p?{^`MX70$yc5EA zM>PK~DAcO7=c^HT~lx0f$p1vM_&PrX6?!Vv-#;Y%bEp_cy(QTr?M7db!wh?(T zBy0pqWL$`ufNX=8CoI^&u8+mP7k&u*F|li}iJmmkwC^~XH{D-bLZomAjTnYwy9YS* z0C%t9qhI*B@iWDGM0y8_b?r+2${+&4b9-|ijj-}HxtPxlg~>0KHyjDpX`Tl zQ{bhlvPMC>yHVz@@s@s|WDUa{WD!p_%c#(%y_9Tgg{dh^R(&)Ag6u;#192y|Ys__v z=eRy{3=mtt8+vo=T|C#bPXLTyVSsFR1Ex>&t~Xi?CXj?tv~?bx^P2H!J#KmvS3a)& zgEjk&JNARIL{P>|C^-ZHyQ?W9IRuV5ubjW&yVts|hx>f$+Lo1Y@C`#j&}Tb)=H68A z&l@Y({{Uip@3Vfo>CWSG;SB~-bJq=g zpUrF4$?CtZCEVi7p&m;!@doR}I_=_lS9g=hq@1aC$3DP;SmXv`*yq1L)}#neOw@uF z+l(GM@n4`Mcxp*#hyVbQ>reT%?LBH_w|4~NpyIBLKH-!zqXCXP4Aqk=h|E#hMJZ4R z%1`B8Ew+VoD~zcb^&Ag=asGals}F}}Z;?|6JPv<6^fjF3lQDJIQ-)Zafsu~ijVOWG zg#)d6oL&*UmOShrNzc%JRnc4gJDN*&+7+@@fyu`sj)SMJaac|LX!b_`())=7@w?VX% z)2M6=XPox-tY?@)Ivo_~PI{jwri(L-x@-(MCp>eGp1#$w2ZowKx*YS!&JX4FudXNf zNu=pE%X_Oy6^PshZl+j4WBfn7AP=T%+Pn|@F?i?3g4%BgczagYgsgD-R-1dMsp*`( zq%x=-!#UbJ5t`(yokB6@Iw|FKC-AGEpO)SgmDLa~c-yxGk(1m10I!o<*4_-%W|di@ z0zCf!7#%*K*X_rE{{Y~V-?rb2H4zq>@UvUAf<;#s8i=y7gbd)9`$X$1fypJK9SG#{ zUuF0Y{s=MqSL-oJ{{RVf$AvsU60!2w>J#71jDSl+YiNs={vamDOd9g5@lF~l(waNr z!}VpaDg4#J@cR2|?6(Q#V1_C==cZ3V>C(CzFNRmaR#>CowBzI}pQ!$w>-z=xTmJwC z`1oT`3I6~Iuf)kOrV+EccymgM*@*;koBNy7kPZRz^ScKe;C_nyJpTZK0)EPRSxc+` z00nrT#Z#aN@2F}wNddzt++nsha52Cq1xoXQkWF}4jxeIMif_>#g+3hRUi6YbfW`Q0 zZK6PK_42lmf)vei!X)H{`|q?Kc>33Uq5LCVLf$*OLbAm&uad}jMIxSsWB`Wr;Gd;` zr9Z-N_$5!le}#9)Q}}Q2XTn|_x)OPU=`E~dn6M{kc^NLHWx>H-I47Lf+m=7G9=w)B zYI@FY3 z0P4TtA=I?%Eyo+Nw%S|CN6XvE2ZP0bp%vG`kA|8I7gl=z0FF$NKg`j<(I6Q6#3p3q zsKFs~)aNy!;r{>vd=a*{NIWxPeShVSsTI@b9C8SaPY0$k?OYFuaV`6)Bchf|hl;vQ zpT)0${{Z0MZ;LlKHp%fb;wOf@Jtow7t*mVBbyqu+aY$#9O+G!!4<(`_)q7Xko&o;= zf(L%eT5a-Nc*Dm&De)|P2Dh|bKTd^#QaE^^mQXhp4t(vyCj<}x{-o}%bpHSi$cv+R zCUy)8{h?#GDuqhmFpJHRat?A=fDU_B_m^d@z>?}3eukyf4rI)ZJwe|Jlj+GHO5v%T zVx{)~0I$3CJ7I^T1*Ctk#QZe)dH(Nf z1Gqtq;8*C+!|&Mhz#j{A$n^gJhJF+9zl7~#`BE!eOGtt$frLf3iC7LX)HVR`Ur|l) zBf^@Tl3rg&2#!U;j@m3L3GuY zo(PPSfaxG*lt<-|j2NlLf18 zqUv$arC332A$3zC2WCGvLRjSTYy8;%0N|$)X}ZVl`SF(K;w5r?W2&A`2rDr{eo4P((*DY>`Zb>AG9SkwV2S%$Q_VQ_0)TMh*@E=}=3hc(Ue3vDYu!>Buai(&Rwzw}XX} zJ-?Cq`q!y3kb@K3I4p2M9P{cx{Cd-*wU$Ly^7zOkGN}ZgZvDBh1yYT-JH4w@addm5ik9!#y#P$F&lNDRF$;BdU&T)NRuGamG3Nj%w+<)nrrV zMyT1wSDrn8{dMN|UNq6XF{l3kYRz*k#0`>jbf&80>^`EoD&V0h za~rAXdWiAF0vR?+%3B}|6UKY?1|3iKDTVSrEoM$nlY zA9Ls9RB1Pe*F)XGN?40wr2ha9_$c{r!~XyX{wMgRDQ>(q;r&ALC=@fv9khiM5y31* z2Oj?Q^cTV(f!`YZAMrCp)wQn(T-{oRS>6kS6miQJKgA@A5&OBroN>tMU!`6+@W;eW zYeb(>_*JdlSZnqQ^2Hhg>v95ZFf>x-2^|E5cQ0ZA#e5IpkK5nl*Ms#((LN!C)@|)( zUDh{HMFF>Xiw0FxjZSlbM4PdU<0m!dVY7PpxJ%fj>8;OBSyc=rSjx1YuS5Du_zQpG zO*2>1ul2n~D?2wwxs0HhkJ@fF#;G%e3A=A9idk`ja6110QQxzz$A-Qs{1dwPE8=}w z_01CU$s;nnNVjo7r{uWvS)+_c8}xYj{^;vp!I!|__$e2Syla1BVc;)?*P8yPZFO%X z(nEU71*m4(zD9xq>Wl#Zw%wy3WSaen{gymM{{RI~_@(i3Ujz7V_8$XjR@N69acL#7 z)HM4+cOu+b$}l9n*%|{GQyDFi2tauF+W9qjR8x~tYr8(amO7PaS`S0@bHZAG#n0JC z#eOOMmNjWDKeD`C8(7{&BnbAmmbWg@t=u_2K)50D;!J=*3~`M7MdHu+Hb3H5h310q z;Gf3`?C&E~j%L@aRpb~#4$F-!*~$)fftGBHocjL3{>Z=ZPK&Eg6kmAz_OJ02-x02E z?asxuw7asm4-QCFN#~MBk09)gL~*Id1d(5z-|$#3_$ar>583Bd(yu>c-y6%|9|kyG zGHXMn+pYD!pqVVDYrA`;RAk1`z-9Oh%tF`F;jzD1pDin2uV#4IoII7??7su!J$L>I zEB^om$@nNFli@bGW#gMWNcSUYUL^A_Wyd=b(_FV?Mkj?&wC%?w#~^GQ{i8X`VG#3;@fRe-&@me{FR>G>Ug5LWoEW{AUnjXFflVG9ANbW^gH6G{1ub- zi};78PiOHD;{O1QwR0MP!qZaJEQu$se!=!A`RYJWG1~_f`IY-Y{6y2fBEFmAIOe;S zZARY{O?@lePb7-_H%Rh4j>PR8vlEWKmHMBDxOu|7xt8es?~E-fag?3){<|L_YL;Pd ze_FD^{@C%b%_jC_Uh3t5V>cNV*y7$cR3*OoY&|_!jFo68u9d6o}1xI zvo*rosg@Urhw@qQf;O^F55Q!gpUI@~6fA5%CYh{{Ru( zcyGiom@aMG8C>mUKXuUYlwr!B!1Itf_=q_59V_)u_Mq@4nd4ssUTer5wM`%kfi%RI zcocr|7aqrU4<9eBepfs`RD}<(7_OX;o^Vwb!HcEIK2qG#DyU`ol!4D92h$`0`BN1* z&tclNEN$buwU#z0n+E;tGC&=NAPk>hYR(8_lgnd}4^iLT8vATD1?>~I=a}cvxo?y5 zbO-s@?U(!!xAu7WQ}I*c&Hn(8{wPUrdmo29r?S&#h3Af2qIZd=8w_D$+P+ri$$iAG z05$m+@V?XG&F6~YgW{F-?w_OWW_fI2ONm1+HjR=HyH7bdKT7_Q{{Un8e`Oszq}Y^oR!OpH~i&^E7{?;+dv zrcjDmj@27R)f^6h3jC(O(O=Dq-s06nPYg{PFd1bl!(;*1uUh?J{{VuR8IQ$(586tf zB)T$|Aaljb;~D%#etlnfY71#%5zQp+BoQ%+Mnb5}na2Yoj)3!D(%d72r!2CEI=-K) zKQPa8Ia0;cjsF0x4;7Qbka+;SJ2!4%<0Bv(V0OSh)z3-ctO7x7o|wVteJkJXtkOGZ zpt)SB-GPi}j@|GDbN9M*DK5x;`S*>ZW$n&uLMO8aCc%tYI#}P@# zLUy0=tM*a0c!`9gn~D?OG6NPb#V)~-oGdy&tf>F?xr{t>%ZGCKtb0|Igb4`Yt~k3(B!n3_1L%A{VY?0Qqh zDpb^JX`$2jZ%(+g(arM$W8P4bM+Hw$T;Td*yUT4pIpH%BT%7GA916SPof69P!t+~8 z5EDO}wo*?~vm6d~^y3tGekrum{hxk;We!Nm7-a2(>0b3Et4pEqnS|V=>Zhrr;7ON9 z@phemD{UzzH+ND&3UT!!zpcN6BN{%FG)c6{6iaC80KWCNbP>+U#0x{YgfJ~N1*=TDm2 zBlG+I3w_{?E5kl6@P~{v-6GZwu~RZ zdvPOzqyV52#g8}~_4TjmOa2SVeWp)o@Z&%e#%_(~t$83;+(cqHpvXNohLJ{jujfj` zPixzOZtw28F+ zBkfmVWEr%K2YDCnvW|9-a(MRZQhZ3&rO`Ae(p@~Oy;O-9WUC|~=NKQt+zcMP@m&eJ z)9r0#n(AR4y|iXvN_BE_+j|VL`3EPCJ-&1CF28r;O;z+?FZl%-?uDN>AMp@ok0n90beZNTQ90CR!XixxyPo@!iYuX>#!Zlmc|63%&81X4*q z=QY(r7eqRUrG_2)&~uZT*Ih34utF2hVOlVFdi~o44hLQj^2a}gbbBP0tj<+Y%kAxx zU#b59vVN;;seB`kO0$^_)|-1VxQjS+N!UpnZ9PK(z!BUL)K}+?r-QC8vD5?ihy>#u zYxEoTaPYOZi}22CyCm}N?{24ChdE-&3vG#hiU~OBl6#u+Gx|v>{b_2xXuY3aoUvlZ&a2s_It=w=K)$a=O66?anC&~k<_$b z5oz}vs6C*8+u3AGJeqm9-;YU-* z;~htC{zv}+1&#QvqI`4xtrA}j>Dq;d{3AXe()1g&)SwVb*Kiw)$09w<%BvOJW9Gva z<vj<^KSJbokfcAA-Iu>6af7{0Xc0+r_VNBe0VDP_(wtY`$@_NUh_I?q;~|WSlUJ zN3_}QQb{cCg4?y#(2rxv zA$Cym%8pY20;`qR1C5P5Qgt*-t;dPMI+T)v)c#7Xycv2_dC)i@_X8&*^~NiH)4;lw z%q=KFfj0~gN%?y5{zf`xzn~8T{{X=ce{ate#4qOfc@KqzYy(N+O;X2CMU93_D7TXN z+tUUxeLXAhFNQzxZr8^xYD-&>9QfPAdOOOI`PRC1mF}*7X6lza>|@3`1jo)R*{{X; zXg`YnhYda*t1pRI{!p~; zhw|&~U)*=YPxv5s!JJv!nmT}o?vONkHe#IrzQt2T_Mh@K=L% z?P^Je=w8Q2zM3Z)=l9D8kO=F`oQ#}fn*RVypNIbdu%E%Ng4WW-;Y}C99u9<^#w{|! z2;Ezj`^tf_f_=?;?Z<%r7}+Y%tN5-aSV9E0NaWI2CcEzWu9J-Dwjrx?|@xevXa zu*oPWsNKJp`~&>&J{^C-8NY7-02e2pFT)#+9t8lr`qk#2WSet64<7H4ddcLK*BQ_XsoiK}0EKWEW=9cu8#yc?TX;y5FcOE@`fWc;HYE6T0J zxT!rFUWcI!o`P4559I6Nzx*4;;Tzl!6!_~|)EzvwbkQ`sTU9HP-EWP_P)0WeB<&*% z2(Qqug}?AYAK7Qa&mG;zfiFB$ETKe)#Cn|5z{*Mc_F;qrBkyfNp55#Bz7G}rP5YWo zfuPy>c?r3TvE#5&g~!+Ot0w;d<36*9KTQ$T_NmPH4 z(N2Z7ibuvj2|wVL-wJ*dM3HzGz`hXhqyy&LVWrzC8R`~85Fti7dXvwu($W6TULMe` zmJbnluHMQmvKL5#L>(l?Lx*6D00SQ&$G-%1*Iod7v<>2YZua>{0Ly(eTeosJh(bD{m~5o!M+{5bh_b2j(Z#oC@hQ{a@i$j>~W0j{@H6EdavE_Tdmeh?#XUqBrklaJn>c&#k1SaZEmb1nUI6J)?{&# z2S}y}!<>QhbL&*OUX~N1=FJ;#9DG%11ilvVzMZA3{n2$dDi3m&;PdNK-haZ{*DQ(F zukEa)aniyG91c2Wa-$=jNc<}r&*Gi1{{T+7Tg^T=0cJNA?Ggd&<|8^4^}`Q()3Nxc zXW~{8XG#fL_zwu4S{Yj3B)YYc1zMIu0?5!7KpQ<6X* z(AP&DyT%K7rd}{ZrtU{@0UVlsqT02Ei#E$>Lauzt(8n3|{p5!O@g}}n@xSeX`!RS1 z@vS~1d{pqZm2jLjsMTPaMPZT5aqoQNx8Yqeg;{9`xAHlqC@+bhRIRWvgDi(=9n|nW zJ#qNen|u2L88=(W#_K|hs#OnAfo3p4u({5q0Y?0iS6_=;r&2=v`5Wmyhz zLQQrp;=$~9V2b!(;y?TxL*YAXc|YM-d@hpbfH&&b-ayWAksYK{0y=T$4*b`CA1cLE zx4LrA8%8=dkLarNL-4xEE2W>=WMe%wFBe^zJRfhN_-AxxD89>SB+jZxZXWL9WM|_6j2Rz8 ziuhmRH~bZ+_U!R0-`Hy37Bw4K3}t3mTE@;rNj_hh?X94aJde8D)wB0VuHW^45xkm4 zamUk^k<|V9{{VuV{{X>qJ~#frU$kTX&wc{%PlkL&C6puU%W|keKoVGRG{dAu0~-jH$Q`sV2Qzbg9vb zzKrLo%{>x4+xA!eoF)CG{vus?H^g5Myfg61&eGK`Wz~FDGeZNbN+S?k$G$H*;DEl_lBX$7q z?O)8EkN*G$9y zxBmbHMDTxrb)(}y9(+rm#a9|jM=idUqSy-Z$gd&$9C}rSkhL*@g!%0Z>jq>%tF_iC z<9*J9hA}xf!NI54-RTLaHNBF_Ev@MCyvNVVk{qjViERkl#0FKBm>xG*jY^$YbEZ|P z-$T}{JYC}%Eg;t~XS19Fz{RI56FiNAPT}UeCOFBCSr`-dO?if`g@QGAwHzn6EVM1c0cP`bNUGcq-Q_Kk1)p&lWa%b);Hyz`9q=uC|sy z-e0=9xhKsrnB(N|QXztW)(*86sjO)J8P-!={?4@1V!u*MGD{Vtt{>!>EJ`e<8&4uf zEZx*{D;CdHUkPcqej8iaXVb27up4NnV3RX`?4+8FywSLmIB6PSssP|*nyFwU_GaH- zR*%-cpobx$av4min$CkS$KvmYRTo&E@Y8& z8)p9iP`WZRl2+daBv&{gk^~@(@4>G;)4XqIbpu6Vd=94(V20Sc-8fJ;$K_egabWSPOB)n_9G~!C{{Y(4O#PxiZTtTKhu#pD=4}JvTrc6MCAyui zCAid#i8ik%m|gxvnmBD$H#e6hys9yr59Wu*my%p~ihblWmS#BnyMKrDuWi=+cd2W4 zS3V`x?;i73)Mk?A^7X#-*EbRwX1O4avq;Y)alvC=1>*uGFwE>3_BQMh&Q5FV8Wa+Y@e!V~BZvU)`M41FlFu{i%hCIpeQtkw_lDhkE)>W1U9Bfxy7Q z&$U{%v}p)G!cNe0&V3I`t`YVYBkR-r;=NZv@R*-chC8jT92j}pR@oA+H#S%TNL2-o z7{ECoQ%)&fPh%>zHj3(fDf=A$${!JbYOjrc82A_P#3SpTB)&^kmBKt0ah<5Fhl~_;LRL1jqjXf>rnn;pT(jX*>g}>rJBR*0u*wp3e2cKtv-Fb7UX zE6)B1d_eyIg>B>8eQxE7+RP2S_$P?ikC@}AQUS+&S9|b~K)(w-Uw?2Fmepfs8TtNX zWn3?&c+Gs?E#0iza6i>BRyi2wr@vnHH7F^J)X;Jtor4^X#XHaQ?eF6T^?MvcLZx#SrE3V8!2 zzf*r?pZF*b?V0f>~4Ow36ek#+XPqA3VeW@E`HsE~OZX_QpaEviAhb-JLAmiw3uP zqFuaRREBf&2T&LgLBSRH{rh5kLWB0#{gS>Xc#Gm?&7O&Qb#HO28SY2!EiY~HBxW}9 zenm_G7#3tFs@<%3h`5xYq{fx4ZVVfclHMuT!GgI71!M8GI(=E(kwMeJjf!8 z%Mvi!%OW{A191v5j&on9a2;B5#ki-fpXh#h#uVy8imJQ*jPXAhS{pmow_vJOitsUz zFagK)?_3?g!ia`>>*?)XKC5=ulSdrk{K37xY?Fx_IS13yxto)M6}dju`kb^@K3f3> zqN5^{tFPSXdF1-l0*&YF9M+g(%tLT{=e8=vrIOxVM-)X&j>@W-B(oF19lu)Tl?5hx zv8&3R#`uP7rPFlnDo^!FjWG9V7>xVi3<~-m_C@fO_M_vAuNgyeBoOG3jq$<@G_7>5 zii83FuL{iE9zu+F9tq>SWwr5L-kyvZEZ@ikbix6iJAB#WwSAxa6(`yBPZC`M6v=UD zneKM1gb!chCchfu3f%STNj)!bxb`wwd$4VJ{raBy;!8<|jPXr>GDUE$69;lstBf3i zcmN;s#ePz1x@5Xup>v_$pWfO?>5w=jPBIVoSaLrK{fhAd+fJWkkS=l1bdheX%^Tu2*7sh`}qD=+&RYO#yxOryMu8n+RB64M}BWf={ui4X{8{& ziKcTJ1_9YdNK$ae2aX0!Z0q_7(r(+yNkzU$jLHV)1Ha4i;POU0el>?2r+99d&I*N&dD+ZlumMbaMo8nFSMzj`?{G3byZ#mTU&5c*C-%1Zli?dri9ZT_ zEZ!l}ZnjAk=AU_EA_LT`B$K45>P8t_FhLdXU@At9RW%M~{I1UjJ)s(u?CNsATQAK1 zt$*O9UNHEp@VE9K)Vxpd&cpjNS@6Z~nRxR;rs@-FBo@~bsZ3@xSpzX70GT9=9Dglw zf5BP5Y-=ef;xB}CPXTz>Q}9l;wB6|!Q`k&UVN2P{*6eZaV_jKJUdU|5dilsOF6vOt)oMKNJ-HzWazMh}@4VbTk`tU_A16GQR7uglr z_KolS6hrpJzZh>G>dBw(Gzfn*foqk&{kVT?jY&#R@e^K9c=<0{Kg(cW>0c|5Ito8b z)kd6GL0tO>MDe$c{vv4By6?m-X8T;dmRpFH<50Pl=2eb8hhTYGil{6|P(bAI&0N*3 zZ{fYSxSG$%-xO^dmLy7|F~|g-I`rs!QShe0Z#)B{v`7OBi}h!~^^A=F0G~?hJV5Bv z#r}&uysHd*q?SO@%0y~6#F;xoA92&Ae?V~iwhD^=>OUXNS2ilEec$GCGooHYvA82H zKLBz+&wN)as2~9D#@)HVKGoPuFp%V^{{VOsgY~TaRtTa51=!gH4th7Y)0*~=9+d3P z4hY4`A1PilgOQBabKy-E-tObhy?u}_=^csNwIfnV&Uoh>_V?*r<*YAvapgtS1Cjo7 z*Y&SY)pVPUE5LI>WdXUpy+pE!007A)!vVE((`W>ePpPbNQ*og=F3j{5UTV;Xc8vW} z-8>|dJ&Xoh6B6J%77Z?bPELOJM;v1r@5d3OY91r;RqfS|pucOliNi_INCW5Y_hnY& zXJAG;S8c9o&7|Af_-x#Lv->x8={Z%BQhsB%(LU;v$sG5uF|o0@wedqu+KN8k9K=d+H^A&cgZM6^9O!-dsJRPK-PhaI;%&Dd6{tAmszA=mW0=>fp!;_uN3lIV485un@UT1N8rQ1xn zw=!gBxX(Q|b;t7hSF->sLDas7o04*|^SGCGn>ePlSXqGRGmq)beuMtOzY?VIRp*Id z*5sONIBb^UD@7wODlrM=?%Dxtka9xcWZ-1i$XZ5^ABf#;tu6R-^I&9vypnqK_4TbU zk9XP^!S4Yl!&_-yFA!_(6~*j89k$wY5<*JAXUw{8On<`)ox|o(O8M8+;Oi~Xcx(Rxqb;zry4ZXqLq(1@h-ml<L`nv&*$J-nzJwF~f$ZV%!l^Mqs^~#MChnltHZCiE=z}!I# zpJH*J!;1c#ejEP)!2y43ZEhC5@hm>G}M*L2S!zUTOT z{{RN;+NGf%6Fwr`#@mx;_AZ7klmpO-H7Raalh^lrsq9E2!NqZQ8Vd1ue|g^y8Ljr^ z{eCC$xnc04^t5tHfj}55&Pi|L1Ot!4y$??KFLubUC7T75{{XCz5}<#*zykoDNzXO? z)p#@h2tE4|cyc|SHN5c$iS8m*%iL=1VH=hJ6Xj^)*x<0>93G>lbM{Yw{{Y~dfACK) zhu4d*cyHmiht2)5&)RNuQFlIK7&*DTk)I@Trvo|q*NFA zgiqpIvti*qn#AyowZ4=Z_H{YO5L>LSdE)^|&lUCm0K>2NCYSAF@fB9D~IJ&*P*stsEJqmd>1oc0K4+ek1 z1;1}k6k8^x7lkz4Ffk?L)-{-5ILAM`vWobU2;^iBt$Ba#3Huy)`}QXBK9BHo;!Xaa zd8l0J$*mKnd97}VcQmLX4KP-bl*Dqt1vwl72(R=kUx)gXsUdsor$){;OCm8qLaxw< zZa~gP(oKIOKk#0k2t(lS+b=`5g&8An2kE3KPzYgjdO8og#y(Mu5tEW>VKbaXdGb|y z4!&DP5psgnABT500TYzj)03ZXexCe_qxMtr9vXxa}8=Q=7V}a>kz+dd$6rb?9baf=6 znTb4r1_37^bUau0hyMTs?eRa@J`wBB8UoLGY|9`6a)TB#o=6;a2b%FSzj>y5SQT^Z z-A_{0{28gi4}knRr|Xs`Bj!&X#5T83?qP;#bqjV`7dxBfl?cW*n(91Z@e|`4$7?5p z;jpz!c*4sRcIzv%fxBkaXGK;R7(#jFPp3V7;+r|Hn*Q!NRX`zd*(Y!%SPn7Q7y~^= zTIa9)Nw34^Ute4`pO}%AlX*E{aDhYa+t-j(oOa+>&B?uy>hi;}?VcIc{wM1;!1#y8 zR+d8Li^FehrrJeU{tdS=`G*HM<#F4N2JGJrMf6~JkHs2{%D`|X0a%aX#&(i>p7rwn zp?PPks#@yaOoB$-%n09{eA$^uIPb#){_SFE$i5zdZG09y2$=e>2Yd_dL8 z`MQ>v?=jkQVInrwVpr!$85vk(mE?XL)mdNNMKt@LQ$Dw$Lm60$w!{Q%9Y%A)kLoe* zDhTy`1=Gy8OtS!Rs>;OUr*3oCo-4x7h%}K1)b&YavxS>1W{Hz#(Ut_J9ciXm#XQogN9Ern*GU_5x@%cp z-S&~mSogRY02=A~wAhtYmZ#E}0zF1AGT{uwF(o|@Sio)LYo!=K zSh~c>N+(cA=L4>B>0cS^9~Ax}ToZ7!cFNCsEST~0r;I)TGT zE-*Oxab7uZsQA~%8dd(G;w@VH#FipEWwi+-t;mtsw2p54J0-xDJOCHUiWW|H?o(GB z6Y5LFK>ZukQ^Qcox3}|KK*|`aG|Ho&y4${1Tn={RSDI^o7XA|2httNPr)s){5&5y( z%{8=hTsc}OxUSre27RmI<6&v3cv9Ny#1Puu-7W0WTP^H%x}=H11u;#mNn;{~ zIotvnfzDZpTIzg5;Ex?ycml>7JD`?!_7Mq+>J0I-2{XU!4`X*0!cG^@y7O>2egjtV z!NF@LVLVG}$sZK{(Ek9p$NUrr;Vic|IuF2?@SdJx5*zk=-D)O8Lgl>cQys8mT#^@d z2M0Wg_y@0M$9P4O9BC_218f^-mbk;OF#qm$|j?KXckJ`4WBj!4#wM9*o+jkNzb+-;*RSCAWvP<)8O5Wm#gmwVPElNQ^RH z?5dz1QykasOd9_Hgzda(Wh=_RYP-FFi6gYIfu-}2ytVn{mhXT@(>g!0uB;=NfCFIY>}f}oJZ$uUZV!No_&a#(c9z^%jvC8%$*DV2|4>3 z_#;Gv#0#Noeki?1xU`iuZAoN>=ZhH1MRTLdjf3|fft+XOLPtMwcq`#o?CtQfeG|jF z7l8B~H%qo?CAWaVZ6jN?zyW|48gzRbdz{9X?K@{1OB`3G>67U`C(tzr?1Iixcro@B zOUG!WP_D`XYTIH`FcTp&#~ICH>*GVV@UETYUqNRCOD&vmU(GxW1P_Hog)eLo5m;l( zTUI#Z8*^TEBCO`GCfU-dsJkQ5zqE&htuK5%avMdoSSEzV#yd+`j3heYUR_SsFPG&E zQDiv6q30Y9lXY>c_*&;e_75o)Q(B@Oe`XNnIvu6<}l^>t>8#^g;zw)buSw zNWRlNX`$N34YZP5-OY7tZv-JXfRkou>~|`L$>R0c0hLltD#J|n9w8nJ@usGi`gWz~ zTJFAoG?@VNqA&K_twB^|V|1}cgTO{+0=Z=qO^z?cekAdpnWI6jc!NuVZA?JX+}Yb* z>2~*%%V@-}kshUO4b;w54b0QWllL+(Pqf=emzvGDiLI`$n#%BQl`pI>t|D0OlO9-- z`e`j)mG*#%97&!?AOLtz2w1~!d#!j^$hEU)jyt&HiKMq_)-q$9FWLV9vI}haca~V8 zUCYb1aJHTw@aKqpAZ`t%gtxkT$z|k)5|)9FnF(25+Tv+;f(9UwV^hbRSQAY_hcVvR zUFg<2to|F*geK$4F^FT5*49HcxyWa2E6sJ>rvfnmCA#Mr6z(rHyIm&c-c31@YbgV} zT%?T21<`JN%h%HUq(%g18_DP9>Tq?B6>1k=8nF10rin8Wubphd<&kB#ZMP5&mnJnh z41B9-Bsuv@1I=`I8rOq7GvaG`?WcwGn}kqq5#x`@7fK2<2qBMSBW(;f%RUCxIV@{c zOw*sSSn7Jk#9C&#rtH0zJe)-s9$P|AKoD6@v8$&7SuWJ$?y&?`o&Kre{{RnOTHH@9 zyt*!=s_PR*lBkB@{F27EH|yzs^0`%>JjlS92&1-pZn zX$s(oQ=Ss8lboruhIGJR%c8Wj+!m#@YiU`5^ zfct6|3NUd)jg0wjbeA_5x`Zj>Nftjc(sK;alPr*lLV>k~&D2POR3M1shjtDd+xUI| z0E2{oYinPM-?iZVoIDeoe+m2~w>B>UiW^jyS=3;P=KA2n09f%2yTNG-6cVX7Dp{3* zKiRj6d|hGiOGNP}j{I~CdIpE2UTROMrJ-~d_Ll8-lEm`e>Ttw2F4vY=V*nnwHU3<_ zKKO&;zl&Zbzwz<{Po$#86yGvT_rex2$yiT7+$3FvzYv#mV~(ZvffNwb~+ImR>C*Y3Cc5tIH2`KkWi-?cWAVdG0v zH^ct`3n10JO@@v%xQ5Y#!=)q4(ZcfIxWthe5Q{~bw*ts(@rU7O!(WI$7yNUj{5|+@ zqh9O!#;Z8Dv$4~llH%s#W(9;$#)`$Af}Yq_1=xjFVg8?A@Jqkgw?X}f{{Uny1K{_D zWwXEW)KY(Kc)7HY(@k-GJ~q2b*P49GVQ8RQ#*)O$Q4bBfpOfO8>)2BEG2TahS%Fn5 zDzjQI%>5eFJX@&Ca~FxUTPBWMHxXPLR8W~9C|hx74c1YU^k5sw5yN|`%6{wtCv9-Q{l(KT_*qYV=_HY$~ zM!}8E*dN3g^{(l3xWa)4i!|Fv{{V&wZuJIuhh%15Uj>(;BZ5a2`4fTlRzLsM{uTZg zXg(tGez4vX(j*bt-4vGexgs$f1i14LVV(!6KbJigfvI?xSkwoG;4(oRjUiPaZB55- z1cTF{0GjD>%bX75y=O}i%{LaTdvnW_VHr7HA1J(U9L6^4rw1O@ z`+NHhd@T5Rr}!Gv;|`H-_E!>Eycc&j`H5q=kNsqk$o~Ml@|<#{A3%BHzaPFP*)^w* zuhVw@%&(p@a-%$UT?qT5P+tIGIQ*)r#Y}&j+5zwR$JRABTQE_=Tp+YvC<2<57yEvZV!#po1*mbXR@Sp5u@n2ljUr^W9FAP{fx{dvuQlpr zxM~<09?ly6W?1ZHTuoIsyU^o&6Z=4Gz6#Enq}oS&c|EC$(s-niMXSxbb4LygZs-Z} z_ks|UivHgJ0N{oChr#cI{{RDL(L7V7$#;3<+eNn3jmGI>HcXpR&Ngw(U^6qbV|%eU zzyN<5@_ZiCei^&jX|Q>%%8RxjyoB^Q94PkUzkWa9l7H|}PYirJdvDph<1L+>b~kbr zu)ch%vy;R!Ncjjz`-Jduj=0alaZYbMMLtS;M(64LBau1{%C@p=?ml<_0D_Eu!V!MN zKeVOa!p{rOHU9vE^$TUy{4a9lH%X?<#YMY29JAR!?nyk9F|!tCV_%RT8*XIN#F}sh zNMmT$H6=!6Wi0%G*kZq|@Axl%oqzuT3cui|#k)Ib=CY4c)ciiPNQIgxE?^g{^Qq*w z*+^b_BpebkU&j9c#7$#b*ZfI!bs~9|;#X+-;~UDT9Ch1_=N0;ogw#E<>o=)huo=>G+xVK1vVAzlbQkWwJPfT=E z-jWf#Die{V+gi*@NASb_GwdsEgY%OnmS8NO+ytm!FwgB2PPvAP& z19NJ^^9&B$XOE{My??+K4;7uxr*K3sHN249Dd2@*1IRc%2Lsn9wQ}NT?5Rql=$Y42 zYJ9N1_FsACz96~KF7<1zN9}vsdF7i5jKOy?U^Boyf5yJ}{fvA=Hihu-#Y0WEw3uGq z#}tooY_d3y$Oj0`orDvf26JB}zl9_FLd)c%4shte;AcI1A7U}voLAVtvlN<-g1j55 z-RZFfd9PR`b40QuCzB9AD*^~O#z#GHMr-g)T~|1^`k$gu!}7G#JI!YDecx`jJDO03 zI0G2q;17P->0Tf4Gf1)4JU28tsSh0SJcXBo8=)*8gZw?g>t3O)>YjX%%<=hZxDK!l zi-VFdx#4r$r?q&tx2XNTiIO_@@eZLV4|lqkm?c=YN4 zuc&`!ui9_oU+nSXB-MNetX$babt*OetWH)tgp8lOI%f^R1zR8jNaDPI$C_rCx6$2c z(IvQ);38~Q+5+H&R$>MSgwB4flOL{p(JuFG6T5mY%4RXWk}CV*Ye^200kxRPMP2z+hgK3 zfu(NpW!5CWKu&PFjB(#_{>gdYAN^|k*Zv8C`#}60*RC|r*^A@;meVKpqhI)UR)LB& zwdCB%ZE}Htk{Q`@xd#q$jO?!_{{Vu$_y$q?YiiT!a;@BVI!==?^3ax0VUi%O0VFW; zdE=oRQ}H2Ft(4NPv}#)UT|53q%5at%czn`$X02I;*1X5*MBF0H4xrd zR>|OGV<*!G1Y_~_s}Xnt{$>%VD%d3TJvwyfp5H@XgD5Lr=jv+>KF6D9*P2{pXYj9R zgW-kb@~m+zn~pduj@*KKS3zs=di~i$uTQ;c>lt##fd(XZ_xkljx6K`JAnoR2L^kB}aPeFv>{J{tIS zsrWa+QpC_m5%mOhGA`Bmp&_=Oamo&~y7XG?8gRP160|a)XoIwDr)cCyj#OmyJog6` z`+pAAQo+)GR{M|2Gu%6;iK!cPzu&3k=?bhn20mWL?_~Gs&+@FzRLe94W5kY5a(KoD zG2hy?-raDg0G{KxtXt7M5i_nb6^Q49*Pl^eMIOh^Ceue-@Xy1s-F>H4z8@+ujxK~3 z87$u53<2&q*p*mA>m!Jc^sVQkFWEt zufn>Jx0)XlYVrR7qsG|YyaTl0F$<7Fo&dnyeQ{9uk4Cw;gZ&oY%#J=;KyE-h436#F zrUx}9tt@(tyb!}`ZJ`(i7v%+)ryP8%(Rlj%VP1UCm77qb%C93ERM*o})hC|Zx0GZh zQw%^Ij!y*o``1a}?GsklRX$TA(X;}B1a;wSrKL!5)!MgHzhhCG#9~t~_XLoU{?w(0wvYPJU-#;{k z?&k7v#{dBj7#wW?59znWm--fkJK9`F9n#1h6p1HN%;W{#5LGJe_eSCK1<2y0ve#k0 z%o^I?= zEQXKpr%Sxlt-)lt)L?tf?X<0*PVA0B>w-Y8*x1E+ zV>Q^+E#R6)GBh^uw9-ga<#%J{&N)9R9-JO}oeCXr2Dj4mJ2_&QD=e-MWfYzr?l(Jf zLB<0feqMsNH47L7AK3cy#$>#3scmS+cW|I>Z{2`TIm#6P0071geCbuQHMc{G()9gp z>{|Cz)J?s)ki^m{NaEX;XITnh?o+@35HbKAtG3YmBjLLn4>s}db*UzaS!IS++5_`| zJffs!PU16x#ySew(mX9bseATs40vNwzMGi$-bNyft^gzn2@AJ@_d&)$42Sb5ZEPMsxh??g2P1a#DPQjoW zs29Qn`GG*NtV#eN9J0Q73<$+lo^OuUR?ju&m1{E3^Q@LPN{Y^)0sEq`6b`sQA#st8 zYUOn93tp#vrT7a~wt@#XaN8+dd6VpNFgO4L?hgcg(0#$KfwZ}-hrBzM5t6MXm0EaKy_b*Y>__L(9O*NQ9#?P+JU&eQ`|CzAE@D z#2TcRaq9_tA&kWo2yN9_7XUJ;RAq8W0Y>0|0BV)D6N^W-%d6@ZYS$K)>1S^umJ1dD z7#Kf0;0$xf!Oeble%qh$PH%uew%F6WSMfqkTS)OPoon_e^?e#>P0L!y#p4Yf#CI0+ z1Z~WW>A80{8w7*(-nZfek~rkB5?Tw8+f1sZNl~|~yeP&q_edj(;H_ZQ6=a$Rkt23v zB3uwdC>*RTuDiCAg4iw2GI3f)PX6xX$u+6|eV@0F!Mk7C3--71r|j9IYL{LjvGA?z z5L)Uti3C>{5!~3@U6zqxNa6(=c8}ym7*KEtHS(D8Br|Oi!y}(i${X;=BELib0N}5B z{(-Fj0Kr23B+sK>L8!$Y)r>DSyR3>Pk4?5oP_ps87;StLf-p|tMSg4Pt7Uf}juJ_c zv5b;Weh0sNSJ`3Hmt@ChoUw2h3M zl1Bhn^jZC}sp-?r9;oSQrY?<6w$x=*DhP8Oq@HW-=MI=*o`;cKF0-sS@pL{O)UP!7 zB%UR6Z+|VuiGI6}$YYM`-fuBDB(l1&Z@um2%Bngeg`LlC(7Zuv*3(|cZDV#$_9o^j z97IV9xmeXcX91f%dmIjH&h=}(Fluq>sMe1x%&HVZ>{(ACqfpM#lwbpO-N$_5wbh@C zEHr&vRlR*y9WGd$s$5wtvAF*LS03BFD(CGkhwm$SB0xIwIOHUG_3aAFM6kP`QITZx zjBwpWe!g64Krarz9kh(Qh~_pf4haMrE{=#{JDt7fir}%=AhXr&+ey5+Qe%6llGqh| zuGKRZ40rw9WB>`{Bvw79m99&p-)gCIb7N^?43e2HH#Ls%Pt3QL*~*sg%G-JE*58Nq zLE(QC9Ya*KxY2C1*wN%?w$*gFz==$ZWSR*MmhqBFW&;C|!-~e!HAU6@C2`^{JX+rU zy5<$1H=Drlr{1kuHauT6il~ z(tH!E>l$D7t)kse5?ZW|@xZApg^ZE2%dSAUk2zh4jzAwSMl!6t4Xwm1ye;A8w0pOa zu5JU!mi}zhoB+p2ytA1Ys)AYUPX2*8LsoG7d=x zT6l@aW??KmCGjJOy^i)aSno7VK1+Q*6^2`xOg7>sGKDWBy0x{v$I8T==`bW6%DcHe z?AMd{i^mq%-cf60xsp#hXVh@OjbjGD6&V>*x^kj$IN2Q4={xBG*5)EzVtYvwCNpt_bIPd6It;IVYAhtrltP z);j+HgY=Cy`%t*Jusdc}GYM@~T!P1WOMO9OQoRev6Z@@$isvSGe_BI+Ov*k?#sADsR zwbhKM;Qa7}8?fR&O4ns9WeM*iina>0HE8t>L&Wpyem%4Bwy}L}b8UTlEKrFhnj5gq z6G*>j^KET>*BH*&F57B@oPcq+;U66QShc#ni%{0}AF@MdEUGo8g<%p#LTxvie43(# zWIRA36EVX9!4;9>FCSUyUmJAo2m5N`#`99W5Zqeo7n-1X8Dk-QyIo2XV{{uFM{6Vw zGDu=U2M@qr81ZD^60da~FT)R~y!w8lZui34!op@j- zJ<(BW=M3DF>~%gC_?M}8$4A$#=8oG?mdfVUXNN$%XKX`ISl4?3-_lx`)r)sB9UlCbc$8~P^)^>VMu#yyzs^;eC*xks>(`BGZ z6s{QW0G{{8{x$IU@SU!gBuOT-eQ~GFWY@ahp`@%fh^m)Oa}D%U+71B`*+0!AHrJ9f z`^Cm}ZK@TMHng2f!S{a6JOxJKA-L>YsM_s&UEDdnp3$gH)spC1cyG#8SP?q7R z5?jfwPvQp*6om&*)|m;D2kxS@YFFm~3b$C8$ErHLPFR<~;@icxmNrvaT}f#K_swk= znG~6Lo;iiCp(AZAfKp_SjCq9Q)!22f75Iz6nqIx8PkE^YhB=yRL3Md#B(q{3KQb%1 zAU|-15L!Fn?au7uBf}aFx8SW4#Uj!v^mNqi*7?@qWbj%%u`!R%n)gLnV079-mA64B zA3Cy-Y70#t{u8fQWs7X+ClRh^F`%oQ%| zZ9Unm;tv@%oAD~e+f}fRdyAN4k54n{vCSkifTwk@lW#hrXC$*rAi?=}fm^ovKB*sx z?sR+kB(;59y_BD1wY#!gmydjrjrxox8-jjpJ6r@nNfDBAIA0LnU0!@B)1dJ#qHHaq zd35WWO`=w)85DXq0oLti9vedjW6}5yC&vB(UmVdK%HOku0yUQ&jTxic~h{cf?$==0t z^BkJ9{{RTZvwTB~#aeqxP0Oo8CAGea(*FQw-6EGr8&7*pMhMPEWpO@0`?<#4yDbC5 z8fS>^yfNY_ZLVRRrm$G=F0Zvu-!>Wvqk~=?w5 zC=Sw3r(9lJBtjxcjUtS*bygtl#IQaa@!y8oyc(9?A=GqfjMndNvG|fmgMY1#0cQGq z@w99*&g+*g%Q+l1Qp^uX({#`HN4#Nq;U5p$e`ea*ZgKl$sA34M| zA1qg!F}1Z_MU=nD>If`P7$>!P{D{ckAn0+Nf1b6}UtC*3AC>`>K6v!@=iiZB;9%qr znQ{HnI(M(q;IwDX&}DMqE*Jnta&cN#mo}nE;=>nW7bpWa(MR{Q_|+0ruGo&?M{(cN z{{XLFW`4#00JDy}`&ItY{tf+}d_8#{n=gztNVItGO{9{vK6ElXlkA8y{^}%l47tI> zH_95h@mP;%2RrO^!k;{pJD<~c{2N>Q63qvQwLkbMhs3K3y&FXE%wO5w54*qeAW8KL zu*r3(z_+N0JTOE>yi-9k1(sRYaTWcU&8S4P+jyHmw7AqQ@0F#J30qK>B1yxGmQU?# zV6K07fFda1=Z-o500#JS*Ws3l;lB-hCALjVLh$~d4wo3Qv{a55E#gS0g=2>PHG7Em zZyiRn{X?W{!+pnXJ;1|!6b8r$pHN!-J?~KbMV<3IepT%=L zbnz8kNi*#5xJc51giQ-5^zRKwZ2Uj}03fYZGG^1(&MS+U!yUUnvu-UeBMBoF-S%c| zZUm0CV&_5ln`?6=sFKONwkwl!tuZ(_$vBof?4!`_(wG(coybGYkfv* zDW-wtYJVSXuua{C$mg7m*{)_wm@Fh~8LzasUJ_$y%V`QwuLVdNf&q@03co?p zyyrV^ZlZtx*8Ww0WB9(!@R60 zF5o@QxsP(jzzCXIK>*=L%)kcg$8pHVZ>4-;u=r>2kHg;#z7mMFKNxCo10l7HNY;}` zLv1W~29%OPQNY37fsFj|@o&Np@N>snCBKNLI;OD*O+!&~&Q?NY+=;o)?Cnyd{vy2W zG$ksMsV_SpL5ao1VdG9Z?Q#DA7W_A$-{}wI$TcSt`M+g`)?f>|#Kl>>%;k=9dU8IM z>K_TdANW4Q!P-BI^(`D-Tx!uup|1dVgDj;3B1pSW-Qj?4;=BvuJ@1OVdtiPl>f$(b z`EHuV(%v(X7*JfUJ79HUa>v*T`xoI|h12{l(IN>iogJ>^SI#!Ovje#Bs&kdcP;1q~ zMcUGLI5Bl2Qm-;rx$H~vQ^!zPjWWXe*G+=ds-!lSkw%+34WnxT?oWTpv2@*A#@fE4 z1&@d)8jRCQ!^-*>Dsc6pDWC=#-VYM=7YO&{{Vj-j~~*#ci?A+ zbp2D}#f8SAo0<^v4ZsjgYDc!*j&gIKt!bI!BTkI!HL_=RRfcu3Fr>P)VR)y&UM$sY z?libGo12-foseY4%o$rdOEaM=M=D7<9Zh&oh&(CdZ824K9UfarSgs?L(7QHxz(V=U zboqyB{Y{6(8YhQ7A^neE#8#K+A@dL1jzCC4+2}HHyQt4RRQ~`FJP0o>B$gD8<7osT z0}cj0R^WoCxI7VGhyMU>UUrh){)?lKB^4;kRDRreH}-q@xBF=S0Kpt}{{VshD1s^c zZKY|k>pCx$a3%?;>FOqmCXy2*DZDPnM3PPzgT;Re(rY{7mrT5f#TpVBPs-?{kgtpo z$^2R5FgfH``Y3K=% z!lOofysaB67M^a!#mbgZk{3ALfxyqddi1}ApKI|JyMM1s3X=u6h7~QeM;ee@8RLb= z=UzD4im4AgdB^bj`)8bc@n58Wv5$aezqHl0twT>)B1o;G8^UAFl(ek4ehUQHNa{1s zYVx!BsNp8xZ_Mq==}X(j>i+=M&n$;X)OE4s>6TA+aKM;iX;T^FbEw)6r+;%^&)`i9 zQSirx4xN2#Et|=6BZ%Y$TW-|=jFLb&I6qqZVjl#?q{m}p3H72HP{BaSP{HJkMiIGiG! zkWS*mJdE@liu0*A(Cw2)3E~X}yiwuSgpl`ED15abWWd=Z@8dpM3P$XnNE}zna7YdW zM-9}G$I`yTx4YBy{VsVG!I#LDBEbup9x|xAj~ubUUZ#czZvxd?s^F6zeyA>{!#ZtF3<8YeZ7>TQ22n#HtrOZh0AS2RI9oco^cj z*IDgnF|hfR;Ep=_8teQgc69x6I5I!de{|W~-~3!~dmDx_$7;+Z-6T&M zI(cMBN;7WSnRAbqBw+r6d{zCJv~4@$73QbmEkJ92AJ>&NoflJkgK+m&7lSbwWsvWA zBOt=Xn7I+F6P7snWA+Kv{Bz--iE~5XPlj>)N;fezrm15rMZeH)M%g22Sov@)iVh2v zEW?JtBEPoJ*kk?)-}?vn)AoMw_lkZb%i<3TTv>^9{{Z+!H62RM2{hpxjO!9h1ai+Z zw)}|1wX6XVYK9=5}z^s%A2bU9TORTH9Nty5HuXk^GZs zf3tS0cN&<|8BmP!6`KRL(mQ*8Rj+OPI@~qDvUuY%vBE(b@0GHDg%@!-^aDMIO8)%( z8T%H0!8g1(xbVlrc>GT+Wb=6q_N>`chWYN?L!9==`eMGF@aOy>pAMn&J{SBrvhd_+ zKpRDdm{emtt-4JpK7iooiu?*$^dqETewY24yPv@a!r%BM&+Sd|Emq^epA39WsUt7} zpG`JVkbC*}q6om95Y19p0V z-r(2u^KNOFMfxS)1f6S~DD+5 zpZNa(75Avud~L2Vigndz5+g>rzn_-mVM0u(c?9RO@zit0YhE1i?Unc2FT9&;p)uUj zNRz%6JQ(-BOnmapNhFLPep;E0d3~beLMvo`H~#>^UH<^!oIkPW!SC2JR`K8L-=^8x zc!R>4zMFk>Ji4s6_m>)fo8;X@rb%Q{RyjPtfmEHmSMkN-vvVb-rkE{EN|9zc!7Utt zhBzIz9uI2%{(s=2z7y0uZ~Fv%R`9-#@Fz{uEcWPgQNc^Y7TryQ7Q7u~h zpZpW%-CEX5eMRlzK4O86hbIHC{{ULEFKig8MqS9m1mm1{^&+&jn|Wr{8Pa3rlbjRB zAKk?|#(QLdEbq?j0(<^H{d)b0GxBLuyE@5J?VcUE*RMWlKo|$*#BqVR;A1=r*JceP% z908G%O+jU+Uh5YrV`m=yq;2 z9Xvs{UQs z^_dPuQyo=7EjqKzKj4(#@J{_A_Ttmy&&C60;?_i$MbWQijyAcI@1^X?F3?~?yP-mV zdg=)v1tz`bT{#nnn{6=G4)~dZ1WPbIP<=A{3B}$voADG(CvEnAYeNxXwxVuM% zT6wJDXl$c}V{bS=)JjW|?EL4+IA; zuOy1j8=dT|#3|<)uT{~08+eW>Ep1}cG`|ek#Ssz7cQluG%z?LXes&)*BeR}$~i>1xstF{aNk)zM#t`uQ^cB5v|o{SApz4%@6Ud|)^kK$ z*Hcfv@HT}o(sVr`5oEzbahDFLdWDfzJ_sMcGuyp+JUYL`&kjv^JX7I=e-)5ERQ3}l z$s+_Gkx!iQvu8MFP!2&j=Cd^~+5_SBpfYOr5@@gmNTz%Ej>U4@LYa_F7UpAu3w6N& zj%g|>YSEjUXTP5r-s*CjB)N(>3OtDlLSsA-6rzPya570@!R?-vpC+qqac-|I;rU&P zu2>Zyl^~T0#BTl^aNW&$Od8C3wqH%ShT6?e8_cI@UI{)7h>Wott!E?_w7$6?T$JXO73U zQq??Akld_-6$q$VH#;WaFmiMJWS&sq9-xkcjJUVE)Z~qA1=MVXveC@T9>oNMg&}$8 z895lQntnO{(SHGc2ir{_#m^mCXpQ&0kF-ZM;K(>p4Y4zFakTCF;2PGQ8Oid+HbOqD ze-TfqmM?L%OO*;7NUqJ)as~+KHw^a`$7?!&gRiU^H4QE$mNK#|rrjrE+>%tO$sKZ~ z*S`Q);|GmD;ITgeyg47+{uOv%#2zWRxWF@BM{lItov3jo%SiIb*ldA_i}c(FIQVbI z-|$!8+V{nAli&*tAHo)8ivwu`>a#957@Be-;B@`S1D?ITC}dUTy0@rH6&0n==@q;_ zG}{Tqz2aNM(s_vXsBJbN8QCW3RY>FIa&eA1JlFEE{{RJ>{il2%L#hhpP%w;UIU)ZVG&ydhvZ27 zqYyxIfB@Tp?ZEotz1Q}9@qg^+@b|~7;_r*z1G4yoVG@yTrO&C|SlYDgK=W=qoyOr& zMivwMvdl|{+PM9J_$mJY1OoWW<85R|@w)EILa;}Sd43$8Eq+PPLm%EX#hJI*zuvZ$ zj&cCUKW+Fs{s>R|6!?F`cY246Zne#3cpb{z&#mgd6>eoK3@Z)fNn^i|gqLmq0C-@U z%MqT^SBux=MI4rC^GTz^{uY11Z+r*fX{~$>;E#k_2ZoK%5ui<7?c-QgfF!Sx=Q0k2 zD=P91dspf|!r$8)$3L`Hifa0;v>pr7;*`Z<8k>y}tMKzkuc&3;}+V*dcy)8Gfg>rV~*9q=`_od&ObEkTrT~%0UMJRP(=>QIL-7*k%F@?fSfu+L znzorXj19!6$+j07m87>ypm03Ih%o?Q1JU|#{4L)Ud>^pUwaW(9?(bDvYpu|DhfZd< zO{94j^G7N~(ww8@x!dz#*PY32&h-Je7C-Qg*jZXfr`hQ;-b%7aMxAq@Oj;khe$$~n z-|qvQgmxRgaopppA-qfDMbNcdyNmlfRgNibrM=R1i*3jY%kp0b)kK&$F_p;xV{bLW z>Hh%moZ5U(vhZ%Lr`ze)I$fegEP5^Wp{BHwV9!2mm8`8ckZ`9mZNUZFx(^P>vs?Iw z;s&>-YFcYf9u$vJv#{1;R=GQ4=Shk0q=Rd2%S^-!jXfeE82L3 zzUa#m0;;Y!#s^((Yr=LKkHi}tKVGx9GikQ)rOnQvrRoqk+I;DAsMxexc~Cu6g>jRd z;1IhLO6M(Yt$)I6;t%*i?5`r7G=+*eBGEO2V>-Mrsd(kme8W6UxyO{Hff&ib0C_dn zgDpHU;TxS^-thQx@@Zg+3MJ=(t{!=0&gQwgd0=-%4e|)Qw>bo0@$?$Ws%!cWiTps< zH+qe%vC4}IKA!;nKmxzwOJ(~}5vT4nsM7tR+IL{e91Qc^VBA4v;V%#A8gGa73wS}(>ROH5TEE+^UIY`~T6o6MR^n}-3kwyuyOum3^=$jL<$D@E zMR?ru?H9&g4)C|bjXp~g7lmfIy;hDFwY%_rz(}JfCOZ_pT=Fv%s;KpfdxnT(CH`EIRmBj7H4%W@A4ppZb}^;xXE58you!*;Tb zC2!`sxKpRK--DKO8mwD>$$FP5AUnS7jJ*Q@a4WZ);byNd#w#BYXnGa5`aD*M`m1PG znvKvg$Cufw+h1GH{9Ar|UFvrfBrw9!vDA{bx)5mkr~D??9x&6s(z?VD!*;*fw)%Ozvqn)g@pWyeA~Cu718J95T2e8MaK2s4 z^W-nh<~PiG4-!Y>4~G5(@b-&+rD@hh?`D=)y1c#cqx`Y$CJ6LJ@>re2_ftuR>s&E;S7*4HhM3 zoDndUoZ4!`%mep_Y2W-LFg4KJ>FeQthq`UVwymNt)WP}Un?dEFV8QM@DsG|o{qmUD z@xj__uDsK&^?!*v75utrzYy436}&duMy(o+>20-TKBUCMkXPo-ZmKk|6P9sr zbH=Ffa%CHz)>lKvf(q1Rx2p{De&<%A*YQRXm8?XFB52u``FkL()dxX zzZQ`UN2&f2IwOs!ovw4w2?<@eo-;qgn7mQpD;cjeyA5wtwY-{N_?T(7n)UCMb;u*^ z)7msB2g)Y$gv>`FkIZsBLw~91z8%o?i*(eq<+-$LNs+b4^z;(Oa1t#$R!93;CjoxY z>cnyHUrp-J4S&Si4!fpVro7SFL?n3a^u1rmbdkExt4V39DPVKEV=e|p8?2fY<>kl3 zEpx*2%7x^Xd|6!I>T5RXW8JkZwJ0UiPCFRoXfwdT$gZF59dF6z_fAmPZ4KXwuD&UYvbv_Z zrp0j!c^8stI@D_n;PPMwJwnnwt-Ewgc;{;=#ddbS2-jfvY})LJsrX^Eq=drv9#)*G z9Flzb{{Vz;YrA$IHzeAa$0q?t7~VMeS#|MN(@56zol1R9;%MZD%+u{Oonp*S2?b=E z?TMvoOmY7JKJtNp+%83L9KxG9XO~TEbDjycZx49M{5_%1YokXr*A`JnqsM>YL8;qKr`+8_*2T@vq-^yl_rObHTb&f7@(>Gc zM^$jd?O=N8c)!9r--o^@*`z9DxVRR#5-aQag^kt26O$0rZ=kr+q>b51u#Vz66;3xR zVT~<&NtfYo#XD~i!J}V8YiTygX0`Cqvq%zrtcwnzqRA%WppaVb79H8t;MYtm!R*dS z(4SO!?}9a77WlhMzVQD5g#190OQ*>d*thWWrOoI<%#9D(d_YTDM0SD+W9BwM+E)j? zxbWTnqv5#^k0!Z)?Cn5F5Y{g5yfDx#>+-BkV&d9aG6^yV@<}6bMjXc-)!oO$pZG-J zk#(CH<(|s!%Jak;XNZNcxki;IzlPvJsl>SfQW#{JNpZah`G$YO`6Tdnjb!jV6Frs2 zw-wc-Eu-oBpN8&QYf@xYv$EBM_6#;|GF-?Hy8RZSxDubNv!&KfA(siF1>DtYYjx@`OB9b4sMJ~2%rj9F{ zcH1GE>%x}-r7TU3Wsypr9>Nl@GP zk{GUFffxWptl1=llpmBDnfjR?Ch*Akaq$Av$37FV(7ZJ)*C~3U4L4TOh>iYBgL|k# z*A_Q2Dm$`AaT!K%K6T;8px+yblqkX4DTTuM3Fmn~6{e}-PZMeP zS~J`!zqU*Gt~@j)xrYlZw|CkUvRgAcZNO0!g+cjb8QT8<8~j4PzVP;&b$vTnnLvO? zcdf@}i!HX@rX4>}xAOOAmEAJSj1eItB6QQVjT^*~_`gE%WVY!$!F1O)uxe4DosZ17 zcXqPf+RB;bl)}hI@nhx%x(-Um;b;68ANKq3AHl!a!e7~c!l|ctkHnuA;=hJ_*1WXw z6n0t?pn~#9FC>QE@;i??(ko#)85P;(CDt+r`T5`6%(wRt>Ocb0Il=&=c5{^+-~-d! z*1u$b;I#h$wbj4vf%|)WLHLoRUOu6q>C=x4=#nst=F{|6k{#^OBx?R;)R$7DU}h$d zIU^_Wn^L-H{PdeA%sAcGzCr4JtM)z)%T*jdJeIOPJ2$V-QstfZOA$tWAVo| zNU;7O032j@qC($w&*CXUlEAh=J#p#Zt$k$XmWQ5^YA6|>843vtxbOye^pNvZg&#=o^!jx?=b#2U|omJKULp3)O7jf|SK4kerSdv77SSb~V>jD$;hoE^Sl zbNqSeF7RrRO}s2oA2vqu~9BQS88^Oc;q~sVapA8K_@hkBvZO4!g3ov4$+Yb96IoV>w5~*WNsjI?Nfq1az8mpv zrK|XU&fiqiEm59Jc%qgaCLcM0G8x+0uaF^8@~MpmayIfSnm#$XzVPVOb>n{<>Fkk1 z6tWvjDIpvro<^$_@tFdQusOlw%wx@X{Dbtts0HVs^bTZ`LBjCa%ArpY5M zb!0*l!FPyAI3olOzmj`beG&EikFo6#1rtFmnx*8iF*&x6>{4m- zt3(FiS{&eml0|wfJ~Q#nq)TgUpggg}0wd3+YMyH<@C=)*-0vfn%dyW^J!|LfQ^vk0 z@qM!Ex>mWa$#)L_`Ur-O32z zByz?6>V?1q7#sup3k9?Ih+F7-Y)^4>b2N7Lr2-E*ow*I1VB?Hurz}7f@Q=p*6HK%C z)8Ws7z7S+)U03aTuahD=OK|gVEM!7*V?}Y-fsB*Eub#^B^Pe^K^FGfr%jv_F^j$lg zhwUHm(@W6231NBR>z}dM>8`fpPcgP$S7NOz$VkZw0Qg zdVPvbqSzp4p>H*WQSA>PQa^g%#U^sy>*vqf-VHlN_-S(-Wdq#Z!mG~X^1keD>c;?+ z&3=@83%b$t&wySChf#{=1YHWi1LesiL7p*|GNSCk?~*q3HSgrrx#7&uD>kH~h_n|& z%U((F-Uq-GLS*~Prh8QEu z7>}R#eweSKVzkmO8B!s}r4Z^a%o(_wpCi#uIf4WkSg?{S<4$l&Lw`@{Lyws^Dt3ETTe z_$T&S@b~T4sCaW%xA3Ql6HJcY=4kDqvWYcjRYnqd&pg58D6u+1tCktSB#*)A+O8KlP&Jlab*?RUq(fyZvaj$$L{iw7noiRN4wHt-AmL?@ZE^>}oa0ouc`g32R z{{ZlCzrk%2jWk-eEChJPkvTRGXtT=Dpq z@c#f?@!q$gABY;1<|~~#c8=aFR6#K?;4F_cgazjqWdP&ezes;34MkG8_=6q4fk~@}dda&O#xc8-PyTkAhBk;Pw1!Nq)?iK0$~b?hBj@ zo_Y~omW^R=KCr*J32z_F<|P4{7nTI|=ng%2?O)jwb4u*~KC4+@k?Vg$>p$6t_E5C= z)vwEU<9X!K^m{l(%+tiN?l~b=HIh;h8S+Ra8-Q>D=D&YF5_|)>T{6SL{{RVm8K~;o zU8e~Sm1lQyUvcEM#4$=sx%rwx#FLV1`aJ&tf=m1h_!;{v{=}aSt+daFx^B7SJ%3Aw zT-SBk;D&kRy}yW0g_c;0i+gm$Q7LW99AIF#U($d0Xm5(T+?O67&|c3*QqcXHYj_w( zv4AW**;V36Hjq{~izH)_o<9}OxZ-uHx^Y@3r}>|v<=i<6P_y5?&*rn?zx)#e_N@4u zV%GXk!i(J!X3pFFe^ZN2w*;Os6Gro~AN9+hP;2rp{t8L^JrCJ0_JqCo8S#qLU-**O zLeceUX3{UnwTj;O+{Tu}&9zpNP}c0s{{SlOU;xfde|LIcjeaTJ$Sk}=;w?h)ls}kZ z@}u5Vk+?OKF@yn>sq+cWa0##C6aEW#7Kv&900ju~g}hC2k;S2C7GQ1??L>xI7c9&f zoEcv|j#r+w;Ze@&*Qv`SzT|bwCWIjEC-=(#0Db=eBl8Fk*|S-9T|(316}w0?mzv?2he@Tdouh^z zzc78QN%?W<)lChZv>$4Q<(zLLX+n$v9q>rcs1@d(E&*(@XE8D7nEL?zPuHj6#d>hY zTixYH2h@^x_LBSm0M6!*!S5Yw9}V?;rSM&@k1YD!f7w#csp(K&+(8ZAVvt$MK7Pw2 zh~-{XLga118+rL#;%|kIf=g-Fc|R|gZoC#MoChRz7{?>;XEp7*09kxRrGd+>)GP=m zc2(7fuckQYYp(sZd=GJ^d`h#|H3AKd&W&L_Oa{{JnG!+l!GO=y3j59n#$L0d-(R@- z&M2WDcDH7Je5JS8;almBWX9wkgdB|5XRYa)W|^zoYJNeU846Ex0F;n&+sf}$O@Y&Z zIV;a6w?A+^$qdti6;48fz#YAR8rs!$tF2GNmNrJ+*eO&071-~lw=nHfz5ww zKd~>0bT1J66}g*0m7aT>*r#=7Q!rPFGN~J6uo%e!81O*JIIr`~{gs*G|wXBGYD{{X=(JV|NeEgw6Tew4}P-Uj5PfRitZrq#Dh{+C|OE$$k}Nb#Ru^ zr$e@7RWOw!Ax`3R%78Iimil(9;yqpMZDTsT;lre%CXQugs63GRo_mK2PLU$A@xC_H^oRh{q8%6j<;@xnYo83cHzD6oX42XCN zLk+KJWlMq{(8RSWsdCbcYQOJGx1Z-toWN}|wN$@+u7OH>YDZ8~m zbCsUyov;A}m5U^t9_QM+T|Y?pHQ~5pg2e^kRmYcWZLu&qb0Mvk3R{^xOf8}DoZk#I zTP+mZ#dJQ?Df11@t-D-pm;z46k&xhl^9Fpk9G2wo-}q0!wwfT+JY8dS@gf;b&)K82 zQJfYfVnR8FF2KEyY#nnMRWMq7UB!wfW$Ri@Z)iz%XSoqD_?^m^wJBgxr z&BdIu+T0N8bD1sK9lbtoJ+WWq)A46dxPtogMVv6VveflEILHb@HMQHu0d5#&Q||IO zPuJjwE}^{x9a3dUvr$FPTBgb2cn@H%94&$%7_4;3%M+i1^In*OmrVt@E0uk1nbtHQcRjP*OO5otavf>t_~lO@Hh@oK7&i;1s_{gfu2%ue4f z(oLmEbNkjieshR8h8P(0%{F~M4dhbBH?6FElm7q&l>Y#NQ~v#3mTcii%D*}Jf46NrUGU$AG)pfGcv{u$bgeQ3wY9UiK^zc6 z5`{?PSjGaXkLzDI=^9>>;Li?Nct65AaMSe7F5PWyV6;0{C}RKw2cYAV-11MYZd+=1 zR*|zu1fhH2Z^-O&Jx91T{N=LVRK2yLnY^uI=sQW?+I@V^;aORoYu| zC+VL`i&*%P;9UmR4O3dW@~y5Nqe(o8o(UaSC00U#GC$wz$0ojdX(QM6cV6l}LgQ1n zlNRd$je_vO4<|g0+i-odYGu?lRfx%NE)Fm=jsf)R*Xj6In18~_yi1EaT`NG;ZxZr7 z(A-<360S}*hY`xs##nsAbYtb`0=y&RXUAWTdN+sl8UFxhUkK=*5Os|-4>G`N?`<-a zVkX@TutHp?8$!NZ$G5oW0|)u9_=o#&{{X>WuKZHg5O^j}2;0h-iWKmb)&nTXAKEwb z%L6y31E)j9e23%DiJ$OQUm9KB*?6w+;)T<)qe%^n7s3<|hD8abL2j5~(GkhVP;1wQ zP^GJ>AH3$N&Y!~m7EOU^IL7~&8^;B*y~%6?QPNOZqtUxd6`(ROpI61DgOWjZT+cc@LXOi@t2A;Uk{>~ zl3h1gw7j`ta{PHoePq(bxkq({kmK))>4pmdOu&HFJy zq$J+~JOQmmtuT!ums-5H(k$bRU*-KMhCO1&Qa^bt%R7lC;ALES;kd%DLVS9ewj!pke|CqV zM+Zs@>L<(EPr&^v;pc}uIpE(3_%};`APa^|zYhNZXhgAK#R-o3-Z|~n*bT&qH)9)# z>OEgU@s^Eq;~U5z@^tMpMv0ZqkKzea&+-8?Zm@v)D=(S~jrcgnp%v6=dSiH!`&IGw zm30r=es8i(;cMTr-A^NYu!M(GiU?LkLAb`!oB&rWI?dv530%L0EbRPgWcsb%sOs3) zH2b!ZV}OQMTa6R#5yK-$K^bw7rvM7zoSS!q{$phLYqn6gN6KYsF=V6-VlF4o_P3Tbmye+v!?YgmgJt!ET%FZX?wpzh_*S;Eqd8 zeEI66oPwQ5?Z#nG4qJG~UbOgiVQb%1VDh+z_1g6wAU2eIf-#R&T}>3U^D z^I=I}@bnE^!L6(K)(g)PX+`Zc2mo8nUqrXP+OUHxj~=9xtf3jW{{U0C5uPfqguHpF z_@Bm8YW^71r@hg1=$2H2O4Y2j7+NUMtYvKCiKjct@}}U7_w8IBgW^95_)Ek3hPA8> zJ5AL!=Qi<4HI3hd?q`wP%$^x_Nt+BOj$S?RUHALtLfTCgW;VQ!}7%igt~;2UFJ($oibmP zh?Fe;90nYWwo#wv9RC0b#^$%Ac%#Ig6w>@E+8xHJ1>BDqUKf3!+!H%%@2FT50%+O3W5vmO1?&lSa;r`ltUpcn(~ zo*nxnv2JXg%@ZN#lh9R~`&(~`8gIj`KUuW4v(e$Vj^j;-QM0?cNhSUBv)SqvS6AxW zS$6*X76g(92D={->DndFkF?DXTzhG*^!r$2W2xBcHrF7`#oHG;Zp}VGc*p|?e(}Kq zw{(9HT-DE!}-w6^O z3_;?1Tg%vEk!{Q#HMKraELud7Y-YeJk!o%i9Ql$1Ai+hSr`pBPByYTOTyltatb{F=(9>3Hy_{GJndez^Bi>I1Ido+2L z^3n2lgCaUe4+n5L9eD8GoZlBUj~2g&*3CXm` zFa>j3Zoe$Q8(+g79`Q}JuAQYAtUs}AJVk90K1v2)`aR8zl0z<7sZn*cmjsY_9JQ{o zVesbrR`{FaT_Z>EWjV{ z*I(iv!kte@)2|}&Y%&W~OI6fi@q#`i1P`)mao88Y1-@xK%nYe425$JK={_I$N5Z;( zpw`|M6Hjug4x@YEwrM_5wJmK{_TGKKF1E}Gjz)uYjF z9!rPb1ll&C1>QpJ1}Ox-X6@M8Mol>;#{2X;k>JI=_}ORTdrbmsm6jVL{ibwXHvBdt z10z*ih_6q%k{smxnHjG%)1>fkiFB_S_$vPZRkQIm#-NwbzK(V2b^C&>b{} z+*vRonT&lUssvmv)x$qxo6apqRwr0nBp< z^10jqH719z*!b(<&HQa7J}eLRDHn9``1u|<Km6E1iIkSl9dw z@jt}=A@JSp_K5*)KGG<@-*=^3c#~aB38K6De!Vba&c3 z9zD~%GvbXOO}Fs?fi5E#R9DG4j|A2~!_p~$&nyAt{MGSe;ZBFDXxc} zYaKdwg7F)(k(H2O72F67H?th+E8QBNAMiz|gS6ijCAN-qh}z!g%ejk8i$hoPShBg) zv^nOPLl3&?3djaaChV!pdzro_{4$2-(i=Y%eUknLE4~jI1Xf1fh16G0nGOi$4)8k? zNUVov z`WwdHEWPk;&xM~{)gbWooR*CMhJ9}L!z6i5302b)JG&`L`*NXTv;$7dwzl9cFEB%)> z9T&t}7M*!LoY%e^)<5AMjyssf*`Sirhw10~pBZe;z zYtky)OZLADNvqxlft*Jz)$W$*y8#}>T}N*PiUQIhOab#82E1g z07e5&@dm4-kisO?zT$9c-(b_N?w7=V6^38f6SrP?jsSS)e{8hTKAu+XBVsz0t~+QaD)UDBFQ5PH;JY5PU@N-H(l?)BH0o zyQyuyh|V5my|)SxTm?zD1OSk(3#^eA4gnoG__cP^ykjidqIg%scfV^FcFAiU&8DF= za^R{@XJv94;f{V|G*YV{n{9LbYAN?EVWjua^?!%D>fY#g+AoH+czm+Wc@*9pyN_9E zS}X>)iXCB6XrlmUX)Kv147Krp{1hwW^k1?U>;>^B;^Z>PZQ@-T&{=q93(HI03TQP= zbt7E40A)P{$ZRw%IQXYhSpy`Eae5@R)8mkAjxZ7x zyO4xKhh-&!9Qt`hDPl1Zr)}AWF4Ump+5EI?Fse?L2_s0M79GTb#dsT=rdgD7PjYy# zD%S3!p88dfgjP7n!Q+qVU0$Q9N#z+-{O2SeUrhcSir{C)O8`JQ_OIWlr3GX2h}Cr@ z=E|oz_ULLTgt0o>>2XZj&a-xr2Wwp~N-J?Zq-A+`FE zSK=IgC)O_fJ*3~=OMhg_WU$eofZjARk>s`0TTG=GuMs<-+!VV8#@hTG{{Vw?{{UjI z4E!ejov*)bFNqqQHn*NKl~x6uOK!Hh2AOd2rNch|04iVY3(+Ga5<@te7~*yE`Fj1i z@dwABgF5b!sA;#0Cy4y@k|-tCQp;Grj7Z9%BDIxOrL$lI?DJXw0EGZ-0r>v_73U8T zicn7RkLLG2_awl&G*f%Cg7Iv+Z-Vvq@YU{@s%Y2iW#`HN00{)qSlirN$mL^ka`Gj- z?2pL-+0}??)D}@%ULrpfyd$LP+Qz%#8(D2Nw0K`vi&u&5R!db9q?aR4iKU8H<&n0! zNck+=iVCr<{B6;u_-8+Yyb0o~eI{rmw9`T@y~Wm#1f=a-HMg@mebuD%^#I8rQZbzH zdcA5th+h(PJI{!EAB68F*R)7p*6QO;w~>viqX^zOBuOnbBnJjT=iA+?HsUh8*M})~ zXQ3%4zK5UuK=EJg0c+vg&l%hJYr}W;Fg21&_TDMIHx~wJlq~Q-G~07CCPTADDdd)33r^?zMYORuE^Ae93PuZDW$zw@llFvaBv&ka7qUC&B*!6ZnA!y+4a2N4_6u zWs37%w?Jh1h*(47B^lud3OMahqt;AnPh_rt|Iqw>{hz)*d<^&%cd2|p)^2B;##WjZ zqZW;5ut_9F@<^NatG{eTo0Nt^aHl1X+=~24{giagOT=FiZ@h1Dd~L>=ZqrXYVcni{ zf>^QY*kaAjNj|mrCcmwGIQSFruKxhQmRh{mI=o&VmN~U6xHpj@EhIK^$hnM6S-=UF z+NUgf1NpJ=m*JO=e0H*F7E(5yqusJK&A|+!R7`-YkG?wKACw+RuNMo1r%sIdU9Z)j zPm9M+rD$n;bUo+e=frytj(#xGz7O~U1950I2*OWnrM&x_qObuP0h6_y0h5+vfIHXv zFZ?P00D>ca$6gBfZ=h+Kw}`w=cV}s7Z(?HeAk)?c3K5akI9_2Qw;N0D&u>rj$^DQ1 z8+<Dp1d_|<%&32GWa`U_w0^(^FJ8ZkFn1W?66>}#Fq}TdlUU*AG zz3~pAb9i7;JmAdUU)m*>L@UD?6iCgLE4Z^TBRnw8bK@}*q03H-wueO@-Zwe#f9V<; zC;Sta{t0F9yG@P@tpeXtNXQ$tjb{5(EbquBBRc|5piB{4RAL3O*g^ zmh(dFtSmH0Ba?a@uoB&>@<_ltuunr?5o@5YhcA{*E?HK z0KvF_1av)CV=a@P4cc2;#T-bJ8!O$&B#`cNv0@}+Xf2HA1Xr8?0Kr;+W`B&o7k_TH z_>Kby_Cxr7TJGA){=`JmtV-7~X|@xeIzKNiP~89v^SGQ6cL81<2sQ0FwoRX@Vynii zu#J38seQ{=_x`8i7wq|}_+#ToitWA~c%I%{7$9c4xU-T_W0lCp$4HOP0sa>GdLGsH z@5lcD62FSS;D#R(Z7tjV3qtXhm2GUhHYkBLXcOe_>Ad51HsoLviuuP(y-(SX_I{hi zch=(1#$GRpbp19jI0++;KsP&u$v`q$j(+H`&(DZowa20@OI;p4pkxg~Re|ILpZR8t z(tQ+;E9Gk!WOVX?RvkV1@t=DBmVe;e4}g9b_@DcDTl_`w#Az;(@cTu(*1mUj#pI!F zEWj!A40*Q%fO1F!pf&wd#5m8ZVW(MiNAi|QgZ9|S#qGKM#p!+!@b`keB)%2XZv~#6 zq1eZ#S;+uPkPVb#HQI=vxFtYfO9KA@FBl~1=$1OYwYvWRW@*}$=AO!uOCX8XdEM7( zzyVcu0HCfiao2%flMjoc(+82@-6kc40!A7LWDm4@Jcz*Ha5BT|S_|S273%^QF{Dbv z<_=3p4;jb{{ut;h`J0wa-D-Zu+KSZVrO|v*M=4SOB?(&Z!cmkebI8rsjFXM`KZZ8bt)vRQX9NyQ_F2?khX1{U|(E z)5$%hz2Io0ocVwV?hp{naKX7F1pKe5>pEV$Byq;37YsKMf>e-COylwWE8K;&JgL)H zvod@+7uft&rB52;W!<;W#&Bg&-s6IEUugdT!9lzcae44-z)UPrn%vN z_$0@SzBqV;;#HT9Zu~1_YvOx2JU^yKZKtF#2-;+td2ViIk~yISm}U)18Q3IjqU&@vgaWKHW9k`h63~ z1XJv@i_QZqghUkvdUCxp_53QM_{&_1KlO0M+>sbK9rDMK>0ic9pDv94g;Gsj85Z6S z@dU_^4Wkfu5L@Q!(EdHfewBQ9BJwmRSD7tbkOtx~PjVNp=UF%YE|J+{)o;bV<{0y( zL_$YpRw0Q4oDBXI%J`4Pz7p`swU=3#!ycd;$dWiGkVwn-u*ju;nLPH+4JNI*aE#@$ zJuK+H7P4&IU-^b3hgf4RwEqAhpT?j4qu~2ce#d7e!><6!#0+|nIThvF*N86krnI)x z^jqysuN8=R3=ia-0tBdLS5m<7fXA-wyl0q?#G03gJV3r4@C^DsvvC-7zJOiKlUmIg z!B&sSaCY;*ESMvW+08?!k)$xQdf55`&&CtQbsUjeBoRhK5(jMj$3FG`X}`AD!x{b( zd~^7-@ZV68;kfYkjJ0Cm50tK)CIN7TMS;PGF_&-@pE;MS|Ae${{QwPdr7(@OZ0;kYM(p}aTI zNqeD8ZLa3~M#WO{M>VJJ`hced;=b?0+kivIwCbLM!*E~zXNl3QwjGnY_& zrhNR)*#Mr{;---P>!X z_S?mg=)w!FR?;N7)2(c+yq_`;nU&1fP8A4HONIGx184Z*{{Vt#csobX^j{qOXV;<) zeWz)Qr|CK6;h#`pZYBe1{p)X3jNp-$z~;ZE@7Oc`3N7$wNxQlIpME|3QPuR>{41#2 zYu0)cI!w~6_UOzav|E@|yo>grvotY)&BS@wSr5ecm6tE5LaevGo|^vvk@ub+VyE?* zlwhx>zPdj%^iINxO@C3e)%9znSk0xZtD)V(Uvl}aZ+RrL24)zI4m**!3{@Rl z!gmE>@Ls)pb|-|HwJG)cE16=23TF`NR=;GnMM2E67G(#Il#YAv9(bzXMDZI-sOvg* zwc>q3D`%SDS+{E;1ebC7vcm+nmaB;j4XfN61Ii z^(bCFTTW>9w)FEgDQ374N03~kO3Zj0x>xfY{v*Jx@c>V-AN)y1mnN3$~F4vEFgN7SRacp;FNJD}FtVPutt(u)vq4+;l0gD-(RD|0Qs!+jF|&rj66;QEra zz2=E_CMvBMOT8hj;dtXxcM-ZWPu?3ytema5`CCVErTB`KZ-UA_X!@EtGcA*}Jr(sB*hV3*1#t_)B~@rD``C?zMGmapL_%6K4jw zrojcoaTZa%uXK1PkqAcH8DpLS_pZu)Gg-ILE&dz9ZK`S?+9?>inpcm+T6v6;4Wz58 z%#*3c8Oe*#WRP)5G=(Mk`T`!rwJH?Ob8t= z9t`+xe`{-|=|TS0HjjOEqRz|?E)3dw+K3o}p<*U)m>${YEn88yxVoR>XT?o6-EFkk z{H-l~Q>I01@wB8EZKTyL=DcmJ;4?pzdE|<>6h+QX&*ATeJOOE^_*=qr>6#yiuB49Q zds|7zioA0Vgyiu?mo~9H+MJHIzh#v8qffWV-0aa5w3o{~ z#<*-KJRU2P)6>E}0ulHtz-PnyM~H5A&3>9i^7wZ49oUjIYbloRayM2Dv=NV}maVC2 zwsgJ`_=T$KE2wzqL620D!o(zY)_NYHZ?0OSGBAonw`ibY8mJEfYaOF+D#B~hA5?CVg^nzoKb16wbVW@)3mu-&14qVI)0m}+D@fY zfht3&+uJ%Q8(LIJ@=3tlK&hu9m9KNPxqlF9J|Kg^DgCK;XJWhVWRbMJcG5{#=8@3d z$tCKvV<>SF`wa6~8dr++ShX!j;w;kHTH9bPlrlE0;rDl8veqYC_ln6R9q}2@rFkva z!wU}x=rQ@1pMZQV zscKQ_vABE5qmsi`wVmAx%A0TJd(>F44>1?r(CztaVjVlTV{NW{L3!YrjnucA^m^0z zEM~T~@fEj}pEQd<_FCEyAMS=SKRs&^HCqc!V$0#Lfp2unJNO&SyRx;_u6#LovrY!Z zYa*8hRr2sbX#vkl=PW<7wXYc+U33jL<4n=vjt{rInvR!zL~n;lE#^x*b}Z*~V{rtr z0CCS>r+h=z^y^(~#a<}5@fExdCl-sYT3KJ|(McgHimj!m+FlsJ2boK3>T}4J(gjT0 zytI8!#CIMr@wJ`KpKYNd6uI#RiDz{jF+eet)P8B&DG2ic#_^tZo+~fIG2LqZ5x?=T zhTuB3qdNpyd?%~Te(OB!Oq0cLs)(&XRVB6%KpX%_0b!ez9;YXVJQ3oU&Zly*o3eaIpdG*lY;o=!5zYlA7WN!gw=P*HgLO|H>c0^6 zFYHvCPTpg=yVGxUyTGs#kw+!v`8L@PfR3sc2jv@itLdxww%_69+_xuMk~`L0P?tn_ zzTHH*X_i}UKX&MfWdVK$;hYW)Z9l|%l-~-Zk3o|D`|X!7LRP}~c-kQXp<-oAvmfr< zjii8B@JY{3U&WgL0EIP3^lt&!Nuz&kSQnP!)_eQE3E!*C0>&e@)4+yRXvxaFp+^Lq z=N98)lv~{u{72$H33v-d(L58Q&*9Gx&8V|pn`>=ePX%8Mghv)E#6CQ+02x8%zVOzwr`}BtoeV-Lb)60!V%^!7%xPfNCNasAxTAuE{G$L4 zvx{#U9c{HeZVwW8OGnZzfp)xyNWZXkwhjB#mHVWW*_F~Uos3(i-gpKQW3s-6*Tb)hJ{9pVk9Aw` z3TyUyi>u1XZ>0E^@?uF=NRf2QrCU39J2A6(j4(XsCb_+8X#5l4d*#<{e$92RvRbVA z-Gmw%yn%zvUi-tl&- zFXI0Ij%4fB8a4gKhN*0~8n1_M(@{mW`J~3HD@AK0dx>OaB|~Q$j&ZePuXxi^yYUVD zmmVb3?1i=L*7pg0rt5Yn7&nzHHn6MQFzCq}ZpYo^^>)x)-H}EZb-(yRd_#NjDt%Qg zQqpG6j4d?zt>Y3Mp?=e(7r1D~{ky{#3)GX7ReUX?#-9tBei;j^p!cvU{{X@(s4lPN z`B^MPJ|j^Nlm`qGE)?QC9&4QVabe-D6XCY4ci=cQ*e=wtOj`B5y{oC^n{DTZ zCz0cA$_SCZT>D^FZ-)LhXudRETj0O^Bvx;&G;`Z|8om9dyy+~B_kPc5dv835fbGYZ z(~R}5Q$g0Rz8UJ4zYlKwA$#Fyjr>O2SI})6P1A236_lmGwx3Wj!6E8yL?LoH9jQ^1 zc1Y)`iiXl>rNcGniF_-f+IW9Znh0Iya~-_4dgOCC1BOj;1;E+{L&nRK*CQ3s>o$>S z-w~kK9&1}zBZkCJD{8u~ktX4ixwEznr%RBh8%W6;PZ+P6G@pnbHTaw2csv_(Kg0h3 z9@%NSl+j&7_N2Piw6t(fn6X`3%96+!2;L+x&BWNjdcG24tUx>A858( zo~qDV={n_&o#u~r0!YYG^2RwvnF&?QZ8gkUA!FECv0YTKP3b0cOE8Vwj8Cbpelhq1 zNBEhoz4iRJG08o|G0i5Er%M?V0u@tJ(Bc2x3oLsN+9eN#lo^+jTV^t~-RQ~Uc>Hgoe`kNfai&`vtCgQk@jlzF zD*1f3z-yFNHi^dg_FM(vAY77%r|L1brz^(0=zSgHO&i4e&Yx?d_*VAY!xq+FdnUQy z-9Xw}B!)lvd9=2YOr68!h2xYnjBZjza~>2LKg4;*g}h^{X+AvEP;wvj_`-^+)p9D?i>NWuo#>qXp z`J(N?J8?6_Rla3iz*l!;;_JEmW8wb*1^A0k@!^M8yq?y^%f(R(yD7t=*vWqYklHIT zJATzTEscb7E20pmv4u(0ZQY)S40?{MZK>%W5_Id>?Pk;2qL0Sr(_GTv`!UNqNu|oB zR%O8QXJA+;WEl)iFN8-;*1S8eS^O#1?6kP=Y?65A_ z`2qR{uRr*C4}^4&9Q~W&diV;$<4>MTYsK*0*`9lN*>Sq=#nh|Hj+j4z)Wk z0r;rf=+>HMlc&2}MIvcFB$r&%t(Yl0r_-m;l^`kth!mA`^I>vF8{uCN=o-Jp2_e_K zM?KBm{k6`N^A8VA;oU}6KRU$)>_cS8N`@n7U`F*Iie%YzDQlk$K0l{|^^f>Z?ffO< zeNxK&U06-y8)wl(7Ho*SUTbx^wfk1!#pj1;Im1SzbFHBG^Iq`J!nVBC?0i|_xO7{V zRbL9Vw}@eyDRU#dQZ#bQHIgwZBWU+0L{K)0?mRK!DEvO(+1?@W?bWWC92Wq-rGIbY z3ng{K8Kg;`8<^ac`A!UNk{gnwS?~;A2EWxWA<}i94#%l#)^7~AULUp^4z02;of1oE z%(&V~Eg^Jd&T)*@a|zwP!S9Nm6HQ!4;lC7W+D+`1w=(PPec_4hh1pplgxdg; zTig^NgRw)NKwN=dr(=1k_-Dg*S^(AcjUL{_8%gYJVUNVJURxmnZYQ{%8heQF+XJ89 zZ8&8lf;zPLC*dtp;(asXMv<#(X>(~LLN668I@OYF0Er&%B3evf;wdB<=U^O~<~|Me1n#LAJqzLCrmJgx;2s#AVnOeAbYK z8L#gP{tH$6WO!@fAN&)4#vUEibwjN9x8h{0;eQXVtEK&iRjo0ql z8I7=DK5XP3xg7TWYnkP+z+fEUnz4Iv70mm>kfVSz+qF~*$^)s-^sm#Fi1}(Tv{EGF zDgp0|i1nY_w_Wh9$B1-a3cUUv(c`xW z&UHJ8qif5pue7vw_Kq#pqdUCWr(ArbSB=CN{{X9DWj~SWW!QCAvWwXM!{+^(JWu-z z>Q>hu1H2t+<4=cHIwh5X(S^6$FSLntyJfb9duvnk&1*6%Gb{n7jz%R_(T>g_{@Zf= zV$!vL3-}-Y5$oMAZ`o(j1--VPsx8I?EK6^6@+7Vsl~T~UGK`Foj-SK76ufot`^|?! z@f+xpTgm1vp0(!LK=BgXTfpZ2H`;NJFU!FDSZqSX{DoS7idUX6El6=j{?tX&ev~rEw{=cnHjWn%y<4=Y>Z>K^v zdE(T}5(z}`hPjRlg~@{E!%}GEk|}($KhZ8-H;gI9ao2cu;_}DDH$EM)(d?UCxd;6k z(%yOA+5{i#4V}Ens#*zR7$Pw;E3e8noS#XX_ICIqtVs@o;BORoH+2Bed}~ zl##52qO>;(_rZ5A<=Y}K#z9kE1-HWA9cZ^wXnr2}h2cG3OP4YxovmAHdNOCyIOaD~ z!~+b04ZCW(|3 zkjbPy93&`xQsLK518cvK`lby$k z^%-qb;jh7QZ+8B2=^A&J=D>bjq(rerJcS3j=Q$jXzcXD+*uph8XWY`2HClYTE&WfR zKj47B4zK<#f5BUHe;4R>){@$IOzN7`MQm zYS&X;S@?NdN_eJYbip^Pz*G?+@}K6BK>)BPp(Gxa{5}5w!KV6khPnG@c+hI|Z1a35 z(r+*c^67F5Zrli5?&bI+BLov)-tM8`&xG3Q#U19Yd#U}BK{Rkfe|HjwiUn*Jv*Oz=0t2sH+edK-@>Mo0X(X(2XEI8X%7Ck3 zRXwwwwa{w64!lpQU(2dnSj;eb4be61@-RqjPF+G8{lEqTlC1{y<8>TOa)HLCOk1uMVQGh`zf0obqE?189 z2K}ZyCGeZVx@&!#;jV((jQ(yTj!XSfnW9*cPRP8tp(TJBV=A2dwf%zI{670oOHEel z=IsU)=0gx8p~+GhjCJR~<^Fyj@Ku)41&%{ zMYj8Edl@7kPGhi$F=&HF9v^!N0aMS4%PUuK!_kf0nn%>|jxP;LwDCBZq@f)z-S+RL z&&0ou9~d=H68QPNN%1;WyMs?;kL)NmCdkQQY=D!*qYEcYsciMHio9#%ZxDP!@h+j` z9}!58eW^@mn8u+|Nx)P=(NPctRqU!e=BDwMxp}UXTefD~EQJA5RYM${zCLE>lbro4 zglEQaU(viJ!j$sbU*Bs}@?I|Dxn`BQqofnjey2sE$ma1$%Bp3FUbz1N3jX=O;MH$~ z_Bs#j@$pOJCaAU{`2LAjX@h5}7;Fvxf)$S15zle3sZ%~M_05!5*wXL(~IAHL?u0FNN z$M~*HB8~TJ{LedvDc-c(=#Sa;YnV>bZS?)TgTHeEKf0qi$9_1f+Kv6a<|Nd$8>{AC zeoP{_;&9l-OCN`^){72zWfj%;S0eDx!`nA@N1R7Kp zQ)(A7S7&V^v3zM&VF^?4HabqYY0!yK<>$v(&$0FWK;Z(q9*CJR5Zm z%@z;d>i0pg$!rGG3~SWOah-}s-CfU|s0aNMemefne-Az^{21{!fm+)K#2O5?PO+9) z*u9vI6_zKAIN%Mhyb;w+ekgvx-|$L1PuXiodA=n0&dU1sc1dm{wQ$h}3-@DH`LGEf zZjhS&n(?2H{vUif(vwxb@g=^KZtwYmp6`&Fx(d6%3Y^^vg2>*-!|_GOb*i|+YNAO#Sz zDGl}|_$+tq>!^GN{kc44Zx{)t_^_vnH0O>Z5;S^|FDz=LurD0pyzDXg z+%X`@f%yvIn&1>6gD7FoaC6?jsxo|Rad>FgkJ?A#xs3-y7g7tO8>whB=sM|!Sj=$9 z!Nlgk~l&g&i-?*=RByAT$9s)2vN`hiu$kiD)^gs;SY;;D2f^xo97Xdtim!e z&pa+k!OnVjuYV)2$tcbAXOo>$bz$!x>G$Ek7wTI70K%O+#Pi&0o*J~ax3_DJM$S7e zLK#6&25rHb>30$XD2y`WIoq1iz18phNovq))(NXvdx3wzhp-B5f zi;0IkGwd>Rj1gT8`|JJ~x%*_XYWilasi|vJvlqJFm2AqYptLLG#Tt$Z#quc`BL=+c z$HZDDoZn#Z=ZfyFTEXN^Qg0R7-CZgoXYO53_E_TpjwZ~4On{_=nz5iu;wu)^HLnEO z%55xpn^n@UlTni8NZ!j9ogtA@NjVD$U~n^<`Dr(E-sRI{tkZNWDeo>J@Jw1>udnKU zW8Z6D9-h-qx`I$ZkTiR)9J%C3?94~qBRtjSnS5(!CxpJ+_RSKyJDMWU_=Ssu>qU}PWOSn$k9EDu`zgZ}`6-gx`NUk-m^PZ`N{`d+!?KNer;ejC0w zR;z2MT3G5gh?2e{cg%tv!X^X=OK)CA+V;49w4sirC~wTa9S=J-qNRqZ2|kbVKZ8FV zSgxgcHlu301p7${bAy0#J$d6lg;?;efu-3(HBD5iqedot9EFU0!*d*Cr#*#I)h^?P z`Gg1}ibiv_SxStOaHUQ#Po;gg`x1O|{gHke>aC~#)1MBP#1j3bWw6(YlHyAP)k4{^ zF)59JWtkD2lZM(96a9mdQH^SmgyfQa&*ePLx?IggyLR04kAXfl@dl4!pnM|MJUx4B zr+i59HjdV}(y4(O-$s}1MMA#UWuA3~GIpaQGX@|G{-^%UU+_`Ozu0^J3IpOT7ve98 zp3A`g9>21%n?vIEQ%Fn%ItLc&lssZB`S2L?xnNxN#F5j*?3J> zIGU5G6>BD>p6Rx~M1M&C0JGQaKl^Ha-@g&>{B`h;M$vSCgi8eWeofjRv|HcqW=puQ zBVxCc^b=ZyiZxa_pLiR(iT)eM;M=9}cZc*HJ-k78JEh&%hc6@0uB}~JTt(#h)442i z&UfIB3gG6vGyVzn{{RIi_z{_2hmLH1Bzz;Z@bPFSHe8?tbr=;-wm{jH77VPxvR#cTxCn#6jsi9~Ws;T_L!Qw)v}h zc&PzgoI|v>;&Pp9Wo(9%H2yNWvemRr2Smg$!J%IbM^AFEh}2kEL}dAk2b2XE1eWQV z%GcrXo88Ny>9R+oXmRe7S8Kf?H7gn8MPHp~mhH(z) z*X%C!>m{|5a{Zs}N#t7=a6yon7=_`+4r-;xi(s2nlf;_Nr)8+TXl2oKe-r6LMP`hg zMvHwsa?E2x)z%_evA7)OtXNriuflq*@0X_Q`aZp`ER(@^X>)zzO*Yx*f1Bi(i*z{U zW5jBRep~jD;AC+!YOCT z+;y@8V-Oq*xC52K$gn%IU9Vh?Hx>y-p;@h>jrVu90N`8&W~f_ z9cxhdb>VBT4Qmq5AUAiO8@ycuP=ewK3r!TUSS+%Tj56fSh9Ik^I_PS)7K)eJZ;17K z>t6_IYSPaSjWn(B7mV!PCZl&R%p_t05gTj=Bhs8xhI@aK(-OFuYxr*LBqvgzTmg9ol;AnmnxU=x|+J}TA({+7!U!Be5+B%;N zYH=A5?~or35)HubW)ZI#O{&Cb=yHPy?cc#BE!CA@2Gt4wvIFh?T& zr^Z%d+aEl03Cv5jQNU(CxjidV9V1S>)o%Pln8*ZdNmp%JkGtkO+((R?YWT>`ne zk~^DUwM=K)Mn37!IZ^?wKNe}9@S0d@ei^sC(|k=N5_#7)de@ql@mv(xo>N(vHn37d zhK+-k1Qi?`R$A57*NUzDF$CTz@MVsS62+}&wutFdY+_ByODq!;=4XstXB-t)$*mmq z6Na}Z)wDfU*TGg^6|=I}uM<@(b8#s0{8y&U5Xg!rlU0=+K**DFpyUjb-WhH@4Xu1g z(>@mXetjZE(p1{X{hO^{XxCBB*lVPpO|gw_nZN;*v*ZkqlpbjKcHcwsmxwj5irT-2 zJU^?=ZwZ|>PYY@iY8r*iaDujMEtUwummLth0De_1T2JCVPS?N}dXI{=#@8mfG3mDU zzBz~f5w*B^LGwYLJ8=GQlp7l$f}s#qpJghCEAusOp~Vov8?P5>b0vC`$s zvBK)wpNM=f;pA@@YPvzY@iJUnJ%+yw9v`}z75mQ(#+;L}mn05fAY_8AjC4@=S5xra z=9g#TS^QV0+RX$~Tpd5l*L1s|GmswVQfcl1i>SblhU7PH-fcTtPaa3%-wogRpTin< znHpNXs_BtWr5`A>$vk(mqsTWFVELDgydt(q<1{Y`*?d9xqp5iJ`wzo9rInxW?XPql zYHOWB_6X1z8cRTl637^B0i|FM-EMeNSJ#Geo>;nHFgT(c@f*B0={y*Dt7 z5tkY2?>;a|l75i&|A_CZzHCXofAHr58ECCqCNe=zR zWY&z_@Gpn7*ghX>dj9~!jT*x8d$}XG7d`;I`#r)3VDnE2)ZREiK-12RxUn*@VMGP`-V8}}|5s|kiBM;)2h-~K4JT;|w z{OFp6)GKp$6kab!vDH<}1X(R&i^&7!C0$np0vA1vpBnsApTPbu((El$#eN;o;J3Al z7Bgu>N|H8R=4FQEWxh>-;gkhzV*s9ONXF(8*d^d zdin_T_wyTY64~84xNtGhwn4x-JaSf=e}=W6jrSLRBfIeat$l5Bu!(QHJ3Y3i9B?K? zV)==Ql^Z1`83Q=vf@_A+elL6v_=WL)=fs{KveNuZuC#_&=CRYREwyj5LKn+3`QBWg zyXR|&nLx(fy(>>f_@nV}MDd({8rJmvS4*Wy=81IE^ZMs_mxE`E#)DS2(KQVcIWOaBHA|gi_ich$z1@?%aF}M!*>*zh=zG@;pBHsI zUmjcdI@jVg)#sabV3$m~(oMOMBFhML>zLXAWRDp_(9eJl7-w}qO}z1jh2n@IJ_%hQ zUTWfYy@OeZMYw{?HluClTeHN#;nZ7AhHm)XT@ix&CoJko?#`QC)ULcmJDCRjliGff-JnkFNcwfbqKM<_7&1>PNmtz&p*Oc0mYX1OiaS%T=m(VoOI;j}m zs~y6QyGR)Yq(W2m3dYi_a`Fp$_ZvA|_{2p(IiN=WIP=DAH? zTTcmVLtVc3esuLZjBNUyuDa=P`?-0qH2(lP)un*}+bnAw?Sp`a85b1R<+&?+r<&=$ zD)?=#{A9G!yiMXi3V4HA({)euD`e51N!Bcx<9L7KUvSP%)WHdinB!2%yCmd;x1xMC z(EKaodkt84f5w^@+BH4m;^SNz-j{M_3^v^)v$e&%S;PFXMzOF8TNwsTf8yZrH;b&S zwC@~Cs0npg65mR@(o7a?g_TTtb>yvX?IzrWkz<(dz##KpfqC%WUkmGB+uk1V+vw80 z__%A^U2;Jq=1sepbiCzT<_cJ9SgY-g%C3r`Vg7ydM{g5$*A71kxw?#PNL zG?Aog7OTuMtP;GkNi3sw!GsFnouVdaZk(Dw1W zfy`t={{S`tV_I6Z#OhYAqq@1ggHlC~KpZOzjW$UlC0Km2Y>k*H`{S_I zp<{uA!~QVw2aNQ)&ll?+2iK+0^$#-2(COySRnvU3NqA(;q2-A|BOX|OrGTwHS4NKe z$1A8yf8xCgSuO%B+N|>G8cpt8KbCeyb8l-DgMg_tOgD4{0akov1T7}4-DuYF}#LFGs}N(dWj1v zfLVXDQ@Gpp5c2QePBK&>E1EZ^3 z^6*7o@lEx()FV2Ycs?uJSdwc!GV)!zlzZ^V$#i3!i8pe;D()j|9C9uA!Z+4$=hp2a z@tx4})X8~k8qZ@H3%};GoKEhlMsXZq1IqfCP28oa#_2G4FGuk^UTAuLjSbzcz@G7R z+xeoq+W!D9>zhHo34m7I=_$cHs_w;hy2pqhhhGwFzAD!*Em_rsclt)_Lky?OW`^?8 zFkryQR(37U8Avrl#P>RWqh^{gi^6wyx3=juotCR1O-WIgB*MifXwFF>vc}5C3`JG8 zy75kdtB)61Y8uw7sC}4hvXa8$^Fdio0W&R={iHG~vhIyh7`EnD6=;TT>7E79{3ouP zpA2a~*$riN=DOJHGM1T0BmFB)xRxjxcCjTk6WwxgT;Gg5S0{x$RSo8|;|qJMT|QvU zFZ5@#jwqyUlM8K9J+Z3@+ZbrDc@9A7)%;w!(|#nyrs|q6fczO1-1cGOxV*W0ElxlI z1ME(*D=RM{e8L9eeR!Ng=SY*Zf7T=ywv?`FC1Qo2DkQW`;5sb>X#+XJXkO zGP0`ylY^SMD{MvE-5R={g?->HT2B*rU&8v#8fsi4v0Yp3aoi`|#QJT{*?hLn@APm^ zK4Nj4%W!9h;$#zzW5JKArP9f16qb#qUdI&R#OB`kf|HXOU>9HvcxGOsE^iO`6XJf2 zrFhfBI$nk2DGHlgolehGz0&9VL@Z;JS)vWGIxr!YN!8JY2_R;?-}c)5qr7YT4t~a7 zF8!o*onBoEZwKiH(!)l-h7CaKTAj`<^!TT{futd&xO5FOOUceqo~|nGBuz~f{{XMl z{uaOBzJInAuk9oId-#jOz7EvyH4g@SH=5(adNpKQY~z7(dmoe(WmPSZ zp(@y~zAA25Fqi7ssj+x_{{mUi5tUU<9BjB;som%pD(5ULU`HxJ0pZ@?| zD35p{Po_Im#xv!KbDSQ%xT#=L(u}BV9I)q++;LvS@6hK`J+t6{!9N##WAWyX@b|$s zPpRu#?d%$En{5KKneQ%0SdoGP<~1jL#YR>oX6;|<(fcm`$=(e8k$wbtJK+BS!5s=G z(X>k#eA~v<<+N>H@nZ(&{&*duwYrw#ZS3WTIT*kLAIGo!8zcK7%kek%h}QoAYRzuj zSdU2fJ*Gz{n)*BoCDUvAzncUGAu#e}Yg-?*OC)R`BH|V)keUAG^qZX<##$z$qgz|W zuIpCVxIgflX|JceR?K+~h~q+X3G*OK9OZ{h{$It}gPm_<(H~WkVcco+%c)z%cRm@? ztj({9H9Kauyg@2ymzr7BEu#Ex|eS_)o%~Fw#=*!d?(MzLM%* z9YasMg4%t+>~+R%pka}yHVY$ewo*mV^QKV@1x^MQRUL4cyCXRb&(QX%8^Vzv* zU~7c)HYvyModLlcK2cL?dWVI6B?){z;#P-zNg1LIg)F#u zRSF&(Jx9emNgnl@*`90TT_eFh4zbfmi9BI>;rnZaW`j?<&Hb&s+k;6R$=wiyw&H;e z%MG~;z{%k){0Z?&?@YXX5^XcY+G8+tC6l$ilyR;Eoc{9GR$Z#08h@1Mg2dO*x(18l zT^moI!M-21J{BHfo;^{n{{XhpHU(CS7P<3gRx8MjlrJQ!16fh&ULWy>oi2mojay6c zTHF&ojlHyILBEpooz4%K&LnZqnGVLtB!Vf(wKjb3s_6O(+P4Kt^gV61z{+@hOVz;|I+?Te{VmBdQZT=_$bGVd>O84iF*!@ zuSWOR%vR&gnn>chxs{AShbBo*Fa|gT0;argfTK84H!nrV>Uw=!`(6Ia`Zk~YZ2Um*SBb9@)*lb*b6DzEF)~LB>M}`?(IzsH z$-HG-4qq*kyEXH_?L*=HdqnWYm8IG=Q0Q7?$847IsY23}KfE|NV0Q7w;BrZ>bgBDD zr>jeI=rJ^^FKK-h+56xB00x`ssi=O~I{Ha>_H#Ccem4|HQermDqhpLQ4BT|bJlFTl zd8Pbv@cShG8(E#6x&HtLjQ+@a?}>Gf7Jkccj|_i_zAo_V{{Y2J;9XivRSbUDrnEQle7>lX3Gm&Bi1Uy@>r@-|wf&y- z$>Ox}KZ&C7%d zzgh8A9vN*~^5yj_D~oGnSzt?L@?La~ONk^5cM=ZO2Q8KW{VxgOD%8hAO%uKS4>vZP zVTY8N{Cw-T{{U+-kk7lz2j<)lI3cibGJQpJfU|tO0gia-?_D2^mL zsa`!TF7T{DEG{ zlit79bKwuff7vtk4gHq>9DdEdI{1s>%@uqV;k`FW32~{x_K75d-3*Q7#lAw!`-#Hy zU*w_i{UVdZmMtb_U$okwjYr>kCPl_M=Yc zz#gBwk>LkDiDUTJ=RA9XjY-qIx@>(n4N2N|T~F=1;*b0mtM(@FcbgA~H9dR9S2pF8 zMWkt_-AUsBPkPqTkTcX|SLaW}ulO&;l3GiB@b}@1NgPUdNv>%DCRINyE_lz^1}ur_keZ!&*4w}5flCi z_3*pIPp$aZP}KZcCECqBj)CzW-%qr)2&G7jN94r!9#JZ-Bbhqn0!?y98$y$A2D8Jt zXxaQPcsoM)-{Q?yK3u?p4UjUy47WjH{EIDw2osIKh&4am!k!9iySp$HXAG7WaBPEdA}sCz@HUFylU%p(UZgxTrEAunCDu%k ziFAENXtc|y;FKMXV)A64b7TnB(RcDoXRke6@#>=z{{XYq(~10Td?^0_f?R*WX8c1v zqj-;4)%-VOGNjVOs`#E(o62FdI{lCB$+)M?7wvJA{98#Tzj8li{{Yx)Mff{lmi|BS zrkCRH9v}{}N-hLmVP(KJbKV){2j<%%TrN5t!!`QL;tz)!pNF+Zhr?bN)HI!MQIRg? zys|^9*~KROy1$(wM2^_`$C4B&$ZYi2ywE&T<9#|03wUQ!(EL|zc?5$}m1cVvMQx+^ zA7Dw8$Aa6*bzpI{fnH5~MOwS)dQnc+Kb4>OI4|KupAb9=;XjF^N&6I3MP@;|n;h z;)})*r;{F_t0bs+^5$`uL%_it{{X zk()lNrr!8k(!$}^PbC^Cp=nz!<-h#$3%4bDzDsoCzS{U3;r%~Mx$tkq4-fcW=Hl8$ zm1VxPx0Xd%!j&sySB6t4RoXEk6(kZ4 zF<+7&Yw>|Q{uCToRYd3jaXzDk_)pTO_#(E!% zyhyTXe-ZRM9dp7SA7#9cP+QBpy;f5Ud1=j+tVn**mSsp{cF4v#I1~6)FZ@GuF0JtY z09@5H{Sp*WYouOj>tZ);++$g;q8C6&+{ndUkeu`wugK5Q7mQmXjiY4&Zjx`VTt8FJ% zylq-2W1aA%S;Z`CweZUnam++`7~la_HOVxc9{<7MI4q4-F4fzVcv(T`GM;Rzoyl zwt0(gf+dLnAW0-4Sl}_q!M$(duZg-RhL^^gPOf}&CG5UZ+1^P8jw6hc9%GIBrQ_v7 zv6jww;*_M)C2Vfrcu&Rt9@j(>cv&^QYf@7z`sSEytX|!)Bif?o;qFAiK4Bv63&0p6 zzGVHWeiwK<_SX0bu71iIXN%KRiso2uCYtHLw5}}gbjjA@7a~b*7i=@f`@$TV22^&g z`$X|4g?u$*ap4aH%`bwyZ97ELS~RbC#wnX}#~9RZg1X898=?bmKtZoU@jjRFm&AS_ zxcGhX2T6leic=AJ%RG{_s=JK!@AxVI z0Ex8!00run`n89KBGI%@w_2+zNhP(pYm%!ZlAOTzCME^j=0l#~`cJ?=h(0v&w~u1h zrSaCg@gv5Uo*g&y=vqy>9SccRgt>=Mc6oltkQm%D!xG#yxZH&p3jMz<;@nfE2~_I5 z&&;uWJwi3*823MEzu<|Vu(!p(23_6!Qt@MHo)U`cEhr=2-F>;`!(B-nDlHo67FlEh zG*y?)WoKQSE`Ok(i$4YY6Y*YCM=>Q?F_AG{}rhG^y7?iEx4`yue}#dg-83pEcD_}1?F*IcvlES3g43u~*Ox3^Y8N1AImkpnEv z7o1hlC!4#K-CrrjnazA`BC1Q^y}oZ#>M|UnmL5`_HSbE-`?Nj?*FRxDh}x#8o*U40 z{{RhY^Mat;TwTd&a|^KC5?fOg_>d_Dg8J9!Cx&!8{{RknHpAe3$A~o*O*YxBby;-~Pkt=N_OI^4-3LAw|-I?Ba< z?1LQF=HA{JtwQ=rG#px_SFU>Yf_a<+gjEV+Gz@D;=!t#1R}h zlxG_VCl#S@uUh!eSGb!008z5K(k)dU$4>Dsi7l1Dc0;xybAIb1yRapn$lRl@Le*=Z z3tauS%fk9bi~f&mw&AsZ3u+eHLQOVVBr@AV!Ja75hSff5vFC=tsH`N=ek=Hpcy_~B z()H~M$7vdGh%{T4)FW$lB35gQ$rU%d6R`Oi!2lOp%i6Akab#9bbUSDGh2zU@S4}r^ zBzEa%H2U6{vxxRD`VFt08bG^pM%0X)1GSE8p3uA%uXs~Le;4>``!@Ss)ILRRFFqWW z);V`?xm!!yi@3bkZUk)lf-OoJY!F7}VpO)X z6DpYtWy<+rj$at#6GgeT@vQe>85>LR*Tl=)Ki%3siKpA!-<3jqr8gJXEg$ZcvyU_a zK?{uF4ydi%kHlJ!iS#`>FNacFKy#= z@I;eYYF;VSJR{;=O?>+sPaIw??WtCOnSxr%o4cy0oR!Ep$6kuxz`B=)tfq%f@MISg z_<|>rZxDDTZ8T(x7Iv4@n^&J39p)CxFI@iRL^2b(LeE3hh@WkxGk{u8T@4hHexjAz&MEi^huZ>C7LTCJoq z6bTp!r@=MKz&7RF%Y`}10Bg&&p9|?;HrDOF5NZB3vcH z7v328ah%iV=vsC=T^da%!@4~FGlm@!ePde_M{PERDA6zXNyW|716jCuivqxv1A=mQ zcV7|Q_{YQk2ZO>orOu_PH;F^7=lmraMBXE`mdpCK-AhfM6(QKScg-=-Hp}WX@3^J9oqQ1ZxQ(Y z^`D2*TQALLsOlyiJ&m{us|A^NR#Lbuu&+Fp9qQEj3zef{)4)3SgtSdJ!ru(^C^WdD zk*2zv!rGO#le%V55iV?QA&`{_+$nA9aly?y#@a@K#!S5iB>0pFc6zfsEaM3!-J-~`IesIkL*5wzKQnsUX2OPMv~?2stj zpU1aS8rZTWQ7q9IL^HqsCrcad6nfQrWdd28Y?=MgNu)Ddc&0pC_E#$>xOrivgpQ3k zFUECsjb|g_8Sev|(@2-8E)h)EDN@)QV==gI7@NOwl%o~#hGwXMd%0GOBRpl*CBcdJ z8=qN(e)2E72hGZ4sHVYhA7<0LP~`qI>RVpPrT&fYY?f%Z7xOxh}EUM zPw37<70q8N7xmX@7N_I5%d33J9@YGe;fvz)54^!Y&YeK)tP_((HZ_OPn4FBO@PBL@ z92kVH+siOE1p*tPqs@-CQ>BSmNHgAj4by`aOMqlUFZ-u+nR@vZ+Q;Ay41S z`Rqw<-<8Rsw??bN%J5uVV&NyeL^$4idDk`$Qc@TnJq z&Jua)j12vhzH$>lT4C{>oV!TTp*7`JYlq6VV+UClT&ToD{rdm#0tB@^Z+EV73=W&~ zt5(^%H}i^o)d0W}M5ZrmW(ff(J2)2et2_Q$lB&w(V{4!6@(9P?VIG2t4b(;7GoX)`&>dOwgf+swdX^xW zD|x?jakKc57tF$ajd(S*3HyR-j0CIhI=Zj*>Z$GNls;ocL{9qy4f2Za@(}~HOl{MO zYLY^d#!lY3x}U$l67?M%{I%J_{-#Z%`>u04Pq*I`l)2cPBAhBxR{Fg(v0s63KDKU8 zd{_eu4kLCu{TFy z@_}v@My=-tXN~T7RRXeO`jFu0H%An^8LsRq!DqF;Ls;_b$xXTo1-4`x%yACk59@X?2-=rk7#P< zu0GSTkQ<;1QNG_6zhX89#W88U+BJ}Sp}Odt=xF0~y9otrZnnnM1H#+@zPu@;7R30i zp-IX4;kkQ*ou!TCUuz*uWG#O{i@aXhTs0Ig_Pw)txX&)$+PO2Q@;m3V`VI9$j4Orm zO208A$fP^fyg94ixO8{5qL4&HRhO@0$w<6~FoiNeS{+2L$bH#5srY&Wz&;dyvB7{@ zd7Zz0K3F{X(Z=Wq7OnppH%c?tN)@W4Ur&s}oz=5dS|1^qzBWea7Fm35Icf=2o!NXH z23zYYc5nZB)EfSD{gTB+q@bPpaaeUSG_eA<^7oGMdT*8=Em~SJxnr{e@w?V=gdQbr zAwN$G__K|D)W18jo{##!%L_p?7|+9d)QZNs(1!5T#JjK3XuvAWR6f(> zaN^0x2)RYMM`Dtk5@m7k!a}GlTDCP}pnO1l$=cAPZ1lHK;PLL)+CRJ*j4sr}<-&?F zzhADlNHrBgIK`+ud@4}|m*A$R9}}0^b)jqTrqjaL-?*xuQ-A6)XZS$y?Jd*_o<1)5%;!#Zjbj~+mhi6zwCpd?M85k|LQt3LTI zT|p5GZA!TQ%szkL9L@Q?iyIVUF67pr2?q^-P2&~0EL5#ZXOkXv@?0^rrG1OaE4^>6 zU{REF0=LNK~Djg z?(FYsGzKeEzEY^F87!&=-Ed=a`{cW_kuFJu)U_FIyG&N~U;B}cRm={t8}P1mS+qTs z)!ISZ!2J)RNVZ>I4(w?j5|GTT6XBd^zn4yUAo&0{@F)wEca6Kjm9ecb6dm@)w3ufUlQRRyd;v%bc`2Ts9$Dp;3C9xXs`_!wRNGo;dGMI~N$;&Zi(cxXG8>(r9;QuuS+|`99dBf)z+|tr z`aOpIJ%woS1!tMWKX0u;CVvbR0hLw4sCD0)Zyrdf(J|AXk@6KbWpmJnEQ6g_lE1pu zDR)!U`q{_G$ZNp2Tv4%IX~zQQO&xRxwSQRRbvg3-$$3-I&B%c$gV{l$C5QQcDy(IB zm3_5Km@eY!b(U%qG8e5K{qG9ev~G#!rf1E&CLdElqc8<7N(g8`nsB3cogejMm1T>1 zZB6Mi{g@RlBhfUz0F7jM_9^n~=ceRG3P}>SI-IPSrY-GYpvxv$XlNQzE9>l%^*U*4 zdA+^D$9n0Nx7R*cC~xBJM!RX$q|^ig!#b&p-py<%Aa+D`T(N#~n>%VUR%*;<=nkn| z>c~8O`hIzGX$N7TMmOB;g7&!WUl*K}Jbr=HjK*4lofJCI2j-$#nFYa5o1oTOufMLa zf1cOA{+ddx(^wj;h+2_9LM1 zgT}R?ue_?pO@Gn3qwVZpZ<{vHZ{NC_*j&%9=t|NO=a!5z+K&A)!38N|%8~TryxgS;RP|vFFCNE}m<^juzvK zj5$H}CnL!?TsEypQ)E^xtyfoKZ|)YFbN!Xmi{FMMo1q1U*p!Hj=M|-pVXOdmIv3L; zTqc68jE)cGR9DDmlE4O5n?3Q|ujgPE16RK_+r6xG+wAc;C4S)%a8rKia-!E0BrHgq$Q70WYyAr^a2HRGUF_+iTHUGRz(uMI;#)CR$*;oS}uT{wJU z%h|p)++pRDFFd8pW)CUak2Tr0nz?v-RC2C*tU$WD%$P_?;H+~xNj4-kt{bOW1iBZU z7w$zYLE80ZL|kDp+U3^njULj>EC1n%?8XdmUbz(3>Ao86P+BK^ZYc!wUAv0tsM%}#6uN$P9 zEks3;z$#3TNhJ_(#E}UX8On1-xGxy0M~YV1eB2 zpX#)#;53Q?YG}Q#XZ5omI7BCH&>0)c3+tv^Bi8GKG+Z~Z=skP~ETfjZay4!5`b+`}hMAM~7Ar zozbzu`el~O>O@hSyjqO{@la2%t7wJd%q79EdE)br9UedUd{GSA>NH>JVYNXFHU*Xg zmPZ#B7C7mlUsRW91B7La%SV!P$8KN9lMAL&f-q$mnOoa6%`H#}H}b@C9bwF0#_PYP zG~N^A#Luw9Y!hw;mo_v4{VT@-q8M>XOg^AMq!%Yxqu0u8nft5AoF1MHVp2GORF++1 z*h%2FocE@wK?iD><=Yb#W{B_^0j}QXeE^H%_LcceX$JEmJm`M!Q5!o*<}~>7^;p-_ z=o6kqD+MaCy+fU=y?3s-@JI5@WnROFmWZ{7`jv(v!Q*nyw?cZUs^eN?k}l%O86UA+ zD4BawsawI&X~wJO{3mG2#*Y*3sY6b6z+@*@+8t;>5ZU@t^Hw+ufh=AsCFCY+lWSn| zQ#YgAd>KZh3TQOoNRd7p21?Y`kNSYXsPT;vmk^JdQVX^Wa|jK66I9dYK_pu0x&k2) zeFX$v&;QI$Y?I|0#*h(gfPZIsU#ig#l1a|8>_Vou<;Gy;f>b=bRJ+OdlzQmeTom9J zZc>D6y$kScK+oqJPL%^?j(t6I~*0HBK}S>#+1; z_oP6i;gSGSIT^LGO&LKIqxcAoZ#d_djWsL*yvrkwQzvac-0W!PgYGv$pIB{qOXap- zr<9}la05b(yH>w~mFUYa%^3JrC*gUYxp^1I^gqeQ&TjcH%d^E_N7r3-zK==uCqRW- zUYlJvjK^NEi)Fm|VJ9VWl5Y-9-YG$@cT!E>y@{{T;-PAP`s()^x>Q5Gk9Q<>qS=Ny zL*I6`IZ)n{3vH$IiW5EQ(Xxy$U3h=$s~meO@z2zCYtIn zeGF-a7Td1^)Eyfc`!#^j6H(>2Jqhl=h(Fa4#k@$ang)AbDh1b(R?ROK-74JThxXf( zcW>qukcfo26+{*+DGLFt_&A@S`AQm=|5{`P2=itJRXABW-XW@ZpI>Lk@v7V4Y{Ra5 zr@GDay!D;!XueCR=kY!O#5nT7V^|Yq)Azv7v`3+Yhy6Lrw!onLem{qyp1$M8)>V|E zQ`@?V;Hg0ig|E+1cFHf@4d=AC*7SPxy?k7$3%V|nC4*^RRu!0tAcn9}~+kP@hTPo79PivI%pN$Oe zId)o6$~6^XOjKyyFBek|5MD_qZ^hd7oGmS%{z?!?+kg60K05QcAx~14aB_m8SV=DM z0*K|!@vAlhCtX@4aRh! zu4^mVxg*3GBI0K9k&u^W?xCz(LWj`}WjSSiHKYxoWtr`OGqucs6pXgNCQM7}vKZ%W zPda^GO!D<~nZg3W+*(#TAE?$Q+eBrN89PGJv;qHpKTbnK`V}y3GYx8gZ1-|RjPlX5H;anFWgsOuGr>!O3hno|G6 zvuus}W3BTfRo(b9Vqdl7I=9^D^$?p?I3=K9cjGdDTC&DsbM1F4kPfy0e{X|VBB4o# zEeU@yG%hz;$8cvG*5l#!|46H#P0?($<89yUpM1&;QNLtz8_&2y{}z}p@8XLiUSS^z zid^rrhXHu{Vf-769Zqgr1C$~p*k+^(%Cz8xQ|3RLk0)XJO?QPAePYM#!%X{zeib)H13LG;)Iw z>syWqxXi|UM&{~EgI>4cYRFuj`e{FV3SHsStfQKcAGT{+X?X2F&14=vd?EcNGz$9= zw~9Bx$@`=)?pVkA|HC7yJ5m-H`>43shNho6yzyH&q%>y>A~_bx=;5Kw@X=)MVg7BE zPg(G*8^6Zo0v4D%eX(3-Oy)LMo!mdJe68b;Z2Itk8P@z zc>S)~YOIK8rCS9sYq{+iW3iL36L0f5{dfgjd~kxr{oab6-jQ-aaq;FoRvyELir;cA z2-&oc+8{j+#s|@J86?nwZv(suQCTp9O14=A4p{)eYfhmrHLX`1ie zEWOz)FNBmpeyGV4MMveTtmv=S1HOt| z{t$PRW#7hzX58EAhXp#g85%ih1v z6pD^hkj;6KWM|*j`j``h%2ye8Q4B83hfduoL_Uv`wGZ+?f$)uVw4sXaA^W|Eve?oC z^;*8PzSPwfD|JbLf1|nr!oVUwr7!JU^CJ(O002{0XgP-bBdLJ1ZD}!CMCe74?7{^j zfD*R4m)TP(*7kg&d^TJ7?}b+WO$ZIhQ)c)g#i9}EDVDE$b;T%cOQfj@&CJ9}IzE3O z9S!w46KXc}PH@1$jZt=`+idz4+y0T#1xS2QJaNtX&Djgq)+e^dy(_4FZkg5Nxx~&e zm5k*J@lrPHA&RYn7%t*2o*}0G3H2rp&b<>}m^6YCUNyn0OPh6e;)ij9>zY3p;R1%s zaeeYlQeyNX$K;4`PY&4e#c^>7r-pq8W?eR0zndN$+JzzlM;kT7KQXXicwH`nk?wJ!5B3hS+#Q8e>)1Numc zv98IF-q<5f#GIM}A71MTFYF(6KgTySZgw%|$Mc{>)1l_GHLL>KJgwp{)DDo@HqYPE-0<28Ul=M_I{d!S zhc|JI9YrXViBg~Z(v%Txc=t8fsl1w`0DkvfD9>!Pu(u);0E0+Ne>_=(LI>WM{G4w~ z307)G-EB2B-Q_A8mm3Ul-M+*%d?4DRgagIoVnf`uIJSzQlC=Jd{v0sw zxYP)XKIlL{>d2HggY9sHn+j{_PmNm=#Jy_g_3Seriu<4>daBpjfXM?^{P`%FAvo!a zWSDcTZ}#TG!o@(4#_D%~htDjy)`@EDr@zmC0I_9af~pT#CnKLk>a0{71_knLi|i^H-WF<#A1(+Uv{1bBwQD+ zjJ+S?S87*vb_gmsSHykN2rNGg^W@eEYl`8L&h;%h(63#!2QFl#T${cEh2mtFCn)Kd zb!{FaylW%M`TNYPd>#xi2nk|9uOb%vaH*VMSAYRb(w9c7gGn-A_?I#I)MxKrRIWk} z#q9^(u?Gh9eB|zY;q0J$F zwa)33rx@1Qm1>BZyXKTG+z(GrU63yrI}H!zP>H#Ja&S-5-LA`AcT!xRdZ@8No1N>S z@SNai^`$KKX8Rx6<#*g8xncwE5+x_-wq>GDo}i;6%}kKnZ)Thv?(!tTt=~K1$6Lht z<*cq?uiXN&u=d|_Ci>yV*W~_R2* z|7^q2C&zz*QwE`5(i=aYOIBZ>luO-_MfUe01f&LyLE zitkB)Xh915%YgX~HDPF?cDb$AnCGur{H>QC@LWjGG!@!O@2i8QY9aebi|CqQvNFVh z9nSt#UbeD&NKrL%xe~$K{g)PCdDL&G=ghzQQz2IIqHde)o&?8JfAnO&O99tKqrwJ7 z*!mFkuN>2EO%34y*&QIS^VzPdKjUxWo}Lj|T&(Zne|Rvl<44t4wMAuiNDgj<#DzL3 zcptt#+Gorf^1X*OS=1F|;gi|$zbkHvGQVwr3np6B~c?=9-t7qcKy5Jxy5 zneLTTy7V?AH4q7VgP*#R5sOEu&A#0krX2R;=dFo~W2Vmy&#gS}Rf}cVzbDytL57%9 zP$3CBfuMKE(gkIZa4r*|OXg4P56!U=nd*s%<7pMvG*&$}P%u#DZ8OsKiGy5#B( znL0a6(l)41CQ@X_4>;{lQeUC#=GFOJIxyv!mjYAwiwftZ?2*P*so%Svp#}@KS8!z7 zgix05SWgi#Pu5;{0g>1qn@^w4j`X;bqaG&j-Y~xmAax<+6%Q}VI z^qd`Po-NuKYZ^Oeo zUesKBvxqjiED)wq#3ZWW(ozm{T3d58h{S(NNCpMG*P6`12I2m0GOS1Lo(?OM`zwqD z3kpN(*67UW25u(5?p&0_Y-P&C3{aq@Q-plTwCTn^5kKB*m8(3}$@Q1F5X}l(+`i`8 zA3+x&0}}5gv8s#rPlL7QO6Rd^H5;pk7VyPK!GGg_WRr~u9hn(fZj%<7;StAGFlZk9 zhj&zjR1k~#C>p!T;q?Yh-|WxYZ-RiB{t|7LUueroi07SMUDxKH|LhDca(lqYXJY&L z483m;b31Y=g0u~a#cnh-r7!x=_FTqyGA@{Y93i&H+|1gz)PiLn6syi?rbvse);+q5 zFJIWl3^PY26?}{)lE*0;}Lg_Jsx0h8IukvMxnr=(%&fF;$w7n+suZBqTL2>eA z)YIB+pgFIsn#YfpdQqEpi=%=g-|r)6nR}86Aaat00IgA8>A{WhQ07Y)JQVsa>EA3f z-6jqd`Q5)FQS9PlW@bwtsubF*!Wqhj5xDg4!g!$+Hof|~_qeiUM2uqP#yF zQ#XmXGS5v~l5P&qjUpMfJq@$K9Ib0)!whY=>9JCDCaqMIVH!V}7L#tG>DRYp$C<3bftm z4eEZhG=_dqlxwh1SGu0njFw4o`rOMt`uehY5!(Cd0ZW{BQ}ZU6!0ePH_$6k(2ok!X zi!%SN2=prx0-ku4e6U=n^;%)vg(sj5 z%2%T8A8o#|;43AV#Jm+yc?$mIcWj?GXOi&ErA;~W+qP__`q~SozQj`Z!LTTVTciMr zE7k?FeA2<=#?Y4T)IMh3@)=redf!U^?!|TfAgIQ0h}IC*uwjq8fBHj6cg;??7sRO7 zh+;?HqpT6ybi6KRDFq@KQSXTvehGvh^Pw7-4Uw%EojP0AAQ6Ef^UDPqn4YTQcx5^7 zQIX}Bx6%7#De^h2I32?z4aUAV6CwHhaKZF%!X!0n)zkwKtP`na#h1EHR-u=8E|!k2 zo$?)g!KkKRldL`McP!9n{qH|KRP-7oV(T*5v4N~dVJ7ufBy=?FaDcr~C7@(NS}+RkDh8cW8rRL&GyX3H%P|pbgeR6MZR^FZ(_adG)1M zR7(2x6`OPGvar#UR0{%Uayw>W-o2PAvE- z7Wu-WpKxL5vWg0u6LjXNk}GuVPB$u8w*tUR^V}Kk8$uBC`4>wZ8Mf0&5?tx@zSWxG~tnH)2-;t{0J$DI^r)s>}6otnkoUDX-SVmd9wi zv~6bhYl`xeXt)7&1v5E2`3Jj;eHE&EB?#BxO-E{@JI-!;EHbrDEEzB>#hmlDS1wPK z74Su)2QkE>L^3E?9;Il8cIri>0TqL|?E9x-#d=7vid|>MQkSB0H7G{T;hBj~AYoIm#g;zze|XjrT4=*?@P}gw zgmcvTAp#UM;%N0=FSSQ(nGYtuuW+n5{va*ELbazPW85WsLWu^k5&Q}j`eU2Sive@h z7vSGM+DG%VbIzp&2lLpovI;9sV?SShrcBgj9J@%AXFlTJl0m-Ft>DQS(ptKAX_ix7z zAIPKaZ{czTX*sRilN5IL!#^SLo~wIdOX7CmcPQi${HGWk!YhjDSffk-(~UJ^(%<*hf% z-O2;WF`*syV{AR6$-C{KS08vhI%h zU`TcNqCkQI=q1Xno2h=z;sy$QEto`KnY2f#N!)*2an$h&X-K0hhgytLR6=qH0)v0# z2`acki@tQ9zRY1+L1A2ui^E@hi1N>?gda+5QzMsvgWQ*~>{^I8uy`^r&kLvfM-$0co*cs{a5eQ`qk|k=ZaHj9dDb&C zcZ`u!za{k<1H|(!q(`8z@*kk3v>{p(Ra#n#U18MpqJEk+si81CuIo%jvOLA6N@YR_ zik`)_Sr+|v!qMueu%xE;5LK)2=12w@b*yUhfqpgJwM-u=vzJA>=&O~4xW?^*DAtUj zI0`HmnWRRXn^w_6S%=vIIK83n+bv=N(i_e(PNxUiVq*iDsW(;V=quzW`L)wIQARQ( z6}zzi_AE6d2H1q8lEB5i#E-TOFCxgVn;nLSYcp>((VUSkpb8)&FEZ;g>dGFYg$w`q z1MtMt>WDJIp>di+h7{6i?DV&(65(xPahl-(q!lZYhq+qh-FDRSn(4{ld1pIrLim-- z`_GGC9Af<>UReDSx(bZOzVT0+^fW+9ZbF86`YVd{DPnw~hL_7bxS98xqgdmRIGF~B zJJ7@u7aF*(9pe2gf3a7vk{{D!XFUxT<*>qeaU7L=7Ya)Hf{Gy$e~nNahq~U1)wn@? zO`ACE5yAe-E9##AkHj9hCMMX>Eb~+;R>k4U6Y{q;dtXXoC zLOykWr%eB~13FFn8vno%dSNuLl7ogm?kRZ5Fw?0$+Q2Dn?OWNi;tn-rj^!-VMNg!7 zg6?VB*NF7H7qQ+Y`VJNLHn10(n?>(A|IzE|@xuXfioM;>*66fh_$}d&oFJrq=B=|$ zV|CH+0MyRdeykf&FW{rkhwT+hq-+sgZ z97e-m;WPeI`hZm{w9~JHWKIs=*bdYKB%|i~c)B!9Kv1bUTFuATmes~*a>n`3WVpdj ziKwhET_Sdvn?(yWLsrB;(1>H(tC~L=&*~U>S)Qw8dlB50n~ey)KTLSwGzd!eB6L?~ zL$5=>TY^|~4;NDn9A|^I8n3M@X|h%xt$L3sq~n76Y;{O|R3egMXTt9lu9PB(wa3@MQw-=JJwKAa*6j%%&SM zPnz(!7phW?njOzU^R59}OIjufri^o7mgbSEn43yy>_vREo|xG}f1vLmHX(?W2- z_O)lNpDmX$>Zo}bDU`@RmENL3@UTG%7VuQ(A;9VT^C5C7DO#VkM@{v;kZ2ura!hS4 z>`r;Mm#NLWg~H3Ns{it;J`%CQXhnM)6med3wtb`_zd$t*60jBe895ez8pFEYUc52M z7N2M?oiD=rhU^SC;$v%nI{p^BH5L3TZ4fke|0I|OgYV{D3Hi-I)PEygy))V+f~rGw zv=ZBPIpe~OuH(KD0xH7U--dncJT+;&tK6FlgW=#h5ZCoyh_Q$yXiTLcb_;UD6nDJ$ zQ0@F7+ClraCgF|8e2>)&r~XiOocOlJLf@Vof@4q;MhAGrhfsnCVY+2xFBfY>(hvVe z)DN_aJWxi%DkQj8xe*t3W@ZT@KAJ z7gS8{_7svpWpYfC7}dc9Z{~VHY)y8p=$saA={3=Lc_Ejaz7BCLXW@zexC)z3SlUJ} zCf?Hlb8t*$$x|wYZ$j2VT(BX=glZxr->Y+5Z&(Mp4}$QtMXt? z2q__6@)BdFQlV zZv`VS3Y!UT6|Lz&sX0Wb)|QAap|nw42^*w-hH(XrOi7`%hqy~6lm!vU2uCEvSH^?l z$)cLm#&FdLcpz^AOGDl2CHU<7J!9%~?Yf_S6O6}AhI9vE)K8`**Ho+{d0XZcpPoLt zEJ0xl_1VP{6ZA19*Ir$A67)NZ9H_KGPT~f%qkBGE)&SEWlI1NNbrp{Se00foFU=$t zcw3pW+>)d84GPWa^b^aiviN%!*2lGj#nmu+~O9GB*7-*Vn=#`E(+?qZQ`jj z2j0uSJeRBpDu)RQ#4#bY{ZcC|Vq=Pdqe+EyOYBtlO<|U}uG|zyG&|`=ZZqaqBM=9n z&tTypxGbYDE?tvKjdX3lt&{^<85C$8#)@cGpUOS{507VE0jIrW+7gQeY@H4$F@o_e zn-gS~Js+O<{o;H_`^0@e8XD{9YUiKI&cNlC>5A?*?SN&OZu!1T^())waJ*FR$E4i? zAU_OHghi<6&6PeU7gAyb_SB*!<9x3qhR~4)RlSj;uaaU6xNAr#l?(%YxO*V{m^XtnJ~t!Var3rHf$4lu!=7D zcoT1pTI?fOnC<0LG~`<~vesv;DenPS_L`i`kVORWg#&69mmOa)q^QZIInpxMlZI4z zhucipTw`>;XSE##kD)cUA6+GPaoOMoQ5;w`WM_cI%-b)RK+g~HxyQ=AzRzTY6s;rs zC!4j1vRGFl6+R1VJ;v#gx&9@>pFcw!v|K#q^p5yf*MhTQo&LJ#PSa|SQvkKQq`Kq} zdP$H~c-1V^o{)IQ)||YB=F1UEP^o`#v1h{}s{!gLgG4lwzOH3HQX7`1Wu?vmQ{9_4 z(Q4&I$zqk5c@n;5uP(SC87FlFTlPAA(ioEtiqe|9VSMbzJh?v+ zRuPS*7V1-0#Js{W{j*Sy(SAd{IMdu=K3`PO*WSj1k8A%Lt+YynzrW7GYe+05^?L9V zgt5R4>qPGR%){XD!+G_Q{MUnkpuwM*pExX}K~E_7sphgO*uF3}#Rl@L4JDpiwEFhXJwVEgU6t81T$U(}s1r zAp$8C3G3pV#z9@M%CixYrhotBRGUM433qfe0|SIE&epDWcBu#mrfL-}fJjq@TL{Rq z4JPp`tZ`pjW?GG3arbV0J*?KBzYcZhAiMDNxv&T7PoU-4YzMg*2AOjEF zv0x^bq012X^h$OCwU-rEG@6AYXw@97J6Nh7{=RDVAnr{?kLn0s@{qP(-y55Eg#9?U z$uDxM1I=Pokwkv2KO@Y)^aQf;cqOTSBa8p^$3k#jF9?v3U;e90^nQ-(ks^Dsd4U<4 zz&df}`U+I)@i6Fhii4I4;-NRZMfx&^_)x(dHv5q~Lu`_jqJ@t!aZprrfjLE^8xUut*>nAuB0#e zBmA{~3!(A1?XiqFAu|JNtspX(E3j<*9SSmbDmSA?Q}Vc{>$~C z7E=^ZawvX?`-K_Ae+NWuFm&!#>5$>@ri!E82(!=l1~0w^|M-5A4jC7Ji#o@Vu;1Tx{6YEwwQU~89ag`(Pr)%$s1j6A z23|>A)QW6d$uVIJigNDxRP&D(zW-bmid}I24=-7+-+hcFkmf}@k^5uLq3M4Gm}gl> z2V0ZL=fSKk9gS1L3=ZYVM-wJbOM`;qGiipQLthl=2gkXIDYfK=7|{P3<$snksIt}= z&I)Zk8?(|py@#tzL|h6?tGF9}Qb8F%J-z1~s(^%9V!+-Nhj8f?^`^C^rnLKembXx! z1M&BXbkD448Ymeo!%FLpptr7?w|2pdICs_6X3()fzdXTvp^ZSknaj}+CZ3N&q?tWA zmYl9ME zKB)M*_SS01I9Fbv>F?XMEp)~=d#5Aif@CjNZm`M!3{miiU>0PahzOPjspR*kq2xz^ zQ=OX9ap$3Pmciq; zK@A~A+(MK(KbY6j6CU=(UFND#KyW`$^s-FkMRu^gs18nv-qzfpRkLiFnC#-Z4tjzp zvV6LdY6*=vm(W)>tv~&7-)Rt_F(aj`#Arwx{A?cEO{tN)HB$()Oqlt;^V!$$uVA(o zsZ~Ukzbpe5_aEu$_3tKAej2)jyY!QNxB8nqVetk{?X-HF&U23&$)2Adf2*!|qao)g z^NgKl@(QqR>$Qn10Zj+EXu6(Q=@w(sR2x$Fu1uqTpIH{&DzZVh^N}>5_eNkdM^evg zgZJuT`S4?oO_g|+rlDglv?>*CUudsa!=dWl7U%kct;t1Rd0?&ov2zkXOOnJYsyM-- z2D(Q^Wo}?FGH~d<{j+P~8KCOEV{-5mm#IMxD@eFxM1*FQ+Hb54(cv>x&uC3^YxM-w zKYY5#L6&hj5u#Qe)}t(PQj+9F=OI!)@QIv8N3`6GNOqmU!E`A~!s2l8OBbP4un5MN zDgsNw1obKfypHGn!3tQ@0#Lpv5O)>HB424c1IRMlFJE1iv6(G{d1nK4FUx5NKbRa< zyw}})Xt0Ws8x;JDiA-2v4sfZnveh}k0S_pm=^>X``b6U|5<%TnIIH2A7)LW+|!Imp)D;T(D)ab(RdnSd* z@sF{zeNwQ}q2W z&i?E-R9Ea~NWg0tYQJx~HdrB?(5l1_c0Vp7A;`5RzlNi`P-+v?LlaJZRVQWsJA1Ck z=7eep$>I9N@e-n7{6)cbgfefp1e)Ca05Jbha^0thwDI{(+|h;FwhHyq8is91{=vMW z92pXXD-eWx_as-ILTixA+*+mDmc=K8B>eCX69WBLH+-kj9o{7e`GTkf{SK*~T%+!G z%a{IC&YAgd8h?Zv^Z3Y9KVZ=9%V$Mhw)?wbyh17=y04pz5=&~nSO?Q;m89OAt);Zk zK7TA{-uvk}kvCnj^*665MLD8aJ#j3@A}+|n-Q;zHp_6a9Bv9HXAhYwK_cN0R=1wqj zFp=ab8^ZqZM%a9V;aF6%j_S`<)kr1Kx@mOe@pW;M8{idt=}N#P{ing~-=RjJ`i1>? z)9LCavE+N&M#i7xjB`uZH~U$Nac?BPRu`x8z9k0Q;F0wF1+;}mBZEM4!P=XNK5ny9 zk~yA{0SuiP4Jhfpy=s1>tkq|}WtM9JUdN2Vf3jf@saPNZFAJZ2Swh+-;CxO;$Y;!G0fp-HF``kG7dBRx3wr@=!C_$ z!uJ9gd)=RN;M!eu|9K$rGB9+kM63R%N{3iG1^3p=-i>Zn?~T>6@&<bh@&xIqRCg;NY#1u}3s%lIpai0PdD?R? z`XauXCa8LVDeru^UU=s2YxbKAFTSW5EiNewOf1@~FO;-1dULUVg9Z{{97gkUq2{i# zb_qx~wpT>^E7MCZm|%>q-QP#S!WExRP^#EGXZ{8G8{k~2py^?E9S=5Ll8X?tKVZk5 zx}iQTTAR7HF2|A5A2+Tu$A}BBE^4$to2ecm2(W~`7x94$$y*wsp+i-x21gi!|ogOWAU{d#(;Gkx_{tEwk zR|$_jF+EF7stf`R<{xFr|It>5fQ8upZX^OaI>nq?? zrh_;d(jGF7d4DJcfqju=V{TqhnzFePXER5V2jFeH5>X#e7p}e7!M&*D6va@pYSZ1I z$>o0QKlyaHHJKhx+)a67TwhHSKHYX?R21Rai|iKq56>kmGOUTP25s7} zLLVvw8S1lGv`+VLRQh)lesDmrP?9V}!6J%L+!X6tjk}A*_})5&&fGtk+Qe|76#6W> zc%qT^D<42anVM)`=*qz{eSp`a>aO6oinVOR+$tGxqHTP%fgqZPhCB!Yv`flA&TKxv zuMW-Y!#>eANhs@voCbAf*|vE?3DAZ3*%$3PyK6FXddRg-#cLx}E#h)7llYUIxO zxD&Nr#96*6Mk5kcRIVdhn6qYXwV9#p{QkUgke#LnN_;`sT|9|6y>yzxNp%aT!5wa> z!yy|SC8xGei!p1R0nHzASKw3!SZbHaEn9od%?QVz@LF3Fs?H{5538>0nT!f+WE$j( z=VC0u(3AmzrI@I}mX(D&>)%?grO&C@B`5tM)_!W}r=p$v}<2B0(kN6}gMHT7^|90dghM7l#5AdRH7 zfRvPibWEff1L+#w-5@Ou5|eI_21)6TQPLYZVDNkI`v)vOH_pB1JkR+)6A`?A9LD?C zNfe1;vCxwJApSuYsE?|9^QQ&sY75^E{anop+B%cLr3#r0~W`qB(G95%v_7R1zF$LZ1Fn-mf zhJcqStTz4J)@>anZ1+Z3SVlGUohF2Lw*-wrRy$NH2c>eI<4Q>AmPCYGJUL){Fh^;f zt@(bq85^<7tQfeS*ZP8WwhkWS8Qms$c{9#0YEHH+C8#(LJ0(-xGpuV+G>Q(fg?|ZQ z!U@lQ(-cwm>X-7y9{u6H;KNHG0SjN6811NqQVQs#^Q1L^Y8dcZY7(J+H24k+q>#0e8)#VD$LtKo!&rT?RqE17Qa zIV~(=UO+OO$e<4L{$<5H7c%3d8-wvMRish>mGoQW{pb~iZWEW}q%l-vDxcOOp(ljy zh@Cq`0_!b|Gj9z}pnAvdH`Ds0DZX84^FJ(n&#=3l#}~b}{>&-8TuN>vMTOjg=Q!V# zIPu`ino*2MWzB_P1bxH|aPe8o&bX<6-Ci+1%za`jDOy%k{a}&&!J>oBEwR?(-DqEe zd(F0=Hp~X#PkMD(6OnjE?GgopsTv4L^+Bo(>4+>+zc0FuiOg+NnFF@w%ZVLJn%|*c zH6W~+Q)%P)jR5z zXtV3s2gSJ2S(MqZ+xGKN#Hfy^eW?TUDS=2nQnT~vnx-A*VzO&v*+0(Pz2-g?CAftf z)j1+C$kVYNw3mFO79-f`d8w{aEjfD`#~eUMnrPzKON17Uzw&QHw>fzr;7fe^D2?Qj ziC;ENa(n6>()wHi>njTW46`bN4L|fFKR)Ar#yUG%JmNWl^xswE|C^GTq6fc-Kakms zKYoS0zM4QY9ee$-O}{pT_P2FZ7e6;H*o;d18X)S1Oz$XE#B~u zLl;3YsD`2pKXrKV<9AJxk%@*;9-ExbGfU_Ke(qC9UpAM9@!8>yc?n6O%G<#csij^$ zR9Ib+XzV2#CX5oilkiyF_%yC)@vH0@+YX$RDmYI1X53fxbBA)Snm@-$`sG{R-5-uX~uHNOoPN*43VQmAVzubAtI znhHMnK!nIC>?X$9K!vq6f51t&j-Ls$T7f;dCNRYflVJ?;6Er>iXsc)I0JOvP5cH9G zwsWGt-2lz&K6`%Rp29bGMcF!)psE%kTH1f{`6n241R(f(I`ffuCJid0>aP!aAQMDN zP#e^82yEQ4tyR%1CWu0PgGABE6xg?kTBAhC1Qk05(;PK=Od%G{Ta)kn-ljVzQUm7y zN~*_2Pid9pG`Pgo*%7#blI)I_n~Ttf91XEGPfL_g`O`072%K2QVz3-Flm(1nBi*1N zNE*s<5FEmQ-_SJYZ$l44%Y2?W+O5yit+ic;c5Wst^`}TM$2(UBq{Y)&#DCWV&$7^8 ziZO_2xkoOqqVEto1E~NM6ak#BOCBy=Py4amDCrSm1H2(0oi@_OfNiMKqg)V z53KoK!jCSuE~a^fO75hKOP=}HSLNC){^Bhm^SRT~J(WH&ZvmXm4ss$9T;3kS_);vYo*UC|H%{a=v^ z#TgNoc8pM4AVj4KA!??V>cTu2ps_P3T*)CTvCw-+n)Eld-RVBVyYZHdBIRXqu8M2p)CVBvLvzu9wbeLj@O*5{_@>eQbA8U)jThe ztjEK4j#(^L>;3XFAjzZ6Wx`N0PFT`A zN~YI{nit73(HNC0ndn^H>0{}URs{Wr0Uw_QrX*%M5R^vVhph4`g0Mx-XHV?kr_>An zj`4~>9PQvrc% zcg>v%Z1@`ZmBoIl*yHSiYlU*wLf+vrPXzw6Q%OsExBX+PCHWVz_*sK-7Wkkn$5E|i zQ3+G3?S!T$&KgVk$e+{QVLj)%$bsUY|K_@j8b!o0O~-T|gSHC%cKBpJm%qs=u&2m^ zNFdULF3KpJG;zM3b%tew0~$KE#fyKLJRfG41kV~qDb)#=NxWM}iRd*A470H6Wg2gA;aNNz6j381mdv0U_H#=|pUAyALe;FGxn7hIaOvZ0+sN*6&cbe;b?zn_+2$+vjOZGcyrz9z_ z`p&B63``t|w@Q`i^Q?ODCL;PWL~o&#Ca5q^haG{pt$H7WS^~l7F~s^fArXC<@!6=62fJ z7JgHf-`#P7HOc?ft*Y~py8}M{!4%^6rydiFr@ak9dUkgYcHcZ<084q>U>DO7B-LbD3fi~MBBoW7}>AA0-s2u2(Z^Ml4bRHHA z<4RgN4`>tyvPAY%k9cCsBWF$&lIC0!{p>B;U5FFjUVQ4&P!Nye9lbZ2#qd>Br;>+9 zUd*EeWE(8wEE|@FTr#9brvqbrceMK4u!5s9gvN+A=Q1443J$!Lf?u_=*%Vs%M0?5p zw$v|5H5JBX1d)_!NuPokd+E7)WvHazcosXj#go~sF`@`+D`O!$6Ao<|=mTgo zxd?y8QCS?Fz&%1H9F4>Xn>`5vUnQW4XGIS8{0nb z%9fEO2`i2NdmpMyO*r(Z6RQgi+1#CH{|k4(u+{W6Uc2WqMIF#f59q#{vnlNS@@@-9 z{!L^!HaQwtn4@>Os+@wcn{B*qH!DJu**z+%Z)}LMO^O$860*P6BOMreCewyPoE#?( z%R*^i$=r7~REW+AOUa$~T^&-}QqcZ=T>swe4gF}IRiaTgY2bj8a3dp+mfyvZ%WR*4 z>EH!^$-|-_4RC`*;)XtO!jzG`<>WtoOO`+VN_(GSY$;h>mt#qsuK3pTgwaFp2MwwV z_hG85_M`#DIonQeiO4e1Rq_GkmEDe}X*a6E6}pfX0fWSI4z_0>^U|5_moTJA&7;W8 zQ+%}b14(?$q1%RQhnk8|p2d@QAeMO02;S<%ZO^mrx7gIT@J+VzJ5dbpbhVHpxvA<_ z>}w;UqxIVTX+u)B)9=Q{NiqUuEo>dvo$Cj)Uzl&u$XFu;l`BnCGWj*-wq9UzY=u z6rI2_hd@>K>>)I|OGk*TwJS0UwehDW4aU+T>80pLs~mZHZ^LdrI5w|Ro{VK|TY za?I~k0|_6Lx_0{T3i@=(kxjxY?)7@ICc(R~PJQ!w5S1 zcv>V)zdPqKDRx$E^S26>-E}vo)*T$>nvfRu>+{YT8ygj`lxCT$gwAGMj!i>Iep3cf zS7jY1Zk#6-F~Z5u*P?+@#wB{uylQ_bJQQYP3BgH-%vW6~ZEepLaQ+dJeEeG9p8}xz ztLEl>{x@y7vXi|xd0V>a5hcxu9+*2co;=+{@!b8AHZAuB5x}56`mm^h`=1v+k zLv9D!w%!)R=`T?ew!0{NL-(pXK^l))ws2kjzBq1yzIK{V$!rvX8QXPGQS(~B_5wp9 z=&^qOoGSiuPF(Yw39(e(Jnyq;ka`onY`?Dp;uH_!C!YQ(de9I5?`iY`RblH?YVOG? z?oLePs-cKxN>cU5__iRf?b9cIz_!m&x`wR5e9HE(4b|3|QMf&5zGJgD!X@Q6Lv9vYE_6vs?2C?ATim~dhTh!- zZZpPT@GUBB?e%6Iv&f_22oq*0w((6~<~)^Lypu}Vq4Tl!N##Ksc4_t9Gg&%pJz;h5 zl^?(wqODT>JdXjJ`z7N*O?S7BmW=x|oF}vX@m9`S{Kkp=+=SdXH}%zr*PcLd77xLc zGB$h>E}lRBNmHYP?_+p%_y zH;yaPoT(yWC;j--P_gn-vJ#cpPKw(W3|x*DR7J*zv!LtB(edva<0kPXpcHf9x_!GS z+{Mz|`SD-Nj0(1|Y)Aj;|VR#y&T_BI5I729=^kNr%Y<~|+@^VpeR||nv>h3wb zB`9coAYJ8)fRXX}y^zX_q7lo+zt^)zsZdKQ)eh^fF;|?`eSgL^Qmx44sD_+f4|7#_ zFwvUVKo%z(_j~zJ#7h!6a*w50D@WQ2l>BoHXWE^cus(Hgg2FdKu{EE&z#X ze+tIhq!WdBFnOn3@pMwhbli6BggNx{GZLPm7gRrM#7grQOKQ)$x8f-Ww@f&9AD#E( zMEv=EqHk@LX(L(s=^PsQ;)n!`tro^t3X_?$?wpZi{Rt!iJ{D1mTQysy@4g9=Fc4fR zcrx^5d4XVyt#yn=13gU{%U#8WtV$2g=lW)UsY(%~Z=N{KbCt}Yy6r)|MA*bLpr`;i znH+>63;oEdrV!YxDx^G#a>6e zE;Rq~BG{;rL}Axp(8`LRiAkco=|3!`KiC1)8TLe3<#3$FNofX1CwsD9ExOJF zF?`IVC&|{v-_*MtA=BX!w9G#ck+pP42`87rJvOSBML=ad9TF`>}RC z@{uIL-SiL7b0r1}R9*-V#?+Wo-Lb5akMs{^FkwaL6Cxd+Z!viL18N_VO)pF`Tx)oE z4Jr!Jxc9)!awD;eT9USW;wKDurVwQr$+|TyOc^(*8}sF5;vXkVpX7uh;kHKs@+9go zvkiS1coR;%iGzsa0J>?+>457-_-)@(H|E6r z4Qt5NiQgIHwK<|RL~En8#&@TBFwsy&HaXslyAdrz{E&EdK2i~eR}YP%HwIy2|6#S| zG0JQtQOJ;BbbPk*0y>4;CfyK!heeVR5!E>5lEedP==S$fw~LL6GG0!Ey%%2wbJnxh z>-m1?#jZcEVu`+*Kq^+%{tZvWnRNStgn| z9#YbwcM>r9)=Ap_P6>Xrk*Z@F$E6uaV1*wCv9`dGsp-D-6IV?50^s*XOVgv}Rqm{N ze2w~12b6jrQ!)`gT&iN`Up!hUc#}uoe5BwN*{lA>HQ`pphp_4>H9=4Gr7gw&F@&G; zxhZ8T74Qgag03jC|jPqkfh z2o8AQLuvLAMR=&eKCzYhaK|TW$WViWdBiFkd7+}J zKOt5ETbgMKQTuuUL9gwYKe2 zNtOL6x0mwDpo^G%Kp*JUO3om>Q;HC#0*ngdawSl$$}kB>fD%>2pO`u7y8N3{g`H6b z|1=p~0)7Tmh15I}H^&HZj}Skh#aQ7}*q~7kiOnTWEVEpH^`>%u4*k#~6~TcDZB(_> z^+agbMwWj~Q;6^g(G$;y^FOVbdQs#k%dyVvYf3!P8}~G=flmQi&+7#Q8GH$i2F%Kf z3L1pMC@LQWZy-%m(mWVB0IWkPa<|xs=ZbK2hy-^uc_$}MhvG^p_@wtvpxN?{bx`Yl zx>IQD&G4A&sEzCY_GuC#W?g`SwAy_s1qNpA_)Tv#mPG&EB(^CZ&D5@KzaYbX-y&lYb-O81P% z66!oS=DtnBvj8~E$@dTd*MOybQHd|gqXlz;5df}i&vx1BLvQF#S7d!ggO3roUbR3| z^622^t$>0VtgNQFC+Ls^CphNS2i#UPJIymT9-EMDr<{z1L5)r4wcE|FWQd-WA^ye(I+AOyAJx!1R6O zE?^BeTypXYWnTC@_?$Q{svip-4!-{(V_79{+5vbja@)z>K-VP7=ccyTs!l9|B&L?q zRL&stk)W}#3HB<_20?rI)akpp7%B0Ot>3infV0i`-z)|RAO*l1*P*8c=;WV3M+_*^ z!mrcTluh+5MV(SUD|&)Slw%Xq)Ka%Mf&k1Ha@SY1kC6ag2&l>=_r&1U%qYHy2l_SkFqk^RM50f3bO7(-&a0rY<@t?F=eeT+ z)eEvGF%(jvo~eQo|I)(Ov5y1nG{DSH+X{Wf&?xE#* zzP6z}wreQM-Sl;Ho33!>=~CChR>TSUZNnBkF0ZyJpuuPv807)Eu8!dkOv++YZanT1 z`dgNAe}!j@T%bB)E8lb$y>nCl88Y_sh?>+k{6YcAMKDz|8VKE?%cb-kL9tS zeQINTL#Ibrh0#HiZ#q&QFfbtF`%TDUn%t4}dBLZbInsYtd82-)I`VR<@Irgt7O)#N zH6Sw>dE8JpM+)0H8G0Bg!O|$dP$n$x)@`neK!qg$Rdr^7hv5G6EczP=tIzyt`@Zb@ zr*i{4Sl*onIvAlB=KnP>;#Ma5sQ&8L3`@LPeP`TFQKplJ8{Heh<@+>Nm8Z@_m7j@w z85%G@JO_2=g^gUNT!%rvRdpi-gE**Au$He(s=sU(Qcor(mjXb)Vy4y?@M@NUvC6%+{1>c!iO`hZt5q_+VB+nf;@>)tQW5WT-(zjo>cHAYM zJMbpno_NoF=4!5A6gDpj-(~FW3dxiSvZzHd2B+-)PID!w{ph%B1G9Htrv zA)jh1GK^UDu@)L8Zy``*;S}`~LZ2(u2KS#=heJdvKs7}aR~7aH_*Oup^)u(!`}oA< zdO>qB%ij#^cYNqU#Pw16mKe-ygS=rHdb-_^r#+RQH+1K}U`?sJytMdS0VGh3E_#$H zppT`qB04ofOpz?c!E03F?=XqBR+@9e;}cSs*G4|>hEhXu@FiX^)bM0{xf9Cx;_rT@ zc;{#Q>TP9?_ceeb3-Qp1ze+$>XLyxnOi5S(ghk9dEq|Uywf+0d5~{1TrAro{8Qmh@R=2*TQ$rNv ziL@tbl9C-Q-c^`ytbWut@to-PE=DYe@|B>yuikPCT_lp)yi*8lZHkf~saeD16ylPH zBlM8=*EE)UgFwimK5tR3$JLkVqiVO1kR3yldE!NmrWfv;`sE2R8^>EW@Gjsf6E6J} zjE*-&N$nloNQE`{(%?&(avl2Q$R9ri;44s*52G5Ikp(0@7#!<@A%U zbswb66ITy8p8PfM0LnJiYeH8tanU3w@MrK8o9x@M{kPv=uI1VKnyhK5KMzbrw^?5) znHR(4k<}@k%*d-!H-rVrNz-3A&4^mu>zrEVwMJBHb%dR z^A)-RLHKdaWPC!1M%83VUr@rb$VdOmus+7imO-lTn*6uj11GB2l;xn4Wu_s_a*=Az zO|>GZ<$+8|0E=IoSMZ{`6(+t6r1(+!+iEvV$<%-&WROOui|`GP(8AeHM$f#@=*8-F zjkz^7-SQu-%?e+6Ez>rcRY(*+{xmb7Iw2@;0{JOJx{xN|=rnQZI-;J5;0Fe@7lLa! zEK#@ua;#*eud_|`8lswIqzWWWU>?5du)7cZ^Cvt$sZWhHyJJ~03Ex$f(BuQ59%`na zXwGNO=vl0RN`e0UBhI!|8D zHb91}JH^mJHVv$=j*?8I$3d}mcg=`~8WK*-Z;H6j$=Iit6Tfe*Ltv~v72Gr>5s^+6 zrp)Urz8Sbns?96Y4LxZoHDC2>xcj!x)RlQI=AuMwGN(R6W9*?ZuRZ*|w&nB6g#X(7 zTC;8Ks5f;seaCB_Y`>VRhHL!tE!O$KQ$Z7#LOeO+==_v7BrfW;GSTb!7{`tHXeD)< z1}W{{+n*KstfB2m{bZ2=$UkO^I>UR{Asok3ogHW%q#_ZP?L@_)6Z64snB->m_X5I1 zssY85P}G25>VASxL;{QCq?2CONd86oaA8p<5Dsq#Q# zF5z;T;Qs-WPTQv+b&<+G?XhzE(-Zhpf0cgLQ=~)aT0@+Qvjt+aspdyz`Ua%hA|>e; zC%o?&?&JL-hw#O=USP$${_2kH$2h#?!#@*!wY$u4F#ksu}P1g^6t9aMEr zneDHA-5SY{-9d(O0((^3d|1}G8h)w&q$xXFnvwl&W5en0E7?f_NM9!~`&%(7O#@JB zQT*YH>8oY*Gn9O}0j| z24BTi6=(j4Qkvn(Js}G@)3Xrhel;|Nql`DPMt;|^af*K3bp%JsbW_|ilvWNBmgG@| z0jLGZfokt3VRK?|-n{#L5FRcl^AA~tcAhZTdP|5oUWJ06yL7=khV!;~#*ZI8vPsM1 zey|8@GpJ{8ovCimprRj&Bml|L?)p|fEDBrAJ)FPK-F+DByG~p+j%2Hwyqsy8Hhl$w>aBSMzy#W8l&Kppbe=P0jVsuGO+4cJ5RTs@`9vY}W8rKnR}Gp4 zxVoPr_l{{1Ge}I@Rcn&s=6eW%b3yW=>4u2RniXTHuaJvflxVgFYgpFxS0CQ|B2Rz0 zR-h{j+=xeK9>fu3#)V&vL8E^}EeMuWWEzH9S2HkvGHc6{%4}Fdqy9&9JQ(-&0ijD(Hsb;!j8gqJm zv}CrW!BigTTHNt5dT+BS$~8k$Hh;fyZhkrg1UY`XZNe=eqN-+)A50=hcjDQa_gz;e z%#Q^Isp$YVW4^RPMfbGJZyNQOh1HJFx0HD@3^l^Ps)W6zx*Pn{ZoLH#za3}+&?&v( znM*ujidHVWImD@ICL~n7`{-ouZDG4sKu}(XcHr5TaH-2C;IY^4Hna>6Hsm9^cw`|$V_R-rv zVa+ks*x18;OCsUL2FqZG^kbApahcKT6Cai;v%-i%3;5Ee_P2i{sZV0U-*WzDaQl46 zgSUG!PNaeM`2myiS$_;8ndNDhE-~pDUXuuhQbiQd`WVf>8OZv*O5XX>**o)MjM5lI zC-TwIGupkv+d=R@ERq{@{YtLfvYSfsju^i!ty-GE{|GOu@uw>}qy_L1}^=KsSICNNBtiRLyP%VZT~u?XDRan^)s_1Yr5 z#^y2*Id=2f5ejV__E#h;$DGU5tly0qHAl{&|L}%{o66VMM+)g>2E=*B`Y4}lC~t%x zCqV5(1>Wn88;I|X=t}&;H)pGmiLk1vsh@`7!j+2E9Zp_N!=amLUIEn4k$)0b6#)iw zEz`g|++YJm^?euJ)SHZL!@Zu<@ar`x;I3J9t>^%<2_&LN%1(KY+AnP1?5uccG}aB^ zaVqmQ6{J3Gi-5^*RpA2-_6o#XYAxN%rO`8u+I7noZ-t6lr!;Oa5o4-X?K7XhFVfV? zx{d)ofTmBgjhBb!9lmE`JsP$-TA<#I$>y{tAzuwDO@qtKd_k*}+Me>9lv|ohEvOVh zUhQZQM^g>OS*8*0xz8`t84Qkk5X3UC)Yk|mX=;yQz?(85(p)I@^ze3VON2V}AxU+z zhpm^Btz+tozFyXt37q8s_K}p0V`{a)tjw-H3ZzJK>AI(c$t2EPVv#CEo$Hg;%ox zCQ$@z5vTi@w*koM88!Qco4hN#qY{Ao=a`_<1GX6f9~@Z^=^|okcdbDeSFDU|BdF{h zK+_ukh8{_xpUVM}UaxS~u7CKA)uOY28;tU!^wOTj;BDE6<&;Eh^4e5b*VKX=tr^7TUYCzt+O7 zsh&eD7R33ib>2kV}u zx4oJo2h#E&^q|$-k`#qEn5+j-?LNW52V+_;pXNEQF|NI^P z8iMt)b&dUdQD83<_BS!?U{fa)-vEH)Z`b1)I|i=w!ajkg8Kux7=~+y z+-m}zal7g~H2xyvC)C@I-+j!oubk&pW__{P#^57(=a{@-Lk-(z(+Q6=q=;lKuQz(DG(8k%11iF|AKP#Ne`%{(%k2IN@ zra^T@oep$(^|FG+SgW~{2PsuFKl?q+i7o3#CEO7(Tx~9%kZ`37p2*)fR`LEeu*;V4 z@i36M(l0_Yd-Xk@K4ro_-{gwZ))^GZtJLnLZ#BoT(SdsE3@PSu$R6 zOhe2WHS&MZq*k?kU;Ywjo0%LVD?GXMFtT2!87AymF9QUK;;SopE0L@i8mn50!Jx_x zcu}*%GQDImecR^DorNgPqs<~u$ZzES8VS-ZF5i7hETe&I(e+R(il5+lBuiWJ_%J=W zhL{vYXi9^mpO|N#v}H8@kOKSa0$O(K>N_wm;|1n{UzcK6m-J#sn-cvRZ4k zTIfmc@glTt8EbyHi>Ava{-rGqTXHFyuSc3CTyq+l#wCW4KD4#82|25R?2K2xqzm=2 z`hvk0TiaA;UTfdSnj;VUw|GQ&70=ImLdRk{!YCS8u4dcqrpiS!E23bsMk@dGn|l5zRH5hg}X!J$57=}#k^d8Qibq90D?1+OA`Dt!ev zrIdcRR#lAwDFlw_4-8h1|Cp%o7$6=<>5>-uLQ|a1C5?LB*fVeIHi*Ki2BRE>!8H1Y zXstW7I&L>A^~zx(ZN`kR%}xEi0q?kzlo(#%Z9Ac(af2}i8@r9|sJObaRgXmN`x zfdE;~4&CE+!WQ=Ze%}nFo?*v~dD8j2cK@S0gCOC7AIEkq7F67$f0CWr#8~3IfTZf~ z&Y1mP1WV2B>4ph=&lx(!P8_$-3uaVUA6L6@fNy_IOj7z)-kaVkHuy_wj^)q<*Rr+l zRJHJ!o9hG}Of)Ye|6yWwwd$0Dr;^&BnmE6$@lBGb{mOUTT}xa>iUpWb{Zs>&Rrp{l z$$Gy%s%kbr%C<1@wr3v8YdLZ!SmezFmR^;oc{feNb>#77l@Wq`?1J!a5qe35$PUHk zeASjQJUo6R&@$j22N|qVG&DXb)#5WDoQPjqAHcu2I{g56RNo#6|I5&~^>7b57~x@f z#6y|rMTyHA{o#?=k|NixW`sW$fKpX&Z|s_@Z@c#;!Ir%C_K6-@X3~2xOrTCU!$g;2 zqsri{2s!b8iJtGJnaf~H1*8|kN?7i_NliQDPfV~Wl@CkGG;IcBNz zv;_u4OU|0lKR=To);E25p9v(bn0M=`YygPOo>s_;$jF#&&Q44WDSmYUi5X4y%d+49 zVYzATPDvdIfK+*>L@@}%WoHE(`wY&|fZpiikvs>O%&(j`XX(n=m@ zo8j$t{zHO~aS^c^5HKO2Mi~E%c>|#8>JK|=-{l%CN&o&VuEO>qsELGwif0Va;KTea z;q>eO^crQT;XJ&Up}&*18dKe_aRJBr)g*Y7Z>h9j#Ux*cdjpnc&0zPXDtI79VzbI` z?uOY3T4c8dt<)}fJbHyt5UEVeRh1QF!nv%rizAs@pC2vjIsAUm&Xw;s;_O*4sE;H+ z9uF;RKad{ZIl?(|YL+T0gn)jKzsM8i2^(hXh#z#k`@HyX`Y;#o7ZM0=Z|-t;Rq0#$ z45uZ1fR7sdXzBh<9jU2{b0LET10zL`n>HJKWi|;rLRwv$o(r0R(x!|8oCPCF8&d|e z&8+0JzY)TMAt)Babs)yltIY3jYv@{5nQXR6W!*wJJFQ7Ifwul|%r}snQiE`CY71f{ z^?hn)%bo33+e8q>by{X#6c0?Dt&&NHrX-iIO6xEEQ?usW*_hYNW@l+?STBw&qOPhQ zPi*m4Am`;#@!dS~awz)B1nT4JhPDhdpL#3qnAX@NuP>KRrdAUYUc7}5%>XJ+5*ogl zmTzVa#JjUEO)b8W>aU-7QD@=1-DwNdXWQ_2boI4;o1ae41MWB5MF(@-tm5lilb(iO z=;OD9^s@X2Y|LA$Py5Oi^77m1?>)lcdC`#EYUTMHq4=p)tIH|?Omf&&Ut|~-i)}hg z-rtxkJYLAsDQ&RXJpJAoFJgEV#G3Nh;*+Tgav2f+yNyuEFnQBDLa+Z!F271oCyh3x zHusFOH!n1eB1~rDHc&U~jD6n!G09TFB)`@eDceqX;`&rzqkNioq`ngP| zceloOh2j{T>;Dp?d1nEmMrol-2Sn3vkJ3>B!;t(X1!u(DVSqglof?SeYEoSTMxAkn+g!o<-&HfS4I>3z92=*}O1y^*A_%3#pttf?d&#N1YJH9MCp3{V; zKf`ptE{#tK9!Yx?FT;pB0rCbo9Olbb@4BA~zwlMqY~~^%k?|1j<;};g^Cwm1mw5MV z9C8zi21KeSDs#XN!H?mpf1=*q_1`01@9r(D1xfR@syrl#!QD(>sW{UWKLzQDF-)Vh zAMo8!E(paONC8@D6CA{2Ikg*ySh_uUcDee2EwnFjDApN@DS!F1w(*Al_sbU?|6y6o zb+v}mAdNbR{>dYD^Q)snTTvd}Ly}4bOD6p;wiWxomhHqtn--T{^+@DB=bDy#%Ot7@ zj&s%RVIOCC9NWRqX8btliNI}*ly7XEBq`I*2!G&;h_0z_J zDE@IG}Y9*4B5KBZO)d=xH*@8ODlJ#8DR1-m)F4Grmh4= zM^_wtVtK`Xz(;T)O(v&M~d1XYwOBB)H z_MJ0hIEqy4&k%B-O_iLWLTcb`o9`vg`7=Qc7{4YAdHU5#4O*44+duq)Kd(nz^ZWFB zm_*T%k)$e120e3gbDT|KXyn`XHz{;1ud#@@C;!88gfHi*Ot0|u<(RsrD8ng!{IYu$ zeAPdtN}+tboUKbTe6VpwK?rb@?N5QKQ`CdpG4L?S`mOXi1dSPCZ5b>n}ZS!v_?4B z&IBtWqI6g8Z|-YkAo18M?=vT&+qFv$I+^O7YTElcXTC;z@Plgbbd{&!=|W5v+v1mS zJoFoY(Pq7+Qxv&@Uv|_$(wxg-T)x=0D`5D`(kafl^7EbEd35y&<jXZ z{NcI5oucQ9lQqVu1bQE}Sdpw8;pyW)vX5XxATF{%MUR&UgOVKm9@3(~7q)vU`ih&2 zWK||RkI~6F*?#?z7}R}Y{r>h#l0;m?uw18j^iHruh1oYZ-y4=|pKeB^I#Hlu9)|em z^2Wr`L={kyo++fqxUTx>p-kJXpd`DuU(zTMOOH55$fK5-n&MyRhpdf(va$hB!33J+ zBxiF0(8j|;YHH>#TTj0? z7$2r`UsN3E>K`G(qTPzji?Fb6ojP`5sUxtIgOA{v>StUf)-B za@jF}phi8d45^zEfY38m;#+{9ZbAL-Y*~_78^Z~TQy6$;Y}J&;(4gb)a;M*alQ&}> zBMe^D-r%isaV&u;qe=rC@XP#G+*t44FFh`om>F{8@;R20s9;K^xfzKT(}FT-#k zWUwP838Bc-fn1W=S1ESS%1nwovTHUf^GllJcK@C^S5_VmGaWzfN^``80D7O+dQ5VJ zAMF@IqMUd%R&ulvFG%~k6R0`Xzj4f;C3y8@p4P9U*&H!s_=vlMV8mRW|C1t?@!UQ2 zOx1tY42lxz#7g!%R{=(@S`Lzgdr3ccV)^{Q^ZBz%q;-v5^P~3}Xu#g^g#A8&uimCz z59K(J*%joLgQsH7kX2wutp4q9Z_|jUSxp*Z0Kq}#w=Gql6lyHMz6u`)11_`f-?|U< z$h5Z2_#i{V>sUCugzLR^zio+aJbhb17^%Y77}ZxJl~tCsH4@Z1@63B)?7&UI0Ms#m zFQTvp$Cx>3l9`jOPcPzqt(j&lK-PBh-f7w*w5L>se$C@v@F|%<;1Ik!4U~EA+U<1r z4kji;#uqivwS;ZxS?EQU%A2WCwcY=6@NJZbS=@0ygg}Jk`K3vZRb!ezu4~ezXZ?|kwd3M2Wj7-}Os5PjPD~OSW_72Sa`aXfZ{|mHMC~O9 zVITKX4RO8MxlU2^hqRg>Gc4Iput!bgOyGlSjO}y`1+{5TVI?vZnp=5zGW$TrsFvH1wMdu*Si^Jb))h-Cu2+7~W+>L!V)LM>^{iMM=A(E@oiAJt z1EYF$=F7xWX*))8X;1{Y&u5@8y^SxOH1l7Yy^c@oA*8$VLIJ0)8ON9|{uAg#!1^08 zobCHZWqx!2H^D4f+)5XsTn4<0Q-<*0t-We)(33=Vv%Yxq4-dClQtwYL2Z|_aUQS&c zRT{QNE&3JV*+(My%>c2cG%E3v;~FknpQX~03qHyyY{-U>UL^_-Jd(^ga?W7M`#vvh z_{;n3z)ibC!frfW4&@Q-ihtL^u8~XDHDe8l{yv$(ns_ZJQP(&PDyp#trGd7SG$3=5 zTqtv0B+qtwrdq5MByrC$xZ(So?k_|VC``p}j=5&r!x)9=)tSivEuZ8unKjPdHx^v3 zWo)VQTstBni24t^m-|an=-J~RI@UpFngTYzB@`D2j3k|Q9fl8-3^B_Q&346x6Ql}< z3wN*znebn?ov&$?*qsImus? zKZ9gKd@e!0TpOyIB#Be%!V(J{W=6~trbPr!LLao~Dzw#zzu=C&I=ZjFZ9I*xh= zMEZ;HceX`d>r-`iGZfN(qdY~osmE7y!zX_6d@DWfVj>LuKLABRy1pHC%~Qi#4zF^S zudP^Gi^*&ynE7zrZDf%`q>ML{8AVcBPmnPm2!7VTvoDQ*W8d0K#UHZw#mJ%l-tg@4 zOBSi6S(9z3Y4>VmvA%@5$+0Jl5XjA#V|8&?(+wX?YZhm8=-{v>yz;0zxdgJWDy(>~uRI0+00#E{ z*4`@6oBkJ1jl4bJ`Ji(my`o-gWt5CA$sjsng$Jl$K(Em&W*C3xRz6mq3a8pX5Mj8~ zZEf6K+b!DlFLN?&VbtO@Ku6#+^sUV{FAlWuCx+E6Zsi$}Si+1l2Rn-)I3A?e^mp*f z{tejhkA&|=t?|d>8teA9(grZ-b{}JT3H}Cqn~T{10O7o_q;&?q`S@%900grC0D@)w zExkS=@n^%?{Amj1yuEUHb!z|t`Ab{AAzXox<|Y)3lis|#nYAVG>UBd3{{Wc&I4)(M zL(`{iSz?|D$=@y2!U9wb@{1T89>eQh$HD&q0(h79iPmB8&+OfyYF;VS-AM90F{hi+ zz~sD$hB${ozF84Qc{w%x+xU0(cm0pNE2@oe;E%$u1=z_HNU7v_cEct!h7tX&K_=0_ z1LfKWIpEi%+UvRpi7lT>_?h9YKf`TeJji02*;d^InHylcv}mJ{Hrz2Ptho7=LFbaD zaP+hehf1Bgzw5F5h5SAL0D?V#!AU%CZy?sb9FGoZYbyD(_@ep*aGlI{wu1I~S#zDM z8jNJ;j(<-+3jYAXzrPB!4O-j8{{XZn#X%%?AG{hrg)HOKVY+18%Cg*F2_<|H_o7IP z5Oc>B`@65`n(nvZYkMskMYoCs%x?;b;AP`$G-_APUI)su0!J#@>yN5{6)|*DC(LcR!Sm9o+bPNb$C_ zqG+0pyV_Y_TtjZg;wc--M=Ks!5zgFX;AG~$eDQCJbUh!%9%iy_<&EMBjUkMnJx+1J z{A=c4jQ2KIe-yk!HOAfl0NS_Z$>>A#&Ns&YSQOYk3L)dl_a}dlYx8m$DS0@ty;zg`%CN} zEFc8VImtK#`+;1?!|&NA_MGwMH3>0h8f z3xD8}zqSvEV~bGnZ-;cx1n&We@gA#f70$*u-5tHDx@Y<`CvlwdT~yv8e(WIr=FH`m z;c0&!rM#O<{tw`OZ$16~p!=*5;`1?p31&F%GlBY7y#CA{F7XF~J}BGS%{Hm3Jion7 zGfnd+xRHhcA{fDtuebq6eAn&|fd2sCl%E2A9oxaL{D1Lgt>Ps}_sji_Xx2J99F55> z=4QFd{njC(9OsPSexm#-_&@NM;Kzq&(S8!>3*h@p;Ha}qf@?`7UXn;RTg@Z5;GZyq z#{}1nXi)kbv|kMar;AGc$-4F_)+n*!8aNWo!^&i zmb#d>)fjwA5mBI)oOtH$T~zyyQiDr^hR8a1M7X#~{Ln1%6ZZSw>Yy z2T#~x)}E)t=9#5TWmj6S<)P@>p0(p|0{Cj%Tk)Q?Hl(oJG*55g8!2^I7XyiJ>(JzbonQ1faJ&vn~NL0HL$Th-@YbrLBr!XuU$zurstZV(z{1xrfB~F z95knr&2Dr%`!f`d_ZAV&48@dl3Dait1Gxu{w+^QS5IC&MZCAn`G`oV`Y&FQdMW;pO z=HA}w!ssFmvG1~wEKZ7gGeYP_FmYbSz7Fvc$7q^O<-VS6W68AETkRfWjIxik#>QC4 zVg~6Vn}9NNlTmo%#oB*{wDs`!#m!DpW2i=nB!o#Wo=NA)7bFOV8=eeo4m;qAbstj? zLj%Q);+tE*`ZdOzV$HB7pMNN})T8-wcCmTkY#f}C2?LJ3d3)c8{t?rBNpIn=h_XG^ z_nqgbo_@z0K0LJ+w^2zKnpg}j)^ek&WD4zM_<5$squoW|Ei4vs6GI-b^IcdfsmX9% z#KS6nP$Xpp0!OV~CtdMggP~sQUlAtKv|B4j1(QtMsKV+_R#lD^GUXY06Gyd4By`Ot z`;IRA2BvJ2Nt6B&og&pQjLj@dsy)n+Y&`j9RASt=esdcr>Ot#R*Z%+%JaysE4q9qn zF4XL_2$;(bZX=HF);t}7e%BI&j7gu}<>O`t1CVfN)?W&IeXTEyd=IbcZ1*yxmm^M) zNoR7XOYCcA^F%E9KQv7V>$|2acEo&3vDe!~x$u>>u9>V#B98Lj^(3Asa87UFxm0`} z-d~yqau>Kwue84L-$TK*kA@y7)D(D!!8+a5+FZuV942SJw^;$qZe(rqMu#W*eW*Nq5gS879^PDg>o~I#%u1)OHS}Nh@p?c zUI+0t&WksPmPKf^iD7|l-c~#O>q$^SG8~`32_TYD)YUt`4EWPT&@Ap{(PHsC!hl@f z+uTQEFksn34WS|@+=q}S3|rf$R4Ut6V&QkW_`~Aw+3!}@p<|->>TOoqay-cG!MakI zJGO{fBTdVYps2z65ngfNpVHGUEu{iUgQUpW}3lu7&__BHZ zqSbA_8pR*>rl)Ndm#ynI_>5e|6{evYD;?Mnjr)NBf6GJed9MQ2d|{}3KDfBe{qCb{ z90lNlOLVunGG(y8X#^$;EqdH9BSzdOT#P%);P!@5^*aqBxJXw%7sR+aeGCxIzJQs%vz_4 zEhN@F3uhmPzqO)ejvKJ>e6hI$7IP3$(X;b;Y6$6(*W#ZPf5A3>FvT_X(@PRfGI@q= zGTv)hr#Z+C9{rBpaKV7b8%J9GDDfBVPw=kV+C2y2o|$p3+JJnky+}qQ-rTu`LwU)M z?;W|v2NiYy0Bw&4cxOSk@b|;{q_LLNNb$AGOL~`KzGaNfb0R*`(Lze(00%*e*AF`J z_)(otX5FNITYNeH0D@FMjPWAT3wY||a{SqsFH?*W$PNlJiS-*NDAQRP7rUB=Vq_p4L7JnRQU)d*5veoWn zQj^-f!ALNAuI;XwBZ9I@qnu>c96ueP_-zC05=(2~`zz#@CA7McLPk^X{gesjRFOv^ zkdQKX9V>d3tIJ0tIH)CfqlWS4?DOE5Py8Z2Cl3VPZ|*Lp`$eQ}BbC}=yBi@n`nKY5 zLGSV3#J_-^6Y!UbR?hRqT1*<9v1XDRi?FvKoRAOMqw^84e*381!0XL=uZ}!*;(rue z87%y3sq2>42wE$9xZ}EzF#yIEXo7&Fou9hhj!7r4hj?e;_r#xwUL_h|jWp}q!EhQ> zH&BV88%B36&{9x=&ekj308dMVd3BQ_ZQ97Yb*x{{V+7h;>bm9R!J)Z=$H`lW<3KqR zA1-FkVy;_`MR``C@Q+8=pgtlNw=afxQT4elC7v0$$R2vEgn|dkTON!>euC&<04CD} zTIYwnL8e>8=oK{SlgzW7!j4VMIF>ih8O)%14z=zd670SRY1%cumE(^GUihNG+kK($ zQr=t5KH6qqnOH^sl8(6l09&vqQTK&Itdkm`<)YmD7V$2V@lV6ItsFXmju6>xjWwz; z7{~gR8?%gUEwuLQRJk)7i?9Q1&Sny z0l7=t&%Lpq<5nd}83QLBmx>R9ykp`!k~~|j{{X^Oy<+n%=38jwDe?f8R}19Iorjj7 zF*)ZZysN?%`UG0^fAFt#`_nzEdDpf=@;GE@$sqZ*(?rEf9*?Z$8m8P4V8A1c?Zsb=PH93Bm-DZ7u@NK`_E$3^6s55xA=IJ(PFmYpH44$(~=Kd@H`U@aCs& zdo(^cy*T>>Z3wjw41fk%CMDt-Jm>hh80u@E@TbC02g~9+-wK&HyEUw0`iy_Sa0! zwKo!DOfV$wt}_1s#af?-{5g4T@fr#IS*gJOO5174Fk72^DxsDerwT~HB~DZTHmRnh z_b*E`$}T<+>(^JysA;|`u-7#Cqy@~;hzmyQ!}l`6%^P6jfHREooL7=~gW)%Yulz_Y z{w`TxS}n!Mf;~g*FL50FswQp~o~`beY_>)OK=9r^yPCaEl()SuyhC zCBex#;}xVUTSjxbZtwmW&+*^vbKp;hC3R%+K9OM}Mt0k4Bkc_+&O!4cA;(NBA;3T0 z>t8i|OaB0bKK}qfi%hZ9#;<>H?6;OSw@r5eF|l3d-uv@)*T05YSM7_WgL_*mC+KBuKp7q{T+A;GacI-l9u_`g%y^}!keu1i5w4M>QpHi?JRw&vVc;fR1Z!>V0VCZ)y zM_jix*HUZwNT(@vXItREioP%L?8)J&HJM)O)RQynf9U;k2FRi(VFy2bM;wFRzU1(4 z!!H{PNaeh;{@>H#j#hUOz5BuC_efDzP+P2R6T1PD!A=4E(1XYRFYs2S4yoeL39NA2 z&XUDNd&vHKU>p&${t3=sJyqbg#)h8~zHL z;Tz(&9vATS?zt37!H7R(U^v_jsc2eN2Y>^yJYaHb=|6@40JW~Y;%^N>rg$p-Gz%S0 zR-bO85}ATC?v^wYF}eq~!@RBr1~Fb_@U;_eS{<>PYR^;a{{RJiM6d_Mll(65%>E73 z<#=Mdw7cCDI(&tgB(U0D7J@y^lP8b@$VOZaGoWw&Dp_kVS^QD)zL%`%?-TF2w1Rtk zn2HV28DcFZyqlQqGb2V7dLhOSo9SLH_-xuHrFY|t$aVF%K#vqw-fP|APUQ)Ev3VF9 zWC3zIn&&(bulxX)##Sxx!^1Z^J@JX<78fmb1TwDD+m4+u`xCT42j2+e7$oG@5rc1C zKd&*#Elww0_@VnYcw<(I!$i<^?E_JJ`*um~%=UH{vMZ=XK2&dH$#TuG$MVf8o=7BP zxAE2A?M1I@7jk%WLeTB>i?t|pJNXTS?d6gW`b~(s3^$AqHUv2sCxf3#m&1P=biMbw z#-ZW+`wN*B-qPg0n(;#`Wn>8hPYk9$aEHobFf)@}f5e;r0NH=TelNJxyia$o+un#@ zZoQ?o?wMknFjO_nUUUqgH{G0V1K%~QB&}pAZh6J`fIby?3q-N;uf>f6SJL&Tia+1! zaa@RO!YC}0-`R;GTsCmbp&SgHjN{fk9cSXN5&Sw!zkwIF{xDc>r(f-R=7#1yvK_+Q z+*+(qtD!qXd0X}nWDZC**XduhH;i=82RGTGE9JYJ^0ADjt$6BtLV{ffD$H`koo9wZ*)aj{Om~Q0Kiik~a`3l^6Gix0s(6#cqA1);X+MyVDIkStWSB>ErCaYSJCXndEqJHL zFWL{lcamOd+DDDP*%Y>>IHSK2%Qdv102npAF^jJ?5;^JPNN)B90S%O*ypbm=N}S22L8vN4kMP&#J(oepuI!p%VL(I?%9!u zEw&%rs3ReVl0g~d5!dsdUGdk%&lg>tIq&As)^c(nxjP*(jD4BjRZ;3A2oG?2*DT%! z@xHi;W22j`Qc!mP0I62_yJ2N>oc{o_Mtv)~I4LFZ95Td3t629h+6MRbCD$~&JtyG? zi?3VV!86Hmdt%UAM>>QdBXztp#vx{H{{XmDkQD&FHv{oa=9A$Y`__}eI$hjcQLB1Z4O>dH&VH_ zSTEwVL?iuM2X?~p0aZR;JMiMMf8v_cZB5PW@+emf`?p84uwAg66+LiCt>?7T5XEV% zyv*KWhTEG!>KE>UTjnJ8;~*Y+s@Ara_g7myB+RkO#kQi5r7_;)Yp8BG$@Z~Md7aq>1w*QuA>b8c|KZ7H#?GkP_T`s zAAzWJY;?kRd!BiHaRc8x7q&A!#C;6grEDIAZwvzC(}V7Ku8J)e#y%sF4vlvd&cG;$ zAZJ!%g+)NilHC;K4)_&v>et315?jklM*yA7sJO_$3xenU-hUdcE}fxV+?gj5gjFI( zfC%F`>_d$6>@kkDi=ek=w@X%iFX1oQ>-IC!JS;We+Are9uNJ*;ArKDaZ={+l$lM>E8@naBa~m8FHaHnq{i1L_jlB5b`$qoDUk>z*FW_gx z{Q?U{i|qQP?xA$C8;L+wQyGFcj#iiN5iP=z*-GZTQ^h~GAMIc92f;TU9M?1_wzagJ zNgbVbZH=UQ2Fk8$qRx>*W(?n?~E-i zWWRk1^2wpSx3-G`=9f;?e4Dg}Ex}!+z*aaU8tlJkEd%!d0QlQsDe%vWbj?!U)QfFbeaW{b}Kk*`xjnRdn&*{Cn_=OQt~XuwLkz<+Q>8 zS8EcQtQI#xHtq#rCm?;&a4R_0s|NX+Diosbw)Fo1gZ^jGKMAzW7vTPv4xRf)d_>TO zhk2E{n&#>?KtaOh&S>+swtzgD8wYDQ86)gJ2maYVu^)%8VzuzE#9Q}BrDj_zi{T!f zGKIq;Ez-!_Sd5{HL}S?FgZTyVNB#-H@RLFD4dt(ZyiKnsie`XZ%Jx_4_R{1HjsE}> z&$Tn59ZXQ5Ameau<6rAH3^B@Ga;HaiGySN(SeYl z;<=qD$zPb&$tIEgM(E$PH--KvEEZZsI)1CALo&QEGAe}$hlrUSCy#QK46<&EBxDeJ z?D($#0O5VJ#pC@?RKL=7CZEfiEk1d*2S!nnwM z7@i{WKCiBPF#gJ!R!i&a(Q7TVjuE`1ie{R~+n2yighE&q1pL5z$HV^siXJ}juZ6F5 zzYgE{cgGvij*}j}s6nA$#?d<|dx`Y;bsITE>H{H0MnR3p2_bkVO}%3O08@7;^#1@O z=-nH{N#cJ79W&tet!LtGUTJdAsx(4YRT)&pjDAu&jDp4%3^?1sstu!jXJ#!Y!@Dxa zleS$uZhPZ)o*N#S0Q&P@ZlAR@k2IIwH1H0w;fQqRO4ezx(G zxulU-%O5e=MO+>m_!}@2Cy%cfKg=0CN2-+|_fwOR=yGeCv@Y-Z{{X->g}<$TQ~%fe za)uue*jZX9hpt-o#Yg;m+uI~Ef!C8fz59$0>rc43(e;Fy@5M1LmE{ep(q0X~fz_ji zTw#CS=hCrnG+j>DQI6w6ou+uk^0#+VfH}r?Muk{*=~Y+7Hb@!#LmYD6J8f68v}bkX z05a*5^*G|Y>{u6Hu~@vDf#;plz6m7q1!iixhMnRmZl=_h z?#1I$43BbPQgO-I6o7y`=Nab|%iS|pmi8O%7HRKgWj`d7h(hFl6j3VfARdjMrg^Mc zd`aP3%|v)xP&Sh5va#~5tS!9lLD*#CD^Rk}GjMPb32wr;r1iTxAs4zu3q;m5%R5&jp<0aQ-iw=}voXH^hl~HS8CvuiLSxm&K;GQ_bq@uc=E)rI{wfpFPNZojM zRMD(0zR2*)W}_rQ43|vtk&}W>-gxBh`c-%>Us4wmFuS&e2M@Jvi|4P(!^{{1o)r35 z>?V!>00cw-0D_BY-XyWJ@jr*gixH8Q0Y0UqMHH@}L%?XTiL_#tQf z6b2jJLtOD6g*-2BV*n*Cul4&#CAbHWRtchX9+-JR?ntSOV=kmQCD-vk5H&3h+Dj{I zX(U5v4i-@~@?5YTc1svnR`g)!=~q&J6WvJfqUuv5F$B-sZ4BXuZHyWh2eD_}^fmjH z;Qs*lBlnDaEvDFbxAwaDyWziv8YT3`+RVWWTAi$~xa~yNVY^Etk=1|W+{eFmLYD(I z@6-PP!8U*3k$(rR;@3ZD{{Y%{;_@ru`DUkF{VM!J_`ClA1QYmSqe-E7L*X6Y#hdGCj8FZU zr=2tRU~h@#)HREU1!S9(~~LQ%n587{#L#m{{X>Mbr0F*$oOma4ERyu z-vVhxR5VuLS%IWkm_&`RW-*x7hcoaQKpyOa_eVvoi!QFzn%X85BMkb-Qy{| zV|{lYiTpirr)k%A@koN+Sthg>C!<9j%rh9!4!D*V3USEJMsJC{U2UUknq`-ZucflR zyJ*rhCFQrcb{m36bvN!N&hNUXo=E37{w#05;HW>hCyabrJVl`VP}MXWMr=c*SzB3u zWWt!omV#&|orZpP^JjKqbGVKM*1i7#1xfz^f~4v;QE7Td#f$9%$Z}xQ>~ux8k%7v{ z$u*71^Ed}^Va^Bu*K9H>t7{%@Ok{0z?tf&OSHt}}&hGD8(AxUQNUo`Es$4X>T1xoL zo1B@VBq;z!?58;^fDLL`c&|~=d^xFjqs5P7;oTQWaSL48-@&O^seCw*rie45fx|TE z4%XwRC;a<jmDEB^q&Nk8D6pAV;tU%&X@Yc{c`23OO(J1gsq+aCE3ozf*;UroKRGCbJWk*&qNVrDtop*J1hFHk_> z{8^w}{5-u~C&yYIqv4C2wRtU9SJLkEx14R%53*~#$(OVVzwyg*~S)1wNj zJ)~>_+79C_xH39prZU;|BC>6?uMK#b?R-THamefmTgzmdZ1N`rG-38gi_w9@4#0Nv z-`Vol_K?(m8~9Eii`r(NFNtondwDPJ^!dfRE!sw+NaeMXR;2p7K?ST%4HPc|fN84nS1{zf4wiw;C3W;u{TH z#Sf;~YEl6WwZ+U!1dJU4u`ddrQ@PdlkG;X%@E{tXrKf zW(GTb+0NGcJnFGbm_IP{PESN5r_NgM?5Xf7O@CSO#qFKvf_2?<3q9_KeLXCuMlEkqo z4QluI-vnT>_?M)4EB0#*PhD*@QH(LbKVxJt%BTm*7V04Q3aGWO@UQl*@ZX5F3;zHB z{6Lpa@IQ()%VRbElX(>Sb^cwNJ+atbL@wu#EAmSSk|*Se4lC>rF7oE^Txr+w*uflS zqk`TTrxQMRg=q#bpbQ+i2aIuCZ^S#_19-o|ZQvh{-Wam+XNUDT%3eWvrbVdfI4pLF zlzEcI=K+EFagaguB;BkiD7&V<@9NL+I8A=v!d^YRPY~!@)#Zk?lcr{}Ef!2ca+fxD zCRE@KF%q0{oE|!F6UpO!2KBXFBf@cN_NeM2mFE!Nf3kL5tWX>i)+Xu685O|zQ}$8Q zd=PHDSMU$w7O&xrH&V7plFsTSxYHxHT(YIT*x9L|Qcsr6b3Vbsrq7nRuM2p);_t)z zEqhY<$)M}LC9}{Z0u4geIIb)&m#aCHMsAY}joa;1@G?yilkR5Jc255Qf$ulEhli3| zuL5|hSr(VqY?oJdnoNe;_2%1#ZSaVZvCwYZtT-aDw0#zR8&00zPD#8=c5eRw(T1e) z+R82Q%x`xByiVc4+8fHq;fNfZZBOGz!Ot3LM*GHEWtNwvnIL#B?;UTfXC<4<`z6A* z#(b*-9APAkg3gQb=d{v(Drr|fA$?=v{{V_ENG_@^F6|+n!<-h{=VZ}0pvw-KBQ>K< ztEKl7DJ7@oK7SP1-FZGK_=}*Xg?FUfc@Ryj3v$aTQ?fW+BrJCWfKo;Tj+n11*020Q zrFe$lSke4Pa?s8j?KZMPFAO=)HuK?G2W+}YfyYYqxqdPHMe*N>?tBg7FNfDA#^LS- zo~>@ynx3A}s^t028mE%-CO0F-aB?z0&3G=Q;ctiDJJaT|_+jHKjRVP;$Pb5Zt>-~6 z3PkWe_~ZbMqA<#<$pbvr^r`hM#I2n*Ay9K8xbb zE5q7MdR2`6I=zAi`*qc?k*8dV6~0!2+GS|njx)4M7bBsrpTfQj_+tg7kB&86B0U1* zT}Or^tU|V0sSTa+#t2ef3hrEP;|-3L*$T0>qq&@+WpAF&DP%`qC zw;^r`URVg^0SX)d26LaSVC&xk7f#hC@K2AtOCFo5U)!wG>F;osk}|mkKh@y3`=DjI z;Eunfx<7|}GvMDA2sKY1-)kBoLnAeYt*4JZIl#E4+ePK!SX7ci*>n)F70JZbBnmQgMdEqWk18|Tf-AJ>~qTx9*4nN&+NzGF9syS zJzgD3?qHH6cRBI+AQyJ)<@6FD9&-`~GOd*Z8%KYad|RUGnl+r0 z>F~a%smNVsM2y-Mexym~uIzFN0GuAF4Cg|73sGB0I{!*d`+lY+sS!-XK5>8 zp}A)AmyyB^s2iT4n0;5RbV({0-{RTTz;eQMGZeI+BxM`N+#t5UF zu0Ce_J<7_&XZS*!xAQtN0m9;5r*!- z=M~_-8u&l1_)El6d@b>K7P{;b`EgomrGsP**|xbyaDHEzS~*VxAdz1}-QW0|!=fEG z!`2bR*2p6!_7=W3E&%zXnm1-coS&K_o}`Z46{4Q8I_bG5X&3wrFNJ(huAA*QO107T zT{dvnAVD-YGlxm$kcpwYZrwQKZ%AdKF3p5iby_dq#zuOvc6jE~7aYI>R7| z(T+kHW66z(!4I+e432o?d^zC{jCz#+0Pu@y)&|lUqGwt3+3mG0K2sP6W5aMaiAei^ z#!GfJ%j^FD4(|Rg-^f?R_p7Vh?TFjQDYu>>ouEsl$^cxE_m(vE=NYXcr1ZF)_P_iC z)pZ{N$Dnu??(TGjSw6=wOPg!(;zb`cGT4J6z;p7mg(?R-d-zv{bgzM)8rCD$ zJQ=HMw^nVHYj~a;t9Np^5-YG14WqX506#E13bW%6gC81v32ASwczViBVi==Tmoxp2 z`bjbaJTW;Dy0QKp%EW&UL0A^|8cl|a3HXC}o)5Y45E~4lEjv?ph!Q|x6ts%GHxguH zkyXnTbj)O)=;$?n9sEkT(NBi8FAHhcIxeb86`aL)Zv365Xm0$uRP+i~6OMr66~Jo$ z02Tfx{7Ufzw)$s?e$REKqT0;{k98S_5Tk{b$;Xmde-S%d801%=_$yV={0*yWqs9Is z)O=NBmPmAqY4oPI)I9y{Ssv~eNaYeKRY!3euF_bX@$(nNe~sD&{{W5bH2(k?{3u-y zU59PmmEFRhv?z^)gEw%P)PuK>?!beCoZ!_uS9Z3gG}i3&KNEP%;@+2{iFN+~4_x@B z&@N~4ZQ_+MxCPoQJGssqY8-+5T`FAz#r`eVygO&8*xK6J=#v7r@X9VNql=PuM+BRk z6(N9A&~(b+*Ta9bAMHcq)t>50PZ|v^+z3{|R8G+xa(v|rIB}A96VDg}7(bs{$Hl*i z`o4#0Z>#ICujz0|(|w)fd+T^zi4MUhW%5e_hVqqL-n%K_9OC5kIp>Jw(Efm|_5EQM zJIzB+w6nN|M-OA z`BfT8ri_deg=O4E1}pO`;U9s1GW=BV?E2TkSYWk+`7vuf<@A?!(U`$k*&7YO0thG$ z0pW?oXjuNvAG9Ubxp6nfo2$EdG`I|r>biC9yIQ;|a>%H{JAhSwZH6`-kY|HS2TeW9 zrHGr-$J5>={j>Zf7o5Hq@U5Pq49%U%Z0PvsZ<`*XBL&YNbh33huN3j`?WN(pUj0{J zw!hG4R9KQNI^`XhDE>QtDXnr(`==ZRAXkv+-?O*EpNN;n9S_E`T;0GGZ*4B65^8aY zIYBM7(A!)<>z)H69x8%c5x$YParq7tWTWZx{tFLn&qS%7S_NdS8J&ZTn07MzPdx z?lj#i!B!Im6W=wwcG7Tz%V9g0jmZSrJ{%e(h`K zWeeWS^qmj>3F+};#oB`Sr{X`0lf=4Z#3kVSSD75_0UHc534E*$M+&TS)DA1qJU{y# z{3bfhzK`)U!TQdTb1mA#4A*v(8%d;Q9$%NJTS;*b&T)m3ME6l#U&6omDL=x$hu$F; zejV^-)|af^F2f97V(u>adxnkBTO_J^4zUtJE0Dc@*RlP%uNkfNeRIXyJXX4hQE_YH zNh7zri61{YC8>C3Dsl3W4_uR6wVokcOC7LK*_~uRvUi63J*-LM9VYumwuK+&v(+t= zP>v?!1BixNWul&uUXPDF5kt)7ae1qB>Pb^=H`L z68J0NPab$yX+9Zh?QF5k=Hcabx`xtd!jo>$T{2A(3UeeS)TucD6I%9`@9_Tn?EW+H zMx0~Q*Uq(`#_C|rDv%6D&R@6~0#u~(A^W6c=D#L9Y5QJ$b@7n6(aqh4ryRpFE!5)l zNHOw}ZZ^xgPds6UYvo-_#Iov>jW@+t*Y>Ib-2+VV#zubZQC<0x3}YZ}+5q;g>EI^y zZpIVDtLT2r`0w_`_(9+p?k{aL{YJx3xo40y)^x^yFy|!}Hi`H-&fYsVdi>k2f5Akx z{eN1PPlpz;X>6(Fmw0BlG6orK_wcEO1~Ii{jDyJCo-5)%40wL`P>x+Y!NNP{XuQD# zy9pU`2|irkxNP;zdwzzx)8Rc@t>{)uRX%xjU4vpxw=Q<*$T1 zS@AmW?OI3n-l9MX!U>G=D;}i3%F!-IAf7laiu(uQ#D5F4d#4FwcdlNkD{d~V{JVac z%QQ(g+~D$1a0gDc^dkPyTJMA)y3;Lw9mfh>qO_w++p(pL9kThYv>^bIxq{(OOxCa6 z_#^rp&S}c>XXiqE4EV1bZr1dRyXNyZtf;?iBaV_hhs{uX5XDDRT%Gs9{{R(7BfYhy z&8*!DBmJU4T%Eh3B&b}2{9_-jewbQ*)zDnpBkJ-?XB?6#dv$p4CfkhT&9@4R$;jTt zz0X?mkBL9DM4lwGlStCMJesY%hzn1BAf58Ojkd}}ih3R-U!gU`F`MmfmdYron>q~wfFU&MbGemr=$!dHF^@h-JJ&XIi>H*ifZ^FGorlSV|G zFI}q2yx{s*h^y%~@hmgTEtI(|-cdHO$hgK2nYjlY)Pw1Y_0O{SNp_=M+}p~E#$<*; zY6v8bW0NkUwpElL%CU573#GT7*F&~9_Q=9PD#&DZ$m}j%m4L@PWaHDC(lJSBijFz7pUZl!0bO?tN;A(QN`GU~4);{;&g5E$}7 z$Ojp(KzZ&Z)eWrZx_#c@Pn&R_ZV1oI3apV5c*ribYVyxlNaFi^^In#9M7Btl0Q=Y~ zsN!d5Vn7NVcN2vi;+@2iu2#mENunavN_ToOk^P{hYsJuZSKy_=_&7t@t)g2{hY? z zev@D_Aa;$dq>POI@nMe~VNQ7R@p+UW`42*l7IxJBTYe9C1HpGUk$98E7MiZFr{2P1 zdp$nRTay*eL5%&X&vb1T11kdPa@j3`gZ5{`pMYNnJaMQ7__^@!RMGVNRU}Dcr^|U7 zD>gSw)-biwDBu>8g$Fx`$sf|+gnwpF**C-9G_}@8gKxYOs99VKm6ql1tu1age69AA z13sHFGL!O5GtWRrA6odM;2(`r)%8ydU0U7gEW%i{yOBPgfE*AbxVLFkj^MSoJ%|W=L4ETeocnU32-dl@Z9`jCJMT33K*A{ot+g`>u zDF!&M1id~`PAli1i+}J^8=1T$zYV?>_;w$O-Y9|{z61WtH#XKjEcm_f)(uW!9Pam4_AjX$ZLX{H-Py$R#z|f2Ce_+U z@U?s$r})$0?}Bu#Lr49cFYTp(X5td^Cb@HL@u?;$CA()3+mMHHN&_g}v}Bs+g=^~d z9ZGAWKNuqL=f=N?S_Sr*;)vn+eQ9Ag+Z#@>(rz@EqkEYf1~BYkW1qfLYh>;!p!4-V z?ECu${?hH@D11?Os$E{_tr3-;7B$M5E&kbg4ba+Sid#Eolrp{_E}8Pk`I9G)7V!uC6_4PIelL#u;a|fW ze+%l1bgsINuOzxdPEceC;?wf7p$G0=N86E*E3INGOWlGdG&Q?N+W!CvJ`n4A4epiV z&xt<{?#=%Ihop{s9b!8Ob?G90KzOw)Mw0FfcV>l73GKSR;BSUnw}dq_1N=dT()0$B zD;<6~b*GJFC5F(g1==A$bgb?G;0`fgkA5iltN#E6L-^NmYpr+>#THs6rlO*IdqHh1 z(KHgS_}&?9LfmX+AgjovFg$Q;xbTPU{{Z_}>l#(IweiwBuM`QGrNyt272H~@`(kUKrOY=GA6$e?gQ zz^~p2u)pwc!kt{(YTAn%mxEIKRn^s)j4t(CO!JMm_o*_T*V^0P>8`~^@zf``cWiD|QqxQ3hUD^*sLx9EPYQm+`sR_T4-x!o*1Suj z&1~`gqHQlqH;_QXBt6aUme*>KjC{uC6bsj;O=bSjzBtri@r}-v;dw4RB#^j~b%l=i zPL1}CH(c2ydT~Z1@o*bGdW_fFpRK$9*@5X8Gbo;^PSz1h1<;Aqf zB#ol}$`-`NMei(VqWitXA#r`zfm78-sS z@d{d4HN0L*jsi-u4WMo$@_u8t@c#hW??1HeU}-KJ1*Isho-u`TG8BLZ3_E%X+v>Q9 z+VV^K^FEzpsl{DCxqqGi03-TY@J5mU00!ay%eR8V#M;h-@!Z@^wk=yz(e=*}%8FG% z^4iiFbjacaHd+*nAL1pHll1q9{{Y~wKk!Iz+0#gW0shS23}<~l%1G{^)czbTzLSX; zBq5qfEN2Y4K5S=neB&+7Kaq&^TTMh8tzr}Bxdew|>T&EzD~xlBvf6g1W<=~_w6~P; zAXYJi^fHhK(;~B3JX^fv%;lIlEt~oO0F{ybPk+Qu8rGFPy{Cab5o>o^Oq+$vYg+Vc zsYD1N8WxTV+h%@;5>2;C}3OMFwUO*ezQL!fwTOOcslL7_o4_m{Zs zGu*t*6tfjLW=nXGk&u|;zb5#E7ut+x3{Zxi&M9IT zVYHTD6!3Ykv2-8!CHL*+t=%C%3$A=Wrtetf*F0Bo7MG;J?7?=;J2dma+~l#g+Hr!w z)k`nK(|4ERXYD4|kw!kh{{RIy{{VuC>-Trse~f-M>o?X{`-IwjmReDYByikY2}-Mac&TcA+i9E&=MjGJQ>c^n5@6 z0E2!%Xg?IgZ>f0e;}?VU!!%ir9}nrb{vwRXz?BPjC)1;J=yDV|BIVx$w5BE`t@Mn}W%^Pg&aXXpVRMw@>6jsf)tW_&)%I zDmA(MscH5AN

0R(=}OKeO~WZ8U2!DU)TKeq2smV{wf!4&Vki?A*fuoB>#RPlq| zK4eRa+4W19^;I#pe3@?z;l2tPerU^K@yM^fejEPE-Vyi)bSv3y zV@S5-FpB)f_c9?Ez?hS`f2_w&d1ka%y%cNgowbwnKhD|@hyMT)E%kdnF4w}^<)zlG zv2CH$G~3GyiP?@oiYvHbc@zRZWjGv?KpC&mABKPMNI&=}=f&+N+fmT`AEfE_*CgcY z`qiuq5S|?t#dQ;~4tXle{oLda?bAo`oELgB-gw5!Zw+dZ&TQTtT!eOFyS7`qw2N^Y zDaw>)Q_e=BL--v!1zFj|C z)GW@QF0Y{}mgt~vh2mFk!?Pu@~RedXW}4|w~-jd`GaT-L7qSh7hKuAe=% znp)jTqvn!nA$cWo6|t6G&(wei73hEPi)#;Lbp0nmgHY1sEGK)3$z7!3$!8&BY;ll- zZU!^RtR;=Dcg-T5B;PYWH~1s}00f`?gT4jb+g^M#@QsDU(YgCm_=Y9%BM`*pW4Y7} zmv+Qpl~i_&bupaR($C;eg_hEt3&)=fbj>c(;u!#zJu=4QOqvMtNPCqZKCjc zmPw1~FhK-IXxo!$qsu73jFbw^*Ky{&KgT}-bgzh>AAKj{j5l8r?VfXUdM)RY7^8_C zI>7ceq8FCfa3qZRo?s-;cKhNw+pEwCKbkfwvGlqQn5VA zwR!2DhO(}8pC0&gO1sxSDzp~D#e`z#O`1a;)FD)xqNt^z>Ng{0h6hnmPI8NFI|=Er zouT|8_+g>JZQ>sa_(#LK72JC+XP(~0bla6#IgAlxe$t_oWmZtRJy4#V9ksO9n(fY{ z{wvk>OBi8c_6yvp^MU^WmRIB^R|g-yP)0BtlZ?L;>Hh!;{wdm7c$-$YZCdffO>L@a z^BAv!BZg43+YQc}!6h7#$pm22HT$25_BWR8;hQKfwArT$@!QTI)NLZ%0GOn#zHCd9 zUNA#+U>F=xI@0b>BxHwaJ}A1sS+r01N3XQ&y`E*%9wLHkxC}vxag2wR!XBM5O7G$;OKN=G@ps@4?1kf$OV5q}019*;7Fylfq`Ho| zYopp;S$T@2NM?~s-YTMu0?~jJaC2W1_~-rz&HE1c-tOmJ(rq>057}HymrzHi>LDx^ zCso7{MRbnI8xNT6a7IoD9Y0od?R#AClrO6MLVH_}Ft%O;6vntjY|jmo9qPc}6niQ4CBJoC+Bc*Ej1 zj5G_Gb!`VvgHy33B97n|8jNK~8=eqJj(&BK9k+6M z1~kd!9@TePe-3E+co%3bmRO3L$>J_wD!F~x+NHt5<$+#@k?6hjqfI!ANah%s>qJGxj5qvRlG+SF6Mz+)!YeVNW{&+unHiB0y%op$c zfB<GCr;hj})5(vXbX*`9M+#rpAdWkEb82OJm&R7g) zxT}v5+Uqf2_-ErUg(tJM5H!;*^ftG;TShvwPNc`Tq08)H-D~4-5B||UE!Xtrj^^rp z0V9Zh<4uhxx56k~JWm{Q07ewD&yQhU>-=ldv`rwv;yY`vvs^|Z7ZZgRGBI4?8Xqta zMpw#_gMD#Uu=dkqIo5YQV*6gwZ7eO@;cm4Si*F19D|>SaU0BJ_%b442#{-2i%y0+Y z=~geia3j;2U0yjgI5tN2O{c(SwpVgN4+h& za^U^0&?$sCAM(;EcdMV_EQ%NoGn)DO^7F?w>2W`bHCd;QA{CoXmF{8~rE5pCAuVH|{vleve`*X5R>2gA=8yX%^D%3R8ac)N~A zOR=xC~-^nQ~NQhTLOjyL_Gwx`4St{NRD!2j0}U?yvFPH&HbRf zc|4ZC4CmCe1!UPHM6je&0r%VO5Te~Fa(6JvUQg1#Ub65WwW(^$;cZP~dE_WQ%XKJ? zh;fh=0X%=Ycs2G1!~X!;_u_|ytVQmhJ4F~x$!_xg;5S^SJFvuc$sU5X`n@=OD8||$bpa(-=Hc{{RaHZ65wyB3#33zSw!-P9hEuH_^i4h?Ky{BoZ|x3#d- z6Z>W>R33H2GOOM;0418=O>JX!Dq#@;7yg`Pgsp`TKT-WwZDY8fq-X;}02IVamJ zo4Tt;eqsRt*O6#H8T4-nXqtpx7uW85Wq8wuvxicWD{`@I3h|qbl5ZoDH>$HA;{v@i z!auZ)?y80#5omGg@M+GIn6(HJXkm~XtEAI@?;H+OZzn%dE}xi6YS;XYo3DhPAkw@S zVc{=^DQ%-Yl=8DSw=9~XNf=_w8$Rgfi4>eE8}J4O0Oq6`LU@0{Q+U7PR+(?12=5xw zYWrK4TyYj58clF8(WgHq(4jCygvVAl+zpw^s&Rn+4mg=~IKj zU$fjd?gFnxf?%P!3E1>LG5xyy58;m!O{jQ3NsnC9?ZOZ38#$tqc#z{OC5&KC5YI#N z1JvWT(2Ym%kmnoQ{v74}75gy!9r5;~Y2%ND@JBtQ6K=V?(eI&(34;ukngwGU9D4>m zG1jHJ{g(V~qxha3M&nG;wapsh;UZP@e!-|(EQ|s>Lp-KA)dBgLC01Y#*vS|_HF($l z3Pa(xgICe~KcbB;4PDl2tKB`K(JzdMKvxql&QBYDbZyJQC#86{o%>b8ad&$(9wpQ@ zTXP6ywY`Ek?;ZM>{P%dExIAH9V;;D!rB1H){Y;yK>VA}KUkX2EEmy>?@TbOikT$(! zT&0ppHH^ySf2_B5^2~%BW#eWYiRQZ)JPCW@Ux!k7bKy+_mYUpO+CEz@Hnd8p79MauRvkTg ztDh0SX@82|Exc`O!X7Py!j?%KP=96W2_3zRk|@C+X^joYF5P#G4g!&bn2j2Pa*g%; zPth6TyzsAuC-HB_yC^&#Yoj_DrPMAhzQTe30M(mm)q)jfIL6g-#~X8BC~7)a!ux*^ zMdD9~+V-Vq6~w~k;wUE<_BVNK%l?mcM1vsbbTSS|;=WArhsTc({5HCV_r+SR*tfii zB)q=UV3$zS(T^K0CC`xAIUhPO0~x_J#_1ol&%qr(`$@hLX))Ve>rkpK-ke$dsTV)L zODS9IyRyy3)gGXbD@bAF^)hlvT>S*J*Su3}aXZD~i@j?|X&1`7(V1FE;%2}%HV&5a zsOO(A&Owf~r5}Sn9LwV?pABffB8S5I$Gr3C?q0 zn+M}J?IC_v9ZOKM)ih*DWLu|(TN~vmj4Rlok|_Zh<|FeSi(fVPxAwO1cZ{^@bYF=H zW#S9lUCRZwxob6z{i);kkX*Lg*c+dfmB&$q0=pxGn~0TY-2R3%o2%~#X}4B)kXZP~ zQF)ZuwbkF)VKEK>kO-uIHE>TkQ_+a(dT$2!`&Iaxb9g6IB~^~1 z;z=VrKvpdgCpj;kYwan2XixY{?>P^Wu|K8Z8q{7nVVaIi3ck`+124&caJAB zh}R4ixvqK8o$UVr%xa?9b^f+LLwr%;pB-6i@_6Ur&&7)?v3YbJ=S$VBKeX(`Wq@U~ zxwa6?3+>2I!O#(sdXwURwIA&h@HgS{x$y7AKLct$B=EGB%fDN{)6M3Kxw@;gA)Xz8 zXUvr&0R7fDCcaDXC&XXai{PJ(?7lmEGoIqg!f!F>)3n`hIwjO{#8gPSV>H&#uyedL zYJ+*{$*-S0asL1Xbolu`tEk8D>gPe$bOAF9cr>UkEbb$Kk`_4<^hhSaz-NhqvB?2? zRZ_xKZ6c?Lh4t6)XV`iNhCE3B00*B+{heUb{4=F%5+b| z0-;!dr{z!$e60_|{{Y%2&(|+LHeOpnW2D<}x4Y1!l4rY7yUY>ToVv-hDOHXZcP)-_ zlkmsH&)aLqKNz)*GS}jgPpAI?WTx9#tTkJUJF9%)<~M+>1=k!0jm|z&+iPb{{kU%a z6?k~Ozi;tor5xbdH;Cr9w}~J)7-x&jX&EqZ5kYS3GhU4>B~`LHCt4CZuY|v`$H4yp z+P}rp*?cwet%a_wY(&?ZLM*ZC>aB+@b!T+3OCABs0}$(+^P2kB^Zp6@`$zl*n)Af} z01rG1Y2*DaILZ4xmaev%$pCd&F2r`Sk_QXtARP$rUld#a0Kr)PHR^sa(R_5eH-|>K zrL30L7gA^`a>5D5*NHA(U$h9_Ko6Nh@IG4WuRm-}e(zD2##f&i^$0KSuHu^aN4v70 z#8xhbMKeclC88wOOh)BGa}<1hhZs7}jdrn13b7%))wW`|z+Uz8SHy4lDF?%i1#Y|}qxdoYAZV9X z%CpO>M!#d5dgYPqZdImrTm9hFVwJq`8hy z3`g@>G_B@I3^8A=&7z~R`GRa^NW5>iTwJj4i~LI-QT2U?fV&BbdP}rtC&QY<#QxL}_{t zhV`rcU%;OqB93o0{{Z4&fA)P^#tn^!jbpWFW{iv!`^Ra)Dngw8L-@zzpY0Fv4^_6* z^G^Q6DLw3$lvR*u<~GTp#nq;S~>6~j+!uIiUl&*MZojm5?>aPdcK=Qtp= zuqx-C2JD_WHRw{yB^_HiYU1Ml6n*F7Z;O5}v%N8VIq^O9`mAWqY-NuUXCSB9*omaa zJo#kw+H2;o5Kc9Fi1p7H$qe5+?GCcW{{VZSie@A*Am`=C^sHSE;clmVt1Ye2L|0KB z_E#J8#$+lsfP3x$cRlOquL^t(&@5q##Xl0RV}>h&7wrjg46+GOa&rgTWGFfr7^vq2 z*Hj!Kc5&3Sta(+hi|n4wq0ww4mK6mGzF~+1kU;qXZg@Bt$6AgpH^CRylUv2%&$G^q z4lLA&e6ztNB@NDT%8`@L74}z$yfOPYcz@3?;>MEt<}#@hXJ+!*O|vvi4*YFm6yO|y zafXk$fKYV%z%_7G#?M7s~7vxX? z{yl_|fVse~*FyOH@v2DfWB7mYwRG!ktU^aH7Z&WNnAJ{M9e}uHdcRsD7 z**d9jvEBoB9!7ZGAC)TSxFNvw;=JeL7l;1UAU79PIZV*OpZ0z36*b-08ex@a2G0ya3>TAJU$JV;W zvki`l&u6NRv)==#lGzv_ugs+OA~HWJ(ll!&cJwi+S`zqHr_&nm$Ik`cUeBTUbHi<9 zLY78^%#(si1w5HzYz{M*+P}EB?Mr^&w0*72(F?fm?e>#9T+0-;gnx9TaT$Qe1w>$uIqzIO(|j_y3#HE2Hdl7V zR{69@!`l#kY%1O5lmLI_nw`9X&raGl1O7w7JcP@OE5sCxZ88JLTJ~LM!Dmmo@@;j?iC{nkWl?RI0-=F7 z3+66Ha!RP@j006Q3vEkMwT9os+DuxSLAhYQn%G^tF~grQ@^WwSb% zjXXOYrm1fXzaWX0@zO^mDuXE}C!r`v?~pT&cTn)=v#IJbjZoZa(<>0`aVlIXJ4gUq z?tP$tvdQ?@v};!SmEDZ3q!R_dm-8g*llg`=;FTFMDCeB7(AU;~v+wLh;-86{N_fBH zHmi5w&jrK-L#d{kCbhUs5(-=fm_Uk73reaRw@g$^5gKajaG@zXBlBa!{{RcTckv^{ zw%!TQpr2Bj?G!W`ZK5y=hlh|%(#f@(8;bBfMSnqmu=ng6`zZV^HXc2{zPQ#rd0`|z zWyYgt7_!o*!DosK9ZElw0OQSKBmn%5fY;EU2Y=w1nof`Kz6kytc>DW*N!3H%CaHTR zsC%S#!i18^7AKA{Jh=Y=a9F6rF>JBqbwBteO^1l|>%aI&wf%h|hG)y(+#7|$ld?-` znpS0qlk%eZdq*BvuQL;k=ekbF^r+B`cRyRa8?1O+R@831H{%iEt!BxsUA+A%-_MnS z0!Zetm`NBoJb8B#?mPq;&0qL+C+yLpc#(9!h7J9Le6ojqC|u8_iC2-ew9BQ92q&({ zf$zG%7(6-u00g-B)1~TGx4r=QlXVV|LT!TbSgz3a-5a1s10CI;1Fe0H;Lm~o02}-( z;Zvi0TGO>35!>mm3^VBs2CWn=3vQ0!K`c>9vJIev63D>vtU8|-aCgz%?wViBevEjV z#(x%kD${BjWuBXTXCZx%L3RC|dJ(x)3v};qGoSByu>S-{4J?YLcJkyrpi#h=O>J{NWfgO{*Z%-(Pl6g}ha}Mc8EUWMeP&iIYk8`O?wO-;kfH2j zbzq$BWruO@#EjR(e-WYhKk<*m$4dR6JPsz8R*miMHK?zJ-LwrF?m=>}Y`=@A?$Pb= zSbz$#$^3r!^W)!+XT^HHjo{A$>AnR`b|5at*}l#viV)6>v)ep&KPN5$LxM@##DIS| zf8i?ewVtT!b*sJ1ai|g-hWkVTOMe7W>~KBGei^S?hFs;X{{T~(ScS4j?dOYrE&jrP z19eHfDe!~h4yaz7*ANukJKo zh2;2h+i<&0K~@JF*bTBZ!pj~*C|%!a49)y2(!NO4e`~+l#(fSi4(tB_53D>|01oF; zvP-u@pkR-*Ixryo%F)Cz>&0^#&+SX_Csv0|(=@#*T`u27REi5W3wS~XEU~4X+%qYR z5^=Pl{{U}*2I~hqYMCyqU60g%g1;ZXWbcT2Wv`31S@k~#TWMurj_oh*Y*q!30(^;J zm5~uoDpFKj_C3E{J`;Y>dXIqgyX{xN`s@5bw0O+geEPJLBf$A8hFheLSf?dV1?9*c z8z#RZe`O!rN8qQ0?QQ1xmoAk)i*;+fM{T0dVR(~ae7IzV*6P*b!EMhO4Z8p?2_vo5 z{{Y~qKd?`Lyk%{9;ZKBDz7o`I?iPRTDD5VeMBB zB+jZ_ty%v7nfk@A{@H&R;nD1Mzly#By=xd`SdO8j&ODfeM7|#cvjJyr|Uq$;uYfa*B zhW;jk+TKDGuB~J*tzc8hM3T_0+-lqn%w+4FV!ofWPXTzpz?-~(;QL)uRJf7SShQ!B z?6U8~G5}+OLmu2SMnW(rZsr3u&Z8*#Sn5e>Z4apQ9U9u!%TBiVOK+!L-)z}5x`F|7 zb07c-4A${W99bNlv1|tbU{-W4U3`F=^aMjLQJ*l@$9+>x7`y_n>Z(&g_9 zT>L!MCbzQihmWtkRTZ25{{Tuimd+TR%DleY=#G6^w{7X%TDb9rloPxW*=VY$#X_;T zb^sh;w>yui@79yzFT_uTo*&dC@V~{chxh4aBdXf#THc=}#gn!-f2>7m;*m3*z$oCA zd;)9bxxe7Azpy`tC2tQ8+7IFVxUeSQ$`izPd0!^qnnHeDW0Fl{8@*yegL@G5AUU z00!ylI;V#rzxa>w!shPcakivE>68RJ`P$SpMw z5-i$gx2Rfq?Wf#Ku-%46!h0yyw~U?H@+HE*BTq zP|vE|w2^dMi-&h-Wh3P+CeT1VMSahAuXvi@#%SF# zz~=<*lPY-m^Tm3WkE81s3*^rho8;^tq)G-?)19!NPE=b^4DnWcHFa~g7T zyWY>p^Sv8LwD6Uy=w1QTFI^*!G_=s~B}+&^URpS=<3e`r`?1J(ard!W7Fy-7)ZbZ` zO`l5CuAS|kE7XoRV1ON}42-jkHw_te+s-q^Wq8NpZinz+P13CXH@(DS;e4jnEiCRL z)80i23r@}+3qs`KQb`K+8~SIf)OD>F#2Pu%-qO=rv2{rxhH|qM;1u$m1|wnA6yw{y za>ci*cP3G`==1%1!`E6py^fvX&2ldbTFdNqBBp22xIP zL00WfoG+vTZ7t7o)P5%X93CT!#r`6fPmcE9CV*U|h;=yCmjwio?^$-|q>SJZ$2hMu zZ;ZN3T7B-Rp=y_&IEpY9DD5rdv(%Ld#@F*6IHXk>$&Uo!5QNvxP_MxqGgQ2T!B#f< zE~7kB3s{3Q=?`(8!Q@B=>v&uQK`w(!-uTi@uGejHeF`|mv#(9CjK zqjp%z3IW=-K=v}sS3lHXB+YorD;QGtqY zDN;Z|=Zq1>a~jvi4-qbzXQO;H@dcNM(bh2RY@!z-=a-b=68 z*~Z^&YfmX|lpL8G?9I1v!TGWC9(IG$w~ie)x)}R!GwpvE{@8yA{t#*};jbENlG|{R z+qCm}cCzp{7;fchBw?PQjmIA#Dr#(RBI5ccUX1Sv^I^B!}Zu0nbQJXdsZQkID~=4UKKW3l?p z;qTh-;*E}*Bm6{dIw&E^NqeI-t8a7(+6;D4$29RV`^=_C*faB-W~yo*wvT|U^!T-n ze^sAR`wJGfm+d0c$vOFud5+#=>Ns-|ZKorr%aAxMafIJ`ULFpJubPw7Y0<)6aQ3BS|S@GY!53Wd?JRF~H{oyRH8K;H?&3 zOf!9_MTWvRD3Q!TDe6(-SCTdZI4VFrEBT|-&&3ZFPc@&1E@8NhS1zfr&IlhcWh$Ux zkfm>VBMP zf45$p;>&w&BjOIZZ{VwjxqF3O4PsZ8cMN5D?5$OnU{Bt|4t)#p1JF5$s2bS1jouIR+Do<7! z1Fb~z+CTE`*p-+|?P`BOJ{tYJbblMAr23wZYdwvWiFIc^;M?U)pC~INteYfIPdiB* zf)7KvYx`UHF3VGn>rnWQY%K2NZ`^AFCzdwJ%G+MsEH?6Fl|@-p9+_^r{(CO|AnE#S z$4~f$sl($LtwKnKNZQWo-+*!FN4iE_{{VV0TeP$gx*}@683udnN z#1Kx_&Lm8pjL6H57O!lZ^fmtwfCz{eu zq~zn|DmlUR`FG+U#vKn-v!6%PE^XFS%Evsmtu(Ny0lryfloG&l2FJYnL08O!!vTSd8^8IB~n^3v8|s&E_39YGNG`ke>SJ>{S*W%W6xS_Pw?=LR`NDFF=A&js;C}fPs4&$Gf%1i(V zDl4PYekfb`c2fqqX$8g7sUymP5-h0slt{-TJm7)}Jvpy4KL>R^KJHyg8D+bKKbz!* zDKK0O%-ix@^c^}?`+peR=_s0Lu%A(8EE?3WpR)&O+`CYwa4=M_9r0K7Xv;{m2~DkT zeL^{Y-MbeBB)G)ttFst8wzC2_$zpjuab4ZV?M?Arqm1a#$!`n(^Drz* zyN@G!vcBWTLJuFO#dh8;u+-A(0So98g)JO6Dvu8F@__NL-h zq+F{gBoaBUCf_e(+Z0)AJxKpI6MI-LqjKU(h;37v7u`QlX4;{`1Fej6TDx{Id3Ek;~-`udj z_jbV9Tg`H;g||$^NxL3|EPXo`SOHroko)x#4NCsiM zxAT=Z$=!ge>ZK?Ct^&}JR zLc-H-2Kn0J;3Y-JfueL$0ULG5fD)Q`l_37~^iY|~}>S9p9e2pVQ zNnRL;&r{bJJf6b2JDcqW;G0WJi0&>eo690eGawxb5=%JbcLWZe)#Lso*R;JF$L*Ha zQ%JkEh-8??RhOJ>FVp>?GCJ3qTl``1eVxpal!Iz7b0%Z523lHmwEfSyox+DIgvarjmJOXB^G zlNwvJGRw1-*cBo!PH;1pDaqOnGwDLZ>#8Z$Z)S7XUM|)3+i<#-vaAaK05!xbBr4eX zNp+2YeCKdd3IcP3+ng}zrq!YFye?!c305f_<#Uc^QW1x!`8dZMVyXC-;s&*-NMSlO zKRFHNTSFw$BWIB4NkVbl5spSHht{urOZ}X(8KHQG?^Dk$9DslU`4%!sAZLOA>&13M z4wBHB)xPG-YThXqk=kok7MAwb{Dp|-Y>t390yg!?&(^%pTEFmiuc#d!<5ar5x|&1& zhjlnHlEnV$5dc;T#y2@{YQVkmJ-zg1Eq?6W+Lp|g?4e_J9N+~c`)=bOjd7EBlG{s? zX*7E)dE!x$u&ae#-L}FoF`m1L&o#7YIGH)N^gWV4i+>dS1qsuA7 z=&;oF3w=H{V5K94?)HK+ktC%+9dZzsW!~`|o^c8m8DZ46 z>{mo-CiNrQ$sKlq<6Ex@>Sx6Ie6wB)c5x%y&Scy_yDG+ZFl_zSTywy$f42CoapJuy z*F_p-_K6*f36@Pd>2&**c|P;BQL72g4@r;!JC3#UhLx)LyHB!f{X0Xqg56&^nPJ;3Q*W{G2=c)|YwwlD59=`UL==V?AHS!C0-8#bNBI!PR|DgD=6l{qco0nr!dVJyT7Wow<$}g3SRRbZsnyt_iH~8-CI{--aEe z@YL<8NhC^I=E;&cK|`I4tXglB<8r)cq;?}UHj|@%&)yK!pr6C~ma7$mqnYe2=YrzV zjH7N6E41?z#lcpHs`n^s=Ud$#(&xoD9wL)Mit(W=(OubHz1u||aGqo?ury~MeS~{S z&$e=nEAE1I7Uz&^{{R*(b=^wx1=BQ}Yt2ZThFu=PqPb^61Io+0%MMEo@3j((mSgs0y*|jYAxL?=Y)u8RN0zJ@Hp&0VB*XTPy1@>DXig& z&PZVjHbUihuTR~L>WOVDu;A0im zLEz6F+grzBu11Lqsc$akXPf0_IXh!w0dA!31&33`ex+Id$i6bt?V#4J<p&Bv;s<0l(mz z-xO}VOKXG(yO*I6dr!*Ae#Mx_$T2OKMf9@@HOl- zIdutzOCzug^c>TiYGXUBU#5M1bg749q{w1U;Ht@T|y zP}$}(ADK<0tsC2wQ@E0iz`z;iysG(?NjohMLKGmQeb37nKj4vZ__oSTH$c{7)NgIw za!5$yU$@VIsX7=KZ!aW#ovSFyMs|kg|xC?+RX*syt1H{GWvbQNENm)b;^*+LCLP$ z;TP;_wmuV@!7bYu=xK(BwN`ssnV!s&Pn8+gsbph*tv(F7mt(M+cwq{{ZbLu6Sn7 z-q%9c?>swuB&`L+7Rx@brj6Oj`PChMi__GKudCPS{@{3s>upclpNV=u>@9KP8N46y zF2+40Qnq%1FQbm{O)WBw#KC{%#{|T5A1MbM5-ZAnC4a#}z6Sgy(zJg9{5kPewyk%j zFWD~NQ8zboOd%g3d2CWOSsSanhs$&r$*;-F{{V{r03S7vJ5sWNFWV5**bW>?E6CAO&c%3Ea=lj?M_8G5+ejxtTFXO2S zc;#iW+UI;W1*LaZ8U5nVw`&7~h3nTC;=Hdx@OF;AE1vtq-Y2vWPi&J)vPj8&y#3gj zFvW_2*LTaFdgiWM{7JU(z0tSvUxwSnSCHJRTFI$N8W^^H!iFS~q}|kRaB;!z2I_<( zvcKTSb(`GwZwx+;WEzKsZmri&yM)N{UtY`hr-@I_W3iA$9E=YbUCd9*&1UKsHoA;) zTIo72r>AKa*^RXum$uOS(s?eBO8o%oxkDdp*OhpK#u}H!?N9qtOVnl4?wU1;;GQ_8 zhvg*gis~_xIVS*>9R@vo*8c#+-YvDbmqVFg)h%t@pgO$j(G97{h`#xkla1-g$I1vf z7aCG)Q#~2!cAgjTjrWY^*R-D;#FE;gEN0jkCwVjXe$gwXvD^kZBph-(S4Xbu`t;FS zTX^5X66&$WU7^%9Vz*Pt8UEz9N`aM!I5`A%ZY#j1eIc#i{T|{ANl*zRg^bV=yyclf zTr#I5e7uec_2~R9&{}GjKiSgVG?I}U{i1a$*bO9RKP>M9AS{RF7(AQ~l4-RK zPAh38*Pixm=gdS;?DW7q;1icVgtr5yYJ2WuYZPugUEp+s`$e?-8#|LA3FnzE3E9RL z+1M(R>J;=mdi8oUc&zEkt@v`*OVKEml`Z8<0!!ejFwM6q!N%0bZPm+bJ}c1lYu0ws z%U!Yt1y&gA=tmAC4=RbM*6bnk~7&mLSBCOA*C?IsPN@e~!FKANIxF@$n7eC7SC}o?$FN94UBbFCw=) zK*!S+#oBnJ-(G!^>S#^CEW7QPTejyIBpEj;<0GIstfQB!B#q~Zi&nAw1Mz?U3k&gv z$5k)jUkYhDKiXe5nvSNUQIX=^^1N`vsL9264e$IF>-LoKex!8I5na!$ zA;<1?t621?Rfrk=aac~(|Yq^r}h2tB*b{IYQ z1JruIhPxB(+B?#5?GJNM8;= zYR`jH=$AUj#B2L2Cy#>+8r(XVakP^o#L9^h^Npd2-;vN)k$i9X`S6>=HaEIQfc$6S zzX8pAZjOJn^-zWsjU0&~g5prIs{#q~*+Y;zkSp**Sn$t|b)N*a-}dyN-FlWvV7_^)df-}##rTv_ouW`>1 zebcD?8u*Xn{Vn`sul!@uyiubj*?@;xO*3k#KRH%kn`pry6)|OEz-X*+v#mG=I zx+PyQw*+BIk4)B0rKg8AyP0BHr;_1Vln&<7CQ`!~+qj=Vr)dMGYt^TRYj^h?RVR0# z-7fRNFlwK^w3I>)_~V{n>&I0j7|vL6&Tvn9`VYf@6?`?SY0~Kl@aD=}epsW+=KBU^ zWL>*^5L?!~GI+dmZE+R4>DKo0GZ@=#zHp=NpEx$=EIJ&Fck5EONEK8eP=|#|Q zBPbf;BGPaKthfrf{v!DwP;*so^&%}Nsq7va*X%BJR2~t$PYtrLCr*ayNpiyn9#zlC zs6hF=wd>C8*JKYTeS?v8tcIIX>EPk~+>i8b$rUMj!SHFy*Ea9dk0 zq*y;bUBolM-d0E%`PG2vK(2eOj>t++W23S7;|GNF$ZvnM9vKxxaIL`d@ zWIb|5))?oJUvYdp{{Vu4e$(C*w6xc}DQTo?7Itv?QbBuhKC^CR1$?=_(fg^FXa#n( zvJac@17Dx^pAfz|_>)Cw?ffTV@*Pwm-*hOUbwK)~(^@aL+Zqi*))ewAncG zUe*~|%#1i<3~G9WQ=I&tB)$juTX^yKD^t>SsTh0H-} zb**#71sD4o!X^v7Ww}0lyUD@X7TDVZIXQ5BD2@^fp)r#qFIr+JvVosy$ntop`TEPuRcqst5#s3+Hvm3~jB{k?K{hgADo*2<|IY zMLmIHAKG?5|I+@pJWZr{2gafZ{vcVU#EMkGaR%sSv~@U1A$5N-Kik0@v4fGtb1-fFZt$*?!a*?0o>g-3umy_`yI%}=U-rH5ABHtOR_nk% zJ8u=}$qe@cM7xD->d)(G)&`d*itY7b$!09IsC)X ze`rf>cff{qvi>BnD?eldJK{i2}pCXM4y*{j59Z>L2yy}i$b z{6QExNwI`1avd&vWJzsjh!E(L8DnBLouGzaTe#Eo{{S4`r~DQr&{=ABlCo+L zxI2t;$)7H0{Z>Fh8R~f*9p<;GXj+Cf*e-9hDZHp6ztR>N&zRUvw~#D;Y#^Vy;hHna zW?FXsVM2EJb@e_R)IKoyTgJZ^@4gIcH@BV~@kA}=+ga*x8!L&{HaU{m1DUczV5V6> z%K!)_$B+IUCy3*UA0O*iGF)2*5qWA1iIIXw$mcjcNey3nc+0|{2mDYikBqDzP&U&U zZKsk@w-)CcTgsaum5=cu1xUy_ug|S_;YWx*0sXSmz{{!Y+6CRc_1sX$cGLMcK4+SW z;(6{ap^s^0A7)8o2N;eez^=t5Z+Pc&wy@Co&s2)X!H{AZE;ZYGDDxci$kE(q0}JK3 z#^Oib8vx^uroI>Wk^6XD{5{nhMz+$gTV_&ADN9Qt=jLyb)wbs#ZXYXtwf0wwJ|*}! zR<@e+QPFN;zk@9ynZ)+)+zs~gtGYPJ!Da}{(Soodz6{rVOQ3k)#P^yUIz@{_fd2a4 z&LU%>JgM2?`kERf23W+DnEE?A<^cFmOogTn?G=(^1l6d%XoFYerUY zJ2I7J2kw!Lo4amKeNSrlu=A${?jy>n&fVGZ&6c;X>vr(?b5^-q^&spsNQpQkWI2%V zPCGFi^fjbD53ao#o55)_T@aIfmi2PXySAAN%Z0#iT=8F9Tj-i$hABK1WNfZofEI{? zSXxqXkmwPEHyJEKj^tN2Y2ZJI3$7-gScD2Ngyy1R#K&Ls#ZU=6)*&2y! z9SrsAc?XBI$-ESr_ciE4!Cwz`?ME7uC578< z*s;mw$1Ar#G=QC+oMWChafJy%WjCcz$JFY`nMhHDb7jVAT#vOhdSqiwHB7kNRG}$0inDO z-eRF7i5B76j!$!reKB69HEj-m4YGVlp3he;^CQ`bMc6IIWN?thpHgsob+13uymP5) zx5I1NC5-8pf=MQtG8g!RDBPv7{4y2C09U7IR(}cg+xv|_M4nrlxmY{tv*O$^i~xSm zIaVwQAQdd9j+q&*c{jX(<`?1CgQaK|z98_d6Y9wvs`nPsT%lhpZ`~plO{H^_mMlLS z(D0wd{{Rw22iW{SqMOYe0+Q+RLiZNU-!e48ut^7Y7<9oMbJbdF@8IoDMACdSCYdQN ziy+?MKLv{!SR|PM4CG_3PkSq^*!bEBE_`ERG}Z$#UTLxy-dGg@?3;980!HZ~l@D>~ zQg>FKp{GNoxV^vlxUuQcUyljQ>Zj~C*9&cC+lWviYhZrwP_iQA6N=$Aj}LgK;V!)J z!#T3pz)sdhQZ@`Wk1)yRB!J`P%ws3-ZK`^&fp6^Z-&zp0+$jS^aV?BqNLC}}-mYbt zTPnXZvJ3;tgNp7vEqANwx>Q#1=tlcZn{)k>Rso?5SO#QiN)60LSITD@=dCq$@allT zNjw|l>$E!r&E_*L&Ad$_Z)^u?B350*?ZaSjKIpE~{uS4VbUS9$r;ktj9B;Zi$VoJw|0@eIp3PP~>uKp|Mj1eS> zCzb-ASp#k=J7H4LGpxpD@39c?wg>R6+ z9Z^>}JDIuZo(^4?+RzV9H~ts!1Q6Qm7V_y=(-I|Dc^S65e4&`hi#w)ne zJa_O)*8Uso`R;7pL63cy^I$wN#gMo_LurW zkVdgY5Xhe|%-(9g9{>3^mo?(*a*36Uzjf7upVYdXdGrC0@;0!QV zW&?wo=Ct36T2G4(ZZwOV$5r0<7V8{x$ellVA2>%GK(;KC4wuB%M7@n)}Ya(pt-+yvU~4Z`_y#;QJ6f17yPyzuOYsjZDCUA$01kM6L~ zAPS}7{W=R4ZDYeQuBEUOIhI2cMUq(m0Hde{GIBuNPrrKUHLnNwXT)-?x{FIo-Twe8 z-0wE$9Ds~N6<~T78+!~_CF1)}7*Od6j$6pV$jhc^yE$)0%yaMx?*_$Kpd66A}DB}SG9svfbI2A2+WWx~8 zyL?6QU&lH;^4RN_cS&oxM%_TI%y$6e43G>*#yBCA_swt@o-ei5G-A4xQ|W-SH`<}Z zaJO7~*b`=y6^`BO~&p?phyZmpnct7B~z*u)YxARvq#zG(nA`$_n7lT{`0&bh03 z%Kk-#%P*G9b!S83~xNEz-pJrKcZsS~Zt&#EqD*7TX1QPXtq-a?R%AmWjY^e`1V$L~GH*NN;Jv+qG>l)))&u6k* zqA;?^Ws}e%k+nGFoSb@Bx9A@N{7)5)&bz7Fjbcc4`Dv%xxw#7caR&oF;re9azLEG> z@KfOjk8H~Lq)0B|$Dd~!2yU(2PB2Qk%^A)KB#&&@HA=9Kh}~i2)z2L8Uw~fMN}pS` z)l7F7LM`WG1bdD_4J2zMiaFqg<2>EQ z01Rtax;4~+Pzzg#?jw|S#Fnyou?+XxLB~8A`8VQ^jM8SeVv=0Ik%0FadPfmi2mmtXLa*xC8a%vVl@gdw}ecgXqqyRI?pIIlbKuYt6^7XCS> z(=Gf&75-_h?%CyyLBT#$gaNemDo@hA96cK=z0N9@WH;L1g)eQ)cCfCXk@7>p?>Im1 zw<!GII>4q1D zu*z>tZ69DqluMF-x<*Iy(Syd8ng)Xg0NWjFdSwZR7YkO9`_kyk9NSsJrMieTb-Ou;d1ptx#@-y6l zR^j-+Z>GsK(?f8Rr~*N91fwU84q7+kuw44$x$Ey8c&k~x`&E#;(=DVZWoaZ4#ULkc z1+gWE9Wk^M*P7_H-lz`a!q(QhW|D5LrMx$2?yUi|`!GNbG9+g!Ajct-c5}#Hm6xk) zUKSUAefEQAE9}cjv65T?(V7zwZls)s9=WeIyiIP&g~7j+!y|5yt!>gdQzty5GvJlu z*#au-lZ$ z{{WU_amG37K{Zu-Ez~JV*++Aregz1ke*!pAOSL{3R=F7l$^88%zY~K&<{5dmQ z-Az8&VU-H2{h|xDWVbfb6Y>^2ut?gcf$y3Z{1XTEYi)MN#@-6?9qyfGvqLxWbTHhpw8`aSzOB-1XfGA|Jo#bR<$0Tn~e&Xcfyo!;nDP24I z9hBt+uDAQo)~#Q_C*a?R?w3vQME)Z1eVXh|N5eNUF}GGxlP;SI!{x`<r2vskDMyR16#``3%yZE;b||nUQ%I1J}hoed9Oyd8W%J ziu5a+J%cjHV_T?+CkwZ7l{<;raG^oRt#3^x(tnAJ)|($SFYQn98^>N7wD@%T9p;T} zk`nWHdRc51H9cNP(UdUc;RBWGO4Z5qABq-}Z~K=3a+FG2}#}YcU6s z7#^H=u0ve-<>B8D-9;yltYxuFnAQ}O%9h}wj4so6)9orp1uO*+yKMuTeuEU|}$NM7wMcuN}2DpP5;iR}9ypJc6 zOM&Z-m5HtVNbx6xEmOi?IML*~vF0#!Gxn1qZm@|Fyv7*dU{C zxrN4dlF-IvZrH?%8-hp6f_F7<5BPh;+EuhZEE9dICl5XTl2#ux`@o}?`AFpYbpThJ zXf8ip(m&4Zpu=I|--y%LUHFpj2ro3i z(XGI?aF-G}+RoxR7A=d6?c@@A05Mv--uE<&dbCBOct=9D)b1wuM{x~n7*MJ zHg<_OF*s}-908HQ;;^i~A$XTlj9Kg2GHbSlu^d`~TV_x`@e8q}k%wX-&PONHvHt*s zR!4*v&kM5?ly$em0)L1%9`)JScymtiCERxYEYhsB=pPnHHX#?Uhao2Nw@4g}B-? z3T+=wYFc};{=Q^7dg#u5ZoV1Z8E?MFcV^7wM3*v)yEO=>cJxCGq-T+raHEcTRxZ8Z z9~OA2+-C-v+dWC9vtwk)Dqra zTM1Ppm4uSKmuvF^K}b%`gpzk{Svq6hyF0yCS@8Y*E8_J2-_zpr72~^&+A!GZ4b-E| z_Y4_I;9z4FjjZ_J!I~?@2nR&JEF|BlV*6trO5$Q2LuV?=gn@&OYE)$OPeG=os|ybi zYgQ0Nq4;RNr5BdpEv?Ed5hz^Zb&#)=pmYpB-5#}VHOYKK1;>Y{w2Mr;8C~{U?q$X} zkH>`DMADI6#i_S$wvJJO%~UbVpUj82~~GoTxqW{6XUX z01WuL+IcnNn!Uq?WA->6IZyx>i;0X!JPZ{G;LM!juoM2#%I&OT+oMV5CmjBaV&^PYc$I47lJU3kaFa+$Q>?6G)|BAFWQ@e5^2H`a6v-69X{?ruYk>&Yp6`MHfa5*QXa8me+ z;>FBfG?P)(VRR72*HI^vv3mTr$}x=KGXi4On9m*jntlB8Z&lMs@}r0k zxwhkbWam4AjC2CIi?0nypm?4e+sjLhPIfqoNp%O7akzZDVUdhcF=Rgo0hB7wmMIYN?i+#t;>^e)Y-vog<0xq@ky z39)G|@R6zsQ0zIuNrP?AP8UA?>xS2L?++%X_Zo(iX>BAcv?&gxo(AQM0Wv!y4ZK#K zsixfCUYESTXrTk=h6ut)PdNrw$U)HKXeZQHIjQQh!=S*Hu!$hDkh{xvjeOo;Oa_0I zNbGlGHQ5N?SE!{=OOW5(>mDPuYs8vj%1DFlC@teqqjuC~#Cx&)InFwolGYtgSRJnB z4G`P=mI*jR)R8Kb+6eV6j=ihfbT5Keo;~n1emL=Fr(xmU6U;3O-@XeS+fL}W{_mhU zEHjP+3|Akkd@}fX;k1tL!oDuF)vj1SpZ%U$P#(#NRxm>h9Q@?<7!{JQ4Ps*$H?i*D z4SpX>@sq=Lrfpeu2e&V7Y;_m8Fz|7JV+J9B0U3$Ab6=)@75@N&ciZ^y!^++(hCMpf zqDdi^D|dMDj0E%ARO$%8*%{7zSLOcz?AhYIF3ZKsq0i;%t#r|#xZWYSExXHS1&N7@ z2FOyUDi2;O^;bstm+`m3lYeITKgAkXimc#>x>{<+ODT_bQGITbKrlG~+YfVICN8AY zv`SgAI(u7Rh&Hz{HPk_c z-<3OCsVDB6h)^Pt+mK1>0L6T<;=hPGUCh&XyTN`W@a5bvLXs14JcXBWARxF6u~@b_ zU^051;;2@e(Thu;HYbGs5W_q(cr!-v3>uc5wpWTM-x({YcpE-eIm(*EiU2RkP-(RHMo%3@e5G~! zO?b}w=1bene@-(@#oU`xa?EmhVoB$@B>I}i)U|=8NV=8GMt6{TjV0Z~LKy!5c~&*p z0VAB86V!oL6XR{cKzx`JkGteR*!It#&OK^nLJ8`KjaShB)BMA;@Sns>!!y8|m)fml z3?h{zEAoySKP5oO>{pI4n)dr!{{W6K7_O!8#;Y9iBDvO9bEioO0AZRbWoYIYA9a)# z1J=Ij_@nzq{5jX5m&3C7ay>@IS9v1HmOn2C1f;taquNG8DJMNE{FQ#f&=HdkhFC;~o zz+u50C)bmXwe0$T{1lJk+StpXc$&u3#TsSG2<5fZu_Qx~L5WnUiGjfkqi`c5-{jdf z-wA!a@2_vJubr6AVT#^q_uzGJnPcyQI3V?^4X!Qrw*C8fK3&4Y<#I049v%3s^#;m)D1N2z#z*HY8(W_g}=(_;qC;fN&3a{p0usl|ES-|f-y z2U;sN&YIDyOx{y{t(iw)!wC`EM?b=SM|%A0wu?;GH2J6Sm6~1MuMzEOEre>M?k^m; z!u7|>Po-~KYf{{#QQNJZz0@F?fiqkzs?CA6H}d0C#yBi9>6)luDaS~m&l=w6(ifi@ zd>^j(g5u-EF*n*J!(7_IZq03TAUqqWBvAy6N{^o+PzFPHti1;7;1;u~P2x`s>G0_u zA&+yzrOPG67VX>2jhiaLfzBf@jNpO69yzbKhi~m5(dE<^PP-Wm2inEAnArJ-LbJv? z5EP7kDQx^P;UTC*pV*#l@gL~-6UQyWAp;<_)V^YZaq^y9rUnLUp<$!bMr5%L+Md~O z@&5o;@FbsO@c#gWEo9Zpt4wU-m?&PLGmb$dbyfQD*6yeA{{TwyZK`SBH`A9wXeKQk z#ht*GEbClIO~-^m3I+eYCb1xn{8ALBT!gM zH1WtxML{I1E-;>l2aMLrlu?wMdW^Kwq=}jI>rtoN#Bd|K#3h4~oT`F*^QfpY;gKXAOMq#)F zXDM`Dfx!L~^KsA|3hA^j0eml-JADsZmh)Y=-Im<)KbAUu^jAM!*Z}ZA0=I=ky4;?J z%X)8)z7}6ze`omWL#I!ZvUtln7g7K+Ll(k8>T!YXUeRy%p9|V0^p<*jnr_cD-lIit zYa~sbubg?YzDL|D+@7bceY5bp;LpQ78(WSq25UMtqZme%*xbBOY7nyEjp+(H&c7}; zsw-!xH9LHE_+@8*bEVtKQ>}0nbh=il!xB-HFhtcRXjo z`YxxdX%?{ft8a)j%jDd!J||e>CxDaW6f687sBSVq&TF!VP4J$8k?DU7B3b2mU+o%p zwRz=BBIkv)crc}x_(a$w<0N9W4yW+@#F1SugJ9D1+xxK`lFtg6;2Ghbdsvwv4nV<~ zLHR}qIN}*T1^Idj*EM(3ZM5rn!%cr{F!DE&HnW9lJ@-^L9&9l}Y z!B0M^Hk#T^)Q-kWi^*b&Spx7DKQJki7#p@30}Qu(Rt~N46T`kXMz+`VsO~ixG8*#0 zh+L3xrIE~NTQT)UEsx@ZKF77_YCb^*bLYa>C5hM(fl(S1%@1o_R!2M*wlp z33cF~g}VKc+V}>0STzY5m2I!Co>!R&0ahtQh!wNIGDh5ba%lFE>Wo!4(VkIt;KjM$ zsQ5DI_=?3teWUl@b0eMrn%)M6NV&?8<$>=N1n(R+aRm}NyvnpyD*Mz)(2mDQL;?gw+ zkom4d%WAS7n3d5E<^Yu?LaF0ExEhV5ejh7;YjZBGHNTV=TWJzPrywbqq>5E~45%Rg z0C-n_s=?q(Z9LrQ)4z!<+lHN8r9h=QS4o2<4Gsy>xaQe*ic!x~2n#)VL z(PUiij6yD>Wh8^Q?BxSzKOiivob|2;;r{@KwQnDIg*-9g4SP}2wFG&wZ>lW%OShIX z=H4k0Jix#cjqQcYFyfqGlHSAQd!x`SWYe@clnHhut`$|~dxwy*`FA16=lzA+qP^Qn z_+4}2wp;AQ8Iv0t{e&qeP0JIIp+(yMGB7{F ztVRbICl!rWw@lJB^ttOAR+aGM;a7=t{{Rbk{`MUgRJk5?qsa3xW@C_JyJi;q^Ty9P zn`jtN0LRmFcuV2VuX`tmHMrJ$&86DmC3F)zjv7@!sUc!A3Be>}Fza6lXcyiax6!1B zMe&t}yQ5qjU1_tzl39izos!zj?IJKE{+YIV=DBT8$Coo$X`USTZ{jF?Ef$>%Lmi4( z7O6h2cRpvLpXpPG8h~|n1$m4Ktm9fD9gO8j1ajg75@txai zo-KP#M)%~DT-z44yI9Eqb8xIfZ{s2TdBt|np!a4qWS#Ay;NCm%Cx@(z@%Up(9D z^ZBrx9BuO(5&{oUS+S04uYvp_;^<5Q?gJA>17H=Dia| zpFmF%NvZrk(scg-63T$)+TsQpoD0+yxRPmOp9eVG<@yCV=j%TV{?I=aXT2W}E$pme zkYF{8tsj}PKsl1sETpL)FDa9ZWY?yrD61olo>SrP*ze-s#XAPqwRO|9_*f{4@>_!? z!BS2Fi(u&40GvF8e=6F${fvBj;f+QM{{Zb3@lBj8tu>{k=+o{&2c&UaDT&DGwWVFd zC3_0@czz#4;d|%SJUge`_*(6ijMnxSQpcw|za`$x??iBUCOwBd70)BeptSMIYv$j_ zHuGvW*K=);Qql;$flKg;@xJSrd-BLv{jpr~ttas4UW-I~@59NyAH$=F{G01PEv16u z4N*SEEu~Tsqw?h28|^tJKtiy?DI=Qo$UJrXId~rCE6pETlT5ldJEi+2yWi>Tcvlj~ zDG01b8{v=~k<<$L3smse!wqZ;4;5WQez5W)zP>U^8)pY#Mv_Dz4oVo$7~>VqKZkrJ zq*%|U{65s?zh!;a*0<8IwL-v#RJll{kw-g}$b_ElgIrUfr7OvrDcO6c-(EWXx;`5C zO6B12b(EL($O5eJNgD4eIrAZ1vn~!sHO&THH&T+_+s~2O>NyIKsL{%$mC$YgD8cE) zcGsQ@_-%BG-Z8yvGieYXXlITcvKGfN7C9Ypl?OdWD&dBvz9n(0?QCZ0x{t^265PqE zF1>MW4*vk&EzE*xqQepx%(1rf^JC`PPjD-}(R52aZsYzDkBR;^y|B8HZKlTV*eYEq z+mjPc%?!*z=gZE~&fM3``t8=A9h}y>4~SO$MJWMokv#W}hT!s+$jSSmLLX6#;<)&J zENZam(%4&S_m>v(w#jXE2T4mhblV%DSeOot8ivn8E34UQ?*2tOm(ic2n(yrY0PqGo zF{StmS+>$O7>GOMh+U=zcNSJFh5rC`m#+tk_;NecwQdY@`KxQY8^6{t3OK`z@k#5NM)J`jncr)x=^(yFOfK#lhS$2vuRv0H3aV`E+69xNSefQd(czsKno6Pa zcC33iM!*;(HhpW*q|(ynXFq1XBhs-BpQBtw6B0jqvD-w%jzMWyWOn&gy(_HnN5OxH z{v}D5O1FmDC2T^G%5I}Z2b|>-X~D<~kUQ~P-VpH?pQ-C{_*+=DvPG4IK`va}$PPF! z5}Yp?97w*k`lI_bd@t~9dL7I@1e#biIY9DLRFYY2hSGO@v6FQY$=Oxh1A;Sx4>l(k z%dLy03(Cjki~i5Q61*XG2BGl-N7bUUTo1PGGuwi4KtEa-Ts@UBw zr=!TRf=L&v`}mjz4p_Rq;epY1T4Gh2fgvGUYcI z!^In{GCFgO+>UWxIi?LdAsvs246=>DZn|G2bivz&*~dKYBh=UL$L%rUjcGm*&bF_0 z;!SGOH(BmvvbDC<=Z`018DUQ-xJ+OH8nMG2xy^nBj}mxy!@Aq)TD12zvq^FL=d0D<2>!2=VM? zCyagduTio16RG(^>%tn$zDl>2w&(X(K&Z)CC6ox%l1?_~BL^b8e}tNyhluppo^3Ny zvxK8|?DuLSbE(Ry9N57OL;05g zukf9?JxCpE2~>;H);6Y@b`1ws@GhBcu1s1v1b%M)1~Dv00|8o7`<6W5Dm z;kS=<`#n>}8V$6U5T+u#ic#l|;ocSsj#ZcC`Il+P$2G%Cq(vORY-IKi*Gq zV=QT=WO7Z!XtPLEj^;GT=Wrsl{4eoiOTN^u^$4_4V}E?Z&%V&%^Cv~e0}`x+ApmaK zwD&ybvy>g2mZQ5;KSKT+`~&gT-m7FZ8?Wu%9`v824!*^VgPY;fr zckK`HtM)bcEemLWv=VAQCe~RR2;tK$rIzI+w*xVqLNJiWFlHYt*C!sF8vLm7Z-zWu ztX}E2$*La{+FZF?ai}%Ho)_W5vhYx^-pb>rBfWGN9y{?jfn4}|;s&wy>vUa@X@3N! z0%Z9>Gs?l(4Bho@J7yc@~)ilFx@y^cHSfXMD z&9NVE5&$=RiSlk>c^i=TBNfGXZ{hEUd}F4QZ*gIJsxwI%*j&slXOh^GSBTa>F~H;? zn*@$X&TG-N87%c3GFd)1TUuN;{l0FOGRb~sJOc5e-%1c|Dm)k~oijVDo|!RcZy3v) zAPk(G_oB~z%S|tIoSzeG*6Dj?b#-qElMy}ZwUkVwqbm)Iw0f!>bfW&xUAvB0yj6E& zs4JI7nPFij183yAR)i72IV$<%HOzlyMX6d%t!WZZWi`$gNcF)YmRIL`PdRxu@^S#c z9l5I-u9>WOVUjrnx2I?lExARuVyB>aw>K(zDwy;lsl7BCPUkT-mx1)#ud&_QE~e`3 z=02rx!qPQPfJTiRP6zj#IS1Oh5BN#7ttD->4+z-3$$j2Hv$Tz^He(D`<8vHf;~S0+ zF_J4P9~1aO&UZx8(@maMX%^)jB$hTEN`0kWLmt}@R?YSIi7vFeFZe|)WRlSlzSVs9 z1sZo!#5*j?zEHsdk|65DZUA(mT3X{$Oyn(_!&kSn-*}rzO(J<=$G0p z!wKin?&d0{8{)U9Msg)B0W#%AaIcK@1XrEwUOm*K)Beq?eX#kV0$Qvj+cZNzIJst9 zbpz!A3EXl>$OgD;jdJq(5{M&$IFNuMl`e7zIml23J7BQIIAPZ26P4NO^Z16_Nq8-^ z*)DGF9JcBl3UO-!2T2|MUB%e1kXF0_`8-S&`t!RIjt z0G;kki=M=OwRtC=OF5>xvJ>GKSC6s&I;z1bD5--dN- zmxD)mE$4vkDp5qk1Cg=7WGn1J9-Psnj|=G2K9!)}MRBOP4J0t-r;wA9U85LL&vFNB z=9{I(Yd@U_iZvN8Uh#i<9PmX9YJF2`1|9ilX6N#@SHs#=M%Ti(%L}MfOKV7l$4uZb zjT;2#sbSojMID!!ENM2McMM`B$C$G@RXG{P&<^gNxIVRKN7D3p?g6q5YV0yr&R2k_ z$2t^y*I-5X{c&(*vTyTMHdq+R}y6WpER4D!QdQz z2B>^WhfmZdwz(G3+8CE?zG+lkZNS^KxqP0S41tn!T~+RdsOU(qq`V?`;cX>H^R7WS z0u+WekhgLRcjCT~@Zap|@jK!dhn?oo=kWFF4>~CwK1Ub_cHAtB^8Ww<8;L(IdRB9* z81BuZNxNu%cxzrB@N5PcZlu1tYzC5BTtgcSAMZ0TQVs|Ylk1A7a})__zv1)Q*?{1#iN7NY&>1!;*)E-c%B$`#3j^& zTe9Palg)N2g?$w{?TV{Ef_@fTUdINntlZqhS9r8jx+au@r#Cn;D8Rr03->1&t$kBW z@%-uK-rQVIZd1%zC3A6Z;kXPQM<5RSRE%~Q#ay@3d_APf*EhEe*MkV>PG69R`^R*< zK|k(gv(mb$$?S8+o%Cq0pzHTu7S-Upv(TZNPKrSCBr==UApjT->+%vgAyo9^HT1WR zJV)WL3~RsI{{XZ90F8V}sp?a@jh|+D?4mpe@~*CcRti8+u&2xXL|1`DK8K@Ro8J!k zS66Pq)u)lt;4TQk5@Abw1Jj!K?JrpHFN{1lr?#zUtlfC7(lH;GrbOaJ3IeN%<&kZW zxX$i=Va6*Ov)5E~Q>Sgt-rO2 zcx-AKECwBrHR@W=bsQi976~LPh8P>SBd8e1IIk46@P3`)4;)-c`%YT;r7kaI`L1M0 z@9nNmzSt19Nnr$`Lc0tG z`7k}Xt{uKD{5pw@e{Hx;ryEgW+A-O1ae>5 zDQTsnSO`vXe|IGFF;Tz+a4I_-*AZvpKL#X*PYd{RZCc_I16&B@aHNh`%$=2TPj%q_ zb?<&P_~+nD8+jw}M}hpeAh67KvHZ{3nopSL__8zfubH1x)^4>~ucWa}Hc3Xpw)YP> zVmbMNL|&Y6l09qaX~VeXJGY_U>E0l_ySaOtdlR8Eg5}9ajmqF+L_x4I>`BiU6%-o1 z{lx2Ou4)r1%;qrhq)?#^&UcXQj@T>>d8`Qfh4R_>iS9?6jCqW3Ut&;!29#!#vxE9EO+W=eIcY=|rHS^d{$br+A;hI(CmPgU+`XJFV3u zzva0gv}#o|+_MZQ++C;iQ%sEw(z_T!U}1U!mHA1#!lCCm5`KEsd2M_)khPzV z{6nTie*iAdBzU8FQ}||v7DD}bP(W^_am7Wh+4y2j9Qb?0TGUs%ilJEq8|9qt`=VsE zR6~vk$lAw@XBC^G>R%9i8KbP%8Wp^j*5t6gSWcZ24nnkpV39VH{5cFp4PhyoI3})+ z>{)n2!a4(MF!**G=8;25b!hhRTf9u!4sK`0)F0hDiS3@-79K0{C!6+vCrH!na2+0d z=ax*jPb{o)C}K0**DIrVOXFvWt)uY{qps;z)|ZUS{)`RU%m~bk>c$`dAKl5^4>;?Q z;{N~u_;$&yt(RW4ot%e;(8~64%6{$_WQ663z*5Hts5#=2)7}={PN!JC@XoV8*)O%* zh$Y9EmT1wVT@c?eTc%duCOA>Xdy&?m(=N1a3FWoZbX%*N%zyx9^B@hKtkTJZ+@r7~ zaqW&q7l%9{;q4OYDKES((Ot+HNSWaU*mAy9uPM)dtO5Ghy6L(uqp!4f+U0~JOjlPi zPj(jJN32Z>N%JW`cX-cC; zJu%2U8tJWm4o3tN>pyL{)vPCsys+4`e`!TO!dK?=0PljQr_1TpUVJLlW(h5o)y}DI z>K03DCyE&Y1M;Yi_q4w49@D|&BD~<6PYPO3sd!UI`z7F#RGM=$iPLEdws)ASBL4so zQy>HkuocYT@(A;-0&f&SXFb<~BD)Yvg>5|BH1SHizJ1F0OmoQu9mB6P{G#iXS@<41x65{bVVTK zD`9(rE6My(@k_wo8}S%|Jx=EIN00RDc&*;k0GzMN&z8mu1}o2~uBgSngIcSex2yOn z9U^#S@u9dPSY>NjEhI=RXpsOj4LcMxlO z6JB|Db20wU)K$F23BU^+Yq69*(}heE#tuoTv`>vc5i}VNqv7uo{gUJlnk!p)ts`QD zi4~BKo=o6jhDqu;uadl1@w4J~z8czl+4L0L&u;{F6T)8ulNHsm_ODi61%dA6tV zb4*Cu!^c;vcE}Yz)bT|X%Mf}>p;ye?oTzmJzd5bpf^BLZ*Im!h4R7O~l{#Fvh=!Tr z-CjlBw-DRF@y^8LdC`Un8MBijuLq0{{y(n#Oz}b_+Lwrgm!uG5dpTnk+ef6E<3c|S z=drJo?VraQwwf+93r5v#OpJ*nQ2mn)j*hG%zyp@X?Bk4{E1I#J$9lb#noY2`vW7+i zMYnf--IzBBi-LLk$DG!&z`{DjC0gg#Px0@?lIcI#db<6$PggruOPMCPgyWwoUAvUy zw&K67d54L7Q)x8N>K5re>oco7GR}}r0AJu#Wmosw!x{UzHRhN429cxN$!h*(t-}DS z&vPn5Q7Hv3H}w6=k^Ni+)9uEDZYJnlm?2Eikedjbw?)PLbB@GaH+dhWBS z>Nc9X;ze~&w%PfIwZ)W!$tq7RBuIuFNd1GrwqUFjoeRi+sYebDn#-Oe5FX;xPj0Dg!=QE%B2MNH0NjNbAHu&FN8G+ zbu9*Fy0nerd2Y1zNtN9}^CJ5{%_IXHZzQo90IoA$o235$!A*4CHP+)r@SJ+=cI-+l zu>%@o<=!Qb5rfxoVZp%WzdL>%e$OxByT;aJ6KVG`mWp(Ax%*NuAYr`KXCb&LgN7I? zKhks#T4|vW5y1qJ@c!RTEpp5M)kVLlZcI0P(>ZBekg1ss>vv)P6&2gG1?aktS zTVIyz!V~Jy+3Hgr`gPftQ&kqGXN?^|KQbe6^khsE(~AA*_@CjcPlqwfuXrZ@D`^RmRj}3M zgY1JUKX)6bS)wZ32Klj*m0(6bEcl)A!qOWRxYq9cn~5M`%WG`cYDW&pW{fOfcn_G9 z(~<$M>@|D)(MK&>cXmH1BGvWzbk??mR+QS~%X=GGORFoML#u(f_2|3DQ(U?y!{uMI z-@yg7>gCUy8(cyek3$$a1p6rf^#E7Dc-m_p5ziK_{hlO7j1Mwrh6z<0oPOhIZ&Q-Q z`q!2>gtYRP68LH2yp34P<--t!j1J5Q51yN{IL8(2({XEE&NiB7eWrX{*8CG_i>K(; zzCwYC?c6MEqZn3FxChsxS6e6T1>tWJSzE>9e-Bz}-cw>Iqbw$XV?Jb83{qIhIKwF< z5C|3XK!?N@*M`nGt*pc&bFA-{Q03ZRkvMUHxV*8uQ955H0W00_pF6{5+inC;v? z{{TKBVU3CCy9|W!*MpzJrc{&M9J-$GsD98g+gVy!cvjv!eUXcB+LVnf(v<_0k&2_g zW9BAF?rW2{_`l+~FQk^|OS@Ta5GR`)?QFz>!2vjC_5iLrfnGbKO{Qq`&;5;I71Ze8 zEycKLg7c0x<%3}V023Z_$gYW|)%9yJ72ca|2b545X1G(_Cr=P`7+w_zC!2_(uwtWWa&=G&a(lTjN>%^{$k#LF~fF?MT-OSvFz zVH%ZQI%NZ7brof_REF>S9(iwm+YlQgg@INL*fMXFDCvT8^c6{L?=&0JZuWaFOKnu0 zU(Za8E0DXk(;1^=WV4nq35{d(Rx-exZY6N*+$#`s*S&jRhregfj~^B+t?sP6AL0jB z{{T~eC~G8BgN=gWm7Pf-5EP8^Ytv`_jsE~??F&i%(D4o42(f5c~YKWi7Q-QFOYlHD8!#@%0+MISC61LTEu6(_)!xgbGl?}@vJ3;!OV0q^`6*8!# zN>27@(zf`Cu2|g1_FYvDCgiJI(5Q8iHwPoH?-7BKanyXI<3CEJrlx?`X2-#^ zq_9X?S|xy+RO1-)A_HR+&)(xdbmukdHcw%sL1>zVt+Wkt7j&e_G()${Ki`6)Zf#Wh;kbAeUfp}BF{{RiOKPy+%?d+Vzm-FM1QQ2{i-T4?Y z0k{&cxDMDgJ9uBjT4tdpqu`61OJRVfW}$1 z-v!uT>H4r?yHxPtj>I^~EM$W*THFkQ^CEee7$kvTpC;$S-XGRh%II1hV(u|7_HE7Q zkR^zoRpE_Y*#O|m*g522@Olpmd^YeVkY&8mHB`_`E0;p(S!`{P7gwxdi2e3 z{aWtm?Do9}{B!uTIs~!96DK z+-p$jw`~%kc8d}XuIJ_j+W~#dj0 z<;kw{v^qpeJ;{drZkB13&~NUMq`Cwg_>WPkIGj7dh`{&I_@ga zIb~*0(IW6NXmhWMbyw15C=%XFBy0BM<1gif*vskWHhN$h={!N;yKn5c?)(jNFNwi6j$k0gGEkBb81@0C@(}_~#B=F9-O(=GRZOxQk2F8hHe^hE;_oWZ?O3S0RgV zyVM?|J?gdAi~j%$`6Sk6M$mOTXjsQQL8X;KfHy}hr)JkYF;zScaw{%L@X z-0pEA6O3*2uVdhw@i&?9?^x3#irK{L5CZP*1<1fQhA|lm00(nAXV(-hHQ07Hp6>VI z(>(Uj-u=tzaf&ow z7igNb%lK=t- z;z$Clc-10E)fA853?6KpU^B9iamndgDs8S?^C3Soy0`H^!W+#pFYLo@pz2OzZ?@jX z_pKQ}!5os7jwQ+9>;!SsCp@;kzZ|r!dPx>*>+5-gJXcn8hPG%_3^AGW0;ta^-|JI& zcz7ShbHx{g?pwmTqy!Obe2Z}$%*O! zW_^%O<-u3p&Ua(3(zzV~z^a9!+WgGs+tD07=fp(4^0dzzMGmI8Wk%BoGb
|11H z4!2IQe6k&Lje zKT`O?6s<0mtm5;{xWw_e>4S4GWZO1Cget7)@E zqutKyi+xV%W8NPbVgr&DRCZjSU~2|jeM7$cxjQ-(qJoa}bN$96yLkPkKHZLjJc47uB=-p?{C zkl{S5dlwJRRwWWS8RO@W4^C@%#>&QScRiN>0OB`-1hPu9+G$c;v39w(fsu|0`8f?C z4UNCM#yz>N3iDO*CFQ{JZj`U7%Pqmkna#9`ka3^10Ol;??gPQ-YtQd~A^1|-VE5)7 zD*8k7O{b>G7%<$pOuL>k1_&PY%4)aTm8GldI(6~9nZu#2z5B(RApWMv5yr z7!V#zNi$;@cA8Mn$mA6vbK1Gtyg>S(7J8iCdM;Kt=R$;W!2(d^;NXBU`PVHirH+|* z2K&fT?8qb{QY4m5px^o|YlLiGfR*<`SZ>JJ)T%`CA;RToKO)tzq127n=2g+naf{drbL& zDhXl87QFi+0Bdtx`o}!)iW(Jw0_?d0wQ>rz(3~gDFFC9S{VYym03{OH!4{FL- z?X7O2{@jaGpKfy`vKdi_sg517ARe1fsU+817J+|q=1W`Ar8DeAMgs9p2P0zbP)AN2 z{x#0e;k5qJj(uiLQte}vn9Q@WhylhL2)v~P9y!T5#apssgta2nHQ$Hn8$)L_hUr*^ zx6_rRb$o991Q(D3p?L27cpJH|Tky`GVQXg$nzx9)!xJtI zjr?)TbgIk7(-S`4NF+Bwk%7UiYE5o$6{d%VYCa6RONO!W)%K@sp4I76{3$t0JvUu=BxDjqiy)hNu^jnmieyuff--&0bn^Jp_rg9Yj`sIc z)UI@d!TUl$s{;doa!GQk4*<5@^O0WdXXC$zF!*tGf*{asqWR*~G`obgkTbCWV54xM?Yof_+Ga^*`4#4jla46?^5UP0rU`D*_F_MiBT z9-gDY`rKY9)7&n}FKy+qxj=aNq>nJo+>(gA9s%e+eDN>FuZVsl*KK?!@ms|*={8e3 zvT5>KD6*6`LgDUXk^XESm~0Ktv92lmNpGRFTD6iteY`WHe#-v<4ZKZhfAIUnv1q!z z$S|=GTI!bD@>v;@L>Dc*j(n-X>zwdw^dI4MuZ%Ul9{%UV_F7c&7?N0TG;=$(z*2V? zwvsfOCdlD&Ht~(5kzdZ&!5@b|v-iUI;_-fy;v?a0O-OTlWNc@*m@Wtp4a9P0J_b@b zBV+yGPHXg+;iv8Osafi9Sa`2f@Xozw%1M2twEAr1XDaV$dAW9f2{ByCRN;c-r;~_; z)Avo&PAb}--SI2J{{XXJ#o45^)F#rV)GY^@C8IUYpQ+loQVfqAs}<6@T;t?L1OhXP z{9OI0{x5tF(e-B1wZwu;{W-qOjdNtT@jOu|SjxgYsUv)S=ZRUE!1=K2sB^1e}5is==wYu9+??bLQ`e`bWhtiV)dcd@Aso-#Csa zklMiLSUJYwy`sPa_*m}kf@)uezp-cS74bVzpGwvs((I>z6u3ik8s1wHIWk0_yi9j2 zrFb}FUZ3FK+kfKcf;1oOzYAH|+*-WB%+r}5lc+Gt=`hAqs0s>#Fn3p_=)blPjs7$1 zHvSa&vjke3i6b`tL`83`B~XjxRA5twZxD)G;%r&1BnhZNn}kKiBpCO?dB zwRF=wF#1=D!lOc`bVsJuviWHN{#eS>&F8!50883#R_*e0p;QigD z>Rb8u3#n;Vw-8DL1;iItQe9e?LUI9k+Zk=3FC5pwZ;UP~@)Y#1oxUdj0Krss?GIHx8PlM7?4ys&lxcGlXvj-07Tu$U zNVCZ*@%IDf#!nSVR;w!({u673{h#&iQ^Hn3f%3+q9;IkxUJPw*arT(v zjE<%WVsLrqjDB?dYxqy9%i`P1-;R-Lz9qAQL>BF*T}0__tU;8$uw;;cpacrcz+~qK z2j-dlf8)Q3Ul8Tje`cQzYQ7h~xl;3s8>F#fkA-;+ygOssK;8S;_~hq|hI~)(@9hC( zXNSj{-lwYDvxL91vziojK*Mp6#-&-+hBJ(GB;vZT_&Fzb%&JvKN#fz~8YqRPfva96 z+CpVmm1K(QA~=k?ItIb-f>yqMfA~qYaWoopOC)6&H22o~RhSY0-UwzNPMtHwcM)rU z3v4uq{5fSFqol9_(d-#0CQerHnpnjv&E$N4(9&u!E&*kyP-vH zWXPB0;jl><$j5s0(y`AZwnjarjG{!p@eSmqF}$P*iym-tj&jU#{{VdY6IuzS_-<7a zjXubR8UFxT7k~R~DYu>v@eDp;@eTc?QSMXaT1L+-n|*RZs2xZLlhckWfztGq`C97u zk;Y7MXYtAt`I;_6SKR;A{Ec6Po+GxlZF^IKD@-=f_N~37E0eX*h5Q>F`&HinXkHD| zwTFY^m-da8nsX$6d$O<1$qun?RL@+M>Con}yhZUZT)sCDYq#xittpT)Nr=j0KYdkq zo;fF+_QiQ*+RuoV=RO->^mkGvfT%%^d1BiS0QDcQf1`wDC8mKbH$LF-map)p+Ua!9 z2tU~Er(#T!hKeFfa0JtS^s()Y7hM{?jflTFltyCj^D_Oo!y zlg7p%5s`(j&RYv#6W!^n9)Waiwq{77Y{(Ib1ngiy!=9Y+*NWEA{1vTu%UFFV$*BFJ z*dyG>3Gx{tm9S+yPIeQ<ON2QRnnraOjE#$ z!LiN?C?~E3cK#Ch(eV$(R#w`+kEz)kdG`_)Cb?BrJ6H#l?IS%`1d8!14}&_3=;K|~ zd`G7xmCB6Ar$8iOjQoo%dV0`h#8*Wu;5u-81YuZHz4h<&NcKg;Y58 zB~E=*9FW$reXP(k3=v*^FG(D7!Q4TQHS^2Qh?i`})P3_;0JB7j4Ze(Mq|=`V->00Djz zS_v$F;Rs7x`*M+7HQFg)WC}`ecX-$`>tgT~??2KmOC79PTvOyjfsx-qkpo{`x6y0yAJb z#(XbW zGFieG#Fx%A=o|N~rKpwR$iXa#jUfXV4H+bLAXiJHe#Y7|Xz{~muik3b_pt+;WRl?@ zY2Y!-`C~nQcC34f`@SEIo&vcqb)y@JZ{V5HqPlq%p(sY}@w0AM0P?F40QdFsj~V#a z#2z(@IIQ5*w5!!(u-)nZ0BK8Ljmc|>j>~y@#z|6pZW*m3nANqU4wqxVtp5OIZ3jWt zE^n@4)Q6U+6ko=%TRevWLZoG7B8YEW9dkokH3H3w<}267dE$1R%j4Qk`W z*II115qOT;<4?R&S~=n}Svo!rzo3mT_IdFC0El%vtus!B-0HefM!mIbi&&a8 z##GA=)|EE~AM(jgdmgpJc+2)e@Ds>C{M}{<+<%9yRjDprD-Omohe57&msV6;gU5Z@F=?eBe)4_kVPm8|K zZQ$=2>Cs+jip*t8#+DgIbNk4qiT=?OsUvH~gJ{MuPakjmJpHHq8{umygLsot)ZhrR zrje+uaY`7SijnF!D;mzBMi+Q8o~-9J_`!Pj{{Y$&N~qdg?3+Y!mb;7+0o{O6RB?=C zs6Fe5nmZ_NUSAWwp>(omdGnI z;E6@X$Zoe2%E=RfjH$>1zaH&;O4@uDx@U(bOY3Aa->FZg2%!@e_D5*fjdcx3#!HBTI>$C7p@m za_<-?p5u<)Yo-`Ttqi9&{pNLx@dLt|mYTYLg>OBT%yQ&Q3~|n^7-WfFC1wkbHsx{n z)E+AE@5IN`2DNjiAMD$Gq+XU&0;_${LI)r(OupW|O>&l6=ZFoPLl&Icnug*_br2@r z{$el#pz5dBAm*xRlHOn58>?2aL_(>ULtHwM&}DX})f{Ih^sb0fOJg2w2=BD%Qr;Lf zc@t2KpOr*xcVmos%JIS2Ruh0||u|i3}ASfNUuRzuH z9}3%DHKwm@>xE_WV`(1M+yRL^xb~J`QM>95RMNGC&>9Eut(-S9+`00S&IHVbNe38T zFl{{Ya1S}FwYMhSPV3?S0EKlw63%qktgaxIM+_QtTV((bn{MV%!F@nD75WR{FZd_+ zjpCVL@dt^n2Adge%+_wwxndQRx){P>XK@+H=nhSMHQ^r^ccUvO$s>_{QPYq!j92P^?A_sS9LJ&DYu^tpCe^HMrD<)n+s#3c2~P~R_qPc= zi~^JLzHmrT;YE0u{8ZzrJFt}Q^xXG-6ZRy~JRhkg{;T0j?-lqZE%7z~0Eh2dTi7Jd z((;ksia6CflI&`5ES zoD+)u3Gtu6&jb8T(X{)Yh>xgg+OCwx>{``@^!9g$&Gs?r(7|hd_S;t`cxIjpd8JI9&k+Ha0b||4;Jh*WPk3X;hgH+Q9}8_Z z<4+qd^ts~l^%zyYP}A5-V=4<|G4h-N$vHT#U#v@Lv!PSw_pNk~D8K!Z{s(xs!;xxQ zO_N;3b8cQeL+u6UXr{vlhUu8d%FBWtJ~M(iuaPxx+0Rh;8LR1bQ+WE*U+}U@#M=0+ zRWGfB@w&%Rg%VfVaq?Ym-JTc`&!m3Sn(vDB-7Y22^{c-P-q@zdE_BIRZCW6?*>k5A zYgZiOI7eJ?!=-)|{BiOB0LMwK<gC~B5gr^8lSkM^80jncz- z0@|+J?E)OT1_wPEbKF#(9kuv{E%>toQoNZFRym4PqFYLdg_+UXNh!u4^X|}W9NRM!0cKJ+b>GKdVl0aW} z2a4_VUkrRF@OO(6=Icka)>_%Im$w$-RHxmInV72#(r00>r^^T()2Wrgm1xg?30byQC--H8}D!eF1gd;0$XfjklMOU5xs-XMvEsy@eQ z*ut%jK?=ieN#krq*WW!pxjJWr{8a-d!rO=}FGDjuwxI!amFxt zIBz~QMXFAgFwJ9aX66RV2raTtPcm?-sKAUW=dW{$;&t0iEj>+ax3LwbjiYD_X%4@l zBPE$o?7;b8f?IHT93A-_nDPMZPHR&C0O9`tgnUwD)MmQV^j&K4!ojCp89{UN6lHhv z;hI(hJ1`DAlS6ptO4qFJ{5RuCAk(hZOi~*)a|*JsRs=cW)ZmbZk_b4apN4)JyV1Vc z;6ZKTDPud^`$}hoJ~Q(yHu+KQ<`(tOYUkB&p`(2f<(?+9_=oW4Thp$!YgdV)$$+O3 z?k6fdrBE4&JF&6HLC6)>f5KaL?F-_s z#4Q(5ZwW(d_M;GMW2UTGWDV2HD6&kwM%|gn?^k?Br+7zFvUvPTkE7}jgo$pPvEDKe z??BR`0!Vd{0{1-hxnE9Y_ZdQ64t~$!Hji{0SMc@pm)=7-iq#lI!*4j6R`ZAqdX8kjv((qA(nxiI@!R1Fi}9f34_N7TU$M z$5@l^ytYy#5U9up5`iSFLGFQjVS1XcY4I9cw2JS=mNqx0%1L1IBw3bdu%Ud(%VtiEu^uyS2DDl<+;-tLh#7582ppnf`a59iS)=~R{sD~ERqFRbHhBa#Gy#X zPEOvWdT(2JH$&F6EkjE1%%=O!juQfxAt4IHoW~Lk?Tm!U%ADl3a4t&ERbw?Sr;Nei z9~O90;yr#_nKX-gLCwve{p8EIMhO@sW2qIQcY5kyLZzL)k#^Bb@Y_{eMl!Km zA_cYH?YM2S49g}#?pb)?j-sumsp1a|*m=`h&t)n38huV?NSZ}B$PVYtB=p=E`toZ~ zXfxXA7nj}|yMjv@A(Y$4q$E*1YS>}tmJcIs{w<|)r+^K4rN*b@TOHT8ixTPJ2UDnR zjwJ`2sFA$V2gZ5ixub-Ph11vNfnh-3g-X;nfuUBDOH_}3$Iec^pl;A(BCCXambGTbab z>4>VsYk?qDRV79jKQ2hka2Njo7d4G`-6Xk%^~+eofg7lFUB5BJ?5ne$NZfOb0p7Wf z5Y6!5=yn>8kFMWGs$LtEzS5E?p5EOfd_wWw$OLZEjIj!iP>s0+Wa~~XX>I!#IIfKD zE_CtXsYbEkyAyqIK=N-Txp{V+{HM#e^5W<}<)M&zW4|AH@oG|&>Du5~?gYopg3T3f z2_rv#AKkGz$IgFB@*Nk)R~{eK9sE5kmaQI7n;9{O{JXdSd*w*Di!8j3FPZ3vDCw>R;C z%F+@$;rjN?Ue@e3;#))@nqqMfjW&mN4&T@uHcFfPvkc`Woo$G|Qg8Rm|Epe)~ zy~mV{#@^DyyO)u$58mSguWIUbzlZvlfWlnGXu2H!NVwf{tGVNqcPpEgXEF7w!sV7- zL-&t@dhaBpWRb{G{OaY#uXCz4`LwBIg^-2&JYHlhBO6?nL$!U&?dKV_fR>wx*Ok}KWe{j5J{p8)7l_!nOA0I1?!(<{CQP7;X_gWPvP2imAayLmo<}sL zE1KCO=-ns78gIlaNiB`z!3@r^x+dt~AfCGbY_9`p1aV&N;BVOlptxKA01s;57`C^S zu3KgYk8omPnO!6WvT*GisOKx5pBd=i5oGXHyjC{1H@+9Z1#O+)J?y}V9xa>_{Ez|M zaqC`-;fOqYuEGa{tTmZbNUpM7I7>`8JDEk>i9q9xpDuHrwS`J@-7`qbvGk|JuMlZ} z3uL;{^@wg3z&Qr;#?C1$*pZSNwzndy~UI7_IcPd#P!uC92>``fO`*c9YeV10g-KsKdFh0lv|^9Lsfc zrAKZqKy()W!H8+s_Yc-vOtJ^G}zO$FSs&E)Qsy87V?NYpPhR1nS z?WK*QC3_xit9(rG7mBqzYwwHku8XoA*7gxaEzRt)ecjSaDl7nJBLFrB9kX4l_=omI z()4?&^gCORHe@nKCO+2j$-!ohJAl!r95yxs*ox+UHvZ6hpT!+2+rx?A+l7KO+Shu$ z%q@TY(TWg`&>)tK!--)&H2ZwBQ%|;+I9f*akrI|{fnc$X57D6{3s z&@GHUGSjRqG_+u0w7R#9iDLr{3dqkXZgN^I5PyhQlSv-8r$_cr4Z zp9T1DTD^zBHZsm7*pXd9bv&sWs-50QxCY^4%auR8Ob~XA0iTw2JyTuRbup@1-o-Oq zH<*bDWq5;az#DSy(BLzYN&G9@Aov}v+=(AuNo;3<48|27!U!LG?&m+k82(kJeieAL zPSeJ#;Qa1So2;z2F@eZ(xIrFI1LZ!Wo+-kWHSq2Z4V+`^_WmNYw~xc#6V%tm zi!bhVD;aKV{%!`(DNycq1oUNLk^t>rTKH@BUj4N1yePVdjW6!B1$KCn)_a>dF62Ez zh4Um~xp~WuI&}mHJ_dMiO8AxG%~DSq8*8|%Wl1D!tCx=I*UVGpsc@*^u>>GRB;fVO z7IfQx+7m|BZag{g@N180Z@^7VeWu-GY;dy77FfXwd1H)n2NX@T?GmnzQ^Vg0yfNSn zJ~8n^$6dGAr41bKbW-BTR_b>kd2>sQ?Z` zTPt~f7Q0&r0K*al6Sh8W`v6%1?eg_Ks(tqFVc76x@c#hBZC_7Yt6OO8QIy7h*$c-q zV?Q$N-mF3F00;Y~veUvoI|u-uQ^r4oO_Oi@2a(66eTU#{yYGk^RQisMc(y)Vd%|ie zpgIBkGUhXaMsPu5Nw0UbgTQ}fkL^Aqia9ne^;qQ~HaP<)I0m`hSA9xQx&PMrOW|ki z2k~3P+8BpQ(luLEFZ=K;#uavgo4)7{PP4A~TTt-_ihM0>klfh#aiJF) zy_WD<)Xg;NZXhZ{MjLSo4$a_!S^oeJ{x8q*-$R?m{{Rf{q)TOJ)#9Cu62Nhn^HH`* z47ktA#~2@XwP*dQIY)EbSS1I{wmxw9aqvU-T-Icp!rDHH{uc2DpA;85hM#swxwndS zA~ljSvXWa2M48xjkO(6cx#Rx;g&(psT5M)LGwiWk+Di-H!8eg4GC2y(43ZG)d#kR@ z;~g$U-A{ zqln?~jLNK}{{R;|xC1;^UcdV;_#eVLUFNay_8VI*YVs)JmsYg4f#ul=$c@tGUoLRi z0D|buhvbkj7111Oio4Y1od`YmKQ^v^VlNo{P1Q}ujkRk_d%Z>hEVmHF6PJ-l$ht__ zeCxZ3HiB0;uS@uK`vz#90MVBB<0;ZKjS|J!N2z_HJ9}va!{ym4C}Qs8Yez64ayFjz z^{<9LHa~*&*S_(Vm#1m(XF9@XR|#o5Oah^J?tF`jdD%t`Ga{92fXhNC@IUBFoXTcI_%cP5Y8xRy- zEzF3;og4<-NNCa3RD!G==RNa^@r^_F8`E^j-s9oUrD1!jiA-x0@=G0)kU!;{KzBy{ zzVT4m+zA}$x9tPq>9h&7?}GYWmW8WLC!cGk>MbeKmANwuUnrHfX#oL{e(W4D4b6Ei z_w8Bn16x@%tGznwLcNG2S5}urV1gp~DE_hk|dKbe9K$*#{hLubzG{d}*@r7Kbmx{ajn?dLS7~2aY*nSy72H zd6E3_zW)H3O5l_8SIXC^CGD}l(r>PA;*v!x1laQ;RRC>CpCOe`@Ri&%jMrrxFE_?= zx|VlH`qx?b&*KZ5DX%;msa;)15k*@&n}wG8ILF95$!8@b1AwF{Za8Csk1x}{Jx_C~ zuAwydLs+#Sg}9P)a;%N=l?p=S1Kn^`V;t9t=(@#@p>&5?Y2^L(Bul9nVMqHXX3{#2 z*yq~4`#`_YJVS1DeSglku;E4gC1983APh?QMg?*Tu{p=Rb^92@rNtVJyEC}5)qF|u zjn9XCM!Jj__p&UJ-D#1mGPI?c)5-aDKOfTugl&`Y>9e=?;aA5+$x8NH{S+CBpfRS`K!(}AKRDU zcDZ4H;cpCH$>2?ER=!*Lb$<=oK+iqIU@x10XK@^hzqCe`qv>9FJw2Q>L zg@eJUp}LMuJ{bJr^8BSIe>5wB!yZB7g(ALQ{{V-Q>DM=M9WPk7Z7{!)cM8Mi!7AJX zlF;5dd4qS}Av$y18u`l8isYQ5oydpW0Kv8vg)>^xqOe6^@;6_KTP|%F)DL=H1#UMZ78#%}~qtkz8%S zCpGyctp3#(b6AT%5ovmsovGU^iS6ZrHrpUSD=S09@sWauA-e7v;C#p8ZAQ=HcBwV( zjkVU9dvz+4cWb?7l0*uD-Yh(FHqZwAq@j49|;DX79v0ie2#k$`@tCd%M^!x*oLzAgUNI_9ei>7E*ywY@pxWu1ll zTk2kD0|rf!0*`-e~Z_W>pDcNf3qYoP38iV+RifgWSizmI`hCdINU3`mL3q+Ov-J& z61qHR#6K52FRa{8CW9QT_%X>1owKydznsX*tW=Tx_6MG8&gIedeRo|;ohL(|NAt)n zJdX<(JOyWjJLG2>-L#Khxc!T^{{Vt)-fQ|40_y(rPP~Vdnk18PBp;qkszTYw`F>z) zss8|mJd5D(fmc)bmR9oKD}W=p(rl)=b%20lxwy1UNRFo=gl@xb2wY?irJ2)?h{0iG z&%}LK;r{@OuCC19I=Q&Dv_Q)p&9qVM=bf)7?;q-RsOf`S_Pz|!{4?RW^)r7R!pyFQ zIYZ7Ee+~q=gezqI&V8%)-@;$BN5O9vh`dRn>u;=SOdoa4hm_FZWEGRj%z*C%5L%#dgovAe1{w~qSCMjS7j6u6ymKe-~~uUb*f5|*WXO`nbydOowO zGPj0Sd65szZ8Tn70OO#M7ijE8ayxNXHFVc4^)%6S3GHlWk+&#q(nz^otU*8&6OPTA z{WjJWz82_`OYvUD$483RUoJbNJTGr>xZYX@alLwxheAOFbiuEXe0%Y8!WV1yH?&<| zCMup>cEUpHa@&N`ZQ8gwI|t#K>iW+&!gETOE3-TcQ}DzZEa5eXpH{UPTX`lwXOoY= zu~RaE$QgDy#s^+G_Ryu%uBE#0bP-2xjg|7F8^JvZGKO*wAo|y1b>S#Ak1prKRuamN zPTOng-Gpd9ZGo}@1QJHuYvX)7*6LixUVOHY}vkc4!7{Yez>DI^KzYdX0=}Y2i%{Yp4RqOTh$_A>GuOGa7Fh>%WgqwduEC z8KcrKg~pk09+!Cwmyob%ypTp0FDV?8-FO3x*9YP+jlL%Fn~8N>12m({kL~cGfUw5V z=9#2epRZhV*FBAUSv9I5Ld04n)~#!G4c@YUv892-J3#9GVN;OOs*@P+kf)xR75YEm z58A@l;h6r}mq@wQE#hz_yN2On^1_UbwzkNQhjtGxK+oR6IsCiT?{)99M3;J`Qd)5S zZL~t-LJ1=TIY`2e%nJiu-@{*rpA-Hdc!Jwk@OPi6JCciYJh6zEXz1}qv8e!hmM7QJ zF*sPwS-TxnDk!_N`Xu-({{RI`v+>pYcwb0@OZSu*Rs#W#2>4P&>gHqDmW`E{f*9A^ z-wk|Yd!hLc;13-g78zwmz3}gd?-x`+jOTk=-^se}7CljJH^-5;03E~m^6-D`Z}CIK zI$T0+a>K*kQndG8T#auEI*hQC2T?H3Jdut-Vb;Fp_)GA!N$}T%AinXxinP667j-6o z4tbJWKEQD8G4jz_h#&@F5Ad_*zEc&4rk#_q>(j4j-*fg);vel%@NeOcu=-EL=x;P( zD_HsrCu7S~E7iI@|RFs%Oo!VU?n-EZRWhCElQ4SPFe22G2is6u8E+9!TIegdUAI9Gcd_&Uw zO!m5@qr>*{1re&lGn8Cy!m_fW&Z;x8^PjDD-X!>0rd;d35!BAVzvQ?^o=Yh3 z31y09i*(0ke&e>>iiYCV=B)gUAvDQ6L*btjcooNstS+si($EmI+QS>aE3uuV>8Q=-fFT2#`g`oEPDvwRMLJVv>))av@xbjDxJE7 zjIizbaBw8sCPoP!SD3%8N}n?KmHfn;PUVeaPlUcK)*_DIP;DaUt9k1P9z>4kE4JZ> zlpN%eNCmPw)tyIAkHHt;+ZLK;-z05@=703biiH3+{H#%iVtPg~$2EbZd?$lK)9s)r$rc5xMz6m!4LjAL+Y z1uSvz*BZVv@Wzwkd&^1m8~8NY{E-0rKm@-G_VcpEDg&0o`Fn%8&m@z()Mm4E&ll<5 zKLbopwpmEQOmf*PJ-l&%Hg1gx2pd0nf}RdajAo{<;2#uQ=T?Q`x3aYJJj;u9lI}~b z#C`Z;a-(3#19G0El0c}mZ*`*uE>&wBy{@-?@aE}Yn#TE0nvn02qO`Y|FI*KfGYkUT zTL&JMZN4acGu3ar?-yy-%W-%J7q*ybqfj=eC78KjpkTyvj&WLACypX*25k*&?lr`f zq`|&;(lu2iGQyFTm2;nz%Pv922N*OB;3QJPP)2xVn?yw5FLV3?mf zV-vS>;1=U)H1@la)RzX%hfMJ1wW3%)r{I~U*DW;wN-f^q@tour7X~>QR1!A~IL;15 zaS-YrB+|7Atm413w-E$av$utuxEyaUF2q(n*>X7hz3I357sW3R&o#D_Yb>)$5ofwU zCRQrj)SJ125Jw;?va!w?NFu5?#N9zHXVJW8abbVt@7#+le`yMrmT9X=*3o&ZntdzO`uZNeevSK4yVp1Qu|LKwq2h0vw*W;L>=LLh+85rls6h zCjJ2>?{aL8kR?rq<4ZVDO5kOj zi8&|upC658N#g$i5%`KPJ6W(zLfuC<7e!&8r#NN96Oqs%$mccV1IFXSw*F)nQCY!; zj%emzHCU-RIC4~Tzyk-;xofW-T59pA(%y7wp8{_!P{rf1YlsRmW$vQr@Wn+I`mu?9KeefZ63T1%|l*%h>o zJzm~MJHFd5+8vb#AO{Md4?&#$0IT*5J?>*EE9hx>i^UoriqMNVUFQ-Xp3W0AO^yNF zKXuf4;1kxbMewWQMD`HsdL5VA5TJNui!$6WMj+w@&jEo4t9mIIHO6WlFpyp?gf}S_ z)WiX8hAF`8v8mhnVz(gikBj^#t$)HrYYm2#bmf(<&zB*P$f`1_9C^qJFh<-CneC@5 zTMGBw`)lC0>^1P`;y`UjT=<&0iH0>>`$#Rm#r8&*sG2*ylOaaPV#(iwUgP6W_$Oz> z--ocKvEbhk+fOCA{{W9&S~%{kB~@T|rgiqT}0!8e9MPyo<{fGo(i6v4!G-G6n-_)wCg{#eUa{N7?2=`%V!-F z`G-4HdJ<_V-PqC#$mldrv-q|d3u!j8$GhkL%4d$@izxv_+r|btB~v|edFQOu%!zKF z@P^pKKHLXgNO{t3z&VOyRy6lMUP;L$X1u~#d_m$!Jfo&t=@20VvdGe0NWFmwqBJ1m zC!FzF*FO(+ElH!k){^5~iZi*OmVMFyq;R&t6(bqORr8FIRn*O^8>7c>s0hsZJD~)k zMQEYX+#Ya1$0yg1Ox5iN!a7HcZ%NhK9}n80Q2Sk`yBv~yw1fo&jGXQRJYd#r+P}m7 z7BZ2^C6&7F2qcfnah?K|QH{qO3=n;5jfQpArJ1}H4xw!Wlk*9Dm7A}~V?R3%`03uW zo!BR1?#~A4{{RZ~&kJdhX{9_*tE7@eacitc`T*FYfg35y5DeOMQDU?*a&5 zRg7(JAIx=DnTFi(n8j{r+NXgcNbGb^CeZE6!dSBxQJD_s1(*}jv)8{B&nRDWZ)%;D zk96_RiaZVDJLt6!hdv%28raxbV@T{^kg+9;MprVBI8anq3#jNi{E}gHIWHk|%yw#) zhiLx%fRYJ43KhVv3jY96@V1w#NqM7cHw%3LAs0>Absk9T<}nI#J=^F=tplxi($7vy z_yI^pGcdPjl;@^nW3Xen#ytg6M(rKHhJ0V(%@*}ziXS3VmN8xzGDtu;23IUg4!v@5 zUc>O)a%NdM*~kR47=VaVGvv5bo+u!ANFHq z-Z~C}Ro39O7TtK@{K7X;?I=*wu1(HPv zfwS(*Ynx9K_>#u=#0?lX7A&#pm-edUZS$a&R#?LnAQHPv0s%NV_zU5m z#=nT34Yyrd-pbxsHi;Whnf$dt8?STk@{U)rue6u;r|=)fKML7Dk91ul#8w7WWbPXvc5NxJ>%yz7@OpspEYcPM1cx@XFsLh|0-$==M9G+ivMbKZk+SkzY*w zZ1~^tXUDqC-Xiekt)|=A!UXfeLYtz)i2%B~BYsA9q<&alyyCuX(mpn7a$1cY^f~oG zb0FK~1(_#opOmrVAQD^W_4ER|X!Bd%B3f#WW5FK;ue?!kb-hY!Z9Y4kNf^?aOLn(r z#x@Aum@2k-Eyp3<^z+h^oMUt7R+>4k8|x`9JVkRShaOOjs|(3- zA;(O@oCE6Hl#ccG7KyI@$QC{ry3lpaE=>U~@Uug#381}Wu4njNZFOy<#_x&dXDN;TvK3MXOq@Fpyag!5SZ~n!s?%QhV+WB2ji*84 zD+^_5CxUClveXrU3S3#rv6&QO@2sv)P6v9(@#pPX@T%iS9t_faEZ=3fBYaoL%m^6- z806#y9HXh{*Vn-{e+?V`5ufWdbR*E_5D zw#!e33maJ0D=Tt9mrzEYWf;H#+a&?+s4-OvN$9mFn%@!F%kewK8q}8>KZRcNO3{wx zg}mpU+(ZY;QEpI88uj^Aii~uvPY8I&!@dw`b-fA3qp3?EfAj)2hmZKot_$ZpovFw0 ziuuo5@h69TO{qtx>Spup3gRf*4b#Kzn6tbo$R}=bjGEoi{uTI@@6DTPQO|0aJx2FV zGxBUQ1IO@?Pp%0S)Tl)-dDvdlvGirX?K|-=#2Wtqo#Sbt(RAY^R@ayJ2q5w!nTn*6 z-#+BMIPa1(jD3B7M)7q15P~ls_!m{O(n}`Ax4O5t znk*7H3c*|hfJsLVMn>Q>QBt2Ox4Cdu&%%fpuIu3xT+1;ak;J0jO)D>e%w z1X2*3{McL*w+9u(cyq&F3H7UWwD9fs{3IIW4kbx00$cfhS{9QrT&xZEQqx{Nz^%)0|#wj|I=#XEz|JVM3{vGK)9=VP`6=|L$hVomBa^PAB zV}*ej46n9MFhJ)Z5z?XYzkoakf2Znc@Y+o`;rEL58>p|Yb&s+~cLn>#0xI6ANZrFA z2+p`T{w#I*&&3yatMJ~(L;aMrJ1-6DI(tto{*`x`6H%6Vcavz(F57krTW$gL`9S9v z!k-c~F{WF|K3i><%)l0b)7LcyV0*O!|i|S9rUhP54ve zkBfW(;GeSo(hZ_`V^J3=Yh!T*+DB)0?176wr{+9o44(d(KHd#2*M$BlczOIk4cld+j z>n%lYGz;$w>Y8MWX?tNR32o$-Kp3VE5VC%!uEOO_;5cXpnK;wO)^{{V`*?xCpNzJVl54ZM&`b2K|Hl~^1si?PI! zFgfyMR*ASc$Qby?Scg>jGowwa-q_!unS_@2F7v~rLf|oTGX<7QAVx@e7~Aj^N`!%5 zzgmaD;rl=QKGf#dH9cG6rj4ZEEEZ312yIJG)8kbPlLwb^*xUHBnzIIl1t5WWuh z&r1>bH&M2l)<~Hye$-kiBaMb6f>^s8aLMx%oM6{wD(O88Cr--Nx<3{*T~EiluA8$> zw7t^q5XLS8A}C)t9fV6Fvogqj?2)mI_TA1r&s;wclSq@p{x;Q)wWhR;l3Go0CNN`y z6sgKE#{isXur>DwjivZK;7wBR;n#(tzRwejwlAUc1IwF{_4yu~3#az1`{ zhEcRA$-vGsPJJuOtb8wTt3sY4n^l70)aMei#^`%(;ZL_5gO9^{NxUfx@XO*IMs>MB zOjAbc#d4qpWC%;+)V@V}aD|nO8oATzpA_`V>u1xV)7}V*l!oTlB5Xm`rTCiaYe@0?$B3Zgfx0$Q20D_X-`2Gx@ofGcwhM1-1?*sNONkEFRcvkh z#Y&QXq#sK4e+=m#9d+=R-WIa7x3`@J*U9$@YWObHvm`n5cD^{=LW3iz|(UW0dasC+oque6JMn>UR3X6%^l zZsje$Giz~kg9%-%869zg4&If=Qya*wYn?RdyI$wTnkVdU`$zbO*vp~#dg9jQf-IKH zFj&h0!HKZ!!1@NSp#BQ}$Di<&>dW9iAKB`fCa-I3X1cSygoc#^1v0@ZowKU5D&-#_ zpP6zoUtIWa{t6%AJ6L4U{AHltXnMRmnCY`fl7?ImB#F`EhEb7`v~}y#72Rn+w}<>J z1-bBq{udgzhwq`56AzgswbzvEx!OCMLXQ-He{_W$bCt>S=a^P{Bz3}__alS;3HRWg zZpOz#)a+)JS$3!){LRCS>Su6s-~C$haF6Q_TI_pR#JGmVW5IBWP?Ovso)xgMOjDz%rhmAZP z;+;t~9dATYd1-5L6~3$GDAb^hiiQzP@w`EPQMp`|Vxx{L$D)_xpMbn&ZQ+|;e$(PN zhc2cqCB@;6*I3hFkd`vt#-3CY!hmfIimX5f4f7i0`qbW%HgL7vPXN&T3Gu(fI?G!4 z`deLNL5drTfvMb?AiqUcLvI|=nAgk_N56cE5tN0(NFRFE@XoWOd`KFWx$xy+ySufN z#L!z44EB>S3aN1ns4E72kiBGe){IG;L96L2gR4Dt97Zx2)Vc27y7v_t>Kb5q?0YT zelys&0=x@G_(SpewU78jel%%5ak+_>EmK2{j*&jqCBxg~wYxI&wqGzD;bbjesMo$F z@s5XgJ?Du0FK<4E@)m~cOY<%#SpIB0wlX5f;xo&b8fHRSZ+z1gX1PNAvsRuh{2b))shtDWv#tmY z6cIiKjz7K2H9tpgXRFSMLPW?7-oHU|1TXBJku~5-S3gvPLTvwiaXSwj7 zfP6<4zry-}C5vN#r{>2_EAn^Z$Lxvm zQ^tNQ`)9-b8X4fh<^pY^Mi?6v7z8ALc@9o_73|T*!LC^&f>>vJz~H_c>0S@9yMw~^ zzCE+Iwv#O$mvN;UD-X*TruYYV-JW#@BIaHxYmy2L;%Ufn$w|35XdxgWz??13zZ` zIQ@r)gQH6ajdjU2Eh(GqiRUtJ4uG?&fU-^HHYL`+3!p{tHw2H3eC;Cc;*!Ua=V1xp5fNRDq{u=8(DZ0DU{55*ohuFw3Y>H;Q zWhan6QoBd+yRxXS({C7l&U$x+yf=OS00>l#bzt&&q8m$xitBHyu0u+T(C)!KyNdX8 z#6BO_ycyyfJx<_DqeQH}UEJ3~2_qcf4e#b0a5#*E>62Pg#WtTKGp$Lcu6ZS{k*9b@ z>|1KORPx=cDx%&>skdtoHX;__?#m3Q=qvO);Md0Qg|>58=zkNhPJ;&TF5Z16$QZa@ z2Bt`PcJY;xF&e2DVsZ)ccJbDM;=LB*P56!*Q(1?U*Y*~OuDHksLhgn_4^6TB$G2+i zG~a>~$7Y(Z!^?B2+s1*~Thx|mB(RtyvyvHMTp!&tTe8lFZwG3|a7N>jE3dTir-yV~RFlX0?Ut6&hjoT| zJo~5v9sdBjZe;6|v_Z6k_W~qQaPh5J}oh22cXIvwFj5xj} zcoRo5Tg`5=+v+i)`+OVh)v(2}C8{`3rX~Z-ibUQ){m8}+Cbz7#-W-)tQ;KcsjoPPKV-xpiGF(hzTewp_+{D#;^*6XcRpr+l2(k{^INb-tes#5#nU zys;NUZ>1PoX#BinH<=)29XA+blNtM_x*<{zcV=ekw61P=r@>m+gviuwty1C?m-lRk z?J^W6B1AGKz~dM=PzxRoDV`korQ_i1Wp8fYAk(B9nm3Df2d}d!{C3o>^ynn<#)2ldmJt`)tz&TP8zT`- zIU$wQ{HO~1F$Xofu6#7Q@ur&(h`tSrm>egR_S%_}3GO8GH|{NV!PLZzLaH~GRCHs_ zduM_?C*Vy=SmW?@^5{BQY zB+q|xoa?@!b`UEw&jbU;K`!dPzQg^9y%jx-46d#(mM z4^O`(@TbE&gp0%ahM=0vbLKR*u}8U9$5ZCW3{A-x%eT{mUSFJaOC_DN zO7D&a9Y8%9Mr*I}SAo7K9da)YTv=VDD|XVuX#@9`LBzq?JIBykr-@$`a zOAm;)nu6XdsJOYbwrxBTM&ahdvI#%{f>dGo2N*ayP%nuff=?0rCVeAPQJCa9bQ>=M zj>pZB+_)VB74-z?70dXy;eUoByq90_y}iz_sVOQgZ*QK`ON0k*=am=Cn4)}&avNyn zjs`w?)IKtJgTdOL_G;Ma{vBBfXSBA8NYS>EyKZMfC6^>;FEIxIGM+_JN=x^DiAol_ zlRK?5$38yN@9s5-XPiTD9oPFwfQCs-gFaxDi{%Ctngu5~&t55b&%|C8@rCurl@VE{ zfq&98!6QbpWSLS}=ZPedR1AlWIVX^M*Om{9ntzWK;PFnEajQveyCJ{RAS;5b2G%RK z6<^{cGL8mEHOgxCemB0jO=H4Zy_A+#aG?`NZcW<-W4HG)5tqkVT;;kku9Zt&bbdt2 z%Tu?#w!VtiIdnFI#I_30iOlVAkT3C*0{KEV2PYm`IrpwM-{V}e*~y?ys107!5UUK! zv@iq!Ayi*A09;{P13gA^PnS20FD3DeUMH~C@2>S2nnN^qL|TtD@$(hWu?vTr9HNRB&FTOwk_I89AvYQ8AxuYCpqojx+6-<-bNK2-iNj8vv_|}(2k?v z*lx6o(4~!@u2rpHE%Q1!!y)bl(d&S7UOA|x*M@Z{<<##U0u)D^2>htoi92yLh0Xyz zkbP^2w!XT12(=vwDYZCmL+p|>CW#&JflY5sJvg0Nlreq^HqJjnjXIU_t~y%7G+ zK0dR)H=3W7;){4fv!oL}z3UI-NS09BPEI}WcJ&p`_+Q3a=8>V#{{RZQ#@5zHm3Nzz zd4VKlnN$=Q7{(orGr%>)YF`+BCml)K#rmD?y`wRYbP=eC7XA>3yAbEO2d*>HqVniC z!5)X=UkvDe3)H5!yVY&9>&ucQlF_6FbPPUU-6zbW8NraQeJh2yu=s4ZA+9Z^zPeUC z#uhYUdNW3n0pA$k$Q<)o7uui1ZyxEwD~o%LQpzmiTkt+#_M;m-+!ND0VzBgG7eVmb zOBTNO@l1`yk`1tyP%yrHqjY?AJ2Cjyu&DZzQWi9|%{RuH45>AqnQt|Ne8M|uTHkUL zpO`Ba8?bYZd*-?S0PL?3Urim}ho(ts=H-NMAy|+xBm&Z$kiGMkHQdedZtBVZ0Mc}5 z^t3%LVPh&|JTR5RIB#4a^{h#LA=-G+ypv|y<3fWE3J}o{NEx`uQb^<;htrCvxjVZV z7WAKl_d2{%Y4_e9Y6u(UnmI(*34xrt z$~eIO>ibrG&X?jXNmyw2Lf3xWLv}^Z6}prvGERDoj1gUqg|29tCa978K}+*kIl!ueTi|cd>YhrAMEcNYACT^$gLYiH<2L? zpWX(Em6)jG9D;h{v9A0tFNeInPvOsr?X^87d@aO!sTOl=;{rLt6_Ysmdy>ACr8M1* zJ(EoM-&XMMsc5_PX)bQKV-ZCrJ{v~*q!#*C?qwwX0@oSi6BJf+H5O*lrt8mE5$DT@MeR&*b zX&xZ(s%iHYG3oHY@3$AQ5jDV2Io#MQ)aRx>J(`bO5ajJ-dCUAW@kPQzu6P30@>3+7 zA~Gsq9z3Rz)JdKJ+!P!io71D$yeX{cjc;Y~67N~Ew=jV3rzCd&05cK_`Rc^T$Nh%q zfr2ZwxAE7CybG&tu-0unLM2V972w$*Rv;)6#&)UW?)5%{im9yr(VisM5`PZ(cSc*g zf<$wytP(bDfR$q7VNP%ZADgFYmgLrjrrxaQ^!+#CXNQr$v2P}|S3YA4(&FJyc?p$4 zQgfZT$R`BXkzQ)P6Vr_M{u|b3BH4+MIB^>8`{dlI2%$MAa6EUdPZIbu#y&B!noUaP z+ROr~QUGpU5xG@QJz6N#9+(e|02BdOx_5!J?+I#9>Q@$jZ6%gFd2f+!88gEO<$T5+ zvG-6%&NZ-; zcA?IG%QGk>k-$uK2h)RF$HqSqvET{3^MM49YEc~sW@dB8CLi0T%NOk;Hh=Ry!dH(cL|$ul~fr~ zB3e3`!Ter^rgX58wC&@*$;#Z&u_hgFNoCZMYpOpM~M@Mz*uq-d#_vTwQ;!is@X+WIUF`FI@D%`kGEETZ&V$ zp{;yG_zbb+5G-8okI|IuQ;xx$4P5}oeoL8gxg7fx(w(y$j7uwy$ z&6L9^x_zN*<|^BYzm}t5Q_kRh!y5AM6r>&!j@w%Be~F^D5I4(ZdmMc12O)N$QaSCl zj~OPn^ncnv;;xFe^XR&4Q%LN)=eM1b7Gxb~mG z)n}K-TE*4wt1^k?x702qS=g!CcNv(p?7uDtnhEMpYUY1v-v)RO;id3T4tR#*=H7VG zQpV;)W+9mG5P3r)H3Ord40r^dw7flYY`dWDFQPCuVF_*Q1B#lP=3=)9K%%_i( z;C0-1&sE~j+2+r{iZ$;6+-b9%W?!_#?{Vf6fE3FiX$N1MDtqFy<=0Z@Hn(TY%cJ;; z^TW3HS`r&z(wFn@Bo?TMN6UrBndWDsA@$;>vC?d;?Iy7C4e8a)s;FWrRaKY}aVzE9 z>c3v5o8rw!UbxbulK%ifhIfWWV|{Nd&R7GL47-^LQNY3iGtaG1(X?+8Et$I2e7z>t zMPjKs{qPP>6mmv5ECKohU5P$h+`>-o0TKx;ru6-4Y(6>Ft5H~zG>WZ zoxQlv9O7y|AMswDr*81vucAc4mQw>0JWH*0q?>6Y{A62M24 z(xj2u#v1{c#-(K`xa6rldChcD=|2)}tuE$~pHI_BD3WOCVL+-$7J$03^s z@~qtx;njhX4MX8>uJYa%X&xlRZ3_c{+sXNu;N?(Y4i6+&Q)#UZg&VWhv_Fm!*(`np z*8DANqeTmT{!4pwkz`=Ich2vbQ#){3fCJvVUsGrJiK@u{8t~o4&|J#g3#ieaQ*&qx_R@||Wgb;EI7CGyK(}MVebMU`K z)Dr8%TA8=Lft6xQV$zVHZNlsvVYwT(jPdlrAZHt_{4&?~kc5<}lLR`D8{i zGIpp^bDZ}ib6i=LuM{#qmo?SnIQhS_ z(pbW_dKqF^)DSw2!20qlx=)LKEck)pt1S!mh0_iFq)Z;mTUmnZdU!$x==_9`w*xuI zVz~!7&3xti7i$;4wEqCX)zqxKN&SQIbHF;1=(p>r#TCYttiq}$x4U~zAIy>m0opBq zNHesq+V(FN>Rt-dwOvENekSof$nh?dwXSU9v1nl3BCGC(YtJlC78@RHNKsBul0zuB zN-kWITBFldoOWl?_x}L2f%q5j?!l(i@BDA0#~bd|ty&n81vx5biDP+{kq8ZlhH!Zq zIO!~Z;H0`P^EDl%f{#6 z+GN8;9jP$vJh3w~w1aMNxL%)mt7VbWhVl}}Nvkw|llX`J3UT9~4^Q^b1K2gIJR5DU z?{7CgGm_tDffWwyWr21aah|pFkBWE8HqfEH ze-2&gu-dsv+GEOqMIyA4qTsUf2Gw2NM(<;u zEB4>P{{XPJ!~5b!v!lnT#;dWUw}~SPPgZ8mTZ4iN9^ZDpj__CPArHcz3hO$L!HqI+ z7HV!_p5I5my7D82H+E^`hU7&vMIsZGiNIM13bKV9Wmh<+m69$v4sC9JA!+{rv7f{r z5v&P$Xtvi=z_Z)jPY6gza9FPIFakzC`w-33AlJG4CjE*0Lt?kuN5nr5Uf9nG61J-d zf?ErS6|uc$m5V~dAT&~<=lB`dCckq&23hzM#-1FO=DEhfJR0M`d;6{o(OLYU3fFac4Ju4E(sSYsg-!zCm14Ab`PASf6Fo*2<$qJ zDw)KsqB>)Rg3+x{$zK8f#hx9vi8Q3pt?q5)+a!@(1XU%J^vPxgfahy-fC$Db-~Jo^ z#C{a;ifUdb_*-pubv23p&*AIaksXJgNO2@mG;Si__Xu3@4pe5o#`tsar{RBswHxR7 zYvN12W5brVOz~(92)uZnF}waH@=GE-QRgv;i@mxHzF(H32dVJi?Mv{x#9GXjJ{AK_ z)qJ5W`bF?aHC0kJq=pE?d9k?!M>J0)ZG*MW4m|pp$5GicqBJ>nS4YO)1JZO)h#n>X z0E8dn28F2TzAj679kdNXCsZLDY4WZEEEg#r1e?%|0=WPjA7}9w>`m~G#ab2ThdfE5 zMR#pE5?vXbnMVOe-KPu30B!(|-RtW;BgeXixYznuinRBTT*2kq-b4~Ri+JT2aUI4< z5@2zL$RrZnSAuG7@oU0*>>6i-b+~oU4N3|y)xKR*OO40Ri6m(a+7rnyB09IsjKhJ? zJY3@WY!aatsy{lkANVHLvGDK4&1>-cUb)nKF$bG0+8ykEY~hL@DPfrhn8*$@7|UlF zIP2uk`#O9s_@k}g_$~ZF{{RS{w`@$eIy|~lY1gR{=vdu?(}T3_c-7Q&CnGib4XOUv z9}E?4JUYH2I`*NY%jMnujELlo&l_C3BPIq55DqxTYsmf}e$w6o@vn=t$UJAES$LFa zmkBM7pqEBlxJV!gZRPUNlmPt7#X%Vj@6`-VOA$ITw!5Doc-Qtc_;KNDG6a`8uA-Z~ ziFHh}$2P)4vdGaEU8j*42e>A_au%LB_*-pjtN0r9>$-iyM%UUM`mNR0nR3T$$){>? z#TmS^L57RZw~B4K5`ubunCkk6#amnXyg<>f#194;joU|}ta@&(V-7$N$zV~I@*)d~ zB}vj`02GQ+a=xK9pWrq2+dK~{2Usqkch1(@Jh?LXKR3#N17D_EAAo#k<0FwOjaV@?AUjdD3h(tt#5xM0$XC zH_G@Ri<>zX6hrcpo$s6;0mgBu&rvHx=%Z+B7JjT__KNss^ZZS_jvp8JXVWreEO`iM(Xz05m;&Vs^{PuUlDvE;JXhJNe#SLo;L8-zjGzxK^q;mmsZfFymxG;E}>4)q_UMP zL5GH^8FI|kxXL`+AJrek{{V*n00KTHYBu`Dp*8N0;wxK)k}YpXxsJ|ATq~bVc zRzk}Af!w?fMR1=H{{U%iSHSwRU#EtfUhoZ!QSH_ATf?YbTU?#qaJkarg=LN`Dq=w_ zrJ1tGfMovwJN_~Fm+@D_+C8n`#cf*eQ3XrKac@792Eoj1K$3QdGM)ib$>SC1zYjlX zpNhU8zO~f6C#^o8Dnqo-8nw)hnJ*lEWW|dk0y0`WgUFvLiUj z6~{U$yFD4L-skC5{w>o!A9!lQ*W$JPm8|n-REJJQ0d1XvBgu*7NSurz3o9!*+JS{w z*1jotV&6qgH6mkWa!tL=7m~_^2gp$y5ahW*$Xu$N0tI|^qKS!prN@fo7f;gaRvSY?;Xb#(a&h;EyHNCf#xat}aj$hE(X z{{R^EK;9L#+FfM09yoT zMhDLo)bA7LCxW4veJdAH_|E#qj|H#6Yb}$lu$AG}8dlVka{PsmksE>m3UJEU;GVTp zPyLnC#9EET{{V*X=XAG-NHmLa5x5(^W982?B9$QE5D6XgU9FFY>@06!@TQkxZD*@T z9@wq*c`n}JW>9gvatIxm50s%QM{;tQUZ*vl?D-c>zxc)R4)*Fl72Mv%cNM`| z_?yFefz|hbt8WNc=Li>O zNtDJ0IU%+jjz&#bi}sAyd<%OM==MvZTt*TT_J)$>_X1Tyf>IVwP+|ZRpL+416HBG| ze(r5P!(F?zn|40OV1*sR9sqd`{i<`hg6+ZR)o#zn4dS^i(^%0Z)fU&~GC40Sau7pE z@y4jFg*jC$9mM${t$TQ!?!DI;Y~_?BU%ny zCS+MkZQYX+F8#U2OQr(;nEhIse&exfZq||Y@_y8R@QTUyp9x-G%XKhjE^boIB*)Ds z$o^v$&p2qoJY%h5+=7sxhu%> z-xYi=@js08!KVC9I%ctc7=Nbe-{{j4F_z9Ylb`MkXSQpP9{~Jeq+C7cgmitdLJ?%0 zAr}|1$iKqIV%&O#MJJ_1DATo`=1O*Y8vZ2FHC=K@^{)-;^7zwBytr;r{?O!%sFzX%@ znm-=3v|;0lOOs-Xjlj$it4Mca1I;n?>)O1;d`0n|mv3{VXwzGGb>NaAHLbi6$15_n z0v|nOmCA9KBya)8YUZa#AM)$*{;p{%jjc~i@YMeR5j-UmY8piUbTS;&ffT7qKkRERRjzTg-6YfxTR9pwDe}Nc3K|ibK>6}_+w4}-S~Z_+C`+e z5=W@%ESK`0GI!4~cLW}Gs&IOam7m}*+C$=P-PP1>X?@}yBKIvUTF~F8tfOvNO>)u4 z5&jZXrZR9d#e9dMd`j^zf;CM-0CEQAdu`GrStEt{(*u$zN`p(^j-C3HrrX^6GJlL8 z1UyZqoij+);qXSG9LTp;V){F+HQbg`EsfH7VY$I6xmUj!&nMy^7wI1jW1i0Q!tx^& z-Dj*`N;NBKpN8M$N995W00FbX=dFH!>c0iNQJ`vf_Zr3hvD59qiDzFjcJgo^0^Ts~ z<_?S(<>+z&AbR(}qi1Vjc)E7Ed3|C62DY9>w~I1l21|C`CKnhjxp~HOQ93Y$eb;kX z%gFalANHg1YeM!q<;{+%Wp^2lXcs0S7*aMyZQeJGGh;22jB=*A%|pZ&+Kefv{3Dj{ zNw~HiY!;WU^G1>mJiB>L1E@L1;gAU!#eG$xe#~(AjtH&(8eLgYVvD2DWDaGrdQKdU~KMu94ABbAcw{L0UO;B2WrOLqtq;e`Ta(NdMkf+^a>5Kth z%{T1vt>4?nr1;Ll%Jn0dNetKGOSzg;%7~&n2_zGNw4S-gTK&8DhvGlkhU3B7hll*C*q^!4!#joezC z68y`iN@KQdt(<9RWSE6jvWen>-Ix&GW=zeyg&~-pdh=gzd?ol>;eUx* zOr9V3n{WRB2_#UtVvlDmn-PIHOQx9~2S)q5U6nw|KArs1gl^7slX{;xL;E!Ni^sN~ zXYpmOpCs`NDbr(dcZ1It3~=4R4hDJvYPFaA6Wijqg`NCUr2IaA7irF-Mi$b^E!DdL zz5K{zcUg!S+#&?$fS_0E=fNL^9}T_&c=P@#-x|+#pxoO^lW2O|+f6)IN~&NM_c6*D z@-v90VUj`KI2F^+{{RIx@E?XX<@k5vEg!@l0=a3*#x<*PWp63lxlO$G4JDk#jydz> zC!NIftYM9BNG2&$P2Hard>Z|fz8QRH@O{T3(lb7Ok?W2r(1$t|0t!`M5tXVb1fEp_i5-uPpB{4DXA z@nwW<{{SAE+!A@hAlrb3^zDv*Fd0~a!@peJCstn;hBYFTebd2yA$&5M!8(ymerx-y zbl$G95o-#vaLXp$>`2c$wkhZ7TAvDTyc1=7+H?rEse$r^VI+7Yt1_IIQcejifKKCH zZ}Cs!4~u+3t4MWSRdj`vyO^wCmRz%w_oQ~&7{g;~@Du^gGHYnMg{Our{@>ws`&_7B z{VL!I#v~YGdLRM6hXHfz*1bqVPB*>GWfh}7z_R!sp!kj_@jrvLPxwc5rcLq5Z*c7f z?&jeYUOXOz6P|wPuIpLwb>5xei|G|zg+HBdA3ipn!*ut4V!FGTVY<7!`(T)EhUpom znQo?=>F%672jk#4-sk-b?$3Qc&$qAdbxFK4(|6MIjd;%&&(hg;{UI^wPehlsbqj{< z?-57XC%6_Y-N*WPFpJv}pZAMGn!@^<&CfrOZ7_KP=y`;v<%7|7%exf!Y6#V!b}=?j zH0$7cr0mS`X+bj93SM4bYM3qqOWc88dsSJ{xU?I@_i@Roo5l!rk1++^#*OWD{_**t z4id6^N#-rlD1h^pN?Y@vt`bNatWa-MA|PgW$CAA_a;IvqD|?T?g6WyI18l!B4Qs}# zM=Y6l*I!ffg|V6X3`|6K)k$aQ+a<{X-kl4XV|~cKo@6iE@|puR>!iOLxRehG2Ds8= zjODLsvHz{A7bW~%*BbS{UA`4RL@6E0WpK^5zMhsQGF zoXl{)cb);7ZByr6}Q>CPhH%Tk{gF2XoaZmx#>(XeBd;l&Qt5IIza=yvkB5y=9=8=|pJS}t0Cb%zFwue5PpVXT ze@v`L8#b$iHMqgqZHoQSx{F%TC0+1eN`RW}hZfmzC`n zW(9IMNL;Gqa1j^h#(2-01`n*&F=rR8l?0OyaT&_@bhGR^5_%5wI@1&`S(lq}jbUfu6 z9wSUcVi@zi;b|MQwx#+#6)ax^4c=KgvO;m(IEuB2o766?(#9I;Fh$sim?-?zKHM1HQwzgp zY>8hy(=_F5uf~Oq!oQt*Q#rlt&N8X zfk2!Q@v7f0^li)79i=UWERhGV1rUHJwrju(NdF5+8=|&f5LDpkqTxeDe!BSYLcrw% zKSD}okJ9)N5lbSY~$=0OJ92H|q*o=332xspG+ zNa$NNBqH%J9*<#{)$Y`M4(*PJyM9(QgL!aFYJpizKCs1FKOD|4jDbu&N!wS$l0(#_ zFVvP!1QNGlb^^1^ChF3S5vJ{*$aOMh8FM>dwP8wKa!H8?>ghQ_S$6NEYkH>k6EVE~ z<=n8?A;CYk**%8@Y+lnZ*WCW2m$gwhLn;**$?;I8e3 zmQ%%%1|pk6Ou}qjj~9i>|5T|@%p=KhVY6B;{UpHou?|jKdE+mr6pX?I;=Sbk0Q;MH zsECG~dNxDf$WLWCGU(>GEZvN86h`eJ^KFx?0o*mf;P#tA4Ey>++^O>im7u#9kcdjV zwMGxJ4t_x(io5=Q!k|3wsvqe5@7C7FR~*ttKRjd(OQ_si(Epq-)E((|!*a7KDDm0Y z8Hu6~Wfm^+#+XSP5C$DelaY={vANC(Vr196IUsh9GpbSe*gWkEfO)F9bl7}Qa z)tqWhu(P-NLJ89N8~6uDf4`CRYNOWoT>r&ekndji9gtB;f9_@{)|hB&Fy_ozH!5E% ziAZ0uqFyZiV5l@D>ddqpzt=YJ`~qq*R@QcPqj@bxWXm zhdKMoN;{3##s{BrceSOK44w9kyE}It`FKZe|J*d@dUieexqm~w^Ff|`%EXN|b-TEF zU>g5I^TZ9G%+mE7(fInBQtDTQVVU7#r08J8#**ic4tqbXTtnU!KGhE|prOagb+9y) z(jFReB($F-sCgKT)r9mO?PEvSy!%P8)?gtz{PlOnz~s+YrgH}kZEZpRi`KSdcGy_R z51&b?`owQc1)s}`%~q)J1`yQkODs)1jQ8w`F6basYs)_U9gj9_JJP+@TlEOUyG5`_ zV#t6>>&ACk{>fDKLF(s>*RKWDgHy95x%w*286oKI%C$rVmBB6%&-kh2Pdg&Kl665 z`Mjq2Z}cBi+uxiDmZ2IHMr&1BS^^~EX`x4=QMf~p#89Q*(#Xp!oEq3bk{m9?tn;}d zvLBsOvG)z3Z<^<;`qnfNOh_P*lWRPg@Xda4GwVK1!Ox~)@W3i( z3E1BC+eKNtEH3){Nc7y(h%CLftZCC^uPebN(6VazkmTz&;q9zfA*O9pZRJ&=mF{{x z+*;VUJT$=uEr;z7EUTA{_m#9Y#Fk~db7yT;dlc@0c6?s-EJHZqg#Fh?>Mb)?<@{X+ zPWDG@mhSG-UlFIgWHVw>;#_Pe0K%-0RDu!arCFa}Lq~icSxccF&YzMdAw`+`_3>&( z1|?WtGP0S55|%2#0w#lMBgmHOtnE3wBx{H}T04;cqeD;Ue*QDj-| z$l$0C5J(4=0$>2nTlE<7b%~mTzW}{a_$d?mV5Ee$5_U{wP(T7|h;z;(&()oD)|~;w zH_`)&Cf%{@&F1s!eoz2|Ce`UCw5~lmAg{vIP!5HJaz}5z18oGD`jL~wFhf{k#5B|V zBR;6!`x3(vnB94jX%m@78A6tvt*u|l2h;6btz`{t;dCHoX6c@EN+&Mu-+#%Hk=j1P z8zp_cjuv()RL6^-T*W5TS&)HV^)z}Y@fMB_j4pGHBq{x=T;4b3mJIscsZyr3Nk`h` z7$+2zBm|-m-4b^KvVNkl&fUIJ_y?5>_P!1mL9K?2?**tJ^aED(B!$;FH^s%?*0n~) zQ1w{;88nCEJLB$pNXEacd`1!M6eB8;E3eB}7UUQ|CXxiWjJY|;h$?pLDE(hRdra|8 zG;e6l&`h7P^aUXV@iy3M_hPZO z%ti!-oR`(_@#p3YNZ)P}^y@MeXrnP&+o7c7%&%_m ztxS0~hm`$<8v^sOWH@hq8#N`YB_|>`o8bmfruJ_kQTxT^@umY=i9u+Ug#y|cE&~3q z4h)7hR>P`vp+TqpH~o$`>Q?2KUpY!Rm=9XopJe$ONUhAHEYhC*-e3#`)J2D&z_y3B zIC>#7F|u0Hs>u?~=U z4CCB@WQ&LWgNrfG@jysZA>k*1Bhg~gd%@(#BLceWp%#=?EfyYEuuWXgq^ z4ic)Y;U|j^SG(#ca7;*H@-bBXGpFk<8%(8oi8x(M#O3yyn^nppGDSk|qFDIQu$<-M z^*$BvVEm3Sn8s!I`GXx-bt`Z84SxQ>xvkQY;6}TpSV@SG@wG7PoMw=t!wlO~^)vqB zC4&1T@MNB|mKiFMlT%$aEp#H zOYet|xi`0>vF0Sn#pCa_|y`8^35Jtj9@opuibfOp<9c35Gb{1O}8JaD-dtPUX8AjJyF zs!;w5L|a0LKL@f3QVPV1uqz>0SvCz#h#ztv+v}YC#-BtpHG~9arPkCET*I!V%D7A@ z&k8^LTv@IUFSg9wA?zFSnb#_W{1Y~6seZIUaQyWY4eiJCV#mQAP5nT>ewYpjNL5+u zAyW<5?%)AbMgc5CMs8^S9fv-3z?z#a%*zOaQ5Qj{eBr|B%AHoh(Z*zp?XKjmP3@Km1+BqFrmUp@mz?;<5H9u#-HZVl=*&n99}`{E*2Ai33QHVkap*U# zatN6g`s7B9jyH#`6)f^gRjiNsm$5nSXNwDb0yOOIKErUJNh1pLd(bZz;$3AyCWaYt zywNi=zb1Pu&=^wt#O~@cd?X{c-a%B>{G;W&e1^zZl{OpWhs z8+niz>t?QsQ<98Yl=kC}{?V-J?Q@jEGv`i-YZxrRcU@6JeO{mOD$p`-;k;%Jr`z~& z@LHw6<`J~U&1Llmam$~MVfwFVTUBpFC-afi6t<+U%F18^dwSvMy-={gu}X$m1^t4x z67zUMCibx>bahy~!AWVak~5M++1qmMBWz+fD_qQN2dguc>0gt81B1+6>T5n*c?$jE zktC5<-O^~OW{zOgh(1FH1Gt2DJdWj&fsMY8PWFD9R#*!oW-5cwt&q5H`j8jfE)YKv z3=3KYv;QPnr?pcH#l13R!bX7Ne=|Lklip)^#&BI(0ZAMc|1wX>}|u!L#X)fTDJw*>qxIW zg+tzx^5l8b#oeP3jfnBNjq$t5I(qpYDK1M;!;&FC?)5U=upBEKECQ+5Ot{YF?2tuE zls#)lzq^aoe7>V{niBJX?wdd^YQ_M*Z>thn&P#$v1H+5q4bW6Z64js`=s$+nf6e}L zS_4==yEU3D+qCPu z%jxOY-4gSGpC<)&2nlxZ>!Zj2-cYyG?!-m7{Sy9LbRRqt+d~dBGfkrMGF)u#gdJ(T z)vX_`IKst!&v?D^naA^WIBL+*uV%^h+`a~zL#gH?YhQJ2=9B<{rSg_hbPILP+=Pk+QmpRbKnUks<-#U3WKV}hH^N%#FZ z;%;Y9;G&S*7LPT5wti3bN=0|8j{2q2+zWww{DjFw7cx29PeQs3XysEZGBnqB2VTm~ z-Vze&d&xuPmtD$`xz=%YsB?X>MbX9a~IY%GP$@8u*F9kjks7_ z3jKFrV&rRuxw4wzc$+5PkNYe05 zMdL}RwF2f~17XQ-XPlEBvQC)#4wc?8;hC2IH4+=URsZZ| zjyD4iBT!1(<5#z%d7XhK;o{_U9H4fs(D!OXr7*>+~mDfSxZo=vS7jR*NJ z@%zcs2;b1O35$D{CA%rHq6; zs>1&16SBIdb(ElZ{VnMh2MfGT#{**T#IH2pb{%r^w(=u_$g#psN>>+yTkuNx-E~h#x(K)9 zruw8h)tY~V4)kN7>SY!?@%)AU`}4Al_mf4^H>O(~sI zyQ{Ef0TEZfh7YABamElsCS(EOF>9e2+S|INDGh`$?ESFY8(5H9^sph{I%8ar7NwpB zc4bt~wk6LfmMeS3uX1*xKd_%2{!a?a)@}8fUgZX(C8m7Sxt@Yt76gOS5t6L~>_4+37`NX76|omw-d^!| zZZJ_%j$-sp)K{3{Az)1P^8j>V;|Yorl9PM*!JnWDYhNC-1_Jo*|5&C^Cf&^wPGRAF zE$yF3MSfdZcYv}+{&*W*>gjYeQpwnX*FZp=-j!@QC7d= zU5Iu*>dO7J1Sod(zL5_01JGs`r@B~(RF{&gf&hvll(!kQx9gqeWS49TcyIjchI^P* z6$UhW+w)Wpx{7{UlK7IAeB9xplG)+>2sO=+>|1{Ukr_B zFUq28+x)EN{&ue;;E;;>7Oot4-Yhovw0|`KAcBnh*MP8V2&ra|_w)QU#pu%JGF3^~ zjp~(n5H(Gbvj^9**e{=dGTr`^t2JATWE9#cmBLytCZj>jCE~(ZVG)+1#Q&FM`*l-Y zB)Ls=T=Vt?-dm3I3-n-od^slB7TXPK=dTnEsjNG1jD+fDIU_nk`wC+M#r#ZB3jW)Z zO5M0hvg-*q$KsiK8!?GpM%s^k@J8Gt!+r#2Q|VIY4?lxhLC%J9eCtt*oLrYv4%g&R zWvw?ANWqa=J$lEC*w`MX*){O@kb~BG*wGfDn~(G9^xf_k9-f_4iGNZ&3TlgER8_Ym zgN2QD5pe1FPXDsG=F+vhz_<;ihGJ}y>R5`9JNB?1`fmppD;o<-5}4f(!GV0Bb#wQ* zU|iINsH^Wz*=!<1(B;NAX6;i0_J1awUwvqAjGLXjL>ql@_#_3yWS@ND(l8Tn4>5!~ z!orTJ60v((nt8)rE_wQgo%>tg!7IG^W{p3Rj#ulgd$|+O>3fBO((Ri#LtT)5=K`Lx zdld{FnJ+cdX-%MS{N6-c&p}EWAFB6SB z8j=jS;r`{s8+djBbMv)jPP<4@GR{(hy@(ltT&Yh%RUwU_Hy5CyrmIMhVO^S-qn z;xckxwig^IT-PP+A=!^o5hU6@!d=7&!Q@_F2{QzE(&nKQztrQ zq+d!bqDfKogXhmo0eHGRY(jeBcy_1oyw^_pspYdWq0xsScv%#8i*(9HCkb_1ZKY=} zb6~gH+LC%is4aO^R%LCc7b}D8+gwhf@B*&yoLLcxwb~*2wSL3D{3ix`)T(~^=W0Tp z7{hdjA2pcs=~z%0!hBQzc9M}jZJvd}l=o>?d)%lB-}nzxJhM7|wLfleiV~_`jd0WD zcIEl0Kypjn{O`$z|h#9cQ7lt+)q$v1~q%*%XWhJAU)krM1yG+V%GwPu?OL7ihA*Pu>h;k zx8}CJ`7W(p(F*X-R{b=@rh%tYhH>G$G#*?ji==;#oe%|1kpF;k3-qmsF~lJ!ykuVd zFG-p@b~%-FxcJ=}u^Q0^8`4V7aL03AuT3%zT(${-LaI~aA;Er9#E_rjyg#&QZI89a z{wrRMD(gm(I6A3BYV}*Fnwo8^lhVQW4_vP8Ia|^}{moyMV94?S$*bwfhd1jA>)9k} z^+sfUm{kP~6+_l}3u(ct_`6edujs2ZPLp$KJz#&%ms^Q_))(BrqU|i;$Y>lps8mAkp*CdN$H&`GG$RsK9aAxq@QTu%1~!s zhXQ=aQc>kqe{RM4xqAd$=)E6-kXV8h*#ajnM#>A#7=!gdw=H__Ni{oG`FcD&o`Bn|`z{(J zpyP?+ph6++K=si^zKZvztgoUBOmCFiOAJ$P(#JP2(`{CM@ABsdnb>mM$^pRV^KDn# zb4Z>4Rxh@P%wf3$)6_rw6wvK#xglOckDQ#B+ZM3?TwJ6_*5@-L<~8M?S^Q+!dzbDQ zg4aUmiW^x=szNxz>{o2G9=8?B!)ztw@1Q>)2_2A4JMq3V&cGmId>_xXn zGTGRv4b{J(_fdPoqx@B6lFdb5e-zY9_(NI3LTLRN?nBVSFD@uN8UJG!T5mF+aJWmW zf3rP=FZ!DzMK9VIP6yp8lU(ca^L_;5(R~cSqgMg|3Nz*tuF+%^R4O+Mevh7^v1V_a zUCQ*u-a(7$23HkYJo4HNq>()4?-TGAbcv#e@rX}|C%5>Ul2nX}hq}uNnoE-DP~P z`)y&!HnhQIVe_OXUZxIJaA|03Jo6K0Y*|NJ+?%0d){}&6$_I}~HT3ZeQ*o!N%rX<~ zJihKr%hTTVHGTkb<+3fk+qVCe-D5q%!~57_1xa)d`JN^05jSp46?1Q#^Wt09r{KZZ zD>~hLgWh<2R%SFvf?b)Iu3ARrRE+||e3SCOT-kQ9)&2qJ60ga0n`BGOV#_9y-JbZ$ z{$%Q!@{>vuXNZAwkPgSFPFpG}s@Lf23MwZx?SEDW!5D|FP=Q$MefhYD=joT#?iJdFn^oJX2-oiHjQ+KCsAkatbKAj29!%Y}HS2~^T0-}aM02%(H zd5~^itKP=6Ku;O+7WsmYncyd=Z=r2kC30-_L56j_`{|_-MU=?9*IRW)VAhT~M)ISu zvP5^Ii+}<>huUuiessb8+?})P(x#22U}lkX)k-*jXIp(M0qSbS@E?Q8_%#-v8|e|U zkL9-)Mr%>2-~DC$Yr5dM^Z{cL#tbAfH!mq&QyL*Igm+>~wgtzwek zpuL!QdzT4ghl=Y%+COV414T$ZWu8am%v}4F`VfG--pJg->db#_?#QI2ZeESV*}m{W z0j&rv0UGxdG(bXw-E4wO+VtNNVrw%d_VT^NA0p@g8Ow&Od7)ZQ#(}8_#Mn@Nat#q# z3%^HW9=^93rHjJ&bWUSVv*9X=?O`xnsjf}vs3LppqVk)3t1~m+biu9Z-wQ~`$`#u8 zPMoxVKOSQ*hRvKQ74N6Ty#h2yYxuwx$Wy292BGG6#AxXqic#?5U z;d47o2@R|y6=DPl{5o6bKgl$d)63MbH^coBh_jX^$2EYu`ngZC%bbFAnrmFW)+wbO z&?4ORVBew-SBYVA9`mr5{g2hXGfT8@hVN67f=-9F^%5#BMM`i+Z*O)*$(ngfJ~4`eV7(vVz( z3KqY@`#EYjtqK;>^&!h~TFQ`rhb{FKGEAwOp(6V0j67ZRy!C24)r{cV-<9Oiu@Vf9 zjSm{et1!>yuXfrh@v$l%kqPkfvhq=u-ts!-S@t@xcp}TE+5N_tdiJ|i?COwGqLyb2 z7NN78vy%B9sxx}^3+rSpo(hxtTBaEtda2RWHi_z2_vJcT3TY9}P%4_;qO&t zm3@Oz$_xpwNK#5kUXvX}z|VZTnD)+i){>)LYkWb{;h&Hl+ebkOJjpVvN8Qo=G!0oJ zIKFqsT9Ed*`49g0(q0uCvcMpmQg$)Q3_zhYO1SE;Pw)@@`4z5*?7)yp%3Hycb(Q) zx%zR!!;MDDA6CLFwEL{QbM@;xfG+P}Wb3=)fv z_1|G6I|}*5@_1j)Z-x2#BM$5>&@0gtHn6ZcKMgQYu^onzotk0Y=9?bf>@2KZ=0PVl zHC(8w?+9sWiKDjmRA=iv^1mc~8=X`rS73dlVKxbwGJO77f!q<2OcF=Dcq5+I(Bvnd z#4|?7WywAKwg*jz|C9%8`hSLTtM85?9LG@y3Ccu6-&$}mduewDORhBTW$6K3Z(YB^ z_`=wnL7O@s!G`&(UjyfvQx;xg4r3?>Hf!4Z-wtRCqH8z@%+SFct!)Oy9R-K$1je!f ze5uBr;ez2d%*F-B1$jQT$r{Z+%O*uKTFNR+r}$g)iF)$!hcBeDu>&@!r0ABz3Xl$N ziSnxr!Q$-;rYb>I6BS=VeNqSMJ|DJX_lRU}aGyFYNw7FjziM2(L24*WftAWGZ|py*?lwq^FLHI{dQN3Q3Uq!T*qJSaXG(q_$4V=FYkGCMjfI)pft~Kq0134F z;Csh-d&%ZFpgiA0xlPr$<@gwZe<)4(tw_%u3kCVY87U9tVxE6x`z-X0&Ae-K{1g(q z_e0i9=FzfE*Ld{vv7~W+wy<#NNG}_)ES9< zRzlq+GX}MuQ;|YExcXc(EUO=YVto8nyZg+U^(~F_gahVtR7cUTT`{FLO^pA7MUAqLFgpXdgS%5 zT#w;7k3UCUjUzN@A#W9t=lsLBhMGkQm>Rm$wnUTZ{!n}NnJD0@_4C3F>f9hB@KiU9 zV2&5i?FYSNu=&$k^s`6q3rLDTgUM?)Z_V!LK0buWziu~(1{0RU=AuB+zFOncO_vPt zXE6?mXAT^g@32+ZqEE{#CoAu5$8;=bd+c)?9ju4p29>y(57HB zUqLDZUuwgnjXK!NtO$VuDFo!(E6%(0to%@O%SMnn{2AKtJ3f*IB-1OT{9unOx)Z$^ zwPi~~of+tTAUb*h)y4GY2h%}|5lVtFq37h1x1P(fCY#JnqUBfPejzB3%#45%hmMvE zI|tW4qscU1saip=N!A+)+|U@VUY~7sQvgG8oE|E$>g?A1-r~2Mx!Khk&IJ{h@^w+F zF-F|BObKZr<2ND)x{`g#4@6?3FCf!jFQAjJbsu3xz>WWgZY2W`8n{>4iA!WdrEzn3 z7hel!f~h-|JFfNALtlT0I~i{E{Zu^Rfcl3a=W>mwWUMe#FNq*vOG81`P9GM%dz{Dz zC>w?pW6yt>Ji1oi%^^ylb4%mXHyPha!@At3kPqsO>^;~LlScrOACEX8nurO>Yo?Q4 zqvU5oCV8?>UNhUA_PgPlT`HB40;*2Pb}W6@vY2RKsr;;@4cRD)R}Vw0`}k+0+dgPL zgv12qL$+k3uX2UHRub8>(9m_~P?(RG8JEr?#%r;>7`A`Oi4{7A*YE&Q5B~e0%9omdTYn=F4Jnj{{!Ao?DJ(6i{%5e4jjiH9K@YMbo^$bW&LbwU zYYn`oP{)H=$s4%{8ix_3GPGjezG@HHI=) zZi&5s(!$C^mjw|T%SPc&TA4bA;PM-Fl4MF4rxuB=KUk;OoZ6)1|9jpRt4x!7EohO(hvq%ZRKjOFdBQT&s#mZdI*!uo@zBCwM0J;88_679% z1@tE|L<_D2oK{v9y1dwXo!K;y<0Vk*lS4Cz_x5k5pdn}6o71J7CJI}XZF;{|3L-4d zOhf172U{ivdr7)AN2jY*yZ*JzxU2$DJ$dn{ARq4J{cKZ&yM3`Iy+sn}op`+8ULisj z2;`S6H$;ypPZEGh?4mt;Hot&oZJYDj`C}veHsDiGw5ig*FAE%*rB%Qzo*LrN8>m?6_(VjgzfHUJwSoJaze%#uPwQl2jkJcQjqG@-l z^5>H^l^<9hnYO3e1b?pe6jU%#+`KULApXhjL;P#rIm2{zCS|#_)MoS{u8N0t6mg}s zl4|;F@d9GdEq(!+S-pVdO8s#G;xPvA!su(qTYCwopC04IDzYnJerVNTN1n#t0+v>z zaveEpmnQnP9feQ;(BDql=aN(duUvcq{k~wjC^Uq`v@X+k1}9}D$<3zWzRrF%mT}5E zc9w3c1>`NT7f=NJuC5l21|Z=D3xQ0W2JzWRr2fwHo=>nyP<)zD1#;M3+GvcyzA}cl z>dKF$|Fb{|{=0g1glT-RUc{W%+qjl1Njok0Nebc|!k+;EF9`oe=rkqHDJ@-UJ4Lhl zp{~7AAR!$a-cv(W^*;)G0WnShP5WQj2$`OCm<@0_qTc7Ax@*JZ^qw;1eN$nW1C)0_ z1l%zk&7fNR%j@o)fYAe6>NVm}Hp5IN&?D};n`MV&h=zD;qQ)UmyCpT^i2d@=S{oC| zjeFEw`l0df?|*f(Zls~3vB57O<66m5Uj)(%D3#|7(Paeq62mzEsC1cnC}Sv6myclN zd%i+4<2A?3O6k-b5(nQ`JV>N~o(rpN#OT=X>Kc|ilI~d-4!A|pia=}Gu-KUA({VJ* z;kq+u6m2gqSbD8Fv6;Q2SR)5aviH1qPfarBA057)v^ig}fsko89RE;OFFu&mK za}HlGjg-;unQ7~4ygUyn@uqpIK0bv~C(m@(F^b2`B#+Z_>2q=uhoBlK&^yn3Hx?)C zN<;8nqtn9|zc``*^SDgQOaU<(dW{7&dyWLE>sg`|#1hRl(+J`UG0z|Q@ne%r6$+Vs zNWj?(2y_AGbqT8cTfmlUI(Pe<0Lw(=@6@0&Yo-r3Z)ALsCus|J-YYey!x;j^QYhr7 zbI~mLbK)&<98`cePuel*) z*3_Q?i~Hl$irii782C+lj-&XpF1iq+0Vp%fS1*~e&7L2$=D*IJBFj?g!0!7bB2Pb_ zzjw^l7IelS8mYH@<;YeIwZ$RF(3xT_3h+KZJ0E{uQh6iDG? zmD}S^;plP&jONM=cxu2fib_c*De2pbioDszO1=J?gx=LGi?LL395!!GXYtP~Gb*aY z?iR?$9OfAOD)002?2JnaQw^B9x)9w>jx`ld!z1~dxlCsM0&+_<+-#MQdB`krv1~)i zR+^yB*fS&UT|P1U_qQ^XuMYI5_rN3cN9f7;-=gSfR^NqaUCs3Pm1jF@6jl5r<-g4}glGIE*hMDbP61_T#SYoMkuagvtGLfxbsqf~Vtqtod0 zdI*1aAS`6-S#o^1kA!&r0!o*Z{I?i-i&9gF8af&oiZ}>;B-ASm8XL+{k&C0|f|P}a za5}y@z56ravPVjm?-PpG&^(()myI&1A!nY>D$sgG_U#&dM=u9%1gqP-`qGi>s7pbuJ_GzuQKtNeSk$zVj#g}v@rzFr^FDqn%<{Amg`TS)c_;vA@mjw#1 zIX407<Jt}NO^3z0M&?~0Qn0#y413}m)sD72x#b*WOMZi|I8&Z0 zWFPUr?j#g&o3ISvh4qq)k?l_j_WNbdI+4pV_NaCK>^xWf-Y~-YN~4g{bv%6Y^HERr z{~ZA13+NNz;7XSfaY@}bJ7w^gN4saA^auU`SjDwT@dtO~>W4yw1+eHYtn5#1mvqT! z{l@3v=?iPoJ6e}x7m{}K#cES}~E(qgXNixz`5(Nl~3cxo7AcD$H(>!V6_dBRG?p?A%IRU<7K8Saij&&A(D4*r18PQC zvtP*>zM+FI&^K(oP6sc0@}-6I&b@Hi*#rIxQWg*&(LQgdHP%43#?|Ai> z=9=haKD_>e>ZLKD+Vre@bo^v%l#ZA6E;5#$F-a3cULS(xYU&qhQBZ6~z0o{jfRWmhnCUk@4@ zxi{a#0Sgb3wqaCpaVx7{x|^4Kz@99f+i2TkSIkM&-%cQ=gMqX zPPeTxl!P`;30X+T=y(RsQV)B{LidNa5$!BDj*Czlc$&E28IwvWDg52`nlNMbRVW6~ z2DzdDi!PQzQL8D1LN`4<+}>VQE}z=DK*W! zbC+PXqp4hLU{bz)lHHR~KUTc$D<*KGMX?TAc_05n9Sc07I`r%$l%;lLAoQBD2gr0C zR8GGhUhx@MwpB5l_fmj0p~s1@js4S9@@<#oQ|bL{k24o}JgN2LEtFV_q|j>Afy5-L zRpOmRaHXX&kHWA4x8obZjWk8pFa>20VM2OR^EiwGp?rA%-1Gukbvj7~y@1BvErg!c zm!}`j((wlyoEKt;+NiHOsbdA8(0A}$^wTyDOk2Y1`9 zs{?YE%Q2hWn#)tq%;%+| zNn4uVaGFQ&M@4y)6r3KjU^BpZ)FX%v6QJ%>=%wwE2ptocliUh@MseIV{h<3^doFQi z{Gbzx&EpIWsq^qT6j3C~DD6@p($BIi-q=W|#q51gf+_V>9Jkb=jx^$R2*XoJmg~;@ zm4;xmH*`M;dun9)R@puFo29izV>xtx?-{SPu~rCNpnqO}VPpNd1nA{& zQ)=So`Uz2+YyAe>Rg6yAIZD1I88A#Ua@rzpQW5L&G*)C!n+WHstsEEpx*JB&f{0Y$ z3fgo-$Q5D(DNyzHBTtv$-aS0PFjx^0&RP$L|F2}pQYXdW3$IA7DSk-$qvl}*3_-rp`32Xu%0-l zh6EA7Vb)CK?}|aUMMg8l&fS0U5d8;&ar`a{|tqn8lwGpS9doZE9Qp~$=ut+z#VlxCp(BFtau(| zBLg9&=Wn_@ZNMd=csmkVyH<8!?piHWdA#Dh4m=HYl6+tr11hEY2h7tSHATY*&t)99f^aJ6+L=zhlZ)tw zjzZ0L+XPuZP6~fyw{-X24$sy7Kl^9fuMktcq!L0O5$~V_E9v%h^$sMBaFa5_w;2^j|nWJ7Q{7VioPE7u3le^teA zR(o7aGoD^i5Pn`^M{~wR#L*z6TP35f>xPoov=CyPl4ni5Q%;$r*<+oJ%JNKB4=DQeM$Qkl-~k`6UpxP8z!>r>kc7mN*pBe<;PT7o{ED5=hlt;!=6RP% z%6M!UHtf9+0&(=ZHVIB(Yp8VOISB8WeDUG}V52nSJyq{NXs_AdaN`UI^4@O$ zrlx=e?{lL5tf}Nfpf8qvOegABDZB0!Wt}<9upD6IT+X?rA@E=BBhK`7a4MhLbFyC( zqH1b?+hn0Ia4a_6$R+etiRzD4(7f{)4Nu#Yd|s8z2$<>Td_X&;`S(PFh*~^SBXJS@ zA!>tSwE|b)-;N+p%JTZ|uogVkv>tb?5nz>)l%PTS&eWq7C3hj`VFO073vguh24HaB tP=2vWpJ7Z|{u2mz!BCNwEZlMV@8kt$Un(go=%hR_5NLj>uB-Vsp{ zN$4PiB26TL&{3)=`d)tj+t#|D?!C{P`7~$FI(z2Koc%m||2OmB65yP%zL7qFfdK$u zI6DCU%>ncPKt{&@X=lN7)|gqDnVFcF*+C!{R!(+KP7Zbs4lZuKb6niK+#DR|1kdsE z3xL63PM-4+K>-M#09fFE24Mi6-NVGp#>~tnz{SBO@c%jf>jLnyGLA6=fD94>Otun$H0k8Gt}WCgA_f`mA^K*>wODFEgLG zq7DncdOz9vTJZW8@u3nh|q;gm!+;q%Un}Yy^cVt>D|1g zZ(wL-Y;A)=+uGSXxVd|HdfoT-!8{BJ4GWKmOh`;he)9BLN@muJ?3|amukzj&6_=Ej zl~+_YG&VK2w6?Xs`_%KfmqhOC9~hsYOioSDe4YKiOr@=?(%05Ee(vob93CD2I{E!S zxERiG{-5!`1N(pA;yuH~$ixI>0{ssz2F9?n6UfWNEUw7Hr(+4a7sM~26vrx{n^9Qb z%_gaA^#kl0{DoZ*cJ2G6pZ^2xeo7qb5g*#C`d7QhK)ID2_OUce2&ufHzx z8zAKNG&q3O-qlkByS+Jl8{E+lu-6qZe@LvGD?F)>Z-Z!V_8W{{iZ*`6!3U`s<3?&w zh<TY=CY^h zYgye50f8>2szScn`j(oPMzYf|q3_fL{M{|1_`V?nbf`hJC~Nm-vG2C9Ll_-gCJ|uE zU?5qdKh`QI;V!S`A@bACtSX0&lJU+xKQ%8ng6SUi=q$QyG{V}Y)jV(feFUG^-CjwI~DQ~j?re?;GK zRFkkjq@T58temM{O6=le#H@`XF8svWEFd4K~M=t5-U#*Y3#lpcTA za70$lJCTf(4YRD%Vn%Q*D5diI>zCcYkJFuo?p8~U0D8fQ8Vdd5rC392xmNaYNa$O? zsh6&JNd*fIz6jv1HT3Sv}VD_T#{TI?k zUU+1XSm*ZAEUgkfxsyJaY2A%X4hj6a5_#IbZdui2GH^e5U?V|{VM3~)RHHOlh~^1T zawZp+Tp7NQEdIo*w7`?;+_QKdpX4l0-&>w~K}zzhNG6xxQ>fESMpjHbfIePhvPE;J zMMhp2?8v3zsTae5B_Y;mM~Inz^x&H}6@QiOTFpl=*ds6z^NMF^-tBUuC%lPJ{1bP$ zrlWp8V$%6bwuXW54@H#r)ys-$h1EvlvL6cAYg%Xs|Bk75%j5N=Ltzk893k=#x|v{m ziDy8cRUwwtLP6RO6pKlD(%Lq!C@9;fmV2$6_;Q815Fo)X~a4%?d00ADg zKaEpR-gE|M@kjwh6GTi^oysLO13-AVoj-$P$hElu!XjC*k9woG1Tx;7ty!-3)MtX z<(f}&iv`gBfvbm_4fL;4_9PKd;O1dz=j>ar8&OEfx+=w}L~2lf_v+isA2%Sb%D7H% zzbA#wN6V6+vX&^6`tv1$Oe3j6G85YAI6S)#p_f|8{g{)&zo5z{DQh69`NyW|RmpGg zp@Dty+%pkX!sN&hEg7_vkXC)S{hd;UaKG$4|5y|{dORZbNB9k;6nbWRm3)(@=D1|b z_u~%tA2O^|NhD-8d=#Wo>K~t#Zgk^Ql#SI~a(bpvTIiMx^r4zS_Mr?DH<=R^9l4tK zqk{!0*YPXb$BozXQ#k-g2-1#oc7$W`hnr60&*l0`2lNFLX6LEDt@Mk*>Q9COQJpeJ zpV4VP2Tn@_H42}SSIoCuXdqmO?95W)Z|^FhuWdSm?53*J`WdALr8A5~b9s=AMf&IFIw)e`S<;)ni~abR(^eXAF!$wvqry>>ei#~2l`V$ z^Qt(i(4yRyatTMMEZLOKr#bM?uHv(jC9kI21~Pe-Sm)80`Yx+@jDk|)Y9bIMBpJ$H zGwt~NJV4=&Do)v!Jv1_3V}Z@H&3W_N${}6FbUwALa8&H*)Z0DxTxC}d6ndb-#=f19 zTI|9w+sSS-{)JcC-p;61P%38YXRjVC!qa)FZFkJGx(WDe2CT^{!mI`l?%CO)-&daZwS{_7{NE+S>N|!qprDZXP@qy!p4~OK%Vy!Aouk^qGA{Y> zkjlGmjBsRVsrBjz3z7&l|6I`_-X%tjOY>P@8h>SPBQ1+0g{w$^d9D^99!iLDBKXW( zdCBDCwCz3eBV^3~?#W!;Dur!JJuqQ#?)q|6Ct|n(*pL>!b#m+{qhQF`zQVEQD>PTp zqnfuVCRuo~T5WMSy+hp{x0wA^_YtvV*bA#1MDD4C`>I^1iOBhEQ<^89+ma+7(%@l5$SeUzTd&0W6n^(PB?= z+gf3!KyZC562)ehF%8`(99ButhUTlP`sQn#2%3E1g2oJMIRa@v;?>(pUc&o@c=0(%^EM7tgR7@?_b>_5uT!DBr5>MlB@{76E3haAAIuA<)^OPY5H+2Jy?#wLkzpOoIv*~|Yp z(nQzKNd0O*X%5XqG98XqOEyY_7;-s7%8XMgjd%Jr(v0I~WD{MIcQmCdoVj{4&r1PX z#?;v}uY-U^VXGBQM1oq}sED-HchsIC2*E^XZkTimqfFjTPU?dxN{YTm_3dzv8Nm+0 zDVgajt?qP4MNV{ah1QZ+iSK!T6UQ(W2b$D^7+v|RYDE^X=I2=cd$ zc_7eT!EJrr)}uYK)S<17n;QYy@u_;5i{&nws??wcZvUe2E%4-eVKH$w*{w(~)Pa<5 zybY9(%Q}aaqnWIvJWnyb&isOk?sSUFd2*hnyGS9pSQx&=WNR)>rO1 zb%a6=$~k9fG1)I_aBV0`%ft&LAYFaTtDn=c+FCv=cs|m4O8f50w$p?`y>ZD_mR!^8 zHc-~|wO{y(ZX~d<@^z1*WCavx-7DN!ez5Dbhc+OZl{`1T9Y_wFZZ@q0FHynH2&Cxy z$s~ZZSf=>{<8Gp8v5ZrouwvZdXe37Yfzv;sQm}d;8M$r+Kxw;;?dJ^_Rhaq#GY2Ap zzb6Lsr*-;*nb+++ctZ@#bJ40JhAl zRXCnBe@Ix=Wq>7v4$4~vX`+9n6^!8IUzD3$esS>#jVXMKK7X~py>c5w95XYa=K|#u zo=~wHU{J^GNPh??hm2wzibvG4Gw$sOCp0les4%6qL0bhrQKg``Ne2852>>duUVgQL zB0e;OEf%}SMp?n#w_O=liSR8UCV!SDIaDNclW5N&*vBw+UFTYZuz#qRLuJU(!*1MX?j)=fpRA*Ha+UX zhZetIXeY0aKB?P%u4cBs!Znt&-@Vq^@RF6ayCG{XG-_g&6g)ewR<2&~!CkqY__2iBzH0vVbduu_uN>vvX7QXdwZ5J*HqV7u zR$RXZkZ|8f)7&cWt8&)I*mug(`vL^Ef2iRNG#%!S|4DY(4q@uk;2j-^hUb5H5?KfM z*{72D4byJ*p{!``t(a?1-~uIt<;?{l4lZIHpAC*xh25~=jcj~nWCrCwK+Sc_UBYHT_4cq4!K8?$s$!k{GpS&r=(W>Yh6Wr+GICeww_By zoqb(CvU)K0gA?|E#yi_UYk@+UscF`Cb&}sjz8Y9y|D6 zpSM!p=j7ImUf*Cm@nN~4oOfI>s#TR~^2oexKV!{s( z3w9arsqu&^iUc27sIP0z!E#9Q7J?lx7dITCa;bWx&7j?bEd<%`@==il`fi-*rq0$4 z>E1pp=Pc$bSm}^HC^ZTU?xk>M9EUT$uS_j!P;|b1@&0IE=+Z+iO00o6_kiPri&yAi zs2>pFX|TJ`oiLv0S_Q^5Y8>788#$u2&PQqFbVWJw@z<~Upk}+fNdm{Taea{RQTd-P z70v8;o-8RL7@kUav2s;{FOqY1fIV4|VTH>=uqRvtxt1mcxDAtAj{ke2*_RI*UkmJJdSU?+6gU7)P?J2(w zd8xp}gstCUs2a@uc}`!*b4M+2P5z$!`IYp{a7oeZmtlgWOKoBdZQO4=q!-Z-+~UK+ zkYC4Sw!5h~Q!7J3lOI4}+^*w!)*5FsmKgue%Er$%OY#E1zFHmC355HreA@Z+jO^{E zak5VoJwQp5csF3G&m`U(4Y1jE`7oqrLH7uVUon~=xouIA-hm4r6;MlzuI|4#sxg)H z+0G9`V4a&sc^A!`+@l2gWKH)i+;ogMA6&!o4=)I&ccHmi~gN|8_h{crm30VXsO&K=X z)3ym1Ab4RUXsW%0 z@5MlcB4^|wS+nvpWMMBIutll_d$_Vfe=w-P0=Jo5UTX4K-Z7JSiTP57p3O@i(Bq6X zM4rm%ze0ToxPAjTOPm!GV#>lHRf5D^=eb!}<)%Ia+eB}h-^q6*)^oaCmvTXpRro(( zUi6b#9={^KmG}>E^<~HXg;A67=c@S2lFUCx!Y}5VTxpi~46RtROnpQ~(Y&Th92}iQ z23VAkhs3EP;amcgUQb;|FW1?)e6Y8w#&yCci*DyHND@jsM1~HVA<^GdBh!pKSYE?Y zQogvC3X8=w6S;FZIOVkh$~t}K)_tCj=zy>n$2&I!wleFNr-{R%ZaQVVPT5Re?B-&q z@`F;1v{yPCo$!D~ii|$s%E^dIGrJ>Rxaea5h!&{tDj zT`{r}&6cCI^q3cjAThIk?UHnhO!54ZsdJ*quv%gi<3FQ;cfsV9REQ>iDTl4tU;_<) zM*ECw+nW#egp|KE;nSXthU>E)OC*q@L=R1yQcQPrxZagHDM)1#1Lv@D9ifjCA6ASr zxfIst?54Cn@T=45oIW}geN*!At_K^h$BEHD@zV{q0+JIx z5^HM{GPax*hEyj4Q%B>q|8(|xEO=9WsA7@S%N;^nQ@Yw9rXQ@U<>vQUmUYP501`aB z#1~(K_TqS4#+b<6c2EVebTTB{` zs~xdE!!5J&-ico0v-wpZIH+4wIz!*&sUg6N<^4&gS>J}KUoP;*6_MDp@^EH%*TP}x z+-H%5Uk&YQ}Nx(VG(wlG@QCy@*ABrn-ePWK?|2uThcgGRK^-prNfVmq+C^UV&o zuR%_NHGfZEE5)sKFZQx-w_usONXyEUiKGt}iq>SGDZ#{8?YT6EyY(l2PIy8a@J12_ zSv@i{G1hyZg-tAD+s5PB8+IcLt@Xp|lebFl+n;bn;5VA>s4EPd82oXeX_I4_Kh`%} zGNJqK@~A_Z0!JT2lU(iNd{cL+VACE`*nbjp=kxDr52PPvI5;~`51(~LtFtDg~;&XW33t&eYx>fBi8w5M$pMzShrC2cz*kvi^e85%J;8apAdPoXrw!4r+lD z3<{0?1-}?=bZD=mm(+D@lrZ_9h&`Ve`d6DIEEJ-SYeG8+lQJ*smBi+i~F7`*o3#Bx&PwCu9d z0({s4uK!7(Tu3&UjdT}1e3$&>HTvUCr~Tld8JoHfTag!y3u^lTgvbjWe%D3YG>Oh> zN+HOWYlYT$9P|CzLcS|fVksjnoI&gBj)Gq@2m88iqzdX^zb0p~4Q1*63)g3|PSYc6 zxDp+&=o&#loqi6V3m9$JEgRdj!q9Sdk8p)CXKd?R-AzgH>o<-YGfvQ^#o9(WKpbJR z)3w8~)KLfl_ViqQ^8l6@|CLz>0wKX&CxPb-nGFXqDg6zBENnsIlJ`d;l4Z@lT^q## zz%qykmr7~3FNnAH9!{4%-u;_6iKkG-qr|`PITsnpgbGP_*Kkia=QIF2C>6nZl~~VK zsMgQQJ@@>!fw=sIRP$9Y8b=yt#HZujh+bd<{$RE7!COr=u9lRnN7VaHt;dbyjR8eG zP0|3&82UvY7*#(Klu{IGuTOzOshzs3ju-XxuRG+Pzh{#JRoI@byUGHkdDWUtjxBt` z?4Q%r79k2-JO^x(#v@b`-(V&Z9VzRQZJckUSM+uBn!%)F7-JDGxqY&Y9k38LmQ74w zkZ=*CA~(m|!C&^Oi%JXzvIvR^W+@i!8*CbcnW~5HPmoki){P+R*CqGo9*h{SaRrUO zx}REfrPQb7le}?e%{(LMV!n#|eJVtZip`p8EXv)6$HiiX?Tl_qbFz%zxHH*fwShL- zEzA*g$HewpUi(__ z0tY<6BB-sleA^`N5;k?gToOtd{`2X4{0zxJ==ZpSX?q{wP%HbridzoJf8xUwtGwv#sdx+Jc@fjL~}VYaYd~=iHB|S!Orq z0dfgAR-x5A+Z^#U5k)jVtNt>o6Hx1^E{d{mZa;^8+QY|oaez4Q!qxk}%Rn1B?0m0D zesWeJV)Smcb-Q(2g(R?C{S}mOIUm1f} z$scTr#G>wZd8@gBuz?s{v%kd2tLLZh84Zl1@Y^aC!C#HA@;YPwq);i2?@=8EKbQc&J>}3u$*)6fJsySQ4G|RGF=UaPr>_PQ|LQ;Ho|gn2xp5^l&mTb&70c6qqk zn)oX5rJGS13h?AlPrAX4e9_2sgmuD*8KUafsxY7*zO6R|Zs69`sYyIUloiH!gK3s= z6ZQW>cCVl9t`Qj(cOAHMGwN4isK&dH#W}{3TE8{=gC)n05AREOF0$ zlnlAo1@Aws%4G zYG(J%Ik&j~E<6?4R9)?VPY?iu5WRwJmxICtleB^bYHmDfiS|`;*3uQR$zgK$?67Yd zRWz=1qOv)=JBl#9Lb@OuOaS(D>`)t7}K z*Q+&>-=s?&73)VM6xf1&d>oWZ!9p1i{5acWRH@8%A=S4go@%R}_k*UbTa&$47)4lO z)*-Yg%8yKFg&Cr1EousRH6B zLHp?ZD&v959-!Dil?}@>3YBPG7(7mf!R6U8rhBB$%kE)ZyypKFQ5oX-UT<~f#*I&eJ z@o>IyX8A=*x*Z(QnlbFMgYeZf|8;27vyYZ7TtN5lJt1pLYWJ zF&a~u(~PUOI#O10Be|uIWiM!V+`dAGEd7~`kO#Ife4@V5_Pr$T8^r*@ zcRpJIAXfqaORCBavm2||cXaE^J;dl>*>m`GV1N{`zLX~3&y+~S2eM$A_?rN1>-H8pq*ar0P37T-4v*Q3ISIKUw{C4!Js2Z!NL0ny zAw=}TSemN0QQGrB!A+c^wY_5@$ETMXf!qdD&2m)@`ufX76Ihf1+T+R*TrVJvm4kBH zb=C5))CiCi3$3RG;is0nFTRS6b5zf*HOpt#J@T;+ZM95~)=Yf>^27{#yK#qq8h}Xb z&)~GOnJKb;-_5>HheL#_k(R-OJ&H7DO+}j;i^oyw`r6$!ZH<{mkcnISw0eYP0`@|M zX1nQJ!(fLxX0~aC)$Y6#XJFJM+tbTBtB;CsYW&XZ&fK;lCQ*GOfq~^w+{yJ*eQL$ZZ zR;+(-+3?2dxW!FN*4MD9=Lba=8~7xCn8+;njc%^{lnl!O#on?^R*1~Be@ zjZewdX2YEK`(}yB!u6xDv8nG7h*2Fvv->^J0w{ICyI}zMs2Dh(NB8dEczF43(-l*} zRT%}q?u3f&vdW?%lj3gPRH(37DD7v0&8n511n|bpJD0o>Nb28h=HHZmKpSZ`|MnX zk#H~&A6z7~!M0lVN?R9n=M%B7s!s_?MjX~8;*92#4F&_1WG;h30g}L!spc>8DuNan zncNESvG?#F=Ra#nWVv4m`P{(!4rbTCZ_p|NH+V~-;oI-NqKyXhUFVAsTDN!|${uz` zgEQgVqlp*mws$JZ^%qt0=6VE!ajn`-2@^EvzmVBGFZ}YBE)gD(N|D$ZdXg!tQ#hc zkGIvz9>4#sZa3lD0z|caxxcK#g+s5@d)$!tO?XtOJS%=7kJSA8;+q(gr#1qTVR|^3 z``?i?QTTSJ+mY^%k){m25+Ah}chD%Vg|LTKA%6sqcu&544%g64W7ZIU5 z2u--U^ds4gWC_KXqbH+*5_o8^%&DY=aM7A`HA5t$c#BCTBAaEE0LMToG$BNUr0PpkrR zyY~UtHHCb~9Z%zKGXt&V3grq2oidd)rJFya@$jgMWTA{pC;)!}NAqZf`1R#Q_h8@D zIwZ}>8VzfCIJxujI8=?;L6_(yewp<{8!cQ{EOekY9?zkS0vv_LkJ z&jVM-de8AWUw@B&Ng=3f)3!}!r4hNd`W7QxTsN`y-c*x#TDu8%!q`M-zt5lZVgpQ0 z0V27_Diyv#P4`}FxC4}jJ@timYb@wu2OL{6TfD_V6&`zgL4E7HC_Z})T)}0)9i^_w z$*5<*04LrxX86FDEMdQN={{;o^|dPPKOJ!+S2|PldhY#Tm}gA3bCE=+4o8O36}zesh7d0>Z$5U7r3-Z)JWh*t_fto}36Ec8SyjZ{~g zbt#ZrG2H;16-GIf$ef~?iZN>gSC=S1&69B`dcLZ&p(75#{7x%z8l2G+i}kiz z(gSo;kMQD0!z6Q&-n%W?l1LbSe>fQZ@*X2LGjiXy_eAJ~q~?EEt5&PpnJTeORMSD^ zNEjhB$w7amV<@4uIVCSM4FgPFxC|4;pwxuCX0Fv0>5ltM(CL>~oEJn>rD94C&L=BX zV({Bg_oaO_=j#^-3)7l#QRGlbj34hbVKP7vkRHZ}biy@XTiS^XdMVR*Ele=fzxYF- zLiq2Ak_=nf5e713Jue<9sQs~vhxzznVG*~0S1Vd1CVSm7^;n{~LoxGS9f-51e7T|@ znkpp~w}%f(nP6Mxc^BN&@5+Cf=iW5Mw)s}@T9DaW2cgsl?K)7mOr!0a;mX%xIct&s z&ZHFPSaw}e zIq>N&y5r2bvdBJlS%R|AKs9J+IRiv2dIKK774)opMl2t$CuJJyn_*Atn4N;}o$WL(`64VThx{_GGY(dQDkm zO|{9%&(GR*3uo2cO59+xb#PT3VN6~s*R;6`RqM6#H)hvdYnpGxj-=>ccjsfHf*tIcUWZFg z{|Bf=!#tn906^1v)-AHqBf2+0wau=}xbpsLeO)E2q0k7!VX6M$g`K4k0Kb(iOtc_? z=5kQ({`7@p$go?42|jC#-pyu zCP%$Hskqa<@8U!aheE0?^hppvffu4nlEdv;hKToGeb78<0r9uvnXHi!|e+{ucfpfcLk}12(kqGh0it zTgsOxmmu-)jo9l#~kJtEBaX3Iv{SL_~& zzHO=cfx2L_BUP||P^)`|niZxh=O&D*n}L>-3|5>kRTpS;TpgzHtlK+0j0&2*rwOlU zh--lh42JZCFm_BdAYog_ax?a|Ic-kv?r)6enkO2b1|?|#Ex-1BP#AjRlV5N(a{A_X z1Kl#DdjY;PLfjYtiE|=QFX`21 z1?t*J((QV0d`k(-iDF$HAmp&d&EjeOda|nxJ-)O;6#p+&7k*B)P?dp=zyYC{oha@5i9Et29n+Z z0L$Og>Z!P=qwRyvfn~ZbZGDL{e%>)Jl$xIkw2vTdUu8t39iowGT!-j)>Qx;*YC2 zd^1@bIY!s*RoqIno;)ZWsw{5gd`L~$b_}&vcq#>;=7_*kCw98nG41Dfb~Jr`rcUcE zRBL#UyJi=dUX8;OXHk0R6nYILrwSu3hKu=&goACISWwK*f<{4czpS2PYs?IWfVR4n z)hv_hD;DHI0G6)F<+;#u`-)$h^M8UzUZMq$1kauN<8G>^X`!#vCuod#Q1 zq^3(OL+yp)EEtfIQ%zT6)9aw+LTJl3pBgV?1DGx^ejTlTc=^@H!F*w8Q3UdE*nQz* zCxK-#ch_N>^;{n!AaY9M`zv>i0I75t;j!|?Rp<J)YbnatR6mUJ8uLYm}&9wl^ey z08~zZEdXLys3OFi=!^+rr-yeX-G47dP0rA-e0aaVZNCyvDk~7_l-`t|LTdbBr@law zLy*Fu(=z&D(8vp&BA?#5wAkLba)+%7a{0`FWPP}a7H?bp^Lk~gnJnX5xMp2`bRGBz ztH&AFv{P`q39!UxFv9GqGWUoasyNYU3?&!{iI_W0$*Zbbchq0Y#9fzX2`Brwy~oa= zAOM`9(5ZCciRJsVJfAX*`%J?{03NESI_6bqW>~E&0nh13ZKnAQwY6GAAi(u7kH=|R z&H9Gj!5#X-*t|tGN$I(YG!15*5f(L+y#&{qc||ocn-#`qU{0#cy{`T^N`wslYPM?~ zUx@!9Jz|7duDs1l*94hNb@Vhx@*nKl%09XpMx578|FpdtJ$7DY@}lHO#araV`po6P z45FZ0tbk|d#M1e#*4jsRr>M=zmvLWNmFIK&XF6jn7dnYw{{wh_o7poL^l6bR?}gVt zCTJ#K)0vhfQzC8GuRhTd^9B@=#@fm^C$($zPinN{u(||w>wZLIfO+4q zKVAO;8p4uJMM7BH8VS1=1{};U>P{NkMQ+Rd_?Q|s@4!q?esIUSLg)L*_&f}aMz2jU z1I3sv3}Mqj%hhr*!uKolP@f~+l?VR=@N-3;92*no4G6a{>{I?0T+_T^GH4&(eMwx5dm5SGD1$aP%5f$M?qr8)yh&B*4WFiiK@Qd`sep*{f#-@Zm<7i$PAIB{zlv9P1B9!%kSe45B+(o zNN%QTQ7$EB;aB0@zPUfmal2Oux9TYe&pSBYtIgS^f6_RD+2;RBxFsq-;XtB(KTJ;J zP+GQs@Mz(~g4*O?_!6nUBs_F+WAdS${}SdMTvh$(sK0!&q&&~D(Y#0S&eg{Ee8!uwk#Y-4SP~cOTI+#0RHt++6F_{xw@P35ToFKqj%XqsFzHz(#Arby@d-bD?>9r zAQpa2_x`)_H~Tc|$EbZCDO#j?kdNF)k^7R;mq6sl!4EdkKd;A~>P5Dw{!YryDJ^@H z_;0KBUe20f&GViQkg03cTHbPxP~1e(c|B;C*7x;MNcAM@lDB$6BHeT?MsbtCqKY2W z@P=8#1#)Ixf$|^=wdHnHlFbi2+s%*U8wrchpR1`Fc^{IzOMONXFPFK1m9*EiH+=US zsTaHZ9`VMEuOB~;VC?NIlH`-#O_z~^G>11%_E6Vu|qU)e{8rf2P%ZS!3@>+23v74ZsiRb~z>&q&kud1#$_ z{S0NsWUfKo81E*U{bfe!9ZUBDfv2_8#XenI^!*!$g*vAzbKQO28k8Mj;T8a!Di)eC z-R##@w3EEl3PsDkk~f;?(S-ZV9mkei<9`rd)cPCUpK3|9lLp-|{>Z2pG$xaYGI(6= zHCr~M^w|xM`Xx+sZxw=KtfMkwddrvD!xU+R`Jd>UugxNVD6}mX;80>%{w?XDMN2V& zz@C{;0_FCzk5eoR!JP(sy{|=0FaHxmlue>Z^Xqg4HZn$RL;4$u2H$|HDA#aJpe#m z=hWvfrQoxnYh~nWL#-4%gA%5aXL2a^FLq^Tb@-U2*VjSnGL>#08! zshcTaqKq!%>9I;-eSNww$mrE z;gECYPOi!<4yhRe84s8b6I}Q``AAXjI5=DY8n>7IJI&a+FDM~l6TN-!GC+Q^du~pu zPzC%+Cc7Vyk(FTbsgbueGlJ<~v$B3BrpKSEmEEH`(r46M;cN&J`t4SNkbz%SH@f&R z? zb!KQXw0YWPLJYheiXgIfEhzx#uv%T2E4=x({~`{Bwyi5u+hGY4XToX#UHzTod4JMv zMX3Ni9+TgbzhGXj=EC74a4jVPQl-M;s|~mgY<{RHGZMNdt^Uo$0iFZyso87eAWyGb zHk_QxPN#(k>mv?`?2Pe1rSEaG3zcjvPDDY;+RG*CYo_hZAQypibK1gBck?P9ZuIJ= z5^=`d81sSrZEQ`TqJZArvs+o>~+`5C1x%BS3dIIK8NKSAjUi+ttr~Z@(hCr8D}^Wd<&M$PS+E)?>hD#J>;rS&Ic`YW<>0s{7C^ z{O6K1li?mA^}Uu($y1+upSai^LZ~cFvfF^iuO-9os9A4UlMGQO$>vg6bW)Rei-(aV zNeTIT+V7gi2{FLFDT#v!s3MZ0T;DlsDF_HGiH6g`Z~Pf+7$1E}f>Jep+v=2cy)J@x zNxq5Pa)esPcrHBk?+KK>9+bDXRfMuCH1vxOuZl)6->0;&X>fl$Gr&##eEMy-#W|>r z^Nh3lZ`mbJqC&4&{SpfJMa*5e&@yJWP$rpHJM|iBlrdeSgy6fhJDiRrk?|e6+GdCK$aI%j4h=m@5 zJspE!{RHjxxRj@5nSo*MKjYR9tBxGI7ED}a!GG6z& z1hPs2k?W@CW3w-DYpwmZy`}l=1ukV=yQcDXfLxboc84nw7N>opNUpE38n`8OwnN=% zCxgFu+g^Gyq@al9Ze9Sn2j8}Csdr8g{txg(2s~jAeX9%zR=OUlbx2-6&(;)RWwy{J zB~+6$vi9x)YzO8gndO|(h(?r#s`Ax8<9LK?cj@}+@YC4SioAMaBn!`p-&09CCsOhZ zIqE}c{(!=|VEw!LiROiUr>X9)S}p?yi`@sm(IKn-*SShM@S(#-M};dyNoI9JNlMgI z^iYqt#muO-jQ`A|DSgjkw8YFjOJo4O1oP5NV`XILNxB+(<$nXuKrp}AbGR zip9M=Hy|h+gWnzN86#*joWI6ND$2RQZs#@4UqTu;LUIpd>s>~qRnhQy3y{RuH+oh^ z-Ha2DyVAMZI@GLknu_g0q@CY*ZR85%wTM9{AP{&J*XmKr<`qyAExU0&vC_GnRT+?w zioE*wtR)_%(PtT`ZVneDw{R;C;IWXLZDZrEiCU~mUO1xnLmo~&fAxUNP)IM1ax z;aK2&yzU3n+L#kOkDKYLOj?~C_JOTjZk{*H0s-hv zVA8 zted%rX4r5s)OywH(5h6AloQl*SyQR<&ryTUJ5=)~3ME^KvM>S9eD$hs+qSP5&wpBc zrI_-2bCFcq<8KE5dit7%%@Jg>Y|q{&j2}@$zsZJRF^piLaGvF`3m)FUSBLye_sTlcHz0XRy@R#;=xVqEY z>s;|0+Pr`XZQ3C2NBhT)eXCdDFM_Nzy-q7lItPx{A?A6qIca2b*Bz_q%{M_?Yh&{H zPQyN9V3GOOkat=GcR4%l9^1iw7+pV3vQ{?e2)wwBc8+`3ka(u%{{UFqDyrYJ4B$(E zG3j2X<4H8TD``%ntG1~lxnHvxcCJ9+5IX%U=8uT_q;OAdaMzl2Z7JHZ!7kQO@6V+~ zrrRf$`EA+EJzOoidBj3D;|H2_g)%l zZFD_8(-JNOx{4W?;Bs?ZPMzWF-yU2|aVD$%oFtUU(XFtKGRLV6*AXC^*v4k!_2w5jSwC&(2Uc3L*S2wKO8kr?GFb_C62cj&vR*LE}Uah3ImeQ`}nI!ph>?_6oH+(MAd`D?>t?0K=-WXtU zi1=q1=-#~dtNN$I&kyML5?X7T-jAVbakkadh3B0zj@idU?M$-pz4n@t%x;?h08gGm z1gJ0o9R@y?DY$8SvRl8rm5+k2ZQd)FNteoCa8M7reJO@9!iUHmtL)zv{>ZWT&h7Ob zBK{a9oA1C*2t9L*dRNcC5Pk^iz5|j-&IMwz-k(QN_YZMV?~OkeTQ~ zHAZwKb;02J)|`E`85x!L=-0;j&1rt46|(NIkJq$Em++$2|xKjP^CRXu!r5&f(jfV!5kXys{4N z2*Dj|TGn9b3aBm*T1{!GEa>gz$m_HYm^GnqH(j7_>OBo}@Z1nVY%YJh(zGqES9k%( z)bt{eHZ4HC;SL5{nRIpYGU zn}N6m9Q7FKRVI`$IN*25{#8|^*nsai1Fu?7W3H3d=8?Dvcv3T*~$&x(16OM47X2eyBuJR%bnh=YP?{U85usp zuf#F9a>RO8O^ewwdtK+|18C0~6{7@xUm%_b89i#OF?^+t2qPa#yA*tnEN=jFxYm1< zWRCr2h`LDA zNmuVWp1=@!C)S?twie@%P6+g>5H{_^gS5Ba>Cd%2e%i>s4h@&!gKpMcndhgyHq+3-D+&lX3xIv8Rg~d+gV!DDL_{&lI3)3f z;<;wiI%87AZkYS0X~r^5NQg@1PS*Z@wDpA>7%b#;z~ZKgAfTuOu{?iOfW=f& zyEt5eGsqx{vuxzQQc!dR^HFr+LD=fHDq9$v71i%XdFH9SWLoA2;sQlbVc1 z0op<1H7XkDWd+Xr?t^9*6DrV#QX zQhf=siwZa1Dsz@Upr~2=;DvbWij4_7LlA*_FQKIK0l*B+lZ>#%cg1u$8s^lBe1OVO zWqanRNEaJ<40@WmaHDol7#YAdLgn`cBjo_#kZYkL97@j29Y@N3W6AH2N<${%4mN}O zQ)C0?Cm=I&8s_aUMxM6xVE3)z3Pmu5~t2h1v@TVo%Dw{>h33CnR)a+Ep~G1%NMY>}F$IaWJ}027MIxP&Tg+Kqve&m-yWSEZABH?cSdvt@M(K;=gO zXQgK;8>loRU@JGtoe~;2L!Mu7=G0)<@sc04anmySob=H#btI33CJCVbIIRhRJt&n zk+cFbqjpF5)_v{5N55+oKA5W4x}!%CDa&WE;;?lMV631iTmX2_Q(0f3(-!P$OQ;24 zTRTTjbL&+`pbKDSn4Sr)eqB*X+&)$u_2V@JdZRRh<{*z=_0vrmP2BISue_iR^&k!h zTI#fm#&rq^5_;E~={k=iYjd7>uER>W7y+LQMoILp$h+=yOUU$FE3{3Z?TnBJZne=^ zUlu6k$j8fodi&P}VL2}wBLk6-t#r0mA=f(KOq8xI7xg)LZ2UFvkRzKj)=yU1Yf= zV*@zvSQo0Qn8^#%cS0)}XlRaaQE?z*asg5T0K_QS8?+22sk;e zb5dkd6J}dE2e}@#gLY?PXFaOKl14&hlr95jlU&vOOk+lkv$G?f)zn=D4S)&*k%8ao zSXYq%nG961=tmf=S~ZEr>Jb@{fF#CFO4mPgX^q@%9WZOQx`9)6NXu{t?N~a6m@?tM zY;p?YZ3lZ4k3%CQFrl&~` z{{SowF_5(AqXBYG`0G1~>=OZ;{Ad2X$69XLo0EH?e1|*UR?mJQl z$Qv7idEJhlly9(_u_8#S+zvrF$?5M_Vci}_1bg~=)CNB-;|B}~9QvAkUpbR3!GR+u zj-9IcpF=ja)Y4m}G4ILT4o==qDmktJcMG|PKpXj*t(J`K`r`gJJpdhv z6>d?x7{hvTRc8l+$AQ|dK4M%-@-ZBO3j2zQt_Cxaoc>hVh{FNfoK;7Cq$wFaOhun!)v?IV1!1g%w^CaD1p@kE@>Kp2E350eE^1M^uF`IEkbSCdJDr z&u>#-dHf#m)E*1ebZB(jUAo=h?<0gLCf*WwEPE5z7113P+2mrT*Se2b_$%;jwBNH^ zYQU?+s@_x)?ZNMsHSc#mAGfrV&$YF*M;Kq;e4r?)Kcr5l8b3?7GsQ&$m%mna^?7SJoObHi1h7b+{-G(XR2Dp&&gPina6%V zO7VXZ{6fFIzPQq?Og6$dDLm2@cU8||GmO`r-haZ!vb|Mr54E&vqA_uEID|UqfE9VJ z_rjhQ9vZfP5qPz<{SMCDw8GsSAefHipPkKNCA*ReP-g;#1$bP z>5E)R>E*}Y@D)qNpBr_H%Q@21M}03&Sej3_>ocvpPCIN)xczI+b?=E#>Q8&(Z7WK% zBHVeg>bl8qEk{5j-TDO;JIii?Jg>KI)E$DtXmaZib0+ zp`If6uj2mzi8{@$k>R^_o*^Mu7Mb$n&sTXmpU$N52ZAiTHDNNx9j2kHMIQUx>6n<~ z80I*}d2IJJ?iN=*2=LaM2ZB5+d#6~$ZQA2i)LuZZ0uD}Fu*Q1V%D)#sX`c-0H=3S_ z;Egz4X%-C84OV5!+_~VDUagb-Myi~t$4Kf!jQ5dGPma%4yl1$(^KIY?BZ?9i&2~Fg z7rExWX2u^6*k4}hclzC)l^&S^ZBl5P%rJ4EDLWYm5W+b1+l;>nrVxFhJ>VFd+Z7{!v zbVbp8GYb_umFfn%x;g950Aiqv#~v!Scy%bXd+YcR{{Y9dY`bxe0a4Ib1E+jwwecLb z_V=v>Rw$T3Dt_#Lz-!d}F>QM#h`#W;=~rd2l3SvW-kfI`>(a7SVJCK9BiS9VfW9+b z>-Scc8up!N*79+(ctbN}A~yI@~)pq)DMUJO{`yNI?TGBld8L1G`B!* zJGz6Op4F7=I5v~}%T&3yM<4Oiz|!4WL91+qZ*=k&<6s-63_eyT6-&Y1AiA}+x3-y{ zNKeeDPYcI-($+4oC$PRtnTvV3U!AfFgTVYNjFwxgO+CXDLI>U+mCZa)Gg=)H!!;Lk z*}N&^sI9dXjpN$R`Q8`+7rs4z8ub4Fh~EUX&xk$+o>}yym4caKyNXOW$o8*`F173H z8`x!vRGF4AilunQa%<@CfnOYFghywpN>GkVkU{`^W74}JUy&(pjv3PLq4AH#?}1ld z1Zi$!o++6KRanUxAB}ia(~dw~Wd6SO`pNN|;bpJHn+Y!Le5;r*AS&w$Zt|FOzuLbl zJa3~-;m;K6dR&Uz?WQ2J zk(#kvp<*7H&URKf73Q zJ8)FsbjB-A-f<=|k;i(LR_67q5`dGytth7Qn}hYNI2hmna0g#{(1KEOqy`^amwkck zZraMAboofeb5{~$xNRGO90S_1QNCO$12twdAV4#N`igrIxw5ya6vkAyuQ(MG-71zl zP5|gR6+p?iZ&GoBDkJg|%2*IOdeD}GxoTPPfpR(yqcuk2VkBU!tCC0^sU?)R%yXWg z`qe2Jfn{O{$Lm5&XeM?F0el~BYPpnxSbW6hlXAOtH~^2vtCu(dTb}r&(X@MyXDT+6 z*V3BO0N?_8bDCt2y9BT;+t#Yi!T8QW^sJh(-%6#diKLLA!6AXQ)1_yw=cuYDr((m#4?)dCg5(Thu1LFi+t;wF?3i7^VTU6%&fbSy zDGUgE44*+#jm*7pM{!9Q1QJ^q^c5s)#~2_D*gRG(+hftEM(Cax%$x$to@)KOj1@Tm zow@X?(E=kGAaT%Dm{C~#VboU~)8*GcV{{ZU>oE04H2d!wMAOV75x*tl!K0aJ-+T9kPEZ7eq`El8KJk=-=;QYkn)4fj_CpZU#*EQP%G=57g+E^OBqM?j z+;9a-W5bu)SRQjp3}HS}aNR+nzUF(e?1e~HJ4pi=rx^B(k+VEuy=lTi`N%l^YR8rs zFU%z*Zoxf1waZd=HiEfXBIE=+J;1A0;Xo!+^An8w(6JLT?F63S9xCinFefd`ACzLe zdXjC}^=QV*0jDSs8TW#^S2E`{QJx|iAF;ZmCe5+EtD(uR{ zfSm~LDs@32Di>)4@&{k7S%xjl3Xl~>7%=42$s1(gsTlO+S7d3l>~T(_waQXP=@4%iN0*-1P!MgRPouFPUMSuaq~8DTe@A&e{0nhL~cujArv<3 zXMieIwv?~jB^2;|v6|G1$jcX&ykSN?ewBP_tsH6s;1Se}kz0Eux8O5E7lF>+wMuIM#fVS8KcR30!C)~Yso7PUt;cWy|=c<6m= zI$hB&RYwQr2D(eTCutXW#~iZutc#e?vnvdYj-!%l?Qq)mGA3oqH{HkZ^~G1h2PK>3 z?m_9rUcG=sq?Q9L2OUjjG>l0|#xuw`$j5qDxs;X5V+FtjfJO;Zz^NU~Q4ze3eq4;z z7$t4PkWP6XwO}y{PScWl=clz~*K-{bCuy(|hynT^fAFhO-ZBvwd^qjv_|{>$&eBFk zazGtDs$`i;k^$i39ffjAD*BeC)f#O!BL$o0W$V`zeWA9TfxG*l{{UK~W!f;KC;;ou z1p;PvBml*_p4Bl-mh4>fi!kZNa!2y3R_~M!TaR;DuuHfR!u~xCT(?mBI&Q(h_pZ2l zoKto*;si693hm<<`c<1%JIO1MK3;k+y=O-IR1MkBRqvl#(Tgsnix%V#1~XmM(dX1@ z?rmFOl^NvlGsSM%$RcG5-*|OAS3PNll~kXYaJlbSZRZ>Y3ykBQYqc#8I&G(OuCurz zFu>y+WcRLy!gf&4oGW9te+uHPZpW3Av;`m^U&^;Euayf7ump0#hP1IAZgpvTR#Y5d zbsyHQ3)KK%tcR}L^{!eCLODo{%MuioAaVHA@mzwdleCYOu~{i{CLB0mgC<9Ou@tudi2W zkAgaT@H*5tce$06uwMMvKYwx#;H<1c`Mz#zE@N&+O?#-QUAb)MC*HZuLsekJVMqH} z>+MUZTnlv^9OHmb8LoQv^CyfBIi}boCGN2Q05}AW!1p!JU+Sc>W`0MZ&M9>Zpt#6x z`8ndTX5_XyW1jh_ie{S=$*L&?DJLzToBCC&E2AW&ILThkSPjgfwuUHhGuT$GjBJ5M z0U6F%R>`7VZ>id8b26fpJoV{ahLv)|80C&K4RHEUV9|n3Ta$rZhMKP$mcSU`5`Ok; zq7O?P(p^sbPH=`IN`QC)SRcl>t!4|j5xWiC*C${QELcH|^c)=5N2d_s7=8ZYx|3%- zk0l4e9DbB^ijT`F9sER$zPZ7|w8 zMsa|Dy^64^7ly!gIo;fv&9u2#*J~ZXSLJQQjGT^ZWum${^&NQrTdwBX zLC$;o*EfB11dO|e!pD=zHKTiV$jQsEHO^k$Fk(z+0CgRWX7@B&85)er-zu)ntO0Cw ztQ)yJ=2zSZ#sM91S~rrJ+_Cvaagr+@=0+s&HZUh~JrAXG%FUfI>}OrhNx>{xKo}iA z3dX%%yQE;Hm-xEY%~hjn`u$<}e~sQ~~SM{VOK%2g<=q zIX#aCweAm^kGl#mIbUB)R(0phz>YhE&6Ga~SzLC3ynHp_}uEJiqG3Q6^= zEeSt3JPZOG2lS|I!Ei?7)SsAks|rNG5^&^p9cd`L=sV~kU9HNNUVZ7y5O!9|BGb{*e3F$bKUqNl-p5%XYjxWM{Ul0tmK zPT)KH3U5=CA(V_9pVq4VNs44MNJ14O<{be!AXN*KzB7}!9;eo{_HB0DqF%qo#eT7lX@28QqSV6;YZ*N}b)&n{slBqi!+6 z&sN$#C*d& zy=fwWhDLLYagO!1q*)m+5CzNj$86SIspxAY(z)!P1bicH4fxburOX2H_rIPvM zYwL(0AS)26qxk_`ZOy04pyLSNQ_p@Re$hJQHw|?6-efR27PfH=uqXJ5JlB)z{{R+l z^*Q0y{6QqM0NGe4kdWTkC#TY*@wMdt0NOWkSxALgY?4G=jQ0vnd1kX?Hmj*bW|5_a z6#*6dpQU4HuT#Dho6#PNfAM3)zZEr`Jt?J@-%5>qrHv46f-m+*Ku@Wz-^ZU8FFZ4$ zLE&!(+&z*7K`O?e%QKI>@=59KUR|NZFNgH>v9#K^mpN%B=v8|PCb1UK{hHr0Pb(Li zeB_g!05tU3KVLHamE~(0J|fgLuMla^qCjLzNo3!7BqaH_j(@&-l^ro%UxhperhGNi zbnR2cU^G)nzS~VwH%b?teHY%ktvgt;_-)|1btx@upuYipxvx2CAF{?f``6AsFY!LN zawOzKh z>7FtN-Z-xwxVxWIx$|`y)tGU(4itI;*jAmFhi5;#2oyIs#xYkQ@I>Zaq!#C(6gj5u z+0wADYYx$@T4NpuR+2vjp@OAfy{#{vq*>%^> zcHG3Cqv>BjcvHZU$EHS(I*u0{5IE`STmB#LjkcMnnIM5e9B>7CcAW$f`A{{w6kakH z2M6A{r6k?@9kIcvBgS=a0ZVrYjV-txz~l0+17Gl~X#|-ecLlbBsb648rQ6%a9dI%` zR~_RY0@`?r$)i;A$R&=`$^2_MB-njk>Dc&#UC?EQPcjrNnE>xR<2CGm0VB~bH8hU? z{M)nOd1ZHCR38w$4EijgRak+R;AB`T5hnwT zS3@FiIKavE^sM`vZMi#nJ%wo()fhPTIr-NI2hywgd$Y$LwXHRvW*;cxsz{h5e+_Dz zW-VBtxELMjRd$bY+*Hn;@q?bUo>q}yK8t}`P*c9MEf^hi{OmIXL8#R+myYD@>$+5de&F z+Nnwg*HQ~{!Okk|*eprTe~T4bROf+}_3u^C6SFhfhTBu=JJesQ`b&uY6oYz9tA2N}gd zCQ0BEk;V;q9LrqSm1D=?vQ9x0k7a(xmhow3}Nkx@u*o8|``R!lO3 z{CPRRtw>y=p<&SEX12NJSiF;NPzmRu3ToUzPBN!rjF3;|SmZw)F?qXDd z{5R#5A`pGoP2`J5MK?W6q=zv;uRJRo>-@+QbZTkMO9^CmX-uCbqc^Us6?K zqzqwk#WWBYXP(E>p$;D}gVU!@lnL+y07rZ`PH9S3B}T>A6(r=Hr1P}Y$~WhLFnP~< zq>`v*8@B*CIq%-B23J)i;IBUQ#ZpbQZwMrd(8A8^kCl1oJJtJm_Ku{5_3KpaZUM_P z7s=x%2c=m{mB;{r>N=75*9@=CW4bbmj*Pt;?K^{Hp1(H~*`3gBURNVM$)-g!oQx3e zq;NY@?kvlZwZOo0QyAJ>oAX7)k|rY#0_Qz@0au}j%H4Mod-fGt0p~J+3Ml7j<29`c zM#HDi#&(<DN5hq0OfikL64rcYm}!>lSS_N*2Zm$t#LE z5z`u%Ryd2>Wp=>mGm<+F)z0cR+ZdC`Jbf#?y3?cB#|A$@MROXhuu+0`>^NNa0;QyD z&T;p#fGXNt7vOFf9F^!Rn~LxN6dWf}pIWsI(Gn_?xm$uWirNsi z$1LZ4j;`A0fJr2eewD8U%H}?PQ`)&pIhSrpUBLCtUxM&&;f_xuHQyAHIHb;)UNWO( zh8V)*u5(t!+Qf~AS7FlwzH6CV{{SQ`Lp*bzTA6P)g#hFej!sT%V~wYKou$5@X2IMJ zI+MrgSK_}-q=vyKBpmQ-i?!5407iF?I&`fFFM}1&ET?xSrM{)W?9%y)WpXlcgU)!X zk!mZGmctxmXRp31k+i-!0a$>2PfD`)!5NM-g~m^M=9I0u+ZY`?>RX5bnF#y9o&l;d z>Ms)&Y_olM#bAv~Dxe0!{{SkEOSURT<=O^tPb60q=8@S7ZJQEmhZx}Q=sl}C<5kRV z0)fv2@@pbXwlU<8a7jG&^s3TNjzg6Ps+#6?F84asHl5LKYo?KRDZpcl_32qRcVal* ziXFWRFH`jPsLaf=5{-o!;d&a(zL}URu_`mh59v{U#;P4zTK@5mR^84sSXVb?mGB7t zb64)z$;cQ0)=c}O$t}A)l@+3Qc4l0b)T41HnxGO``c*Wxb}fPrzo)fblMdZ6j(YW{ zMHwlA2^s1KTCQs#QPiTAq_OPY)umx*#fasPN_j-QWOk&*IQv3jx1$Dry!{4Yn{_A!2=}jJzG8Nrm#S82i+WG9&=rAXB6XPbed>V zs=jC$`LacH7GuehFgg*~S36)g!!OD?We1#&--UI>yBxBK(ppY1plt;9 z?^_mkDpX*yp4{U#%358k#v30mJeB7)rE7K+Dkvv!LOTlSO6Mewx+#oDkB~nZ9gR5M z><;Sa8F=^Ru;aW87Y@fgiS(z7;k%a@Zg6X%Y;xOO3zu^P83!MSdcwSwV-Lt}-0iCp z&E^8k`2!@9c>e(PRxS0|ExQV%Z*h~FhFVOcaOF=p13W0tO3S%#UA&UaD%HgKV4;h1 z*F8S9pC-UE#H4`Za6N09US~pX+7vDrxeXyW3)N5EHIaF8thp?B8(ScbYW?ixX2Y;- zak);t~qO|){+wezE$P1RrxAO3Nm=W$;C|UrDWJkDxTbCqAIFS&O4KiwGr!cRj7-! zp~2jv)ODm-h+(-|K{?BIsjZ>}p5I!PKuI__4c9-!Yee1jG_}^Gkbsaj&{;vr%}EN3 z;GLk1jidCYM!8INBer;}3lfm3eozNE>qj+#MX_VrEzto4_1edrRIBq$sw2V1Hv`(6 z2uS&00UzESL8+G@N3dge8|qWu#apQpm0bPp-93#k#0s_)bNoFjc_Z9L+=lDH;QP~L zO{%A$9Ag>AXiUz{vL%aZtZaMt;}uTvFnI$%(yx=qZ6+NPgfDNURCt$h2&8AQ^{lzl)aX>(Ga4)c z3Je~(?^UJ9V5IFQpsg1u7<{|?5!`)iKIM^DDo7)d-l`7L*iIHj3y`c^<@tv^cC4F{ zLon<)#cEu?85qguBl^}&+?+RJM+Ee)>SrwCWlfUVBQ-)oMn*dKJ*wrvATtBF4%zQi zr6qD(xUSgiq36mvkq??d&N0W~SFIQ393D8uQ%%mxfC(LOTJ~{%P3>D-K!RvnL1!r6|_EFj6)SabaBRO)BX_nQs-XPZQ9Zmx&emeCmiSXq^!`Zq^Y)j>F|@n zkoafAI&!9GxVeNFgARc5a%%Rl$6eQzqCs(Grj4=37~*S#-MyB%-vId5_WuA0aIO3&k?uTE(Vmes`SDwE07T zI%N8q<32L@n#M0LXu3Yp#Hu?H?kcJ%b6c&NU{^q~2bw>~cMr-qJ`b0xjN zj%)Ls$e12d`+>kd)#dtsg{ND{i{>!T`@-CiJ*(AVzqq>&ZZNXU<(#m|Irpr|Q_h_& zkm60*b|a_KyjfLUkE^AGaD~~->N+{OyOwl~VvR6^)UG)-Np&c+%bm{zY_alVZZdJj zc9X1D8fK@tNtICGp6Ae4pIz8$8pWKKDFVdKKIv?lBQAYWlquNrj~VKE$Bni5bqm%E z6zqZ_#BiwWqqcjRqiLgFDMPaW9^;Db?0gfd%8b%RtFf?mH$z*Hd^*-lNbx`}Gn|Ou zILB(~oa~jEy(&&xvj;%XZZeY?npe*ONayQaHm3}Bku2ib<8U(B9X}eIP0}@c3$3qV zENY|f?a$?1CaryF+FmY!iLf_EAjUIZeMzgHq$#Ui8QvY&?V;MPLasr<{+0D7!mHt_ zLk+xSs;~gGm^^l`mnQJ-h3%v>+sFaVN0-q3E8hMKU$VfV8*?r`_`&*DHAb{(?53Me zEgwcTy^fU^pK$*GFyN3gx~I`K7(7F%n|M*mrUA&u(w6-FyG6Ho+Aoj-*f=E8csfgK zJwYx!_>%yzQr)YeN-NOhnrbbjsp)a}NZebiV*?xt$<}-^EsSXRARZ1Ysn9j}@6-V_ zWkJ9KuIoExh;%F7y%$c6N_^2iayYz2D%`R@bNH9x(|2I!XDFnG{{R~CuLAv>Pw{2B zxcLgpg>l&CzO2@CsI_hJNVvv&b+0edeh@(xv2w1mOAE=lB*r$J@GF9nN?IG^?I@`q zLb4(7&)Ke|5~Q%(xVyrGnznqH_6wSD=ctN43Dn@_*p zIw9K|4o2Kp$v?Iqg_>^!>sNLz%@W7v9j7hVo-5iD=J1Clhd5$p*F>_jD#ngj%&R&9U1CtBvO-WoMRtO)lOJIS0EC4j%!h( z8-_4F$fzZVWP(5k__0ki4mRA!c(;ZG40{@*acr^Yzdijcb?vqxTyjTHDwLMY6y-_h z6{KX|rV?EYNh4KacK7K}NUN4Otx0Y2a8FJ#ROCRU=O9+;+{VQ+yySH6QpdZfAbL>X zXCM-KVw)J}9V+(`I!H@($gL}BHth}3t1UWr1RA?;#EcSsg#$M1=|D#~}Uztl0RChD{cqx2-J+9TGN&?0X+fd+Oy@4aBlwqN>{n6b}3Bx z=Q%j463d=W8HcB(UAR!7?a0RmjAPofXBccL!S$)Kxux0NO&6Bg8Dt!0sk~vy#sKYF zQbqvX$s8^ZYJkM7y@q`SeEWS*)zZGj1%Q$moPpM)5e8yaNF$yqV+r1I&rWIE843-g zu_P1PvgWzEvP3|X3Jx%TT6}I*js^hr=MGd+fR^v?imIo}KQ}r3DoCd>9FxHQFG}fKSsyiswe>E<%7bVG_UTvc!DQrT z)0|d(PI3-M)K!S?8*^c>7+?@Mts2uq%6c#*6Dr#8WLjp2RGm~8I_VYGcdK{&wA8NJZ@d63u7uz7^^m@suYFGk3q@xtI7!ZcCc-~ zbc%KkYZ((+4>7Vf`tey;_WLFZ+;8U|wbV(Cps+01IrOYs=#dx<61dJ!N}i?O<^|Lt zGr91)cYFOS67f~!GHv-l`B?S*YoWP>hEO?;k1PdZY6?S~XDz`erxaQ4Zx?2BH|iZY zTmn1sn&UOKNR^os5xbmMZG2Kq#zJtU^U}CoV1D(1Am_J1Qte|{$((Mmt>y5_InNz& zT+RDuE4vB?eznnRvI#ff{JPGoOIeT;25Va|J0hW96fkLz6IcVuK|gWT1a?l&GA zZaLAc1WIV4f%4hZYt2C;?C(0_{*;W%dCpIQ>LA<4ZBsOG~KQ}d1vdsWEp za#hJ0AB|u`bft5j$KI^Na?C~!-($vWM71+Ii+j=vw^8Yeyl;1|cal1E#tn0DTpR$$ z<~-H7?%R=^FRgBjcRDEb68Of_K;)i+o;4AZkGwjMPkO>jhFmE87<8%Bjy5*a+$DdRuPt%HMe7PJxM?iTM zA`={goMWX+cF=K&W+1s82iC3I{H!z3RScG57X? zHh@7S9OQJZ+X;^HM=W{_aa`53fl7jK2R`+&1?H4 z$-qt63-4NX*DRzy?Zgg;y>m9VJCv~`FvrS9e+tpJmuMg`$3E5985~lDj)E(#(EZ|= zI3b6BYQGiKh@wc?vkn*)z!lCV)(|%YuLKjEeQ9C6V&w1VfJv^nzUL(4bF58GP@qyl zAf5+WMb!Mylo9hD-Nkb!QZW`(Ja;Gg3S(-@FgF!Z(-qRJ_Bk3!oh#odk9v{^BWJy3 z-d|w`cqIMcd8~KUF_Rx0;FfL$Qf*O6ICdUjI3uyClD2}?>S)V&8+J0lV?L+ev#&1< zdvZqq0OG;p6_+lmv6evU820a3bLveZlaKz8uYrVu{ z9GcC0oU)zS9)_WL31U>Q%zEazWVyFDOxuoL17s3?IH}^fE6#SFx%Cx=9m53!Zh7OH zk*=CNg4n>%J#$>nHd-A~lG5g*T?ba`NjN#L)4 z83O^isggZ_C?T*2BON}KLNm8&@J9!cnz~3u!3P7A&mPoEO@yO;#A{_6Uy;smO#~7M zBoetf1E{5U5|S4xpnTl)rzB!EZPFDaagX7sl8fefAkOns%G4EAnCvMfqQOF?vb;l{W9nhwtX3rU8^Dzp%U~^PuQnELFzjq(j zwVgsUN%w{^$9l`Ujf!O85=KYAYLz8>Hg-mx*2Yc5j9Bmx0zp32n=|ZS2^l^6R;|ou z4l(m*9QLf6r3bDT9B0J&#K5iahzzvl7`b zPTU@PR^5bT62TpNYdVG0OOA;~ZTc@cimZqhgdjf;=6VsaMEMpFY z;~44%W9jj>6pRda#dNkY6St;){{Wp#-j_S+K{EBNx@|$(mpCJlTUM5=gyaC_KpMXQ z9DVQURBV7UDxvv^!994byBJ+d1|^6i1bSC1yE`eERq$#0QMfHx;-mdy3_QgxbqHLFsg)ef;lzHTI+WfdMZPBv6*Hup*Z+u>g!TzUQ*MvyO{m%+}C;Io9Q8+QFg$N7{~*6-Z2f9dpom-Vh;Z1!6S!@Dl(w_z}0O&4bh@w<@Owk==41m zq`8eFj5-sOyRYG1M5;#GA6*!`YGvv(&vpz^K*lFIES#Fk9tqdAC>6}hr`u)YnhG*1oaQfnHt%WZ7{+{@I+dn(tTYaTJb@ny1K+`$f!;pp6~K4`?VFVeNc)QU|r zlC38lR)?Buz8HqZ76|-RB$Jf|fl>h^{{Sy~lSI*VOD9oft-$KXfh46#KSFAbyW$TI zq|z0E6a$7}=W`F~S##@}9n@?s_d_ENK5XKOoULeSNw~gddW<*v-kB=3<(Ziff?6(e zJDk?7-OZnfbcxu-46)=4O1O3W>(2Zo}0PP2{f*I}1xfSmO8sc!F4rOMA+(sdhKtvOO>mE}MJ zhHvLx7l*8b_Xxwv3g;t{kUi_md^uwU+(K5BR>CzO9S~X?hg?eVA=RpJgmVM-8 zg^=VQYV@4B6_1>i6;h`&{7EftyxXS=ax;TnFNR=Ci>ZU6vVp({y>K>{E*P|ovB)`L zU6+J4A|?ZZ9P&kTQL3Xf-q#*sLHj3TOIooJTFQf+xyE~n`0w_h@f=B7qLOL*VRjZOru)rC`ga&V;l)ww;EB_O`-y!w&tPzzNE4mccS4%NRs zv^~f{Bw{>Uen6)tt2Lm(908sMZ?&En?gXjhIO$M5rsgAP3cLyww6Y&1%n1>biP1?_DM2 z2nG%}5OOOP;w1%#AP)Yu(H3(_E~Z@lFhB>f6$e5yoOY{nu10ax<2~xDC{_c|*2$S{ z>^)x}De2ayjDX9}T7?H*hNOvrZ18<5(+e8b?URGpb@i+E?VNFu@6xdmx*pljUbUrc z`*!olIj5*t*^W+I<|=yC=p`g>1Oj^>TEs~REtA)!Xxqp^9Wl@hRJ}-O+D6_3W3~<{ z{4#Je`1Gm~%m^otqp7PZ;EXBTk(|{feT5wIxESFU-@K^r#ApP9O3 zb52GikDH8R>+e-#KOxT^6@Fka4l~o~RAp1Q9AmF1ty`LA2k#6Hpw&6XUpYDb0H#q= zQW$^-$_M9FB~Y)m2Vc^!TsO_h5G91Wqps_bgO{NYdPaY;NITOn`&$n8^fjX8|t7|SWj=k=(W z4gmfTa5<}!LzCa|s0hbZAoVq^%qvM)fuzPkI3(cEA8FgOBOOWNqm(KfhRCRt;h2IL zb2Jhfu@4Rw{>cpO|Nkb5l-8 zV8yp*xTwT@6SM+*9w{cneTY_k0;B*z&myhEI3SXscHBGE3m)<^Gm+ChDn`lx0pqR( z4JV-0&3kDJ50q~tdg88$Jg|0Mm$*db(s{P%}216c(wzSssFw6=8*bcSn(uw5Mz0Rt^%!n{maB{$O zHMwCD9KK7T=NJGA=51U=5q5-9G6Cf0wQpI<#o~P6^&>q`(z@dwrz>Y)r>^~k9-{zy z*F$X*ov1!h!N47Bo73dl7nL|UAx{FjjXe2_^KC9b1g|82g<4iZ^fj#E*%s0PVoy$( zsoY~2Y>+#iY9x6gLC$@;jMUA73EQ;e<|K8gx?@PMa028KGI5M_HI1k!iE!kCTO*E! zweB#hxb8XSR~;)p=#@?eNnG>JD>m#}MjhZ1m-$3%o&g`{n&4`&P#GJpT?r*isvb;OSL3pjhM~{QCudvcPUf=D9KaYel^@`ueoxI zjE+F)Yl+q)Ra1l^`Q4MjslQffw#N;sjf}u&EHVZ<*B7cF2d)U@jC8KURQZ{J1QEf> zt~*d)^_z}yoDTG_*okU#+KZ5{jCJZO9pw4&dC2Cq^#H^8ai2=Un0aLLxxFhzW?TFFMsZ5&m&<|+W=81GimDVLI9a=9S;Q`TZh!RmUCN~)6j_o;K9 zgz<`xP`fTcE;jAwj1Wa#g@T1*103|M$mIZr$Qj^z)z}GMxa9CfG$xHpX*L2dI(Id4 z46^qbIT`I)_Tf%9ci@`83;B*WFl_PZ(vDzJF2>ur&Oq(!R$_~L?mPwVf2CH1c)-ev zxf<Y&vrEu{kUMLzrc9UdeXNnilZRo(Ek8B&xht-4npuit4ibzppI}xO>xS7 zPN-A4ryauyLu76o=QyeS&@;I2Zo|^Efg@`IJvijmoy_2s#{>XRaaUB}k@@q+g(XBQY)Q)|zh$s}}pAv4QDa^m>qX zTs8>KPB;|T)W9Dzag)Y+n(K==B@?8$ye_zoQMaxIQ+-4Cgpu;DGgyOBgt!MGPgc+L zsGCtLa4_Zac=YX>)k@nl7f$A++LG>*=FU&AK~(3|-N6O6_fwDOSu<)pkV2PVL)Nn{ zH6T7*u}?unrn(XA>}p)xz;&cv08X=Cf+Y zTx6M~mktW>3XXSwO22P(ag1Zps})B%86&9XtU*2r%N}|t?O3$VgwBc!gPexo4nQ@l zX5=7pdag2i*Ewf%k^wj%@y2Ub+UI^(U^qLEYR$H37~I#2U8})3Q^t8UZDu=%A1OHJ z`PO_g3Cw(|J8(z8)~zFMDx?F?KU&WAG)cY7TbnDKHr}VcX~Ef5n*n!@{i*}=6bFX- zaaI;FfI;AqipFob*$CZ|Our%_`>Y0iDyt(0akK?ff$566(3t_q;2ey9DvaZNd0qj- zWY;s5`ks_2c^UIYa#2pwM*cbxS=S5#u^1f*3VNE>o-#{dWkJFB6-Mv`!t5gefw!;H zi0F4lK1K!dyPm89o=;AHtz=!Dr(*)R?_0M}<+qcZ4WJQ_YZ~Wrl3N9RE20*N%?zprSkTsPCbq*sY#_{&y^K>v6k?MJPZNXJ?m!7F;S2do(UD50uTo$IrQSS z?H6dk8PC6J(MrZ{m5psVN6LpJcgK3^tfTWb5D8@fV2+;%f^T%(ccRvrwhgY^6caqDuaO?AQu38urL%SQ8ACsH_Yo_p|Y?`!f z5<1G*0|%+~u3ECSIY`wAqyPmK#(G2Ejb-Tsy0o(u6c z&x^I2=+Lv?$n7(%YmAb50bf7qpAvjcqIj=d5!}f35!^AEA0#h0#eEgJ1F zpKp<7P>B!9zh6rI9`Rz^>iQSB$3 zU0H=#D5V|Q<7G6bI*(0{F}4X9P#0sfV;JdOR+Xw)SRw-IC1d-rM^ztrkFc%-Rl1fZ zSkX_KIx)`!*0}E#_|j_|YujsACRx+$?I=8|cgNPeD5oc7x%L$rrmT+7$KMt#yici3 zb95PeIis(aZ)^_JbuNDS9+?&MPmlg9YMw3Bn`S6hoUvx$7T zv(Gss`kM2*n}Z|Ds5dhA^{-nEhe@>02Nzk%Su!jC02yCzD#Y#j=kca#-ZzpNa?Wyi zBQ>9KZ0a{inBC3^suNl(4dVd&SFajW`FUEg;JMUJ_dP#F@q1i2e>ZU*KtI;I9V^6* zJcWdfjE|Q!;1z@tWz`^Da zCRm_3OtFCI>QO zcYJph!R(fnJ%x*IFPQt=;a`lc?`_B0n+Uw0n~}$A^&boPhfcIVZk{xIagZ3k6dL^0 z(|$Hy=r(MY^APRSo^$I_#rs3}m_v1K1^j|JMpZ&*6`G@0XrQZ$rjOL^JL4CFW7L|? zNn?&?f8CbG6nnLHHl8BWH57R?>$qJ0@_uJIdd62j(m4e}i8W z^$!lD78-<8#)oJ=Y7kH3Sfg4J>>RbP~$IT_!Ia+fPTXv3K2&9ex;L=L^Zs;rjBZtQ&7`@Z$o z7y^^>jlGAZP`S5Q&Pf3EKT6NqU1~pMlR4XqRx6eSud4r^J_ZBN@d#kxI>6u#b)h&5reYDahaePpPc9obmw#dm6Hm?Z<4@14$bY z%y2japVF^e$W@7Mdk$+FIj{!P+dP`PZz&v>&`?#o@=nzQ3?TpgDV&aLWl~`wsHv z%gD&*-@PQ04(F+=jXvjMPI%^xcP(l%?p49dW%~X#l{*$1{nwM_VwEf`O&U%j3HS3wW9D9WKy67IRI1-ZW)LJp2D$5S3S07MDer8{eT~x zR+c8*0T{^72DK%HnT`{$ap_spC@s4j4h3{1b5xt$&zd5k@s36gMO0-Wim({18CPnY zuO5c0$r%lupf)=C3h#yNcsPl3F_z7OIXFJmp>7%Wj-IC#UKqf^Jb{DHYQGBLbvZri z+vPZxtVKfPY-6|_X09$%lB}n=Z|O}~_H+w?aqdlAw}Dg>oad5B=CX1rBtx@t2Wcmt zml&x8E(R17Uc6P9q9I5Hfgdm!&otP;79rIDln+GaR zJL8jA-X`St-PgTEDwaSUzG8W7^r`AnOsVCnfX9wA@ANdP127=2;lTh@(yrh#+-DgA zpm1RwI6dj5eMX5xkbYcaJo?qxix%Mj0KiWbK@=UFbJ+AXp=~DABpjdij&W9(V72!& z?S@nZ$ZYa4+O};;%7B?9l+F};m=fQ`2? z3cZ2v+O#cXXk4q0lW5M~$FQvHHjY_v*kAy{nzy({<%50haCrS|t_>Vj8okboPhic7 z816%~u06QLbQTho+M$pfckB39C1)c>;DhEyPe6Ot&Yw7CJxrO&&ls+iE{7zM+-b7! zY(8+uZ(c@gsIt2e6;u2p0~``Zt#CSR-U|)D9ycD9(^+5O;Hb#wCy;2lDmFViNeBzm z1Z0wO1zNVBF@^w>w=6TjIj(xz=@JpOlmX}uKZRVjx(aYM`-9fAax_T4F%5-Q$sC;X zn$Eb&q1l1YMX3$(F`(YtpLR$cFZuHqGkO45LQQI1!qK?i0V1WJ$S7Pc)n0ZKp%S|vXskEvoF98 zTo1;y?Sr!9U;;9Is->VfUUsQGan`Oxg@I9%yS-s5t7deta=ckYca8}lvilMCj)OaGP&E5R{-QNG%}Np z4({3S^rx&Oe54({$6Df)+>%FpA#H#Xk@YQ;ka64mDTyg=HjTZ?^r>Y+%0l3Ci~-1{ z5ClRIhUV$VTE(NOM#Tnj&9RecISY?kfXquTW#kizvexARCExwi+M;-5#?1BT*~V+9 zLl+x8H!FZcj2!WfKb<|4ToMQ&v)lElg}zwu7~|%}J{Suq2nfbAj!CUlZc9!(l-dQ$ zF4Cs}0tOG@YE-r=#~>avlh^523?IKc)P~2*zMhz=1+Zq$NY6u>*O0-bt1>2w5w>y( z8RM{~3sE7A{JW3gT2Ba)KrxIQWPe&oVaq88Xa|xy8qr3}Ln&Rdn%k9SRwq46D99ev zOKpReVpwnoV_E?+epblr!kcw!LwScBuLnPO@T{8Imle>@*ti9c&5Vpx#$ymxC#C_w z{#6|0f)rx_V3X42d1Y4;ndY@)AzT0*yTW|P5%HGF5gVnLtGF>)5$&RKeP}@CxOZ0 zx}xv7gQd->xVN7%Jn@Wzc%_>65%=WuQ;vVavDSA3A%M*iT(-;;&tfXMk~W4Nv{|^g z+_)LR?mcTh;^mP>I+KEPie#4sg7bq^q$RVvC+qy`Iqn};<<`SyxL`*(B=f~pdCQPM zC3*u^3i*wKMsZM>FoUldtfeyE8Si4hn8~}IPfU)K838%={u5G&ak!8PKBkz^6-PBHbV7BZ})@Cdl9L}Adq-qMW=)j(o zZaa>7rxe|fOmIao@)Q)>PkLM2&evna0J7?so4ow+O~2ZA%F~d z6|D?}4dVcG#&KNiEwuv%;PuUFLn%033aHOK=B4d(rYz`fW;@iL2V5;_L$o*pr(Q=& z!_#N=1&>YzZCj}-!S`+?V~(P^DmQbkl1&!8hC(tnhxwNWty>WZ8+OyP0D^lOiWu=A zW;q~Z2cW9~RmySqPSKOrux{^D(xn?{iGr%J0JaVR2fq|FOUYawoO{&E8n9jDs0WO5 z%`cVaE*KIyiFL&~cJiqWGe;}|`S zUxVg2Dh4}_m0gr8k{c(W^{2tZk~;d1wXZ|YlS^}RPsacp`IXEAsZdl4s&RaP-+ni>x=CRQUSy=3CCMwFz6-ur#*EQStF6kiD1kdvb<7OBh z++*In-pcDG9hfg-E3fd=ytfj>fH`F)O8qO&#P*YC)C(m zO|9w#8;gelwyEkpDo+5~X!<9`h_sy}Q%zFAN}R@Ws@dsYJ@Mzn2TjpDJ*8V*Y;?CW zuuP4+k39Ob0iv+rMv46yh* z)?%$3NfM}4{u7G%N5uNE)u)hpZtW86VdFa>mCDk_aAv&hP*I#k@It#ex9%GUBM%CWL4^2_qEKVJ3X9})aN z3{Xro_YZLL5Z+{1(3_8wdFt<6`!5crPJ$!h|^5&}cH&g#58 zlXj7&F#0Nq;^_7<)1zS?a{bi4A~Vd0yn73Cg0 z@O*aSQ*CM)Kf8{CzKai58#H{iYE>q$6TzapxVw{S`J3LfZ#1imNR{w9bgfgQTH0JJ zh_0lQ@|8SOHB0uIKPUl!xXvrKr0&#Wk2W->z9w@<>&zQ+0Uc`Ymdz}iM+c^Av>I`m zemMkZfnCRiJS8lyvJ^NyF;O=dqM)TI9Y@1&3O=)}!D?e*WII*ykUi_a_~Y;^RQPk@ z+kJla28&IV)-{?!K4v`t?_DRtJrL=7&8*W#xCC-|CccRMw)`1$@fY@L(66i)%%4fU zN#w(FS8mdA?O#7RIag77CVexbY%e<0{MNm`hTR782c~OJ!(JxQ?5-nwh?ZgxOyeWg zurPdHrt>g%J1*+GD{vTcH8j?g|U>NdGO7@-}pC#oh9vrL6GEOkL_CLWbQ^1}r zF`YKz3oS)XJej}+z3b`td^@J=npfEGE#sErB-*ls0E}0_{vG|E{6;PzyzuF0=0Y7n zJfHGw>wkq8o-($(w$nUEVD?uE05YlKLEo=>`P@}G!E-`aN6_KrPP}bWJhDA2!5V$m zjU2Z7*xT>GEP7X__&WVRwV}u?H{gsK%!@>gErh#BS0J2>l4|dQwYW5&6n~|lYe*61 zc_hHv6cd{96eBv3Pr5s1xp30-K0*Ds{x4hj@5fgD8`Cb@TWtZ^B`J`6=KJ4Va4Y9a zc-7_H7j`&ojPqOmGq@Vxj(kCSr0lrVCzTn8EUd(iew9``P(UQ$XOKJB=#heiR98dg z=u^CIFk6MdE4X7JLEsv+umO-aeF>?dhht-5Voge@OXOrJ>bU7yPhAg6FtX7Uq==Zo zDmLSu^+s!Kh7F8$1PmJ5bSl7S<>`aov##R~Ng#v$-1Ml2W<;TNrsq9z4hd|K-9=~W z3PX-gFfwslcQIhG9Wj!70a;gDyuwQs0AMaFqAOc7&yv$a0`4Vk_~4VCD-!w)l6#&x zuAbr^U!sc2xVGTuj>5X)tZ>df&Qk6KRmLkC`q(R$UzfH&!n(`bP7ene@7}R4ZApB7 zHPp2^UXJHAE>(ye{uLRQZyoCG>@1SxV{K7khRGQ@#TM=)(?gZ;`A{%>Vy8*E20UavaGqJR8#z!~@tvAmI$8Ip*rn6S(dj%j8Io(nMSdG}mdsMbG zwa~Qh^ENPf=M@nrXE+Vl*wT#jBLlW++Z%IXcXr7YJjJ9{aeD`UQWD*phH zv<~DRqrD-Apx_?-cC7h}O52{iup-Dy4cOz3^(2vQLA7y=<0Gw23aK0^J%@8v1Ykr0 zoOdAq0807UJ#{|DCV!R@RfbgOoEl?Bid&LbIm>kwc1StKNM13X^#rScppn$@d(<^^ zRZ_-Xzzkb-g?Yyuiq4$}@G0PTt(jp1Y1+Y$8OB9Ynj#&E8+q@|ZyijfE3HgfqHX63 zxZ|d37K!6vNEjV!N<{`nJp8TLaZ&jtw+xe>;=5rd%Vc?V756E_8S_3+AD*VPppExn zXC8o3+SvjbhSl}&T9$UBAm9LfDavLrhjIvD#6ij)PB3cB7Rtc=dY+YOVGIhcTrVJ= zm2L}087GCtKt1axE>vq6tCWfaDpYJeI^KOR@s3?ZOj{|A6je6%MMujcCBM|pq09mm3N#4+B3mDjZ7Td zh2&?U1Emg%xoi=+dBFWCyGGRryNdEN+P9C94?@CRg=O3b-Oejk(&PZS`Ac!lbLltD zoSny@J!#>mt*NglkK=}jEXKBrM_K7Lhjn@AvMlUiPJ&&v2-p-ppE_hb%)jA!39 zsc&$lfZy|x#zucy+6^rZd8EzDORz|0%C11l9Fbc#mv~*EHv@uCYUJ&%)d^ly9tZ=a zYFk~ZbS&U5G1UENy_v|<*z6(Hn3V;C5(3~J4R7iCi@Z#O=TpHWsINbc;EXWZ2>{_c zR>iKMVZ#!<@-x@zQ#shu5xMA*>Zlw2VP*hum2m2askvVgcV6|v+G-8*vk)+O_p6Jj z61jE&=K%MvXIG)rsdYzKbz_6Q_Xaos)MHQ|bhD@($9m?H{{ZG2akr@4dQ;n1BY(F! z>?;>lJ&uUbwye<9ulF-<8CK-(>T93W@77rg;}Nb7(0MeM8jzA)mH~G1$DEqQy}S$@ zkC(pT*06PLp|`OqY8On%Km|j8jdD8F?96gVAm`;ZuX%O1cFdOk@NO%aylv-m;m@h# z6e^mM>F8l{V56@zyP`H50M@6$EePIo-|k!a`TjDSxh zRZ|f`#{|}#(G|uA2CGQ{$RPKw)hisfxluE?o^U$iuEu!C3f)dXx#-ZNI>QrO$u z)}_AX#y2OoQMdz>o}(3|3}yIb`7_RHtP9_7UbsCgM$|R{UUB-?4ocb`5rQql8QNQE zVmZcZQU%_JTYjYQgd;1>k2>JO)C+50gCP<3;h{{VN|k?hZyS+GVnkF9mW zIs;1D7}45@+l-dv5PMY7TL7DazDGqJPrYfrn5qHJGuH<-EOv<)&gO0Ff=^oPj1{yw z-(wZ6j|0p&$5-UkK0z2_eq}iC*Veij?a;hsc{uE9yw<9#f)3z&n%kDgGmiH%Je4Xz z1a9Q8^{D)j3S~y+CkNWK8VL6StQa1dJu0k_5O{uZ&U;bI(8kTPJwR`|a=`QhwGyYw zdiCTJ$F*78v4AHXy{WRqBmk)Zp0y2Fn;7?TZ99nhk5FqaXFJFDK+ZANwl0yQ8Qe+E z10L0jbPi7A&Uy;w*xPhQ)3GIY1t;FA&u~E_KQFFo#>0RLVD_rBL&*$BUX`RCT zUS^t6NaSF9cdHS{yr{!x*k?6nJ7#dqbA=zBV(NL1 z>Xt@hSOyD|w4R63n)bm#!Q(uiTHl7)%0BKm2cBvPtiWKZpd90+VeM1VoVBpJXkope z1FjD4N3}%Ie9R9x;8x6#1blFKKS5Fbm}C-04;4|Ums7*7Nv2}S30=5eIvgAfQ+%Zg z1_}1fYBUNTrqPe$=~0_-ISfh9euA}g(DUU?%zJa5nFqFMf*i;&JNB&^pvxTaJ!&aX zfB?exrFWpcpK`jP>ZI{jRauyvg(TzK6&fe6I6ZS$BV01B0playp2)Z+iEY7BqaAvR z(zYZ5yLiaQxT+}@7+_$N*0inA?H?!^;2xEZ$5b@f*0cw^F|=iCHJ+fWi;Bjp{9PSHC2tIi202CQ4FqwR1HpvNMl zMBN^9oF1J&3Q>c(-3hesb1MLa3XLL@)EdjahkCIpzMFg3MR2FfWB?yvFy1mmSpxKF*2h8;z0%k#T&k8I|%ZURU0?I%2QT?#F96sHtr$THa( z4n_r4Dr4+elY{G5r;BL-4X4tokT6{Qq;&(mbwyb4aT3167twmMg&T?_CL99z5q_{i^Pe4cG+oPg>fuW4~}a7u7^=$aftwbmB$K4AlG^D{`_0&dzrad*abZww-tk&}O|~!u~w)UyUc!Kd`N( zeOls1FAQo%+#XANSIf_pLzV7&Ri!CZZOb z69dx~*=XKBiYT2WxOKGy7-sM7>0FnKJUMe@yN@$bi+oq=tTHPE} z>Bco3IsBRQSHiD{x|WeVQr@WYQAXW}0nd8&Jzmnnb84p9T3x{v*nGeqPc`HI3i!=- zG|0{Ektoa9;j7TRIjGI9TthK(pyaG=*9(5n#E%zg>#-s^rMyD!6cr?D}83_NM$SE0OVq@=e9h^GDdPev0cU8uyoy( z36G&aTF1N7cP`Eg4x}~!_O6Jek20)v(Bp0+VqCLyKXd`e`qowC860vm&$V?E*tAG- zj1l~KkMW0TZvlbkoV0^!sYHZ<#yxOzP{v5l)}9!t@>%7fV`2~n_@*c(_kED>r=%ET%LnD&vETq5!i#Y?8ztct1(#Z8R|*>D>mC^ zkzGniif~GvIK@jGZJxNu#xq(`*k!rOJu-xKcML$N*HrBN#XU4WQujlULr?_l$TfgOi@7 zm}A3W-~o<(Ysi#reH?Y%&XPA}RaOdm=dDzDlZ6U0asea0d)Bn}`{Ya@9YG}YH3ZQN zg}^-I0(q?UIvQ3orHO*$s*lvw8--FwCzGB#R;|+R&K!-Sx2;i?L%&YhBONPMqtxb= z?9rCVx_mLf9COm6XpAQWmL&A66EE)w+A*9FO&Id|{{UOtK9$ve%X5*dj<$_qi*oam zoB_>g+9ETn1!4ilI#n1$DH!HG+0T4dy`!qixCdw(y}ufl?&>J6XJX~8r3VVNz zUKZs}K3sGItuits17HUu<{dbzk*H9+Q~)qJ?eAQ!S4UckLq)J|cJtUCGf|sy96{HN zpI-H3$AxX@b_m7~YOiuKqJQ1Fi)dzhJX*&&)PdqYiIP5^hK`zm>5?>(l1~6%@D@eo)k7LC)O}oe4 zV~z+tD|Bb2uE~pUjAghTx_&iC`TY>`;PUFUL$gOWUb1btk=QKz8jbf!NgukYk?R>q^yComD&XK^tTMSp}1hM}KchmJU?;W4}!Hsc|)raZ)ccc6sBF1_fJ*6t5~nV++S3nQIELS0sk} zz#en>RrsT3P+1s%4oK_oU2sY47jJU9TkK^R1@q9Jl{|K&le_{zBxjnr6kC{J1}CTk z^{JwSIr9-pl6VAHZ8)c(-iWN0XjC$&Rz3QRQa!3C%GeSffDUVBI{`Eyc8#RUYc**NsecUSG422yAd)5B{1$-p&SH!(h zE&My7UFvL3!qFiL2<}+$1!VPC1yM!&%=4F0SObyB019!kVmB(}6V|@c@NfJRk6V&B zu6$wQD@_(Q$!j}d=gJ4va4XgH@7N#VC5j-ry4N)~AZ5(LF`r;NR}ArVX=y1M!OB}j zv-4(4Mv&owByva40s*tSmFygfDbLlRB%E3s_Yi{BP`yzHBypmQ_`nO6IVkL z*tA5PZqM+JhOQ;CRRxr&BO{8kw&@TU9JfqolUCzbP*s@Z_U($~l&)N;+eT-tsAm{o zsP(Akk1ox;fCoS`&uY;St2Ws%FnZ^jt2AJ62Os@@^!%tvb3)a6R&mZ7PTAjl7Zdt)7as;t(Wu~j(fir(@GQWbjShV-g4Tk-|~W7LYM zRJ+t2Bx0;%Yi9+!FR80A+qn!0QPU&6IiV2D2s=)C8h%s~z>?>jR??Q58OAE(qWk4Q zKF2kqV*wzjDoE~sol}4olXEFRc^p=~ji_{GU>gTM#*=o`yIk1Qtu`nG9Fd%g-Lkjb z6fqq4AI`JvVh#WpKPdy80BcgqbRd>bnQ%8_h83DgyQ01V{k&lYcZ`FA2+nHsF@kVI zV0Il1MFdQNTO;P_fmb#d_7XOdaB_J1(aQRpMx;v%YRteMK_C!MPs*%D4hohGs&F>0 zKRSpIGEg56~Mo-Rh*rs}Ynuden_3)3rh4uJ1~L*p?Ur zo=?)cqfHr9#Wb!}kQ^VIB%j8qPtmcH@9S44ak-C7_o{O-DYd@n&q~!o$nt996^in{ zR08*_rz@Ht*F$6!0vSshrMcIZ84JP|i$h z7oVkfUIf*l({=l4r8^*286nY!s$Q@*z(f2Ncs@_=qGPkf@J?-x`7`&yOhu-4{ zZ|7WwhpI#HXT(-M2C-Nt@dmKUbqT=B#Tn!-duP3R7m2y=@$yN4qqnVc+I@uAlCDN0UUB?}E2H>;%J-K}3G)kXVbJZ)Y99!pzY;Q` zP^+AE{Oie+(D$PrV_M5du!#-L>`^nkMOCEezTkS-i}>5Yu-$6He<9u{a;l+nK+hHI zx`w@DqFpRf0e09WfLw&Ga(!YO!2bZdMm)A#qF1Aae(tB4ijAdYc^%Kf>sxhqk|aT% zbM*DEE%9ZQu8_v1+*^wI)5Ln*v)q)Bk{FZ8$jy2R zQfouc<&8aL$gg#r%zU!Qfc8DBHeGR|I1ETU44TQhzua9&&%I&Id}MV19faWL`PR{g zwCHlqx{l2Dzl7ctzwvd`P8I&i4-Y3npG?=gXdVudEkY-Vouq}qU)H>H!v6rY6|cgd z2>Yo>M{R6}oItARq zDazqXw?UkAu2WOiVDTlR*vP_5V0dCVW7yZ0{?s?tSH2e>Hq+%;H4AB%%u)B94^S)5 zej9vFmr{};B+^SEATVRW`d2QkbH(V-S`^^brnNoCz@HCq#(lblBjM*<+lHxr|_r+*MzMUuV9tigrL8oRX*K&#A7L#-b2NSkDSa2a{Dalp9-| z6Kz#dJjYm_u4eu4FOKHC$Kzh1C7;8mMPDx7@^p;@E_X0Jv*}*5saqwjmzuzg^=^cF zSI=J^wTU$?Lrv11neDAm&H_GEGY&ZQrIu6nn55rAvDGAAjAa7Y~Dtt5Gr9z4R{y{qpUPebP^tu-WyH4(AK!a4c-)xjAU z2dNzGJXC23+DmO!Cxy*jf*kyyj&Le@7UyDi(4%u|r!3f0+@60Lp>b?T+!?ob9f$t_ zTDlLiorQTqI4XIl-o=bzytByeI5lyN-i9t+4n|ubQI6D*aNL}eS$FmxVK0!M-u~@& zQ`su=zc68d02PmMVYvtifH^y=;~u9SN$7BQ)(oVwJu-S%cYk75 z7k#({=NQ2s{c7i~tcoAFFiuL3O4>0@0Q^7$zu=TkM()>Y9;B%1PjgszcDo!8yTKX7 zc6WAT3<=$n!kWdrvo22nHts7$=5S3U&MxxY6+JoXFgbr+7q#n4UXImqL@;bPMus$2c{iZ2)1CGEXBNs@yg*ZouFkySe-;O4~~y0#oPiJ626IXp|$d zGBA)2ls9TwY|9q;aqZf=7&P;RY+xrQt$${r>%+Gok4nN`XG~q5hZQBFx5`rouQgSy zvc!@Ge(4>nues9l&Bj|DPfE?X)5Kv^F5{o2Lf=DV)6n#lMJx+rBckJv#-frl=Hn{J zMS}Q^WXZJeJxHu=uBP{9huSiP1R;Rw z>+e;jknPK5heEYZOO_-O!47a#pURoNfh^21$i{ja5@T|G2(Jn{H#lL)6^>*Cft|0O zok6Qta;`{kn4Yzlb1%z)M%-kNdg-TRjyjFDI^sY<-MFwljZ~iSV2Fqd>)x3q!hlEs zG3%3A(_M@M>(aYaE^|7L)-+?eGGLyg)1juxI_@95W<5ykwUU>GT#etmCj-)~OK`4$ zu0rFbZxv%FI~N{NB8=lW>JL*)Sy@2{cHrQH-l!$cByZwV>UvX!%W}EO`jJ}M*yfb( z?r8}DY;+y6YEe56lpUwmvKn(2`$x7#Iyv_S^(XPB-$6NP=4~y{BoZ4vYH03DmgH?t zfL1z5xZsjUcInMqhH}R_7$X@KjGHxG&Dbt-!w=MPjw({Vbg|trMfpHpxKo2y1@2?qWXU7~#Cumc4bsR6-GD#5YEyJd zsSHUx;dol}=dP!t30WNgy1-Wh=OFS;HLkfNzV4fVtdV|FAH9OZZ}~LwcFGU?Lv961 ziz-q^u#mG5NyGG@OB#We`GDjo&1XkBDgi7AIQPe`J~;%8dEcCZ_{Kn~IgoyqlI(_ISM}RdWx+yWl@R2Q;(H;)g^5K>SSEJrF^sVHhOzj zJ;Z#-8IA@B2fcI?p+mcCV2-}^m3?j*uvb3dR`8NqBL@rW%u9%hHWZfk?TXF0;FiZD zfN{lb-Nm;(Fgn&%+kh~71I|0v>Q`W?-B`q)Hpx8?U`158jFmX$vz}{Cal80UvhS ztU@Uh1D=GAl~URun|CP(8T!_}ss>(E9P}htEar`42Xghg0kn+gr?zWa6=f@u17oNa zp8!z1dFQ4(R)j0T1&eOyK9$DzyGK%1*uMxWyfT3;m ziMESI$)Vj4QUK05&ow+CxmH26zm3^*_q5+QF6t6SQHDI}b{<^=CrA zE`_UDisK5qdNJwwR-L`D1Lhq#;7X zIKcbdb60HbK3;mBgE+=2ac&m_A&BUxG7Vpb9k?tG?@V|8mC%}q+ij5jn{)#h3xkk* zQ{%MH%DCH%f$oZgz8UCx|5#=ziq9{$ytJ-*B% z5g5nI#den#8&%{Z+~9T3eAXVJV;eU;58cIMQcGizG}*~qTOG)%069E$&%H^fczWwr zy-QnUGRZO78b+myx26Sc*yuBBSCLvlZ6uS(L};KS&1&S z`4loMDBElZ=s()`u4MV7qKvE8jVQLyDgB@S0AR~`Y@pUWex-w|v|3^u61_(A+v#6d z#h_{44bs<5(X{)0Gf!2*!)X{nk8|F;VxS7-yk~y4a1zG@zrACauVqK`JbDwATFaQ@ zb({2yA#)}_yI4)ADI2)ZR&e)*i0Pt(zs&b}& zg@>aB9a-4TsV>2>@^SZlD7V$2WmI-3#F5WR;%+s+^k>@W#IVGG=U^RdVC^kNhD`FWX0y=;|F?jf^r6;A_s{*RHha?q;6j?NO^` zQZ>UM&rf>MkHqaVx!OQ0@-aPrl~I$io5QE0cPsec;D5uPhy_34An?_xx++jz*^eeb z52;?JzH{-%>>Y1sGf&{Z614G2|Ppa+G7X+wAu~a zncwT=xBkj&*`bT4U)_33vDXa3OWm%t?!Qyy*?vX?Y z1J{FAqFt&+Fh@h{QV7GoK{)pnN!hC(Jz6~1L|+G0_C~;!;(EJ2#>_cb;%Q>F8@j^w3(G_K^Ta zNl-W>)~%o{512=m4$vQvN3};|Zm;vR5u6>k>sr?E7GIbzm`@9i-qlFwl6#@i0;O}g zmjf(4eJb6mvxQa39Fv~h(p$mhuq>nyaDByFjTxV40EFd=oaEN(E$WQ<`B0%4N=|ux zrN5Ovh_DG#>W5w&O6qs2SUVZa=NGnCn3oh$EHnjTCAa^Wl1cqEOMBFf0knX)n%qi3G=5tQT}gSn_=mBTQ` zan#gJxP#6ITmxGz5zSW;+ZI_c#Db(}t!Lc6=G~GGJ$R{Skg&iwC!xhsoSp`A>yGtp z98;}*2t(~|a7TKU)_gfU{{V$ll!hQ=4`E9TgMplL)7GJ@oc`6{Lu%RZ!~#xx9@VRF zaur5-^{#p;Laq-^pGwoTljho5uxi&eg-+(}yo!yL{A)%j{Kwo_k~;ChtO#c1esDqK z+O*)gJC~4hE1FBG*9pCjhfkZ#UBrdwpzB(&-T7l(ABAc` zbO|8ie`asxQDke)bz|8+5o*6_@ePfw&EDu5;Z&eFPPO;n#vcz_UTHc-)yGzykP~UyU8fp zinuSl@1%SD70!5Th6*4)x@E-;A6SvD6P1O_=`ahzA6&O?$SRtRx7eOio82 z^vztc)Ggz@k91MUr;gQx2so#h!WKABP&pitm=v(#m^_ggl^LN53`g zfAG&g5xh4Aymr}gbH@_DznskX{k|7!r00RTkzSg+A)29*7 zaETyS`4n^m>s&Rl6r9?WO=$Z{ZAmlWuO9pw@m-g~=C_kjlJ%IZw>z>}sOX}+Q^8uS z+6JhRHt3{{z~r6>(!W%^e`#f^X#PdKD6vR4Fj1WPX1_SUX%7vr!|#Z&T*n!T+VPG+ zA1GiwYnzr1Jdr(!;^w5L-;wr5?Ee7cBXes4Z`&q5@o&9~_N`Y?k5c{XfUHhK9)i9i z{hs_{w{03_Re{Z0NC^yuc)#-=t)~vT7uur!vaPKJvcROEyEUIz~?7DKPqj!MT2~~JP@Su^sNY?U`_|$ z`EUnMrB$>o&ibx|P}-iag;T&EPW3ZepL2nX59wCzpc3JqiBEEKfmUOTv}K6M?UB~B zidz};X?Anh8hWuINJ4S9BcG*Y-096Ak#=Vu_14R5WNs5Edgq)9u5{K{8$gT>Mk3_V2Jd58G&_`Stk{A$=e=!O*a3n-2N>e5 zSy(X~rac91S?Q#($jXkZS?cs1%6583$_n90HKw{6BV=KKUT`@TwPmKFgOz4)O!TcN ztW>xJG84&F>s4zq4qbF?f+HDRXC!k{!=^Se7mk<%y>>C_zGyps>afokCay=LnFj!m zY+wp3n44!KZ=xPJJBB{#&TC%VOaS9b{5|p=!+Z5U)$0=INrz&p8CQZwBC{`aHDcU>xMPLred=6p zXr}r*(xq8R++*0Sjy|7SWRj`e&Vf!q$T`J6TWnONrJ0j+2`JeL zRE%@m-nsom0rH*_Mo?NI~dY`_Pe-&|C391~X5=B`fG#z_<_ zp~tUId(zy6T(DxLxgScc=M1P`8~fPlP}R|e`kGKM!N==b79^`5;a;OPk!w(10y|8y-Pf~8*#=!1e&FG#7M^&QZbQK zlf8v0b}C%m?_vid9epbR7Xe+taGaQVXg-ErtEtw%$Ry>%{ zhO%DdWRrjaq>gdP-Jeh^MdoA7_cs#cGXkVGe@c_>ISi+v<381kYT03ff;~-NSqB4? z#a_&t+`Z>-Ae`iN=9MI8^%>yRWFvNeg!k`OA`OBw)}gV-eG8DW@7J~}#8T~EIQI0a zXyq6yboK34VmatZ9V<4wol#v)B#=2=;Xv*^YSc5~$2g|11XrV>{* zgoW`N<&<>5{VK?l_ln86Pf&51%UKvNEJ*LoTS-U&oa4A11w&5S7b_ZVD-F2F1-@cO zTAn5j*ibus%luVB7Y{Oq49a;Gb|hs5m*r9U=Cf&AaNU+d?@R(02dMR;WWm@8*u-NT z^Gm-ms3i3Hi*PBx$UwjvX ze=lfHMdzB#mLq_!J@^=;mr~42n8-#2K$khobjB+ZXo`|sn(4I|h&WM}#~zi5EJfLh zsBHDe0=gqDE;g;Pg?STX07xBi=~U%Rlg>}AYE1{4CQm9#Ro%)cmd z2SeQTs}MoTWOM0Ln8rZBBy-5CFhkg!f-|2=QlDa@1h*x&fCHVtzolP=2H;c?pI+4z zFe>aja@_$O)#Zf6ka~6eYmu~Q;)&yrEDkWkjyu)cbSnFlXD0`aDhS3KAmlLb&0c{> zJe)TeQAO;uFt(+ec%M9HuN?HPSX6mqIV0|jj&oEk)6Brj1#YL(v>_o>pku%TaOa<` zbG=@M@zTtXL0V}M6IRm^@t+&dgV{06r@TwMZ2UQF?CpjmN!nSQ8-c$^AA8u8FPf0wZlw z2{{KHD&%TI0tR@`N_DzAC|s7n!KoS$tCr3W3%9*(8MB9JZqHtTjAPcNjsv%l19j)US&G=m zFzloNKJNt7?;q~*otgY|TBgy2wMGrvD=;sNDLGTm(y(r=yw>vA7F>dOuCD4eXrDhI zBe2bMKMTAhHC)*2f0KxclMqd%@2@`4( zWLrkq{nBHp9;Uvy)AcAV;zW_c40R`p`L^H2vFV?>g_F!Fw`u4PZ_=`Le~VLLr2(P? z_iDmSfv&DF?#h(;O4XyElRKgsbo`SrL ze0;oH2rZ?!2(I`n-7r6hd*-;$i5@&UmWcOu?q`xW**uQ?xIW#5dNr#&WL3^eaa3i@ zbGrWkjwF`pz@Ng~`DL9aH~d_fFn$Vp4cPVBEw!l~PM zg`kILov;Ig-oAcVX*Xnj9vc|qw>`T~@za>`JB3_)tJ6O9B;F%^Nm?kbJgxx*^%de) z{wZ15qp4O=!cGw?|usah=;7W1fcwvi41_ z1?~BsjP>iE6s5zd4Y`o<>G;q*NeogeN5=92!Rwypyo%q%%XCz%g;)%@9A>n%`}rm4 zGRO$tI30K=v5r0ls%J$QTm&&*5EGnc@hCltCo1I~Wnu*1S^RRVb;r zj1z_kJNniAA6$`OaEHrX$7VmfQz&bAob#Ht$6fJ{;Gc$mIO*}DIMO^=0sA(tploB= zCG`i_wR|PxPX_CM4?IP2p!kx-tTh{VDh7U8hi60lN_}hYD7;hV!xH(8s}tc@Yb<)rfLz0uAq>i>=<@o^{TN94tWEwdh{ixgK|^V5`xTt zZpR>xdRVgTS=VR{v}Tq+kTF*vx3+Op>{W~?8-eF1wF%iA?G)x^MI}M{9<`r5c~>j) zYE^n15%;T3ZEj`CtWmXv*bRh+ZvDk{I$RAKs^>WV@T}{5W08j2*;ou8t!&xA zRDi_gvCq=FY1y208&^6=5OJ2?03Uan+KBmcec{01f!e5P=&p(plLX`{j{WOWDE|O< z10D(EBbw}tQ#hwS#PLL}mv+j5kf*IZBl!xT01?Os2A7b)V+4#6M<0j1Iyknr6kr38 zPgC`+(z4Lyl(iim*ZZev#sK8w=~EmCh(-BxgOks;YBHsm0H*_tgH{XyTzuP?`=^T1 z=1N6Ng%V>9!{x?13ZXi*s&YwPxzA&Y)|DE6m2yh;+&fl$;kJMZ?NNi8qLkl5lhwx6 zA1p&w;p0(ZU3>_7Mj!Egr&#iHqlAJJ;VQ#-K02*mK6sjDY;hDHO z2lO=*t%U@x78-eZbR^_uHz~u3YV%uPq$6n%; zmtt0ft=kR@9AngVtr#T;%HP9|qqo+nZ3BV{BehQ-0D+(Oaal(h*ygWCjii!$U{?)G>~_KFp`!)cqbbKh)OM{~SM$^}WPm}(^{g~j zVUT%Vm^Iq?Bg5~i+!9ogox8}vtf|MD*ttzoi)`8O?br4_^_}d5t4APWbC5-Tn0zkP zEj~5;1iH}RmpZ1Jq2Sy;eg@=W_2gIQ&y0LG4wI!UDnncqOPg~`#7lC9ulo7Rp+_Y{7196iLN7f+B=w^47uf7zG-|T z=f0ORF!MfL%5%uT=DkbyqOsC%ykBGDtqORj(ELMcw+SMB-T35j-n`?()?pGzeBH4W zI|o6GbUwB4c`bFK1615Ju1U@Z>sY#^ ztd}8x?ZK;-*Bi(U^LtfoWM^k#(>2!x%RNt?sVl7+6j$>}wMRJQ*GZ+`mnyF=&#P7* zn`paoKKE~W>U8T{8~urK7(zJSI3l`VD@`;yQIp-BPLpw9$Qn!Z0YKoMPqli_g1jN4 zc$Y?+Td*|8a$7js_|7@$Tyvz} zwB~yBlD90N{ZDH6ZQx6rKN_r-^B|Tc`^E8}U@PkAu62tI9b=YeB#=bAw_?5*(R?TI zBIe;_wp)alu|n$DC9}|1(Ov?&_@SgmuwJA$(Z)n;6plBOjqvs{{U{ide2$#&Vh4jFcQOd0gsKxJxzKK#rxe;OYo(|`@9zS zh`wWmeZ^0vYvHYX#9F6}@6q)ehPm3=RLLqU3>_)Ds-u4DJo>Mb+?MS1ui5j(kEiRg zM$(AYn=Hqj-rrjL+rwJBTv)>>-q1Q{4au+0--Vh~_WISU0h~BEBc*-i;3?1cTm+UV zhz7&l*Dfute{-qyIeeOhL<>UcQX-qyUDH*N4# z1v|ELyB$Sp*~R8zh$^Fy0p!+wkMQM`xA9}4_N|*VRA54{CzD>nS{!Pvv}OBcXwG&P zayEc_)~qN%10NwAcfqQ5@ovNX*~UQ~1#CfRp+Hl?&@RP}^GHGPCT#eZ$rZH8@a?Wj@Z*ilP8@BBXer8dOn#{h^ z3hr>lf#9he{uSu+>E(B4x!{r7v9Gl3?l>TW*A=CkvS$>fspXe?Sdqy9ry1MRHO}g~ zF6DIvQ@(MU^((z39$87_8>9iLIkLHz4p&q!7cv~>WIY4VWWj0m($L^mAspO?S2T#rdrBlmmp z)vKeouqP}47ny+`j;I?EqshBD=~pAORwsZFM?=MF$!5|F zk({5_og^WK7y^3%=~+E>HB82PG<0Bu7Gd00MdqHO4i_DF4_fM)J&*E%j)3F&)^?)^ zb|8*}0FHSzHtJo^1iZchN}}fJlfsdDpmNz4iSw~e| zcCJLV)cc2IX}gpFte8?WjDD2KRdQ8^SMOwZ_Nv>-ae;tFIX&qa+{y_f<@KcWVv~wP z*<<8#tOr6X15}Kab0ZMV$jRtED{|sNxG8*r)Q+`{tV$Uif*C+6GC8Sw9(`9GQON2X zfJZ70Gm4JHm74p{X9swq+qN!^NzbU5V8kQ8IMeB=&AX-6w0V`v*& z^A(?Ehh+h>d+}PaL(F`Z{3&U$t)A-!+GNtW=Os2O!q<;}S*z!RTui`GZH2F^;@fTy#ey-OhI9-;hWmv*eSK za0sjS*bdu!)^U#;h|Whitrn=`l!%%4V;qA_kLGN5^r<|P+wsLo3_D3~K^*cb*Hb$- z$HET$u;#R3Y_=G4k6Nu93?FbjH)B_$flG1MKYF3NmZHhx8@BcXAc~zhAhvP_DX~T6 zft)iF&0JJaLmkIC91bd5MzMp~M2xZi_pqx7g#f~2wm1~^iM+65z2}IBg)~M-vhlzINd#qFbp&)!6E#Ex@}0nRHf z=FqQU$m-S7%?M$%fX6(X)_ue|AQ8atR;QrcTGbr1)@2y8jlhiNqI*B$Dg$FGJ!^92 z-)RFOf&T9~s13LX$^Zq1ao)9azT;H%IrgI`LQS)W$NH<%V|ED?|^F>%6xBuNeF){kn#6`u4}xxtq49Qe6!Q2bPL{ zQU-DlU#P2A#~Vu$OLfgujx3$3GoR9~MmG{%kjIafvv$8Rs=Lt_V~Rc7iOvpjo_?Q7 zwPpLik(lo5(*TN~vA*w>fma;!^HuvWBiX!^AMG6A)~SMe*yt@|Q-&BDdB>%1S)p}8 zq@g1pUwX>ZSygcRN#h)ITb7o}lB|K*SPWws2Q}3V45aU=ZXCwT&gDNY5A&&*7T8xf z3xJh1x+{>ONj-Dx?N$;KwSXAKbj4~b+jdfd7Av@N$DtLX(U{a7#4sR|De+kp$^h9U zeFaW!%0@~MM^o=xrFKh_ute^KPDW2VNXa6c#9%g9s094oGHMAC;4IQB?!el>@IIoe zYMP9CJ&Rn;nIj)6#GP5Y3{q~|mXcetwC=iwr){So2hCyTg+DTZQ`=Z z8+6zc8~_e0hVW{sf?Y~b@ai+M4I7MYjC2FiscYKnM{y?BK4H$?rvtAw^LWX+^W|G3 zfz-aJ>h-@I{gV^#T=t=_ zyeM~l_2>cbSkt1>&gATOmtHM+6lr&=2Xg1qvakGFF5_ldofWxcC+0Qh&*G+* zRz{O{>|`J1Syy-KBk;vWPtBf1cfu`cYa^OX8P|_hV2UW#!s;0(uqxlsbDgxp9?_g>}4=&cL%VImaW~v@ASaQC1_e z6(A|w&wqO2-t02%L5EebT&R{{XFCMPYp~X#2mfQcZVSJ^b@cmv5AE0X-|stsY5S_Av8vR=Kwd z$qmK=k&}=+AE~aYIEAr|7tE3e1Lkw<)DOm~>CnY<8njWBz#Gf=KDE@trIxnd7kFF^ zsx!CVxi1#2-1MVPT?H1RGNLxvMgSenZrdu`B7|jApW_?>ilZIBm2tfglp_JU^!BY7 z^!V?so6crz9zZxF6h&LzGOf&u)@ZZChZq?J$QT&*t5Mp_XJZ>-c`B#x3X@B-S#IIC z$(^K+a)2^G9YLuuYww>EcrtshsRzKpTK@PyV53BOL>^j zXgyT&Mr+WaRm$|_wk8!hUY$w^-~d?gI^wL2iohKD@l@d;$e<5h)oB>8A1HIvX&5!p zd+2th)N{kUh51yjI3028QT?6M8H}E|Bcb=JgUjEv42*TfH5U$d6Uv_1#UzR|Wwgu4 zGtY7=Hnt22EJ-5@4?~K-BCvENvVV&_W{`p9jmkp~fdUpmwc0 zC>Brqpn@^f|XBR#ZGL{(6Iq98}B>(%aQrZ$rt?)}f6G z2PKcpGm(mL0yQLtagIUiD%7>HHW129x}!G(oaen|UCh!WVL{2j=lRv!jFkm?W0f6E zXI(P5Dn=OdkZSHLYI9n+#!C`0`3mE8MFcs{bAerz_%FN>kVnmv&2tx)-eVMX+qaB& z_NwM7D;%Vjrbb|(mM0YGY>KJMZO0rA^}4nN!tx75E9+ZXqq1_%m%Or zr#rjVX0jV_2JU|DIr`UJnpd2woS_4t6sJo7AD1BTGAggp3Bj|7I(8&{t|SAU-j!Bc zCQvYNap_&t=!_d{50v_hel=3(N#08NWCi1=S}xHJHrAaEQ2HUs)F1`04i96#wkiJ&c#PVI7PH{B(UH&K~pYsfyrL|jc!=@Uhh+o{hbRi z843kr$!R5|j*@~f*dqrco@(qrH1~g|x)bBv+FjLXX}y zaseFInfMP;(d_lhFte$d*a5g>2fl04ek|(IX?`2Jx3rrVGa-*?Z@ZEAt7>%lb$gsu z>q<`D4>s|C#UVe4u6&Dyk~MO?p0&(sI^XumHz>-JDacXCz!l`z@jjn&+vna}Il;|s z=(jSgd*W@_80%YgIjqll@F&L${X4_M!n&M`E|`*e){;tcZP@iup4IAF2a5#KKuIiw z0!MGHd~2v*7A=v;uR!>B@d!9J^GL}u4of!&74!L?7uqE_pI4LPRVVCa)am~KX)74+ zwZnA~VzMrJa1JZy$!#P?B>r{oemwCkI{l&-i>-09`IvM^xQFD&2Wo<;uv zWlvh|yeaWlSkZ1Hk|sD_17`;v>b;kOqQAIU%ZR}1xb^&N-#!_B&6*ys_E){RW?2T} z6l5-Y*ELyFdPwV`3{+j<^cnEm<7!W-!xiKX31U|b-`c*8(7aD?s-j68O#=na{-FC- zex3U&>7F07No=JOstw`ZFi$n9r}#_kb2JEyo?bFYU!~CD|lz& zrT+kj<-3aRH;tpr3IJ$CGla|KUfV3T#(Js}>kV|uc_}9H_zBZ4- z9u{fAU$kur7(>y>z&P};Kk%NLCZBKitA+B>er9vn59eHei2O${iS=0}az(lwkfNM! z9;8;-Oh49V%=!^5L#I!cJC%~+Y3`<)=2JY=67J`zbJW(ogulE1$(F`I$4bhyw^VF` z*atlGirKPN$Rp(nI%AS+?V)6>j}fk?L28P?lAtN%@yR~b&{@R$)v=TVCnK-5VOftb z0M1wribqd+=`7X3`J502M;vClW2!T|>TT(ff^tYiJb=fF+O>rzMc{1*A&IQ5FamHJ zEWa_wTIj76q*IWqoMRnxTI?L}p!BmSp-Ehye${dtVIL?_(bu3IDp;W|QTBvU&Iso< zWe_3U+dv%gS`Evw1Q3NQ@}4@MTF{QsB-+6cp2Ix-X<>u+oQ??@0#OBUy~bc<{!i{)H^a1BQGN_V+H-!=nQvC!pmIIGPyaB?@O z91iu5b*6$jBy?_i{x#m-*bFxV*bXZq>d57HF7DYqt6N#+c=f0AbVLG_7ri%z5SBmJ>J!7~|fk&7y-M z;45t#w=aMH09ASo#*_%#oveB3S<>ljnIHlI#~ju(O2bUz{{XUflaL5JhNgdKsEoR( z+IR=jyBTazL*aIgr`**;rkL<{s0@Cf8s}=}iCbe5(!{DU0Z!%U#~jx~W2fy2g#a$eOLpcQCFyl3Y?tQDMoNmkD z%0l2_y9PWOf1aogGF!OLYA8YTH>+fkwNQ(R>T!+$HK}til*b!gMt45?1me8LxD}&b_ z^zj=u*3L&9`c|c#vgaiO6WI3rYZmMgT-uJdwM1n^Qb6aP^{oZD3^F)8jQY`OYRbg( zjB}2)sF9-$gkUi1+)`H5e5EqG0UI-(;8Xm+F+eC=9B^^@)SQM3#y536eT7FLc@{!c z9jET&HC(!A9L>E-Q%0bU!16_B+R^U;B!=N}k zR9xN=B1o_}8LQAh_(DKX4nZEi^(1krAPj9@q*cX$kho)xGg(SjH&kC?k=l{Y0XfMv zVmov!dgB~suEPmu8%NORtx7GvT$8lty<+8ct2T`%bVUq*FvswZ^O}E@rO=$Q>UN$7 zwLC(_v4FVlJJpEbEOIu1>+4+ZO>B1iHGPQI1t&i?SEy={A_~~shCM;6NSg$B%N~T| zIjNR4B0a~4?s}XW=W~5dn9}OX=>mi#FX8?*VIXpTVaOT4^s7r28wa-_RfsLw$t471 z9YrTS4OKl#(Z&}ej4xdFs|jzl*K2QK$pBQQ1y|bKwh0;O#ap%mW3d5T9jl(T9PZZ% zQ@(@k;6T6wl2?qJl$3xcPBJ<0W}@Tp)FDuOe(p0yI4p5g4+ zkwO%O+mKHgsYSaZGdh(GlA@%JEud}(@aY_f)6bCJOuR!zhvLz98%GgqNDXvdN;#OzR|zerDq9wGk~B1diJQS&@sj_j@ZsCLfQgcWaI#PclW51v(Xb3ec3&tmwc7M>6+Jq z6*(+as}2u5nynh5;~=w-m;udb+b&BkR1D+{S3T~o&bX^e!`9sGP@gVwobiF)tt2d} zRY?juXQ?$ZD((4sVmVSXx~#3cJ7;c31LhqoK4#8@mGmHrCT9+KY;--VQqln+1dQ{X z;+Y+|bqIF0e>&2&v@!+xh|7J`+O2D1T5GAJX&SNIIw#A=sr9aw&MczmB#t?(TU&7- z&4Og-f&9g8>8Z6Ju3aZw&q=-j#sGY)g`QW zh>T<&K5q4=V{GxpsQ4SON$4@{U66~|<#Q@SYbMf8G5kPus|#$YkG&&u4{T<&A+-l` zjua8s(yXSM9oRU|akWSDu8V3)n|d7lwxQ!HkK`dCy(lab9jVqNfXYDy2sK!#7@) zVzDHMpqn8&9f@^aAIHF}a-VY@9$F*WzTr7?9g>OTiMN4U!>V@AszXvGT$2``poEJ=|A2`n$ z9jhwV;wLyQht6<##dKDeq!l4T#(nFSw7F6`Df`cI<&58Da(+=ZF@`(}>a;uN3=83< zI2Z%o`x??Kf=&K28o z#!lW(=}Hr`)Vb1AJ=a3ftuJmNySI!b!$uUcovNe171qb$Ssva6)1 z&UkUu;~for`TJS?6_dnT47!e$3ftY-ui6nvC73A50QwsGr(e^p^ziQLBOv7M;5QYI zb8~HfXe7FfT*ChVxY|xf%m~S?9Zp&EUt`aW7T1yaI~Bt`F}ua(GpNr!4PK3iT&m&J z0C#ogy>@>ZJQCgt_=$gI}>7;a(WS3O9G;NsvQ079%_}o7y+HQ z>JLL&aGyO@SOK@VI3!ld^*H3WOcb`%EW~7ixO5+dSdKioNj~u>80bKzhzxC&Y!kt5 z*rkdle0371-5Kl6U6R=6YT6ef3Jc*>0l){GdeRaZ%YaVaFnST|N-^duY{%4erWtnt zI<|RH#yvT#oz}%$LcG{mNfm`pZgN*GJtP{*90B66cs+SO~L><^pgnQPD zL_S;eIm^31t0@4MAH|MMV_n0vlDJ}i_o({URe2Yb!9m^c^{PK&-60`B2d)QGO+9R1 zv%bb1o|MuIwsJWDe-&(5Xre`3h|U}DXN-H+{-37q3ZyplBLFY+u9DA3BqL!+&vDLc zq6<>JowPXkv=uO}W8e<`NzDhtT1QF zwPE3?vuK}c)OO%bqYD}7#a)tGoE2&)>{?43Ux&Ia%19LIvw$53{IgF(?_MY3eNiR3 znb0RafbU(8h-JRj?hKNRN(oBGY+!4}-l(b-^;_i;l^H80%=;k9f?~k9}vMsfVB9#5zr>7O! z{0Y^xeMb6E3V4DvO*d4;DR%z=dfrlgRvy)r@t@%)vG8NX(&>7cGulXe%UiS`B*=OY zYl?VC*~$p-hBjAR%&q;VCyA65B=tD{RW6%!5>XyO7~qkbcl!sCuBntfX!A>;L=O+=#Oq0UD=Y9mdx zW2f82v$eLrib<80O_B!Z(AUx52>fOCc&#l<2g9k{Ivn<|ku=>&q@U+u^Bm_D?SBn5 zCx%OZEr$#TCmed$4k9frPg@6CGj_T9G2yQeL8@3ayv&ROf`>Rb^sLLYf-OPGLFG3) zn}83cc$b7cRPZ{MkjW9q!nO$>wcP3cB$HFNKfN-G5hB;CR4TTOpWZBzYaVOyL&Y&WrjushEJHFdai8~it~S9FX=wosk;m4l zOLE$U?31=P$WC%QX0&ZlBz7`4m(U9R6A4Z4l%<6(ATRQGtX?@($$w{REFDvU5Dvi4WVQhT&n^y zPaultZKgqj12FCm20oRq1>yxDXCb{fu9)qigK3>MoZF>R%be$N&j90!+puPm1aq{r zcRtn2TV4QmmH-@JcP6(j?v#PF{oYTZtrNJlM^9}bQH9zEOoP-_xEMPfpl=7}4b5TO z-4B;1p1jr6oJPh)BWMFBJk{(?-*ajf+JG#BEZm;8rEj&>fD8#IXdT62L2$&4r7%AC z9E#Ypx=>Vu!8J(EPockQwc}oJz;dIws}d;M3aA)40CQBWY-0FRHe3%{lqP}@}%?>-oj5) zl%CC9oD@GMGnL?qsd1->$pSU!sV9-`iU7_)F~8}Y1>P~8y^ngUbpUafUzqcfYo2ED zXDM^1#;VGL3_H~s;oQ4-Hy9g8u7dV3cNI8c>D*Ra+fbY_a0fWzxtXT>6>aUlRSSUK z*!9h9+FR`kNXNE+oo5Ldl}ds(@#sx#Y2{>G4eOs#!8IC)zh`VhsM;5EU?1;Rg5f|M zug!y=DPU01hTFRs1DqOHk8@?V6~{R}4L+Kgx7eK~9%c?Uj;g$7hmaWx;{!d8MNwOj ztW=}!St4gULEY-j-k~~pWn#xC9Xe*TZKGyYG5`VOkZM_NcLpG29+^EW zHjNSa^AzGjr+3YafNK03Qv`BPIjo2QR!z#I9DP0OQrwa<4l~I5*B>$TvZ&q5sUd<@ z&In`l=hCEiX;*NScXRTws0_>r+rzF2`ufyUOSEs`u{j6R)|{kdqjH_x$7|t>HvU2F z?O8VtPVPt_P6cUPmUqZwm(PAGkxm#dB#ivs39R|KV-{h7!6O8pN=I;mb_WNu%=yne)F5Gj z11m$4%v9$C@}&hv(YNubDl4Nbq&eC^3ObSNR$z@!A27(r9jdf01CH6~D*dkHIl&`8 zjb|ow(~XNzpow;ZGmh1M2+7)6K*`1c{c5~8`@Ij+wJoAx-zX=HEG*Cp=DVY0CXJ(TBx#~H?YS*1zfjqO5h$2Ngb(N zu0?%|(MF(+s?v|;R^TCBcBwe_s?hBx3fsBPSG8Y5<)ZF2>!GxjoHWy5A>ngQI66}40Yn4!ez^G&f6F{RXhQ-eeC?!g-+_GSjIV5D7go4Me2YtgdYBH+_@_@jNf%$+LNxnnVKQ|}&RxPWtm%{BMLxMB*oUjG1LzXxZ0-&9QCMkgN9J0eqqIEuY0ow z$n9ZhcWnR$U%Gk&T2SIL2j1W*Va-&6G-hT1DC%n6v|(KGcl17lR&^E4De7q;5wT-} zNB|6DHG18EU@*Dup7lZ{EOxgG(S1j~Xrw#9&ngHfJOFXpxn1qC(;dwSLrfO~C5{gq zX0%ZVIXr(+-l;>CQgebjW2S3D!Ib2DtPcR5)GJ=rI=Wo7ZxXY7tHI}RJYu!&qK%kH z1~I_EP4y1RjOLR^_$=|}ZFhx|eDJ!V} zN~4Sc!Odw~MofGVN8PP$pm`#msirQX_LlfNM1n^_&hR~J8^+N~bA2l$ahPIJx3LGX zuCo4bBHeQ6hB(Ux+m=yWKB;vz)vz<7u$jPST#g5N`CP_N+gqVAZcgW&Yua;O-$`(v zC0WYt zt3vr-G0bZ4J;1kxMo||fv5cPl)YvNCc2B(L7C$Vw@BoOL4EYgCkn`O2cYf&(6F8sM2H0S@ju0aPZwd1P(Qc8)+a zkm@D~!31YLd)D#MSm%;YQPMMsm9hip1of*kYY!4K2FidrJY(rqJmi$CtAWo$)ccxS z>r|33+kwxZ{&fv0C3Z|xT8_8ZXJfcXRfkL-D`!sEC0Ue6yMqI{nDnk)V=%4)0tY!1 zqfwGEz&sueb54X_sU1+N;?>#eb{hWxl>sj>A1*#c+()HuXxiiCN*M~04(Sg=>5B3D zZDgFGU8Ba@T(*QM!NWL_kdL*4Ch_e^cU9<}CFR(I%q4LY9E zH;%`ocpJs`mYS+Y>P`*|5s(M1eRJV0YfU^4MR=w28VANZHDqOw8TaT{{ZXLg;h&M)aS1#)s30zUL}xUgB%5AjAZ1u z0=(x~GTX^-Yvq(I%0N(d*F`RgHk~Z98^&Z)jh#WyTDNUy4yProjDOX?&;HW-Qk{Q) z-ZZ&V8?d=b-*UR!u`2z%>EF9 zQy)=$S^b?JMjDys%>r-1M=55Cy_5T10uN#?h(1YL7 zu0a`)smN97(~8y!qX{xwNQA8Hpp%{lrE9?o?ral}am`kSA%)K;2aTj0`&OO3qb}Yu zJxzB-XmZQYlG~LT^~h7fts7<`#s+=QCZM-X%bnRHJxA+Ol42-9${yIQY+)|N?Zkqh zhEu}|9YK%)`+@8SdaRP7F)Q-($v(9pkc@O-2h3Nws1xvLcQ_Yx|6cdmmbCx7AncPM+|*2pT?g)o{&erP!$t{Avnce z)x1Ne={BrceAiKupl~b7wf_Jb-D;AQP-Rv)NYG#b??P_&NaC+v6qTDb^?wstTQrFh zHO!#r3cbB^Tz0kMR=bsBnO<1f5|BVL0rV!F=195wRI6?3iOD|IN^Nr2PF>dO7EVBX zb6Uk7HC#;PqAo`)^PqES4Wt?W09W@*4*9Ix*{tCiQ65{6$DO#Yb5`+`jvYor0DgYi zt}jsWMa9IKFP`0cRovrCyE;vCTuG&MN8OGXpXupbb^XxfIZeT8J)J2Amv4?s;V%1q_8 z0`Zg1Dx2ZKUB5qwF%_*>(P9b-?pc*-~u1^zY9D@{qo-4NoQgS=9% zZz9P9kXsxdaad7481Ds0;=e+^B=|o^_~+njFBWMjADu0zl5K;dcL&^8#hxYje{rQ< zy2}>U4S-Z}_}4vpNv>$E3#VU|@=tS(W4jIaPCFWKV89mRjNi?xLOx-;IL~U~tSp)CPylw4I#pGSo_`Tu75hf`s%<{v*TddSwvt>ZxozrInE}Q})YsJ?0Q?B|I>n{Nq`NK1 z<~JPuDu?_Oli^HP{sz%Jccqs}T)&pfxk>0ftBPHzfsk;dlaPCg+0*aciZcem94mTy)uj53OGCV|)I8D2TxTCz zzXih;%VQk|*6Un_o#E=l4Y}G$$ib~EeL)7+3^p$$uOhVAm#Npobff@;;PpApYFb^Q z73ITZa0W-#xT`%)T0&K@n|Zu?HgtubCN2Iv27JJPyErTB~|OpdgM11Rm8F+dQyPkU_xR zRqn5YE1)W%{GrEso?)oc%J&;P zYfFqN2up2jVxJ_3Vg-HbPds!!^???o!AQtX`-Lo;(&T)_bH@UkeF@OemiIDb3=mEU z?d?=$)D%cd9uGn9>sgXsl-!}Vvz&~cqt>$LzsJkw{D5^QB-BbnOH-kJMLMb3jCZS0 z-kD2{!*SqawQzH4*l(1L$BY5T)~xAT+cyMmlyQ?;w{hE3)2$~_7tG4aIKv)IST}xP z#tP(Q4%N#+sttToeW1dZ?W{o&*Sp0)3 zIU~}o`PmpFA$>bmG#49`eB3Qsg6+3%UzZ~s_pMh}%-*)KaIj_FoM7^C{V9s!Q00i{ zimUd5kDb60zcDoo*FI>-g7r4+phhg&X_83{_-| z#&AY*axgnpu%rwDyPn;v8Cf0F;%HhG>NAd<~FxS`IS12*@C9 z82tU}R8W}R_s%+siBmFXx1sf`NTuAn0lT0b!0(ES*wrhsh>Af9q-d^TiVRHlOg5L1ad_+_5{Fxn4SsgQN*Q)AbgoooO8ewsR$BA%;j0L&Nw_(hKuAWg=3A`^!BNvg=BISF|~N^F++M-#!>E8+FmiW zLcVfCb;UeH+f?VCfC2i|u@G`b?Bf{r9+a}C=geS+=zGzk(HKb?#?TZ}PT~$vwN$yc z+UF~{k9yj8ixoVm-Sr=-Q1sTsyJ!?bAE3|?0 zo}QTZrMHQrRVjiAAfCWfyFEe|S`pg75g#O=QHs>HKmw@w_YQDrw&{_AIX?V@>sPJD z*vhc?+&k7S+1%`ebtJYVjlg3kspv&(*+dzD%7Vj;{82eq~i& zgDNmZW}V&9*p-&0=)*bt%m_Fwj+L)yf+hfSj(2y@xT+TIm5g*G*9b51%N7RbR%h0aIoSMAFAWlIJmjIXUyw2%ZVwUjZyJw`=cw>w=!e8)NIz^_J{ zIHv5jHLL4k@cM zV#0Dm0NYT4Mh7F(wb6h@YRN;fOsTr6b$7pblu z*Y>;kd#QMe+d%OB#;M}HTIs}a+i5n1mPS5>u}R9x=5BCHR(js4BWjaOBAMh!Wb(F{ z_Nw4=Gv2um7T>L%v`M%lX`P_r;^{)^3 zpM7_w>en;*MN}rn1o9i&zDpB7Z8>s^8b4C5sp4B`r!qW_N#R!<=DEwi90~1YgClCN z+<52luQt`ZOLeK;k1=)t#X##<^h;wk)KKg!@?jAba$UOsJ!^^(x5n*b(xVqWj`qeu zcWX0TFc7#KS&29wO5wHn?IH7$*5FEtalt<;cLu(=u+)AAd=JoUYD0n}`%`6&chBRO7sap}pXAa2}dY0;I8nr1>N=e@%61azkPBWNJw2Z2#W-e}5z za60F$Iyt3cV<0MLA*-@WO%Ee|StKe`InO;Rc<{>tAa`Sw*$s0=;-^Dsri42 zd)6&XcrA?{*#2R6oCp&gzAE`C;ufo3+>lSwq z{h|%F`*6VEZEnLA;Nqm(c1O|Ro7(5A$KyX0XqORvufwoQX(EWl!E`bN9Z6dCUj^!# zM!N&t8&9#{0GQkI^PaVHz+M>Az9Z;u4dgcW^Mw1Q@ae~Fbj5MLDDdKZE!Cou8Ffu2 z*`!nrcs|o^+3C+!0=eT!$~@87+ErIOr!CKDOW3Vol`fTco_7qIj?VhnF%hyV<0O;T zsOa5 z9QXrKj?UmlEMgY{k#e#!oZ`L~f;N#C1cxiTYY$&a{WJK%;ilHU8|$`~HtiezfF#f-%%rxtIJhT@N=GD7PL|Wi`1fati02q*O_7`^77`HV+lM zb8F@&$_~-X9N^V4ZdHa`1rOfFe_HhxnzMHseK2eMiYB|6=Fg{WT1J212krj=tilqdy|r z&eD*0Z<~>v3bb4_JGugR$gIS64gf$`8<_VMJIIRyZf&^E257dH6_At8Wn}=a!ZXc8 z*CtmA#JR=-j+v@2aKVdo{_=yGvuf;5DVFqJzgo$rcT$PdX(MXnVoYOwW0sa@YZS0rK#0Ln@wh8-n#t6B zV4_n8ny|&m2R(k3=bkC}ffk!#Ec4yR2vi|_(~^G*@oiJ%8c96iz1h2sNz^#(M zyw3*}jd$gx_nyJy9~XVKr_;2MZYNdC9Dg%jb9b!YTdbeDAc2-*bIxmmm*R!hv@^wG zOfpZ8GEhEaS+`y~mN-LACUe&v>vXJ-5#CzJ?QS*ez*rT0H_l6co|Tz@;w|xvxeh^W z4_e^wejvT%hiTZOjyuv@c$F^X%)o8#4|-Q@YAn~EUJ^4SHtZbj^v!2Ybgs>o+s|WF zu9^vyG;%8Qou;v-n4@hOBzN_z8qu0_UX8_%Ur|+Lyeu1&yBNtms)v}tNo*Q75JCIg zXK13sxipsm42*p<)|g1(jy(k%S9URz)EY)3%^-~N)OV(=E8j*_nliu+q~@|OqF~$% z9PmYNk+hOd4{TLSn_}fhcK7K*)J84cvRy9aAe9*)de&r-x{w)2HK%o9C9|0$9S%tK zt2S}LsxH&=&FSwz%tdBW1`pPl?#keUf<5b@o5R}yY?a4)&6`ai|NpwP6fNOwrRsD+N;M(?Ct&n-1sj^jkOzjEu=sRV!Uqj z&qH6SnlFYljV9PfV{066{{XYjX`UqTR*mA#9^*^Y^!RQp=3T6Xa7Va3jYVuVD$(U< zeF~MS%bMu??eLF>2ieh*;dYX78EoL!sQ6dG)|T=at`~n91J<@YS^G2G{4&-J=71&C zbZJ|5*$HQdo`=`#Q`n&p1(W1u2dT!Jev@u%Q-iZnrkCDxlPTN{#(u$9l^1$?V%e0*Q8T}^2IbO|Sv zyLt_$2lcP4J`w0I;g1LE7dMKqLa;NFpWzBgCpGjLTqO%w;Lg39i%)a$w#nKkHb_sM z8!|8--cnSL%vPPWggTZuEIafSmuY2yBy#)*7)<^ADAJFzhPYT(ES=%CN^?-K(M%NH0KXAzEpaZ$&53bPVhur<=OMlQ!s9q!4AE;`_2 zn$WYYEfe zP&P{=X{u%sl1V3_Jt=c^3p3DS)RdJ+ZgZbnvu~)97z30! z1XmSlf3cQLXg?|vpZDO1d$Qu4w*b;8jkApe5ZP;7~Sn$Yg-}Q*oSfG5Hvq-Md4s;*GxV(P)vP~fWp;RC+wJ*+ zkj!{2F;4q2W4L|u$0x06$!uAlfZvZwQ4UTqli!Y&<8#>ho=cSy(_%OWpy|-(>rpg8 z!6kO@>sreLNx>(cea%$4w*-!GdK^=XwA9nt-qtakZ6|K!40%4LnYFO&OJla<(yv`a z2p?RKPkPV0k#}_=xWF8ad)Gu~bPko4g=@3EGswW_1FdCCHY6t`WMoy_#xcZqXCtWx zt!7N%<7otAn(L+5;;RP5SK6Q+PIK0*LvqRj^b5!YX0ln%?Zf3imTE{v6_MI20l}s-DAdDy^@(5gFt;GsPiU=fk&uYR_y1Cm9qQ*l7Y_`GB z_o+p}ErM~J9z{j9V|MwD!g$Y6F-Oc4f_HI`#<^|WZcCv#k-WwPa7Q>jO*vJL0O~m; zo+_r|j56K20oJBfSyyuqqXRiTY8vVaSJ1MHi2?H#G281_@Cq&%1}6=UwN@nE9__;! z;~lEV$2e`kcckQR5_Cin_1<%znx=_M4uxDEq~rlmD~THo$s7agRulzX1_l2B+V-TG zs=Im?Ae2bC01c-J>f*H2)$6Jyk0LP4TRA;FD%30c zjF-+jo|&dwL`-UU&qAJtr-yO_xrWkmR~4*ao`x<~V5_STV&V=1Y|o3u>@zU>{Z>$Amf}?HgY=S%MpnY z^1{0k19Z(@5xlq=CZkXHxs+BwR&{vO>{VFV=(JS zE1I@QR#2tOWDl3Q_pOUITVNt~-%6a01UI%Gg>k$afz*eFtj2CZ`>oFC;OP za=0@Qx!%1MPF+t5wqqStG;dGhYb_St?x!!YcwoZ-ekY}Twegqs-PSE$_eAhZL1z$s zuc-wN`0fyq$JV(|+FRpw)RVV{beW0(Gk0*OmOtDdO8KDkTBuy}-D;B6vrk>|9;xD; zNw0N%Lg!PvbC#OsLhbt3=#Tsoi{V5c+M4ggO-9{Y`#pZerxs87YUAf{eTD^odU#jE z8i$H?m^AG+)?3TTm@IKbyyWyX`@Qfx;Z#2dehc_(!j{pjH#YW*J>0+Z(&5*0=jeF+ zE6L3(`x;!a-JMt}UfQm^BHxLutf0NRdDjw7-eFLAC)E4b&Yu>%4JG-xo)Sz+E=K%i z)w`V6uz1^ExV3p2_T3_5_tkP$$4<57TF;Mk`w7F}OoldH&acS|y)j=K3XUr0vFy~P zW5l1~MZKiL>7s4Kmyj_#MtUFqwRAoR(d@hzr0ZTWzKlsGiyIp?kp|N{0wdu1b*%3f zd{mC}P?{(TvN7 z(2Uc1*z$Sx9ZTXbiSF+%WowD&k))ndGJct=H*I&}$1MZ81vm_(f)Ay4{s7W{u)H)0 zM0@MvTXOXZJ63;>wf#F!wHC{547YA_z;aM}n&{?=lBX4PcjEC9NzJmy!~Xz`R|#~m zPR1EgNGt~usp_V=kBGWur}mm%S|W(2JOpfo&nCF#)LKnINRg~c5AfCPPf&Y{WL?>e zbB{{hQI#sHRP;}De4Zy0?c&t+B|1p6OceyM&pm}!lH7TQG3k&y(QSM7t5ST@tq|Zb z&hKiuWiXN%EhC%YHPyQ<=>x|}jf_b$h%Upui|rnR>W4#XG?dsfby<9q!kDG**u zD>2Slx)JGF77^}*EX(_#Hy%1w-}p^DBjLCeH9)SyP*no%W?uEilAO6zZ|GEIDt`Ah zjmsa6`u>2?-Pmi=Sl#D1fkJKJ9PBx*kZu&THMm(&df&o*hXpdv+zt$Uy|LJ;COyG7aPdf*U1G z2<1YSCnuf0l_M3EIsn}de0>deB%8I&TkB$Y;YLtc6jRVIHK%J68L$D`2PL|J>sj{E z?Njp(2ORbK*5#;Kn63z5ypvrq@-dMjtFAVNE6}!i9+f|kQ3xyZDd%-bZ3C%Xl~4iT ze-&9sN0dMX%AT3sU9s54*E85GYqXu*@LHg_&i6eZ(>$!rwB z8P7`SNXhG?D9R4tNl>722V+&G3-aWL2aY>c`P>k}v&hKE=~UeUC}zVFbMl|(TS6(A zw`~YZ6x)DLW0O^$W?*`dJ072fO?5KFbDio2cCQ51E!66x01_3rIpkKnx71wr*qNqg zk+!zbPXyJgNdg=@X=T{)S0!0enF*p^JqkT=3^xWZEb_BVx#>JpR`QHn3J4~wbeyEAPhk&gXqQYnSXZX^yk=aXDaQXRncA6~V$q~2wK zW5+?B-qk7gBh=nh-}8)|oM6(3@DE)5YDj@4Pw<}Jl{3Y$f`ENNswZto0_|Mjjos4E&}Nt8Z&(?T-hL7oSvYUsQ<;z@in zs6%j!-)D9SEMVgU*0BU2k~7Ub##$pA5P0wW>Q=0oM(Fzs!CLe)TElk)!IXehp0)IM z!(WIk8G$_CEW`%{<0ij9{s{QYEX|}(H}4naP3PsseJA1j+4RZL!;+^51B%j7)7?Dg z+Lf92MuqW;`@@n(%&m2&%0XO?*UqhP<-7jq zj57E7R%W5B>OKVVAbUw1+dJ9cdGnNO$WjHE%n&(1hN~s zjkBu=;@h+Wzuv}iUL*0x<2{e;Wv5&Ca@R+<)jU^nrdX`*vV6Rs^|IrhPeEKbj5pS8 z&9XiG207teq_=4L!{BGbjSs^b1Kw)aDQl-$jLUgD1VY{NO>>{LzwK>lru;7Pou`Hh zLt!3^JV|`x5hyd~3M_+2dv)7_vkj&%FP)uwaf)8rZmV7AWa57GFQ5_D)D9v4L zSQjiGEW}~D=k%vZd%-G49C~qB*Ea#3TX|dz=hmq{rO(QH{MEXV$kol&zmg^+j=g;; z<4`VDnStklo-3Re${eoa^9%xd(;HCv!3qkf?r}qs+?1?!E2%U{cay13>9#t|PFIb{Pf~vEbozd@BwyWyGBG#i-^@+qQ*nLE~`m zTt%Lu@AGdC-Ff2`cxvq7TOd9;!Qj#DIvOf%>UIl%u&{{d2a!!}O;jvDOxqd2L!6Sl z@l$=VvUagioZ}o;POhf@%KDl&S7Q^3mFM4HVo|?nq;?b_-+V4Fg=A) zxtOsh%p;yqb{@4p(Y=av(nF<{LqX)}j8=AY^17 z=QU4PemPcl z-MxtD0-RSBWpG4C=If8W(zfkq+5+IH0DRqzWSUxB*+a52O`9nr0PXLdzmIwfh~z31 zGh>sUm7>udpfTuix4jZZdZ+XqDy$CX?#HEN+=%|}c9Y(;CEo9Y(4K;*+#RE7!8z^TyJ1sfj;HQrmT2%XgIRcAY|v0?aej0 z+@yU4KHWUCfWeo)y-v_|Dh>y?O3qR_IUCWG9y|bdbDo2}ShBV_IOE$DVmm+?SRcm~ zcG}$}1OiKrPZ+6k^Eu@aBC-I3!+X`ZEP-qy1@$=VR&A_#+~au913fEB*4VfJ4ZNHW zK~r?}Gq=5FQ-;L`NIkF{(ya&!NUGTiah4US6^a78kYTvTsi~gZZUV$SymYLkC9&BC z(U#W#03cEO!ZUz-)mbeFkd#0WADchYw~ec^kgmN7(e7Z$ z5Ov%!l6uq(ODeZa^&K%<(m)v)?D=~h^_wNTdMR=v?FUrQI}F)$d0Y%v+nQYuJ} z@)Y9)WFDC`c{xS>-q#(+y-Klc+mpG^N?IaQWU!;NAm9vVj@4~ZlmsAc9FhlmY%mg3 z@z1Xnptj_$0-WcAo`cq~aysMOfIuB{ij{F8SjkW`%O2De&T+$Z$mo8xJWvd)s38gH z!>v-ZtZNoY0{o?v0(xyc3{}`il1s+9>ON7NA4)A^Rsg9W9?UC77!)$F0eR&5QFd;n z%dtiU`S?Hx=Q+(>F_Dd=v0?Xs;2z?h@PTr`1I8QW9jb?owZY-P56Ly5GG4jD06s{^ za((N%l{r*u$380+3^hJ!J$oHBhKTbj`LaiBt;lIxX~sP&*zm!U7h7#c)_Ig=BWA$% zubUIb`i;%|%O%vT@?bvZ$-wPe_8vLDigkGI8C_0r8`8aK(2RN?8;H#7)s^Dxk45pG ztFL%z40l>}xu_q_$d)E&yilvevGF^eR^M>i|T^yb<9wpE&Z1n@Vq`^X`ED!$x zUc572ntcycAsnKv;L#3*S3-j5k2Te)CwWPo?Z?Dh`)iCP>wbM|B>10tNZoPrXCY5o z;qGoN7AWOIy^3eljOL##dks{my0h04;$6ZhB7UO@f9E($hn+HDJ^r?TeAwiw(oYMAidbA$aqw0GN&%`UHQRT|2af}ZB=ANGr z?#d#G2;-BUbL(C?CaSxGdBpLIdQld*MnlM~x#FKBdLeVoFNpS|@d|frHNy}SfO)~G zzu`ZX*b^eYc^Uq-<7upu+#kk+tjajw#ZR(Zq#n|L%+F?r;u|3$=g#lpUWTkiX4j+o@SQdsl`xh_HE(;Bm(os9(jb2GJ%#9<^MFYhqNUx}Qba`1aC3Rkb4z z00SxHA57PGq)Bam6mr9L71l-zBkuZF#WU;vXa{iw9_PJm_*=y~t)k3CPIs5L(s>0StdE+;YG>Tab{*o>ljXWqVikKz^V zo?Me%8Ks3bq4hn@S8Z-hYf%j>gr27-H7T~q<+9xO&kXDL-YwLA(SAhml{kcOtK4F$ zd|~*Qo*3}8!&_XRJjJ#!2N~(t+PU2aS6fXXWQtZ!2e&ohpBy}0F1M=-VpK;E+p&M0 zTUOX2`JK-$x{`fTc`m1sq?%SP?%a}1cHam73HX!aAB*igKjG~@YwOwFSNCa^jY;L3 z&y2eVmVnH zVknp2BxbdYZ60ZC6P@I#$A8H3KiRYPW$@0E@o~HhqO#s6g_UnSQ>>hBxw_z`x*p08 zTzXgA`m#eWoC65TbRhSynEwFanl`MT2sK|4>ZOje;!DXd?q`(bxGr(erF{Y7eL^_Q zk|`3#!0mP9d)L8cmHn+(2q*zZ37QC zDA6eeP|xB^KRA0sb4E57k|vkVryB%ke$ z*#rUi7M=VOKZ!C8T6#8+}y5%X#JjnI1a!{z0jlTz*Yp)S%7KS`E zy_0zaaszGP*UmcJvP~n&GwztPZgP6_(!PiAeY~xIs_J%fmV0|vkRT&1k<%UP<~?Up zXP!ZTs#%WS;OCJ~1s6HVnPv?rSZo&-_L3w?^UEu3+1T*Va z#%@rb!y~0__^pI-f}m~PjMuMGT5*SSJOh5=b16J(BkS6jQVeOiOMnYm! z#u)dlnPCi>+!PLj6y{$^{z!$?)?s5V)yEKKAqtCyl>)|^cb!m zOM-M8CYECYEw41Fu^Pm7U3 z9^u(a9iyG6wS5ur`^6q5@LrDh(pt#`?-`B=*|GBxjGX#ZpRsm_rKYy=YVdi=()ndu zk1UbY*T4SQKM$ht*M|H#tZ4CW(_?0~yD{{Pw<_)IdRGF9ty`4!qY2^Zsb6#DnA6A- zd9rzvNX;a&6OztONi?2dkdOvPKuGF+Y5XSa5E0Mb1FyY87Ys)%Nd6l4+)`FJ)3dmN znB%EnFh(lGJNA`G%PRH)ltQc(80788PIFVo3t;jvcq~s}rE{%UnX9<_FP(!IP!A+y zk?matsOCM`IV7AK$8@QF$vsO}$R1Z5Z6mLG zv>#+_yoMYA02%eoHW9Kdz%B{E#YSh_87{2AoNXBSYq}S)7fW_kyTgUslpy5s)K+w} zF=j@>=LJe&kH)laz;Jf9&wO>Pxue26?&x{rt#v^qxtr5eR%y30so<(C`$PVIiz9+e_)W8Va4)~spzOp^{5aAS(u zIxAbM?p$LWRajR>5~oa@@x?(qkjFSt)1LLKdJ-a3eeSr&O3S$c`haoQH7upMpDtLQ z!1t&?Juov?if4W}%|o+=;~DC6N1P@UoPRnt5#4B$ zHQK74SnW#)3}kbPX^|y#P@c%ZRSp}T z-Kk=;e1gN|=NYXA8|67tp;+YOsi_tuBw!Jb){V6@PQ@!&_DlD74D_v-0oVY-r>Lre z91t;(4{r5jHF}~ zVp`pcxZW&+4`v+&7PBoj(N{y%#pD}H{`p5kTOmsZ9eBwCxI3{XhpRTwRA)H$uHQ+y zTj*9`dUY73{t3t9Jq~$R?^4n&*7iJden$O*09!NoOQ zp?{PX$;K<1@NTLEi5cKvaNRktR`B1$tyji(DISq%_mfNaVug1%N{2NSc{5iRvN`r~ zEIW{{;qTJ6^bKfU#pFAU=y*Nb+OVe6r_!$GwY!Q)CXJaHB~q*yb;zu5ww*IpUD*i4 zVEfeQ)zMVs)3-y=uJ*cy13aEH(z0g+w(P46=N;=-$wl>|&Yegk?Z*bH$!5F4sU>+V zao^U5D`<6cC#jvhmnC-r+i}fBxtHbY3CSeYwz1om8DKdEqL$NWW+%&RbMH{t)H{`o zgf}@Lcg;m4rcKxYWAv*A+>ww8=N!{g$%Kah?L6ldle$+lQM(x&vJ}G(I}uuz&H~1y z6VD`ds_@&L%yG3q{LNd2ClUo8QV*qJChT-ZDA~7|u)xS3o$*_Cm!B+UNnCy5-nrXU zZ!BPPeesIhvH8aJ1gEYrc&=GblT)rVlRA4#{K=ioyQWCywPU#`dJw2M+%gSwG0JyF z+qhAR(v8e<77B~k^D!u;Zhrs9$GeM?di=>iT1hPcN5>z ztlOw^j80erPDd0#sR6B!)_6>*<${{USv(1%^4_(;nR zewCyq<%vhRX~7314#uiWY_cI@4_}+N z71>T!TC;|5zK1VymB0v~`0j*o!A<23`X&PdKrQVng(rYfo3^0s)-b5&-5+l|WEU;^Zf=iaXO zGxl!!8EF7O02{Y)&0K&502Kj;%bfJ58*{ymbCZl>rGgFu0rK?(jMg(Yj4bR}idRPxz%lKC&{YeYbqnS(3b^Zz_4EBtu!LQ-F{Jt91eGLrJPOIX z`IxXc0Of|>Ptexo>c$Qgat}Ce^_z6A>yi#Y$Uu0lW1;BMrE{O2d~h2qynq(7=gB7+ z1pC&l+&?P?%H(o$(y3f3upz>d00oCy(lX`H^=aYnxruWV7X%PldJ3a*hap&nBy{gv zS0XeU_Z_*&P%;H&&a3x!lEnS&bgr0A;;V*MGhPA|jDW!4W|)z)9fV|pImKFF5Mw+q zKBqM3pxO=r$nJaBd}n2;^BBB6t~%Zv#yICccQt-HH<3o(amLo~Ot*{$O~WnM*0kdr z*>iwMBacs7Ii2)AcBDB`1U99%khoPOW4NsgXj{vC5IG!TrH(Q|Q_2o^4&tuc4fqO6 zMg~dSiqFlQ6Sq);*Uc z9JRUC2&Jkq&9eeAfd2r*bIA1Km&*A}OaneM$6?;IUeVkJ11E2;4_X$`hx@xgAd}Pq z?^53~vo?ogE6MW8y7b2&R97(W+$`#+n(T_1y%bjVG8Zb79ch`rS9JS z3u;YR#1L=bsZcs&oY57p-pUe3W7id`ZB&OUO5>@=T8=ndc-xG7j=t2JR|J_HQ8ANg zCj$+G&086E01Tdbe=1asqU0x2lkO_VlrpSN;16H%td5kLAxm|L;Ejiz_vuqXAMbZ6 z^cf!crZh!lFR->r00W%W>m>6^WNrWuPI>#y=7(?Dgr>pYf>FNy0(7k z9sARC_*F;E(DWqs?N;?s_G=kkNn_bp)}1uxH56+)k);PYJrT(Gm&R~u7j0)8izVX_ zLT%CT<9*@(02CcU@mLujED1cH{{U5X{v+`e zQ`qTRl1P$8B6I-ZJ7T_~l_f?tTc4lfF%repb8k+EmTNkDY1)jjf>-9?`%=Rv?!g~; za2wXQylWJ5?7e`3vMzWT3Vkb@hC?6+g2S-ly1dRC=xynD>>Ma;9OoqTu7BbhE69?e zIc}phZ6+4vuml`>^sJu|NG_qtegMZOnwO~7r!xi9pb@#;y|YkE0c@!t9%)?pP6!zk z$z%+;3*V`yYiMYl#u*A{4T>7&f{f&hcJENI1Y`r&n8zca$n~eP4N3gMq-P`n-i5l! z8R^tjQ!W7quc)RMBm>_$q&ryA8i3DSFdZq6d`KZ&k4mAu^(UWSS`>qVa&b?fSh}}T z44ucGywkqg!AH%z9nDc^IW&2MCu2xfWyZdFSmfj$)eN!&fwe_eW+VfgQf}aIPADCT zizZZbu7=WY?5#91w&`vGP=C6;&2yH~?wNCw>zcPc$-0zrgi*CddsBN^irSq8mGck@ zIRmA4UJ+llTn)n@^!Kkd(~)I$8Nl|hOYr6`?Z(~+9Fy9fiCoS{ajR=74xUhg5;Ver z$3Nj;J?pnGsOz%Ho!gtBuTAluu(rBeD#*aFBOcY}dK67_sTK-djyS2+$ZFA7XWk$1 zOOM(9A0PZnJ}T8MeEl~>Rhdz_Up)QK%NehwKj5+cHj_d9o-_{*Lgfy>tfEL2vHQ4` zz6kWrE7<=4;Ge$=7vZPCJsw$L@^7_e7jteOF$eC~$zSkVyYkxK!ksy?(QUi9E896y zTh?~>S43xqPD#!6W~b~|@m?(#;NFpCa;#D$5s1)Ws08Dl=Dmww@!hVYr!>~Li7DhX zQT(NSh_4U*ko*s6sC+Q7wYemTHps5Y_X#}nUb*8h+1l$$)2=lQCsZ-X{Y##hxsJ;iO+IERIQ&a`eZcuOrv>Cs`S7=2=xTSSxY`T=7)8 z-kGahTU}Ys0Uvz^QGxCop{_D(3(LnP$&YyIBP9Mb>C)2Xoarw!nehjS8dhcVS7Mxu zb@~%f=uIW$>`9PAD&j>0KQFkhxO_KTh7(9uRdK*5S2a@OMY_>iNZ|xR20@Gy)7H9d zqJ*6HISD4!v|HmXmBYkiKPJ}3MPOWLHXb39$h?g$FWrgrmwlV>Xr!DMK0Xi{RBRmy9N{VuH+vU*Swd$>PI88^vx|WlyeT?8b5LQh5 zq#lNk6gt~O6b$Io0y0qNHDg)TZZ1%nC1+(E5_?xOVp~zX4D1Ri&&(^wej!7d%MqwzBbikRQvgjZHEOh+m8L~>)(oAC%4e_nWfq%kr#Ol z(;r&+S5o+iZ>IRJ^Ix}UlWQ}w?(2@6`ikgrjk#B;rXDFTX*H`iJQ3kx;a?Bl-E_RP zm=;_B{C6YTzLV0m%_rk8!i{UfGPj=hmJE?dtGJ0k%8~RUzA%sCB)&4%HGA7v5&fg* zniK>A{WJ8%d(Z62;&$-Qi6dtN?1os-D)YFXPtLm{r3opfIYm{cD7ES8e7yyv@S(H0 zS2uR<(8!^1Lm?pkJXC>q9l?usBN)Yd5A9vxRPaZ~oqh{{lTOv6O-|e|8S^-N9^BWI z+u4L&qyV7c9Q76SmcH!yXw6#b4T=HI(~NVT{{Wp_hS+&yKfG>mame-*9s9Qz7R02a51md+L zMG@zc263J~xvtuANosR8Vax0 z1Kyk3$mN=AQl_P+yc&F)a3qm?4nXc}$^1T|{bB9%Wmn*WP1ho=rjHIz_=i9ASl#Co?pQSc71YnYWohmtrT=o>wFL92< zjN=*YQIWwtM^jH2b_bv|fdY~-l6q!~_7V`T^xd}tog878fuF*aRObVydYUXOc>YwH z9>^`XMIaz#gIT~UkzE)Ft4#c*(-BRC+Qwb1x}_v}|Gl6H6FCXSjC)y}}Va$7k$ z&H?F93^wlEc{!~4ro`D_x_RcLX+&U>cZgZY9^4i04kGe2%Tn?9YvfzyO#dp3i8??3natY*& zX1vDUU6cO+_32-sY2QO{#nvBcv`dK8VHc9UPu8+LEv+?#{!wGGjtD&~sI}XuY3&OT zaywTCs9QFjdh4`|9x9sbR9^8tT|k~QBJw*CRp)@I9CYdJR6G%@t@Xc=BZ9+#4R6gB zeqsounKJBVK)YNL0UfG&Wt!6B9YV?cyQrKxu;Anzii%|Nm~O}6S{hZD)nyj!-dSbg z&p>I%MrTsaC+ZjMuIcg9!5U_*3`;H4UnR~utco$uv9GJ4_*0=jfi*7;YPL~LX{bcW zIyeV#BEKyE0ALS^n*RX9kBfisjYV)phNgVKEzS_Q=ttJSSY*`Z@kW<5x|3~T0l$PtCG1G+&#hPHTV(n{{Z1F@4+97o+a>Y;xQU^(}_ejtiJ={kn% zi%m~Sp4uoexmrdTQTo@(b~o*3IOlFS_O9l`Q%n6krwrkGn&eI09de4b(Z<}{NucUM zMgkBSPCD1nUjhCv>b?TjtaTkKb%IUXB_#pM5rd5M;=BXIQoYI|l6ORKze?5cR;(hL zH8NuzaoVQ6DAbFLwqvC!MJBgL=?BLjgFY7c!SFj(_=n-EV{76K8WC}G8gjDd0OuU^ z=D$4k*j9Tqn>i{ok=)nNz5)HEej)rh@SdTeczaN@{>aoKD=n?e0q{u1?Dgqh6{*1T z+e#;p%Pa0@Yyp-hCb7fDRH>?R>9g_^SDTu9@9K1Z0lf3A3N8s?AM5X3J;jPMWFu#; zYtOzK%JAIArBnrNg-Op9>knl9X55BupFv7V#h$GnekVIEwmEN=M;NFjvE$`d1GafJ z)IFZ$A(S3)8z0W0HZB#uP}l%u)EmkPY;zB$1!Wxd$>$lNEV9jyr~}fvfu|Q=nKxtm z-_nvTF%XfBw6#cT2Rbaut{HKT#-&SNoR#HI1Y@Oa{{UugEP8XuIp&45DS^tbOjJo5 zL0GE<3m6Hv=57uu?X+24gpi}KJ?YWgC~`5L>%1T5SM8u1*p*TPbnjk$M4is+cj!qJ zTn2SG0|YOvYuiD%s67gR3v*Q9M2(Z?{o4NkGg@#acmp9wBe50c&2rkurB7YS64)We zSg|LMZ(6@;Y_4(}ws`4HfP$pp0(czutJZe>3@#UvKDEx?wr5gHp<>qEBXu8mvEwV= ztV<9i`D^n6cMy70lX8W@RpW!yVx@hhf|G`BNl}50)i%4h-ip>Uni9xLrqBj??@upGjtyveDcOSq#!2XECV@bFWOm#JKPv2m&UwQ^izz5jjFIza9Zght46cE|8R&Ws z#$)z?;BZpVp*^c?K3@!fhN``p4Y_~^9Ff|#OhA#8 zF{+$n)~npaLiu}wgWU8U)zKO0V?Sh`tjKFL+kuVG*Ko~IxwlqK00GAvc*pqG+?Ko* z#!tEVt3G(bpd%oIzy~?}D*2u4jg=VRV+P_D`BbX{cv8F?8W9@>bpz&lR<0Y+evT<8x;xjw=phPh+7wks^XNhCB>&lTj4&l&MTGW&o;; z_3K#oa14^l0V5~d>07A1k7kRp&0Jf8aItSSn|6&HA1*0wJoV5gxYo_p3w?lI62_|lEy zv^wfWE*B}!7$`CV`qNZx#@<2cR$c<@wv<3BhY0($*x zM$z&H6nnO79vhLpLn#290&8B*U6KHJLGuHOnW-dsGJ4qAu?1Lk13aGfy>AZo4y5h+ zb6Iw11|-}u-`cjJVxcw_2r+^ilj~fS9=AFvQj^@CHB*7G4mcokYGDX@4n9>pfKNe5 zXrz<#9ofOjtH~OY!GOR6k%N<1%3CCNs#esdw&N@B?^W*N2qY@GJ@%d}VMre_10-M! za%!Z-BvuMB*_YO;EKMlvW6B(>Yzw=&TB^}tm&gs!@x^J)Xi)A&41c@Qt1vqOz#I@V zGen|_gRw?fcY+Dy+tb#f^1vmE?QiybRk;z7a(3sYN2Nw^qjxIU!9KaEOQI{ET}=6v zSY!syae>Avx0b8BX(55g=Cr3qUzS2hT=f1~st*andY|EA(vwLQdyP4ek`4|yQaSq6 z<8Wf2oDq^p?^Vex${Rd#K&v+LsSXB9V2}q;D;Al&TDf5&!-3RihUi6TS;7!6%DD%y z>rq)@`T>k`Ta0^((})5@VO5nuIUeGeAqlgrbY$bvw4mXaNu>#dt1c$gxWm+0K+$Fs#?pX&B~U4l~79-NzUrtV@7M9=^3%N6b_V zdehvrAx!h0I*@8P2e0W?UNRI8SRT~`quY^4Wmd^8j?~6EH6*;8W15Sc9QL4SGI8u_ z;5Ko`N{=}Lo~_Tl0u+!q<2dU~-cJ;aFlbT8=9ug+QN=W5W1MEB;d;>!-s(TjDIi3p z{J3&H@!(_9mM(e4MJWbDj(sWduH*NxPo*1)o0?$S!IuEnbEU~E$M=9xD~8j@k`b7L zkJh?)bv7`pai-tKuSATRxd)4Ga{&y&i0N8?0Q@ZeGWd<5>8L+>cYPFU4p$m6|!aeu*Pygu5e!|wy@=($}p zQFvuwGl3xEC-A76ZjPIOEd0>+c%0vJmbso(%sK6v7|ZC>L;izoLk>fm4iGx%2zsrZbDB-J@{cf_7GlJJ-joRN@BZ?~_m zD=%E}EVnB0szo9{-dQ^Rb6m*2K4VE52-A)f*JU_K@3G4&%V?gH;Yn>Y_@KSi3~Hz{ zHxftF70rA^)-3&CRP}rQgYIard~$#!Y&^gEjoh zspgXGbC7>Z@s)kDWDMuIuCK!PY`01?z#vy9B9fB1-Gjt9%_j6d$oOTgE4F2klx-{( zyPEnZz}ov`bI8WgesYcSh4tpXO894bR4NeC@Gu7eoOQ25)&3=0>DFmy47;L24b+63 z9{69Sc(A*khuK1^sRGySJMp^v#FCE+_-<=h;!KN~nXw~h_(!HI;eQpTs%Lm7^Ca@v zpvFKXb^6u@t?>fp@8Tx6sZVml?HAHAC^#E4^8SLoGT+A%)5ZQtW})B#2K1PqlsH`#4CxD)C;KZyZ|$$0~N= z1`h|@iui-YkEUMSM|Eo&tct_u9mh;p)nBq-j1G?_oy;;U_cq|5UZqBJfNE;gge^N8 zVO2)uO8WFW{{Y*r?g#k6pxQ#p#@|b8ec0=@hJPbpKEPH{7#V&PeN@N}k>TyaBb3I3quauISUU(B{kyV=oSw=`b#a|;wY!bbU%Y%Y)2tAmNFq$`^N)!yr12K3G^@G%)+Isfj@hE^o<1q^ z=yErglU>SKoS*Ac>H0kPk`0PmJxJ!gFTmdjFFZ>+m4KAaa(NZ)x);GcKS9|g#>OZ& zHhX)0YPnKco>QpyNb^sIUkuek-(LKYAl`5}`c=P(-ZG9CYiK0eh`Rt@K3eI1BKYPl zBG3Ci$VC#SLlonu^{<@u&1PHsmUd!Ksqb00%v4piBh>X-Zlz`{dEgG%tce`4C#N)x zHW=q5dr}9>bvQn>BxtP3k%&Tf9Qt;qvE0CPJ#$IB0QEnG1cuHJ&DbAGHry=GV6eg9 zgTbLj*4jR9;F?l52L5!N#~l9vo+^Y*w~@|ERI3IP8NuugM+0$?xC6CZRt19*oO@E( zG-OcjQ-Ux$eQRe*h)E*kfJyFaILDlvbK4bSAkavJatA+3O$*TKZ&9sajlm@JHCydq z<1PvHroGe8r!tapf%(=$Ugk6;VD$ZHZCc^8(Cjo@lrA#Eis&51;I`e^I3)4KaJntX zNw$tbARc(G(WEj)2nqqzNU9PEU*ZOLVZph6wakpo4 z&OAS;ww->BlY&66WkN4>_@+z%!;Dvl-az(OOr&6`;=Nnp?xrua-z~ml9u7Td>JnF1 zC!P|W2VT_68Ja7lAYnDcLO6MA;!rRf~0;``p4iuivv# z=s`*<&t9jWQcf+aX!bs^Td1!sEj&SEh6O>CDErEK*X0lV76-zRXnz8YC0p26Wu^}n$~$Ee0f6&|T%P|2J&dI8(sxW55wB_|O#3QtdZ^qb}T zO}e<}jz9f%jh)9WH8K1rVAi^;GICgk;Pf@_G3ooDhJ1+}sL323dh!hpPqS(XyK3$u zCAwF(>6XhatR^fra1PVJtm^kYEF-dMl)l2O&OTr{Y~#H~`c^pFr_`U~9<|XWpDe_2 zo(TT5^s;3jfx$fceih1|w>#}~pBi$3xFN@(=bDCHHH)3Aka;JjE2@smouoEFKQQf4 zn~V~_sNc6Z&ougi|r3+bdC&b))1XQg&iY2`^QrC4K(b*LJ{db5Gg zCqDj_R=%T+(aNo_2Mi5K4Z9ggW*;^<$0D~TvoNmWgRV#;JXL3bLmXfO-1Qa1QZ1d- zwkt;-?Xm4amMxYU;;zC%o?GP{61{4+)58M5e@c-ZjCmvH=L8z_Y9^0DosCE&210X{ z9Anb7>U2>_f9f0bZ)WSoJ?0I~F|miMU~KQ>7Xj1kUjf=jDA6E{*_q$?F5U}rh} zO-~%V@VO#FPdVx5wUd%SIfi8RskQU9pl&LzYzfl+liJwOMh_I6s9l;Uh4CFBs=N$*WSJ zDnUgBhw*k5S)&eA9zg#0YT7Mc;!4P&ELaCBT%3H}2&xW%J4jg^x1#i|YluSO9d|GB zDd~>YNt5Qv0;AP0HlX~-C)TYP{O4nX9Sf(2VKVp6#zDB!T?@u@^) zNQa`gQAxWM%iWRMKnp2#7#}FlrEl9s9^_7gsLOC`E)b?Y!4+691Rg$>uVm52Mpx!; zoOG;`yCufT#VA0JcE;S<$UoMtAS|JsK-w~MR8@g~2-*e#&N!_|6c$c*2IB|Q-m&&p zTAIQ|1uBxB!(Ux*F>4ZQBEP1Q2iqXU$}SJ*yam7CnDji9@OF z)2nSxCi=(BjN=15;F{)c;#lQ5P%^j$dRKpa0;pi1G6S?T3|Bd*+F|g68++q}ShZ(+ z9gezdQ=5`7Mlpfhj`dPT%V2`Lxd8XAE2xN&Alkm9)fr=0$v{aedCz*P22cht4r(Q)h7@TzGqSS0C08sk!@u}e)wR-; z3K(u|4CDD%8*OlQL;)p$t($qyF(+OEkz5k9-1Xr?QQaL!o>{(NPdt0pg}GTUxxfp@ z*15Q^3}rVGL14J+k8@h_U8vo)caefQ#y#shTcObCw2j62R}9Jl$36c53X0-Ask?v= zL*AhmD(i#8u%Jvf=OYd2OX-XoL8tNhfJO+&cYTWAeQH!TCUR= zm=*!>KH>YnTF%1ea@2FVlV}HaIU_wPyB)~Vau^(Bj)s(6G^_v_J(PE|i|!k+Ko}cMbIn_0dQCYR zuw@9r+>AiM{RL>nGQzx{Eb zx7DNE+lk(yzRHwUIHeQw+SMIdIZLU#;yrz1)MpTrgZB$_kEJ?IakV`#rSt>!89LUPEs++N0|H#h0Unv9@a(1F zmeOnza#N0TT+XlJS+BKS(#XxU{JA_2YQJTDiPGmdf8qesc5G_nwHqh2W(4OXpQp8Q z`hx=GW1m{@b(rJ5MV=$c8D*{mTSks@56kqV(k*OKkBJ9ElSu3Yt(knm(2hymY1msCa;_UB zV~k?2TtXD{j2@Lyi392DNfd~H#V<}rQA^da)KX+n09faqIixtc!H~K#~e|xir1GOb;j<4Kb3u({{RHi zv5s$!I$gx7cEcOUA3YClE8>ij!yJnHpZ*EKXvy&&D}Znk6M_yhbNs7HOW(1}iP8t` z^`*#?%rds%MnM(wpZpX9!D;aa;0KCy;*TZFFq^0pbMvPl*U^?2&mEvHyGB0>^WTYg z7TS)Pbrk5XB!%O{bU7HV$y2hC=SF<8O#F2HpL|awejj@_c1KPCJq9}0v02=`$pD?( zc**KBUp`&w*Ix>LGWdtXjJr{8*`ik+QI0_e@vlm?)pZ9T?Z7`cu~ZM=Tn7Xas_$rr*U&| zU4bpTC#`eBoRZYqtyHxp)NfK?TY|vyLB>U7T;HPS3%EBOYL?opg)sLfIV5@3C<6tYei-W0YNQ{V-=&djJ>0J z8fkp9C`5!3I4z#FpLYkA#J4#es$&$8;RZn*mOUxAjpW2bvX�n#M^%*&1_Q8RYE% zfPHXHA-EFbBN*cq6t^B^U?~7+9MoQFfI#Ppt2r%9T)Gl8n8@dEQAv_ta^$b2HXvCK zT;rjjkf&~PJqJowGriG%(r%!s&vWZnT5PE~7&zcoQi6-McsQ%k5g}94j=18nlosfv zC$VnQ0P=tVHR#?1@SNJT=&KRkk=viZ*CpZI6Z=Bx2lu*j-n~ZS#nAX)O*+7q=Tl8%!mATt`9qxb z&qH0_$wmq>*&I==I+a|e?%C>IN~mB4;-@TqD2g{AV4R*Y?N&UNk&vkhRP8w6Vw6Q9 z6%QaT4)1EHK^<;P5*J1%sNfvc!EAi8TL9zkf!2mbbOp#?W7vKbWk5ii%Rb|`0CCp3 z)OJRYnnieFbA!2X1_lQs>+4c9M222lBRM0bXj|FD7&Kvn5?P0(NcQE?g8l6Lze+h$ zXQ55I+~@A?2GYSYHt|?ez{oPYombRn71v)|D=A zP@`vddJ9x@lfk?x^8WyMM+e%fJONfhc^z?{mC{LUk`a-i={!Gi0y>76s;&vi8RS-H!_R^(aPVH=6@uRdhnRW%X?`Slq74H708WIe zu_dG+4tVLB=4bfdb#~EN!zY&#fU3FTsx8GIA&pK|nz{5z)il2jSf#as#EhdGm-^S9 z_}}6L%LK5YP!C2^_r1+?k$8u7iq9p+SdNU`{uSmvHt~s^kd2)2k3mVc&M3u1#_?vk zCFRSLDLmI3Ex_J>ZhP}gy}xN{-xUx%40oxa zjIsRnYhMOJx2nNBp@_nj5H!3^&nttz; z^aP5H13O`uVvjO)Xtfb8M`51zY9WO?vJQBv(6`IObM&hrvbpX(XkzMGiwvN3tEj5D z3~F?-=qt-+e8}CoqQ7?9Q5!)RI>t0!-60!rHE7p8COqNJF10S6XmU-um^pgyWDy9_oqv7tRH2qmOabf9>4EFb=rms@Lj|{1eOef&SU>oE{I=Zj0Pn zE+8($?)nO~IYLTNCzdk2C$l<_*(2j%*Zf7Q=$e|PuBmN1*<8Xi#E8UkUr|pCI)mS7 z7qKjs*AW#c+=fONV1GLJ6U8GRG23+gfN+jMQ@bpc2Y-1lYpc!=h zvTN}o?k$CuRp6XwxHW}E%#}|2`khoK$o-q!rsFov=CddO ztQ-MOJ`PJ`z9X1ROe*~0G5SPrJSsv~+)v(sXWj9`Jtfs< zWP}H+9z{lkNEkQHcmtlMyxFFDF->YtCL|CDI5-$RO)PT|9gB528&4x1wMC`?=WjW| z9{#l=$r_9^1>~MO*96s(yjiVnI;m5;aXjastq3kbauu>iLDLn_T1dO!Bw={>tIIH0 z(;qO&+B=HLMH*b#Nt{MDF4K}r6M$;Kln8?DCHiAMs{~AOfWZcN4cqBbq=A8E1wieb zf-6|QQ9X7yKGWrg1fC8;ukh4uD#;@a*%->M-lshCS>ja)HpV_~diALmV zAH~|MJW|N**q?sB^>A&F9mTPp3mRs^SZ_Hz@=a;XnY3eafsX|`h=CxUywf8al?#U6 zy-!-a@vLA48$jd$I2BdgFIEama(#WPH&dafC7syhRR9t;16D2WL3R0o0Ps3ysXWSY zmI|xa2Lhvx?jTN1Tkj0?#Y|6nxlkP)}Mi zScTpX891nMR?zw?71Bjp$cw^8(pdBIo|Qu8)nrrmfl+;*(Hn=)G{7_ZX1twPO7J#aqh?rRp}$gpjr35<_XSjJ0Z zixX3w9E)3IvoIi!T7${n<97?5)zRGPs#Fos^#s)!Y}t8L2OTS>HC3#9>!jR_!yeOx zz$2|#hC-{7yc}eIT9M)V$AD@zjDeClJ$)%UnjW4T31@a`T1xMQLd5m?dwpwa#~xyZ z3{;Gttzg@sP#K6`K)^ilTXu5s;8I-jje6-+uv=wqEsD(XorPs)1J+H<(>KwE-}JwL{% zTig<31=|d}S(N?JR#rwvS0m-g&#CmPgK6q`wW)H6D-r=GB=e7IsVb=X!x-B+l;fO314J05%jF>!#9IK6bj)sdNK|P}yK|N3Lp)UNqVQmSc`;{E>N*24V>4 zaBym{iMBT5haCYKrf{WqWL6(5kcSPDGQ8%iN{4m$} zm;hAu+&z5_M4gS}sZkUm0o1yKz|R9UeliHhZ2c;!Lfd&Ef#J}8o0*1q9OIzpgIrgL?`70)i^#=<1cXkj^{(RX{w*`Z(@Nz<%!6uy)km#) zq?dUz6NV@6de^yx!+RyoEzg?A;@x@jN8v=0UauPuAamH)E2v&HO@Fpuouem_iqMkr zeAeB@Iu;nsbNZdZX&mErL0n_CdRIKO?qq3tnV#9BCmT*hL#bc1aI&e!Mh#%t-1%3s zp&$(Nj@32g(R`3jI#ljmx|ugEiHm0=(yiI*CM~Bude(j1j7jwUY7~&8W1#h?sLsS| zZ1E@!+ZYwiYPV)dNiDkus(qk>GRLoKhFLO86T6yS2D_5N!Q2jlr{!cPILYbjRW0TK zpOkg)RuVw+Imq464#JuSSqZ-*k~-vZ%`At0K?gl4VSMloJ5;#y2Htt;LlH)Fz#J2f zl~!fL6O0r2Rn}sKhJOmJDxBo>J*c*Tilugpk5AT^K;ZgQ++q1AibK%SMU6Zkts;QK znp_RMo-sfp9FCOz20k?dCY;I$2ACfW?@tT~IN)}tu_SDH1Cdwlaf(|3V`~ApoF4h8Bww4Zev~|=m=ba+fw=&5qWXdsCEUQ1^{>A_;GK3j@f5PK z`?078J$bKy;02`sjx%4O{{Zk#h!wRv&_GbggDiR;z*h7=Ba0V)O%K!boynRc)QV+8({-dkSc&Nki~jB(z%9~53As|hlyK^Oy*UGr(H9#*XU#r?WG z5hjnXXnr%cT#(;nx+*cZ`>aptUU%XD02NJrVH{~F-;ABSSK42+kA%DRlN>MQfF!1{w&NcLAMW4MLPZ@Z`?f-B{*^pw@tL+bL}Ltf)&ZLQz5Gc;@i z5(9gG`s1k$rH79*YnoxdW_>)k> zD{dr{wwUTpLXLCJO(RHEK*vGOD@3k&@^>|@ZsXneK^f%M)xFy81xFahYmMGPBpy2T zts4p2LAQA$&>C(nt8^hWZ0dT{N{XR!e(pykn!>xB$}m9Oy-jIJYN_|OjP%D^qjhS7 z_;MS8iiDAJli1G;RP63KsS}VoZ$7mlg8_Kw>DHQKY-O?3f(2HL$diMf#BokVCxE%< zze@6SAUtjIx=NlEU zig+V(dW?($E6I)Y_j3Zrocy3yaiHI8R<`D2D?PfV3rap!A6oV6VdSG^Wg4@Kyjk=Y zf&Tz&{YT*Tqis0Q*6gSsypCYnQ(T#PdPRD zxvFbzeQ;ZFlOw3<_*EY{q&(vX(v&jAI9g4ncSkyvFODiq_0QV7$66P}KaRS#t>KG< zbE2iZHqgpf0F_>G>MO~HbJrr6pyV8n!k-!f%rbG-z4}o0aEv=2GghOmQdJtY*|(wG z7^MS_M}F1x5A5q=ZN{$wLW-v;&jWWtE9XloRyoj-mE(b5TYL}GL>d&=U?^GDS2*jA z*{$!Tj3M!((4mLR0Jd32d}QXOjyTn`xNYbLIr`O!Vpt*#8EgT`6y}G_eZZW7oOaDP ztq%K+wzM1AZvjs#2U@oh29V$!DDFAMHq-8a4UwmwX&A-~<6@97IQ10< zr)9PhLljffo+|V^_pal?{VF|j9NCsq8A9$Eu6}QIo${k(d{^w8KRm5> zQeLM20E~6-B55t)lQBj%e5Z_x^ZV=OxMIYTX>}XpaXUF|?&hcyxRadH-*Y<&xa}nM zry$%1bLm2%Ve9NF^`)7Rj491l<79AKATCAu$4Qa%!A|EuTuzy_672oDe!zQ?oM@k&1@KmB{3BNWdrD(!vNN5$;VP zZw(xhbYPT`Evu=*Em$CyCK0D6khHtb@`@=sdMx04Tye7((VLdmlkz%=eAQu0{{ z$ioBq)!RZugQo|PM5?@P=bRpTQY$V--VI9BR$WfI*2doM-Ei0-V~X?t02o`FJxn1a zarLg6&O)(~86b><&OxqY#!l}%EW{JH1kxH>l(fk=ONKr3UZJ8sbnr3=7~;H&!eWwx z!LLs6+}>`TA>?csBz3CT9riQ)TX0RQH&r?O4SBm|oRf}gv+Lo;_=eUepy8 zWpiIg8IcBd=c%t*@XHpn3Zp!8it?=;OF9wLJ?qwdHyW+GWA5$a-ktgxy~k@4D=FcD zQMU)Zd8deU!y$Q*4neO%)Ya}Bmkc?^O;9>foNO_c?i&;<5v}e^@Z$K~Y0m+WZ;u2i z^sb)f468UfJ-{6+gwT}e9w&_$ksQYxk3ehDOs5#nd~@7V+d`e5r#Ennk}xN7k;k=n zKL~y$Y99>zO>3%IyGP}?E{cBg=aF26`~f6#FzUpU{?D~T5ZfR*!RdqPPo<3YKVx;D z4`_e3XTWP4yD9c-cCq=^cVLalC>;(uWQu?M<8qL%(mFcuda0Py-Yw^$UYsVJa^r$U#w^^<%OP`oT z5aT}e`8R*8$*FkDTDzIcokH$u*pZG|$^NwKr%q3^D-R_&tM0C1Xu<5Rz+srfbJ*g$ zy-Lw9^hrokSb$WHYm_jE?ptsqAH>&ZqsV5nNJ$$=#t&-cTb*)x9&qt3-L#Q7A(ta2 zv21;P)PJfBJ)x)%oTbJ}$2#-XRA zgy0--Kb=>&Lb5wAVcXKVuK{ZY;?zW2cd~s5Ljz!6j#N-Tn>ur0Y5% zw2|{IvD>$)D~>DBc9Xbclhb{C)$qSwJ|6K+yplG{X?{r#dS|9<>Bhv5kjn8eA2N;` z(z&Lr?0c9=^Fqv9RAavbBNZ${LJrZ$4M85ZS=R3sbw9=!%lW zt$shTS8BPesK_gl124?xrays+grh`!gpOk{D^0_z|u2J<5F;r*g zE6(0CP5#kj$0f7C=QKFCpenq&ot#>eWDXQAa(#Z3_^#QD65|Ku$mHg@==EhFBOm^? zUH;QCzX3?jK2y}zysxN6qHOGK?$pQ>63U|#ua-taVUw2}$wn&TtYiZgA=7|&nO z)w^9x?%W#~cTb{J6CU0!Nn-lw#Q^vrkvceoMd$CR4!c^(jvi+IO7=i&1X-j7Ril($>ifTn?AL; zaL461Bc(*8t4YaexnM%evKCOn^aPKJIahnxgui z6pi~-au2<2PG=QZ*zEM{q{1K|w_mMv7I%FVuuw-NbIoviW%`rLbnWR~9ktd_fJY}a z^iDcl_$bwBo6+4i@)?+O^KKkt)~+jw88OqY7mU^B zcJ|){WR~<8toO4L0NC0wf=5GJ)2J*X86P%IYA+!MKsm`gj`hhWV2t_BUh2THEPg@N ziR5Opr?SEmj2@U3)ZE6(tfXWCz$2;8TF$$*ZN#tuAp4+n%^cbq(~55AGj(PoZbk+gq_* z?a2oK6OL&kgA!w489$w0v$x&CoO8%y(v|Fc5J@2|;Abnih9F~}m2TQLQh6jE26|M< zZM6|qG2cJqQwTTaBF7oQ+t=`k08`T$>5r{%%zVxv*hyd*m6@}Sa0OP}KvJu_cgv75_?pm^?k5495rgb~ zDwN@mEJ*_cX*fBqh&5|i`MgH=MJaOgG7K>X<-x(KpDQyrA(xOr<&9os1gi`ZPBD(3 zjaL!wlz^p4I0X7tw{@ZC+o;kEZ{{-;+kwFQ-&$nKlBmw)$546`R^AX$V;i%K;F>`L zZO075pOj;0u30{ON4H`4RMuAJTZKkq>ANkS;-_fT?k95d)b%v&0wb*5 z>aj$@zR3}1!RUPly>;QK^VHX&#f+)!so8Zo4-@N#YpGz50V5;2&@&kM7`Mg0~yVI5?re1%U!p2L95*uow9MxGJAb1I^yf?Hu=vu z#zCsu_1K!=kO|Itta$DkEcafWdQ_*-*K-q7Z?ozyS2#UKy(3(+q3!L8Zm@GWAaZ#= z)m|wuxhFMfcM?YV+pz}*p^?<^M@m^3l1_cOs2t=9T}Fp9ow*%7X^cqcy$E?eo|L%b zC$Fsxt%ZfM;;dS!XK+1pRbgyN_|>a7+c9CDyyA}`GU_fEj+o}ARy~^;JbKh_y)nj@@ZHB6kvG401Eae&(w?ZpU-!il2KPaq@xG_o$c-cJ(yCt0In188uv~9rIT9 zWSsOJ>IYMdfk|jl67V{H6cd7Y{3+eY&(foFj(XEf2MwfVmvI~p2ct2TH=anUG4;t|e zq>@cBbSkBe2;#pz{{U!T2oD4NP-uKA%JEh^u(cgIrhLAa&36!Oz>+!&&$(didyLj@nRdx; zH+;Nw?^o7iBxG(FJb{i)eBD@b%INy1E@X%$OcBQ&aZorLhdB1D6J>$#7~9gS4E z&z^Y!dkplfy*}yVO^v`j0h+sqEyx5`Qd;O~eTI+$I6MlgFEV4){&hQ!af}01<^_u5 zgN$%$64W{mW>&xe4)oPUX*1J~c>O9M%N_t-$D!t@f;XAK3UYc?B!hDPl><+7s4#tr z=DPH=Hqaal^Mj7{Q$W$=g#_}UPcRTT=sV!n#;1QBm6J~UTWaA{@y&2mpS@_Soph<+ zR$cggpdS!;iD8oxTieGa)X9wEIMA^_rF`pua{At_cY1kg=4Zei26!KpeV6+tcvkns znum(6blXP0)J~Z!FvI~rGx32|zisb;I!D8Q4*W6HZ8Up*6I{DVt@Rjfhyp;02x7ju zBxbv?mElfqo86v{BMT=QT)j`5ei8gTo5lVhw}|vDz#}6yq4CSb7FT{EZx3iu#|D+9 z0SwYY*%woR^S96*g1sY2u#e%t!5wqObAc7}sIsssO0kwsGQUBSUo^v(kCIChmUGme z=j&dk8ueuAOJ%V0ig3Wvr-zTm)!*oYX;#M=6*?dt4m0X)g z3=47J>raoKG1tC&RC!V{w}5(zv2Pyq;0hZOx*o0NI&CLCYt?=l{8iLE39V?m?y=$f zr!ikf<035eap{WklAt&}N#d{SF}(gBaTIZfj??% zXnZLoJ|%bL%^K5m&HjOw_ad%YR$0t20 z%%xud518}2fVCV9$gSKrIVXTSijyk;05CZKaxt9Mrgu}lv@2Z1rHDBT(<7d13sQt8 z?65rX>s=+n4<)$yNX|RgH>t>90~^5Tr;o<4oS#G6!#lGs*bgv(ps_jNb*TJ9Z~p)i z@P2HC#`QSB#aq*3CPJYw}zZA8_lm2^|ayyFSH4BflTsvcTHKN%3+OP2>yvptK zvCRmepK4Wzv8|mqNs{5$YHsxyrj^WFNQ+Of^9bXL>~x(F1P;mo?eAA~KMxp`7Xh=+ zBc)BMYKf<>mI&UX0FlzP+e16KJ0d%`iU{`Y1FsixG zWKr1E-IZ-LBe|HgjK`e%R0Z%cO7G4w&q`*}r<#?f1PANF7B~ySRB+ocHURPsokV+HVv>;PIBmahizuq2&JnR0M13isI2^cO0GvKq~w| zzz8RhGeW-Oww52_4Mo~$C31d7+yeHmO42nGf26@S1D&7}1$m9i8@L=}01TchW58PP zlMJ>}wn`t9*ieOS4#M(WKgK7Af*AQOYf|Eu^&Cu8DZ~PPY_Lhg@_rufSeQM=Sx2DD-B1ZY7L%=opHL1XE zEu>S<7a$Kx=)Y(G0E_yU!M}(aX1ifARY?NOqmAjFL8jMWx%#2;kKxq*Fz_deJUX(p zz9O`krKmZ1rHu4dAKbzHEAf}c`X&DWg1l{cr0Ouez15||LdfGBgNpsd@iw&o01!MC zqkKZ}j-e&H-`a~i$F?JQ4bz|DAousL&u`n8_G-J*ZgfpX&_{W%-G+OM8+S8K6l3Rn zWaN`x)#Hf5-L-YMe>8Zoz8@3Hle7E0kCk?;s}zR?i0l3pwWP`-lY(0vtJZ!K{=`2R zejsTU_ZI#Uy0*4+v`=}67oPZTarjr$o(ul~f>vMKEE4#)#&%W>fLaR}T31%|EO0Br zl;Z_<&wW#r-J*}rZx?C)eXpFk+=RAqT)oA))GTwfpO=tDe$QyX@J+9S+C`(wtDR$8 zz}((s{#yI|sw%zD>|y&b$QMqR!-mdTNZTUb$gPvp1fS_#uNy@@S=9_JX|Hi?elB=c z@wDw#RfYjL=Dl$b+EOwXB1Fy?n*CDnZ~POx_HdHsOY1KW-D-;yylgJ7o#V>&I2f&I z^k3P-;SADQP2oKj)<{?gp)eWce@>*-&lg4(lX^2+cpOD*!ZB^HBl34o(Qob@TF%nq zR#w0;M9fFv1${yA_San3{5fW1(seJj#GX?FXC_>Z_{DwQKgS;gcqd8Iu11MrqrKvn zw$q}uL36h_1(Xk$@vLtPe%IDF+Qy+}cc;f1z~G~y87uf#H1iq>tNur;hsrULPEn5D zX!*ZO_;KR>RWk>MbxBkvShcGS-%9GVui5*@dXfOX8oPiJ2I-2f2jXk#h`(w-4ZKCR zxn%_GUJvD5_lo}jY_AV!78A5~@=HEHcDe8GNoE*(T|dCLfX6q>$#0rGtv_V{03W*K zTG*=NJ7v2Y$@~T?o}v3Gd_xeQ_?Fqszub!Pay|XcR+Ih;1$6dEO|7#sGX^UBo_#Bt z*MH!ldb?UqmUi&K<^EZ7mQn5vWb-JsbryB3vlSO@su7ExN;bU#Ic{p zt#Z1r?6vVm62-g|dng?m7e@UvQd)n(LOeldt$EKX%^8WJ+H?1hMyNFp+rPznq+;GH zr?JZdXN`t`$*kp$O=|{rLockJl_PJN<-RQVdE(CrZ~ofwzOknuY;T&$_JiM$2*r8* z(+PL9Wn?)kk<~qoeO;jb+kPvzaF-f{T8tn9%WwnacLzLIORfAp@Q>|j;Z0Xi)AZ{< z4*2fI+3sMxut5^skTd1TqlV5p^%a`03CSy2owLGLqfOLylveh7{l~<{;Q2}e93FCN zoVU1Nl;gHd3n7vmwo*Zql05y_2Mj*~D>n78K?HHwno^Ft98;Cm(8+ZyU@DW?d(i5NeUOBJ%|1B}N|UtYbv4Z> zH>v4Tskd{hwbk6XOatk+oK>-?JE#&3{)0T`xj3D{Eg0uG1Rg5%S5LI%xIAMy{415K zGufuineUhh2hADVxRKhZ+}<Y^tT{mD#tt?dR1mS7%mPm&fe9m71`z1scW0|H*6U0+k!H2 z-!;+c_e`J?LvlJEab9_8e9)`z1p6B3^y}=z2_;D!`+8T>#pr&1&i2mtdjqr$r=uQz zm7`}nImmLG&rY}^xcdv$3$=P2Wc26Ox(f@U3mhqJ-91eWqiS<|Go*q~Bq+&dURj6n zeib^+muFn>as~nR^sL3XW``KVoRWKWsY{}kEAuG9HODyhJvvoiLdDG6NXW_QkaJ6l`T$E?1dj343a%-$*7{wViT2-0FQeHRUJq?(?azqJ+_}( zxFBPUji3zUrb(&AzwXo!H+8N_q_szNH{9ngEsAs-hy?ST;#zxQroL;E0k|j&T5s! z0Y|CGImH*$_UP3zz?_mFlT54?aAdy;N#M+_Bio1 zRkSC818&0<$;r)56+-0>2Ve(Eh%OPBx?pw#y-K@*B@_U7t4*y>nypiv*ayk8f8IG@ zDhVOV`Jco%2en>~GFR>b%aE;{nrxAS5|XZ<@^>D;g=nX3Pa33Lp2l3U;4pFWw;+m% zZG#>GBLf3G3cDl)gYZ`b^4wIm)>}!{sUGz1K9qh~rtrcTuZU#A0de=4J{b_9V zFB7p-2rh#uf(}Lp(AMvX?v16wLZE^ax5|Ah$h6&1+iIG`Y#f>6Cu+Cx1I2wV1`>@v zchvc8!fDmyT?l+rH+N2WZJuU^7O!y6bV$Gib*hsxh>JEj=nr~LMnD@ReAuju+p@{GkbQk> zx|BOn8}P(nA4<(v*h=$SF}I&&GtSdgnR3})v}_Aejq>%+)|%fk^yAi_Ajs*+?NL8a zGm+>hWm@cr@lU}Z8e!zpj+o6CMOY93Pc=QO6@I;YReF3!4l~k$G;Bu4W;+h_j0%&E zbJm#)mP`y&v$FsUue};>K%IE#eN9CvDmdWcoL?j6Jawk9EDy{GrM7^`h}-}N4~mXJ zLUWp)LBf?kH9+K*Jdy<)kEWzddFKY0Mowwx7|uDSIH$2lYIVkNY4`+Yn;6R&Lx$kHRky4NE|^ZO0`P?j7-7kvyMiRtnC< z42|8smF#>`O&(44vpo02DIzq0?b=BrBavQb;%z-HEu=?M!PJanJXf%5aEAnOzc>f! zTpqn>+gr@rg!`D}n)B;wtc{@T_dhzmCiqGG5#x*7bsy-LE$6_&a158u?z<>vG_OkU1R)J@Z}_eRV#I3sc%c&!K3Y z!vj2e@l4oqJLkPUn&D zTFkRh-1P6qTJ+zCnqrwm%@JeNC46M+OcJ>v*Yx=w2 zHa8I(o-tdVJ-PcWzp0<38a>>t8eH zJYy%)yMK!R023wn!SO%EJ}|q8LN%M$o+yX>vyY-L;4xfOmfl&({447y!Bc{X`G!9e zEM_8AruAVNBy-ojO3nAj8%IiwSLMMtKD90nY#uSqZ)9)?F_D0J@=a)4L|F*H$81(} zKS09+KZSHUJ)oQcj!$X?^e9aOB4_V5Ju2ao$?)y-@wrncv8@~XBBSLQ?bfJkN6PS( z?iZ7^9QNj=jwcGYURoI$gUEU*t2XSw7d_7Gn#+PFmB9ldwGL4T{U|SErSR^*bEMtH zacgrWwZu#c$rE7b{443-h99*tFCCYRArhe>S5km{z#azW_OFlNx?~`Mf-7%N)L0O}w&Pdp0xJL9K=uRLw>en@TN z^AbB`8&GE_*0%f^`!mmKW1m!Cv&7v;QU!EIJd*}PRhNKn!sHAN zm7k}K$%B$lp{=OlE6``qgGG}Ep;7`>vPlc*Pg=VSg=OTDr>C_<@B+Y$xjjdv1~M2N z9CpPU30g3bXs2a-VMzLNM0{{RHS z_^snhXzn~g3V2sdkcJI>_?K_{gOOif_#^%aOW-XIINMtB7OARE7i)+%#H9YD(o#q{ zUqka6cve`6-MM=BW10!#{!^2>3T`BwiQMV6d7+3oYXf)5r(+ob&ipN}`Ir#;LpNc$e%? z`#$*F;eW#$?H9+l5Ve+|fTonKx=1oP+{1uDO-xaif2u6O@aG~>oUp#zN{jOsE&6?XumMJCN!y&mVj+phYDyCshDX8ju7<{^= zTRT}EiR16uV%x)d&E$ce zMTr+AlEn6|{{Z55#{FkOvbTd!nO%`YqSi$$9OEiT_NNSFqokr`{f`vy3XJ-W&&M7g zx6y8VJ)*}hi)DCOQdn5D*<;iL(>3w$#E*x5B={HcPJMsGH<9>@Pt+qn**b;VVk=|c z%wPHF2SHk21wJ-Gu4=~O=0atXOztC<`g&Knd`h_R@5J8&Y91EU7_2uRyn^O@JHs&L zU-Ru;v9EU3N9p8ywCFl5Iy*i5{%6Uz{w~t*?(Q{P>&Kc~f$~dnA!+CO58+>G;$iU{ zf5J-~`d0A^lBA9S72o(1_DnS*~v;5yU{)hk`-wO?e;0ZxdZc zZWcR$%n2D&k_C4+I-a58yC@d-Q=9Cs5Md3l3xG!))eAoWcyVM$mfCBk94We#wokYn zYpGyrJ#l!Bm_x5%K~{eqHM1XIj1h!>33lV+A}%t*EP#7-HTCn&B(e@&*+;GoM)TP#WTs`n>m$avDB^q0NL^w^Pi=CDf>74Rn>eMZGC;HCC#eG8ze!~X_2;^ zS6O|$i6mDzKYN^$_}3e%+nL%%EOw5^t5+>-H94M+9}!-a)0OUdz0@94&oSmnWoev) zk-do|ezl8p%k#Gc_O6>bvgCt+2m3~(78oQh7x;63U&zo=tBXLRSny?NvU=Tred-0OGZSN1IxUz1hme zbG}2nf_WZ-wCwLU5DcUa)q*lxB<1tQ2U;SzE6zqM=o9RIe$}@-yKDZVa3F!5#~BsV zY5IZ;Fk(=i2(L1*yun?+H>qRS+PZ5URE(Y2R_X!hYG+aGIIgFmTVIJ-97x2RZBvYU z)}_3|1UMy$?cTVJKKU4ZLL zT+OiqWlwhe>n`qP0hyQ*&4I^$!iDB^#!4j_NlZTBk~z-=4|=6=%<*vBlhUtE<&Dff zX&Zyie+tQ+nU#PHG5jK<(Cvj@RzWK$`F9*Atr4>ri>Vx*Gt(5=bpEPaS2{^#3@hZuaoRE0I=sjyH+5E;;Jg6AKttfs;7D90t z>}nc!vCTYJIxaxP0nf|<;B*94!2ojHkXUi``c-IMWE+&@10)gpRhX5SV36U585|G( zy({UfI9#sDBaE)x5s}}w<3YC%g;3ZB029Fc>dY3;8YsZv`t=ogAxRWA7jYf?R#BN2 zze6G&BM9FwKXjk#Q*`W`DML%M z99uV%10CZyBOPfcg>9iE;S>VQ{dmP|3t{srfXaFTdRHa!M^^sA@a@4;8RP;rQR;vC z)zt*&H4_@GQ>NznBZ7xg{@vHL39tlFMpixVk&Jb(Bi61O!(J{J0PIwiIO4h;4_X#C zm(eH)vLFn@lC{V9y@bELPP$O1Re_XtvPicwAx@} zaOqhRu3IBJedyJKEm_JG6?2s%pzTmJfgkT4J5*|XvJQBr$TyO4iV7#f$@2rpr9&cs zPSJpAisc&>H=3*u9MiaLSI0D?tt)Uyrqva=9+c(hZ8Y?!50tN^O@Wu~$QaH?sTDH? z`G!x|wLxrn#&epUR@{Vi0CuTu>_(R&W(PRyP2LIqp0tePZwDvZnlXdUDS+(brxeK) z7AGHGX-+YYp7iVuq!al1(+;E#es4}cI)wD8oqC>_s5?-!DIw7uII9m6Gj;1#;6J<^ zX0`-j;|CaQU{gtn*vgO)*lvQODo0M0eprUp7{hlJS7}^&7U9^YC}p#BuBF^^gR z0;^^hI=_c@ziZbKNC#?`{Bd5PaUH}8kF-ds&(gS0g*L)#`x=I2|kM9W~6>(e59-JQZHm`Tg)$#N)#{ z;#!t;ibvcRr#<SGl^>FDBG3-Ol@qmoD%1CZcopj>mNj zGQH+|*NVSsyRR8sO0YyUy*LmOFTE>|R&VKCrmc0RYxbh#i0yC3K4d&AkM@VHc?M28 z;-wcdr~pXV`qvb&tD5PZ>Kv}|JBxI^yhHX|Nv=B^k`WB@P^rgJy=WcrXgH++B=2DtV zN2&B@!GDGF>e{`rSUlK+bA!n~)uH=i+$6sc^j`+rZZ=*Pum(Y#gL5~`bM6-czK{Ku zEOeQCCf+QZBszVayWGy$;RI?~x%?~SJyQN}8hD3W@ipNsb*$dWdd0ehWAZQHxT}ig zh1usuTDh~&CeWIDe5fF^2@$1A5wu}8ZMy}D+-@8J%cYpK{gXD=fFM;Onwcuu$B>+cNB6nAQN zNIpg%bdT|`Ncbh<(PQFEIOJt}V8?Q|7~}a@&Sn&67`UG14y+|4-o@XFKMA!RM_BW; zn^`VN2QfwoKZRj&djjib8x!?j^t;peQBj##dfPBE0$r_5BXB`HRmYA>Ke zCI|rW^s7aSD9PO2u~d*0eCL9B2CPRQT(RMQhZWOp4o1hu*ngM}!@g-=UJ2V(u8Brj zI4jWOwL~Pw0b+ecD~!JrZfqQe^fl<74X__&Y20Ihk9zZa8AOhuyB?M5ULM;XmK%0k z)YG-dY_F_B5>~)2qd8D{u4~2T$?)69z-Dp4uBTIN%1$u8zV*iV+EBW3FFD=@PebcN zzd3_IT`v?3nov}r&CqAVHgljFgw-QCC)at9e$OV zvUddIHEARsT8!d=*5^TCeDTOI!~@1Frql0~U;qiu2t3ybZ76rf2pIdvpsU~83vD9# zK(mkeX`VArKs^@UR(}uNCE8ufV=6ZFk|!$PfE{ao>-MDayIHNRrllFc+4iYn4K6xY zi5thcxOrxko>G4E8q?FF^PC>0r;;t*l?s!m@Vgyd-^4!_>TYJb@gAh^f4eb3pP;Wo z@E`3B;{N~&N9EjJKD%eTEopq&;~4o{sQhcmF6BrBXs!uvpjQ2lhS~I6jao~{E$yLT zjEx}k8IM2~mm;-|KCw|gWhQ!jpBe0-@ssP8c5Z@YR4~ZA#(Vv1U&Wua7Ay)ym6g=w zi#~D3-wN`l|z6Z@s;GBHN68jpLTQe zaa?S(yzjV@Pipijyd|dIMZ+`%3}tz&seB7@Zs5d4%xCU}7&OwHC;-(z;7O2~G$>1CySBbTxk4!Z$`jf+7#CYsybIbDLXbiY@@aHB#xo91ttClfb&H z;}JkeKAyEe{50_dtO3)QsXV^mFgRfuCDc~TF(dOfd&JSsp5CC$u* zfl;^sSHB*$+xUOsHKw~d!>4$1X$+0DOFNPW9Fy1&tx}E5ol97Lc~x_mtF#8Q+x4YEXp9DsKoIR?1z7Wfmx`<_c3 zEfoR7Ff{o}bGDW~a=3s4xG6aFt~*_Q^JB2Dr>uS%Lw&f55R1tV$Of*<@KeK(LK^DI zIj1XJT&*AY<6#zM!=D38CFxM#E1_c$j~DsN>$Z^zVXx z6}0mfZ7pwGIauUkM^Dz8OSn4Cy%F)QnQ7)qrG^w9eXFXH2@E1QQb`yN1Zqxe>xg_G zp=pSs*TdRd=AkeU zX&3Vb>9XC9e@gbxgjzs)+r@0#n}>!TINFB<=iKI|rZL7XsHNz?&fy#!Z+XeTpZQq%6T#o`N^gbU8`R^v z)b&kMT#7aXTgX-*-|rF0uc-V9@Snk73+)o$!yX&cw0O{C%z_cT{ao|<*I$!a_p!9x z0Cx2?^HuS+9m#8+nA0?gV}dUWf5s zOLh2oE@UZpqsG9i?#bkFniQnfq;{$?j?qWUA0K`sB?w+-I6E5|WqVlc}2AP$2*)#G=1(=?<9$Q?6Xk);-{iuRI= zw2p&Y*KRdkpJM~tlO51>D$Rnl)rF%U800sX5XN^AGe`$P7@fV1E zqC<-a5*VN9oRBM>*ZdLiBTiJg@Z#!`+yH#J<-06*b|ebtYYzS?YIROk)KZLWyvz>? z>i614s;1Fc_B#eT*Vf+zFX4ju6U7k!0BBvf-5spQ3l@IsA6`3G!B<`&o5NbpTIKGc z1eoN{QJyni<>6nBHae~SmY^8h>LL|}e=)#T3OFLT>e7nUA=1Qqyu9A0=?{h?x6*Zc zYlEiRUFlEqJ-aUB7(c`_)~|d@_(P?5lGy6{OmTVeDKBX_#B(7ZGV$$R0q}$NgV1Nw z#Fv+;bRh(?@4^5&2`9l5wvxsKu5)*NljMUDXV$4c6@nE+Mql0hfEaUbweZ9dK) zkM_x-ytaqz5nVN=IzhAgDjCi7+l0^neg>TB!iIm(i9>U>O`VOCIG z8R+S9+7aNYTDmBulhpmiR;)qdJnX>*e5a56g|dZw+Z z)fxShcHbpR?LiowRxPVv;;B7%nv8}RMA_yeAwJDGBfW11{EPrEsm7nQ1}Y>2p$DyBu(1&{2HM9N1JbnZ^qr_xp#@KD;~n!!%IfBk zi|leUY5Ny(2Ml@v$g0M{*|2$RbtkB<+T%#%5(xt*o`$M6k_#ySk<$Ql^`@rYjF$&{ zBy$4A!5Hcfdax~oWCmQQ7|QZ$5?FW)ZG2GP9>BIor&OqdmS+=)lWjMRo!&@d~ zDggsO#5!iE%WE2zCxAO1xUQb+)GtMhS+207JA%ES0h7H#Z*i&8*gZN25 zF!!d1M?ry*qXRUXmwyMK!0t_4cRB0Fsl?q|m?U634!FUr=VEsO!T#-b`ks}z2EZNo z>sIl#sM{OW?+GLf7DMQK_< z%F2o{(38{aM5bd1+3Ig;?5YE?Jn?{fS4nAehTxXS>)yFrSPCg1`JKgX+fC)B?BSQF z70E4K?s^oLE~igrJ41|seNAuJ-U$HvsxjY+;$XQ}RSCEpgUPKZZ`%O5O}W9(3{Pt2 zlpAMsDp>4=+BzMff@gvT25FMr0Q|PqJqYy`gKMZURwSLh0bWLGJG=87x!=!Xr;ny9 zHBI+ARa>o1Te-k*nDTfBu4;ty7T~cb(RyaHrPR+m@OuH?t9`v6aRjjkjC2&-9lhfH zv{TTxH2k`^P%ol=^1C!k}T^Tj`9q8r4b(UM6) zK*OGQ5Go*~kg5ha3_U5aO@(p0APk>cnhcg4246$POHFBGCtsUZUbP-r0CL3gPds+5 zI3f8?;6f?LInGUI+K>}F!oT-VQ&*9cI;ka!9(^)vk;&YpR_exur_9W~9dJ(mhc%@R zBm`|Lky{J*Rl5_@-mf7?IbWGnW7O9) zmthGzv+KrB-IMj^!V5IN6k z_3NETTg5ZUC{h%#J9rqcmb_c6NvvyEx2(kZ(jL8l;18vHnRPB|yzimmX4U24$Q56!%CE9x!O`0b8K?mo{W zD~G7(v^uPG}3Z?{b>wxw?IFYDk6|+p^EqEPTqD5el=c#d)RE43Q6Ls6H`vv z9P#Z?9%*h|M@Pf2*18De4H!7b70&~L52bWgj!Y;+XCtAeDUeGZM<1P2X!_xDYg$Ah za#s~fKyq*~L7v-`N?>4sdE%N>1JLo>tU8cEG=Xy7p7kYmJxAdEnQI+FeS>IaRb$uI zy=z3ax0Ws2eA#-A+iQyWL7<~}HaVT04aLJ?s-Ktv>t2a#XqIDQ#Ljl-E&Q`wl_1=% zdDFX_vCivPZ)2&079wT&fd{2C;t!9tKO5QD*j`JwO|n2%(SGZ=>h0=kPZ8M^MV@HN znF;DEl)aoNRg7ajMP(X#TIEF}lhzV6ib+BR+A=UPUSTsAm6)95lbZB@6&H(7lP$}2 z8Lv2G+t#}2t0T6Zxki-$3Q(n~Pk(BBXi3g7{cBTOk?c7Gv97zpz7CUIoI5Gj@4yHU@~#)G1Sz%wPbVFsN!-~ z`UG~$V6cG$V{1q{sq__1tPh>#D8N?QTRAz;HSJ#%JSV37EYft_RguoEbi6>z^Lw#A zmB9F0!&iFM=#CgzDI+qT0Qan67tN{1&vkR%d^h6@j|u+Bn%04AS#JD8sV16-aLU1r z(=hhH1M#mdYea(AYJnI~6?TroyMGw{mcq+L(rrO#^ynLWglEi^y0`SMHbs*8wv{KI zoib}gx!!@z(d=aXiKm&c3l<<(Pp0^G9WKl5my6|-_jh{wn%MB~h2w$-ySdCrI6?HT zYsY>pE?Uk4268z4Xq`n$Hd~3=`W#=4^+eXDc$GZJSLO#F%Dn1pnY=%$FWFn?1919R zZyYp&A6Gr=2gDX)dw1HWa`H3Z>0FXIhD>J|tD%I$ z!Slj4XQHdrsT($Sf>8T7C)^Gyq|kzgBbv{#zHD+%4;if`kT@&bu4~b`gSuv9pvE?k z4{=Ou$^$ZjBTNcl#3b+0|~<88EsGq>l-O6(U?A`m;U&P93OjPE44xj^4^5=AB|Yf}Yg zVx$vHX6FW*M_P9_u;C#^#%gqsvhvtpYO1{{#k+b?I~PsK=s1rZ1u95e03Hu+^+XoQjyPhs5E+f8-k-nQ1@0l_tZ>QkGUFvIu($)~2PapW`+ zH`C}$3;ZB)f8P^ zH{7t#0Q$N9lx$6n%_`F7><6}o$%AlTwA{W2)K^b&K7--I98=#uqo@qCf%3%|^aJ>- z%(WjCU+Whr(#eQpBg%m|NWka-tcmXCl6H9{EbMYkE3&uT_9(s}Y0foe{41QF z$6CGnZ8NyfTp!YxMDQl5txS=_Y~;2v#e0v#kJ)7WvuvyWr8J|a>p24oaVi&;qUCN;y)CXmg?49NJck#s&Z@UF9>`z@UOyM7HvxF z?6E@%ZfUMCta<7)UN!NH_RbL{#+Ts3_H=xs+!MQZy3u&YViz>^y^g}y;Mama7`!K1 zxVN~zWXjtH1-kSnwRJZdgfi*E+Q}i)?WFmYZC9$1(4O`2Plr53d8>GWZCg{glIrF8 zFDn+WQ$G`~gb_s}NUIXzSRN{GB6&4E@iZ+0(#!2v*T}x9jJf=44^#M!;cXc}vsq`8 zA#!o~{VV407W`4UxMxVD&lpxG`d2p=yLGEvD3U^VbCc^)@3vnmvGm#apW*wN8+=nk z)CBo{Zv0fbhs2);+*ve2-c$ib@Kk5;_2#~Mj@-_Fix}z+U7FS{0MtM$zdF>r73qPv z_p4mVU9nj=W6|dLo284{W3aTE8zGRbHw=FY8{%D>EMn3YGNHpeu7Aj{9ygzCpKJ~7 zpIYA1JU4l$JZbZ7C!RXz-i}uFxI0H<(!b$8n(1~%p-9eEN8ZV;Ehghq)XStwM2s_* z$qUUh!+#Gx%@Fe<+Q0@UJb_-RpxD{k!?GofS8D$7_03X>irmf1n`B}5f5EpCt6bcj zfG`D{lHRzlD*dQFD;oelA9#JlQrv};Ql1~b6n&%mo~FHj;-AEuUjTSE_ga?WM6nE8 zz={60L&B*3wegpLEN(ni<5a%X{&ekT2-^}mR-8RfN2KF#yx!5>rUT)wO0R!upPErs0dzy*Gw`u=s}`tQa4PWtf% zo1w#D0}&dhR#VU)%DVjl*7XfS`t5Jg zBS00mxWcE<)(^*D+8MN+Iu8%{Ys4o0C{cdd47@3PV~@kVarZthx!13?tv24~Nz39w zs@&l5fmggm`!$VIOSF4?X)X0XFMOtF+mJ``6I7uYKO=s39k)D7!ny{n@r%bAXNUCI z8TB1j0I@c<9BqC6r$=&-A?9zieFCf_lNDR>?4Cpg%@m&bC*83 z`qkfrUlo?yTeiQ{ZXr!SOdmUlIIh@rH}zZ;HMmz3|nwwWo)~g4We7qw?dCQ-dI1bDUQ$3X;Xtl{@X! z*1j&T4$^;dtMPBg&GFCT{-@%GO^`~l+gheDOfm2Y_a`;kX>yBeXq6WTsk8;{it{f9 zd2vi#uw+s4Te7Ht> zq6X;{EL(faq|)D*i}|jh5!O` zTR7-w7a$e{V?QuZc>2(n8SmJVae<2FljVAyaCcf2FKoQ{fCC`llU)7wo+TIx7aS9u z*HUB%2?T(8sUOO!Tw9SY#bDeGqzrx)=TzTQ(5CdU$IGL}&Pc&O)7F}{2OE?z8OKhw z*QS~bhvf%$GRL(k)0tcpQZen;xaCsr&7CR+6_%EP_b4(R;I9I+G{CQdjzQ@dF0wz1c4eB{uV#bRaq?;E@Wk`15Z9HW5tsBK< zh?2XB;A4SVB0{Ge9!>~371OI)9)upmoCPXK9{9#7=;hQfQME@co`#@p&OUAwaB<$J zk0loYdiJVU>P=LdX65FdHn}Rx_-7~5urD;FZKI9Hj8}1aY!2iO$~R;Haw{V8$&`_| zdZ#}x^{t^c_dY76G;wli%ES@K2Mv+wQ`^R>gBS#!3F<{`O>WXRIbx#?^5UY76-O#i zV>t)CU$oP64`$S~BZdMG%DaK*6>=yWBP)=4;P5IyyCDD!73Yrm^{2+|<)_G7w@OM^ zRz)b=az(-&tB;uabf7w+0Jj;=eJYvUt+iO5dgrAQOY^Y=GWW+!Vy-KyWH}>IZp!0m z+wZ%ADy_})jFw_YY}Q}e81TTekOv2kY9^Cm1Ak6BR*5B{4o=&ajpIy7XXCKWa%t;w zSTmOkx1c@iAl!v>f+@dh05Z1LOa=>7<8@;mcA2KvBn{?4$@x{W&$ULbC1!SFSardw zFEEjnoNiqFyk`cbNkXwX0eBpCsFkcrsM^%hh>V3@%00Roxfmft!wt{Mq+klqhC#p_ zoaExHq^jRD5=KDzMRCox?-Qmj`#l+e5wposs|tZq=_X z#dmT3HrI`YfpF^~M`YLv@_DbGZ)GUQlFQegwdj8kwWheeTSYq}jdIxQ^{+d-A&Kq= zex!TX)L?1*I(L0fjm+zNYSvpL4_03?A~CxorD9p$U>KN?I%Cqduf}uL#@uJ7E0@$2 zqnNs$dz$o?r{{;H2l; zpK%zJ0)PYUR2(`#JReGYGwdfPudN#n!x>sOBR^Vj!>=3>RvIxl$RnjwWihr7QA=`% zGHv3RYE>cAo}!{4gU=M~2udj3nkkJ9UZSObX@FpgbL6+RD+4UsuGU;*p0#0y0mom^ z)ktzjO1LscTc@QYNfWWbBy&)1=YdkrI6Rzs6H$ORjz2nnV%%^eraDsY135oIQLY2^ z?M`yI&ssppWjP|4PSHyL0D77Cg)~ywQ(Xp@!|dn?9Xgulfz58|BpYx6`Mv4fBwcsh zPdpwfv!>#E4wYfRSLd8m4tHY*IL%AbLp>Q)qXf22b4?0E^ffCSAL~*+-$`$Q_MTApRpoD{gtOjfmQ@u-aX-f!J3Qbr=BSjGF7dAjjqSnp#cE@6fT-Ju2VI}nO#}czUovJ-gpr~NAZNLygJvgtW ze`gPbuXq0d6)p5Y#T)FkiI+KV@7AgpBCSqDU-(t>bRe>%C^lDgdTT%Ap1ct^w;?YwE?87`jAId+|bNd{Oa z)~cie(%;E~B#JVC5}*v@-mJ%G6|`>3$1E`H+~?Y`wVVF{GD8i#t@8|$M*_HG7WHQ8 zIvCBm;$yjFXKblGc+GVl8}PdEadI}G;R7CjDv!hZIc6^TB~CDS{JlkA@fNopk7#zV zX$tTcsHZxSp&NrTz9jgDEjHFGSRAB;?O?Uzvpe0p?gX*0Q*@i#nj~smnnzjN{U^AiCT(_3xb4Jd5SI9)_oaPTA(WdL0JSO0AL9_X46U zPBYi8MM37C#BKwwDtNe1Nd>-@M6CuDGH^j2hJ@UD^Vt0=a=dlq9@LEvBrTE;wJ#uV zsSUIy3&5k0d0OeX}%Pc0$M&YI^w$+nSbFQA%HkxgXn4MOEaaN zqf9qUR}JC^FMLLN3hQpCA30-@o-^9GokBw`beXxz{Mz@?<6rUQb3 z)Z^GxqT%tfJfJ?64(=HBKJ=?A0mU1N!^9FZWP|?kr;w-!b|=|tCYjr-5v|HlV=o?@f+Jk(Eb0)N!GafFFsgNLCcuy?eDVvM)sIWQ3~*pB%UMew9t8GBIXs zvE-6^)Md%{WxzZNWKccK!B>`J(x+q#XBq~>cNN)uInk}9 z)RJ4ks}g*~vB1SNglKpN;J&Zo`NKnJBSb#)?KmFhy$1hbGp7alYr+9v|`7 zg}gekv?!CxTP=)W3cIWR)E1r{vu3h|Qd5#*Uz)&-s%!gz_T%L zT=uNma@~T8Nc5kH{{S5I?}(RDeY;SeXm=6ivtdB&Fe}gPL(eGV6`qN_gRw2~f(?KCow+GjW;uN`T14O&p8QJ5!8=ZsV~1ndZ>b~SKyNquJpowMMmDli5p zx+eG7=Oak%5IJ0`jFX;8_O2^l)g{$$yts3-C*`ew66*eA`Dua9Fs7^ce!|u%8s_QO z4ai=+)bm_Q4Qh424}2cFfQ_dK5Ko=F%Jyv@*GkeYQF5PX$=#pQyc%yFMW>cn+#+<( zx20v>d`h^LfHw!q2RY6SM>j(GF10?f8pWA>j3PGxZt2HgrDbZKB7)qr#W^JiAO_%Z z*1lA@_^WdpW;=%^NB65TZ;CpV!niXw)1BDpX)C=Llq_!iY4JSXF!5X(dd_9EvHjyH zCuxztz*gVE_#?LQ9gMqqN#n>PJZF<$W1y3K?>msvJC`}HPViCkE>=kRWMv}<86vTi zxnzv0c0XnQ68uGq=fk!U$mNR02K@CHuSK86_He|m#~|y;*n{cD$ z4bWG!MdE2Llt<>0n|NWyYtoBSS7_q9j+Z)b+9$;k_^06Kk96Ben{1jjzzj!M3!cA) zd>h~o9Y+2Q(ym@qQT901fH`77$2In3(4ALL)U2kHd{&Ry93+Ya-_^NpeZgCp~K77g}x5xwVy`nK&mW<{pN% zPm1NeR%?ecI~*MJs&SV!*2h*C7fvy8p65lZ__Fs_Y#Cgr9Byvjm6aZYrr&5@zy1mf;SU)2H$m{;r?21HJ8RdH zTTN$T4#*J`Dx3SC=U*0CAVO6})X2h))%Oql6z}5}pWtz-;@(*2wF)5>9e`E2Y!5 zagH)DI+MXQ!dqMzfdF6;)OM>{X0e4_mMjlZ*jJ@3j!0>t>sDH-bs;RO19CCJHPTt@O|Xr}BZHd7T-`RA=&))EH*Lr)GQ=DLYVuyJ zak29oEB8^GBDhUARSg+d`G4B%YhF9S^8DGz6`Ng6>CEb2zIfb{-3NZTHE3L`pj?no zSjTGTBE4>`M&sAor(4h743o5j$m(%il_SixJLyl9=sfmsnTOruVxHdBGP*F|xq}hk zll81vFAx%{aks7;zok!Yd=HovmpSN9f3117cK44)l$zBWu*N)+6}t2Ft1A;O58pfj zgzzgm3&9WqPMGA5d94Vp2kyeQ% zjos@Sdy+G_;Qsl>{?JuXsx3{6iA;eKLIbL(8@r2`al%zj=;uBPQ!zDLa6zs9+( zO_i0GZ#n0lwcQVYV>Il}OHd`ZV3p4xb*#pXkTUuV=Nx9X?@%eaoz^Y=#{ho`+DVu%Pr7vGR--Jh<0bLb zR!P0a>57{hutJ0#Fb`Vf zQg-)F^`b;=8zgyB6}x|pOACY1xC5pzGfZRzw?x~JK|IrzB9W<9IP4pmiAl9?$3$e- zgW2D1?hkR!e=2-e1{C84De42;Jt z*~kO2sU>p^Oy)oWsbiDS)j==^f@mYQ=)RxlA$W{Y*KK0c~mpf5LRjVp+a%TnN zJLSB;x}DVg!>R9FeD@+YVi^8)tE)L7lzhMwh92z1x8Oav14h;j0r#khCyv9t zNXh~0k8D&7ahx1>=972ORvb<`5ubWO!vqo2r8Nl21ZIX|{{Ysa$YgELIHVM`l=>Ec z*4~OYTaY<9tV6)9jT7gIlirJpTN-kWtMZY}Rh2Q%w`#R=1ZWs%t}%+L?X-Fm+KZVo zzaw`&`4zR`>ksWO5$Vt{1sdT2By<_gW3rFE$u-*iD6?CS6xxDV1O`$Ee02S2xTLf` zhq}2%X-M0*JOk8Xxs7$c(J)3}NcpnIrD;j`814DJ+o?6qBvWe|nxJgPI6HaGaz;0H zJpN{#%bp(a!|RvGQ!g29pkSQ){VVCOhhG7e{5gNAUw-bw(%M^@WXB4srHId_1ycQ= zz7xx?Us}c?ml81CF^#KVN&ePPYw(lyMe*I$qZp-=Pg@l|8+@l=$KXwRv8JI;<_q_g zQa=`TOW~^Om)Bv73GO73Nk76^ zE^K%o!k0cN)~z(lXqm1q)uy>%sO-oN z;t2KYUq5RQwAYU!0kriay<;`Z?#Npl=BIq@-db)apCt#vG>DPsF(Fj)LG4&g#**e`b1jk2UfwQ^5or{8-BQ?NhxLzD-w6(fIU97$7)u|4)X7eVB~kNBBOO3-otMD7?Z|W za&g-SzrAz%uByP4hdyHDVR{O?s_G(nogmIS7U^94=ddhE_3K>G?Dj4#Eg73Tkt-9H zVmf+Ob;i&MIAi*cdegc&QUVni<-Kc~*R=?&p=CpdJxzAOJDWv0Xi~q`7Tr!5H@#q7 z>Wujj=N_i1++4lOlE;q1q9+yT#tCkA#vG!OSdw`(|M{IL>`4;>K75)cT&(miEDzagsS16`;{=3d%+Z!4&VI z4}C(yAtF3y4Y(1N?OII{hd^lGc2$Q?yj4vZlwNE&T#o%STRPsuq&$iS=5BC$Qbe08vF0SOS%KGzG}! zldrWjCQPygTmjE|YPZTqV^QatDIl>T^~Zl))TqH_9XP7{jVN+6%>ruXo8A)R+lsWl zAh>}%xZJKqV1^^TF7kSwDO%xTY40Ssc4uwJJxx9-L|}T=Q;J>Wccu*%0HO!Sx*djscokxJEU?w z0ih9QceXR;QMN@~9`)#c8q{=c0et;R_8DX!FhxCU#Erxe#97)=#|yt4)>X~g%6c;} z=u2jyr_gp?Pop)zjk;O}hSor+2Lruv+Q-Fz?Eu?lI3u9qynSR1_m@7kAu)hR0251> zKuI03@kP0t%z1J#p1or!-^!V^b;n%f#M~5Zy&UrgVUKL#OU1Uf zaylscM7hQghjV{{iIUaeX{{UGOXbU&CNZ?m}466-^(oW3taT%5)4QB;(%0IKe#2CC6;)cK1 z?PjqH2ElbY6haBkK{f3E02RM(O((-X9l5vg)yx{_h-Zxbjyr#v8*}~n8{4q;ubjMV z;@wxqwo<`&JXY||n?xed5|;(w)RMRbbe?uIxdZ(8 zy>}TTD~>QdY6`x=KDyBn8b=0a#)^DE6}9}a~hQ`Vk`Gu z?5nrs1E&N30Iyl>zGRFul6n!=p}4(SSdovHjiBU;ulCXkbMh0FE9+hAozk&CShl8` zI7s@EpniW!zil8`Kw*>}7t*m(QnC<8!*icXy%gcFNp5)NslwY6EK-xy^tmBrW+w-q zuQg8U*h?P^#^0M9Vzfa%V{F|RIZ@AQdPlvOZOV`V&rH`FZE8YkocH!*VDAN+k4JHu;xs)yX}o<*k-U&&tCb5yMd03u$C#$EJloyS+y_%~)+W$vdzx zPdTlmgbk-4ob!{?p4ROeI}}pS!3uH(GG!YawZ)dy#?8E(isv;eCIR-0FFX(NuHyIi zY86xhM?!O-dgiqhF03~4dE=q0QfpI^X=|b9cky{-0tY z=S{geZgoM$EeBd*4$uiC@Ns}C7B4RgjFFrGI{j$#6!5LJk8E>NCfM==Hc1CKti762 zI^!j4G$^)b!5}kZAc0ZJxj~jJgmMQMtATc_s&(M5Cnsvr(Ef+jq$K#ZbmATZi%o#27&H3ka?n%oYP6gJCrCKQArwIUD>4`bewGi0p9AJU?52+ljwyI}BhNLxK|_)rL9_olZU`K3KLrkVyN6j4mD zJu73w>=mTQ{A(Qat#1#o61GJ&OW54Fk8)%j_No%bcel4aD_!9U7{)p7Duk#>$tN_o zGAZmlV*@>fdRM`~w$o}CZZZ`FN>5yq+Pv~O%MsXm*P?h9L2IPAP#4UTu6ZC-%S1&_ zg!F&6JnmVS^{2H~Sm> zD2MwR%Kj~)GjBYsR74ZlB6{F%R z-gp~F@cy3ew%1ykESq2Ce=HNyzeD~d_`=)5o*cSGogH3! zjeraCf;v~oU+_&26() zk?@yCjTEGstn0Q5afXnbb#H1_B^ay7i)*7eT~fnO_R1t`aanXE^><;(E@v zb*kzzUfgXnIXF}N1Jb3}{6&A_Jx)u@>DpNspOwc5I{{c%R|^u!^JH{9^Ij!dj#q4@ zwb;pm#tXZmjAt3*ywk-R$Der18*NNrAUPSWpAu`pM9@O}n-~KeW7q3kYTGpOvyge` zBDhrDO^Z(Zl;$EvIDBO1r!CD&+#U&94f&W0C_m*E6sH)=f0hZVpJY8R`tzc@1|u|I{~bf>0YHMa;vk_ggH=v zW}F5n0~79ME*iZewRq=0xl{VPvSl$HbDvUA$4S)brbkjM5X^wMMDEsR`n{AB5UXhP&gfGf&iF zX{~2+uBvgK4JkV-o27jb#~+ui74#p8e`h;esU(+0yK7Qhg@WUh1M6O4to$SKE{<5f zp7TZzF9*yq>sZ3HC$lt_DZ_Meba$&)8g0$16_yu`hfvB#^Qb(h#&(~|w@k|8DQR{Q z)MA^`fViadNktS4QfQ>3y&(!HJW^7gY3Kn(B`$cR7Zbe)Gzy*H@GYC4al z9e5pR1UPp0tB?T_XN&{QRa6J-QcExkwNE|i+?FNG{&boC5!=$CCz=?Xnr#CojFktI zO_hfj^r*q~?Lj%CaImkJvRt8!+pSBKu#h=+j6* z!aLzt9)ud;{6YI$c*j|lTTMtbX%)9eHMu54J%~MxYfcrU({VJ7I*F+*534>0{?5AU zc!J{h#;{!YZ%NVB=hv_AlgozE1m)Bd!!u(8>t8DV)Sn(Me`p_!_r4^xhwPpmutn3f zODteRWgp6%dbi#`TASjZ?N{-0;~$3Yd^h6{9N+4iJP5vJl7)pQBcYV29$_EtSC*oJ zoTwaQ73w6Dx<_r=Tk2uzyxcPmgw-0+)l?*y#sz0pmljyCQrB^T>s{8PFt@P`&9{Is zD~r*g^KYXB;|Ddj;(M%f3mY(fFPMpb`Zy=b6V4{>?EgSlMzMNk1BRP0MtGzXk2L zh5rC*4ShUQrX|ON=D(6;x`hM4V|fnWpP=B3b6)5Bdwey~e`jxn+MkI02Wcgvc!mY= z{lsy$VTL(*GG{$V1OY}P6+BGR=Gf+;hlK^qqrUgt{JgLO?67?9z^OaEM{0%c3hGg& zKJxR5w-u_~C^`+BYYqzXj-xcIO20d*-lkHRn5PThxPw`j@5Z}6A@+4 zPH?B1^J`sQhI6}<(477iq`n{#$kLbdSd3)xiu0#N$E{AXX*(V7_K@0mWGR8P^O5UN zTa^P4;H|h7QG=eitQ8w-^*VjJ=yg{bvD(hg z6k|Jk{V`ehdWkBeu^{qD;MX%Yu2q5|9P!hwQlD0kkCr)1bjN(vYrRN7MlKc|?j%KIWD3W%X4~^Q&T-POi5;>!ocHvoloCaIXM3~Q65z8b8QKr0QC5+Ia^PV| zInD)hF=`B`3Eav~(8T=0v~Beb&+_D;<1Bigaa=Q^&q|$W*yv!C`J1+*Gyebx;{)2X zB4HUPIXL8FIQKQr+upf9d;q7Z?N=bVK*6wA8Sjjm%~96RV`W>bH&Q_=mkO-Few~d% z?8LUif9MV+vb`z_nrENzj56l>q{w_BAS1EYg zBFSycM%}};Z(U4G>c?)y?0<@~uFx_#Q@jE=Tng!hI~;SUn{4IpB36`~pP6{if5Ne^ zEymIn2~(0!Z1)w@+{1>!#s?cuU&6Adw_w-<<|Ln+rFJMLaLx2*F+4+WJQ2=18n5O3 z(4c1-+&X5quWkI|aH_-}xa(BKubBS;GV-UK9AMWce#tGiB%VQ11snw5MSE2Z$Q%LXa?YAn0$tAm* z<2*vB@`(dvbs=l2j8svn3N#Z~1<;+EitGRG_)RQ0YUZVo;f49rU9R7RRf-XN(p3y5Jv7n=}axaJYs@QF-<3$014)h^ra)- zjAns>>q$iifk48wGzP`XbBqk;vVqNN_80L{LW33Tl6zhU|)oJ4j`Qrn% zTxnHe+l|DY2c=M)lh=|>GbPxmIT7-|rFsvCqBgo@3__q~+@l=i*PX6syM{g+a{xg4 z*QZ3kyF?)EMG6#ltSUB^h7~2LyW!BXYA_e%?#?lumHN~BCD`A$z%yl10?rCUK zNgrf>!Z%KQQLNjl-E=1h&jS$SM^HS(wY6U0QH5j7~Kl}Zg3 z;Ox&&m>T!*iT?l`#+&wyE5n;)xKEN>r5`ML9glNbx{4_Csx3=@5dJ3VUkt4p+QJz1 z9eL9W7I3it0JQb;=DDoiYr2%WhM{oR*RqyposL51_*c+X?NUi5xRzfnr+z$Cw&Uy*Ubzid?3jzhdkALxzaT| zNJR=p-sYC}{6sPrgPbmZTF1MUnlzSR2s{IvVyS7Ns#hs$x97@WPSnP5PbR#_#MbCMJY{O>)T8bMwwDZMo8^eBY~nvVB3eF9<|n7 zXvEGwRRbNmesxjw_~Hcu661v(MSSe9D<4TtNa3|F4{5rR?3%*b<27;=PmQ=Ix$!nD=WdjzOcmaQH$sr(7i zE`D2U%Zq34xCRx&>pueZOI^)m(cT~&fN;d;j0*ZD;@C#S5IxB6QT>Si>;)>=&hK2- z@WoV)t>|&f15LkoL*yM>!aAmnF^=xWXixh{aanC_=y1aaIIpyx%~gq!JluiAsvG&% z9-;96072FFi=846Pt4LrHZeK%73f157PUC#fuGrD$jupCoOGt*zKqsB0_gW`BUoKc zXxaOc4azwD>&^U4@QYE=BlAR%%Lisw=kl(&R)tT6oO7j8pTv(k#yikPaaN|Xk|$(k zJCxw!qG%md5~uO4Oj9DFoKS`H<_=}T#F_T*)-2fawzO!O$_R)j8bk^V(NJW)!G89PTo>sh*9qKz7d zAfH;*kZ<{iJC7&UksBD=m&896M{6F6z9iL`OGgg2e{6p&Fz7(fYW3@%j2=Dsm*Fi3 z!(KP>74EO${W4!SPqUEmyzm*w<*}YK(!5W^c4cHsl)x%^uh9Ph*vr6P0QlwbLs-yf zwu&DR+FZwU`je?Hw#&K4EjW1E{X6$381h7kH@ML8e=1wwCtkV+NmUNmW?#rOO^mF$9zE zS@w4S0J>tp0#8n~*4JlUCw`|q)*D-s^JmhSZ88ZN&(gYyEId$Kuta!%b zk~3D7nR4uBM5P8;_pJz=?$5tkYQPh!o<{?%NgNq&bA{=e(#f0>c9Cr^FfdLBw`!PQ z>~aqT4oy=$Bsuz&XKam1mU7XtYr2z*HvZ!6$B1>s}X8BkTy@LtN|(Ah0i>e6y%-9&5{5(BXx61ns+** zQKO}gQwrG(Ksg!BNSe6YjlkgJHP0p31$ke(delpFLXMuiS1hQ-^g5wgtqz>)3|J6J zUAR2be`<0t+#kxhq@0!`IVPn`l12*<0plXGR5^4;s_TFD)b3g9oV9M!n!LX$OZ z!!FUtepPU!9On#rd)B;H zLPb{HyKx|a)K?t@>x9b#(Brq#wzT~~DddMk$OE7?o1(9=vaI`^TFRgyC6$jj9+e!# zHze&K@srxJBD+~PD@e@WPKVNo@5xXUly7lIvNUy7iKZ=b+jnDV+A?@GR%rIOLWijI zB9yTl7svaEc*DqB=T#tMcTd8#jXh3xJMLtQZQ(vpGENS7G>c{#$r0rD=k=|~>_eO+ zd@e!br@dN^?Oi*Q>Q;<{83mpOqQn~#-A1RgLkTDFmx2JD@|zVG*jD@Hv$6Y}7y zwlF&rR^ro07%RUAoQ}UrUd%nBGgi!%1Z4(taoVKV6L1XOdY*Aw^IEgH3WbOrbC1)# zGFx4%A;4BoQ;L^OmZjLmWCgN6A9L?jqzjpLj(cwP9+jze5apL?Q`C;W^_Oy`k%kqA z0dQ%1F18msC|piX0ALvBe%{8YNiWJiTpz7kx^Pgcfq}~r$I`PUm@e{8=D+|Rf~ir{ zLS0g5qzJ?nQX8C+)7qpAVMmatZHt_KHBuXdBq2gL0E5!1c%JobbSsjik*z%@o$QS=CyesZbS9Y@~oEEjnZRwYa3A?eYo6DC$4Kg8>U7)RHb7* z^)=^U@)PUV-l?}F1KYJaIX5Rfd-~HR8QR3)&`pvW_~?3`)lz852iG+*U|eSz9VsPz z@)x}xg)Is!0%wjhOB{qaY+{_`g4n<$)F&Wy=9^=&vl^|`JvkkPW=j~0DnSReTGB4W zQiWr|HE&t4Xz?gtmAT@Ac6Ko~m0`|)wWDblnuH%ps8GlTMhK~GSjQ(lDFYjE6C#nG zJ5@z(r=@FN0VH_5@90F)J9RTFl2%yc3!EMgY*kX1y<-<=6HCRAUic%z zUM#oN!o`1YXC!!G!E0uWQFSYuxNQ8z z9J-vd1N6mwW1&J8*)n$LsPA1eQBU1BrF8|(xx1Xiyx^bKq_ZSku{aD*9-S!=10FWu zbPL{{5(tzr8_wr!eSV?mhHuqUYYtifv1IRS7EK7U_oe3Hi$d*{M|%KGEdp_cfV1IXWJgPbNl zcJ#$@>Xq3FCJwJ^(%(9xkg+K&2Oxem#@}jt%SQ>x2L-WOn&*h7yLPsN{T|o?7yblSz2is+~to`OrnbQwmSlfD5iiYqKX>=#kwg{ zNX<4#3iRfmX)?wfIqkxVmfWs>asc$rX7prgnMC}(c%wP1w@V91cDQePtnif`N8wF7 z?p4H%@ZgSXs?aXhNmLW_WMhitf!cVjEiU8Am~)RqRbpB_TH0~s<%q%SS(=^Kn2rpA zcJvs@HCES9GFY+51Eqe9{3`HA{30I-{4u3@ zTV*z@bu%dIB%d)S^A+&-!M}$y{8sp%;r{>(+YyVqm1)W9u&2!D@)i1fsp+V$*6@O` z#-ccebG1}peuBLGQ_{@qpzSNPd8Njayd39_$E{~cXhswO7@i0`*Li(pV|)M?wgKdt zz`L-(`?>Gan(?OEdLG;+w?`L$Z@HK@5S}pI>zvdr)R0Ed!=~R)O6>J3Sp#EW5%lM& zu5$V$0pnId+XJn0FMCy?=tie9If)@}J0k=(PB^3xef*XLFdJC)t5;EDXt;F z_|$GdRSXGnkC;{SChk*LS0OO;Qo)EnFBA>Qy;*QOlgB>QYBmL<8*oP%=}OTT&iMlc z&H+4TsY=_4uEt#PGcO2pgOk?0{{Z4DCBB&Ddf=RU*Qh1COBld7>OmL=xbKL%QruWD zC{Y+>W2e1$(6=&Z;ly8-S3Yagbh&SDTg*~&-*|Vf^3%dLLeZMyC3V}7aniT^JKXKqDo*;?fEF1^iOm1BxLDWp3IXAR;$IOI#)J8MV3g5iCSpgUXRS4rv{-pL4pODmy;-`U16p{E=s@g_+jqsaxLgy52BcY_Cnm|2IbN>MAr>Vin;-eWHcc=1k)97gklkc!O z`qato$p)ZM2VB(A?j!-h9rH@$HL0g*@{y6(sI6EqBFhZ=fmwE}9IqMeT8zRwa!9}f zjBrgc9LBRI`?5OveF10S#w;vO!2Y7 zHQHU=CA4wILB=`}#be!RDR2yt923qdEl65!RI}BZFn64a+qAq{QAy-`*D*bmw(LU# zO@ih~7;4cKgSJ?e*mb*VffRICv*X zmp-AgY9$pYv?p`*t$rwcKJj11Ek945!sAbaMY}qEq&HKKEismsLcNJN0=S=q+AaM4 zD%Z5jUlrZ>cShH5X11BE)CQ8=xxfVUBp$WrUMKMGnBF_md|3viX{krA!upSs1A^AO zRAjL2laX5AvxkXoJbm#UEv~OTn^tEw;dn0NILYiYT=f%Hmud1tHK>JwPnJ*B9)kzs;KH@cuW0Jp7lDRV~W zRvouKR#Ng6)Zutuerl<8Y6&4np}?)vYzbWagaC8anJfZ84#G(8D%0q6&$-LBvnkFo z&kK%~EKuV+j^6dBJ*zB-joBQL(xVp2jCUy@4x+b5n<;B_WengK+mXjTF;0;LL>!Ed zTD~9+k=G-wSCZKGBL^Q$(q?N#c4W^MbKDc0)k3>QGJmCOO&MTL7XzuPQpQ5yV}aE7 zq-_~AQDRd$2R*5!vJNqk)~tDMiZXG8dpGYMt2V9ySDn2otolyS%tDR5 z0|I$HswA(mqQ6nsW>Vzs7$cub%X!<)Jd9(~wdYnS<+3tygW9n!QDZ!v+4QejQdaXh z)mL&vyYk8mH$3j^&1u`;?Tlca4sd-d060K;t8f7|IF*P1bsXeVcy#&}tF@`q{{U(h zISjji1Ofeh>UGqM?;-bKec$U`s!qTXcrVcPsXXixlZOOg4@x72x^^4HX>+AbU+xT& zJMBE5YH7GE3EYDxqZ~Ka-mxRN^Kk4oG2ny!D)p!e-#!RD1sy=E`n8Dm_cR28cfx=h zpeMaa9BvsmIAC}v2WpBqRBXXG2OytX(TWk~gko{joK<~Z#$MY}&72I*oP&-)&JKO+ zW-Ck(q^WMa9D$mS!q4O^LX{m3ImL7qmf|Jaz?5tpw^~Y#_6J(aQ!#YWw35Fvw+C-u zV^$MSV81XwcR5~w`c}V^xY~Jb!)ZN*SQcEdP&aeQrlDjnX>$qfo>>8GDbH5)s<)Py zIbL@Ty+&((TTQCXj1b+fZOJ^jG^wljcFKwyIUPFZb>JUNiShFVN`EoKg zl1J-ZPN1JCzyldL70z6#Op-}+*X8M2MxSBIrH+2~JhAz9sNj-%`c_QnR|K4n0Rp$L zV#}4=xiQWcPf(pGuOO zwxu*X8$HE!+BNJtt(2Ctf2`YuK9$SM11Q`%6t;IyV>pp=4I%Day0m*MhmFA)Jx_Xt z(_sX12(9_mHOnzN@iL4P(y+)eGbk4iAPQPaDBJ>&jMC9XfGYNm3A|h! zRT{gb=j}rSBzB-rW22CP5a;h_pGwNSY-b&PI@XlTLJ`mk%GlR8aYm|1W&?^(xtZMf zKG;X5+oXs6V;5pOS922NmLce6J6_9VUosK z=W-e-<6Lw7=JuPM6W)tSZfDqXs9AsjepTZ(N&=`~nMN=VTIsf@Tx9uUb51S0 za?D8Uno!`NTyylRI%b(JrF8|Y3i8b>g;>>dl6wlZ;$IAEJ`>b_&wmG#CQcV2v-pnm z)3ExTm%^Wt>gklI{oXRekF9zzneOCefO4k=++g;tI>%AQF*t3y=O z$PEHE2SbYUUm0twZeG$BCT8K=upX5(sk)BF^x)?7F?>y}oB8~da?qzATEQJDNX-CZ zy{JV+Sst2Ps=@(^S`Pw%b62rJMHC8X0*WZ61C@6$9<&rVCm931HkDYeK7H zN-r;qzfLnv=x1ZxKVtoN`yck2@KbKuF65TY{*HtX*1t?I>;xWcL3vgxL-p%llAp8Y zpmhHLj-CR!EJ=@9x8!@cAJ)HM^{7C*yWFH7F^?zSy!!CFHH@r|8&b25P}^0bBZHnu zu6FGeUO+}d@r}KB>s^kexJE%@3GeCbE1J58C|M3h-gkAcHgL9$9k8oy&KpsMm6>+n zH}N-0<}acMv7ameTJ zHDVa|1wBaTrAHmVn9R}u9G)>(&=C*HxyJ>4Dy*ACH*G`gv z7YZ_IKG35%0xpmKv0H;4>aaI0mlE z;wPO7A>0m58`6{4R2FW>G4USdB%UNj-Hzj-u4cwQGq*YISA0!7++U6S06J85Y9j}s zIO&=T*=k99If(+PI3tl*(?0Xlk6N{PIdV98(bRXgrD!N?eRki(*kl z8OPVP0~|R#@lV_jBPSlT&;cFuP9LQN4^Rq&&U@7G=XUHIXQe@q7dgr4#aXo{jAS2r zEGBB(0@%+^xB{isE`nMS(Bp%iY4%a_3E)62OiYcg!M3+rEB|Rt7N_{awq_!j0F0*h~cHpBOdQsR-}!Q z(aRiHF%m#OTFkuAH-b=L_8F-yu8V&T-M+o+YSM8i`=Fk~(w)fjM-F0^)2y?+eibYH{BJ@bFQ_$OQE!!j<_DQ`-?oNbinP3 z!En|HuNWx#w#yd zw`syi;PJ@T=&?9)#+*&~`tc?lzpQ!<>GxmJcBenQ}Y}ZdDR%YSj zFvE~rEw>zUFPS79e_KHPwFCI@gN4L*xAyQ}HIVb*c$077HOJ6(WQHc7-1O z>z(jKMfA4~f;Wym!0TBkIIEjcij0jz5G)HEWOAdY(xQ7%7^3{Z^}wx9FKOg)*QY~J zuraeXbByC154BWP*|gn}i13oeGNXm+I28c50B#?5xfQ24+A+KD4oDxRM%KX`k;uk9 ztEZ)z<;mVxMp%hY0Asl)wO5|me9ycy?OKvA497c(03T|vMZOe(!8l5?f%e z13krb^XY9LBMiKN4QI<@q%H?j&S`0}tX{jAhBg@r4=1OkI9r^r9Czox(zIU8=Wgr_ z@wcE9!E8&FINh9&dc~u%oV2k`ZO$@AIp|L{Ab_aA!98(T5dxeA1bWh<6(A6H`jJ_* zbws2fEs~&)I(yZqpe__H2X1@Q;{mc2PbaArr)z2$5}cl#)V{-HmNuJX$?55e>GaJn zcRLi1)~jijIbEy>+%cSF`kpN2J812S0iibXie2C@0;UC-zY;DcTvh&SQzR=KQ zD9?YTdK^}fxIFFv;A7UR+-R``We9PeeJD=*gNft!5cz61A;vli#J+%qU{?fmuA=(b zyA=zN2{;~=&t1j=!j4Ed$?acV?DRY+&dkI)G61X_cNjgfP6)^XkU}0#4=0*g0cK^v zAaywPr^bplq9MYJeATRJ*O=$|2-MF^noL04?0PV==2;-0|8W8M9IUVvVUf8sQAO^wYjtw_O zuZW*%eGO|>#Gf%3BY}?9t!|*mjl7eMrzW#4NK&L7{CvTewDdX+ z9uldr;NX#;QT48Z)&+D>Lv3HW8nE=jsoAj0%8s?sTDB2OWPyRn=kcy-D?6=P0MalRAi?>7m#X_=*4vPIeYk#;a~v{*-#H3{;KEi zCGtX?a<~OR&o$IvL<=w{4aoqGb66KKc_m38t~vwx*71n%)aS1X%+AE+w{kiYSQjd# ziNhf|&U;r$bh|hO%X%M5!?=XZH*5eC_ksN@u9Dp{FR7kVFvZS$53jv*o-c(=C<2hg zpO~L+E30J7o(h67gIFFfw=wCfDFK!#75@MX@^f7H+I-X{ZidunYKb$zH3+<$rzG+i z=A8){$OPx5YIx61YkfvZ6;GDOa{VhF)a0BK^skLxXScaKw;qGJ^s18g9DPBlrf{I= zBi^dJZztdNqh<{ujCTQm0QRcV4cN~F)bG#?;Coag;jw~g+d-lR`@~Qkm%a^HHrx_= zR1Fyb`+Cy3cTBSd*=7pduhOhdY_j~r)6%mC%UhZgM{=eo#{6(i43@hUH2c@ExW;mX z^Y2<0YaPn@u;VBAYKEg_A#*C$!bg}Qrw6&>yAKL{I=t~VrE6hh47!Eo z$NH;#W!%NPXVSi%(m!OAYpPj2!^Vi9eXZ`7<$!$$Qhh6=`!5kQKqM@vhBb~`;dFO#YWJ_NN-NQGFZdG+RQ7O~lJLHLp;4$f%_dO@W z(0FT3c&EQ!zi}oR8pKdt=vFTi&iRIL9+hmMlg|ZAW zDR0L>3t3?yZ?_HmsU$*7n+I~M-+V5veT`m%&obtqzUSV$$2XX!{TJ1k+$VvU9 zwWuU48Vh((?>RZ!pX*$NDZPL$2Gd-xbJM4Y&N6*-SpE<3Ww(Z|`vFC8(f<1nfYvMnTY`j-3 zlv-uGL1hdmuErpqV-A=dKop*`u?qW_{u;@3<<#3&@cx!V{D z9F;ly)%bbg`-5w(>C=OThUzB=9*zxu-)k2FNn)1+62TOoQe#tG^<#T7Tehbys6JGJ zG8FTJ(z)GGEO|_N|glLi!QfxaSM#YZ*4qTKQPdo{X#b z8s;yfJDB7yN6ZQ28tSz-1h&vNo;M2SH4KCsPnbv{f%F~gj!{r|JqmJY>Pe3lw)aVIZDdp_Ua;rNY(MQ@JY!B&{MDG z`#rPoZz|h4rEHeT#wtS;k`aJ^D!*r;O*v2ywtM25LQKexHg1YdU$xTYl&{KpJuzL5 zhu|6J8{H1x_!ZE}q1xz{!RJtjI0W^pvWwJ?ZFD%h9}T|U(t;PiAJVV1R@!_$2I6=< z@lfjiD~2~IY^MW^;djkq5ZVLpFnLG-cs@W419>w#2iH z)j!g&gJ9qf&YLR~LC|{o;;i0bZ(DCfJ9AmLcL2L`eL7V2(6XGbxsN1(YARm2q@#00 z9cTi8F-khq1_Wa?&@eMk>PV)JDT?Ei@&KnvmB{NzDZr31*Xu|JA>0pR>sKuhHaW?v zaf6n~&1hIC-lNynnJn9lSz`_x01Rgtu47cXH&;yA$Q)+6JNtoaVowTjz|C?yixN*< z)T@mOgL?own(QrN7TRhIeg12ev5(B1N4<2{cLVn@j4sd)XzCJu3Kv$q?mQ7xO6?1b z4sp$I21%pH1+Y(1S@$pl{5a1_d!s$IBDav|2d`Suh02h69-{`cRT?~Gez~jA@^hLH zT-Snr04F#el>3WBxI#b-aniHmVls2|=C4C8(mHWeMVW(lro0Y%u6d{0%jL3;K^4_p zTBYnfD9>UmpS-a!z856-z^Soovp1l-^DR&lyN-jkMS48g{S?-2n?IC7oM3dVONJ6c z#1K1yO+MsFHhl;C7G2ME;y)31YAH4iV@aAdE5R(hfm8ngV*#ptx-hFGM`W4K@+KI3hko_XyVL&+GU{B$Azxpk>rxz$Rmbq;fo*2v3?&R8Xopt>%WnXhOK--jB1u! zw=b3xDvB~e^ri3}x0B&pa`{$fW+S-gp{%5=j;JThY)B5x1%f^a(asf0xd!xGu-I3lEl74X1r9dJD;OdXbne3tvj3ZNgoxc;?T z;>iv&gFP!Px#5=m~E z9Ar~ON~+oAy5t;IO&#!BmEnV9U=xAehu*D?IusZIc9~9Q*42b6Bn^{* z27mh1o7~Z)wntN;TYy6aJBI`??d@KJr0Ma=9A~b3j{! zvnq|MMh1OBsB$xsT}=pWnFa{oi~zW;1&1uVkIc+>5zykF2rz=ookjpBky5HSA1>4_ z(6aRPtW#~=$?3R}!mY87%B18kLQP2a?JB+(1v(DocB=|LMr8^HcHy12|F0stKkT-GJi1KU!ccb z51Cd=Zy95T60IC3v($Hs}&jJu6tIMt8M|f z9FdA+TA0@?u2>xOsn#PC4Z!@z98}JtXcE^$O4>kGK?H(+nXRii{J@~+YWjBetQd!u zNFePEmC5PFXj)tZ1$N-$Ao`l-=SG%&PQOdI0!0`B`;Thgx4cv?qn0cKAnja5z3$`; z+_yvT_|}vft1l$=<2m)LB#T~WVFlti04U({K=1EUo1-%2akw0FT%3A|+wzsl_dV*$ z>Ws%Nf)5#9#-Y11(z(z+pfShZUZn6om7jHXv9l6(x%=BqX346vI~F@a`Y-dWtK0qT zGFxw64O~{YY=)B1n(lceV&f`=5L*?`PjDo8BXE7V@9k1vTsds3I)FzFRb;u!NCM%B z%N6cDt9Ue)qLs8L-ENSa1`-9p2C}YEi3;2hdyYT-b*V1WTW;jXBb?+_3DawZ#~ctr z%GIi=YEy${Bu|v?Dn<`Zrn3A~7@tACKa)&*urqg~REyP|%t!|fUzzxK50I!K%k8al3=3X_G4goAV=CfRU2 zWG!ib_EuxeGIRs$S=zsWF5*)qx~-D$x0n%)|6Yz*j ze*uou?Gd%b;-*;0{OmtLUzk4!d@~=6ek5pGRKR&t-XWB?-3S;K&$cV~cf&si^q+=4 z9@O;PDOPM6&h(uw{{Xs{LU$_3j!kmEv`4~E5?$!SS@4U=afacc9)Y+1$J$RgQGn7DeHPI^>Gzrxyg4=Nz0|wmoA|@uU~G zJ3P!GVTFirM|!Dyt3(x&7cm3d8O{$wTvnTHI!?;5^PCN%kEpH5&{)Q@O8Xl-h>+s| zQ6|xGwuk12?E!w;r|li%Y2=g5D{vyn?S%ssgJ_}f4?r{c()>Zb8mGn|8)^!{)=f%K z8G2wG3{qR30}c*3#dD=|)RxH)m}G_E^VhXT5AhMtKssaIqz(_u>Uk!S+l-zDdeONr z-C7;r!_gdRbCNd|9J6B;wX14{B4tB?{7es8=6oJp#WKVhn1O}Ha&ywJYbq@yf3*;s zMs}L=F-`NFRTsT;CjRH_+wSgP%wW7A^zEAR4;^@c^&6ikIU_k>PI2vBZ;N&IOKC1- zAR_#TqxioH@|mSG%JMhfW0O~hrFYFEvV;>{%w-f(6N>as&;t~-Q9=ODXrK&HOah81 zpasn(6a$(!5V#$v+qlwIi%WL7eXG~4HLC{fM5sx}deqQ7E2?;(RJYS~n=6e*{&vK6 z_R*JiKEP3JVBc#mSN>HJQ1V2og(Z0Mq4WWoRIOi45-&_y@X5Kl;Bi69xzNIG2 zo9Tkc#n1vc2R$p8)TRK)5V~&1&DOTHIaxStD97GB^{#ID_DJBI^ec{l*Cj=%+fG(7 zE?hG)1Qy?%5ztjq;1k0UkXw!vdsU0~Kf*V2oQ^V0Qz{}%zDQA$19U&uvGqG*?zAU_ z9__ar24j!N`Y`&|Y)_q|j5j-g!Tf66G8OB+0UQOv2b#4Np8r}- zGnS0`R{6ZyODh)l^{TPypJ{M=lTqAU6beVuuIU=QFtGCmMtU0dk>tCxb3yPxymdQ% zP?7<6-fVS|AZ^*g9jaX6n8>nZxaTERV zRh?ZHS&F>x&<2eTG3`~MH@?KuHKm&7S>st0=V&!YRPjU*TZpXkWnuFYI0xFXb*%`j z0(dtS1CUNgtXq75K_pa^yO!QuM#LPG$v*X&D;9E1Se{I9K^dxj%>|?elYvV`I~E3t zD4+$!1EHXr2NW`FIR<$qo}g}~0#~+0Dg&J3npP7gjaauIm#u5rE(aqZdscO_Vn9bV zw{8WzZ2f@iLUtj$nY#VJbx4BazH1t+UtBfC%dt5FfmP|6tr4Z5#9r}EGDkdB{Xw_g zGh=BTy($d|2;-Jae86$_spJ__#N?j7l+wAIdXBZ9%NRX6in0|ZQ;g#ko01fBx7^k1 zSpmY3c<)BU?m;9R~TKuY_)xbR-7 zb>e@7UJ2AF8 zqp39{($JjTXFIXaTC)tlI2a&w&14&Z8MC#N`iiv#n?^I&(zNtBS~jic45fN|`qkB# zv5mJdJ+Nyc8OO@JH#~~CW!|~K2N=SdT}azf@Y)$zmuBawz|Xx%ueHDjX&{V_n5uD2 zj!KZKIS0K?vXyU`h9Kkv!KOCRX}IADWdLVBwLU`f6r8R(>MEK@8FB_n>R!0#G zpMH9Q=}yGj`w4UcsUY!;b6N6`k(?&(GnG9vR_2fh+z@feJHMr8$syRdDo;$F^^~r5 zQGW3R&A8woQ_xfjzr6XBw>T9frCGs0HaN)RH4C;|X~$fUJBsIREe}>Bk&Bi9oP&&3 zmYBpo*B}AV6V|eUS8NV2GgfUb3`_~Z!Sv4+oR*pz-p8Y8a+N@2Wh@Bk+wu7JZn%A_sP(Ue$+;Yc^)^7Sc9NHHl#!ln)_M}xP^zv-k6hVWiO25psMCui21_iatT023FMyCf-^jYn4Bof5rdOiziD{~ z4hScwa%)#kjE$~Wp~>2J0Bi15T@MO0w7GiD&nKW^fx)edeK-KFLgWsV+c-nGBb}p` zZZlg}&|yI+3jY9m^{gjdwWB&6EsBFlNj8v#D)ac!CW;12@q@yipr1hmLz1H4mj^o!d0;0y;SqXOhy?`7*)ncN|{!%s;1clilpQYIO=oF za=txL{{RS{q{MF96CMUDt(|u&s`5xbFzH<9jiK{ADXFN*Re_m-?~h8(rEX|i=8&Y6 zu6*luyWkv?kO%~ORvd_8lOxu;OVQ-pBV^!?GxV-z;!W&*b6!4&V%!qxk%0_&t&Km# zcUJDDW^9g!HGl?ADtmazI5`8>f!Q6bJ}$m>3iBBg2j!4%$LBuS!5nPqs&E}l$ zjNd1y@Aajy(|0h4XwN(mPLf|RVTU8VX@64*?eaz}2|T^epxqKYXL zCfZOjMHFlSMHF|YfGDDh04So0U=&eBAPOj=ivXZzwJn{av?OB{mOAszYQUvJ1;#Kq zp`$q-yZbxawYIH$ss$c$#pJ^w>WofF{Dpqk_;bdqQxvb|#xVE>JPcBK}G9zd274)w;*0eZvuNB35wul)+258i8E$dw; z#tEjg)Vw`&0y88`@s>GLGwR7GZQ#>*=95c+#lkQD)TLxE*upD|1t#wCZ(S1ic7EIt_xaOE} zI4T%)`c#r~oPIc}<#D@kBp#Tm+(&=l4N}iV({*dRwO5sVv2F?Hc5-oC?}>GpHG9df z=6#FH70&=wT|I}8LB`SP{cDQx_lXRSY#?9<{Io0ltIeDo>e$Uw=a-Sr_>)>Xwc!YV zaGWUh_pE&8iU7rTMJTHyy(E>8QJhgh8K7ZB6jQhb6j9AE3MitB08yOMiZB2e>q!Vf zz^htx=H6}Nt}#_>LqpS~)b$u`p#i0pf*>aYB;uXzsHLj0dGDmTx>sg7$m`a=i}1`I z7}tC+;Q9P`y7b-~(2R%5)29dbOxv9lH$um_uN}GYL_Qz!Z}v8$t!eSk8z_0L)xWJMC}#3Fo$+_pS1!hwQw&JA-q zh+()PKKbcgb?{NUcF?K@?DhO>mDEG|jDTTONBJFBwNHYP()D^*Q!6UCH2g&lJ}C*z4=ssbaT6!^}HH9z#QPy=LOVQ0d4WYku2aw0Q|@cHGC3 z#-OvZvuOBVoOS1|NhP(vn~2o=mZxsPbGp*3EOm`NnL$IehEtF8n)6Q*X(vX#bsVWb zDX#wj!o{tuZEAH3Lvf7l=tVcFo~hjPnOw-^ns0CGLv{RVDIEbt6i`rO z6j4SvrXfJ*tr+K}B`S^FP%DqlK=!AqN1&%}whFJi9D)-Ofcd zlRXCNo(BYo$=IWh{=F-q$pGMXKD3dOE4*jQ0OOP1k|y%l@_h$1_x|pBcc{)=QG+ z5lH|l#OL&+hv+><2TW$BwV%p3-Lz*k#zDhmjL^{ojxt9)(Yh7vYl#H0{GXwzK~Z6|iLxK=n{Eygl^fu|PJxThP&$oF>E zuI0Sa-uX-@ig6=H%laRsb=FX&*MS!n&uz?-^UzYe=wu{{YHIkmNG-EzNsaiYj!~SbWjt zQmH7**~@;1&6<_j66ZUC80U=EMAsn*1oi3dU023kD0pMyMz^M+EfjNku$%#y+m$%3 z1+JGPIVaM*jgHl-v3agla=frT#Z3jk;Ee7)vsle?OMIkt;8c-b6lOb*Urv;hDW; zfES?0>r;KG7|SpLalt1C=}zq+QP8(>KJmHOa5^5fkuA{PSAm0^0Z`mte7%Eo=eOfo z@?H}rLkylXP}#K=XQ`lWatH%z=f8SB(!8<82><{&&1bB;m>lHw;+)NZMt=7o=cQox zJxJWpl0d3HeWTQWDpj1Taq{P|HAu>13xEzV2fg0elWV@aLMAx}qf2k7?5NTn(GDq0tDFs|uYTPtlzniq7<`{x;8tbdh&HKbV1)G>jYEylT_b|CxZKH*pm$vG zE2GmGDx3vToD^O&?kkP5c-2cYe3{1`PHU{v&y*17rx|Yl09yBFR@)gyn%L>I@09}J zE=NPwx9k{Tf;JwCxUO#1RoFH}nCjW>>04H^6jf*8LCEJP9jl$yZ&ODEbarmU{O#$6 z#wyFi4qb7z^SEdAs#a1bm6c;5UP$d%7BMyha7fP4&{s67jYO$+2?0(w6a?TLgIRZx zmgtNNV~=X+fE!d3^YS(kS@)2Uy%n>Q@~{5@Ua|J+F{)R39G0e`iFCsJ+j2RqONVR~ z93Na)O??WWVYm~LdSqs>?jl|pNiKLiaC7ff?cZ>8wbap#)p zG@F*&w{}?c1azm)9J?J9E)=PhfWD))YW=zL_OIVPM(;|?w|4Uovj!N+z^zCn*mo1p zraRVdc6*v7+|ae!2pw6vr?;(4aKxMf4^i9NtH2sUNx@8Kj@4yI2wVPH4^FxIRxY#Z zX&8-ds<{ib`kZG2+PVJ#6e7vty-zKYU}evtKHinmyoF&TDI9mMYsE!j@W!bqQ0^fo z(`d-%quRDxg^!(lRXW2QQ7AkPeznSz-n6y*(|35x4nLQ3U;?hQ$7HAsi<)=~#J?MYl|Eyy)H2ePhkbIG`9S?kd6rP%#^<9zf{{R;6KeYU3E%GpETm(W7L>K^m z7_XrJ0N|hABacb=uc+C{KEbHzj~NZ~1}btZh$4~lkgq|seLB@IANY@0_>tlZonKqk)xnVcn@@#y#R7H%)cTs^ zTFTxcPuiuGNI3zx8T77t)!h9GOO&%LAj;l#*ld6Z4NRS7T$7Lc{YOYiNhmSOp)}Gk zI;5pLr5j{41C&ry7=zK^=o~4{2&EZ~bc>AcMg;x*_x(NkKit!8ulv5PbDi@(U4Kk{ z+9CG#g=C95PkodrWo~{ zCae0d%;CKs5^GjmTnAR(spG^fp(hrWY{z0Ffe&v|9f=xM(qDgx+4Wq}I8NV;D-`Ev zKtx69i+#c4ikWAr;CX}aYrkH{5;c^~DE{&_o6E3y;54$7_*plhXq@`R7mnN4t$)li zb@=r7Fq;j+L(iMS)q|$Tm`+ymXklD2|HpZtS+7noIF=zROq=25g*&Yp-6}kdz$%mh3SXSToLt) zoXyZY%Q~PRbGNULRmcD^@5@D1chK*r*{=R8an=0J?zT5#_o`yS!3QFKCk!IM9f1Q2 zM-aFO?4Tq5QCvbyQEJyimBUxsbuJ4B9x-5Sn4A=taQd5iS$mY2_>CkAVJr5HmgVOm z&zrs1C2dueJ2JF6#~Swgp6iJV+Ak&Zn2%~1_|!}HZk`afoF=!Moo#--%V{GqAj z{@Bz^Z~LbYc18mC7P7dz`N|R@7eprWtntFeWZ3x=w}^l1jd;8m`Qa^pJ>CEv_`o*z$}z9 zCKPIH@}$%!`b2&)DZl4p1Ro~!dLZHNEQ`v=Re&u@-9{LC0qsodPQ5hQz|x!-eHmKD zRfh5>+31l)VVZS3`0Vj(dRcC9Z-JUZnGzMu^3c6!_nRMBd**0HU8YV;|7~>&4JmCy z)X1X==-VkZsq*gLFsL!2)(DAH$~qS_l!l9f<)B8Ic!^?1%f`90PnZ(`GRMstm5+G* zyEX&Il%H&Kroi4d`Oxvi#I&ROIcv04&bDH3UAFpT!ma?ejHTp_AaXM)uJX%rx;i=e zMt1RP7he$y@4f44nS8DqnK_E&40hZDySQy;k%icuOo83Q#*sw>434q<~n6fa36Y(2Xq)D?=p8xik>HJ~vMeoG$V6}10 zm%;0&6SgV9SRk%Cz=N{<@PA#p1#|+yI37Gqz7o=?4(X5tb4X<^jgLgl)mY~HV+MVA z>#wS17zXZX<4Bw@`N(y~g}^mF8%AQIYo*PBp|4DrmB3uzOYA9A3C;8?|9Ei<`>Z>; zpwuM4@%b(Nbh3h^AB>q|)BfP+@n58@)zqGV)*t1DR=|WK^RaU8O3b6}TY%Z7b?MlW zcrudBzr4?1 zPc*;}pqGsKGDHK4*GC3c7^Oi%_~vY6LA}P#QlfGI$tHQXu3q)48qQY%SYTkkB2v&l z>c9~9AALyq7Zn;GAkTA;ijKOhW_%mw2ek{VI_-|cmO*ozT@v3wW}wp)&BAYtXoL)* zf#M1FzFt>&><)~&Ay>e-a^QoSPF(pZO;PH1Y zc`DC$r^!oei|lt=z3lD}17E=v$S&Q@(h+Y|CUWRqpP_I4@9f;Diaq`@&g9a^hI%s% z?7F_#RtpihqQ5AzA-`N2_^$SSy_gm6({K6Xe{)1nBwG0V;Gc6Ue{}x`V7i}9_8v>t z1+3>bO&AJNcB@(6yzyIXWr4(&iT@lH(YSir@M>W|kLW`l%!#SHoyBx!u4P6dZT)D1 zsL}qL8HS56c*1fk-JxR>b#Um>W3wJRAS6BgkdNg(6GB9YNrKy|Cv&a@}EUFn96fiYn_liqe1qUxNNRh?rZ|QHN*G>Lm;&`rV#2(zQ7u3;*`Rx zCjae3zWPALluJtp5S34Wr#QkZ6nf zZ>=WsMi>U!ow$+qc_tfFN3zTxQ?4aWFW7yB@La_h4`#=0hoS6n^!5Z@R=V;=Rs!h% zU8<%*r+r|N`Tfd#BS^b4Hl2xHtYrOl0p)@VNA-wA` z#RYIvz_9bh(`K@Pk0Adrs7YDgh~codV@htC@~_9Qq;Rj9tUcSX@}kN84+7b0swk^Rh~otR4#yy`Seq-m+>C$Gx+NGGr_nx6ElLv_mnLcE}k*Vk)_|Lk8ZBR z9|SV-zfT*X_V1QY!z`m{!h6MM-F`fFHbPQOwb2N>9W>q=w;Gz)I>N6y-4x@u34(Oj zbwmD`vVI|P1>nW)3Zy-X+!}$=$W<~{>(JKuJ=zjfT)&>s3NbX#>7-nbDA!0G8i{oV zM%`CzvWdikj-&m5?pfEjNxo!oWWH}T^s5?_3p1lSlU`r^Eox!cZ8l6sJcFhGxl-Tp zFnU;z1Co7Sn{cCzyA|wB|6)f$_=_onT_fM+xN6FKUh?Xg-yU<1^>xG@;>amq>u#qa zBLh~XXV&OW6PK)xXvtreDOgUPz4_#Qxg{|-E)v~IUs@*GJ~=Wcu5rlWY%4s-a`3sW zBMc?^Sn5c(^SkGwr0_j&8_K1tXe|4g({q01~mocqHjbo zou7!ZeJHzHffrGF(Oj**#WYqWVl}FV3Eq$Oj_~-+5LI1p2KYPap*b<}78N-(dCp{@ z<+6s6Zp*^^tjzGST{#}a@{`1!r&7x(F;Yo?7>gT7T<%bLZT}@PJ1d1!Tk(n|cVR^r zohz=mj?2)d($c6W<4P?+Zk87$}xN~F)Q$kH>TWJ!DD4^0?xX1@(9n% zVCBC#mlI?mRK4j_9(7i>s_wvyxms!$k?#T%){&6c;D23t#2deB&kj@@I}(bg17^J~ zcOPJupYW0%rT4#mAFg{U{{`#0XWi~T@2i6@`5480ggCNN7Mxc{e9qY#4!^)A##++D z48*_fn6SM75D}wmnEyTE^9)%vb!3WIk0ajZ>pUBLf5k_k8z#{n48z;nJXJig7WkJI z;iBO#$@UBMXP$iJOS_tofwsy+{ppR^SHyBs9F-z`2I#MnLDkGT@;;Jv0xNh3{rD#i z>g-OeR0p;J)85(iG!tpw{ay5!aMM(k^@C=%0aH=YK`s1s>De!kbf6QIEK2EeP#iiSFwK7NPI9Mk4VctY!A68;_|QBgS)npQf82`u zQ@E}ANF=3&Sc-o0KR`WYWVqZZm2P@~afU^E_Ek6Vkl2IvH`x{G&txJ7C@X-?@?wW? zsNg)1-L8y;n3wcQ>ZamY59IM!F(?^KSL;i(^$@<2J5ZQz=)~6oi&zgSnfxK3)gpYd ze8HA6SH7YQWNrU*d}689z6q!?CQrfLgX7E>iW-8B%cROPgr^^R{1~@gmM|}uY3RYi zOvBSbj-98BZg*qbAy<75}J(;9_z>Xiq{l=ew0B!Il3)s6W{eVM^x^- zh?9N^wXNyloCIV#_y=LXtfgN=Fa|5_FS^u=u^qg6WI-k zm1sU0%+3`dUD1=DWwtku7QHGlA%o7>ByxC3t0W2knJTm+=$0yV-3GvBt0T!j3=u>b z`E-d1uHa!%sbLPa)fTTj&b0nsROQS)~5BsJ{Yyw5!BvPRAuIV z+P~}!Vr%%iM0ZI5oc$3bsC=liu!Uy zzbY0(kFj=)j_@4)0KG}%^j7J;5tWeSA!eQ3e3~L{cZu5@Q{;HmE!Cp5ms0WzBgy<= zM()^hqFC)`O6b(+wP_cB1FIAUzt z7aU39Lb(PT=Mw+Y@@srR&e(@+QU<``-WxWa)&(tM0p=;*sZSLGHG&FgrBC7c`%c$x$;YczRrIYu6K@0$6Jq-v>sP-IaF#AF7Jg7Z4!fBj?u2Xtv`owEN_$#^dcXpOty_@Vnx`Za}bH zXC*GDnR;D(Guc|nI(8=%QhhInU3MtZ8kWde>#V%;zt%c)D%+4Ra^-V*c;HJE!u%0> zZ}ax_}}O6s63&u83AMyt}wO5=%+!r&IA*7MA}iNt`h*A&JL{2MKz9N^~ z^XPQ|X}6C86CuJC!lAJq%!oqV3N(tIMd?$Tv!Jp0l_Wtql`**<7R83I>M*xGDN#56 zrgX_MV8;9XRs&c}%!w&sC<_EJSO0?rV#<{;tGJxDz9YLivrA{k-OIXA?qISYomZkm zvKNahK3{5Rm!>}o(B8uBbklHmHEPIw4G*n?5Zmi&o}aDsMoHl9CXvkIoaMOtvX~e{ zwQg$`Ts$rY5$CS_r3FYZd1v$^(=lgZbEm`&u|2ak>M6}I;wQR7R>^zc%N#kORmZyt zvwP73ys5b1Ul=(0`BAIVeBdwnm(Y0W59={E2?5y=rwd9l8atpSh{X8vN3MN7Q^s?p zaUDOdm8-;;(}6k_SrE+8{DF;N1~jUh{0Rs-pT74e|C94q65t~%Q_F4OR>Qc4T)%d^ zC0+H*wSbI~sWG5t5^r2Od#$Ns1rD(JzIM!q@d{TACUW6N50@2GkGy^)Bb|1u8_LyM zD_s%XCP&7@3R{YTeTa2tM~dWXvNq1Qn`Kg?k}{>Z%6`v3dt7fZv_jN_mB^vntUOa(Z*qC9S@mD_t8zB;l>Q zme<$@U4C9d4Wd#a9xLB>uu{p%9b|Si1q)KgG-4I7Hi3P%4oSx4j);+VLrhX?c=pC4o}Y~pC=``PY{ zc6cCm+Pyv7zXGOZVRLI+B?p;ZTShtC(BF|@(qy#}>xoNdUmS4lK zFL&N*mL+H#agbimqUiz^Qh@N#fO#N&UReQu-1m^nnz(_5s^TP4`@9mn zZubImiGASuri8y?;R8*2%3DeR=ek3UCaK`CvG*QgDvMB?)l#WhzU0%mu=Pb8RJ2fV z7~904n^OF(be%fysZ-#al8|8dwr4wINm3}TJ9hD6;>^(_)&3uV`d(woniCbab0%37%j;0vA#?T78Olr!*=kPzhp*rMDUi{*cNI%h`xHmTm2|SRw7hz)i(sc9y^6I zH}46qlG;hhr&75%?tgf@5OfV-t2V}z1PN+7ChKaJK8mBz-{&soB8rMvr88(HA>wf> zO!!}jlpjFEEtc|VPHuoniF55l)1J>CqfUK?kQ(e%y}5J^qw5xx%2ZkSE}%D$=+P{bU(+C!=jI) z26Gn7kk@ycWH)+DzvHH?3`8*rUy~~|7;8Qa`Q<(^-2(N=3wliUS6SUA=84NY>-k*m zNHUEYZG{k04z^5H-M>2HXPWzF;NWaNPqOah7usF(^uWNp-uM)+d+>S3-b{O4{p63< zyl|uthzp)oE_?d=Ze<;4%7J$1|J-dGa?CrU>&zZqq0adiYDQi>8p*;43T~zmt!h?-ulY(#aA-kt`5wf>qVC;Ui}_`8lQiY z&ZR0b1@k3~2RNbaIecCi~gw-cr9I7K>_BtITVHb7Kc^_dI_ zcac;+xxUbWHe6Iuswh`#jF~iQJMPtmDu51+#Vtt$xiV8UYz$!5D^zLMylAn(1>e^M z$Vy7$frSTPz;jN)^#$=DRw3S&VS3AO4%eh&_4%K50^)w~mhAJ8s^kKVU$VyE$EVGi zvOMdx6uZMwk9&1viR8e`DT6aJEy1ew0K*(LN~zNQ`7wdZF+p>&ECa+1F3v7Dnzb&W z_vZCL$1_3K;F%L7IN;fZ1(L=jay{|27U3--sE*~+AQ9>bO~J}7Se7hYr&*Cw3Mg3} zSQd|kb&R`4Umx0gp2LN2tIXB@-g$MqgL39NR-SUbn zdHTFC1gWF{bWF6zPmG_j6I?VqI=}!OD}9awMHmJNmwqJNs@KG&23AaU`oDse&g~o8 z*IERBGcY%dFwp zzz5%=J1M%7^<^H6D~{_p?wdZcT^vp(`n!?l43Y_{dAXGAb_6jh8xD64bHDcd3*6_?u*~JFa$;5EZ59eS z(89wE+BTZIApK+DkX@es*mFp-C)lJR>S|1nuGLEzkbC*#G5@bl(Gx==PICW+7z6$s zVid~g<|BTlo7-N>rxHS#V-u7$v6@>|S;W{0&(@hNHBhELBEf6XMpm6m>PAdJ@~PGx z%hg&PXZ^7WfYMjaQ{D$z(-TP^O*lb zFoD|}C2-w?WIU0Dxsk}fJTcj6#9GM~(C&xS!j~xV&vD9iJ1KKz({$7XW6)~Zn?~k8SE@MLIv}-C zM7HB24KArePH+h!$Cv}neP+hs%n{^f zo>ouhu@qaiw+&j~PY5zs$0_r9qM9`p!C9X&+-f=wuKhu^$h9HVc}lz>dxPi_)-a11 zeEMBlLS6w&ANEBka=2s(XD{BwLddc zl;@hR6>w4z*sA0c-H8S>+>Ali>qMik>Gd+YHcG}(+6A}gUHBm!ofKe=0biVFI3vA& z$$Ek?hU#G^=7yNA&uA|uakx{hjmHYFigFzWgul8TFc+6I=>r`Cm;;3|>IxxIzJw~Nh}qhP_< zsrR~NNH*J<0LuJ0?Ob2!>=qLK6}+ga0`uW3(7d>qlM_`3g`V=%yz*2|Je>5hk~#%} z1o|g?G`b3Obsy^JB7Q}NCrcwN#QmGOzvieyknmmWuFUk@)4n{RuS%FfEqW3Y(6(INxLk$t8WHwJ&lmdFJxlYo zLBiwhbtl*Fbd^l!IdZ~=B8Qxa7M8F{#}Ox*2uk68&nJ@?`~7R1|KDVQ!EkDj#j zP;DjegQrco&x<4psWHP?X9xGzPrPd00nOA)M;f_cex)}{D` z(Bpc_d|FeJJ_D1Qo2&~*W}9Cr7oC@I7XCfMWvlm!#(=@^US{Pvvsu54r_9uH^E?#Q z*`L4DgRV7ys1Hf>1-Qy)W14us;4w}xzCN(LhHKq(y8 zJ==tN!9uF^H*tC6PbPx~vBy4b_L5cg0X?#{x{ZUjT~*M!Q!z?`Z3ikkzwkM- zXBX_v?7_0d_ilkN^;`??6$k`(T|Rfq(O0IrW$CFTTBZ>W%bO}hl?69E;c1fmxTPFa zu8^P{fTrdF0TM08pR_!}`^+rAopLnIhL!KgHk%Y!nUAA~pE-Gs? zF#Y1nLrKtFlh~1t>J;rb75h!$4Sl2ZhzzqQL-0_Ob!5;+`MzZy@|C>hEmy^Lzn-W{ z1iN#N$udm1pR`Pc=5gB1AAsN6RK7>=K)DgmhmVQLUKs@%ETK^aZca`FP46t;ei$w@5rjw#OCf*{H9*N z9;wqL`|!`2wTym1g<0?E4zqFEICwC>+Q|kXMBZ+mFbIsh0B=1H<<&XXE4Rg5z*<+R zOym-;N*LcV*zKUp4W1^<$S;zB$<+L>P@Jh9?>X(CRMHx*B21s_gSyxsO>i2jOND=$ zA+NQa@gg6lWO#Hi&cpgHlRJ?kKc(oxX{GVz3gj@ly9O z13mFYt;hG)bLGybK`1x}N8 ziF5J?k>HN{vHpB{GrHz)A=m5cmS-a`FE~jA4>rG_#5y`Rviwy zp#|>3jQnDue!^33BADHs!B@(}XVGU9&^v-_rqck#zU4_=T*@_HV149H6eqIfpETd5 z6+^aXDnS%tobU4Ewf||}J;!UJ%aCpT3~2nf0zV=p=2oR!s<~F-G%$~XXbZ9S((I+& z<3NhmIs#uTZ0=PgLDKN*XO?eGJ{yXC(PY$Nxf&I=9(|e^&fJh%{M3~|*IH8q<)rx| z7#s_mo_(@4Aroe(vHr`i0(X+jX0&4HBFG;U-&Oi)!Rl4yB3xtPjMu79LI}B?=0tl9 zD6tcCv+1gGRzX8UmW2G?-gFxJfxJKf5f$2{Hul;=k+*w=f|3IkR>YH6H@FC&8M!0w z<(A9)-iTj!Y+xy-1~+ zG0XQC9)&*m*nGe#6?8k*Or?~jIo~FY+AawX@WS!LVt(`jgrtWIJ?t9b?M+MZbujr% zXg909m}Qqz;6<$@J5R-@r!$Vx;S4`IvwV(8C>9>buDaUVWj`%m|GKVXP@I&K;sMn zL~hqkv&>71DdC&m(i2 zl7=pPbVln!$R_-%w$RgM&V+C?pkH9%QmjT+#*fm#{Dnt>d`6)644VPM=5OAoqU<&C zThX6#drmX!sm;3vk+Q$s;?;a^q#jxw#DmohCWxnM4$^sQEyeZ$txDtX;@aFBS7i+C zXsNa5DNF^zInRKDy!B9?(q{Uj>O|U*B}uzpLSJ|ESim>wAK&Z69q6{4n7_rM^My?u zOJwr@FW#-;5Qzhk-f}>4Z!0Q7_jTO&KH(NMuBPWIFXWfQo_i%Nhlo$mXmVbO*Vbrm z6)Jb$dJMzm7Nu}dnq%<-hj|Y>NpbrOs$9r?Tye!PmL<%k21HXNXPThl{p2V!b;S%W zY_O`PND-4V`Nw!}Tw?QnwYAiShoI(HqHXF6k)VbADPZ7caJ7sY=c%9YRQ&`o%iV0d zJIs+N^1Hw8O9$c`kLpqT3>dz;fvJ58lr`8XxY*d#Hm8;DIafitF>3v6&v2!wq(6-CQsejO%Cx`Zn=-(%M!bev$^b z^zV-p&bbeVIf^Gs4Jf8QzKY*(s+&J=tV$ekD8DMp&GAe_9X%6T z0TIA}+mXp_3ija~Wn&v`AJSRbKhRhi_DDFdqd>79*uWez3-7YlCcllRR({BN(L08F z?bmeS0_~@re}5!siTTKEQYGC5R=x$EapDMP6a!IJn%jLfj2`V6DPu*vQqB5a;4T4BcN?Yeh^WuEZ~Ij@;LGhX<>cdfE-u zOjBgra^IVc*0C#a0jP{DI}Kf@f9>-&PP6z(0v|QPvnuQG0Y)q9&TtMWNe3~Ce zd)^e{M;(7iw-n~zkyR|`5<~LCfYN@dRR$n(yOacv0N6q>m;1h@WOBedTG~H4Y#N~z z=|7Ikd&~CtuV7(QUjy~&fVC{9V20_>c2DxVKkbPpsTzfF)B%}-{K=LDOjP}xnT9N`DYUgV__WX|BJ&Q4451W zE_Yn<#bW@m{q@LwCIUp@RgRIBmS@yL?ne-B1tF%V5eHQl|&MF zmaGo-936_Cw25(o_bLn{8EQy}tOJ1m0zMfxZ%7EP*idLBAvY3I~!pDTSmW`@@2W=}H4g35`Sz*EgTs|yu<%lj@VH4Z^|^#SENagp zBdPKM7ctx%5@UIABLztEhn;(rjB=|W;Y4~sO3&u>Hr5DF;O6xn$cF>I?#EnSwH!|e z=OPPcJBEAuZLh9=UXeP~aT(QkE^f0LMF4PJ>#Wc`%(}=-@HuqSEUU!VU?4!|e%C@> zzvdT(kz6#oG_8T1_{(D!{r}9+ST&FfaYK@cJuZKN22jkYhc{0vP+M(|aZRD9=UHCP zuDy8YcX=MQvbMUMfZQE;TF>t?Ye||}E91$YQGt7~&=w*qYXwQWRPfOtdqyHOapt;k!xXV7e=I|wKhPZy z3SVxmXc`ft`Rnqkj(9FVNo|1_f202RY4T#%(1M$GZ#8P9EeK*R1q2aFW?yD*|Wy_ z#dgLqS{#FMm5=WH51>BT-_#LQcYHn%3>Y)|t@O*&hD!ZT^^i(nb%4!xH|DGOE?=ZAc$?6Q9#~u0=cVbUKD@rW@S>SMzoY>Spjvv(`@tX6SvX~4bFP&B3A z5FM$GVW^2Ybk3yh8t@I9dyDcie7f4MKN}a1PQ(!-#mVDvBk}{5-C`~}K%;>ws~t{} z2GBL8(jI9%08UCzv8KMD1LU7+bh*YEf{P0x$)~clcorTMJ!_jHpPqDr-INq?y*qu~ z{Q~=r)$NalQS))2sa$kS`d_}wwxg%ON* zE68uadq(41gpoWCyTQQid`Ropo^Oz=y{V4l;CG%!Xfo!NV2Y zn;3suRhypX-03v8@QZ=|f+BL$rh)3hy}XYXPzn#&e3YcRWf2y%yo8JcM}x4dp}p+o zZ)uF$!Y4R>@W)`@=>Gs6n*j`eJ!cAx(A)E6-k1{S^NShOI_26dX)ABpiF4bGk#zxQ zvQ`WG_0tBf4!-A;x}q_I_?_HC!LsP*@kXn^nvZ!G{@sXw2@Xcq0K9fioR|tPa(3P^ zdf5f^8{|gRBRbyIQ~jQUjc?x2LQ?S<62Abq%bNzatl8RUt&lTRp-5t8%?X0oF%vTi zVLPEG(VF`YRN{Cv(44;*DD(04>M9zB54?pzVcz#|OS2pJ;pV5zTm*%)ESB~Zu&Iwr zeUG{aS<~L*IxHJqokGg`Y|JfW%Krmcu(Q+ND`;I*mRy*yfidHrMjMY&*GJB?YLE1Q zbE;Lb+TLYN!D~b2+t7VegGL1WQsg2TV#>r%)Wvx-=#lVI33fNT*ZL`O1b$iG*BlJl zGZ3Ywdh7n-6J2cD$n~(TLG1cki%OCyn7F} z1`U|B&vkV|FZ|!^(}DL9YpOD2+s;k1eNxPy*w3sapeuX5PBfuh+2xLgqEX$u?kZlp zolpaps0LJ1EBq`7y)Y_d?(69=$DFzN7_1#FJfAqw`X%CSe!fPt=0>UGjwN1CVRew; zk3s#jr5mu`XQ)vpQc{)%DM6)I9h!p!6;-O2FWc~c6mI2Ew>C>_#am_()x8n_w8UK) zqjdK7SN7txdirP|mH?NA;w2n`{+S!8U)me(y66*i|KWw)UW4<}G%UrF*&JpRxx9z*3`mW!ztUKtmfN#tMswt$CZzVXtdF`%pVqF)wHf{9q#OkM2; zIiRzq@&hq1B|9}4Zeyiq-t|4u13EPaZO5lh6nuC7&MNs{R-Hh!_Jr+$lPvAhw**|0 zh4~%r?^wQa0o4XEl`+QKSGk4X*iIX(T~M+w2loBUDLwZlVt>-}%j!82WmVC|?mt@w zR^`)p{^)i9DZjlQJryMf_3;h*|H$hy`zQ0;JKRU0BMKjL95TLk1!j&JQQ{{x6(jA{B_uxT~X+0o0c2h$U3`BD1vj*<1g&&uqE2eBECd2G+~AuyBcQOU|FhPrBb&y1WoO6 zV3*~o)Y?%>KXb=8nmRrEc1ems_o1?v?IsJpng?3>64bT2`9b;4X0Qh**;WN^Ra)_1 zW7=yi8=o!t)lnBcEcgSOn? zm7R{w)$NoKljrJMhy{_$o}KRo`k<(!RosNlhfFz73fXS9fitCl6?0C#zo$|%ctt$c zBK5Q?)OEy=^6=Vrmi#?d<}5cqu4xe%=Ho}f+S?!(G9WWu#3&1o*jIm0vcoTf48B4U z+SNzhKsV>P1haEjn4EsR)F+Rt2RZW%zMe{Eh~eFdnQ^IAq4~NzE`u}^t4UNwzA698 z&In3Ym%6I1bQ^x!jEBoK7{sx+>Z_MxyQ-Oeh)7Sy`QnP+IQ}SO5f}Hm5Y0FABVOk# zYD|h8eyb{bd|eYX7L>fRMYwKvRsBVilv_w71SE@REgI}M(CZfl?Q`Q#xpwb0`d=Ynpl%<7P!X!X&Z7$IKaT|($cpF={nRMbzOz_QfwlvvH{iBWWvv*SPx6bQd_byAVMUI$7psv}mnp#I) zOB$F7XhCQ>bHw2&y|NPuOFz#TsGcGACskIjA|iXy+nAOQUpve!r%E6!9AjW!_v>>v zezo<%3)HIaUH!+@pd5YqU0k(5ZDY z_%Qs?YB{aFHAT$N=PPWMTTMBtMS22n4gE9^v?nEATF%~vC2=@l26wu4JqgSa+6cxK zTxOpBspflLzva$9rcULPTBd4)hLr9Lo#e}-VqrZ;DdI?AEJk)myoe*zIxiJD%c#$> zqCm~P(1@sHusU!i?`8MzcfwI>1THHjg6&-f)}&o($9x)|n`InxJ1lfl_LgHV#d)RA zlB9O%4k}#|8d?ai%SDWH`^ZdrWYV$rNG$72yCNXBNgvz=9_&;Y&X z7hhtl&1hWf`UQ_Q{(3}y@m-xExIZxLSud~#QuEP~pUaqJ7W4j$rO@dw=(tf`_3yc= zW;P?IVv%;8LSH5||3GfE3CKxW?XZ8hjo|r1Gialh@Q9H);vjKOtlMMbC75EtOtr$M z;uO7ZHz$f}ON5LI$2sZ_(9_kF0HcLepMmTO;|zuhA}f#d$iQ+HLJ&h<+W9=Kpnb(j zG}lz8UFDM7SEPmp@cVzo({ciU%FiZ@fTr5QB z=fQcaYdJlIxfT#riJj_(PY5si!+amOY8Lkg`s6D+$eBx3TE4BHhkCtYNTw!79ZW!< zwdd}UWd%#%gu=IJ7%Y9^!g>=TMjrM){S)iow%I4sYtMK5FlzrRU*idKOv@PI2W+)t zJBcaQX8bz^=}-JF|8GHuiMlgw`8keRu@Q>!5<20PL&LO-7Rg)Y+D zI=Ehw-*^hzv=08*GwMCdnGv&73*MG5C8g3X@63(9Ie!A$5tOXBRr&FZ_|w`J3)WXbssMm6xPIYZgcT;UKQ!M}xesQcY!NyrlHM;8_C%yM{bMBtI|hYxkMVa=&PzQv@q*ovTmpxrfg*&Z#D^_OB;!hD%)VB-Ddz3 zBnvfVix3R=~c z7=A$}Y09VGe{L=b0?Th_uC4MM_9Y4NWbVY_Ld^onz8>Z^jFiDZs1cbk4zNqZDJW~b>`>BkDsZ2F7Oj=sSdp^Dr+jv zA`4>m2wIsE++xz#g=p$3N|J9QQ15i(A=p}Io_Rzi(Ehv#X`Jn3ujHU6oSXI~0-PNL zCZwG)n2EPZyJ($l4|ayFUzpH^ugoRn>z5_=O%Yi%Wj3m6t_&n^t;u4y=`dX@LAk2( zJ3GGlIScpPT;2xsoy(Y@%cC*NbbF>0SP-UE9lx)5nu{X|g~LB&)|T%c`*o!=$W{zr<@d25h6$ER&pQP>z(28z;v+_* z?!6?OW8eVzkCP6l>ai2^!GKN#2K46(DXkBV-B%u8Q!KxY=!nz!WQPMl zW$v&E{yZ&nq@ku15w>VR(QFtXiYD zKj?Id$S`FSuT%`?Qu-U49Seqnkh33TNZ1hrawv11`g`w=z`fe_{7pbC0DAy5a()<1!r4MMt03}$fnnUL31@`Sg$(?d zJ}}w!+r0!YZY=b}anG4>0={-P@ZK@6RSBD@D~N^XM-2b~fs@e7!Us!;9W;t4Lv4Si z*?vj|ZDTHRb#{)o8uGI~9ODw}4rnf-M77j!=c5mk9=?0xhf7E<>6|1~4k}H9?mdux z4>&j92D&__IpUQ_#}P(q<%?gd{D3V7dCfLE)KKDa-W>ATF8P+{1A(pzSaTsZ0fRgO zM#7kx*{d%M4!;d>@Sz$;kMY(;7IMS&&752IN4c|=o{+6JAL|-Ff9uui>nYK@$=+@D zb+{CLnY&-N(S!d9=OuQt@Dk2I(+obwXKhd_Xsh6c;ktcutpGB(-#_T4uZ|txm=-5W zFWd0ab4L8i?JzNFa|vW}Ufz2$s(;j^O}|q%SMMWez~&}p{}ad*v3Y2W>SiRQoj9-v zHo{}w2D3wQoA#!1vvUR9R-2y}g#X}JX6J;L<$TxL`hNjN3c2;XL#dlfHi$7vXB>v= z!`8H2&QV#5JDh@g3gY}ts4k(a&6nRAksc30j-*#*JIUyeDzoZWvPZbMRE&T}Jo;AN zp>USb1zv~c!K#{cD{XEZl3ltH)C#SsP9%35h+*8V_JbG4af=kzCJY%6W;^o^U z@M*G6u`UNpezjM=qe!JA`EfFkH)pL;k-5kjtB)U=OP)PPtyPl*0tN@CrA3SAFdr=B zas4Y}Nx2go9COFLViI=}4;3}O^w{|-Gr<)0)K14!b)@;ww1~Tvdx2c#?w;{vk&*3P zcZPL@kcg!t(1YH!^{cN~m#6lnBDcjb5Jd2V>YVayg)6WOwy62|DCZ&GU7mpveItIXFG4NXhTl)}Y5i4;1kG zdUc}0+haoAhye8+Yi`o+eknedL>*o~Jw!O+Y-b zPeJMHRmSUygZg%>n^Xgqz#N?S_NCC5V;4}jc-%Q&8@Q@1g!Ey4Afb^=Db|I9> z7X)=$FLOl9>M3a{mN_}}tz8>On^wF?V@4#Ak~>s-Hj_51c924)-2OtngTnqBheeD9 z1I)+H-`+Kp>T)BRr9~YGG`$scyJfdTP*?9CQ&*4fq{#VPb``0y6c9p_oE+3rz&=nA z1}Aq~!=C$`_wIN83BMEAG(;B2$s}+qqClKUxa6EE#c{t5+D4jma`Frel^j=SBQXQF zlh`+>ttVrdX>?)VMJ=t%?S2kd3&&cnx`V=Sz_>w=mx|T9kSybmmHh`pQB4$Yv;Y`) zQPkHo;UToqj&)ad9N~!I;;W5HX^4yDaPY`Jl-rBP= z1#qQpCO+TC(NZQnOjQrV**3*2-S4f>nk5h88KjAPomD?2r9 znFDWcTJ$L|cSEggQb-;_o|)r;(yo7JkCjS*0LQ&uNOraX0|y^kvlIbYnVT(~Vzse1 zT}&6WSe1YTamH&S#Hl1&Eb{Fj5CJ}!uH@Pz(i7J!laAHQd_`4{!^8}My+%7z(#1z~ zd0iAyQ*)>&>p|v<06;N88Kt6?fGDDZGe+PBlOK&O6u>B=iY@^~2Z2H9OaRa*rOhD; zj&nt5TU$q{#v@kQaz5Za8``N!3q>O<4Xh3bG_C?hyVwf*SN1~in9{s8d95<9nRepd zFjtbe%N+hy@&|@Axpi$?3wRT7lR6}cgMXRTUAL%0r}-9=`l zV^svSXFqjkwU~j%Fb+A#YQ(*@VxaJ+zI)eYEL(}%r+@H~+PQ1x%et2VvVLrI_N`*y zLpjGoi`1f3mA4XclhU~>;k}2=vjgsd&NE$}u1A}}1axERT&?n%9gp51mGm{}(QZ1k z%BxK~qmtAa<1V<|%6Z39eNAFtPQlLR%w&AK1~7f=e^65!Nrm~Fg3HwVS0jHjt3F8` z%J#*51qXdu^H_-L(N6O0Hg+%_w-d%|Dp`vhmjIKJSD(t6d3aTZ8C0mr&1SvPcR3&v z{oa-A(^?)iLr6-^fH+S70RFmFZO*s>vyezR#ZrQ8xLgCCxbId_20MY;Gtdn4T0M3- zJx5$hsUB6fc_~m$7~iXO?aznM*GB9t~me{1fJCi zbw;sT7#gc);lMfOvQJvSs2#Go&Rf1}hH`UMYZPf|e>`pu03Nj*%y)8g&rwt9sgxu5 zjV0V$55KUfxi4~t-nkj21K0|fCU){M-kvSPjt^=FaxJ{$0;Tdn!RIvf8wOZ_J2f?> z}xkhh)u!jYo@gDznYG(IXp1q1Jbj#9|K$5w2_nm9FDy!Zq_?}B_mIjnMQJlioUb7khID| zx4)sP*I{!XJ9w8ulR?xT7W;>kZ&6wHsAP~XGtGLBk94o?UByvU#5iC(SDo6x(8xy8 z4NFF`mYSLl4Yip#1d)MU{NF5{Fnud->TpPlJTF|;8KCn%VbFJ^RnbXCSRVPS>+;4} zjy43L5|;AJw(||vj*sK+NX|2Wyz(;U@mA3B3;A-N^ zi_~L2ln9z;7~Bc`J!z`p*PQ1)Di1hggWvR~sVyPE+O+o6EKa8c;EzfOBygabbYLP4 z%ilF3k&yd8F2bl*AxNJijD2ZdKb#)ItZo~+FOyRc~&DMJ%twP z!tD-Oe8xer1K%~Huz3XFeQP=+vQpXaz%_kX{hi#f;Yp^Q*qKYhu!*&6hwiz-6|1P( z09cW^^PXylgCYP=H++ye^{uN<<24hLc>Hugan-%)(QCPo3rBpPp+6rYzq zhL%QEPzV`3)J9Bl!zZW%y=Km+`Hx()79TJBB=xCA(wOb-Rs6+alh&jT4gmYZ^rN)RLmA1CV!3U)+oy7c$5zCTOU|=wFp7m}v3gZBR zdjZaBns(0zJuy}mO~R( U)TTw=tivogI5l>*#uwE8*<&4X;s5{u literal 0 HcmV?d00001 diff --git a/application/yolov5_example/data/scripts/download_weights.sh b/application/yolov5_example/data/scripts/download_weights.sh new file mode 100644 index 00000000..a4f3becf --- /dev/null +++ b/application/yolov5_example/data/scripts/download_weights.sh @@ -0,0 +1,21 @@ +#!/bin/bash +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Download latest models from https://github.com/ultralytics/yolov5/releases +# Example usage: bash data/scripts/download_weights.sh +# parent +# 鈹斺攢鈹 yolov5 +# 鈹溾攢鈹 yolov5s.pt 鈫 downloads here +# 鈹溾攢鈹 yolov5m.pt +# 鈹斺攢鈹 ... + +python - <= cls >= 0, f'incorrect class index {cls}' + + # Write YOLO label + if id not in shapes: + shapes[id] = Image.open(file).size + box = xyxy2xywhn(box[None].astype(np.float), w=shapes[id][0], h=shapes[id][1], clip=True) + with open((labels / id).with_suffix('.txt'), 'a') as f: + f.write(f"{cls} {' '.join(f'{x:.6f}' for x in box[0])}\n") # write label.txt + except Exception as e: + print(f'WARNING: skipping one label for {file}: {e}') + + + # Download manually from https://challenge.xviewdataset.org + dir = Path(yaml['path']) # dataset root dir + # urls = ['https://d307kc0mrhucc3.cloudfront.net/train_labels.zip', # train labels + # 'https://d307kc0mrhucc3.cloudfront.net/train_images.zip', # 15G, 847 train images + # 'https://d307kc0mrhucc3.cloudfront.net/val_images.zip'] # 5G, 282 val images (no labels) + # download(urls, dir=dir, delete=False) + + # Convert labels + convert_labels(dir / 'xView_train.geojson') + + # Move images + images = Path(dir / 'images') + images.mkdir(parents=True, exist_ok=True) + Path(dir / 'train_images').rename(dir / 'images' / 'train') + Path(dir / 'val_images').rename(dir / 'images' / 'val') + + # Split + autosplit(dir / 'images' / 'train') diff --git a/application/yolov5_example/detect.py b/application/yolov5_example/detect.py new file mode 100644 index 00000000..c699a749 --- /dev/null +++ b/application/yolov5_example/detect.py @@ -0,0 +1,257 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Run inference on images, videos, directories, streams, etc. + +Usage - sources: + $ python path/to/detect.py --weights yolov5s.pt --source 0 # webcam + img.jpg # image + vid.mp4 # video + path/ # directory + path/*.jpg # glob + 'https://youtu.be/Zgi9g1ksQHc' # YouTube + 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream + +Usage - formats: + $ python path/to/detect.py --weights yolov5s.pt # PyTorch + yolov5s.torchscript # TorchScript + yolov5s.onnx # ONNX Runtime or OpenCV DNN with --dnn + yolov5s.xml # OpenVINO + yolov5s.engine # TensorRT + yolov5s.mlmodel # CoreML (macOS-only) + yolov5s_saved_model # TensorFlow SavedModel + yolov5s.pb # TensorFlow GraphDef + yolov5s.tflite # TensorFlow Lite + yolov5s_edgetpu.tflite # TensorFlow Edge TPU +""" + +import argparse +import os +import platform +import sys +from pathlib import Path + +import torch +import torch.backends.cudnn as cudnn + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[0] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from models.common import DetectMultiBackend +from utils.dataloaders import IMG_FORMATS, VID_FORMATS, LoadImages, LoadStreams +from utils.general import (LOGGER, check_file, check_img_size, check_imshow, check_requirements, colorstr, cv2, + increment_path, non_max_suppression, print_args, scale_coords, strip_optimizer, xyxy2xywh) +from utils.plots import Annotator, colors, save_one_box +from utils.torch_utils import select_device, smart_inference_mode, time_sync + + +@smart_inference_mode() +def run( + weights=ROOT / 'yolov5s.pt', # model.pt path(s) + source=ROOT / 'data/images', # file/dir/URL/glob, 0 for webcam + data=ROOT / 'data/coco128.yaml', # dataset.yaml path + imgsz=(640, 640), # inference size (height, width) + conf_thres=0.25, # confidence threshold + iou_thres=0.45, # NMS IOU threshold + max_det=1000, # maximum detections per image + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + view_img=False, # show results + save_txt=False, # save results to *.txt + save_conf=False, # save confidences in --save-txt labels + save_crop=False, # save cropped prediction boxes + nosave=False, # do not save images/videos + classes=None, # filter by class: --class 0, or --class 0 2 3 + agnostic_nms=False, # class-agnostic NMS + augment=False, # augmented inference + visualize=False, # visualize features + update=False, # update all models + project=ROOT / 'runs/detect', # save results to project/name + name='exp', # save results to project/name + exist_ok=False, # existing project/name ok, do not increment + line_thickness=3, # bounding box thickness (pixels) + hide_labels=False, # hide labels + hide_conf=False, # hide confidences + half=False, # use FP16 half-precision inference + dnn=False, # use OpenCV DNN for ONNX inference +): + source = str(source) + save_img = not nosave and not source.endswith('.txt') # save inference images + is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS) + is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://')) + webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file) + if is_url and is_file: + source = check_file(source) # download + + # Directories + save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run + (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir + + # Load model + device = select_device(device) + model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half) + stride, names, pt = model.stride, model.names, model.pt + imgsz = check_img_size(imgsz, s=stride) # check image size + + # Dataloader + if webcam: + view_img = check_imshow() + cudnn.benchmark = True # set True to speed up constant image size inference + dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt) + bs = len(dataset) # batch_size + else: + dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt) + bs = 1 # batch_size + vid_path, vid_writer = [None] * bs, [None] * bs + + # Run inference + model.warmup(imgsz=(1 if pt else bs, 3, *imgsz)) # warmup + seen, windows, dt = 0, [], [0.0, 0.0, 0.0] + for path, im, im0s, vid_cap, s in dataset: + t1 = time_sync() + im = torch.from_numpy(im).to(device) + im = im.half() if model.fp16 else im.float() # uint8 to fp16/32 + im /= 255 # 0 - 255 to 0.0 - 1.0 + if len(im.shape) == 3: + im = im[None] # expand for batch dim + t2 = time_sync() + dt[0] += t2 - t1 + + # Inference + visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False + pred = model(im, augment=augment, visualize=visualize) + t3 = time_sync() + dt[1] += t3 - t2 + + # NMS + pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det) + dt[2] += time_sync() - t3 + + # Second-stage classifier (optional) + # pred = utils.general.apply_classifier(pred, classifier_model, im, im0s) + + # Process predictions + for i, det in enumerate(pred): # per image + seen += 1 + if webcam: # batch_size >= 1 + p, im0, frame = path[i], im0s[i].copy(), dataset.count + s += f'{i}: ' + else: + p, im0, frame = path, im0s.copy(), getattr(dataset, 'frame', 0) + + p = Path(p) # to Path + save_path = str(save_dir / p.name) # im.jpg + txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}') # im.txt + s += '%gx%g ' % im.shape[2:] # print string + gn = torch.tensor(im0.shape)[[1, 0, 1, 0]] # normalization gain whwh + imc = im0.copy() if save_crop else im0 # for save_crop + annotator = Annotator(im0, line_width=line_thickness, example=str(names)) + if len(det): + # Rescale boxes from img_size to im0 size + det[:, :4] = scale_coords(im.shape[2:], det[:, :4], im0.shape).round() + + # Print results + for c in det[:, -1].unique(): + n = (det[:, -1] == c).sum() # detections per class + s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string + + # Write results + for *xyxy, conf, cls in reversed(det): + if save_txt: # Write to file + xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh + line = (cls, *xywh, conf) if save_conf else (cls, *xywh) # label format + with open(f'{txt_path}.txt', 'a') as f: + f.write(('%g ' * len(line)).rstrip() % line + '\n') + + if save_img or save_crop or view_img: # Add bbox to image + c = int(cls) # integer class + label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}') + annotator.box_label(xyxy, label, color=colors(c, True)) + if save_crop: + save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True) + + # Stream results + im0 = annotator.result() + if view_img: + if platform.system() == 'Linux' and p not in windows: + windows.append(p) + cv2.namedWindow(str(p), cv2.WINDOW_NORMAL | cv2.WINDOW_KEEPRATIO) # allow window resize (Linux) + cv2.resizeWindow(str(p), im0.shape[1], im0.shape[0]) + cv2.imshow(str(p), im0) + cv2.waitKey(1) # 1 millisecond + + # Save results (image with detections) + if save_img: + if dataset.mode == 'image': + cv2.imwrite(save_path, im0) + else: # 'video' or 'stream' + if vid_path[i] != save_path: # new video + vid_path[i] = save_path + if isinstance(vid_writer[i], cv2.VideoWriter): + vid_writer[i].release() # release previous video writer + if vid_cap: # video + fps = vid_cap.get(cv2.CAP_PROP_FPS) + w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH)) + h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) + else: # stream + fps, w, h = 30, im0.shape[1], im0.shape[0] + save_path = str(Path(save_path).with_suffix('.mp4')) # force *.mp4 suffix on results videos + vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h)) + vid_writer[i].write(im0) + + # Print time (inference-only) + LOGGER.info(f'{s}Done. ({t3 - t2:.3f}s)') + + # Print results + t = tuple(x / seen * 1E3 for x in dt) # speeds per image + LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}' % t) + if save_txt or save_img: + s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else '' + LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}") + if update: + strip_optimizer(weights[0]) # update model (to fix SourceChangeWarning) + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5s.pt', help='model path(s)') + parser.add_argument('--source', type=str, default=ROOT / 'data/images', help='file/dir/URL/glob, 0 for webcam') + parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='(optional) dataset.yaml path') + parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w') + parser.add_argument('--conf-thres', type=float, default=0.25, help='confidence threshold') + parser.add_argument('--iou-thres', type=float, default=0.45, help='NMS IoU threshold') + parser.add_argument('--max-det', type=int, default=1000, help='maximum detections per image') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--view-img', action='store_true', help='show results') + parser.add_argument('--save-txt', action='store_true', help='save results to *.txt') + parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels') + parser.add_argument('--save-crop', action='store_true', help='save cropped prediction boxes') + parser.add_argument('--nosave', action='store_true', help='do not save images/videos') + parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --classes 0, or --classes 0 2 3') + parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS') + parser.add_argument('--augment', action='store_true', help='augmented inference') + parser.add_argument('--visualize', action='store_true', help='visualize features') + parser.add_argument('--update', action='store_true', help='update all models') + parser.add_argument('--project', default=ROOT / 'runs/detect', help='save results to project/name') + parser.add_argument('--name', default='exp', help='save results to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + parser.add_argument('--line-thickness', default=3, type=int, help='bounding box thickness (pixels)') + parser.add_argument('--hide-labels', default=False, action='store_true', help='hide labels') + parser.add_argument('--hide-conf', default=False, action='store_true', help='hide confidences') + parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') + parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference') + opt = parser.parse_args() + opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand + print_args(vars(opt)) + return opt + + +def main(opt): + check_requirements(exclude=('tensorboard', 'thop')) + run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/export.py b/application/yolov5_example/export.py new file mode 100644 index 00000000..595039b2 --- /dev/null +++ b/application/yolov5_example/export.py @@ -0,0 +1,616 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Export a YOLOv5 PyTorch model to other formats. TensorFlow exports authored by https://github.com/zldrobit + +Format | `export.py --include` | Model +--- | --- | --- +PyTorch | - | yolov5s.pt +TorchScript | `torchscript` | yolov5s.torchscript +ONNX | `onnx` | yolov5s.onnx +OpenVINO | `openvino` | yolov5s_openvino_model/ +TensorRT | `engine` | yolov5s.engine +CoreML | `coreml` | yolov5s.mlmodel +TensorFlow SavedModel | `saved_model` | yolov5s_saved_model/ +TensorFlow GraphDef | `pb` | yolov5s.pb +TensorFlow Lite | `tflite` | yolov5s.tflite +TensorFlow Edge TPU | `edgetpu` | yolov5s_edgetpu.tflite +TensorFlow.js | `tfjs` | yolov5s_web_model/ + +Requirements: + $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu # CPU + $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime-gpu openvino-dev tensorflow # GPU + +Usage: + $ python path/to/export.py --weights yolov5s.pt --include torchscript onnx openvino engine coreml tflite ... + +Inference: + $ python path/to/detect.py --weights yolov5s.pt # PyTorch + yolov5s.torchscript # TorchScript + yolov5s.onnx # ONNX Runtime or OpenCV DNN with --dnn + yolov5s.xml # OpenVINO + yolov5s.engine # TensorRT + yolov5s.mlmodel # CoreML (macOS-only) + yolov5s_saved_model # TensorFlow SavedModel + yolov5s.pb # TensorFlow GraphDef + yolov5s.tflite # TensorFlow Lite + yolov5s_edgetpu.tflite # TensorFlow Edge TPU + +TensorFlow.js: + $ cd .. && git clone https://github.com/zldrobit/tfjs-yolov5-example.git && cd tfjs-yolov5-example + $ npm install + $ ln -s ../../yolov5/yolov5s_web_model public/yolov5s_web_model + $ npm start +""" + +import argparse +import json +import os +import platform +import subprocess +import sys +import time +import warnings +from pathlib import Path + +import pandas as pd +import torch +import yaml +from torch.utils.mobile_optimizer import optimize_for_mobile + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[0] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +if platform.system() != 'Windows': + ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from models.experimental import attempt_load +from models.yolo import Detect +from utils.dataloaders import LoadImages +from utils.general import (LOGGER, check_dataset, check_img_size, check_requirements, check_version, check_yaml, + colorstr, file_size, print_args, url2file) +from utils.torch_utils import select_device, smart_inference_mode + + +def export_formats(): + # YOLOv5 export formats + x = [ + ['PyTorch', '-', '.pt', True, True], + ['TorchScript', 'torchscript', '.torchscript', True, True], + ['ONNX', 'onnx', '.onnx', True, True], + ['OpenVINO', 'openvino', '_openvino_model', True, False], + ['TensorRT', 'engine', '.engine', False, True], + ['CoreML', 'coreml', '.mlmodel', True, False], + ['TensorFlow SavedModel', 'saved_model', '_saved_model', True, True], + ['TensorFlow GraphDef', 'pb', '.pb', True, True], + ['TensorFlow Lite', 'tflite', '.tflite', True, False], + ['TensorFlow Edge TPU', 'edgetpu', '_edgetpu.tflite', False, False], + ['TensorFlow.js', 'tfjs', '_web_model', False, False],] + return pd.DataFrame(x, columns=['Format', 'Argument', 'Suffix', 'CPU', 'GPU']) + + +def export_torchscript(model, im, file, optimize, prefix=colorstr('TorchScript:')): + # YOLOv5 TorchScript model export + try: + LOGGER.info(f'\n{prefix} starting export with torch {torch.__version__}...') + f = file.with_suffix('.torchscript') + + ts = torch.jit.trace(model, im, strict=False) + d = {"shape": im.shape, "stride": int(max(model.stride)), "names": model.names} + extra_files = {'config.txt': json.dumps(d)} # torch._C.ExtraFilesMap() + if optimize: # https://pytorch.org/tutorials/recipes/mobile_interpreter.html + optimize_for_mobile(ts)._save_for_lite_interpreter(str(f), _extra_files=extra_files) + else: + ts.save(str(f), _extra_files=extra_files) + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'{prefix} export failure: {e}') + + +def export_onnx(model, im, file, opset, train, dynamic, simplify, prefix=colorstr('ONNX:')): + # YOLOv5 ONNX export + try: + check_requirements(('onnx',)) + import onnx + + LOGGER.info(f'\n{prefix} starting export with onnx {onnx.__version__}...') + f = file.with_suffix('.onnx') + + torch.onnx.export( + model.cpu() if dynamic else model, # --dynamic only compatible with cpu + im.cpu() if dynamic else im, + f, + verbose=False, + opset_version=opset, + training=torch.onnx.TrainingMode.TRAINING if train else torch.onnx.TrainingMode.EVAL, + do_constant_folding=not train, + input_names=['images'], + output_names=['output'], + dynamic_axes={ + 'images': { + 0: 'batch', + 2: 'height', + 3: 'width'}, # shape(1,3,640,640) + 'output': { + 0: 'batch', + 1: 'anchors'} # shape(1,25200,85) + } if dynamic else None) + + # Checks + model_onnx = onnx.load(f) # load onnx model + onnx.checker.check_model(model_onnx) # check onnx model + + # Metadata + d = {'stride': int(max(model.stride)), 'names': model.names} + for k, v in d.items(): + meta = model_onnx.metadata_props.add() + meta.key, meta.value = k, str(v) + onnx.save(model_onnx, f) + + # Simplify + if simplify: + try: + cuda = torch.cuda.is_available() + check_requirements(('onnxruntime-gpu' if cuda else 'onnxruntime', 'onnx-simplifier>=0.4.1')) + import onnxsim + + LOGGER.info(f'{prefix} simplifying with onnx-simplifier {onnxsim.__version__}...') + model_onnx, check = onnxsim.simplify(model_onnx) + assert check, 'assert check failed' + onnx.save(model_onnx, f) + except Exception as e: + LOGGER.info(f'{prefix} simplifier failure: {e}') + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'{prefix} export failure: {e}') + + +def export_openvino(model, file, half, prefix=colorstr('OpenVINO:')): + # YOLOv5 OpenVINO export + try: + check_requirements(('openvino-dev',)) # requires openvino-dev: https://pypi.org/project/openvino-dev/ + import openvino.inference_engine as ie + + LOGGER.info(f'\n{prefix} starting export with openvino {ie.__version__}...') + f = str(file).replace('.pt', f'_openvino_model{os.sep}') + + cmd = f"mo --input_model {file.with_suffix('.onnx')} --output_dir {f} --data_type {'FP16' if half else 'FP32'}" + subprocess.check_output(cmd.split()) # export + with open(Path(f) / file.with_suffix('.yaml').name, 'w') as g: + yaml.dump({'stride': int(max(model.stride)), 'names': model.names}, g) # add metadata.yaml + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +def export_coreml(model, im, file, int8, half, prefix=colorstr('CoreML:')): + # YOLOv5 CoreML export + try: + check_requirements(('coremltools',)) + import coremltools as ct + + LOGGER.info(f'\n{prefix} starting export with coremltools {ct.__version__}...') + f = file.with_suffix('.mlmodel') + + ts = torch.jit.trace(model, im, strict=False) # TorchScript model + ct_model = ct.convert(ts, inputs=[ct.ImageType('image', shape=im.shape, scale=1 / 255, bias=[0, 0, 0])]) + bits, mode = (8, 'kmeans_lut') if int8 else (16, 'linear') if half else (32, None) + if bits < 32: + if platform.system() == 'Darwin': # quantization only supported on macOS + with warnings.catch_warnings(): + warnings.filterwarnings("ignore", category=DeprecationWarning) # suppress numpy==1.20 float warning + ct_model = ct.models.neural_network.quantization_utils.quantize_weights(ct_model, bits, mode) + else: + print(f'{prefix} quantization only supported on macOS, skipping...') + ct_model.save(f) + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return ct_model, f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + return None, None + + +def export_engine(model, im, file, train, half, dynamic, simplify, workspace=4, verbose=False): + # YOLOv5 TensorRT export https://developer.nvidia.com/tensorrt + prefix = colorstr('TensorRT:') + try: + assert im.device.type != 'cpu', 'export running on CPU but must be on GPU, i.e. `python export.py --device 0`' + try: + import tensorrt as trt + except Exception: + if platform.system() == 'Linux': + check_requirements(('nvidia-tensorrt',), cmds=('-U --index-url https://pypi.ngc.nvidia.com',)) + import tensorrt as trt + + if trt.__version__[0] == '7': # TensorRT 7 handling https://github.com/ultralytics/yolov5/issues/6012 + grid = model.model[-1].anchor_grid + model.model[-1].anchor_grid = [a[..., :1, :1, :] for a in grid] + export_onnx(model, im, file, 12, train, dynamic, simplify) # opset 12 + model.model[-1].anchor_grid = grid + else: # TensorRT >= 8 + check_version(trt.__version__, '8.0.0', hard=True) # require tensorrt>=8.0.0 + export_onnx(model, im, file, 13, train, dynamic, simplify) # opset 13 + onnx = file.with_suffix('.onnx') + + LOGGER.info(f'\n{prefix} starting export with TensorRT {trt.__version__}...') + assert onnx.exists(), f'failed to export ONNX file: {onnx}' + f = file.with_suffix('.engine') # TensorRT engine file + logger = trt.Logger(trt.Logger.INFO) + if verbose: + logger.min_severity = trt.Logger.Severity.VERBOSE + + builder = trt.Builder(logger) + config = builder.create_builder_config() + config.max_workspace_size = workspace * 1 << 30 + # config.set_memory_pool_limit(trt.MemoryPoolType.WORKSPACE, workspace << 30) # fix TRT 8.4 deprecation notice + + flag = (1 << int(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH)) + network = builder.create_network(flag) + parser = trt.OnnxParser(network, logger) + if not parser.parse_from_file(str(onnx)): + raise RuntimeError(f'failed to load ONNX file: {onnx}') + + inputs = [network.get_input(i) for i in range(network.num_inputs)] + outputs = [network.get_output(i) for i in range(network.num_outputs)] + LOGGER.info(f'{prefix} Network Description:') + for inp in inputs: + LOGGER.info(f'{prefix}\tinput "{inp.name}" with shape {inp.shape} and dtype {inp.dtype}') + for out in outputs: + LOGGER.info(f'{prefix}\toutput "{out.name}" with shape {out.shape} and dtype {out.dtype}') + + if dynamic: + if im.shape[0] <= 1: + LOGGER.warning(f"{prefix}WARNING: --dynamic model requires maximum --batch-size argument") + profile = builder.create_optimization_profile() + for inp in inputs: + profile.set_shape(inp.name, (1, *im.shape[1:]), (max(1, im.shape[0] // 2), *im.shape[1:]), im.shape) + config.add_optimization_profile(profile) + + LOGGER.info(f'{prefix} building FP{16 if builder.platform_has_fast_fp16 and half else 32} engine in {f}') + if builder.platform_has_fast_fp16 and half: + config.set_flag(trt.BuilderFlag.FP16) + with builder.build_engine(network, config) as engine, open(f, 'wb') as t: + t.write(engine.serialize()) + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +def export_saved_model(model, + im, + file, + dynamic, + tf_nms=False, + agnostic_nms=False, + topk_per_class=100, + topk_all=100, + iou_thres=0.45, + conf_thres=0.25, + keras=False, + prefix=colorstr('TensorFlow SavedModel:')): + # YOLOv5 TensorFlow SavedModel export + try: + import tensorflow as tf + from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2 + + from models.tf import TFDetect, TFModel + + LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...') + f = str(file).replace('.pt', '_saved_model') + batch_size, ch, *imgsz = list(im.shape) # BCHW + + tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz) + im = tf.zeros((batch_size, *imgsz, ch)) # BHWC order for TensorFlow + _ = tf_model.predict(im, tf_nms, agnostic_nms, topk_per_class, topk_all, iou_thres, conf_thres) + inputs = tf.keras.Input(shape=(*imgsz, ch), batch_size=None if dynamic else batch_size) + outputs = tf_model.predict(inputs, tf_nms, agnostic_nms, topk_per_class, topk_all, iou_thres, conf_thres) + keras_model = tf.keras.Model(inputs=inputs, outputs=outputs) + keras_model.trainable = False + keras_model.summary() + if keras: + keras_model.save(f, save_format='tf') + else: + spec = tf.TensorSpec(keras_model.inputs[0].shape, keras_model.inputs[0].dtype) + m = tf.function(lambda x: keras_model(x)) # full model + m = m.get_concrete_function(spec) + frozen_func = convert_variables_to_constants_v2(m) + tfm = tf.Module() + tfm.__call__ = tf.function(lambda x: frozen_func(x)[:4] if tf_nms else frozen_func(x)[0], [spec]) + tfm.__call__(im) + tf.saved_model.save(tfm, + f, + options=tf.saved_model.SaveOptions(experimental_custom_gradients=False) + if check_version(tf.__version__, '2.6') else tf.saved_model.SaveOptions()) + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return keras_model, f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + return None, None + + +def export_pb(keras_model, file, prefix=colorstr('TensorFlow GraphDef:')): + # YOLOv5 TensorFlow GraphDef *.pb export https://github.com/leimao/Frozen_Graph_TensorFlow + try: + import tensorflow as tf + from tensorflow.python.framework.convert_to_constants import convert_variables_to_constants_v2 + + LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...') + f = file.with_suffix('.pb') + + m = tf.function(lambda x: keras_model(x)) # full model + m = m.get_concrete_function(tf.TensorSpec(keras_model.inputs[0].shape, keras_model.inputs[0].dtype)) + frozen_func = convert_variables_to_constants_v2(m) + frozen_func.graph.as_graph_def() + tf.io.write_graph(graph_or_graph_def=frozen_func.graph, logdir=str(f.parent), name=f.name, as_text=False) + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +def export_tflite(keras_model, im, file, int8, data, nms, agnostic_nms, prefix=colorstr('TensorFlow Lite:')): + # YOLOv5 TensorFlow Lite export + try: + import tensorflow as tf + + LOGGER.info(f'\n{prefix} starting export with tensorflow {tf.__version__}...') + batch_size, ch, *imgsz = list(im.shape) # BCHW + f = str(file).replace('.pt', '-fp16.tflite') + + converter = tf.lite.TFLiteConverter.from_keras_model(keras_model) + converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS] + converter.target_spec.supported_types = [tf.float16] + converter.optimizations = [tf.lite.Optimize.DEFAULT] + if int8: + from models.tf import representative_dataset_gen + dataset = LoadImages(check_dataset(check_yaml(data))['train'], img_size=imgsz, auto=False) + converter.representative_dataset = lambda: representative_dataset_gen(dataset, ncalib=100) + converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8] + converter.target_spec.supported_types = [] + converter.inference_input_type = tf.uint8 # or tf.int8 + converter.inference_output_type = tf.uint8 # or tf.int8 + converter.experimental_new_quantizer = True + f = str(file).replace('.pt', '-int8.tflite') + if nms or agnostic_nms: + converter.target_spec.supported_ops.append(tf.lite.OpsSet.SELECT_TF_OPS) + + tflite_model = converter.convert() + open(f, "wb").write(tflite_model) + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +def export_edgetpu(file, prefix=colorstr('Edge TPU:')): + # YOLOv5 Edge TPU export https://coral.ai/docs/edgetpu/models-intro/ + try: + cmd = 'edgetpu_compiler --version' + help_url = 'https://coral.ai/docs/edgetpu/compiler/' + assert platform.system() == 'Linux', f'export only supported on Linux. See {help_url}' + if subprocess.run(f'{cmd} >/dev/null', shell=True).returncode != 0: + LOGGER.info(f'\n{prefix} export requires Edge TPU compiler. Attempting install from {help_url}') + sudo = subprocess.run('sudo --version >/dev/null', shell=True).returncode == 0 # sudo installed on system + for c in ( + 'curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | sudo apt-key add -', + 'echo "deb https://packages.cloud.google.com/apt coral-edgetpu-stable main" | sudo tee /etc/apt/sources.list.d/coral-edgetpu.list', + 'sudo apt-get update', 'sudo apt-get install edgetpu-compiler'): + subprocess.run(c if sudo else c.replace('sudo ', ''), shell=True, check=True) + ver = subprocess.run(cmd, shell=True, capture_output=True, check=True).stdout.decode().split()[-1] + + LOGGER.info(f'\n{prefix} starting export with Edge TPU compiler {ver}...') + f = str(file).replace('.pt', '-int8_edgetpu.tflite') # Edge TPU model + f_tfl = str(file).replace('.pt', '-int8.tflite') # TFLite model + + cmd = f"edgetpu_compiler -s -d -k 10 --out_dir {file.parent} {f_tfl}" + subprocess.run(cmd.split(), check=True) + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +def export_tfjs(file, prefix=colorstr('TensorFlow.js:')): + # YOLOv5 TensorFlow.js export + try: + check_requirements(('tensorflowjs',)) + import re + + import tensorflowjs as tfjs + + LOGGER.info(f'\n{prefix} starting export with tensorflowjs {tfjs.__version__}...') + f = str(file).replace('.pt', '_web_model') # js dir + f_pb = file.with_suffix('.pb') # *.pb path + f_json = f'{f}/model.json' # *.json path + + cmd = f'tensorflowjs_converter --input_format=tf_frozen_model ' \ + f'--output_node_names=Identity,Identity_1,Identity_2,Identity_3 {f_pb} {f}' + subprocess.run(cmd.split()) + + with open(f_json) as j: + json = j.read() + with open(f_json, 'w') as j: # sort JSON Identity_* in ascending order + subst = re.sub( + r'{"outputs": {"Identity.?.?": {"name": "Identity.?.?"}, ' + r'"Identity.?.?": {"name": "Identity.?.?"}, ' + r'"Identity.?.?": {"name": "Identity.?.?"}, ' + r'"Identity.?.?": {"name": "Identity.?.?"}}}', r'{"outputs": {"Identity": {"name": "Identity"}, ' + r'"Identity_1": {"name": "Identity_1"}, ' + r'"Identity_2": {"name": "Identity_2"}, ' + r'"Identity_3": {"name": "Identity_3"}}}', json) + j.write(subst) + + LOGGER.info(f'{prefix} export success, saved as {f} ({file_size(f):.1f} MB)') + return f + except Exception as e: + LOGGER.info(f'\n{prefix} export failure: {e}') + + +@smart_inference_mode() +def run( + data=ROOT / 'data/coco128.yaml', # 'dataset.yaml path' + weights=ROOT / 'yolov5s.pt', # weights path + imgsz=(640, 640), # image (height, width) + batch_size=1, # batch size + device='cpu', # cuda device, i.e. 0 or 0,1,2,3 or cpu + include=('torchscript', 'onnx'), # include formats + half=False, # FP16 half-precision export + inplace=False, # set YOLOv5 Detect() inplace=True + train=False, # model.train() mode + keras=False, # use Keras + optimize=False, # TorchScript: optimize for mobile + int8=False, # CoreML/TF INT8 quantization + dynamic=False, # ONNX/TF/TensorRT: dynamic axes + simplify=False, # ONNX: simplify model + opset=12, # ONNX: opset version + verbose=False, # TensorRT: verbose log + workspace=4, # TensorRT: workspace size (GB) + nms=False, # TF: add NMS to model + agnostic_nms=False, # TF: add agnostic NMS to model + topk_per_class=100, # TF.js NMS: topk per class to keep + topk_all=100, # TF.js NMS: topk for all classes to keep + iou_thres=0.45, # TF.js NMS: IoU threshold + conf_thres=0.25, # TF.js NMS: confidence threshold +): + t = time.time() + include = [x.lower() for x in include] # to lowercase + fmts = tuple(export_formats()['Argument'][1:]) # --include arguments + flags = [x in include for x in fmts] + assert sum(flags) == len(include), f'ERROR: Invalid --include {include}, valid --include arguments are {fmts}' + jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs = flags # export booleans + file = Path(url2file(weights) if str(weights).startswith(('http:/', 'https:/')) else weights) # PyTorch weights + + # Load PyTorch model + device = select_device(device) + if half: + assert device.type != 'cpu' or coreml, '--half only compatible with GPU export, i.e. use --device 0' + assert not dynamic, '--half not compatible with --dynamic, i.e. use either --half or --dynamic but not both' + model = attempt_load(weights, device=device, inplace=True, fuse=True) # load FP32 model + + # Checks + imgsz *= 2 if len(imgsz) == 1 else 1 # expand + if optimize: + assert device.type == 'cpu', '--optimize not compatible with cuda devices, i.e. use --device cpu' + + # Input + gs = int(max(model.stride)) # grid size (max stride) + imgsz = [check_img_size(x, gs) for x in imgsz] # verify img_size are gs-multiples + im = torch.zeros(batch_size, 3, *imgsz).to(device) # image size(1,3,320,192) BCHW iDetection + + # Update model + model.train() if train else model.eval() # training mode = no Detect() layer grid construction + for k, m in model.named_modules(): + if isinstance(m, Detect): + m.inplace = inplace + m.onnx_dynamic = dynamic + m.export = True + + for _ in range(2): + y = model(im) # dry runs + if half and not coreml: + im, model = im.half(), model.half() # to FP16 + shape = tuple((y[0] if isinstance(y, tuple) else y).shape) # model output shape + LOGGER.info(f"\n{colorstr('PyTorch:')} starting from {file} with output shape {shape} ({file_size(file):.1f} MB)") + + # Exports + f = [''] * 10 # exported filenames + warnings.filterwarnings(action='ignore', category=torch.jit.TracerWarning) # suppress TracerWarning + if jit: + f[0] = export_torchscript(model, im, file, optimize) + if engine: # TensorRT required before ONNX + f[1] = export_engine(model, im, file, train, half, dynamic, simplify, workspace, verbose) + if onnx or xml: # OpenVINO requires ONNX + f[2] = export_onnx(model, im, file, opset, train, dynamic, simplify) + if xml: # OpenVINO + f[3] = export_openvino(model, file, half) + if coreml: + _, f[4] = export_coreml(model, im, file, int8, half) + + # TensorFlow Exports + if any((saved_model, pb, tflite, edgetpu, tfjs)): + if int8 or edgetpu: # TFLite --int8 bug https://github.com/ultralytics/yolov5/issues/5707 + check_requirements(('flatbuffers==1.12',)) # required before `import tensorflow` + assert not tflite or not tfjs, 'TFLite and TF.js models must be exported separately, please pass only one type.' + model, f[5] = export_saved_model(model.cpu(), + im, + file, + dynamic, + tf_nms=nms or agnostic_nms or tfjs, + agnostic_nms=agnostic_nms or tfjs, + topk_per_class=topk_per_class, + topk_all=topk_all, + iou_thres=iou_thres, + conf_thres=conf_thres, + keras=keras) + if pb or tfjs: # pb prerequisite to tfjs + f[6] = export_pb(model, file) + if tflite or edgetpu: + f[7] = export_tflite(model, im, file, int8=int8 or edgetpu, data=data, nms=nms, agnostic_nms=agnostic_nms) + if edgetpu: + f[8] = export_edgetpu(file) + if tfjs: + f[9] = export_tfjs(file) + + # Finish + f = [str(x) for x in f if x] # filter out '' and None + if any(f): + h = '--half' if half else '' # --half FP16 inference arg + LOGGER.info(f'\nExport complete ({time.time() - t:.2f}s)' + f"\nResults saved to {colorstr('bold', file.parent.resolve())}" + f"\nDetect: python detect.py --weights {f[-1]} {h}" + f"\nValidate: python val.py --weights {f[-1]} {h}" + f"\nPyTorch Hub: model = torch.hub.load('ultralytics/yolov5', 'custom', '{f[-1]}')" + f"\nVisualize: https://netron.app") + return f # return list of exported files/dirs + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='dataset.yaml path') + parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5s.pt', help='model.pt path(s)') + parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640, 640], help='image (h, w)') + parser.add_argument('--batch-size', type=int, default=1, help='batch size') + parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--half', action='store_true', help='FP16 half-precision export') + parser.add_argument('--inplace', action='store_true', help='set YOLOv5 Detect() inplace=True') + parser.add_argument('--train', action='store_true', help='model.train() mode') + parser.add_argument('--keras', action='store_true', help='TF: use Keras') + parser.add_argument('--optimize', action='store_true', help='TorchScript: optimize for mobile') + parser.add_argument('--int8', action='store_true', help='CoreML/TF INT8 quantization') + parser.add_argument('--dynamic', action='store_true', help='ONNX/TF/TensorRT: dynamic axes') + parser.add_argument('--simplify', action='store_true', help='ONNX: simplify model') + parser.add_argument('--opset', type=int, default=12, help='ONNX: opset version') + parser.add_argument('--verbose', action='store_true', help='TensorRT: verbose log') + parser.add_argument('--workspace', type=int, default=4, help='TensorRT: workspace size (GB)') + parser.add_argument('--nms', action='store_true', help='TF: add NMS to model') + parser.add_argument('--agnostic-nms', action='store_true', help='TF: add agnostic NMS to model') + parser.add_argument('--topk-per-class', type=int, default=100, help='TF.js NMS: topk per class to keep') + parser.add_argument('--topk-all', type=int, default=100, help='TF.js NMS: topk for all classes to keep') + parser.add_argument('--iou-thres', type=float, default=0.45, help='TF.js NMS: IoU threshold') + parser.add_argument('--conf-thres', type=float, default=0.25, help='TF.js NMS: confidence threshold') + parser.add_argument('--include', + nargs='+', + default=['torchscript', 'onnx'], + help='torchscript, onnx, openvino, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs') + opt = parser.parse_args() + print_args(vars(opt)) + return opt + + +def main(opt): + for opt.weights in (opt.weights if isinstance(opt.weights, list) else [opt.weights]): + run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/hubconf.py b/application/yolov5_example/hubconf.py new file mode 100644 index 00000000..011eaa57 --- /dev/null +++ b/application/yolov5_example/hubconf.py @@ -0,0 +1,160 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +PyTorch Hub models https://pytorch.org/hub/ultralytics_yolov5/ + +Usage: + import torch + model = torch.hub.load('ultralytics/yolov5', 'yolov5s') + model = torch.hub.load('ultralytics/yolov5:master', 'custom', 'path/to/yolov5s.onnx') # file from branch +""" + +import torch + + +def _create(name, pretrained=True, channels=3, classes=80, autoshape=True, verbose=True, device=None): + """Creates or loads a YOLOv5 model + + Arguments: + name (str): model name 'yolov5s' or path 'path/to/best.pt' + pretrained (bool): load pretrained weights into the model + channels (int): number of input channels + classes (int): number of model classes + autoshape (bool): apply YOLOv5 .autoshape() wrapper to model + verbose (bool): print all information to screen + device (str, torch.device, None): device to use for model parameters + + Returns: + YOLOv5 model + """ + from pathlib import Path + + from models.common import AutoShape, DetectMultiBackend + from models.experimental import attempt_load + from models.yolo import Model + from utils.downloads import attempt_download + from utils.general import LOGGER, check_requirements, intersect_dicts, logging + from utils.torch_utils import select_device + + if not verbose: + LOGGER.setLevel(logging.WARNING) + check_requirements(exclude=('tensorboard', 'thop', 'opencv-python')) + name = Path(name) + path = name.with_suffix('.pt') if name.suffix == '' and not name.is_dir() else name # checkpoint path + try: + device = select_device(device) + if pretrained and channels == 3 and classes == 80: + try: + model = DetectMultiBackend(path, device=device, fuse=autoshape) # detection model + if autoshape: + model = AutoShape(model) # for file/URI/PIL/cv2/np inputs and NMS + except Exception: + model = attempt_load(path, device=device, fuse=False) # arbitrary model + else: + cfg = list((Path(__file__).parent / 'models').rglob(f'{path.stem}.yaml'))[0] # model.yaml path + model = Model(cfg, channels, classes) # create model + if pretrained: + ckpt = torch.load(attempt_download(path), map_location=device) # load + csd = ckpt['model'].float().state_dict() # checkpoint state_dict as FP32 + csd = intersect_dicts(csd, model.state_dict(), exclude=['anchors']) # intersect + model.load_state_dict(csd, strict=False) # load + if len(ckpt['model'].names) == classes: + model.names = ckpt['model'].names # set class names attribute + if not verbose: + LOGGER.setLevel(logging.INFO) # reset to default + return model.to(device) + + except Exception as e: + help_url = 'https://github.com/ultralytics/yolov5/issues/36' + s = f'{e}. Cache may be out of date, try `force_reload=True` or see {help_url} for help.' + raise Exception(s) from e + + +def custom(path='path/to/model.pt', autoshape=True, _verbose=True, device=None): + # YOLOv5 custom or local model + return _create(path, autoshape=autoshape, verbose=_verbose, device=device) + + +def yolov5n(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-nano model https://github.com/ultralytics/yolov5 + return _create('yolov5n', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5s(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-small model https://github.com/ultralytics/yolov5 + return _create('yolov5s', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5m(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-medium model https://github.com/ultralytics/yolov5 + return _create('yolov5m', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5l(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-large model https://github.com/ultralytics/yolov5 + return _create('yolov5l', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5x(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-xlarge model https://github.com/ultralytics/yolov5 + return _create('yolov5x', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5n6(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-nano-P6 model https://github.com/ultralytics/yolov5 + return _create('yolov5n6', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5s6(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-small-P6 model https://github.com/ultralytics/yolov5 + return _create('yolov5s6', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5m6(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-medium-P6 model https://github.com/ultralytics/yolov5 + return _create('yolov5m6', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5l6(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-large-P6 model https://github.com/ultralytics/yolov5 + return _create('yolov5l6', pretrained, channels, classes, autoshape, _verbose, device) + + +def yolov5x6(pretrained=True, channels=3, classes=80, autoshape=True, _verbose=True, device=None): + # YOLOv5-xlarge-P6 model https://github.com/ultralytics/yolov5 + return _create('yolov5x6', pretrained, channels, classes, autoshape, _verbose, device) + + +if __name__ == '__main__': + import argparse + from pathlib import Path + + import numpy as np + from PIL import Image + + from utils.general import cv2, print_args + + # Argparser + parser = argparse.ArgumentParser() + parser.add_argument('--model', type=str, default='yolov5s', help='model name') + opt = parser.parse_args() + print_args(vars(opt)) + + # Model + model = _create(name=opt.model, pretrained=True, channels=3, classes=80, autoshape=True, verbose=True) + # model = custom(path='path/to/model.pt') # custom + + # Images + imgs = [ + 'data/images/zidane.jpg', # filename + Path('data/images/zidane.jpg'), # Path + 'https://ultralytics.com/images/zidane.jpg', # URI + cv2.imread('data/images/bus.jpg')[:, :, ::-1], # OpenCV + Image.open('data/images/bus.jpg'), # PIL + np.zeros((320, 640, 3))] # numpy + + # Inference + results = model(imgs, size=320) # batched inference + + # Results + results.print() + results.save() diff --git a/application/yolov5_example/models/__init__.py b/application/yolov5_example/models/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/application/yolov5_example/models/common.py b/application/yolov5_example/models/common.py new file mode 100644 index 00000000..17e40e60 --- /dev/null +++ b/application/yolov5_example/models/common.py @@ -0,0 +1,771 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Common modules +""" + +import json +import math +import platform +import warnings +from collections import OrderedDict, namedtuple +from copy import copy +from pathlib import Path + +import cv2 +import numpy as np +import pandas as pd +import requests +import torch +import torch.nn as nn +from PIL import Image +from torch.cuda import amp + +from utils.dataloaders import exif_transpose, letterbox +from utils.general import (LOGGER, ROOT, check_requirements, check_suffix, check_version, colorstr, increment_path, + make_divisible, non_max_suppression, scale_coords, xywh2xyxy, xyxy2xywh, yaml_load) +from utils.plots import Annotator, colors, save_one_box +from utils.torch_utils import copy_attr, smart_inference_mode, time_sync + + +def autopad(k, p=None): # kernel, padding + # Pad to 'same' + if p is None: + p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad + return p + + +class Conv(nn.Module): + # Standard convolution + def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups + super().__init__() + self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False) + self.bn = nn.BatchNorm2d(c2) + self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity()) + + def forward(self, x): + return self.act(self.bn(self.conv(x))) + + def forward_fuse(self, x): + return self.act(self.conv(x)) + + +class DWConv(Conv): + # Depth-wise convolution class + def __init__(self, c1, c2, k=1, s=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups + super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), act=act) + + +class DWConvTranspose2d(nn.ConvTranspose2d): + # Depth-wise transpose convolution class + def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0): # ch_in, ch_out, kernel, stride, padding, padding_out + super().__init__(c1, c2, k, s, p1, p2, groups=math.gcd(c1, c2)) + + +class TransformerLayer(nn.Module): + # Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance) + def __init__(self, c, num_heads): + super().__init__() + self.q = nn.Linear(c, c, bias=False) + self.k = nn.Linear(c, c, bias=False) + self.v = nn.Linear(c, c, bias=False) + self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads) + self.fc1 = nn.Linear(c, c, bias=False) + self.fc2 = nn.Linear(c, c, bias=False) + + def forward(self, x): + x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x + x = self.fc2(self.fc1(x)) + x + return x + + +class TransformerBlock(nn.Module): + # Vision Transformer https://arxiv.org/abs/2010.11929 + def __init__(self, c1, c2, num_heads, num_layers): + super().__init__() + self.conv = None + if c1 != c2: + self.conv = Conv(c1, c2) + self.linear = nn.Linear(c2, c2) # learnable position embedding + self.tr = nn.Sequential(*(TransformerLayer(c2, num_heads) for _ in range(num_layers))) + self.c2 = c2 + + def forward(self, x): + if self.conv is not None: + x = self.conv(x) + b, _, w, h = x.shape + p = x.flatten(2).permute(2, 0, 1) + return self.tr(p + self.linear(p)).permute(1, 2, 0).reshape(b, self.c2, w, h) + + +class Bottleneck(nn.Module): + # Standard bottleneck + def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = Conv(c1, c_, 1, 1) + self.cv2 = Conv(c_, c2, 3, 1, g=g) + self.add = shortcut and c1 == c2 + + def forward(self, x): + return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) + + +class BottleneckCSP(nn.Module): + # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = Conv(c1, c_, 1, 1) + self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False) + self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False) + self.cv4 = Conv(2 * c_, c2, 1, 1) + self.bn = nn.BatchNorm2d(2 * c_) # applied to cat(cv2, cv3) + self.act = nn.SiLU() + self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n))) + + def forward(self, x): + y1 = self.cv3(self.m(self.cv1(x))) + y2 = self.cv2(x) + return self.cv4(self.act(self.bn(torch.cat((y1, y2), 1)))) + + +class CrossConv(nn.Module): + # Cross Convolution Downsample + def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False): + # ch_in, ch_out, kernel, stride, groups, expansion, shortcut + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = Conv(c1, c_, (1, k), (1, s)) + self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g) + self.add = shortcut and c1 == c2 + + def forward(self, x): + return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) + + +class C3(nn.Module): + # CSP Bottleneck with 3 convolutions + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = Conv(c1, c_, 1, 1) + self.cv2 = Conv(c1, c_, 1, 1) + self.cv3 = Conv(2 * c_, c2, 1) # optional act=FReLU(c2) + self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n))) + + def forward(self, x): + return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), 1)) + + +class C3x(C3): + # C3 module with cross-convolutions + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): + super().__init__(c1, c2, n, shortcut, g, e) + c_ = int(c2 * e) + self.m = nn.Sequential(*(CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n))) + + +class C3TR(C3): + # C3 module with TransformerBlock() + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): + super().__init__(c1, c2, n, shortcut, g, e) + c_ = int(c2 * e) + self.m = TransformerBlock(c_, c_, 4, n) + + +class C3SPP(C3): + # C3 module with SPP() + def __init__(self, c1, c2, k=(5, 9, 13), n=1, shortcut=True, g=1, e=0.5): + super().__init__(c1, c2, n, shortcut, g, e) + c_ = int(c2 * e) + self.m = SPP(c_, c_, k) + + +class C3Ghost(C3): + # C3 module with GhostBottleneck() + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): + super().__init__(c1, c2, n, shortcut, g, e) + c_ = int(c2 * e) # hidden channels + self.m = nn.Sequential(*(GhostBottleneck(c_, c_) for _ in range(n))) + + +class SPP(nn.Module): + # Spatial Pyramid Pooling (SPP) layer https://arxiv.org/abs/1406.4729 + def __init__(self, c1, c2, k=(5, 9, 13)): + super().__init__() + c_ = c1 // 2 # hidden channels + self.cv1 = Conv(c1, c_, 1, 1) + self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1) + self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k]) + + def forward(self, x): + x = self.cv1(x) + with warnings.catch_warnings(): + warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning + return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1)) + + +class SPPF(nn.Module): + # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher + def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13)) + super().__init__() + c_ = c1 // 2 # hidden channels + self.cv1 = Conv(c1, c_, 1, 1) + self.cv2 = Conv(c_ * 4, c2, 1, 1) + self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2) + + def forward(self, x): + x = self.cv1(x) + with warnings.catch_warnings(): + warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning + y1 = self.m(x) + y2 = self.m(y1) + return self.cv2(torch.cat((x, y1, y2, self.m(y2)), 1)) + + +class Focus(nn.Module): + # Focus wh information into c-space + def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups + super().__init__() + self.conv = Conv(c1 * 4, c2, k, s, p, g, act) + # self.contract = Contract(gain=2) + + def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2) + return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1)) + # return self.conv(self.contract(x)) + + +class GhostConv(nn.Module): + # Ghost Convolution https://github.com/huawei-noah/ghostnet + def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups + super().__init__() + c_ = c2 // 2 # hidden channels + self.cv1 = Conv(c1, c_, k, s, None, g, act) + self.cv2 = Conv(c_, c_, 5, 1, None, c_, act) + + def forward(self, x): + y = self.cv1(x) + return torch.cat((y, self.cv2(y)), 1) + + +class GhostBottleneck(nn.Module): + # Ghost Bottleneck https://github.com/huawei-noah/ghostnet + def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride + super().__init__() + c_ = c2 // 2 + self.conv = nn.Sequential( + GhostConv(c1, c_, 1, 1), # pw + DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw + GhostConv(c_, c2, 1, 1, act=False)) # pw-linear + self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), Conv(c1, c2, 1, 1, + act=False)) if s == 2 else nn.Identity() + + def forward(self, x): + return self.conv(x) + self.shortcut(x) + + +class Contract(nn.Module): + # Contract width-height into channels, i.e. x(1,64,80,80) to x(1,256,40,40) + def __init__(self, gain=2): + super().__init__() + self.gain = gain + + def forward(self, x): + b, c, h, w = x.size() # assert (h / s == 0) and (W / s == 0), 'Indivisible gain' + s = self.gain + x = x.view(b, c, h // s, s, w // s, s) # x(1,64,40,2,40,2) + x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # x(1,2,2,64,40,40) + return x.view(b, c * s * s, h // s, w // s) # x(1,256,40,40) + + +class Expand(nn.Module): + # Expand channels into width-height, i.e. x(1,64,80,80) to x(1,16,160,160) + def __init__(self, gain=2): + super().__init__() + self.gain = gain + + def forward(self, x): + b, c, h, w = x.size() # assert C / s ** 2 == 0, 'Indivisible gain' + s = self.gain + x = x.view(b, s, s, c // s ** 2, h, w) # x(1,2,2,16,80,80) + x = x.permute(0, 3, 4, 1, 5, 2).contiguous() # x(1,16,80,2,80,2) + return x.view(b, c // s ** 2, h * s, w * s) # x(1,16,160,160) + + +class Concat(nn.Module): + # Concatenate a list of tensors along dimension + def __init__(self, dimension=1): + super().__init__() + self.d = dimension + + def forward(self, x): + return torch.cat(x, self.d) + + +class DetectMultiBackend(nn.Module): + # YOLOv5 MultiBackend class for python inference on various backends + def __init__(self, weights='yolov5s.pt', device=torch.device('cpu'), dnn=False, data=None, fp16=False, fuse=True): + # Usage: + # PyTorch: weights = *.pt + # TorchScript: *.torchscript + # ONNX Runtime: *.onnx + # ONNX OpenCV DNN: *.onnx with --dnn + # OpenVINO: *.xml + # CoreML: *.mlmodel + # TensorRT: *.engine + # TensorFlow SavedModel: *_saved_model + # TensorFlow GraphDef: *.pb + # TensorFlow Lite: *.tflite + # TensorFlow Edge TPU: *_edgetpu.tflite + from models.experimental import attempt_download, attempt_load # scoped to avoid circular import + + super().__init__() + w = str(weights[0] if isinstance(weights, list) else weights) + pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs = self._model_type(w) # get backend + w = attempt_download(w) # download if not local + fp16 &= pt or jit or onnx or engine # FP16 + stride = 32 # default stride + + if pt: # PyTorch + model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse) + stride = max(int(model.stride.max()), 32) # model stride + names = model.module.names if hasattr(model, 'module') else model.names # get class names + model.half() if fp16 else model.float() + self.model = model # explicitly assign for to(), cpu(), cuda(), half() + elif jit: # TorchScript + LOGGER.info(f'Loading {w} for TorchScript inference...') + extra_files = {'config.txt': ''} # model metadata + model = torch.jit.load(w, _extra_files=extra_files) + model.half() if fp16 else model.float() + if extra_files['config.txt']: + d = json.loads(extra_files['config.txt']) # extra_files dict + stride, names = int(d['stride']), d['names'] + elif dnn: # ONNX OpenCV DNN + LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...') + check_requirements(('opencv-python>=4.5.4',)) + net = cv2.dnn.readNetFromONNX(w) + elif onnx: # ONNX Runtime + LOGGER.info(f'Loading {w} for ONNX Runtime inference...') + cuda = torch.cuda.is_available() and device.type != 'cpu' + check_requirements(('onnx', 'onnxruntime-gpu' if cuda else 'onnxruntime')) + import onnxruntime + providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'] if cuda else ['CPUExecutionProvider'] + session = onnxruntime.InferenceSession(w, providers=providers) + meta = session.get_modelmeta().custom_metadata_map # metadata + if 'stride' in meta: + stride, names = int(meta['stride']), eval(meta['names']) + elif xml: # OpenVINO + LOGGER.info(f'Loading {w} for OpenVINO inference...') + check_requirements(('openvino',)) # requires openvino-dev: https://pypi.org/project/openvino-dev/ + from openvino.runtime import Core, Layout, get_batch + ie = Core() + if not Path(w).is_file(): # if not *.xml + w = next(Path(w).glob('*.xml')) # get *.xml file from *_openvino_model dir + network = ie.read_model(model=w, weights=Path(w).with_suffix('.bin')) + if network.get_parameters()[0].get_layout().empty: + network.get_parameters()[0].set_layout(Layout("NCHW")) + batch_dim = get_batch(network) + if batch_dim.is_static: + batch_size = batch_dim.get_length() + executable_network = ie.compile_model(network, device_name="CPU") # device_name="MYRIAD" for Intel NCS2 + output_layer = next(iter(executable_network.outputs)) + meta = Path(w).with_suffix('.yaml') + if meta.exists(): + stride, names = self._load_metadata(meta) # load metadata + elif engine: # TensorRT + LOGGER.info(f'Loading {w} for TensorRT inference...') + import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download + check_version(trt.__version__, '7.0.0', hard=True) # require tensorrt>=7.0.0 + if device.type == 'cpu': + device = torch.device('cuda:0') + Binding = namedtuple('Binding', ('name', 'dtype', 'shape', 'data', 'ptr')) + logger = trt.Logger(trt.Logger.INFO) + with open(w, 'rb') as f, trt.Runtime(logger) as runtime: + model = runtime.deserialize_cuda_engine(f.read()) + context = model.create_execution_context() + bindings = OrderedDict() + fp16 = False # default updated below + dynamic = False + for index in range(model.num_bindings): + name = model.get_binding_name(index) + dtype = trt.nptype(model.get_binding_dtype(index)) + if model.binding_is_input(index): + if -1 in tuple(model.get_binding_shape(index)): # dynamic + dynamic = True + context.set_binding_shape(index, tuple(model.get_profile_shape(0, index)[2])) + if dtype == np.float16: + fp16 = True + shape = tuple(context.get_binding_shape(index)) + im = torch.from_numpy(np.empty(shape, dtype=dtype)).to(device) + bindings[name] = Binding(name, dtype, shape, im, int(im.data_ptr())) + binding_addrs = OrderedDict((n, d.ptr) for n, d in bindings.items()) + batch_size = bindings['images'].shape[0] # if dynamic, this is instead max batch size + elif coreml: # CoreML + LOGGER.info(f'Loading {w} for CoreML inference...') + import coremltools as ct + model = ct.models.MLModel(w) + else: # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU) + if saved_model: # SavedModel + LOGGER.info(f'Loading {w} for TensorFlow SavedModel inference...') + import tensorflow as tf + keras = False # assume TF1 saved_model + model = tf.keras.models.load_model(w) if keras else tf.saved_model.load(w) + elif pb: # GraphDef https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt + LOGGER.info(f'Loading {w} for TensorFlow GraphDef inference...') + import tensorflow as tf + + def wrap_frozen_graph(gd, inputs, outputs): + x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=""), []) # wrapped + ge = x.graph.as_graph_element + return x.prune(tf.nest.map_structure(ge, inputs), tf.nest.map_structure(ge, outputs)) + + gd = tf.Graph().as_graph_def() # graph_def + with open(w, 'rb') as f: + gd.ParseFromString(f.read()) + frozen_func = wrap_frozen_graph(gd, inputs="x:0", outputs="Identity:0") + elif tflite or edgetpu: # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python + try: # https://coral.ai/docs/edgetpu/tflite-python/#update-existing-tf-lite-code-for-the-edge-tpu + from tflite_runtime.interpreter import Interpreter, load_delegate + except ImportError: + import tensorflow as tf + Interpreter, load_delegate = tf.lite.Interpreter, tf.lite.experimental.load_delegate, + if edgetpu: # Edge TPU https://coral.ai/software/#edgetpu-runtime + LOGGER.info(f'Loading {w} for TensorFlow Lite Edge TPU inference...') + delegate = { + 'Linux': 'libedgetpu.so.1', + 'Darwin': 'libedgetpu.1.dylib', + 'Windows': 'edgetpu.dll'}[platform.system()] + interpreter = Interpreter(model_path=w, experimental_delegates=[load_delegate(delegate)]) + else: # Lite + LOGGER.info(f'Loading {w} for TensorFlow Lite inference...') + interpreter = Interpreter(model_path=w) # load TFLite model + interpreter.allocate_tensors() # allocate + input_details = interpreter.get_input_details() # inputs + output_details = interpreter.get_output_details() # outputs + elif tfjs: + raise NotImplementedError('ERROR: YOLOv5 TF.js inference is not supported') + else: + raise NotImplementedError(f'ERROR: {w} is not a supported format') + + # class names + if 'names' not in locals(): + names = yaml_load(data)['names'] if data else [f'class{i}' for i in range(999)] + if names[0] == 'n01440764' and len(names) == 1000: # ImageNet + names = yaml_load(ROOT / 'data/ImageNet.yaml')['names'] # human-readable names + + self.__dict__.update(locals()) # assign all variables to self + + def forward(self, im, augment=False, visualize=False, val=False): + # YOLOv5 MultiBackend inference + b, ch, h, w = im.shape # batch, channel, height, width + if self.fp16 and im.dtype != torch.float16: + im = im.half() # to FP16 + + if self.pt: # PyTorch + y = self.model(im, augment=augment, visualize=visualize) if augment or visualize else self.model(im) + if isinstance(y, tuple): + y = y[0] + elif self.jit: # TorchScript + y = self.model(im)[0] + elif self.dnn: # ONNX OpenCV DNN + im = im.cpu().numpy() # torch to numpy + self.net.setInput(im) + y = self.net.forward() + elif self.onnx: # ONNX Runtime + im = im.cpu().numpy() # torch to numpy + y = self.session.run([self.session.get_outputs()[0].name], {self.session.get_inputs()[0].name: im})[0] + elif self.xml: # OpenVINO + im = im.cpu().numpy() # FP32 + y = self.executable_network([im])[self.output_layer] + elif self.engine: # TensorRT + if self.dynamic and im.shape != self.bindings['images'].shape: + i_in, i_out = (self.model.get_binding_index(x) for x in ('images', 'output')) + self.context.set_binding_shape(i_in, im.shape) # reshape if dynamic + self.bindings['images'] = self.bindings['images']._replace(shape=im.shape) + self.bindings['output'].data.resize_(tuple(self.context.get_binding_shape(i_out))) + s = self.bindings['images'].shape + assert im.shape == s, f"input size {im.shape} {'>' if self.dynamic else 'not equal to'} max model size {s}" + self.binding_addrs['images'] = int(im.data_ptr()) + self.context.execute_v2(list(self.binding_addrs.values())) + y = self.bindings['output'].data + elif self.coreml: # CoreML + im = im.permute(0, 2, 3, 1).cpu().numpy() # torch BCHW to numpy BHWC shape(1,320,192,3) + im = Image.fromarray((im[0] * 255).astype('uint8')) + # im = im.resize((192, 320), Image.ANTIALIAS) + y = self.model.predict({'image': im}) # coordinates are xywh normalized + if 'confidence' in y: + box = xywh2xyxy(y['coordinates'] * [[w, h, w, h]]) # xyxy pixels + conf, cls = y['confidence'].max(1), y['confidence'].argmax(1).astype(np.float) + y = np.concatenate((box, conf.reshape(-1, 1), cls.reshape(-1, 1)), 1) + else: + k = 'var_' + str(sorted(int(k.replace('var_', '')) for k in y)[-1]) # output key + y = y[k] # output + else: # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU) + im = im.permute(0, 2, 3, 1).cpu().numpy() # torch BCHW to numpy BHWC shape(1,320,192,3) + if self.saved_model: # SavedModel + y = (self.model(im, training=False) if self.keras else self.model(im)).numpy() + elif self.pb: # GraphDef + y = self.frozen_func(x=self.tf.constant(im)).numpy() + else: # Lite or Edge TPU + input, output = self.input_details[0], self.output_details[0] + int8 = input['dtype'] == np.uint8 # is TFLite quantized uint8 model + if int8: + scale, zero_point = input['quantization'] + im = (im / scale + zero_point).astype(np.uint8) # de-scale + self.interpreter.set_tensor(input['index'], im) + self.interpreter.invoke() + y = self.interpreter.get_tensor(output['index']) + if int8: + scale, zero_point = output['quantization'] + y = (y.astype(np.float32) - zero_point) * scale # re-scale + y[..., :4] *= [w, h, w, h] # xywh normalized to pixels + + if isinstance(y, np.ndarray): + y = torch.tensor(y, device=self.device) + return (y, []) if val else y + + def warmup(self, imgsz=(1, 3, 640, 640)): + # Warmup model by running inference once + warmup_types = self.pt, self.jit, self.onnx, self.engine, self.saved_model, self.pb + if any(warmup_types) and self.device.type != 'cpu': + im = torch.zeros(*imgsz, dtype=torch.half if self.fp16 else torch.float, device=self.device) # input + for _ in range(2 if self.jit else 1): # + self.forward(im) # warmup + + @staticmethod + def _model_type(p='path/to/model.pt'): + # Return model type from model path, i.e. path='path/to/model.onnx' -> type=onnx + from export import export_formats + suffixes = list(export_formats().Suffix) + ['.xml'] # export suffixes + check_suffix(p, suffixes) # checks + p = Path(p).name # eliminate trailing separators + pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, xml2 = (s in p for s in suffixes) + xml |= xml2 # *_openvino_model or *.xml + tflite &= not edgetpu # *.tflite + return pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs + + @staticmethod + def _load_metadata(f='path/to/meta.yaml'): + # Load metadata from meta.yaml if it exists + d = yaml_load(f) + return d['stride'], d['names'] # assign stride, names + + +class AutoShape(nn.Module): + # YOLOv5 input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS + conf = 0.25 # NMS confidence threshold + iou = 0.45 # NMS IoU threshold + agnostic = False # NMS class-agnostic + multi_label = False # NMS multiple labels per box + classes = None # (optional list) filter by class, i.e. = [0, 15, 16] for COCO persons, cats and dogs + max_det = 1000 # maximum number of detections per image + amp = False # Automatic Mixed Precision (AMP) inference + + def __init__(self, model, verbose=True): + super().__init__() + if verbose: + LOGGER.info('Adding AutoShape... ') + copy_attr(self, model, include=('yaml', 'nc', 'hyp', 'names', 'stride', 'abc'), exclude=()) # copy attributes + self.dmb = isinstance(model, DetectMultiBackend) # DetectMultiBackend() instance + self.pt = not self.dmb or model.pt # PyTorch model + self.model = model.eval() + if self.pt: + m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect() + m.inplace = False # Detect.inplace=False for safe multithread inference + + def _apply(self, fn): + # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers + self = super()._apply(fn) + if self.pt: + m = self.model.model.model[-1] if self.dmb else self.model.model[-1] # Detect() + m.stride = fn(m.stride) + m.grid = list(map(fn, m.grid)) + if isinstance(m.anchor_grid, list): + m.anchor_grid = list(map(fn, m.anchor_grid)) + return self + + @smart_inference_mode() + def forward(self, imgs, size=640, augment=False, profile=False): + # Inference from various sources. For height=640, width=1280, RGB images example inputs are: + # file: imgs = 'data/images/zidane.jpg' # str or PosixPath + # URI: = 'https://ultralytics.com/images/zidane.jpg' + # OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3) + # PIL: = Image.open('image.jpg') or ImageGrab.grab() # HWC x(640,1280,3) + # numpy: = np.zeros((640,1280,3)) # HWC + # torch: = torch.zeros(16,3,320,640) # BCHW (scaled to size=640, 0-1 values) + # multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images + + t = [time_sync()] + p = next(self.model.parameters()) if self.pt else torch.zeros(1, device=self.model.device) # for device, type + autocast = self.amp and (p.device.type != 'cpu') # Automatic Mixed Precision (AMP) inference + if isinstance(imgs, torch.Tensor): # torch + with amp.autocast(autocast): + return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference + + # Pre-process + n, imgs = (len(imgs), list(imgs)) if isinstance(imgs, (list, tuple)) else (1, [imgs]) # number, list of images + shape0, shape1, files = [], [], [] # image and inference shapes, filenames + for i, im in enumerate(imgs): + f = f'image{i}' # filename + if isinstance(im, (str, Path)): # filename or uri + im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im + im = np.asarray(exif_transpose(im)) + elif isinstance(im, Image.Image): # PIL Image + im, f = np.asarray(exif_transpose(im)), getattr(im, 'filename', f) or f + files.append(Path(f).with_suffix('.jpg').name) + if im.shape[0] < 5: # image in CHW + im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1) + im = im[..., :3] if im.ndim == 3 else np.tile(im[..., None], 3) # enforce 3ch input + s = im.shape[:2] # HWC + shape0.append(s) # image shape + g = (size / max(s)) # gain + shape1.append([y * g for y in s]) + imgs[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update + shape1 = [make_divisible(x, self.stride) if self.pt else size for x in np.array(shape1).max(0)] # inf shape + x = [letterbox(im, shape1, auto=False)[0] for im in imgs] # pad + x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2))) # stack and BHWC to BCHW + x = torch.from_numpy(x).to(p.device).type_as(p) / 255 # uint8 to fp16/32 + t.append(time_sync()) + + with amp.autocast(autocast): + # Inference + y = self.model(x, augment, profile) # forward + t.append(time_sync()) + + # Post-process + y = non_max_suppression(y if self.dmb else y[0], + self.conf, + self.iou, + self.classes, + self.agnostic, + self.multi_label, + max_det=self.max_det) # NMS + for i in range(n): + scale_coords(shape1, y[i][:, :4], shape0[i]) + + t.append(time_sync()) + return Detections(imgs, y, files, t, self.names, x.shape) + + +class Detections: + # YOLOv5 detections class for inference results + def __init__(self, imgs, pred, files, times=(0, 0, 0, 0), names=None, shape=None): + super().__init__() + d = pred[0].device # device + gn = [torch.tensor([*(im.shape[i] for i in [1, 0, 1, 0]), 1, 1], device=d) for im in imgs] # normalizations + self.imgs = imgs # list of images as numpy arrays + self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls) + self.names = names # class names + self.files = files # image filenames + self.times = times # profiling times + self.xyxy = pred # xyxy pixels + self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels + self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized + self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized + self.n = len(self.pred) # number of images (batch size) + self.t = tuple((times[i + 1] - times[i]) * 1000 / self.n for i in range(3)) # timestamps (ms) + self.s = shape # inference BCHW shape + + def display(self, pprint=False, show=False, save=False, crop=False, render=False, labels=True, save_dir=Path('')): + crops = [] + for i, (im, pred) in enumerate(zip(self.imgs, self.pred)): + s = f'image {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} ' # string + if pred.shape[0]: + for c in pred[:, -1].unique(): + n = (pred[:, -1] == c).sum() # detections per class + s += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string + if show or save or render or crop: + annotator = Annotator(im, example=str(self.names)) + for *box, conf, cls in reversed(pred): # xyxy, confidence, class + label = f'{self.names[int(cls)]} {conf:.2f}' + if crop: + file = save_dir / 'crops' / self.names[int(cls)] / self.files[i] if save else None + crops.append({ + 'box': box, + 'conf': conf, + 'cls': cls, + 'label': label, + 'im': save_one_box(box, im, file=file, save=save)}) + else: # all others + annotator.box_label(box, label if labels else '', color=colors(cls)) + im = annotator.im + else: + s += '(no detections)' + + im = Image.fromarray(im.astype(np.uint8)) if isinstance(im, np.ndarray) else im # from np + if pprint: + print(s.rstrip(', ')) + if show: + im.show(self.files[i]) # show + if save: + f = self.files[i] + im.save(save_dir / f) # save + if i == self.n - 1: + LOGGER.info(f"Saved {self.n} image{'s' * (self.n > 1)} to {colorstr('bold', save_dir)}") + if render: + self.imgs[i] = np.asarray(im) + if crop: + if save: + LOGGER.info(f'Saved results to {save_dir}\n') + return crops + + def print(self): + self.display(pprint=True) # print results + print(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' % self.t) + + def show(self, labels=True): + self.display(show=True, labels=labels) # show results + + def save(self, labels=True, save_dir='runs/detect/exp'): + save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) # increment save_dir + self.display(save=True, labels=labels, save_dir=save_dir) # save results + + def crop(self, save=True, save_dir='runs/detect/exp'): + save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) if save else None + return self.display(crop=True, save=save, save_dir=save_dir) # crop results + + def render(self, labels=True): + self.display(render=True, labels=labels) # render results + return self.imgs + + def pandas(self): + # return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0]) + new = copy(self) # return copy + ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name' # xyxy columns + cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name' # xywh columns + for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]): + a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)] # update + setattr(new, k, [pd.DataFrame(x, columns=c) for x in a]) + return new + + def tolist(self): + # return a list of Detections objects, i.e. 'for result in results.tolist():' + r = range(self.n) # iterable + x = [Detections([self.imgs[i]], [self.pred[i]], [self.files[i]], self.times, self.names, self.s) for i in r] + # for d in x: + # for k in ['imgs', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']: + # setattr(d, k, getattr(d, k)[0]) # pop out of list + return x + + def __len__(self): + return self.n # override len(results) + + def __str__(self): + self.print() # override print(results) + return '' + + +class Classify(nn.Module): + # Classification head, i.e. x(b,c1,20,20) to x(b,c2) + def __init__(self, c1, c2, k=1, s=1, p=None, g=1): # ch_in, ch_out, kernel, stride, padding, groups + super().__init__() + c_ = 1280 # efficientnet_b0 size + self.conv = Conv(c1, c_, k, s, autopad(k, p), g) + self.pool = nn.AdaptiveAvgPool2d(1) # to x(b,c_,1,1) + self.drop = nn.Dropout(p=0.0, inplace=True) + self.linear = nn.Linear(c_, c2) # to x(b,c2) + + def forward(self, x): + if isinstance(x, list): + x = torch.cat(x, 1) + return self.linear(self.drop(self.pool(self.conv(x)).flatten(1))) diff --git a/application/yolov5_example/models/experimental.py b/application/yolov5_example/models/experimental.py new file mode 100644 index 00000000..cb32d01b --- /dev/null +++ b/application/yolov5_example/models/experimental.py @@ -0,0 +1,107 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Experimental modules +""" +import math + +import numpy as np +import torch +import torch.nn as nn + +from models.common import Conv +from utils.downloads import attempt_download + + +class Sum(nn.Module): + # Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070 + def __init__(self, n, weight=False): # n: number of inputs + super().__init__() + self.weight = weight # apply weights boolean + self.iter = range(n - 1) # iter object + if weight: + self.w = nn.Parameter(-torch.arange(1.0, n) / 2, requires_grad=True) # layer weights + + def forward(self, x): + y = x[0] # no weight + if self.weight: + w = torch.sigmoid(self.w) * 2 + for i in self.iter: + y = y + x[i + 1] * w[i] + else: + for i in self.iter: + y = y + x[i + 1] + return y + + +class MixConv2d(nn.Module): + # Mixed Depth-wise Conv https://arxiv.org/abs/1907.09595 + def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True): # ch_in, ch_out, kernel, stride, ch_strategy + super().__init__() + n = len(k) # number of convolutions + if equal_ch: # equal c_ per group + i = torch.linspace(0, n - 1E-6, c2).floor() # c2 indices + c_ = [(i == g).sum() for g in range(n)] # intermediate channels + else: # equal weight.numel() per group + b = [c2] + [0] * n + a = np.eye(n + 1, n, k=-1) + a -= np.roll(a, 1, axis=1) + a *= np.array(k) ** 2 + a[0] = 1 + c_ = np.linalg.lstsq(a, b, rcond=None)[0].round() # solve for equal weight indices, ax = b + + self.m = nn.ModuleList([ + nn.Conv2d(c1, int(c_), k, s, k // 2, groups=math.gcd(c1, int(c_)), bias=False) for k, c_ in zip(k, c_)]) + self.bn = nn.BatchNorm2d(c2) + self.act = nn.SiLU() + + def forward(self, x): + return self.act(self.bn(torch.cat([m(x) for m in self.m], 1))) + + +class Ensemble(nn.ModuleList): + # Ensemble of models + def __init__(self): + super().__init__() + + def forward(self, x, augment=False, profile=False, visualize=False): + y = [module(x, augment, profile, visualize)[0] for module in self] + # y = torch.stack(y).max(0)[0] # max ensemble + # y = torch.stack(y).mean(0) # mean ensemble + y = torch.cat(y, 1) # nms ensemble + return y, None # inference, train output + + +def attempt_load(weights, device=None, inplace=True, fuse=True): + # Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a + from models.yolo import Detect, Model + + model = Ensemble() + for w in weights if isinstance(weights, list) else [weights]: + ckpt = torch.load(attempt_download(w), map_location='cpu') # load + ckpt = (ckpt.get('ema') or ckpt['model']).to(device).float() # FP32 model + if not hasattr(ckpt, 'stride'): + ckpt.stride = torch.tensor([32.]) # compatibility update for ResNet etc. + model.append(ckpt.fuse().eval() if fuse and hasattr(ckpt, 'fuse') else ckpt.eval()) # model in eval mode + + # Compatibility updates + for m in model.modules(): + t = type(m) + if t in (nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU, Detect, Model): + m.inplace = inplace # torch 1.7.0 compatibility + if t is Detect and not isinstance(m.anchor_grid, list): + delattr(m, 'anchor_grid') + setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl) + elif t is nn.Upsample and not hasattr(m, 'recompute_scale_factor'): + m.recompute_scale_factor = None # torch 1.11.0 compatibility + + # Return model + if len(model) == 1: + return model[-1] + + # Return detection ensemble + print(f'Ensemble created with {weights}\n') + for k in 'names', 'nc', 'yaml': + setattr(model, k, getattr(model[0], k)) + model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride # max stride + assert all(model[0].nc == m.nc for m in model), f'Models have different class counts: {[m.nc for m in model]}' + return model diff --git a/application/yolov5_example/models/hub/anchors.yaml b/application/yolov5_example/models/hub/anchors.yaml new file mode 100644 index 00000000..e4d7beb0 --- /dev/null +++ b/application/yolov5_example/models/hub/anchors.yaml @@ -0,0 +1,59 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Default anchors for COCO data + + +# P5 ------------------------------------------------------------------------------------------------------------------- +# P5-640: +anchors_p5_640: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + + +# P6 ------------------------------------------------------------------------------------------------------------------- +# P6-640: thr=0.25: 0.9964 BPR, 5.54 anchors past thr, n=12, img_size=640, metric_all=0.281/0.716-mean/best, past_thr=0.469-mean: 9,11, 21,19, 17,41, 43,32, 39,70, 86,64, 65,131, 134,130, 120,265, 282,180, 247,354, 512,387 +anchors_p6_640: + - [9,11, 21,19, 17,41] # P3/8 + - [43,32, 39,70, 86,64] # P4/16 + - [65,131, 134,130, 120,265] # P5/32 + - [282,180, 247,354, 512,387] # P6/64 + +# P6-1280: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1280, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 19,27, 44,40, 38,94, 96,68, 86,152, 180,137, 140,301, 303,264, 238,542, 436,615, 739,380, 925,792 +anchors_p6_1280: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# P6-1920: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1920, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 28,41, 67,59, 57,141, 144,103, 129,227, 270,205, 209,452, 455,396, 358,812, 653,922, 1109,570, 1387,1187 +anchors_p6_1920: + - [28,41, 67,59, 57,141] # P3/8 + - [144,103, 129,227, 270,205] # P4/16 + - [209,452, 455,396, 358,812] # P5/32 + - [653,922, 1109,570, 1387,1187] # P6/64 + + +# P7 ------------------------------------------------------------------------------------------------------------------- +# P7-640: thr=0.25: 0.9962 BPR, 6.76 anchors past thr, n=15, img_size=640, metric_all=0.275/0.733-mean/best, past_thr=0.466-mean: 11,11, 13,30, 29,20, 30,46, 61,38, 39,92, 78,80, 146,66, 79,163, 149,150, 321,143, 157,303, 257,402, 359,290, 524,372 +anchors_p7_640: + - [11,11, 13,30, 29,20] # P3/8 + - [30,46, 61,38, 39,92] # P4/16 + - [78,80, 146,66, 79,163] # P5/32 + - [149,150, 321,143, 157,303] # P6/64 + - [257,402, 359,290, 524,372] # P7/128 + +# P7-1280: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1280, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 19,22, 54,36, 32,77, 70,83, 138,71, 75,173, 165,159, 148,334, 375,151, 334,317, 251,626, 499,474, 750,326, 534,814, 1079,818 +anchors_p7_1280: + - [19,22, 54,36, 32,77] # P3/8 + - [70,83, 138,71, 75,173] # P4/16 + - [165,159, 148,334, 375,151] # P5/32 + - [334,317, 251,626, 499,474] # P6/64 + - [750,326, 534,814, 1079,818] # P7/128 + +# P7-1920: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1920, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 29,34, 81,55, 47,115, 105,124, 207,107, 113,259, 247,238, 222,500, 563,227, 501,476, 376,939, 749,711, 1126,489, 801,1222, 1618,1227 +anchors_p7_1920: + - [29,34, 81,55, 47,115] # P3/8 + - [105,124, 207,107, 113,259] # P4/16 + - [247,238, 222,500, 563,227] # P5/32 + - [501,476, 376,939, 749,711] # P6/64 + - [1126,489, 801,1222, 1618,1227] # P7/128 diff --git a/application/yolov5_example/models/hub/yolov3-spp.yaml b/application/yolov5_example/models/hub/yolov3-spp.yaml new file mode 100644 index 00000000..c6698215 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov3-spp.yaml @@ -0,0 +1,51 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# darknet53 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [32, 3, 1]], # 0 + [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 + [-1, 1, Bottleneck, [64]], + [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 + [-1, 2, Bottleneck, [128]], + [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 + [-1, 8, Bottleneck, [256]], + [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 + [-1, 8, Bottleneck, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 + [-1, 4, Bottleneck, [1024]], # 10 + ] + +# YOLOv3-SPP head +head: + [[-1, 1, Bottleneck, [1024, False]], + [-1, 1, SPP, [512, [5, 9, 13]]], + [-1, 1, Conv, [1024, 3, 1]], + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) + + [-2, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P4 + [-1, 1, Bottleneck, [512, False]], + [-1, 1, Bottleneck, [512, False]], + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) + + [-2, 1, Conv, [128, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P3 + [-1, 1, Bottleneck, [256, False]], + [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) + + [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov3-tiny.yaml b/application/yolov5_example/models/hub/yolov3-tiny.yaml new file mode 100644 index 00000000..b28b4431 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov3-tiny.yaml @@ -0,0 +1,41 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,14, 23,27, 37,58] # P4/16 + - [81,82, 135,169, 344,319] # P5/32 + +# YOLOv3-tiny backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [16, 3, 1]], # 0 + [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 1-P1/2 + [-1, 1, Conv, [32, 3, 1]], + [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 3-P2/4 + [-1, 1, Conv, [64, 3, 1]], + [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 5-P3/8 + [-1, 1, Conv, [128, 3, 1]], + [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 7-P4/16 + [-1, 1, Conv, [256, 3, 1]], + [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 9-P5/32 + [-1, 1, Conv, [512, 3, 1]], + [-1, 1, nn.ZeroPad2d, [[0, 1, 0, 1]]], # 11 + [-1, 1, nn.MaxPool2d, [2, 1, 0]], # 12 + ] + +# YOLOv3-tiny head +head: + [[-1, 1, Conv, [1024, 3, 1]], + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, Conv, [512, 3, 1]], # 15 (P5/32-large) + + [-2, 1, Conv, [128, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P4 + [-1, 1, Conv, [256, 3, 1]], # 19 (P4/16-medium) + + [[19, 15], 1, Detect, [nc, anchors]], # Detect(P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov3.yaml b/application/yolov5_example/models/hub/yolov3.yaml new file mode 100644 index 00000000..d1ef9129 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov3.yaml @@ -0,0 +1,51 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# darknet53 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [32, 3, 1]], # 0 + [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 + [-1, 1, Bottleneck, [64]], + [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 + [-1, 2, Bottleneck, [128]], + [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 + [-1, 8, Bottleneck, [256]], + [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 + [-1, 8, Bottleneck, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 + [-1, 4, Bottleneck, [1024]], # 10 + ] + +# YOLOv3 head +head: + [[-1, 1, Bottleneck, [1024, False]], + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, Conv, [1024, 3, 1]], + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) + + [-2, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P4 + [-1, 1, Bottleneck, [512, False]], + [-1, 1, Bottleneck, [512, False]], + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) + + [-2, 1, Conv, [128, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P3 + [-1, 1, Bottleneck, [256, False]], + [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) + + [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5-bifpn.yaml b/application/yolov5_example/models/hub/yolov5-bifpn.yaml new file mode 100644 index 00000000..504815f5 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-bifpn.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 BiFPN head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14, 6], 1, Concat, [1]], # cat P4 <--- BiFPN change + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5-fpn.yaml b/application/yolov5_example/models/hub/yolov5-fpn.yaml new file mode 100644 index 00000000..a23e9c6f --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-fpn.yaml @@ -0,0 +1,42 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 FPN head +head: + [[-1, 3, C3, [1024, False]], # 10 (P5/32-large) + + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 1, Conv, [512, 1, 1]], + [-1, 3, C3, [512, False]], # 14 (P4/16-medium) + + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 1, Conv, [256, 1, 1]], + [-1, 3, C3, [256, False]], # 18 (P3/8-small) + + [[18, 14, 10], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5-p2.yaml b/application/yolov5_example/models/hub/yolov5-p2.yaml new file mode 100644 index 00000000..554117dd --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-p2.yaml @@ -0,0 +1,54 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: 3 # AutoAnchor evolves 3 anchors per P output layer + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head with (P2, P3, P4, P5) outputs +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [128, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 2], 1, Concat, [1]], # cat backbone P2 + [-1, 1, C3, [128, False]], # 21 (P2/4-xsmall) + + [-1, 1, Conv, [128, 3, 2]], + [[-1, 18], 1, Concat, [1]], # cat head P3 + [-1, 3, C3, [256, False]], # 24 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 27 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 30 (P5/32-large) + + [[21, 24, 27, 30], 1, Detect, [nc, anchors]], # Detect(P2, P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5-p34.yaml b/application/yolov5_example/models/hub/yolov5-p34.yaml new file mode 100644 index 00000000..dbf0f850 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-p34.yaml @@ -0,0 +1,41 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.50 # layer channel multiple +anchors: 3 # AutoAnchor evolves 3 anchors per P output layer + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [ [ -1, 1, Conv, [ 64, 6, 2, 2 ] ], # 0-P1/2 + [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 + [ -1, 3, C3, [ 128 ] ], + [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 + [ -1, 6, C3, [ 256 ] ], + [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 + [ -1, 9, C3, [ 512 ] ], + [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 7-P5/32 + [ -1, 3, C3, [ 1024 ] ], + [ -1, 1, SPPF, [ 1024, 5 ] ], # 9 + ] + +# YOLOv5 v6.0 head with (P3, P4) outputs +head: + [ [ -1, 1, Conv, [ 512, 1, 1 ] ], + [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], + [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 + [ -1, 3, C3, [ 512, False ] ], # 13 + + [ -1, 1, Conv, [ 256, 1, 1 ] ], + [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], + [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 + [ -1, 3, C3, [ 256, False ] ], # 17 (P3/8-small) + + [ -1, 1, Conv, [ 256, 3, 2 ] ], + [ [ -1, 14 ], 1, Concat, [ 1 ] ], # cat head P4 + [ -1, 3, C3, [ 512, False ] ], # 20 (P4/16-medium) + + [ [ 17, 20 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4) + ] diff --git a/application/yolov5_example/models/hub/yolov5-p6.yaml b/application/yolov5_example/models/hub/yolov5-p6.yaml new file mode 100644 index 00000000..a17202f2 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-p6.yaml @@ -0,0 +1,56 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: 3 # AutoAnchor evolves 3 anchors per P output layer + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head with (P3, P4, P5, P6) outputs +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/hub/yolov5-p7.yaml b/application/yolov5_example/models/hub/yolov5-p7.yaml new file mode 100644 index 00000000..edd7d13a --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-p7.yaml @@ -0,0 +1,67 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: 3 # AutoAnchor evolves 3 anchors per P output layer + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, Conv, [1280, 3, 2]], # 11-P7/128 + [-1, 3, C3, [1280]], + [-1, 1, SPPF, [1280, 5]], # 13 + ] + +# YOLOv5 v6.0 head with (P3, P4, P5, P6, P7) outputs +head: + [[-1, 1, Conv, [1024, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 10], 1, Concat, [1]], # cat backbone P6 + [-1, 3, C3, [1024, False]], # 17 + + [-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 21 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 25 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 29 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 26], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 32 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 22], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 35 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 18], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 38 (P6/64-xlarge) + + [-1, 1, Conv, [1024, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P7 + [-1, 3, C3, [1280, False]], # 41 (P7/128-xxlarge) + + [[29, 32, 35, 38, 41], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6, P7) + ] diff --git a/application/yolov5_example/models/hub/yolov5-panet.yaml b/application/yolov5_example/models/hub/yolov5-panet.yaml new file mode 100644 index 00000000..ccfbf900 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5-panet.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 PANet head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5l6.yaml b/application/yolov5_example/models/hub/yolov5l6.yaml new file mode 100644 index 00000000..632c2cb6 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5l6.yaml @@ -0,0 +1,60 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/hub/yolov5m6.yaml b/application/yolov5_example/models/hub/yolov5m6.yaml new file mode 100644 index 00000000..ecc53fd6 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5m6.yaml @@ -0,0 +1,60 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.67 # model depth multiple +width_multiple: 0.75 # layer channel multiple +anchors: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/hub/yolov5n6.yaml b/application/yolov5_example/models/hub/yolov5n6.yaml new file mode 100644 index 00000000..0c0c71d3 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5n6.yaml @@ -0,0 +1,60 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.25 # layer channel multiple +anchors: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/hub/yolov5s-ghost.yaml b/application/yolov5_example/models/hub/yolov5s-ghost.yaml new file mode 100644 index 00000000..ff9519c3 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5s-ghost.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.50 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, GhostConv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3Ghost, [128]], + [-1, 1, GhostConv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3Ghost, [256]], + [-1, 1, GhostConv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3Ghost, [512]], + [-1, 1, GhostConv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3Ghost, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, GhostConv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3Ghost, [512, False]], # 13 + + [-1, 1, GhostConv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3Ghost, [256, False]], # 17 (P3/8-small) + + [-1, 1, GhostConv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3Ghost, [512, False]], # 20 (P4/16-medium) + + [-1, 1, GhostConv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3Ghost, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5s-transformer.yaml b/application/yolov5_example/models/hub/yolov5s-transformer.yaml new file mode 100644 index 00000000..100d7c44 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5s-transformer.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.50 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3TR, [1024]], # 9 <--- C3TR() Transformer module + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/hub/yolov5s6.yaml b/application/yolov5_example/models/hub/yolov5s6.yaml new file mode 100644 index 00000000..a28fb559 --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5s6.yaml @@ -0,0 +1,60 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.50 # layer channel multiple +anchors: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/hub/yolov5x6.yaml b/application/yolov5_example/models/hub/yolov5x6.yaml new file mode 100644 index 00000000..ba795c4a --- /dev/null +++ b/application/yolov5_example/models/hub/yolov5x6.yaml @@ -0,0 +1,60 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.33 # model depth multiple +width_multiple: 1.25 # layer channel multiple +anchors: + - [19,27, 44,40, 38,94] # P3/8 + - [96,68, 86,152, 180,137] # P4/16 + - [140,301, 303,264, 238,542] # P5/32 + - [436,615, 739,380, 925,792] # P6/64 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [768, 3, 2]], # 7-P5/32 + [-1, 3, C3, [768]], + [-1, 1, Conv, [1024, 3, 2]], # 9-P6/64 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 11 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [768, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 8], 1, Concat, [1]], # cat backbone P5 + [-1, 3, C3, [768, False]], # 15 + + [-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 19 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 23 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 20], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 26 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 16], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [768, False]], # 29 (P5/32-large) + + [-1, 1, Conv, [768, 3, 2]], + [[-1, 12], 1, Concat, [1]], # cat head P6 + [-1, 3, C3, [1024, False]], # 32 (P6/64-xlarge) + + [[23, 26, 29, 32], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5, P6) + ] diff --git a/application/yolov5_example/models/tf.py b/application/yolov5_example/models/tf.py new file mode 100644 index 00000000..b0d98cc2 --- /dev/null +++ b/application/yolov5_example/models/tf.py @@ -0,0 +1,574 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +TensorFlow, Keras and TFLite versions of YOLOv5 +Authored by https://github.com/zldrobit in PR https://github.com/ultralytics/yolov5/pull/1127 + +Usage: + $ python models/tf.py --weights yolov5s.pt + +Export: + $ python path/to/export.py --weights yolov5s.pt --include saved_model pb tflite tfjs +""" + +import argparse +import sys +from copy import deepcopy +from pathlib import Path + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +# ROOT = ROOT.relative_to(Path.cwd()) # relative + +import numpy as np +import tensorflow as tf +import torch +import torch.nn as nn +from tensorflow import keras + +from models.common import (C3, SPP, SPPF, Bottleneck, BottleneckCSP, C3x, Concat, Conv, CrossConv, DWConv, + DWConvTranspose2d, Focus, autopad) +from models.experimental import MixConv2d, attempt_load +from models.yolo import Detect +from utils.activations import SiLU +from utils.general import LOGGER, make_divisible, print_args + + +class TFBN(keras.layers.Layer): + # TensorFlow BatchNormalization wrapper + def __init__(self, w=None): + super().__init__() + self.bn = keras.layers.BatchNormalization( + beta_initializer=keras.initializers.Constant(w.bias.numpy()), + gamma_initializer=keras.initializers.Constant(w.weight.numpy()), + moving_mean_initializer=keras.initializers.Constant(w.running_mean.numpy()), + moving_variance_initializer=keras.initializers.Constant(w.running_var.numpy()), + epsilon=w.eps) + + def call(self, inputs): + return self.bn(inputs) + + +class TFPad(keras.layers.Layer): + # Pad inputs in spatial dimensions 1 and 2 + def __init__(self, pad): + super().__init__() + if isinstance(pad, int): + self.pad = tf.constant([[0, 0], [pad, pad], [pad, pad], [0, 0]]) + else: # tuple/list + self.pad = tf.constant([[0, 0], [pad[0], pad[0]], [pad[1], pad[1]], [0, 0]]) + + def call(self, inputs): + return tf.pad(inputs, self.pad, mode='constant', constant_values=0) + + +class TFConv(keras.layers.Layer): + # Standard convolution + def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None): + # ch_in, ch_out, weights, kernel, stride, padding, groups + super().__init__() + assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument" + # TensorFlow convolution padding is inconsistent with PyTorch (e.g. k=3 s=2 'SAME' padding) + # see https://stackoverflow.com/questions/52975843/comparing-conv2d-with-padding-between-tensorflow-and-pytorch + conv = keras.layers.Conv2D( + filters=c2, + kernel_size=k, + strides=s, + padding='SAME' if s == 1 else 'VALID', + use_bias=not hasattr(w, 'bn'), + kernel_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()), + bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy())) + self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv]) + self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity + self.act = activations(w.act) if act else tf.identity + + def call(self, inputs): + return self.act(self.bn(self.conv(inputs))) + + +class TFDWConv(keras.layers.Layer): + # Depthwise convolution + def __init__(self, c1, c2, k=1, s=1, p=None, act=True, w=None): + # ch_in, ch_out, weights, kernel, stride, padding, groups + super().__init__() + assert c2 % c1 == 0, f'TFDWConv() output={c2} must be a multiple of input={c1} channels' + conv = keras.layers.DepthwiseConv2D( + kernel_size=k, + depth_multiplier=c2 // c1, + strides=s, + padding='SAME' if s == 1 else 'VALID', + use_bias=not hasattr(w, 'bn'), + depthwise_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()), + bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy())) + self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv]) + self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity + self.act = activations(w.act) if act else tf.identity + + def call(self, inputs): + return self.act(self.bn(self.conv(inputs))) + + +class TFDWConvTranspose2d(keras.layers.Layer): + # Depthwise ConvTranspose2d + def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0, w=None): + # ch_in, ch_out, weights, kernel, stride, padding, groups + super().__init__() + assert c1 == c2, f'TFDWConv() output={c2} must be equal to input={c1} channels' + assert k == 4 and p1 == 1, 'TFDWConv() only valid for k=4 and p1=1' + weight, bias = w.weight.permute(2, 3, 1, 0).numpy(), w.bias.numpy() + self.c1 = c1 + self.conv = [ + keras.layers.Conv2DTranspose(filters=1, + kernel_size=k, + strides=s, + padding='VALID', + output_padding=p2, + use_bias=True, + kernel_initializer=keras.initializers.Constant(weight[..., i:i + 1]), + bias_initializer=keras.initializers.Constant(bias[i])) for i in range(c1)] + + def call(self, inputs): + return tf.concat([m(x) for m, x in zip(self.conv, tf.split(inputs, self.c1, 3))], 3)[:, 1:-1, 1:-1] + + +class TFFocus(keras.layers.Layer): + # Focus wh information into c-space + def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None): + # ch_in, ch_out, kernel, stride, padding, groups + super().__init__() + self.conv = TFConv(c1 * 4, c2, k, s, p, g, act, w.conv) + + def call(self, inputs): # x(b,w,h,c) -> y(b,w/2,h/2,4c) + # inputs = inputs / 255 # normalize 0-255 to 0-1 + inputs = [inputs[:, ::2, ::2, :], inputs[:, 1::2, ::2, :], inputs[:, ::2, 1::2, :], inputs[:, 1::2, 1::2, :]] + return self.conv(tf.concat(inputs, 3)) + + +class TFBottleneck(keras.layers.Layer): + # Standard bottleneck + def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None): # ch_in, ch_out, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv(c_, c2, 3, 1, g=g, w=w.cv2) + self.add = shortcut and c1 == c2 + + def call(self, inputs): + return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs)) + + +class TFCrossConv(keras.layers.Layer): + # Cross Convolution + def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False, w=None): + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = TFConv(c1, c_, (1, k), (1, s), w=w.cv1) + self.cv2 = TFConv(c_, c2, (k, 1), (s, 1), g=g, w=w.cv2) + self.add = shortcut and c1 == c2 + + def call(self, inputs): + return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs)) + + +class TFConv2d(keras.layers.Layer): + # Substitution for PyTorch nn.Conv2D + def __init__(self, c1, c2, k, s=1, g=1, bias=True, w=None): + super().__init__() + assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument" + self.conv = keras.layers.Conv2D(filters=c2, + kernel_size=k, + strides=s, + padding='VALID', + use_bias=bias, + kernel_initializer=keras.initializers.Constant( + w.weight.permute(2, 3, 1, 0).numpy()), + bias_initializer=keras.initializers.Constant(w.bias.numpy()) if bias else None) + + def call(self, inputs): + return self.conv(inputs) + + +class TFBottleneckCSP(keras.layers.Layer): + # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None): + # ch_in, ch_out, number, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv2d(c1, c_, 1, 1, bias=False, w=w.cv2) + self.cv3 = TFConv2d(c_, c_, 1, 1, bias=False, w=w.cv3) + self.cv4 = TFConv(2 * c_, c2, 1, 1, w=w.cv4) + self.bn = TFBN(w.bn) + self.act = lambda x: keras.activations.swish(x) + self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)]) + + def call(self, inputs): + y1 = self.cv3(self.m(self.cv1(inputs))) + y2 = self.cv2(inputs) + return self.cv4(self.act(self.bn(tf.concat((y1, y2), axis=3)))) + + +class TFC3(keras.layers.Layer): + # CSP Bottleneck with 3 convolutions + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None): + # ch_in, ch_out, number, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2) + self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3) + self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)]) + + def call(self, inputs): + return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3)) + + +class TFC3x(keras.layers.Layer): + # 3 module with cross-convolutions + def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None): + # ch_in, ch_out, number, shortcut, groups, expansion + super().__init__() + c_ = int(c2 * e) # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2) + self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3) + self.m = keras.Sequential([ + TFCrossConv(c_, c_, k=3, s=1, g=g, e=1.0, shortcut=shortcut, w=w.m[j]) for j in range(n)]) + + def call(self, inputs): + return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3)) + + +class TFSPP(keras.layers.Layer): + # Spatial pyramid pooling layer used in YOLOv3-SPP + def __init__(self, c1, c2, k=(5, 9, 13), w=None): + super().__init__() + c_ = c1 // 2 # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv(c_ * (len(k) + 1), c2, 1, 1, w=w.cv2) + self.m = [keras.layers.MaxPool2D(pool_size=x, strides=1, padding='SAME') for x in k] + + def call(self, inputs): + x = self.cv1(inputs) + return self.cv2(tf.concat([x] + [m(x) for m in self.m], 3)) + + +class TFSPPF(keras.layers.Layer): + # Spatial pyramid pooling-Fast layer + def __init__(self, c1, c2, k=5, w=None): + super().__init__() + c_ = c1 // 2 # hidden channels + self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1) + self.cv2 = TFConv(c_ * 4, c2, 1, 1, w=w.cv2) + self.m = keras.layers.MaxPool2D(pool_size=k, strides=1, padding='SAME') + + def call(self, inputs): + x = self.cv1(inputs) + y1 = self.m(x) + y2 = self.m(y1) + return self.cv2(tf.concat([x, y1, y2, self.m(y2)], 3)) + + +class TFDetect(keras.layers.Layer): + # TF YOLOv5 Detect layer + def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None): # detection layer + super().__init__() + self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32) + self.nc = nc # number of classes + self.no = nc + 5 # number of outputs per anchor + self.nl = len(anchors) # number of detection layers + self.na = len(anchors[0]) // 2 # number of anchors + self.grid = [tf.zeros(1)] * self.nl # init grid + self.anchors = tf.convert_to_tensor(w.anchors.numpy(), dtype=tf.float32) + self.anchor_grid = tf.reshape(self.anchors * tf.reshape(self.stride, [self.nl, 1, 1]), [self.nl, 1, -1, 1, 2]) + self.m = [TFConv2d(x, self.no * self.na, 1, w=w.m[i]) for i, x in enumerate(ch)] + self.training = False # set to False after building model + self.imgsz = imgsz + for i in range(self.nl): + ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i] + self.grid[i] = self._make_grid(nx, ny) + + def call(self, inputs): + z = [] # inference output + x = [] + for i in range(self.nl): + x.append(self.m[i](inputs[i])) + # x(bs,20,20,255) to x(bs,3,20,20,85) + ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i] + x[i] = tf.reshape(x[i], [-1, ny * nx, self.na, self.no]) + + if not self.training: # inference + y = tf.sigmoid(x[i]) + grid = tf.transpose(self.grid[i], [0, 2, 1, 3]) - 0.5 + anchor_grid = tf.transpose(self.anchor_grid[i], [0, 2, 1, 3]) * 4 + xy = (y[..., 0:2] * 2 + grid) * self.stride[i] # xy + wh = y[..., 2:4] ** 2 * anchor_grid + # Normalize xywh to 0-1 to reduce calibration error + xy /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32) + wh /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32) + y = tf.concat([xy, wh, y[..., 4:]], -1) + z.append(tf.reshape(y, [-1, self.na * ny * nx, self.no])) + + return tf.transpose(x, [0, 2, 1, 3]) if self.training else (tf.concat(z, 1), x) + + @staticmethod + def _make_grid(nx=20, ny=20): + # yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)]) + # return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float() + xv, yv = tf.meshgrid(tf.range(nx), tf.range(ny)) + return tf.cast(tf.reshape(tf.stack([xv, yv], 2), [1, 1, ny * nx, 2]), dtype=tf.float32) + + +class TFUpsample(keras.layers.Layer): + # TF version of torch.nn.Upsample() + def __init__(self, size, scale_factor, mode, w=None): # warning: all arguments needed including 'w' + super().__init__() + assert scale_factor == 2, "scale_factor must be 2" + self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * 2, x.shape[2] * 2), method=mode) + # self.upsample = keras.layers.UpSampling2D(size=scale_factor, interpolation=mode) + # with default arguments: align_corners=False, half_pixel_centers=False + # self.upsample = lambda x: tf.raw_ops.ResizeNearestNeighbor(images=x, + # size=(x.shape[1] * 2, x.shape[2] * 2)) + + def call(self, inputs): + return self.upsample(inputs) + + +class TFConcat(keras.layers.Layer): + # TF version of torch.concat() + def __init__(self, dimension=1, w=None): + super().__init__() + assert dimension == 1, "convert only NCHW to NHWC concat" + self.d = 3 + + def call(self, inputs): + return tf.concat(inputs, self.d) + + +def parse_model(d, ch, model, imgsz): # model_dict, input_channels(3) + LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}") + anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] + na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors + no = na * (nc + 5) # number of outputs = anchors * (classes + 5) + + layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out + for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args + m_str = m + m = eval(m) if isinstance(m, str) else m # eval strings + for j, a in enumerate(args): + try: + args[j] = eval(a) if isinstance(a, str) else a # eval strings + except NameError: + pass + + n = max(round(n * gd), 1) if n > 1 else n # depth gain + if m in [ + nn.Conv2d, Conv, DWConv, DWConvTranspose2d, Bottleneck, SPP, SPPF, MixConv2d, Focus, CrossConv, + BottleneckCSP, C3, C3x]: + c1, c2 = ch[f], args[0] + c2 = make_divisible(c2 * gw, 8) if c2 != no else c2 + + args = [c1, c2, *args[1:]] + if m in [BottleneckCSP, C3, C3x]: + args.insert(2, n) + n = 1 + elif m is nn.BatchNorm2d: + args = [ch[f]] + elif m is Concat: + c2 = sum(ch[-1 if x == -1 else x + 1] for x in f) + elif m is Detect: + args.append([ch[x + 1] for x in f]) + if isinstance(args[1], int): # number of anchors + args[1] = [list(range(args[1] * 2))] * len(f) + args.append(imgsz) + else: + c2 = ch[f] + + tf_m = eval('TF' + m_str.replace('nn.', '')) + m_ = keras.Sequential([tf_m(*args, w=model.model[i][j]) for j in range(n)]) if n > 1 \ + else tf_m(*args, w=model.model[i]) # module + + torch_m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module + t = str(m)[8:-2].replace('__main__.', '') # module type + np = sum(x.numel() for x in torch_m_.parameters()) # number params + m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params + LOGGER.info(f'{i:>3}{str(f):>18}{str(n):>3}{np:>10} {t:<40}{str(args):<30}') # print + save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist + layers.append(m_) + ch.append(c2) + return keras.Sequential(layers), sorted(save) + + +class TFModel: + # TF YOLOv5 model + def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, model=None, imgsz=(640, 640)): # model, channels, classes + super().__init__() + if isinstance(cfg, dict): + self.yaml = cfg # model dict + else: # is *.yaml + import yaml # for torch hub + self.yaml_file = Path(cfg).name + with open(cfg) as f: + self.yaml = yaml.load(f, Loader=yaml.FullLoader) # model dict + + # Define model + if nc and nc != self.yaml['nc']: + LOGGER.info(f"Overriding {cfg} nc={self.yaml['nc']} with nc={nc}") + self.yaml['nc'] = nc # override yaml value + self.model, self.savelist = parse_model(deepcopy(self.yaml), ch=[ch], model=model, imgsz=imgsz) + + def predict(self, + inputs, + tf_nms=False, + agnostic_nms=False, + topk_per_class=100, + topk_all=100, + iou_thres=0.45, + conf_thres=0.25): + y = [] # outputs + x = inputs + for m in self.model.layers: + if m.f != -1: # if not from previous layer + x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers + + x = m(x) # run + y.append(x if m.i in self.savelist else None) # save output + + # Add TensorFlow NMS + if tf_nms: + boxes = self._xywh2xyxy(x[0][..., :4]) + probs = x[0][:, :, 4:5] + classes = x[0][:, :, 5:] + scores = probs * classes + if agnostic_nms: + nms = AgnosticNMS()((boxes, classes, scores), topk_all, iou_thres, conf_thres) + else: + boxes = tf.expand_dims(boxes, 2) + nms = tf.image.combined_non_max_suppression(boxes, + scores, + topk_per_class, + topk_all, + iou_thres, + conf_thres, + clip_boxes=False) + return nms, x[1] + return x[0] # output only first tensor [1,6300,85] = [xywh, conf, class0, class1, ...] + # x = x[0][0] # [x(1,6300,85), ...] to x(6300,85) + # xywh = x[..., :4] # x(6300,4) boxes + # conf = x[..., 4:5] # x(6300,1) confidences + # cls = tf.reshape(tf.cast(tf.argmax(x[..., 5:], axis=1), tf.float32), (-1, 1)) # x(6300,1) classes + # return tf.concat([conf, cls, xywh], 1) + + @staticmethod + def _xywh2xyxy(xywh): + # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right + x, y, w, h = tf.split(xywh, num_or_size_splits=4, axis=-1) + return tf.concat([x - w / 2, y - h / 2, x + w / 2, y + h / 2], axis=-1) + + +class AgnosticNMS(keras.layers.Layer): + # TF Agnostic NMS + def call(self, input, topk_all, iou_thres, conf_thres): + # wrap map_fn to avoid TypeSpec related error https://stackoverflow.com/a/65809989/3036450 + return tf.map_fn(lambda x: self._nms(x, topk_all, iou_thres, conf_thres), + input, + fn_output_signature=(tf.float32, tf.float32, tf.float32, tf.int32), + name='agnostic_nms') + + @staticmethod + def _nms(x, topk_all=100, iou_thres=0.45, conf_thres=0.25): # agnostic NMS + boxes, classes, scores = x + class_inds = tf.cast(tf.argmax(classes, axis=-1), tf.float32) + scores_inp = tf.reduce_max(scores, -1) + selected_inds = tf.image.non_max_suppression(boxes, + scores_inp, + max_output_size=topk_all, + iou_threshold=iou_thres, + score_threshold=conf_thres) + selected_boxes = tf.gather(boxes, selected_inds) + padded_boxes = tf.pad(selected_boxes, + paddings=[[0, topk_all - tf.shape(selected_boxes)[0]], [0, 0]], + mode="CONSTANT", + constant_values=0.0) + selected_scores = tf.gather(scores_inp, selected_inds) + padded_scores = tf.pad(selected_scores, + paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]], + mode="CONSTANT", + constant_values=-1.0) + selected_classes = tf.gather(class_inds, selected_inds) + padded_classes = tf.pad(selected_classes, + paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]], + mode="CONSTANT", + constant_values=-1.0) + valid_detections = tf.shape(selected_inds)[0] + return padded_boxes, padded_scores, padded_classes, valid_detections + + +def activations(act=nn.SiLU): + # Returns TF activation from input PyTorch activation + if isinstance(act, nn.LeakyReLU): + return lambda x: keras.activations.relu(x, alpha=0.1) + elif isinstance(act, nn.Hardswish): + return lambda x: x * tf.nn.relu6(x + 3) * 0.166666667 + elif isinstance(act, (nn.SiLU, SiLU)): + return lambda x: keras.activations.swish(x) + else: + raise Exception(f'no matching TensorFlow activation found for PyTorch activation {act}') + + +def representative_dataset_gen(dataset, ncalib=100): + # Representative dataset generator for use with converter.representative_dataset, returns a generator of np arrays + for n, (path, img, im0s, vid_cap, string) in enumerate(dataset): + im = np.transpose(img, [1, 2, 0]) + im = np.expand_dims(im, axis=0).astype(np.float32) + im /= 255 + yield [im] + if n >= ncalib: + break + + +def run( + weights=ROOT / 'yolov5s.pt', # weights path + imgsz=(640, 640), # inference size h,w + batch_size=1, # batch size + dynamic=False, # dynamic batch size +): + # PyTorch model + im = torch.zeros((batch_size, 3, *imgsz)) # BCHW image + model = attempt_load(weights, device=torch.device('cpu'), inplace=True, fuse=False) + _ = model(im) # inference + model.info() + + # TensorFlow model + im = tf.zeros((batch_size, *imgsz, 3)) # BHWC image + tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz) + _ = tf_model.predict(im) # inference + + # Keras model + im = keras.Input(shape=(*imgsz, 3), batch_size=None if dynamic else batch_size) + keras_model = keras.Model(inputs=im, outputs=tf_model.predict(im)) + keras_model.summary() + + LOGGER.info('PyTorch, TensorFlow and Keras models successfully verified.\nUse export.py for TF model export.') + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='weights path') + parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w') + parser.add_argument('--batch-size', type=int, default=1, help='batch size') + parser.add_argument('--dynamic', action='store_true', help='dynamic batch size') + opt = parser.parse_args() + opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand + print_args(vars(opt)) + return opt + + +def main(opt): + run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/models/yolo.py b/application/yolov5_example/models/yolo.py new file mode 100644 index 00000000..e7a9fa1a --- /dev/null +++ b/application/yolov5_example/models/yolo.py @@ -0,0 +1,360 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +YOLO-specific modules + +Usage: + $ python path/to/models/yolo.py --cfg yolov5s.yaml +""" + +import argparse +import contextlib +import os +import platform +import sys +from copy import deepcopy +from pathlib import Path + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +if platform.system() != 'Windows': + ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from models.common import * +from models.experimental import * +from utils.autoanchor import check_anchor_order +from utils.general import LOGGER, check_version, check_yaml, make_divisible, print_args +from utils.plots import feature_visualization +from utils.torch_utils import (fuse_conv_and_bn, initialize_weights, model_info, profile, scale_img, select_device, + time_sync) + +try: + import thop # for FLOPs computation +except ImportError: + thop = None + + +class Detect(nn.Module): + stride = None # strides computed during build + onnx_dynamic = False # ONNX export parameter + export = False # export mode + + def __init__(self, nc=80, anchors=(), ch=(), inplace=True): # detection layer + super().__init__() + self.nc = nc # number of classes + self.no = nc + 5 # number of outputs per anchor + self.nl = len(anchors) # number of detection layers + self.na = len(anchors[0]) // 2 # number of anchors + self.grid = [torch.zeros(1)] * self.nl # init grid + self.anchor_grid = [torch.zeros(1)] * self.nl # init anchor grid + self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2)) # shape(nl,na,2) + self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv + self.inplace = inplace # use inplace ops (e.g. slice assignment) + + def forward(self, x): + z = [] # inference output + for i in range(self.nl): + x[i] = self.m[i](x[i]) # conv + bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85) + x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous() + + if not self.training: # inference + if self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]: + self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i) + + y = x[i].sigmoid() + if self.inplace: + y[..., 0:2] = (y[..., 0:2] * 2 + self.grid[i]) * self.stride[i] # xy + y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh + else: # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953 + xy, wh, conf = y.split((2, 2, self.nc + 1), 4) # y.tensor_split((2, 4, 5), 4) # torch 1.8.0 + xy = (xy * 2 + self.grid[i]) * self.stride[i] # xy + wh = (wh * 2) ** 2 * self.anchor_grid[i] # wh + y = torch.cat((xy, wh, conf), 4) + z.append(y.view(bs, -1, self.no)) + + return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x) + + def _make_grid(self, nx=20, ny=20, i=0, torch_1_10=check_version(torch.__version__, '1.10.0')): + d = self.anchors[i].device + t = self.anchors[i].dtype + shape = 1, self.na, ny, nx, 2 # grid shape + y, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t) + if torch_1_10: # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibility + yv, xv = torch.meshgrid(y, x, indexing='ij') + else: + yv, xv = torch.meshgrid(y, x) + grid = torch.stack((xv, yv), 2).expand(shape) - 0.5 # add grid offset, i.e. y = 2.0 * x - 0.5 + anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape) + return grid, anchor_grid + + +class BaseModel(nn.Module): + # YOLOv5 base model + def forward(self, x, profile=False, visualize=False): + return self._forward_once(x, profile, visualize) # single-scale inference, train + + def _forward_once(self, x, profile=False, visualize=False): + y, dt = [], [] # outputs + for m in self.model: + if m.f != -1: # if not from previous layer + x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers + # if profile: + # self._profile_one_layer(m, x, dt) + x = m(x) # run + y.append(x if m.i in self.save else None) # save output + # if visualize: + # feature_visualization(x, m.type, m.i, save_dir=visualize) + return x + + def _profile_one_layer(self, m, x, dt): + c = m == self.model[-1] # is final layer, copy input as inplace fix + o = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPs + t = time_sync() + for _ in range(10): + m(x.copy() if c else x) + dt.append((time_sync() - t) * 100) + if m == self.model[0]: + LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} module") + LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}') + if c: + LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total") + + def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers + LOGGER.info('Fusing layers... ') + for m in self.model.modules(): + if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'): + m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv + delattr(m, 'bn') # remove batchnorm + m.forward = m.forward_fuse # update forward + self.info() + return self + + def info(self, verbose=False, img_size=640): # print model information + model_info(self, verbose, img_size) + + def _apply(self, fn): + # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers + self = super()._apply(fn) + m = self.model[-1] # Detect() + if isinstance(m, Detect): + m.stride = fn(m.stride) + m.grid = list(map(fn, m.grid)) + if isinstance(m.anchor_grid, list): + m.anchor_grid = list(map(fn, m.anchor_grid)) + return self + + +class DetectionModel(BaseModel): + # YOLOv5 detection model + def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes + super().__init__() + if isinstance(cfg, dict): + self.yaml = cfg # model dict + else: # is *.yaml + import yaml # for torch hub + self.yaml_file = Path(cfg).name + with open(cfg, encoding='ascii', errors='ignore') as f: + self.yaml = yaml.safe_load(f) # model dict + + # Define model + ch = self.yaml['ch'] = self.yaml.get('ch', ch) # input channels + if nc and nc != self.yaml['nc']: + LOGGER.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}") + self.yaml['nc'] = nc # override yaml value + if anchors: + LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}') + self.yaml['anchors'] = round(anchors) # override yaml value + self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist + self.names = [str(i) for i in range(self.yaml['nc'])] # default names + self.inplace = self.yaml.get('inplace', True) + + # Build strides, anchors + m = self.model[-1] # Detect() + if isinstance(m, Detect): + s = 256 # 2x min stride + m.inplace = self.inplace + m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # forward + check_anchor_order(m) # must be in pixel-space (not grid-space) + m.anchors /= m.stride.view(-1, 1, 1) + self.stride = m.stride + self._initialize_biases() # only run once + + # Init weights, biases + initialize_weights(self) + self.info() + LOGGER.info('') + + def forward(self, x, augment=False, profile=False, visualize=False): + # if augment: + # return self._forward_augment(x) # augmented inference, None + return self._forward_once(x, profile, visualize) # single-scale inference, train + + def _forward_augment(self, x): + img_size = x.shape[-2:] # height, width + s = [1, 0.83, 0.67] # scales + f = [None, 3, None] # flips (2-ud, 3-lr) + y = [] # outputs + for si, fi in zip(s, f): + xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max())) + yi = self._forward_once(xi)[0] # forward + # cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1]) # save + yi = self._descale_pred(yi, fi, si, img_size) + y.append(yi) + y = self._clip_augmented(y) # clip augmented tails + return torch.cat(y, 1), None # augmented inference, train + + def _descale_pred(self, p, flips, scale, img_size): + # de-scale predictions following augmented inference (inverse operation) + if self.inplace: + p[..., :4] /= scale # de-scale + if flips == 2: + p[..., 1] = img_size[0] - p[..., 1] # de-flip ud + elif flips == 3: + p[..., 0] = img_size[1] - p[..., 0] # de-flip lr + else: + x, y, wh = p[..., 0:1] / scale, p[..., 1:2] / scale, p[..., 2:4] / scale # de-scale + if flips == 2: + y = img_size[0] - y # de-flip ud + elif flips == 3: + x = img_size[1] - x # de-flip lr + p = torch.cat((x, y, wh, p[..., 4:]), -1) + return p + + def _clip_augmented(self, y): + # Clip YOLOv5 augmented inference tails + nl = self.model[-1].nl # number of detection layers (P3-P5) + g = sum(4 ** x for x in range(nl)) # grid points + e = 1 # exclude layer count + i = (y[0].shape[1] // g) * sum(4 ** x for x in range(e)) # indices + y[0] = y[0][:, :-i] # large + i = (y[-1].shape[1] // g) * sum(4 ** (nl - 1 - x) for x in range(e)) # indices + y[-1] = y[-1][:, i:] # small + return y + + def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency + # https://arxiv.org/abs/1708.02002 section 3.3 + # cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1. + m = self.model[-1] # Detect() module + for mi, s in zip(m.m, m.stride): # from + b = mi.bias.view(m.na, -1).detach() # conv.bias(255) to (3,85) + b[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image) + b[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum()) # cls + mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True) + + +Model = DetectionModel # retain YOLOv5 'Model' class for backwards compatibility + + +class ClassificationModel(BaseModel): + # YOLOv5 classification model + def __init__(self, cfg=None, model=None, nc=1000, cutoff=10): # yaml, model, number of classes, cutoff index + super().__init__() + self._from_detection_model(model, nc, cutoff) if model is not None else self._from_yaml(cfg) + + def _from_detection_model(self, model, nc=1000, cutoff=10): + # Create a YOLOv5 classification model from a YOLOv5 detection model + if isinstance(model, DetectMultiBackend): + model = model.model # unwrap DetectMultiBackend + model.model = model.model[:cutoff] # backbone + m = model.model[-1] # last layer + ch = m.conv.in_channels if hasattr(m, 'conv') else m.cv1.conv.in_channels # ch into module + c = Classify(ch, nc) # Classify() + c.i, c.f, c.type = m.i, m.f, 'models.common.Classify' # index, from, type + model.model[-1] = c # replace + self.model = model.model + self.stride = model.stride + self.save = [] + self.nc = nc + + def _from_yaml(self, cfg): + # Create a YOLOv5 classification model from a *.yaml file + self.model = None + + +def parse_model(d, ch): # model_dict, input_channels(3) + LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}") + anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] + na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors + no = na * (nc + 5) # number of outputs = anchors * (classes + 5) + + layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out + for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args + m = eval(m) if isinstance(m, str) else m # eval strings + for j, a in enumerate(args): + with contextlib.suppress(NameError): + args[j] = eval(a) if isinstance(a, str) else a # eval strings + + n = n_ = max(round(n * gd), 1) if n > 1 else n # depth gain + if m in (Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv, + BottleneckCSP, C3, C3TR, C3SPP, C3Ghost, nn.ConvTranspose2d, DWConvTranspose2d, C3x): + c1, c2 = ch[f], args[0] + if c2 != no: # if not output + c2 = make_divisible(c2 * gw, 8) + + args = [c1, c2, *args[1:]] + if m in [BottleneckCSP, C3, C3TR, C3Ghost, C3x]: + args.insert(2, n) # number of repeats + n = 1 + elif m is nn.BatchNorm2d: + args = [ch[f]] + elif m is Concat: + c2 = sum(ch[x] for x in f) + elif m is Detect: + args.append([ch[x] for x in f]) + if isinstance(args[1], int): # number of anchors + args[1] = [list(range(args[1] * 2))] * len(f) + elif m is Contract: + c2 = ch[f] * args[0] ** 2 + elif m is Expand: + c2 = ch[f] // args[0] ** 2 + else: + c2 = ch[f] + + m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module + t = str(m)[8:-2].replace('__main__.', '') # module type + np = sum(x.numel() for x in m_.parameters()) # number params + m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params + LOGGER.info(f'{i:>3}{str(f):>18}{n_:>3}{np:10.0f} {t:<40}{str(args):<30}') # print + save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist + layers.append(m_) + if i == 0: + ch = [] + ch.append(c2) + return nn.Sequential(*layers), sorted(save) + + +if __name__ == '__main__': + parser = argparse.ArgumentParser() + parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml') + parser.add_argument('--batch-size', type=int, default=1, help='total batch size for all GPUs') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--profile', action='store_true', help='profile model speed') + parser.add_argument('--line-profile', action='store_true', help='profile model speed layer by layer') + parser.add_argument('--test', action='store_true', help='test all yolo*.yaml') + opt = parser.parse_args() + opt.cfg = check_yaml(opt.cfg) # check YAML + print_args(vars(opt)) + device = select_device(opt.device) + + # Create model + im = torch.rand(opt.batch_size, 3, 640, 640).to(device) + model = Model(opt.cfg).to(device) + + # Options + if opt.line_profile: # profile layer by layer + model(im, profile=True) + + elif opt.profile: # profile forward-backward + results = profile(input=im, ops=[model], n=3) + + elif opt.test: # test all models + for cfg in Path(ROOT / 'models').rglob('yolo*.yaml'): + try: + _ = Model(cfg) + except Exception as e: + print(f'Error in {cfg}: {e}') + + else: # report fused model summary + model.fuse() diff --git a/application/yolov5_example/models/yolov5l.yaml b/application/yolov5_example/models/yolov5l.yaml new file mode 100644 index 00000000..ce8a5de4 --- /dev/null +++ b/application/yolov5_example/models/yolov5l.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.0 # model depth multiple +width_multiple: 1.0 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/yolov5m.yaml b/application/yolov5_example/models/yolov5m.yaml new file mode 100644 index 00000000..ad13ab37 --- /dev/null +++ b/application/yolov5_example/models/yolov5m.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.67 # model depth multiple +width_multiple: 0.75 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/yolov5n.yaml b/application/yolov5_example/models/yolov5n.yaml new file mode 100644 index 00000000..8a28a40d --- /dev/null +++ b/application/yolov5_example/models/yolov5n.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.25 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/yolov5s.yaml b/application/yolov5_example/models/yolov5s.yaml new file mode 100644 index 00000000..f35beabb --- /dev/null +++ b/application/yolov5_example/models/yolov5s.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 0.33 # model depth multiple +width_multiple: 0.50 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/models/yolov5x.yaml b/application/yolov5_example/models/yolov5x.yaml new file mode 100644 index 00000000..f617a027 --- /dev/null +++ b/application/yolov5_example/models/yolov5x.yaml @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license + +# Parameters +nc: 80 # number of classes +depth_multiple: 1.33 # model depth multiple +width_multiple: 1.25 # layer channel multiple +anchors: + - [10,13, 16,30, 33,23] # P3/8 + - [30,61, 62,45, 59,119] # P4/16 + - [116,90, 156,198, 373,326] # P5/32 + +# YOLOv5 v6.0 backbone +backbone: + # [from, number, module, args] + [[-1, 1, Conv, [64, 6, 2, 2]], # 0-P1/2 + [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 + [-1, 3, C3, [128]], + [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 + [-1, 6, C3, [256]], + [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 + [-1, 9, C3, [512]], + [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 + [-1, 3, C3, [1024]], + [-1, 1, SPPF, [1024, 5]], # 9 + ] + +# YOLOv5 v6.0 head +head: + [[-1, 1, Conv, [512, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 6], 1, Concat, [1]], # cat backbone P4 + [-1, 3, C3, [512, False]], # 13 + + [-1, 1, Conv, [256, 1, 1]], + [-1, 1, nn.Upsample, [None, 2, 'nearest']], + [[-1, 4], 1, Concat, [1]], # cat backbone P3 + [-1, 3, C3, [256, False]], # 17 (P3/8-small) + + [-1, 1, Conv, [256, 3, 2]], + [[-1, 14], 1, Concat, [1]], # cat head P4 + [-1, 3, C3, [512, False]], # 20 (P4/16-medium) + + [-1, 1, Conv, [512, 3, 2]], + [[-1, 10], 1, Concat, [1]], # cat head P5 + [-1, 3, C3, [1024, False]], # 23 (P5/32-large) + + [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) + ] diff --git a/application/yolov5_example/mqbench.code-workspace b/application/yolov5_example/mqbench.code-workspace new file mode 100644 index 00000000..7e768222 --- /dev/null +++ b/application/yolov5_example/mqbench.code-workspace @@ -0,0 +1,11 @@ +{ + "folders": [ + { + "path": "../../opt/conda/lib/python3.7/site-packages/MQBench-0.0.6-py3.7.egg/mqbench" + }, + { + "path": "." + } + ], + "settings": {} +} \ No newline at end of file diff --git a/application/yolov5_example/requirements.txt b/application/yolov5_example/requirements.txt new file mode 100644 index 00000000..10620566 --- /dev/null +++ b/application/yolov5_example/requirements.txt @@ -0,0 +1,43 @@ +# YOLOv5 requirements +# Usage: pip install -r requirements.txt + +# Base ---------------------------------------- +matplotlib>=3.2.2 +numpy>=1.18.5 +opencv-python>=4.1.1 +Pillow>=7.1.2 +PyYAML>=5.3.1 +requests>=2.23.0 +scipy>=1.4.1 +torch>=1.7.0 +torchvision>=0.8.1 +tqdm>=4.64.0 +protobuf<=3.20.1 # https://github.com/ultralytics/yolov5/issues/8012 + +# Logging ------------------------------------- +tensorboard>=2.4.1 +# wandb +# clearml + +# Plotting ------------------------------------ +pandas>=1.1.4 +seaborn>=0.11.0 + +# Export -------------------------------------- +# coremltools>=5.2 # CoreML export +# onnx>=1.9.0 # ONNX export +# onnx-simplifier>=0.4.1 # ONNX simplifier +# nvidia-pyindex # TensorRT export +# nvidia-tensorrt # TensorRT export +# scikit-learn==0.19.2 # CoreML quantization +# tensorflow>=2.4.1 # TFLite export (or tensorflow-cpu, tensorflow-aarch64) +# tensorflowjs>=3.9.0 # TF.js export +# openvino-dev # OpenVINO export + +# Extras -------------------------------------- +ipython # interactive notebook +psutil # system utilization +thop>=0.1.1 # FLOPs computation +# albumentations>=1.0.3 +# pycocotools>=2.0 # COCO mAP +# roboflow diff --git a/application/yolov5_example/setup.cfg b/application/yolov5_example/setup.cfg new file mode 100644 index 00000000..020a7574 --- /dev/null +++ b/application/yolov5_example/setup.cfg @@ -0,0 +1,59 @@ +# Project-wide configuration file, can be used for package metadata and other toll configurations +# Example usage: global configuration for PEP8 (via flake8) setting or default pytest arguments +# Local usage: pip install pre-commit, pre-commit run --all-files + +[metadata] +license_file = LICENSE +description_file = README.md + + +[tool:pytest] +norecursedirs = + .git + dist + build +addopts = + --doctest-modules + --durations=25 + --color=yes + + +[flake8] +max-line-length = 120 +exclude = .tox,*.egg,build,temp +select = E,W,F +doctests = True +verbose = 2 +# https://pep8.readthedocs.io/en/latest/intro.html#error-codes +format = pylint +# see: https://www.flake8rules.com/ +ignore = + E731 # Do not assign a lambda expression, use a def + F405 # name may be undefined, or defined from star imports: module + E402 # module level import not at top of file + F401 # module imported but unused + W504 # line break after binary operator + E127 # continuation line over-indented for visual indent + W504 # line break after binary operator + E231 # missing whitespace after 鈥,鈥, 鈥;鈥, or 鈥:鈥 + E501 # line too long + F403 # 鈥榝rom module import *鈥 used; unable to detect undefined names + + +[isort] +# https://pycqa.github.io/isort/docs/configuration/options.html +line_length = 120 +# see: https://pycqa.github.io/isort/docs/configuration/multi_line_output_modes.html +multi_line_output = 0 + + +[yapf] +based_on_style = pep8 +spaces_before_comment = 2 +COLUMN_LIMIT = 120 +COALESCE_BRACKETS = True +SPACES_AROUND_POWER_OPERATOR = True +SPACE_BETWEEN_ENDING_COMMA_AND_CLOSING_BRACKET = False +SPLIT_BEFORE_CLOSING_BRACKET = False +SPLIT_BEFORE_FIRST_ARGUMENT = False +# EACH_DICT_ENTRY_ON_SEPARATE_LINE = False diff --git a/application/yolov5_example/train.py b/application/yolov5_example/train.py new file mode 100644 index 00000000..03aa2509 --- /dev/null +++ b/application/yolov5_example/train.py @@ -0,0 +1,727 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Train a YOLOv5 model on a custom dataset. + +Models and datasets download automatically from the latest YOLOv5 release. +Models: https://github.com/ultralytics/yolov5/tree/master/models +Datasets: https://github.com/ultralytics/yolov5/tree/master/data +Tutorial: https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data + +Usage: + $ python path/to/train.py --data coco128.yaml --weights yolov5s.pt --img 640 # from pretrained (RECOMMENDED) + $ python path/to/train.py --data coco128.yaml --weights '' --cfg yolov5s.yaml --img 640 # from scratch +""" + +import argparse +from ast import arg +import math +import os +import random +import sys +import time +from copy import deepcopy +from datetime import datetime +from pathlib import Path + +import numpy as np +import torch +import torch.distributed as dist +import torch.nn as nn +import yaml +from torch.optim import lr_scheduler +from tqdm import tqdm +from mqbench.convert_deploy import convert_deploy, convert_onnx +from mqbench.prepare_by_platform import prepare_by_platform, BackendType +from mqbench.utils.state import enable_calibration, enable_quantization, disable_all + + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[0] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +import val # for end-of-epoch mAP +from models.experimental import attempt_load +from models.yolo import Model +from utils.autoanchor import check_anchors +from utils.autobatch import check_train_batch_size +from utils.callbacks import Callbacks +from utils.dataloaders import create_dataloader +from utils.downloads import attempt_download, is_url +from utils.general import (LOGGER, check_amp, check_dataset, check_file, check_git_status, check_img_size, + check_requirements, check_suffix, check_yaml, colorstr, get_latest_run, increment_path, + init_seeds, intersect_dicts, labels_to_class_weights, labels_to_image_weights, methods, + one_cycle, print_args, print_mutation, strip_optimizer, yaml_save) +from utils.loggers import Loggers +from utils.loggers.wandb.wandb_utils import check_wandb_resume +from utils.loss import ComputeLoss +from utils.metrics import fitness +from utils.plots import plot_evolve, plot_labels +from utils.torch_utils import (EarlyStopping, ModelEMA, de_parallel, select_device, smart_DDP, smart_optimizer, + smart_resume, torch_distributed_zero_first) + +LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html +RANK = int(os.getenv('RANK', -1)) +WORLD_SIZE = int(os.getenv('WORLD_SIZE', 1)) + + +def train(hyp, opt, device, callbacks): # hyp is path/to/hyp.yaml or hyp dictionary + save_dir, epochs, batch_size, weights, single_cls, evolve, data, cfg, resume, noval, nosave, workers, freeze = \ + Path(opt.save_dir), opt.epochs, opt.batch_size, opt.weights, opt.single_cls, opt.evolve, opt.data, opt.cfg, \ + opt.resume, opt.noval, opt.nosave, opt.workers, opt.freeze + callbacks.run('on_pretrain_routine_start') + + # Directories + w = save_dir / 'weights' # weights dir + (w.parent if evolve else w).mkdir(parents=True, exist_ok=True) # make dir + last, best = w / 'last.pt', w / 'best.pt' + + # Hyperparameters + if isinstance(hyp, str): + with open(hyp, errors='ignore') as f: + hyp = yaml.safe_load(f) # load hyps dict + LOGGER.info(colorstr('hyperparameters: ') + ', '.join(f'{k}={v}' for k, v in hyp.items())) + opt.hyp = hyp.copy() # for saving hyps to checkpoints + + # Save run settings + if not evolve: + yaml_save(save_dir / 'hyp.yaml', hyp) + yaml_save(save_dir / 'opt.yaml', vars(opt)) + + # Loggers + data_dict = None + if RANK in {-1, 0}: + loggers = Loggers(save_dir, weights, opt, hyp, LOGGER) # loggers instance + if loggers.clearml: + data_dict = loggers.clearml.data_dict # None if no ClearML dataset or filled in by ClearML + if loggers.wandb: + data_dict = loggers.wandb.data_dict + if resume: + weights, epochs, hyp, batch_size = opt.weights, opt.epochs, opt.hyp, opt.batch_size + + # Register actions + for k in methods(loggers): + callbacks.register_action(k, callback=getattr(loggers, k)) + + # Config + plots = not evolve and not opt.noplots # create plots + cuda = device.type != 'cpu' + init_seeds(opt.seed + 1 + RANK, deterministic=True) + with torch_distributed_zero_first(LOCAL_RANK): + data_dict = data_dict or check_dataset(data) # check if None + train_path, val_path = data_dict['train'], data_dict['val'] + nc = 1 if single_cls else int(data_dict['nc']) # number of classes + names = ['item'] if single_cls and len(data_dict['names']) != 1 else data_dict['names'] # class names + assert len(names) == nc, f'{len(names)} names found for nc={nc} dataset in {data}' # check + is_coco = isinstance(val_path, str) and val_path.endswith('coco/val2017.txt') # COCO dataset + + # Model + check_suffix(weights, '.pt') # check weights + pretrained = weights.endswith('.pt') + if pretrained: + with torch_distributed_zero_first(LOCAL_RANK): + weights = attempt_download(weights) # download if not found locally + ckpt = torch.load(weights, map_location='cpu') # load checkpoint to CPU to avoid CUDA memory leak + model = Model(cfg or ckpt['model'].yaml, ch=3, nc=nc, anchors=hyp.get('anchors')).to(device) # create + exclude = ['anchor'] if (cfg or hyp.get('anchors')) and not resume else [] # exclude keys + csd = ckpt['model'].float().state_dict() # checkpoint state_dict as FP32 + csd = intersect_dicts(csd, model.state_dict(), exclude=exclude) # intersect + model.load_state_dict(csd, strict=False) # load + LOGGER.info(f'Transferred {len(csd)}/{len(model.state_dict())} items from {weights}') # report + else: + model = Model(cfg, ch=3, nc=nc, anchors=hyp.get('anchors')).to(device) # create + amp = check_amp(model) # check AMP + amp = False + + # Freeze + freeze = [f'model.{x}.' for x in (freeze if len(freeze) > 1 else range(freeze[0]))] # layers to freeze + for k, v in model.named_parameters(): + v.requires_grad = True # train all layers + # v.register_hook(lambda x: torch.nan_to_num(x)) # NaN to 0 (commented for erratic training results) + if any(x in k for x in freeze): + LOGGER.info(f'freezing {k}') + v.requires_grad = False + + # Image size + gs = max(int(model.stride.max()), 32) # grid size (max stride) + imgsz = check_img_size(opt.imgsz, gs, floor=gs * 2) # verify imgsz is gs-multiple + + # Batch size + if RANK == -1 and batch_size == -1: # single-GPU only, estimate best batch size + batch_size = check_train_batch_size(model, imgsz, amp) + loggers.on_params_update({"batch_size": batch_size}) + + # Optimizer + nbs = 64 # nominal batch size + accumulate = max(round(nbs / batch_size), 1) # accumulate loss before optimizing + hyp['weight_decay'] *= batch_size * accumulate / nbs # scale weight_decay + optimizer = smart_optimizer(model, opt.optimizer, hyp['lr0'], hyp['momentum'], hyp['weight_decay']) + + print('wxc1 lr0:', hyp['lr0']) + # Scheduler + if opt.cos_lr: + lf = one_cycle(1, hyp['lrf'], epochs) # cosine 1->hyp['lrf'] + else: + lf = lambda x: (1 - x / epochs) * (1.0 - hyp['lrf']) + hyp['lrf'] # linear + scheduler = lr_scheduler.LambdaLR(optimizer, lr_lambda=lf) # plot_lr_scheduler(optimizer, scheduler, epochs) + + # EMA + ema = ModelEMA(model) if RANK in {-1, 0} else None + + # Resume + best_fitness, start_epoch = 0.0, 0 + if pretrained: + if resume: + best_fitness, start_epoch, epochs = smart_resume(ckpt, optimizer, ema, weights, epochs, resume) + del ckpt, csd + + # DP mode + if cuda and RANK == -1 and torch.cuda.device_count() > 1: + LOGGER.warning('WARNING: DP not recommended, use torch.distributed.run for best DDP Multi-GPU results.\n' + 'See Multi-GPU Tutorial at https://github.com/ultralytics/yolov5/issues/475 to get started.') + model = torch.nn.DataParallel(model) + + # SyncBatchNorm + if opt.sync_bn and cuda and RANK != -1: + model = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model).to(device) + LOGGER.info('Using SyncBatchNorm()') + + # Trainloader + train_loader, dataset = create_dataloader(train_path, + imgsz, + batch_size // WORLD_SIZE, + gs, + single_cls, + hyp=hyp, + augment=True, + cache=None if opt.cache == 'val' else opt.cache, + rect=opt.rect, + rank=LOCAL_RANK, + workers=workers, + image_weights=opt.image_weights, + quad=opt.quad, + prefix=colorstr('train: '), + shuffle=True) + labels = np.concatenate(dataset.labels, 0) + mlc = int(labels[:, 0].max()) # max label class + assert mlc < nc, f'Label class {mlc} exceeds nc={nc} in {data}. Possible class labels are 0-{nc - 1}' + + # Process 0 + if RANK in {-1, 0}: + val_loader = create_dataloader(val_path, + imgsz, + batch_size // WORLD_SIZE * 2, + gs, + single_cls, + hyp=hyp, + cache=None if noval else opt.cache, + rect=True, + rank=-1, + workers=workers * 2, + pad=0.5, + prefix=colorstr('val: '))[0] + + if not resume: + if plots: + plot_labels(labels, names, save_dir) + + # Anchors + if not opt.noautoanchor: + check_anchors(dataset, model=model, thr=hyp['anchor_t'], imgsz=imgsz) + model.half().float() # pre-reduce anchor precision + + callbacks.run('on_pretrain_routine_end') + + # DDP mode + if cuda and RANK != -1: + model = smart_DDP(model) + + # Model attributes + nl = de_parallel(model).model[-1].nl # number of detection layers (to scale hyps) + hyp['box'] *= 3 / nl # scale to layers + hyp['cls'] *= nc / 80 * 3 / nl # scale to classes and layers + hyp['obj'] *= (imgsz / 640) ** 2 * 3 / nl # scale to image size and layers + hyp['label_smoothing'] = opt.label_smoothing + model.nc = nc # attach number of classes to model + model.hyp = hyp # attach hyperparameters to model + model.class_weights = labels_to_class_weights(dataset.labels, nc).to(device) * nc # attach class weights + model.names = names + + # Start training + t0 = time.time() + nb = len(train_loader) # number of batches + nw = max(round(hyp['warmup_epochs'] * nb), 100) # number of warmup iterations, max(3 epochs, 100 iterations) + # nw = min(nw, (epochs - start_epoch) / 2 * nb) # limit warmup to < 1/2 of training + last_opt_step = -1 + maps = np.zeros(nc) # mAP per class + results = (0, 0, 0, 0, 0, 0, 0) # P, R, mAP@.5, mAP@.5-.95, val_loss(box, obj, cls) + scheduler.last_epoch = start_epoch - 1 # do not move + scaler = torch.cuda.amp.GradScaler(enabled=amp) + stopper, stop = EarlyStopping(patience=opt.patience), False + compute_loss = ComputeLoss(model) # init loss class + callbacks.run('on_train_start') + LOGGER.info(f'Image sizes {imgsz} train, {imgsz} val\n' + f'Using {train_loader.num_workers * WORLD_SIZE} dataloader workers\n' + f"Logging results to {colorstr('bold', save_dir)}\n" + f'Starting training for {epochs} epochs...') + + model_name = opt.cfg.split('/')[-1].split('.')[0] + output_dir = os.path.join(opt.output_path, model_name) + os.system('rm -rf {};mkdir -p {}'.format(output_dir, output_dir)) + if opt.pre_eval_and_export: + import copy + print('鍘熷onnx妯″瀷绮惧害') + results, maps, _ = val.run(data_dict, + batch_size=batch_size // WORLD_SIZE * 2, + imgsz=imgsz, + half=amp, + model=ema.ema, + single_cls=single_cls, + dataloader=val_loader, + save_dir=save_dir, + plots=False, + callbacks=callbacks, + compute_loss=compute_loss) + kwargs = { + 'input_shape_dict': {'data': [1, 3, opt.imgsz, opt.imgsz]}, + 'output_path': output_dir, + 'model_name': model_name, + 'dummy_input': None, + 'onnx_model_path': os.path.join(output_dir, '{}_ori.onnx'.format(model_name)), + } + module_tmp = copy.deepcopy(model) + module_tmp = module_tmp.cpu() + convert_onnx(module_tmp.eval(), **kwargs) + del module_tmp + model = model.train() #prepare鍓嶄竴瀹氳鏄痶rain妯″紡锛侊紒 + # exit(0) + + backend = BackendType.Sophgo_TPU + if opt.quantize: + prepare_custom_config_dict= { + # 'extra_qconfig_dict':{'w_fakequantize':'PACTFakeQuantize'} + # 'concrete_args':{'augment':False, 'profile':False, 'visualize':False} + } + + # print('named_modules:', dict(model.named_modules())['']) + model.train() + model = model.to(device) + model = prepare_by_platform(model, backend, prepare_custom_config_dict) + # print('prepared module:', model) + enable_calibration(model) + calibration_flag = True + model = model.to(device) + + for epoch in range(start_epoch, epochs): # epoch ------------------------------------------------------------------ + callbacks.run('on_train_epoch_start') + model.train() + + # Update image weights (optional, single-GPU only) + if opt.image_weights: + cw = model.class_weights.cpu().numpy() * (1 - maps) ** 2 / nc # class weights + iw = labels_to_image_weights(dataset.labels, nc=nc, class_weights=cw) # image weights + dataset.indices = random.choices(range(dataset.n), weights=iw, k=dataset.n) # rand weighted idx + + # Update mosaic border (optional) + # b = int(random.uniform(0.25 * imgsz, 0.75 * imgsz + gs) // gs * gs) + # dataset.mosaic_border = [b - imgsz, -b] # height, width borders + + mloss = torch.zeros(3, device=device) # mean losses + if RANK != -1: + train_loader.sampler.set_epoch(epoch) + pbar = enumerate(train_loader) + LOGGER.info(('\n' + '%10s' * 7) % ('Epoch', 'gpu_mem', 'box', 'obj', 'cls', 'labels', 'img_size')) + if RANK in {-1, 0}: + pbar = tqdm(pbar, total=nb, bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # progress bar + sample_size = nb//1000 + print('sample_size:', sample_size) + optimizer.zero_grad() + for i, (imgs, targets, paths, _) in pbar: # batch ------------------------------------------------------------- + if opt.fast_test and i % sample_size != 0: + continue + callbacks.run('on_train_batch_start') + ni = i + nb * epoch # number integrated batches (since train start) + imgs = imgs.to(device, non_blocking=True).float() / 255 # uint8 to float32, 0-255 to 0.0-1.0 + + # Warmup + if ni <= nw: + xi = [0, nw] # x interp + # compute_loss.gr = np.interp(ni, xi, [0.0, 1.0]) # iou loss ratio (obj_loss = 1.0 or iou) + accumulate = max(1, np.interp(ni, xi, [1, nbs / batch_size]).round()) + for j, x in enumerate(optimizer.param_groups): + # bias lr falls from 0.1 to lr0, all other lrs rise from 0.0 to lr0 + x['lr'] = np.interp(ni, xi, [hyp['warmup_bias_lr'] if j == 0 else 0.0, x['initial_lr'] * lf(epoch)]) + if 'momentum' in x: + x['momentum'] = np.interp(ni, xi, [hyp['warmup_momentum'], hyp['momentum']]) + + # Multi-scale + if opt.multi_scale: + sz = random.randrange(imgsz * 0.5, imgsz * 1.5 + gs) // gs * gs # size + sf = sz / max(imgs.shape[2:]) # scale factor + if sf != 1: + ns = [math.ceil(x * sf / gs) * gs for x in imgs.shape[2:]] # new shape (stretched to gs-multiple) + imgs = nn.functional.interpolate(imgs, size=ns, mode='bilinear', align_corners=False) + + # Forward + with torch.cuda.amp.autocast(amp): + pred = model(imgs) # forward + loss, loss_items = compute_loss(pred, targets.to(device)) # loss scaled by batch_size + # print(i, 'loss:', loss) + if RANK != -1: + loss *= WORLD_SIZE # gradient averaged between devices in DDP mode + if opt.quad: + loss *= 4. + + if opt.quantize: + if calibration_flag: + if i >= 50: + calibration_flag = False + model.zero_grad() + enable_quantization(model) + print('close calibration') + else: + print('calibration iter{}'.format(i)) + continue + + # Backward + scaler.scale(loss).backward() + + # Optimize - https://pytorch.org/docs/master/notes/amp_examples.html + if ni - last_opt_step >= accumulate: + scaler.unscale_(optimizer) # unscale gradients + torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=10.0) # clip gradients + scaler.step(optimizer) # optimizer.step + scaler.update() + optimizer.zero_grad() + if ema: + ema.update(model) + last_opt_step = ni + + # Log + if RANK in {-1, 0}: + mloss = (mloss * i + loss_items) / (i + 1) # update mean losses + mem = f'{torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0:.3g}G' # (GB) + pbar.set_description(('%10s' * 2 + '%10.4g' * 5) % + (f'{epoch}/{epochs - 1}', mem, *mloss, targets.shape[0], imgs.shape[-1])) + callbacks.run('on_train_batch_end', ni, model, imgs, targets, paths, plots) + if callbacks.stop_training: + return + # end batch ------------------------------------------------------------------------------------------------ + + # Scheduler + lr = [x['lr'] for x in optimizer.param_groups] # for loggers + scheduler.step() + + if RANK in {-1, 0}: + # mAP + callbacks.run('on_train_epoch_end', epoch=epoch) + ema.update_attr(model, include=['yaml', 'nc', 'hyp', 'names', 'stride', 'class_weights']) + final_epoch = (epoch + 1 == epochs) or stopper.possible_stop + if not noval or final_epoch: # Calculate mAP + results, maps, _ = val.run(data_dict, + batch_size=batch_size // WORLD_SIZE * 2, + imgsz=imgsz, + half=amp, + model=ema.ema, + single_cls=single_cls, + dataloader=val_loader, + save_dir=save_dir, + plots=False, + callbacks=callbacks, + compute_loss=compute_loss) + + # Update best mAP + fi = fitness(np.array(results).reshape(1, -1)) # weighted combination of [P, R, mAP@.5, mAP@.5-.95] + stop = stopper(epoch=epoch, fitness=fi) # early stop check + if fi > best_fitness: + best_fitness = fi + log_vals = list(mloss) + list(results) + lr + callbacks.run('on_fit_epoch_end', log_vals, epoch, best_fitness, fi) + + # Save model + if (not nosave) or (final_epoch and not evolve): # if save + ckpt = { + 'epoch': epoch, + 'best_fitness': best_fitness, + 'model': deepcopy(de_parallel(model)).half(), + 'ema': deepcopy(ema.ema).half(), + 'updates': ema.updates, + 'optimizer': optimizer.state_dict(), + 'wandb_id': loggers.wandb.wandb_run.id if loggers.wandb else None, + 'opt': vars(opt), + 'date': datetime.now().isoformat()} + + # Save last, best and delete + torch.save(ckpt, last) + if best_fitness == fi: + torch.save(ckpt, best) + if opt.save_period > 0 and epoch % opt.save_period == 0: + torch.save(ckpt, w / f'epoch{epoch}.pt') + del ckpt + callbacks.run('on_model_save', last, epoch, final_epoch, best_fitness, fi) + + # EarlyStopping + if RANK != -1: # if DDP training + broadcast_list = [stop if RANK == 0 else None] + dist.broadcast_object_list(broadcast_list, 0) # broadcast 'stop' to all ranks + if RANK != 0: + stop = broadcast_list[0] + if stop: + break # must break all DDP ranks + + if opt.quantize: + print(f'epoch{epoch} convert_deploy') + model_name = opt.cfg.split('/')[-1].split('.')[0] + output_dir = os.path.join(opt.output_path, model_name) + output_dir = os.path.join(output_dir, str(epoch)) + output_dir = os.path.join(output_dir, model_name) + os.system('mkdir -p {}'.format(output_dir)) + model2 = deepcopy(model) + convert_deploy(model2.eval(), backend, input_shape_dict={'data': [1, 3, opt.imgsz, opt.imgsz]}, + model_name='{}_mqmoble'.format(model_name), output_path=output_dir) + del model2 + + # end epoch ---------------------------------------------------------------------------------------------------- + # end training ----------------------------------------------------------------------------------------------------- + + if not opt.fast_test and RANK in {-1, 0}: + LOGGER.info(f'\n{epoch - start_epoch + 1} epochs completed in {(time.time() - t0) / 3600:.3f} hours.') + for f in last, best: + if f.exists(): + strip_optimizer(f) # strip optimizers + if f is best: + LOGGER.info(f'\nValidating {f}...') + results, _, _ = val.run( + data_dict, + batch_size=batch_size // WORLD_SIZE * 2, + imgsz=imgsz, + model=attempt_load(f, device).half(), + iou_thres=0.65 if is_coco else 0.60, # best pycocotools results at 0.65 + single_cls=single_cls, + dataloader=val_loader, + save_dir=save_dir, + save_json=is_coco, + verbose=True, + plots=plots, + callbacks=callbacks, + compute_loss=compute_loss) # val best model with plots + if is_coco: + callbacks.run('on_fit_epoch_end', list(mloss) + list(results) + lr, epoch, best_fitness, fi) + + callbacks.run('on_train_end', last, best, plots, epoch, results) + + if opt.quantize: + model_name = opt.cfg.split('/')[-1].split('.')[0] + output_dir = os.path.join(opt.output_path, model_name) + os.system('mkdir -p {}'.format(output_dir)) + convert_deploy(model.eval(), backend, input_shape_dict={'data': [1, 3, opt.imgsz, opt.imgsz]}, + model_name='{}_mqmoble'.format(model_name), output_path=output_dir) + + torch.cuda.empty_cache() + return results + + +def parse_opt(known=False): + parser = argparse.ArgumentParser() + parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='initial weights path') + parser.add_argument('--cfg', type=str, default='', help='model.yaml path') + parser.add_argument('--output_path', type=str, default='./', help='output path') + parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='dataset.yaml path') + parser.add_argument('--hyp', type=str, default=ROOT / 'data/hyps/hyp.scratch-low.yaml', help='hyperparameters path') + parser.add_argument('--epochs', type=int, default=300) + parser.add_argument('--batch-size', type=int, default=16, help='total batch size for all GPUs, -1 for autobatch') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=640, help='train, val image size (pixels)') + parser.add_argument('--rect', action='store_true', help='rectangular training') + parser.add_argument('--resume', nargs='?', const=True, default=False, help='resume most recent training') + parser.add_argument('--nosave', action='store_true', help='only save final checkpoint') + parser.add_argument('--noval', action='store_true', help='only validate final epoch') + parser.add_argument('--noautoanchor', action='store_true', help='disable AutoAnchor') + parser.add_argument('--noplots', action='store_true', help='save no plot files') + parser.add_argument('--evolve', type=int, nargs='?', const=300, help='evolve hyperparameters for x generations') + parser.add_argument('--bucket', type=str, default='', help='gsutil bucket') + parser.add_argument('--cache', type=str, nargs='?', const='ram', help='--cache images in "ram" (default) or "disk"') + parser.add_argument('--image-weights', action='store_true', help='use weighted image selection for training') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--multi-scale', action='store_true', help='vary img-size +/- 50%%') + parser.add_argument('--single-cls', action='store_true', help='train multi-class data as single-class') + parser.add_argument('--optimizer', type=str, choices=['SGD', 'Adam', 'AdamW'], default='SGD', help='optimizer') + parser.add_argument('--sync-bn', action='store_true', help='use SyncBatchNorm, only available in DDP mode') + parser.add_argument('--workers', type=int, default=8, help='max dataloader workers (per RANK in DDP mode)') + parser.add_argument('--project', default=ROOT / 'runs/train', help='save to project/name') + parser.add_argument('--name', default='exp', help='save to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + parser.add_argument('--quad', action='store_true', help='quad dataloader') + parser.add_argument('--cos-lr', action='store_true', help='cosine LR scheduler') + parser.add_argument('--label-smoothing', type=float, default=0.0, help='Label smoothing epsilon') + parser.add_argument('--patience', type=int, default=100, help='EarlyStopping patience (epochs without improvement)') + parser.add_argument('--freeze', nargs='+', type=int, default=[0], help='Freeze layers: backbone=10, first3=0 1 2') + parser.add_argument('--save-period', type=int, default=-1, help='Save checkpoint every x epochs (disabled if < 1)') + parser.add_argument('--seed', type=int, default=0, help='Global training seed') + parser.add_argument('--local_rank', type=int, default=-1, help='Automatic DDP Multi-GPU argument, do not modify') + + # Weights & Biases arguments + parser.add_argument('--entity', default=None, help='W&B: Entity') + parser.add_argument('--upload_dataset', nargs='?', const=True, default=False, help='W&B: Upload data, "val" option') + parser.add_argument('--bbox_interval', type=int, default=-1, help='W&B: Set bounding-box image logging interval') + parser.add_argument('--artifact_alias', type=str, default='latest', help='W&B: Version of dataset artifact to use') + parser.add_argument('--quantize', action='store_true', help='quantize') + parser.add_argument('--pre_eval_and_export', action='store_true', help='pre_eval_and_export') + parser.add_argument('--fast_test', action='store_true', help='fast_test') + + return parser.parse_known_args()[0] if known else parser.parse_args() + + +def main(opt, callbacks=Callbacks()): + # Checks + if RANK in {-1, 0}: + print_args(vars(opt)) + check_git_status() + check_requirements() + + # Resume + if opt.resume and not (check_wandb_resume(opt) or opt.evolve): # resume from specified or most recent last.pt + last = Path(check_file(opt.resume) if isinstance(opt.resume, str) else get_latest_run()) + opt_yaml = last.parent.parent / 'opt.yaml' # train options yaml + opt_data = opt.data # original dataset + if opt_yaml.is_file(): + with open(opt_yaml, errors='ignore') as f: + d = yaml.safe_load(f) + else: + d = torch.load(last, map_location='cpu')['opt'] + opt = argparse.Namespace(**d) # replace + opt.cfg, opt.weights, opt.resume = '', str(last), True # reinstate + if is_url(opt_data): + opt.data = check_file(opt_data) # avoid HUB resume auth timeout + else: + opt.data, opt.cfg, opt.hyp, opt.weights, opt.project = \ + check_file(opt.data), check_yaml(opt.cfg), check_yaml(opt.hyp), str(opt.weights), str(opt.project) # checks + assert len(opt.cfg) or len(opt.weights), 'either --cfg or --weights must be specified' + if opt.evolve: + if opt.project == str(ROOT / 'runs/train'): # if default project name, rename to runs/evolve + opt.project = str(ROOT / 'runs/evolve') + opt.exist_ok, opt.resume = opt.resume, False # pass resume to exist_ok and disable resume + if opt.name == 'cfg': + opt.name = Path(opt.cfg).stem # use model.yaml as name + opt.save_dir = str(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok)) + + # DDP mode + device = select_device(opt.device, batch_size=opt.batch_size) + if LOCAL_RANK != -1: + msg = 'is not compatible with YOLOv5 Multi-GPU DDP training' + assert not opt.image_weights, f'--image-weights {msg}' + assert not opt.evolve, f'--evolve {msg}' + assert opt.batch_size != -1, f'AutoBatch with --batch-size -1 {msg}, please pass a valid --batch-size' + assert opt.batch_size % WORLD_SIZE == 0, f'--batch-size {opt.batch_size} must be multiple of WORLD_SIZE' + assert torch.cuda.device_count() > LOCAL_RANK, 'insufficient CUDA devices for DDP command' + torch.cuda.set_device(LOCAL_RANK) + device = torch.device('cuda', LOCAL_RANK) + dist.init_process_group(backend="nccl" if dist.is_nccl_available() else "gloo") + + # Train + if not opt.evolve: + train(opt.hyp, opt, device, callbacks) + + # Evolve hyperparameters (optional) + else: + # Hyperparameter evolution metadata (mutation scale 0-1, lower_limit, upper_limit) + meta = { + 'lr0': (1, 1e-5, 1e-1), # initial learning rate (SGD=1E-2, Adam=1E-3) + 'lrf': (1, 0.01, 1.0), # final OneCycleLR learning rate (lr0 * lrf) + 'momentum': (0.3, 0.6, 0.98), # SGD momentum/Adam beta1 + 'weight_decay': (1, 0.0, 0.001), # optimizer weight decay + 'warmup_epochs': (1, 0.0, 5.0), # warmup epochs (fractions ok) + 'warmup_momentum': (1, 0.0, 0.95), # warmup initial momentum + 'warmup_bias_lr': (1, 0.0, 0.2), # warmup initial bias lr + 'box': (1, 0.02, 0.2), # box loss gain + 'cls': (1, 0.2, 4.0), # cls loss gain + 'cls_pw': (1, 0.5, 2.0), # cls BCELoss positive_weight + 'obj': (1, 0.2, 4.0), # obj loss gain (scale with pixels) + 'obj_pw': (1, 0.5, 2.0), # obj BCELoss positive_weight + 'iou_t': (0, 0.1, 0.7), # IoU training threshold + 'anchor_t': (1, 2.0, 8.0), # anchor-multiple threshold + 'anchors': (2, 2.0, 10.0), # anchors per output grid (0 to ignore) + 'fl_gamma': (0, 0.0, 2.0), # focal loss gamma (efficientDet default gamma=1.5) + 'hsv_h': (1, 0.0, 0.1), # image HSV-Hue augmentation (fraction) + 'hsv_s': (1, 0.0, 0.9), # image HSV-Saturation augmentation (fraction) + 'hsv_v': (1, 0.0, 0.9), # image HSV-Value augmentation (fraction) + 'degrees': (1, 0.0, 45.0), # image rotation (+/- deg) + 'translate': (1, 0.0, 0.9), # image translation (+/- fraction) + 'scale': (1, 0.0, 0.9), # image scale (+/- gain) + 'shear': (1, 0.0, 10.0), # image shear (+/- deg) + 'perspective': (0, 0.0, 0.001), # image perspective (+/- fraction), range 0-0.001 + 'flipud': (1, 0.0, 1.0), # image flip up-down (probability) + 'fliplr': (0, 0.0, 1.0), # image flip left-right (probability) + 'mosaic': (1, 0.0, 1.0), # image mixup (probability) + 'mixup': (1, 0.0, 1.0), # image mixup (probability) + 'copy_paste': (1, 0.0, 1.0)} # segment copy-paste (probability) + + with open(opt.hyp, errors='ignore') as f: + hyp = yaml.safe_load(f) # load hyps dict + if 'anchors' not in hyp: # anchors commented in hyp.yaml + hyp['anchors'] = 3 + if opt.noautoanchor: + del hyp['anchors'], meta['anchors'] + opt.noval, opt.nosave, save_dir = True, True, Path(opt.save_dir) # only val/save final epoch + # ei = [isinstance(x, (int, float)) for x in hyp.values()] # evolvable indices + evolve_yaml, evolve_csv = save_dir / 'hyp_evolve.yaml', save_dir / 'evolve.csv' + if opt.bucket: + os.system(f'gsutil cp gs://{opt.bucket}/evolve.csv {evolve_csv}') # download evolve.csv if exists + + for _ in range(opt.evolve): # generations to evolve + if evolve_csv.exists(): # if evolve.csv exists: select best hyps and mutate + # Select parent(s) + parent = 'single' # parent selection method: 'single' or 'weighted' + x = np.loadtxt(evolve_csv, ndmin=2, delimiter=',', skiprows=1) + n = min(5, len(x)) # number of previous results to consider + x = x[np.argsort(-fitness(x))][:n] # top n mutations + w = fitness(x) - fitness(x).min() + 1E-6 # weights (sum > 0) + if parent == 'single' or len(x) == 1: + # x = x[random.randint(0, n - 1)] # random selection + x = x[random.choices(range(n), weights=w)[0]] # weighted selection + elif parent == 'weighted': + x = (x * w.reshape(n, 1)).sum(0) / w.sum() # weighted combination + + # Mutate + mp, s = 0.8, 0.2 # mutation probability, sigma + npr = np.random + npr.seed(int(time.time())) + g = np.array([meta[k][0] for k in hyp.keys()]) # gains 0-1 + ng = len(meta) + v = np.ones(ng) + while all(v == 1): # mutate until a change occurs (prevent duplicates) + v = (g * (npr.random(ng) < mp) * npr.randn(ng) * npr.random() * s + 1).clip(0.3, 3.0) + for i, k in enumerate(hyp.keys()): # plt.hist(v.ravel(), 300) + hyp[k] = float(x[i + 7] * v[i]) # mutate + + # Constrain to limits + for k, v in meta.items(): + hyp[k] = max(hyp[k], v[1]) # lower limit + hyp[k] = min(hyp[k], v[2]) # upper limit + hyp[k] = round(hyp[k], 5) # significant digits + + # Train mutation + results = train(hyp.copy(), opt, device, callbacks) + callbacks = Callbacks() + # Write mutation results + print_mutation(results, hyp.copy(), save_dir, opt.bucket) + + # Plot results + plot_evolve(evolve_csv) + LOGGER.info(f'Hyperparameter evolution finished {opt.evolve} generations\n' + f"Results saved to {colorstr('bold', save_dir)}\n" + f'Usage example: $ python train.py --hyp {evolve_yaml}') + + +def run(**kwargs): + # Usage: import train; train.run(data='coco128.yaml', imgsz=320, weights='yolov5m.pt') + opt = parse_opt(True) + for k, v in kwargs.items(): + setattr(opt, k, v) + main(opt) + return opt + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/tutorial.ipynb b/application/yolov5_example/tutorial.ipynb new file mode 100644 index 00000000..9fa338b1 --- /dev/null +++ b/application/yolov5_example/tutorial.ipynb @@ -0,0 +1,1151 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "colab_type": "text", + "id": "view-in-github" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "t6MPjfT5NrKQ" + }, + "source": [ + "\n", + "\n", + "\n", + "This is the **official YOLOv5 馃殌 notebook** by **Ultralytics**, and is freely available for redistribution under the [GPL-3.0 license](https://choosealicense.com/licenses/gpl-3.0/). \n", + "For more information please visit https://github.com/ultralytics/yolov5 and https://ultralytics.com. Thank you!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "7mGmQbAO5pQb" + }, + "source": [ + "# Setup\n", + "\n", + "Clone repo, install dependencies and check PyTorch and GPU." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "wbvMlHd_QwMG", + "outputId": "185d0979-edcd-4860-e6fb-b8a27dbf5096" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "YOLOv5 馃殌 v6.1-370-g20f1b7e Python-3.7.13 torch-1.12.0+cu113 CUDA:0 (Tesla V100-SXM2-16GB, 16160MiB)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Setup complete 鉁 (8 CPUs, 51.0 GB RAM, 37.4/166.8 GB disk)\n" + ] + } + ], + "source": [ + "!git clone https://github.com/ultralytics/yolov5 # clone\n", + "%cd yolov5\n", + "%pip install -qr requirements.txt # install\n", + "\n", + "import torch\n", + "import utils\n", + "display = utils.notebook_init() # checks" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4JnkELT0cIJg" + }, + "source": [ + "# 1. Inference\n", + "\n", + "`detect.py` runs YOLOv5 inference on a variety of sources, downloading models automatically from the [latest YOLOv5 release](https://github.com/ultralytics/yolov5/releases), and saving results to `runs/detect`. Example inference sources are:\n", + "\n", + "```shell\n", + "python detect.py --source 0 # webcam\n", + " img.jpg # image \n", + " vid.mp4 # video\n", + " path/ # directory\n", + " 'path/*.jpg' # glob\n", + " 'https://youtu.be/Zgi9g1ksQHc' # YouTube\n", + " 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream\n", + "```" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "zR9ZbuQCH7FX", + "outputId": "4b13989f-32a4-4ef0-b403-06ff3aac255c" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[34m\u001b[1mdetect: \u001b[0mweights=['yolov5s.pt'], source=data/images, data=data/coco128.yaml, imgsz=[640, 640], conf_thres=0.25, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=runs/detect, name=exp, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False\n", + "YOLOv5 馃殌 v6.1-370-g20f1b7e Python-3.7.13 torch-1.12.0+cu113 CUDA:0 (Tesla V100-SXM2-16GB, 16160MiB)\n", + "\n", + "Downloading https://github.com/ultralytics/yolov5/releases/download/v6.1/yolov5s.pt to yolov5s.pt...\n", + "100% 14.1M/14.1M [00:00<00:00, 53.9MB/s]\n", + "\n", + "Fusing layers... \n", + "YOLOv5s summary: 213 layers, 7225885 parameters, 0 gradients\n", + "image 1/2 /content/yolov5/data/images/bus.jpg: 640x480 4 persons, 1 bus, Done. (0.016s)\n", + "image 2/2 /content/yolov5/data/images/zidane.jpg: 384x640 2 persons, 2 ties, Done. (0.021s)\n", + "Speed: 0.6ms pre-process, 18.6ms inference, 25.0ms NMS per image at shape (1, 3, 640, 640)\n", + "Results saved to \u001b[1mruns/detect/exp\u001b[0m\n" + ] + } + ], + "source": [ + "!python detect.py --weights yolov5s.pt --img 640 --conf 0.25 --source data/images\n", + "#display.Image(filename='runs/detect/exp/zidane.jpg', width=600)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hkAzDWJ7cWTr" + }, + "source": [ + "        \n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "0eq1SMWl6Sfn" + }, + "source": [ + "# 2. Validate\n", + "Validate a model's accuracy on [COCO](https://cocodataset.org/#home) val or test-dev datasets. Models are downloaded automatically from the [latest YOLOv5 release](https://github.com/ultralytics/yolov5/releases). To show results by class use the `--verbose` flag. Note that `pycocotools` metrics may be ~1% better than the equivalent repo metrics, as is visible below, due to slight differences in mAP computation." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "eyTZYGgRjnMc" + }, + "source": [ + "## COCO val\n", + "Download [COCO val 2017](https://github.com/ultralytics/yolov5/blob/74b34872fdf41941cddcf243951cdb090fbac17b/data/coco.yaml#L14) dataset (1GB - 5000 images), and test model accuracy." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 49, + "referenced_widgets": [ + "c31d2039ccf74c22b67841f4877d1186", + "d4bba1727c714d94ad58a72bffa07c4c", + "9aeff9f1780b45f892422fdc96e56913", + "bf55a7c71d074d3fa88b10b997820825", + "d8b66044e2fb4f5b916696834d880c81", + "102e1deda239436fa72751c58202fa0f", + "4fd4431ced6c42368e18424912b877e4", + "cdd709c4f40941bea1b2053523c9fac8", + "a1ef2d8de2b741c78ca5d938e2ddbcdf", + "0dbce99bb6184238842cbec0587d564a", + "91ff5f93f2a24c5790ab29e347965946" + ] + }, + "id": "WQPtK1QYVaD_", + "outputId": "a9004b06-37a6-41ed-a1f2-ac956f3963b3" + }, + "outputs": [ + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "c31d2039ccf74c22b67841f4877d1186", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + " 0%| | 0.00/780M [00:00

\n", + "Close the active learning loop by sampling images from your inference conditions with the `roboflow` pip package\n", + "

\n", + "\n", + "Train a YOLOv5s model on the [COCO128](https://www.kaggle.com/ultralytics/coco128) dataset with `--data coco128.yaml`, starting from pretrained `--weights yolov5s.pt`, or from randomly initialized `--weights '' --cfg yolov5s.yaml`.\n", + "\n", + "- **Pretrained [Models](https://github.com/ultralytics/yolov5/tree/master/models)** are downloaded\n", + "automatically from the [latest YOLOv5 release](https://github.com/ultralytics/yolov5/releases)\n", + "- **[Datasets](https://github.com/ultralytics/yolov5/tree/master/data)** available for autodownload include: [COCO](https://github.com/ultralytics/yolov5/blob/master/data/coco.yaml), [COCO128](https://github.com/ultralytics/yolov5/blob/master/data/coco128.yaml), [VOC](https://github.com/ultralytics/yolov5/blob/master/data/VOC.yaml), [Argoverse](https://github.com/ultralytics/yolov5/blob/master/data/Argoverse.yaml), [VisDrone](https://github.com/ultralytics/yolov5/blob/master/data/VisDrone.yaml), [GlobalWheat](https://github.com/ultralytics/yolov5/blob/master/data/GlobalWheat2020.yaml), [xView](https://github.com/ultralytics/yolov5/blob/master/data/xView.yaml), [Objects365](https://github.com/ultralytics/yolov5/blob/master/data/Objects365.yaml), [SKU-110K](https://github.com/ultralytics/yolov5/blob/master/data/SKU-110K.yaml).\n", + "- **Training Results** are saved to `runs/train/` with incrementing run directories, i.e. `runs/train/exp2`, `runs/train/exp3` etc.\n", + "

\n", + "\n", + "A **Mosaic Dataloader** is used for training which combines 4 images into 1 mosaic.\n", + "\n", + "## Train on Custom Data with Roboflow 馃専 NEW\n", + "\n", + "[Roboflow](https://roboflow.com/?ref=ultralytics) enables you to easily **organize, label, and prepare** a high quality dataset with your own custom data. Roboflow also makes it easy to establish an active learning pipeline, collaborate with your team on dataset improvement, and integrate directly into your model building workflow with the `roboflow` pip package.\n", + "\n", + "- Custom Training Example: [https://blog.roboflow.com/how-to-train-yolov5-on-a-custom-dataset/](https://blog.roboflow.com/how-to-train-yolov5-on-a-custom-dataset/?ref=ultralytics)\n", + "- Custom Training Notebook: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/roboflow-ai/yolov5-custom-training-tutorial/blob/main/yolov5-custom-training.ipynb)\n", + "
\n", + "\n", + "

Label images lightning fast (including with model-assisted labeling)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "bOy5KI2ncnWd" + }, + "outputs": [], + "source": [ + "# Tensorboard (optional)\n", + "%load_ext tensorboard\n", + "%tensorboard --logdir runs/train" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "DQhI6vvaRWjR" + }, + "outputs": [], + "source": [ + "# ClearML (optional)\n", + "%pip install -q clearml\n", + "!clearml-init" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "2fLAV42oNb7M" + }, + "outputs": [], + "source": [ + "# Weights & Biases (optional)\n", + "%pip install -q wandb\n", + "import wandb\n", + "wandb.login()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1NcFxRcFdJ_O", + "outputId": "bce1b4bd-1a14-4c07-aebd-6c11e91ad24b" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[34m\u001b[1mtrain: \u001b[0mweights=yolov5s.pt, cfg=, data=coco128.yaml, hyp=data/hyps/hyp.scratch-low.yaml, epochs=3, batch_size=16, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, bucket=, cache=ram, image_weights=False, device=, multi_scale=False, single_cls=False, optimizer=SGD, sync_bn=False, workers=8, project=runs/train, name=exp, exist_ok=False, quad=False, cos_lr=False, label_smoothing=0.0, patience=100, freeze=[0], save_period=-1, seed=0, local_rank=-1, entity=None, upload_dataset=False, bbox_interval=-1, artifact_alias=latest\n", + "\u001b[34m\u001b[1mgithub: \u001b[0mup to date with https://github.com/ultralytics/yolov5 鉁匼n", + "YOLOv5 馃殌 v6.1-370-g20f1b7e Python-3.7.13 torch-1.12.0+cu113 CUDA:0 (Tesla V100-SXM2-16GB, 16160MiB)\n", + "\n", + "\u001b[34m\u001b[1mhyperparameters: \u001b[0mlr0=0.01, lrf=0.01, momentum=0.937, weight_decay=0.0005, warmup_epochs=3.0, warmup_momentum=0.8, warmup_bias_lr=0.1, box=0.05, cls=0.5, cls_pw=1.0, obj=1.0, obj_pw=1.0, iou_t=0.2, anchor_t=4.0, fl_gamma=0.0, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, degrees=0.0, translate=0.1, scale=0.5, shear=0.0, perspective=0.0, flipud=0.0, fliplr=0.5, mosaic=1.0, mixup=0.0, copy_paste=0.0\n", + "\u001b[34m\u001b[1mWeights & Biases: \u001b[0mrun 'pip install wandb' to automatically track and visualize YOLOv5 馃殌 runs in Weights & Biases\n", + "\u001b[34m\u001b[1mClearML: \u001b[0mrun 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 馃殌 runs in ClearML\n", + "\u001b[34m\u001b[1mTensorBoard: \u001b[0mStart with 'tensorboard --logdir runs/train', view at http://localhost:6006/\n", + "\n", + "Dataset not found 鈿狅笍, missing paths ['/content/datasets/coco128/images/train2017']\n", + "Downloading https://ultralytics.com/assets/coco128.zip to coco128.zip...\n", + "100% 6.66M/6.66M [00:00<00:00, 75.2MB/s]\n", + "Dataset download success 鉁 (0.7s), saved to \u001b[1m/content/datasets\u001b[0m\n", + "\n", + " from n params module arguments \n", + " 0 -1 1 3520 models.common.Conv [3, 32, 6, 2, 2] \n", + " 1 -1 1 18560 models.common.Conv [32, 64, 3, 2] \n", + " 2 -1 1 18816 models.common.C3 [64, 64, 1] \n", + " 3 -1 1 73984 models.common.Conv [64, 128, 3, 2] \n", + " 4 -1 2 115712 models.common.C3 [128, 128, 2] \n", + " 5 -1 1 295424 models.common.Conv [128, 256, 3, 2] \n", + " 6 -1 3 625152 models.common.C3 [256, 256, 3] \n", + " 7 -1 1 1180672 models.common.Conv [256, 512, 3, 2] \n", + " 8 -1 1 1182720 models.common.C3 [512, 512, 1] \n", + " 9 -1 1 656896 models.common.SPPF [512, 512, 5] \n", + " 10 -1 1 131584 models.common.Conv [512, 256, 1, 1] \n", + " 11 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] \n", + " 12 [-1, 6] 1 0 models.common.Concat [1] \n", + " 13 -1 1 361984 models.common.C3 [512, 256, 1, False] \n", + " 14 -1 1 33024 models.common.Conv [256, 128, 1, 1] \n", + " 15 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] \n", + " 16 [-1, 4] 1 0 models.common.Concat [1] \n", + " 17 -1 1 90880 models.common.C3 [256, 128, 1, False] \n", + " 18 -1 1 147712 models.common.Conv [128, 128, 3, 2] \n", + " 19 [-1, 14] 1 0 models.common.Concat [1] \n", + " 20 -1 1 296448 models.common.C3 [256, 256, 1, False] \n", + " 21 -1 1 590336 models.common.Conv [256, 256, 3, 2] \n", + " 22 [-1, 10] 1 0 models.common.Concat [1] \n", + " 23 -1 1 1182720 models.common.C3 [512, 512, 1, False] \n", + " 24 [17, 20, 23] 1 229245 models.yolo.Detect [80, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]\n", + "Model summary: 270 layers, 7235389 parameters, 7235389 gradients, 16.6 GFLOPs\n", + "\n", + "Transferred 349/349 items from yolov5s.pt\n", + "\u001b[34m\u001b[1mAMP: \u001b[0mchecks passed 鉁匼n", + "\u001b[34m\u001b[1moptimizer:\u001b[0m SGD(lr=0.01) with parameter groups 57 weight(decay=0.0), 60 weight(decay=0.0005), 60 bias\n", + "\u001b[34m\u001b[1malbumentations: \u001b[0mBlur(always_apply=False, p=0.01, blur_limit=(3, 7)), MedianBlur(always_apply=False, p=0.01, blur_limit=(3, 7)), ToGray(always_apply=False, p=0.01), CLAHE(always_apply=False, p=0.01, clip_limit=(1, 4.0), tile_grid_size=(8, 8))\n", + "\u001b[34m\u001b[1mtrain: \u001b[0mScanning '/content/datasets/coco128/labels/train2017' images and labels...128 found, 0 missing, 2 empty, 0 corrupt: 100% 128/128 [00:00<00:00, 7926.40it/s]\n", + "\u001b[34m\u001b[1mtrain: \u001b[0mNew cache created: /content/datasets/coco128/labels/train2017.cache\n", + "\u001b[34m\u001b[1mtrain: \u001b[0mCaching images (0.1GB ram): 100% 128/128 [00:00<00:00, 975.81it/s]\n", + "\u001b[34m\u001b[1mval: \u001b[0mScanning '/content/datasets/coco128/labels/train2017.cache' images and labels... 128 found, 0 missing, 2 empty, 0 corrupt: 100% 128/128 [00:00\n", + "\"ClearML" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DLI1JmHU7B0l" + }, + "source": [ + "## Weights & Biases Logging\n", + "\n", + "[Weights & Biases](https://wandb.ai/site?utm_campaign=repo_yolo_notebook) (W&B) is integrated with YOLOv5 for real-time visualization and cloud logging of training runs. This allows for better run comparison and introspection, as well improved visibility and collaboration for teams. To enable W&B `pip install wandb`, and then train normally (you will be guided through setup on first use). \n", + "\n", + "During training you will see live updates at [https://wandb.ai/home](https://wandb.ai/home?utm_campaign=repo_yolo_notebook), and you can create and share detailed [Reports](https://wandb.ai/glenn-jocher/yolov5_tutorial/reports/YOLOv5-COCO128-Tutorial-Results--VmlldzozMDI5OTY) of your results. For more information see the [YOLOv5 Weights & Biases Tutorial](https://github.com/ultralytics/yolov5/issues/1289). \n", + "\n", + "\n", + "\"Weights" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-WPvRbS5Swl6" + }, + "source": [ + "## Local Logging\n", + "\n", + "Training results are automatically logged with [Tensorboard](https://www.tensorflow.org/tensorboard) and [CSV](https://github.com/ultralytics/yolov5/pull/4148) loggers to `runs/train`, with a new experiment directory created for each new training as `runs/train/exp2`, `runs/train/exp3`, etc.\n", + "\n", + "This directory contains train and val statistics, mosaics, labels, predictions and augmentated mosaics, as well as metrics and charts including precision-recall (PR) curves and confusion matrices. \n", + "\n", + "\"Local\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Zelyeqbyt3GD" + }, + "source": [ + "# Environments\n", + "\n", + "YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including [CUDA](https://developer.nvidia.com/cuda)/[CUDNN](https://developer.nvidia.com/cudnn), [Python](https://www.python.org/) and [PyTorch](https://pytorch.org/) preinstalled):\n", + "\n", + "- **Google Colab and Kaggle** notebooks with free GPU: \"Open \"Open\n", + "- **Google Cloud** Deep Learning VM. See [GCP Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/GCP-Quickstart)\n", + "- **Amazon** Deep Learning AMI. See [AWS Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/AWS-Quickstart)\n", + "- **Docker Image**. See [Docker Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/Docker-Quickstart) \"Docker\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6Qu7Iesl0p54" + }, + "source": [ + "# Status\n", + "\n", + "![CI CPU testing](https://github.com/ultralytics/yolov5/workflows/CI%20CPU%20testing/badge.svg)\n", + "\n", + "If this badge is green, all [YOLOv5 GitHub Actions](https://github.com/ultralytics/yolov5/actions) Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training ([train.py](https://github.com/ultralytics/yolov5/blob/master/train.py)), testing ([val.py](https://github.com/ultralytics/yolov5/blob/master/val.py)), inference ([detect.py](https://github.com/ultralytics/yolov5/blob/master/detect.py)) and export ([export.py](https://github.com/ultralytics/yolov5/blob/master/export.py)) on macOS, Windows, and Ubuntu every 24 hours and on every commit.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IEijrePND_2I" + }, + "source": [ + "# Appendix\n", + "\n", + "Additional content below for PyTorch Hub, CI, reproducing results, profiling speeds, VOC training, classification training and TensorRT example." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "GMusP4OAxFu6" + }, + "outputs": [], + "source": [ + "import torch\n", + "\n", + "# PyTorch Hub Model\n", + "model = torch.hub.load('ultralytics/yolov5', 'yolov5s') # or yolov5n - yolov5x6, custom\n", + "\n", + "# Images\n", + "img = 'https://ultralytics.com/images/zidane.jpg' # or file, Path, PIL, OpenCV, numpy, list\n", + "\n", + "# Inference\n", + "results = model(img)\n", + "\n", + "# Results\n", + "results.print() # or .show(), .save(), .crop(), .pandas(), etc." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "FGH0ZjkGjejy" + }, + "outputs": [], + "source": [ + "# YOLOv5 CI\n", + "%%shell\n", + "rm -rf runs # remove runs/\n", + "m=yolov5n # official weights\n", + "b=runs/train/exp/weights/best # best.pt checkpoint\n", + "python train.py --imgsz 64 --batch 32 --weights $m.pt --cfg $m.yaml --epochs 1 --device 0 # train\n", + "for d in 0 cpu; do # devices\n", + " for w in $m $b; do # weights\n", + " python val.py --imgsz 64 --batch 32 --weights $w.pt --device $d # val\n", + " python detect.py --imgsz 64 --weights $w.pt --device $d # detect\n", + " done\n", + "done\n", + "python hubconf.py --model $m # hub\n", + "python models/tf.py --weights $m.pt # build TF model\n", + "python models/yolo.py --cfg $m.yaml # build PyTorch model\n", + "python export.py --weights $m.pt --img 64 --include torchscript # export" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mcKoSIK2WSzj" + }, + "outputs": [], + "source": [ + "# Reproduce\n", + "for x in (f'yolov5{x}' for x in 'nsmlx'):\n", + " !python val.py --weights {x}.pt --data coco.yaml --img 640 --task speed # speed\n", + " !python val.py --weights {x}.pt --data coco.yaml --img 640 --conf 0.001 --iou 0.65 # mAP" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gogI-kwi3Tye" + }, + "outputs": [], + "source": [ + "# Profile\n", + "from utils.torch_utils import profile\n", + "\n", + "m1 = lambda x: x * torch.sigmoid(x)\n", + "m2 = torch.nn.SiLU()\n", + "results = profile(input=torch.randn(16, 3, 640, 640), ops=[m1, m2], n=100)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "BSgFCAcMbk1R" + }, + "outputs": [], + "source": [ + "# VOC\n", + "for b, m in zip([64, 64, 64, 32, 16], [f'yolov5{x}' for x in 'nsmlx']): # batch, model\n", + " !python train.py --batch {b} --weights {m}.pt --data VOC.yaml --epochs 50 --img 512 --hyp hyp.VOC.yaml --project VOC --name {m} --cache" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "UWGH7H6yakVl" + }, + "outputs": [], + "source": [ + "# Classification\n", + "for m in [*(f'yolov5{x}.pt' for x in 'nsmlx'), 'resnet50.pt', 'efficientnet_b0.pt']:\n", + " for d in 'mnist', 'fashion-mnist', 'cifar10', 'cifar100', 'imagenette160', 'imagenette320', 'imagenette', 'imagewoof160', 'imagewoof320', 'imagewoof':\n", + " !python classify/train.py --model {m} --data {d} --epochs 10 --project YOLOv5-cls --name {m}-{d}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "VTRwsvA9u7ln" + }, + "outputs": [], + "source": [ + "# TensorRT \n", + "!pip install -U nvidia-tensorrt --index-url https://pypi.ngc.nvidia.com # install\n", + "!python export.py --weights yolov5s.pt --include engine --imgsz 640 --device 0 # export\n", + "!python detect.py --weights yolov5s.engine --imgsz 640 --device 0 # inference" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "collapsed_sections": [], + "include_colab_link": true, + "machine_shape": "hm", + "name": "YOLOv5 Tutorial", + "provenance": [], + "toc_visible": true + }, + "kernelspec": { + "display_name": "Python 3.6.9 64-bit", + "language": "python", + "name": "python3" + }, + "language_info": { + "name": "python", + "version": "3.6.9" + }, + "vscode": { + "interpreter": { + "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6" + } + }, + "widgets": { + "application/vnd.jupyter.widget-state+json": { + "0dbce99bb6184238842cbec0587d564a": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "102e1deda239436fa72751c58202fa0f": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "4fd4431ced6c42368e18424912b877e4": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "91ff5f93f2a24c5790ab29e347965946": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "DescriptionStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "DescriptionStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "description_width": "" + } + }, + "9aeff9f1780b45f892422fdc96e56913": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "FloatProgressModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "FloatProgressModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "ProgressView", + "bar_style": "success", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_cdd709c4f40941bea1b2053523c9fac8", + "max": 818322941, + "min": 0, + "orientation": "horizontal", + "style": "IPY_MODEL_a1ef2d8de2b741c78ca5d938e2ddbcdf", + "value": 818322941 + } + }, + "a1ef2d8de2b741c78ca5d938e2ddbcdf": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "ProgressStyleModel", + "state": { + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "ProgressStyleModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "StyleView", + "bar_color": null, + "description_width": "" + } + }, + "bf55a7c71d074d3fa88b10b997820825": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_0dbce99bb6184238842cbec0587d564a", + "placeholder": "鈥", + "style": "IPY_MODEL_91ff5f93f2a24c5790ab29e347965946", + "value": " 780M/780M [01:10<00:00, 10.5MB/s]" + } + }, + "c31d2039ccf74c22b67841f4877d1186": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HBoxModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HBoxModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HBoxView", + "box_style": "", + "children": [ + "IPY_MODEL_d4bba1727c714d94ad58a72bffa07c4c", + "IPY_MODEL_9aeff9f1780b45f892422fdc96e56913", + "IPY_MODEL_bf55a7c71d074d3fa88b10b997820825" + ], + "layout": "IPY_MODEL_d8b66044e2fb4f5b916696834d880c81" + } + }, + "cdd709c4f40941bea1b2053523c9fac8": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + }, + "d4bba1727c714d94ad58a72bffa07c4c": { + "model_module": "@jupyter-widgets/controls", + "model_module_version": "1.5.0", + "model_name": "HTMLModel", + "state": { + "_dom_classes": [], + "_model_module": "@jupyter-widgets/controls", + "_model_module_version": "1.5.0", + "_model_name": "HTMLModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/controls", + "_view_module_version": "1.5.0", + "_view_name": "HTMLView", + "description": "", + "description_tooltip": null, + "layout": "IPY_MODEL_102e1deda239436fa72751c58202fa0f", + "placeholder": "鈥", + "style": "IPY_MODEL_4fd4431ced6c42368e18424912b877e4", + "value": "100%" + } + }, + "d8b66044e2fb4f5b916696834d880c81": { + "model_module": "@jupyter-widgets/base", + "model_module_version": "1.2.0", + "model_name": "LayoutModel", + "state": { + "_model_module": "@jupyter-widgets/base", + "_model_module_version": "1.2.0", + "_model_name": "LayoutModel", + "_view_count": null, + "_view_module": "@jupyter-widgets/base", + "_view_module_version": "1.2.0", + "_view_name": "LayoutView", + "align_content": null, + "align_items": null, + "align_self": null, + "border": null, + "bottom": null, + "display": null, + "flex": null, + "flex_flow": null, + "grid_area": null, + "grid_auto_columns": null, + "grid_auto_flow": null, + "grid_auto_rows": null, + "grid_column": null, + "grid_gap": null, + "grid_row": null, + "grid_template_areas": null, + "grid_template_columns": null, + "grid_template_rows": null, + "height": null, + "justify_content": null, + "justify_items": null, + "left": null, + "margin": null, + "max_height": null, + "max_width": null, + "min_height": null, + "min_width": null, + "object_fit": null, + "object_position": null, + "order": null, + "overflow": null, + "overflow_x": null, + "overflow_y": null, + "padding": null, + "right": null, + "top": null, + "visibility": null, + "width": null + } + } + } + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/application/yolov5_example/utils/__init__.py b/application/yolov5_example/utils/__init__.py new file mode 100644 index 00000000..da53a4d2 --- /dev/null +++ b/application/yolov5_example/utils/__init__.py @@ -0,0 +1,36 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +utils/initialization +""" + + +def notebook_init(verbose=True): + # Check system software and hardware + print('Checking setup...') + + import os + import shutil + + from utils.general import check_requirements, emojis, is_colab + from utils.torch_utils import select_device # imports + + check_requirements(('psutil', 'IPython')) + import psutil + from IPython import display # to display images and clear console output + + if is_colab(): + shutil.rmtree('/content/sample_data', ignore_errors=True) # remove colab /sample_data directory + + # System info + if verbose: + gb = 1 << 30 # bytes to GiB (1024 ** 3) + ram = psutil.virtual_memory().total + total, used, free = shutil.disk_usage("/") + display.clear_output() + s = f'({os.cpu_count()} CPUs, {ram / gb:.1f} GB RAM, {(total - free) / gb:.1f}/{total / gb:.1f} GB disk)' + else: + s = '' + + select_device(newline=False) + print(emojis(f'Setup complete 鉁 {s}')) + return display diff --git a/application/yolov5_example/utils/activations.py b/application/yolov5_example/utils/activations.py new file mode 100644 index 00000000..084ce8c4 --- /dev/null +++ b/application/yolov5_example/utils/activations.py @@ -0,0 +1,103 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Activation functions +""" + +import torch +import torch.nn as nn +import torch.nn.functional as F + + +class SiLU(nn.Module): + # SiLU activation https://arxiv.org/pdf/1606.08415.pdf + @staticmethod + def forward(x): + return x * torch.sigmoid(x) + + +class Hardswish(nn.Module): + # Hard-SiLU activation + @staticmethod + def forward(x): + # return x * F.hardsigmoid(x) # for TorchScript and CoreML + return x * F.hardtanh(x + 3, 0.0, 6.0) / 6.0 # for TorchScript, CoreML and ONNX + + +class Mish(nn.Module): + # Mish activation https://github.com/digantamisra98/Mish + @staticmethod + def forward(x): + return x * F.softplus(x).tanh() + + +class MemoryEfficientMish(nn.Module): + # Mish activation memory-efficient + class F(torch.autograd.Function): + + @staticmethod + def forward(ctx, x): + ctx.save_for_backward(x) + return x.mul(torch.tanh(F.softplus(x))) # x * tanh(ln(1 + exp(x))) + + @staticmethod + def backward(ctx, grad_output): + x = ctx.saved_tensors[0] + sx = torch.sigmoid(x) + fx = F.softplus(x).tanh() + return grad_output * (fx + x * sx * (1 - fx * fx)) + + def forward(self, x): + return self.F.apply(x) + + +class FReLU(nn.Module): + # FReLU activation https://arxiv.org/abs/2007.11824 + def __init__(self, c1, k=3): # ch_in, kernel + super().__init__() + self.conv = nn.Conv2d(c1, c1, k, 1, 1, groups=c1, bias=False) + self.bn = nn.BatchNorm2d(c1) + + def forward(self, x): + return torch.max(x, self.bn(self.conv(x))) + + +class AconC(nn.Module): + r""" ACON activation (activate or not) + AconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is a learnable parameter + according to "Activate or Not: Learning Customized Activation" . + """ + + def __init__(self, c1): + super().__init__() + self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1)) + self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1)) + self.beta = nn.Parameter(torch.ones(1, c1, 1, 1)) + + def forward(self, x): + dpx = (self.p1 - self.p2) * x + return dpx * torch.sigmoid(self.beta * dpx) + self.p2 * x + + +class MetaAconC(nn.Module): + r""" ACON activation (activate or not) + MetaAconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is generated by a small network + according to "Activate or Not: Learning Customized Activation" . + """ + + def __init__(self, c1, k=1, s=1, r=16): # ch_in, kernel, stride, r + super().__init__() + c2 = max(r, c1 // r) + self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1)) + self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1)) + self.fc1 = nn.Conv2d(c1, c2, k, s, bias=True) + self.fc2 = nn.Conv2d(c2, c1, k, s, bias=True) + # self.bn1 = nn.BatchNorm2d(c2) + # self.bn2 = nn.BatchNorm2d(c1) + + def forward(self, x): + y = x.mean(dim=2, keepdims=True).mean(dim=3, keepdims=True) + # batch-size 1 bug/instabilities https://github.com/ultralytics/yolov5/issues/2891 + # beta = torch.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(y))))) # bug/unstable + beta = torch.sigmoid(self.fc2(self.fc1(y))) # bug patch BN layers removed + dpx = (self.p1 - self.p2) * x + return dpx * torch.sigmoid(beta * dpx) + self.p2 * x diff --git a/application/yolov5_example/utils/augmentations.py b/application/yolov5_example/utils/augmentations.py new file mode 100644 index 00000000..b00519ae --- /dev/null +++ b/application/yolov5_example/utils/augmentations.py @@ -0,0 +1,350 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Image augmentation functions +""" + +import math +import random + +import cv2 +import numpy as np +import torchvision.transforms as T +import torchvision.transforms.functional as TF + +from utils.general import LOGGER, check_version, colorstr, resample_segments, segment2box +from utils.metrics import bbox_ioa + +IMAGENET_MEAN = 0.485, 0.456, 0.406 # RGB mean +IMAGENET_STD = 0.229, 0.224, 0.225 # RGB standard deviation + + +class Albumentations: + # YOLOv5 Albumentations class (optional, only used if package is installed) + def __init__(self): + self.transform = None + prefix = colorstr('albumentations: ') + try: + import albumentations as A + check_version(A.__version__, '1.0.3', hard=True) # version requirement + + T = [ + A.Blur(p=0.01), + A.MedianBlur(p=0.01), + A.ToGray(p=0.01), + A.CLAHE(p=0.01), + A.RandomBrightnessContrast(p=0.0), + A.RandomGamma(p=0.0), + A.ImageCompression(quality_lower=75, p=0.0)] # transforms + self.transform = A.Compose(T, bbox_params=A.BboxParams(format='yolo', label_fields=['class_labels'])) + + LOGGER.info(prefix + ', '.join(f'{x}'.replace('always_apply=False, ', '') for x in T if x.p)) + except ImportError: # package not installed, skip + pass + except Exception as e: + LOGGER.info(f'{prefix}{e}') + + def __call__(self, im, labels, p=1.0): + if self.transform and random.random() < p: + new = self.transform(image=im, bboxes=labels[:, 1:], class_labels=labels[:, 0]) # transformed + im, labels = new['image'], np.array([[c, *b] for c, b in zip(new['class_labels'], new['bboxes'])]) + return im, labels + + +def normalize(x, mean=IMAGENET_MEAN, std=IMAGENET_STD, inplace=False): + # Denormalize RGB images x per ImageNet stats in BCHW format, i.e. = (x - mean) / std + return TF.normalize(x, mean, std, inplace=inplace) + + +def denormalize(x, mean=IMAGENET_MEAN, std=IMAGENET_STD): + # Denormalize RGB images x per ImageNet stats in BCHW format, i.e. = x * std + mean + for i in range(3): + x[:, i] = x[:, i] * std[i] + mean[i] + return x + + +def augment_hsv(im, hgain=0.5, sgain=0.5, vgain=0.5): + # HSV color-space augmentation + if hgain or sgain or vgain: + r = np.random.uniform(-1, 1, 3) * [hgain, sgain, vgain] + 1 # random gains + hue, sat, val = cv2.split(cv2.cvtColor(im, cv2.COLOR_BGR2HSV)) + dtype = im.dtype # uint8 + + x = np.arange(0, 256, dtype=r.dtype) + lut_hue = ((x * r[0]) % 180).astype(dtype) + lut_sat = np.clip(x * r[1], 0, 255).astype(dtype) + lut_val = np.clip(x * r[2], 0, 255).astype(dtype) + + im_hsv = cv2.merge((cv2.LUT(hue, lut_hue), cv2.LUT(sat, lut_sat), cv2.LUT(val, lut_val))) + cv2.cvtColor(im_hsv, cv2.COLOR_HSV2BGR, dst=im) # no return needed + + +def hist_equalize(im, clahe=True, bgr=False): + # Equalize histogram on BGR image 'im' with im.shape(n,m,3) and range 0-255 + yuv = cv2.cvtColor(im, cv2.COLOR_BGR2YUV if bgr else cv2.COLOR_RGB2YUV) + if clahe: + c = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8)) + yuv[:, :, 0] = c.apply(yuv[:, :, 0]) + else: + yuv[:, :, 0] = cv2.equalizeHist(yuv[:, :, 0]) # equalize Y channel histogram + return cv2.cvtColor(yuv, cv2.COLOR_YUV2BGR if bgr else cv2.COLOR_YUV2RGB) # convert YUV image to RGB + + +def replicate(im, labels): + # Replicate labels + h, w = im.shape[:2] + boxes = labels[:, 1:].astype(int) + x1, y1, x2, y2 = boxes.T + s = ((x2 - x1) + (y2 - y1)) / 2 # side length (pixels) + for i in s.argsort()[:round(s.size * 0.5)]: # smallest indices + x1b, y1b, x2b, y2b = boxes[i] + bh, bw = y2b - y1b, x2b - x1b + yc, xc = int(random.uniform(0, h - bh)), int(random.uniform(0, w - bw)) # offset x, y + x1a, y1a, x2a, y2a = [xc, yc, xc + bw, yc + bh] + im[y1a:y2a, x1a:x2a] = im[y1b:y2b, x1b:x2b] # im4[ymin:ymax, xmin:xmax] + labels = np.append(labels, [[labels[i, 0], x1a, y1a, x2a, y2a]], axis=0) + + return im, labels + + +def letterbox(im, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32): + # Resize and pad image while meeting stride-multiple constraints + shape = im.shape[:2] # current shape [height, width] + if isinstance(new_shape, int): + new_shape = (new_shape, new_shape) + + # Scale ratio (new / old) + r = min(new_shape[0] / shape[0], new_shape[1] / shape[1]) + if not scaleup: # only scale down, do not scale up (for better val mAP) + r = min(r, 1.0) + + # Compute padding + ratio = r, r # width, height ratios + new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r)) + dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding + if auto: # minimum rectangle + print('auto') + dw, dh = np.mod(dw, stride), np.mod(dh, stride) # wh padding + elif scaleFill: # stretch + print('stretch') + dw, dh = 0.0, 0.0 + new_unpad = (new_shape[1], new_shape[0]) + ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios + + dw /= 2 # divide padding into 2 sides + dh /= 2 + + if shape[::-1] != new_unpad: # resize + print('resize2') + im = cv2.resize(im, new_unpad, interpolation=cv2.INTER_LINEAR) + top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1)) + left, right = int(round(dw - 0.1)), int(round(dw + 0.1)) + im = cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border + return im, ratio, (dw, dh) + + +def random_perspective(im, + targets=(), + segments=(), + degrees=10, + translate=.1, + scale=.1, + shear=10, + perspective=0.0, + border=(0, 0)): + # torchvision.transforms.RandomAffine(degrees=(-10, 10), translate=(0.1, 0.1), scale=(0.9, 1.1), shear=(-10, 10)) + # targets = [cls, xyxy] + + height = im.shape[0] + border[0] * 2 # shape(h,w,c) + width = im.shape[1] + border[1] * 2 + + # Center + C = np.eye(3) + C[0, 2] = -im.shape[1] / 2 # x translation (pixels) + C[1, 2] = -im.shape[0] / 2 # y translation (pixels) + + # Perspective + P = np.eye(3) + P[2, 0] = random.uniform(-perspective, perspective) # x perspective (about y) + P[2, 1] = random.uniform(-perspective, perspective) # y perspective (about x) + + # Rotation and Scale + R = np.eye(3) + a = random.uniform(-degrees, degrees) + # a += random.choice([-180, -90, 0, 90]) # add 90deg rotations to small rotations + s = random.uniform(1 - scale, 1 + scale) + # s = 2 ** random.uniform(-scale, scale) + R[:2] = cv2.getRotationMatrix2D(angle=a, center=(0, 0), scale=s) + + # Shear + S = np.eye(3) + S[0, 1] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # x shear (deg) + S[1, 0] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # y shear (deg) + + # Translation + T = np.eye(3) + T[0, 2] = random.uniform(0.5 - translate, 0.5 + translate) * width # x translation (pixels) + T[1, 2] = random.uniform(0.5 - translate, 0.5 + translate) * height # y translation (pixels) + + # Combined rotation matrix + M = T @ S @ R @ P @ C # order of operations (right to left) is IMPORTANT + if (border[0] != 0) or (border[1] != 0) or (M != np.eye(3)).any(): # image changed + if perspective: + im = cv2.warpPerspective(im, M, dsize=(width, height), borderValue=(114, 114, 114)) + else: # affine + im = cv2.warpAffine(im, M[:2], dsize=(width, height), borderValue=(114, 114, 114)) + + # Visualize + # import matplotlib.pyplot as plt + # ax = plt.subplots(1, 2, figsize=(12, 6))[1].ravel() + # ax[0].imshow(im[:, :, ::-1]) # base + # ax[1].imshow(im2[:, :, ::-1]) # warped + + # Transform label coordinates + n = len(targets) + if n: + use_segments = any(x.any() for x in segments) + new = np.zeros((n, 4)) + if use_segments: # warp segments + segments = resample_segments(segments) # upsample + for i, segment in enumerate(segments): + xy = np.ones((len(segment), 3)) + xy[:, :2] = segment + xy = xy @ M.T # transform + xy = xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2] # perspective rescale or affine + + # clip + new[i] = segment2box(xy, width, height) + + else: # warp boxes + xy = np.ones((n * 4, 3)) + xy[:, :2] = targets[:, [1, 2, 3, 4, 1, 4, 3, 2]].reshape(n * 4, 2) # x1y1, x2y2, x1y2, x2y1 + xy = xy @ M.T # transform + xy = (xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2]).reshape(n, 8) # perspective rescale or affine + + # create new boxes + x = xy[:, [0, 2, 4, 6]] + y = xy[:, [1, 3, 5, 7]] + new = np.concatenate((x.min(1), y.min(1), x.max(1), y.max(1))).reshape(4, n).T + + # clip + new[:, [0, 2]] = new[:, [0, 2]].clip(0, width) + new[:, [1, 3]] = new[:, [1, 3]].clip(0, height) + + # filter candidates + i = box_candidates(box1=targets[:, 1:5].T * s, box2=new.T, area_thr=0.01 if use_segments else 0.10) + targets = targets[i] + targets[:, 1:5] = new[i] + + return im, targets + + +def copy_paste(im, labels, segments, p=0.5): + # Implement Copy-Paste augmentation https://arxiv.org/abs/2012.07177, labels as nx5 np.array(cls, xyxy) + n = len(segments) + if p and n: + h, w, c = im.shape # height, width, channels + im_new = np.zeros(im.shape, np.uint8) + for j in random.sample(range(n), k=round(p * n)): + l, s = labels[j], segments[j] + box = w - l[3], l[2], w - l[1], l[4] + ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area + if (ioa < 0.30).all(): # allow 30% obscuration of existing labels + labels = np.concatenate((labels, [[l[0], *box]]), 0) + segments.append(np.concatenate((w - s[:, 0:1], s[:, 1:2]), 1)) + cv2.drawContours(im_new, [segments[j].astype(np.int32)], -1, (255, 255, 255), cv2.FILLED) + + result = cv2.bitwise_and(src1=im, src2=im_new) + result = cv2.flip(result, 1) # augment segments (flip left-right) + i = result > 0 # pixels to replace + # i[:, :] = result.max(2).reshape(h, w, 1) # act over ch + im[i] = result[i] # cv2.imwrite('debug.jpg', im) # debug + + return im, labels, segments + + +def cutout(im, labels, p=0.5): + # Applies image cutout augmentation https://arxiv.org/abs/1708.04552 + if random.random() < p: + h, w = im.shape[:2] + scales = [0.5] * 1 + [0.25] * 2 + [0.125] * 4 + [0.0625] * 8 + [0.03125] * 16 # image size fraction + for s in scales: + mask_h = random.randint(1, int(h * s)) # create random masks + mask_w = random.randint(1, int(w * s)) + + # box + xmin = max(0, random.randint(0, w) - mask_w // 2) + ymin = max(0, random.randint(0, h) - mask_h // 2) + xmax = min(w, xmin + mask_w) + ymax = min(h, ymin + mask_h) + + # apply random color mask + im[ymin:ymax, xmin:xmax] = [random.randint(64, 191) for _ in range(3)] + + # return unobscured labels + if len(labels) and s > 0.03: + box = np.array([xmin, ymin, xmax, ymax], dtype=np.float32) + ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area + labels = labels[ioa < 0.60] # remove >60% obscured labels + + return labels + + +def mixup(im, labels, im2, labels2): + # Applies MixUp augmentation https://arxiv.org/pdf/1710.09412.pdf + r = np.random.beta(32.0, 32.0) # mixup ratio, alpha=beta=32.0 + im = (im * r + im2 * (1 - r)).astype(np.uint8) + labels = np.concatenate((labels, labels2), 0) + return im, labels + + +def box_candidates(box1, box2, wh_thr=2, ar_thr=100, area_thr=0.1, eps=1e-16): # box1(4,n), box2(4,n) + # Compute candidate boxes: box1 before augment, box2 after augment, wh_thr (pixels), aspect_ratio_thr, area_ratio + w1, h1 = box1[2] - box1[0], box1[3] - box1[1] + w2, h2 = box2[2] - box2[0], box2[3] - box2[1] + ar = np.maximum(w2 / (h2 + eps), h2 / (w2 + eps)) # aspect ratio + return (w2 > wh_thr) & (h2 > wh_thr) & (w2 * h2 / (w1 * h1 + eps) > area_thr) & (ar < ar_thr) # candidates + + +def classify_albumentations(augment=True, + size=224, + scale=(0.08, 1.0), + hflip=0.5, + vflip=0.0, + jitter=0.4, + mean=IMAGENET_MEAN, + std=IMAGENET_STD, + auto_aug=False): + # YOLOv5 classification Albumentations (optional, only used if package is installed) + prefix = colorstr('albumentations: ') + try: + import albumentations as A + from albumentations.pytorch import ToTensorV2 + check_version(A.__version__, '1.0.3', hard=True) # version requirement + if augment: # Resize and crop + T = [A.RandomResizedCrop(height=size, width=size, scale=scale)] + if auto_aug: + # TODO: implement AugMix, AutoAug & RandAug in albumentation + LOGGER.info(f'{prefix}auto augmentations are currently not supported') + else: + if hflip > 0: + T += [A.HorizontalFlip(p=hflip)] + if vflip > 0: + T += [A.VerticalFlip(p=vflip)] + if jitter > 0: + color_jitter = (float(jitter),) * 3 # repeat value for brightness, contrast, satuaration, 0 hue + T += [A.ColorJitter(*color_jitter, 0)] + else: # Use fixed crop for eval set (reproducibility) + T = [A.SmallestMaxSize(max_size=size), A.CenterCrop(height=size, width=size)] + T += [A.Normalize(mean=mean, std=std), ToTensorV2()] # Normalize and convert to Tensor + LOGGER.info(prefix + ', '.join(f'{x}'.replace('always_apply=False, ', '') for x in T if x.p)) + return A.Compose(T) + + except ImportError: # package not installed, skip + pass + except Exception as e: + LOGGER.info(f'{prefix}{e}') + + +def classify_transforms(size=224): + # Transforms to apply if albumentations not installed + return T.Compose([T.ToTensor(), T.Resize(size), T.CenterCrop(size), T.Normalize(IMAGENET_MEAN, IMAGENET_STD)]) diff --git a/application/yolov5_example/utils/autoanchor.py b/application/yolov5_example/utils/autoanchor.py new file mode 100644 index 00000000..f2222203 --- /dev/null +++ b/application/yolov5_example/utils/autoanchor.py @@ -0,0 +1,170 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +AutoAnchor utils +""" + +import random + +import numpy as np +import torch +import yaml +from tqdm import tqdm + +from utils.general import LOGGER, colorstr + +PREFIX = colorstr('AutoAnchor: ') + + +def check_anchor_order(m): + # Check anchor order against stride order for YOLOv5 Detect() module m, and correct if necessary + a = m.anchors.prod(-1).mean(-1).view(-1) # mean anchor area per output layer + da = a[-1] - a[0] # delta a + ds = m.stride[-1] - m.stride[0] # delta s + if da and (da.sign() != ds.sign()): # same order + LOGGER.info(f'{PREFIX}Reversing anchor order') + m.anchors[:] = m.anchors.flip(0) + + +def check_anchors(dataset, model, thr=4.0, imgsz=640): + # Check anchor fit to data, recompute if necessary + m = model.module.model[-1] if hasattr(model, 'module') else model.model[-1] # Detect() + shapes = imgsz * dataset.shapes / dataset.shapes.max(1, keepdims=True) + scale = np.random.uniform(0.9, 1.1, size=(shapes.shape[0], 1)) # augment scale + wh = torch.tensor(np.concatenate([l[:, 3:5] * s for s, l in zip(shapes * scale, dataset.labels)])).float() # wh + + def metric(k): # compute metric + r = wh[:, None] / k[None] + x = torch.min(r, 1 / r).min(2)[0] # ratio metric + best = x.max(1)[0] # best_x + aat = (x > 1 / thr).float().sum(1).mean() # anchors above threshold + bpr = (best > 1 / thr).float().mean() # best possible recall + return bpr, aat + + stride = m.stride.to(m.anchors.device).view(-1, 1, 1) # model strides + anchors = m.anchors.clone() * stride # current anchors + bpr, aat = metric(anchors.cpu().view(-1, 2)) + s = f'\n{PREFIX}{aat:.2f} anchors/target, {bpr:.3f} Best Possible Recall (BPR). ' + if bpr > 0.98: # threshold to recompute + LOGGER.info(f'{s}Current anchors are a good fit to dataset 鉁') + else: + LOGGER.info(f'{s}Anchors are a poor fit to dataset 鈿狅笍, attempting to improve...') + na = m.anchors.numel() // 2 # number of anchors + try: + anchors = kmean_anchors(dataset, n=na, img_size=imgsz, thr=thr, gen=1000, verbose=False) + except Exception as e: + LOGGER.info(f'{PREFIX}ERROR: {e}') + new_bpr = metric(anchors)[0] + if new_bpr > bpr: # replace anchors + anchors = torch.tensor(anchors, device=m.anchors.device).type_as(m.anchors) + m.anchors[:] = anchors.clone().view_as(m.anchors) + check_anchor_order(m) # must be in pixel-space (not grid-space) + m.anchors /= stride + s = f'{PREFIX}Done 鉁 (optional: update model *.yaml to use these anchors in the future)' + else: + s = f'{PREFIX}Done 鈿狅笍 (original anchors better than new anchors, proceeding with original anchors)' + LOGGER.info(s) + + +def kmean_anchors(dataset='./data/coco128.yaml', n=9, img_size=640, thr=4.0, gen=1000, verbose=True): + """ Creates kmeans-evolved anchors from training dataset + + Arguments: + dataset: path to data.yaml, or a loaded dataset + n: number of anchors + img_size: image size used for training + thr: anchor-label wh ratio threshold hyperparameter hyp['anchor_t'] used for training, default=4.0 + gen: generations to evolve anchors using genetic algorithm + verbose: print all results + + Return: + k: kmeans evolved anchors + + Usage: + from utils.autoanchor import *; _ = kmean_anchors() + """ + from scipy.cluster.vq import kmeans + + npr = np.random + thr = 1 / thr + + def metric(k, wh): # compute metrics + r = wh[:, None] / k[None] + x = torch.min(r, 1 / r).min(2)[0] # ratio metric + # x = wh_iou(wh, torch.tensor(k)) # iou metric + return x, x.max(1)[0] # x, best_x + + def anchor_fitness(k): # mutation fitness + _, best = metric(torch.tensor(k, dtype=torch.float32), wh) + return (best * (best > thr).float()).mean() # fitness + + def print_results(k, verbose=True): + k = k[np.argsort(k.prod(1))] # sort small to large + x, best = metric(k, wh0) + bpr, aat = (best > thr).float().mean(), (x > thr).float().mean() * n # best possible recall, anch > thr + s = f'{PREFIX}thr={thr:.2f}: {bpr:.4f} best possible recall, {aat:.2f} anchors past thr\n' \ + f'{PREFIX}n={n}, img_size={img_size}, metric_all={x.mean():.3f}/{best.mean():.3f}-mean/best, ' \ + f'past_thr={x[x > thr].mean():.3f}-mean: ' + for x in k: + s += '%i,%i, ' % (round(x[0]), round(x[1])) + if verbose: + LOGGER.info(s[:-2]) + return k + + if isinstance(dataset, str): # *.yaml file + with open(dataset, errors='ignore') as f: + data_dict = yaml.safe_load(f) # model dict + from utils.dataloaders import LoadImagesAndLabels + dataset = LoadImagesAndLabels(data_dict['train'], augment=True, rect=True) + + # Get label wh + shapes = img_size * dataset.shapes / dataset.shapes.max(1, keepdims=True) + wh0 = np.concatenate([l[:, 3:5] * s for s, l in zip(shapes, dataset.labels)]) # wh + + # Filter + i = (wh0 < 3.0).any(1).sum() + if i: + LOGGER.info(f'{PREFIX}WARNING: Extremely small objects found: {i} of {len(wh0)} labels are < 3 pixels in size') + wh = wh0[(wh0 >= 2.0).any(1)] # filter > 2 pixels + # wh = wh * (npr.rand(wh.shape[0], 1) * 0.9 + 0.1) # multiply by random scale 0-1 + + # Kmeans init + try: + LOGGER.info(f'{PREFIX}Running kmeans for {n} anchors on {len(wh)} points...') + assert n <= len(wh) # apply overdetermined constraint + s = wh.std(0) # sigmas for whitening + k = kmeans(wh / s, n, iter=30)[0] * s # points + assert n == len(k) # kmeans may return fewer points than requested if wh is insufficient or too similar + except Exception: + LOGGER.warning(f'{PREFIX}WARNING: switching strategies from kmeans to random init') + k = np.sort(npr.rand(n * 2)).reshape(n, 2) * img_size # random init + wh, wh0 = (torch.tensor(x, dtype=torch.float32) for x in (wh, wh0)) + k = print_results(k, verbose=False) + + # Plot + # k, d = [None] * 20, [None] * 20 + # for i in tqdm(range(1, 21)): + # k[i-1], d[i-1] = kmeans(wh / s, i) # points, mean distance + # fig, ax = plt.subplots(1, 2, figsize=(14, 7), tight_layout=True) + # ax = ax.ravel() + # ax[0].plot(np.arange(1, 21), np.array(d) ** 2, marker='.') + # fig, ax = plt.subplots(1, 2, figsize=(14, 7)) # plot wh + # ax[0].hist(wh[wh[:, 0]<100, 0],400) + # ax[1].hist(wh[wh[:, 1]<100, 1],400) + # fig.savefig('wh.png', dpi=200) + + # Evolve + f, sh, mp, s = anchor_fitness(k), k.shape, 0.9, 0.1 # fitness, generations, mutation prob, sigma + pbar = tqdm(range(gen), bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # progress bar + for _ in pbar: + v = np.ones(sh) + while (v == 1).all(): # mutate until a change occurs (prevent duplicates) + v = ((npr.random(sh) < mp) * random.random() * npr.randn(*sh) * s + 1).clip(0.3, 3.0) + kg = (k.copy() * v).clip(min=2.0) + fg = anchor_fitness(kg) + if fg > f: + f, k = fg, kg.copy() + pbar.desc = f'{PREFIX}Evolving anchors with Genetic Algorithm: fitness = {f:.4f}' + if verbose: + print_results(k, verbose) + + return print_results(k) diff --git a/application/yolov5_example/utils/autobatch.py b/application/yolov5_example/utils/autobatch.py new file mode 100644 index 00000000..c231d24c --- /dev/null +++ b/application/yolov5_example/utils/autobatch.py @@ -0,0 +1,66 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Auto-batch utils +""" + +from copy import deepcopy + +import numpy as np +import torch + +from utils.general import LOGGER, colorstr +from utils.torch_utils import profile + + +def check_train_batch_size(model, imgsz=640, amp=True): + # Check YOLOv5 training batch size + with torch.cuda.amp.autocast(amp): + return autobatch(deepcopy(model).train(), imgsz) # compute optimal batch size + + +def autobatch(model, imgsz=640, fraction=0.9, batch_size=16): + # Automatically estimate best batch size to use `fraction` of available CUDA memory + # Usage: + # import torch + # from utils.autobatch import autobatch + # model = torch.hub.load('ultralytics/yolov5', 'yolov5s', autoshape=False) + # print(autobatch(model)) + + # Check device + prefix = colorstr('AutoBatch: ') + LOGGER.info(f'{prefix}Computing optimal batch size for --imgsz {imgsz}') + device = next(model.parameters()).device # get model device + if device.type == 'cpu': + LOGGER.info(f'{prefix}CUDA not detected, using default CPU batch-size {batch_size}') + return batch_size + + # Inspect CUDA memory + gb = 1 << 30 # bytes to GiB (1024 ** 3) + d = str(device).upper() # 'CUDA:0' + properties = torch.cuda.get_device_properties(device) # device properties + t = properties.total_memory / gb # GiB total + r = torch.cuda.memory_reserved(device) / gb # GiB reserved + a = torch.cuda.memory_allocated(device) / gb # GiB allocated + f = t - (r + a) # GiB free + LOGGER.info(f'{prefix}{d} ({properties.name}) {t:.2f}G total, {r:.2f}G reserved, {a:.2f}G allocated, {f:.2f}G free') + + # Profile batch sizes + batch_sizes = [1, 2, 4, 8, 16] + try: + img = [torch.zeros(b, 3, imgsz, imgsz) for b in batch_sizes] + results = profile(img, model, n=3, device=device) + except Exception as e: + LOGGER.warning(f'{prefix}{e}') + + # Fit a solution + y = [x[2] for x in results if x] # memory [2] + p = np.polyfit(batch_sizes[:len(y)], y, deg=1) # first degree polynomial fit + b = int((f * fraction - p[1]) / p[0]) # y intercept (optimal batch size) + if None in results: # some sizes failed + i = results.index(None) # first fail index + if b >= batch_sizes[i]: # y intercept above failure point + b = batch_sizes[max(i - 1, 0)] # select prior safe point + + fraction = np.polyval(p, b) / t # actual fraction predicted + LOGGER.info(f'{prefix}Using batch-size {b} for {d} {t * fraction:.2f}G/{t:.2f}G ({fraction * 100:.0f}%) 鉁') + return b diff --git a/application/yolov5_example/utils/aws/__init__.py b/application/yolov5_example/utils/aws/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/application/yolov5_example/utils/aws/mime.sh b/application/yolov5_example/utils/aws/mime.sh new file mode 100644 index 00000000..c319a83c --- /dev/null +++ b/application/yolov5_example/utils/aws/mime.sh @@ -0,0 +1,26 @@ +# AWS EC2 instance startup 'MIME' script https://aws.amazon.com/premiumsupport/knowledge-center/execute-user-data-ec2/ +# This script will run on every instance restart, not only on first start +# --- DO NOT COPY ABOVE COMMENTS WHEN PASTING INTO USERDATA --- + +Content-Type: multipart/mixed; boundary="//" +MIME-Version: 1.0 + +--// +Content-Type: text/cloud-config; charset="us-ascii" +MIME-Version: 1.0 +Content-Transfer-Encoding: 7bit +Content-Disposition: attachment; filename="cloud-config.txt" + +#cloud-config +cloud_final_modules: +- [scripts-user, always] + +--// +Content-Type: text/x-shellscript; charset="us-ascii" +MIME-Version: 1.0 +Content-Transfer-Encoding: 7bit +Content-Disposition: attachment; filename="userdata.txt" + +#!/bin/bash +# --- paste contents of userdata.sh here --- +--// diff --git a/application/yolov5_example/utils/aws/resume.py b/application/yolov5_example/utils/aws/resume.py new file mode 100644 index 00000000..b21731c9 --- /dev/null +++ b/application/yolov5_example/utils/aws/resume.py @@ -0,0 +1,40 @@ +# Resume all interrupted trainings in yolov5/ dir including DDP trainings +# Usage: $ python utils/aws/resume.py + +import os +import sys +from pathlib import Path + +import torch +import yaml + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[2] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH + +port = 0 # --master_port +path = Path('').resolve() +for last in path.rglob('*/**/last.pt'): + ckpt = torch.load(last) + if ckpt['optimizer'] is None: + continue + + # Load opt.yaml + with open(last.parent.parent / 'opt.yaml', errors='ignore') as f: + opt = yaml.safe_load(f) + + # Get device count + d = opt['device'].split(',') # devices + nd = len(d) # number of devices + ddp = nd > 1 or (nd == 0 and torch.cuda.device_count() > 1) # distributed data parallel + + if ddp: # multi-GPU + port += 1 + cmd = f'python -m torch.distributed.run --nproc_per_node {nd} --master_port {port} train.py --resume {last}' + else: # single-GPU + cmd = f'python train.py --resume {last}' + + cmd += ' > /dev/null 2>&1 &' # redirect output to dev/null and run in daemon thread + print(cmd) + os.system(cmd) diff --git a/application/yolov5_example/utils/aws/userdata.sh b/application/yolov5_example/utils/aws/userdata.sh new file mode 100644 index 00000000..5fc1332a --- /dev/null +++ b/application/yolov5_example/utils/aws/userdata.sh @@ -0,0 +1,27 @@ +#!/bin/bash +# AWS EC2 instance startup script https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/user-data.html +# This script will run only once on first instance start (for a re-start script see mime.sh) +# /home/ubuntu (ubuntu) or /home/ec2-user (amazon-linux) is working dir +# Use >300 GB SSD + +cd home/ubuntu +if [ ! -d yolov5 ]; then + echo "Running first-time script." # install dependencies, download COCO, pull Docker + git clone https://github.com/ultralytics/yolov5 -b master && sudo chmod -R 777 yolov5 + cd yolov5 + bash data/scripts/get_coco.sh && echo "COCO done." & + sudo docker pull ultralytics/yolov5:latest && echo "Docker done." & + python -m pip install --upgrade pip && pip install -r requirements.txt && python detect.py && echo "Requirements done." & + wait && echo "All tasks done." # finish background tasks +else + echo "Running re-start script." # resume interrupted runs + i=0 + list=$(sudo docker ps -qa) # container list i.e. $'one\ntwo\nthree\nfour' + while IFS= read -r id; do + ((i++)) + echo "restarting container $i: $id" + sudo docker start $id + # sudo docker exec -it $id python train.py --resume # single-GPU + sudo docker exec -d $id python utils/aws/resume.py # multi-scenario + done <<<"$list" +fi diff --git a/application/yolov5_example/utils/benchmarks.py b/application/yolov5_example/utils/benchmarks.py new file mode 100644 index 00000000..d412653c --- /dev/null +++ b/application/yolov5_example/utils/benchmarks.py @@ -0,0 +1,157 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Run YOLOv5 benchmarks on all supported export formats + +Format | `export.py --include` | Model +--- | --- | --- +PyTorch | - | yolov5s.pt +TorchScript | `torchscript` | yolov5s.torchscript +ONNX | `onnx` | yolov5s.onnx +OpenVINO | `openvino` | yolov5s_openvino_model/ +TensorRT | `engine` | yolov5s.engine +CoreML | `coreml` | yolov5s.mlmodel +TensorFlow SavedModel | `saved_model` | yolov5s_saved_model/ +TensorFlow GraphDef | `pb` | yolov5s.pb +TensorFlow Lite | `tflite` | yolov5s.tflite +TensorFlow Edge TPU | `edgetpu` | yolov5s_edgetpu.tflite +TensorFlow.js | `tfjs` | yolov5s_web_model/ + +Requirements: + $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu # CPU + $ pip install -r requirements.txt coremltools onnx onnx-simplifier onnxruntime-gpu openvino-dev tensorflow # GPU + $ pip install -U nvidia-tensorrt --index-url https://pypi.ngc.nvidia.com # TensorRT + +Usage: + $ python utils/benchmarks.py --weights yolov5s.pt --img 640 +""" + +import argparse +import platform +import sys +import time +from pathlib import Path + +import pandas as pd + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +# ROOT = ROOT.relative_to(Path.cwd()) # relative + +import export +import val +from utils import notebook_init +from utils.general import LOGGER, check_yaml, file_size, print_args +from utils.torch_utils import select_device + + +def run( + weights=ROOT / 'yolov5s.pt', # weights path + imgsz=640, # inference size (pixels) + batch_size=1, # batch size + data=ROOT / 'data/coco128.yaml', # dataset.yaml path + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + half=False, # use FP16 half-precision inference + test=False, # test exports only + pt_only=False, # test PyTorch only + hard_fail=False, # throw error on benchmark failure +): + y, t = [], time.time() + device = select_device(device) + for i, (name, f, suffix, cpu, gpu) in export.export_formats().iterrows(): # index, (name, file, suffix, CPU, GPU) + try: + assert i not in (9, 10), 'inference not supported' # Edge TPU and TF.js are unsupported + assert i != 5 or platform.system() == 'Darwin', 'inference only supported on macOS>=10.13' # CoreML + if 'cpu' in device.type: + assert cpu, 'inference not supported on CPU' + if 'cuda' in device.type: + assert gpu, 'inference not supported on GPU' + + # Export + if f == '-': + w = weights # PyTorch format + else: + w = export.run(weights=weights, imgsz=[imgsz], include=[f], device=device, half=half)[-1] # all others + assert suffix in str(w), 'export failed' + + # Validate + result = val.run(data, w, batch_size, imgsz, plots=False, device=device, task='benchmark', half=half) + metrics = result[0] # metrics (mp, mr, map50, map, *losses(box, obj, cls)) + speeds = result[2] # times (preprocess, inference, postprocess) + y.append([name, round(file_size(w), 1), round(metrics[3], 4), round(speeds[1], 2)]) # MB, mAP, t_inference + except Exception as e: + if hard_fail: + assert type(e) is AssertionError, f'Benchmark --hard-fail for {name}: {e}' + LOGGER.warning(f'WARNING: Benchmark failure for {name}: {e}') + y.append([name, None, None, None]) # mAP, t_inference + if pt_only and i == 0: + break # break after PyTorch + + # Print results + LOGGER.info('\n') + parse_opt() + notebook_init() # print system info + c = ['Format', 'Size (MB)', 'mAP@0.5:0.95', 'Inference time (ms)'] if map else ['Format', 'Export', '', ''] + py = pd.DataFrame(y, columns=c) + LOGGER.info(f'\nBenchmarks complete ({time.time() - t:.2f}s)') + LOGGER.info(str(py if map else py.iloc[:, :2])) + return py + + +def test( + weights=ROOT / 'yolov5s.pt', # weights path + imgsz=640, # inference size (pixels) + batch_size=1, # batch size + data=ROOT / 'data/coco128.yaml', # dataset.yaml path + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + half=False, # use FP16 half-precision inference + test=False, # test exports only + pt_only=False, # test PyTorch only + hard_fail=False, # throw error on benchmark failure +): + y, t = [], time.time() + device = select_device(device) + for i, (name, f, suffix, gpu) in export.export_formats().iterrows(): # index, (name, file, suffix, gpu-capable) + try: + w = weights if f == '-' else \ + export.run(weights=weights, imgsz=[imgsz], include=[f], device=device, half=half)[-1] # weights + assert suffix in str(w), 'export failed' + y.append([name, True]) + except Exception: + y.append([name, False]) # mAP, t_inference + + # Print results + LOGGER.info('\n') + parse_opt() + notebook_init() # print system info + py = pd.DataFrame(y, columns=['Format', 'Export']) + LOGGER.info(f'\nExports complete ({time.time() - t:.2f}s)') + LOGGER.info(str(py)) + return py + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='weights path') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=640, help='inference size (pixels)') + parser.add_argument('--batch-size', type=int, default=1, help='batch size') + parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='dataset.yaml path') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') + parser.add_argument('--test', action='store_true', help='test exports only') + parser.add_argument('--pt-only', action='store_true', help='test PyTorch only') + parser.add_argument('--hard-fail', action='store_true', help='throw error on benchmark failure') + opt = parser.parse_args() + opt.data = check_yaml(opt.data) # check YAML + print_args(vars(opt)) + return opt + + +def main(opt): + test(**vars(opt)) if opt.test else run(**vars(opt)) + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/application/yolov5_example/utils/callbacks.py b/application/yolov5_example/utils/callbacks.py new file mode 100644 index 00000000..2b32df0b --- /dev/null +++ b/application/yolov5_example/utils/callbacks.py @@ -0,0 +1,71 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Callback utils +""" + + +class Callbacks: + """" + Handles all registered callbacks for YOLOv5 Hooks + """ + + def __init__(self): + # Define the available callbacks + self._callbacks = { + 'on_pretrain_routine_start': [], + 'on_pretrain_routine_end': [], + 'on_train_start': [], + 'on_train_epoch_start': [], + 'on_train_batch_start': [], + 'optimizer_step': [], + 'on_before_zero_grad': [], + 'on_train_batch_end': [], + 'on_train_epoch_end': [], + 'on_val_start': [], + 'on_val_batch_start': [], + 'on_val_image_end': [], + 'on_val_batch_end': [], + 'on_val_end': [], + 'on_fit_epoch_end': [], # fit = train + val + 'on_model_save': [], + 'on_train_end': [], + 'on_params_update': [], + 'teardown': [],} + self.stop_training = False # set True to interrupt training + + def register_action(self, hook, name='', callback=None): + """ + Register a new action to a callback hook + + Args: + hook: The callback hook name to register the action to + name: The name of the action for later reference + callback: The callback to fire + """ + assert hook in self._callbacks, f"hook '{hook}' not found in callbacks {self._callbacks}" + assert callable(callback), f"callback '{callback}' is not callable" + self._callbacks[hook].append({'name': name, 'callback': callback}) + + def get_registered_actions(self, hook=None): + """" + Returns all the registered actions by callback hook + + Args: + hook: The name of the hook to check, defaults to all + """ + return self._callbacks[hook] if hook else self._callbacks + + def run(self, hook, *args, **kwargs): + """ + Loop through the registered actions and fire all callbacks + + Args: + hook: The name of the hook to check, defaults to all + args: Arguments to receive from YOLOv5 + kwargs: Keyword Arguments to receive from YOLOv5 + """ + + assert hook in self._callbacks, f"hook '{hook}' not found in callbacks {self._callbacks}" + + for logger in self._callbacks[hook]: + logger['callback'](*args, **kwargs) diff --git a/application/yolov5_example/utils/dataloaders.py b/application/yolov5_example/utils/dataloaders.py new file mode 100644 index 00000000..184558fd --- /dev/null +++ b/application/yolov5_example/utils/dataloaders.py @@ -0,0 +1,1158 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Dataloaders and dataset utils +""" + +import contextlib +import glob +import hashlib +import json +import math +import os +import random +import shutil +import time +from itertools import repeat +from multiprocessing.pool import Pool, ThreadPool +from pathlib import Path +from threading import Thread +from urllib.parse import urlparse +from zipfile import ZipFile + +import numpy as np +import torch +import torch.nn.functional as F +import torchvision +import yaml +from PIL import ExifTags, Image, ImageOps +from torch.utils.data import DataLoader, Dataset, dataloader, distributed +from tqdm import tqdm + +from utils.augmentations import (Albumentations, augment_hsv, classify_albumentations, classify_transforms, copy_paste, + letterbox, mixup, random_perspective) +from utils.general import (DATASETS_DIR, LOGGER, NUM_THREADS, check_dataset, check_requirements, check_yaml, clean_str, + cv2, is_colab, is_kaggle, segments2boxes, xyn2xy, xywh2xyxy, xywhn2xyxy, xyxy2xywhn) +from utils.torch_utils import torch_distributed_zero_first + +# Parameters +HELP_URL = 'https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data' +IMG_FORMATS = 'bmp', 'dng', 'jpeg', 'jpg', 'mpo', 'png', 'tif', 'tiff', 'webp' # include image suffixes +VID_FORMATS = 'asf', 'avi', 'gif', 'm4v', 'mkv', 'mov', 'mp4', 'mpeg', 'mpg', 'ts', 'wmv' # include video suffixes +BAR_FORMAT = '{l_bar}{bar:10}{r_bar}{bar:-10b}' # tqdm bar format +LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html + +# Get orientation exif tag +for orientation in ExifTags.TAGS.keys(): + if ExifTags.TAGS[orientation] == 'Orientation': + break + + +def get_hash(paths): + # Returns a single hash value of a list of paths (files or dirs) + size = sum(os.path.getsize(p) for p in paths if os.path.exists(p)) # sizes + h = hashlib.md5(str(size).encode()) # hash sizes + h.update(''.join(paths).encode()) # hash paths + return h.hexdigest() # return hash + + +def exif_size(img): + # Returns exif-corrected PIL size + s = img.size # (width, height) + with contextlib.suppress(Exception): + rotation = dict(img._getexif().items())[orientation] + if rotation in [6, 8]: # rotation 270 or 90 + s = (s[1], s[0]) + return s + + +def exif_transpose(image): + """ + Transpose a PIL image accordingly if it has an EXIF Orientation tag. + Inplace version of https://github.com/python-pillow/Pillow/blob/master/src/PIL/ImageOps.py exif_transpose() + + :param image: The image to transpose. + :return: An image. + """ + exif = image.getexif() + orientation = exif.get(0x0112, 1) # default 1 + if orientation > 1: + method = { + 2: Image.FLIP_LEFT_RIGHT, + 3: Image.ROTATE_180, + 4: Image.FLIP_TOP_BOTTOM, + 5: Image.TRANSPOSE, + 6: Image.ROTATE_270, + 7: Image.TRANSVERSE, + 8: Image.ROTATE_90,}.get(orientation) + if method is not None: + image = image.transpose(method) + del exif[0x0112] + image.info["exif"] = exif.tobytes() + return image + + +def seed_worker(worker_id): + # Set dataloader worker seed https://pytorch.org/docs/stable/notes/randomness.html#dataloader + worker_seed = torch.initial_seed() % 2 ** 32 + np.random.seed(worker_seed) + random.seed(worker_seed) + + +def create_dataloader(path, + imgsz, + batch_size, + stride, + single_cls=False, + hyp=None, + augment=False, + cache=False, + pad=0.0, + rect=False, + rank=-1, + workers=8, + image_weights=False, + quad=False, + prefix='', + shuffle=False): + if rect and shuffle: + LOGGER.warning('WARNING: --rect is incompatible with DataLoader shuffle, setting shuffle=False') + shuffle = False + with torch_distributed_zero_first(rank): # init dataset *.cache only once if DDP + dataset = LoadImagesAndLabels( + path, + imgsz, + batch_size, + augment=augment, # augmentation + hyp=hyp, # hyperparameters + rect=rect, # rectangular batches + cache_images=cache, + single_cls=single_cls, + stride=int(stride), + pad=pad, + image_weights=image_weights, + prefix=prefix) + + batch_size = min(batch_size, len(dataset)) + nd = torch.cuda.device_count() # number of CUDA devices + nw = min([os.cpu_count() // max(nd, 1), batch_size if batch_size > 1 else 0, workers]) # number of workers + sampler = None if rank == -1 else distributed.DistributedSampler(dataset, shuffle=shuffle) + loader = DataLoader if image_weights else InfiniteDataLoader # only DataLoader allows for attribute updates + generator = torch.Generator() + generator.manual_seed(0) + return loader(dataset, + batch_size=batch_size, + shuffle=shuffle and sampler is None, + num_workers=nw, + sampler=sampler, + pin_memory=True, + collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn, + worker_init_fn=seed_worker, + generator=generator), dataset + + +class InfiniteDataLoader(dataloader.DataLoader): + """ Dataloader that reuses workers + + Uses same syntax as vanilla DataLoader + """ + + def __init__(self, *args, **kwargs): + super().__init__(*args, **kwargs) + object.__setattr__(self, 'batch_sampler', _RepeatSampler(self.batch_sampler)) + self.iterator = super().__iter__() + + def __len__(self): + return len(self.batch_sampler.sampler) + + def __iter__(self): + for _ in range(len(self)): + yield next(self.iterator) + + +class _RepeatSampler: + """ Sampler that repeats forever + + Args: + sampler (Sampler) + """ + + def __init__(self, sampler): + self.sampler = sampler + + def __iter__(self): + while True: + yield from iter(self.sampler) + + +class LoadImages: + # YOLOv5 image/video dataloader, i.e. `python detect.py --source image.jpg/vid.mp4` + def __init__(self, path, img_size=640, stride=32, auto=True): + files = [] + for p in sorted(path) if isinstance(path, (list, tuple)) else [path]: + p = str(Path(p).resolve()) + if '*' in p: + files.extend(sorted(glob.glob(p, recursive=True))) # glob + elif os.path.isdir(p): + files.extend(sorted(glob.glob(os.path.join(p, '*.*')))) # dir + elif os.path.isfile(p): + files.append(p) # files + else: + raise FileNotFoundError(f'{p} does not exist') + + images = [x for x in files if x.split('.')[-1].lower() in IMG_FORMATS] + videos = [x for x in files if x.split('.')[-1].lower() in VID_FORMATS] + ni, nv = len(images), len(videos) + + self.img_size = img_size + self.stride = stride + self.files = images + videos + self.nf = ni + nv # number of files + self.video_flag = [False] * ni + [True] * nv + self.mode = 'image' + self.auto = auto + if any(videos): + self.new_video(videos[0]) # new video + else: + self.cap = None + assert self.nf > 0, f'No images or videos found in {p}. ' \ + f'Supported formats are:\nimages: {IMG_FORMATS}\nvideos: {VID_FORMATS}' + + def __iter__(self): + self.count = 0 + return self + + def __next__(self): + if self.count == self.nf: + raise StopIteration + path = self.files[self.count] + + if self.video_flag[self.count]: + # Read video + self.mode = 'video' + ret_val, img0 = self.cap.read() + while not ret_val: + self.count += 1 + self.cap.release() + if self.count == self.nf: # last video + raise StopIteration + path = self.files[self.count] + self.new_video(path) + ret_val, img0 = self.cap.read() + + self.frame += 1 + s = f'video {self.count + 1}/{self.nf} ({self.frame}/{self.frames}) {path}: ' + + else: + # Read image + self.count += 1 + img0 = cv2.imread(path) # BGR + assert img0 is not None, f'Image Not Found {path}' + s = f'image {self.count}/{self.nf} {path}: ' + + # Padded resize + img = letterbox(img0, self.img_size, stride=self.stride, auto=self.auto)[0] + + # Convert + img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB + img = np.ascontiguousarray(img) + + return path, img, img0, self.cap, s + + def new_video(self, path): + self.frame = 0 + self.cap = cv2.VideoCapture(path) + self.frames = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT)) + + def __len__(self): + return self.nf # number of files + + +class LoadWebcam: # for inference + # YOLOv5 local webcam dataloader, i.e. `python detect.py --source 0` + def __init__(self, pipe='0', img_size=640, stride=32): + self.img_size = img_size + self.stride = stride + self.pipe = eval(pipe) if pipe.isnumeric() else pipe + self.cap = cv2.VideoCapture(self.pipe) # video capture object + self.cap.set(cv2.CAP_PROP_BUFFERSIZE, 3) # set buffer size + + def __iter__(self): + self.count = -1 + return self + + def __next__(self): + self.count += 1 + if cv2.waitKey(1) == ord('q'): # q to quit + self.cap.release() + cv2.destroyAllWindows() + raise StopIteration + + # Read frame + ret_val, img0 = self.cap.read() + img0 = cv2.flip(img0, 1) # flip left-right + + # Print + assert ret_val, f'Camera Error {self.pipe}' + img_path = 'webcam.jpg' + s = f'webcam {self.count}: ' + + # Padded resize + img = letterbox(img0, self.img_size, stride=self.stride)[0] + + # Convert + img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB + img = np.ascontiguousarray(img) + + return img_path, img, img0, None, s + + def __len__(self): + return 0 + + +class LoadStreams: + # YOLOv5 streamloader, i.e. `python detect.py --source 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP streams` + def __init__(self, sources='streams.txt', img_size=640, stride=32, auto=True): + self.mode = 'stream' + self.img_size = img_size + self.stride = stride + + if os.path.isfile(sources): + with open(sources) as f: + sources = [x.strip() for x in f.read().strip().splitlines() if len(x.strip())] + else: + sources = [sources] + + n = len(sources) + self.imgs, self.fps, self.frames, self.threads = [None] * n, [0] * n, [0] * n, [None] * n + self.sources = [clean_str(x) for x in sources] # clean source names for later + self.auto = auto + for i, s in enumerate(sources): # index, source + # Start thread to read frames from video stream + st = f'{i + 1}/{n}: {s}... ' + if urlparse(s).hostname in ('www.youtube.com', 'youtube.com', 'youtu.be'): # if source is YouTube video + check_requirements(('pafy', 'youtube_dl==2020.12.2')) + import pafy + s = pafy.new(s).getbest(preftype="mp4").url # YouTube URL + s = eval(s) if s.isnumeric() else s # i.e. s = '0' local webcam + if s == 0: + assert not is_colab(), '--source 0 webcam unsupported on Colab. Rerun command in a local environment.' + assert not is_kaggle(), '--source 0 webcam unsupported on Kaggle. Rerun command in a local environment.' + cap = cv2.VideoCapture(s) + assert cap.isOpened(), f'{st}Failed to open {s}' + w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) + h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) + fps = cap.get(cv2.CAP_PROP_FPS) # warning: may return 0 or nan + self.frames[i] = max(int(cap.get(cv2.CAP_PROP_FRAME_COUNT)), 0) or float('inf') # infinite stream fallback + self.fps[i] = max((fps if math.isfinite(fps) else 0) % 100, 0) or 30 # 30 FPS fallback + + _, self.imgs[i] = cap.read() # guarantee first frame + self.threads[i] = Thread(target=self.update, args=([i, cap, s]), daemon=True) + LOGGER.info(f"{st} Success ({self.frames[i]} frames {w}x{h} at {self.fps[i]:.2f} FPS)") + self.threads[i].start() + LOGGER.info('') # newline + + # check for common shapes + s = np.stack([letterbox(x, self.img_size, stride=self.stride, auto=self.auto)[0].shape for x in self.imgs]) + self.rect = np.unique(s, axis=0).shape[0] == 1 # rect inference if all shapes equal + if not self.rect: + LOGGER.warning('WARNING: Stream shapes differ. For optimal performance supply similarly-shaped streams.') + + def update(self, i, cap, stream): + # Read stream `i` frames in daemon thread + n, f, read = 0, self.frames[i], 1 # frame number, frame array, inference every 'read' frame + while cap.isOpened() and n < f: + n += 1 + # _, self.imgs[index] = cap.read() + cap.grab() + if n % read == 0: + success, im = cap.retrieve() + if success: + self.imgs[i] = im + else: + LOGGER.warning('WARNING: Video stream unresponsive, please check your IP camera connection.') + self.imgs[i] = np.zeros_like(self.imgs[i]) + cap.open(stream) # re-open stream if signal was lost + time.sleep(0.0) # wait time + + def __iter__(self): + self.count = -1 + return self + + def __next__(self): + self.count += 1 + if not all(x.is_alive() for x in self.threads) or cv2.waitKey(1) == ord('q'): # q to quit + cv2.destroyAllWindows() + raise StopIteration + + # Letterbox + img0 = self.imgs.copy() + img = [letterbox(x, self.img_size, stride=self.stride, auto=self.rect and self.auto)[0] for x in img0] + + # Stack + img = np.stack(img, 0) + + # Convert + img = img[..., ::-1].transpose((0, 3, 1, 2)) # BGR to RGB, BHWC to BCHW + img = np.ascontiguousarray(img) + + return self.sources, img, img0, None, '' + + def __len__(self): + return len(self.sources) # 1E12 frames = 32 streams at 30 FPS for 30 years + + +def img2label_paths(img_paths): + # Define label paths as a function of image paths + sa, sb = f'{os.sep}images{os.sep}', f'{os.sep}labels{os.sep}' # /images/, /labels/ substrings + return [sb.join(x.rsplit(sa, 1)).rsplit('.', 1)[0] + '.txt' for x in img_paths] + + +class LoadImagesAndLabels(Dataset): + # YOLOv5 train_loader/val_loader, loads images and labels for training and validation + cache_version = 0.6 # dataset labels *.cache version + rand_interp_methods = [cv2.INTER_NEAREST, cv2.INTER_LINEAR, cv2.INTER_CUBIC, cv2.INTER_AREA, cv2.INTER_LANCZOS4] + + def __init__(self, + path, + img_size=640, + batch_size=16, + augment=False, + hyp=None, + rect=False, + image_weights=False, + cache_images=False, + single_cls=False, + stride=32, + pad=0.0, + prefix=''): + self.img_size = img_size + self.augment = augment + self.hyp = hyp + self.image_weights = image_weights + self.rect = False if image_weights else rect + self.mosaic = self.augment and not self.rect # load 4 images at a time into a mosaic (only during training) + self.mosaic_border = [-img_size // 2, -img_size // 2] + self.stride = stride + self.path = path + self.albumentations = Albumentations() if augment else None + + try: + f = [] # image files + for p in path if isinstance(path, list) else [path]: + p = Path(p) # os-agnostic + if p.is_dir(): # dir + f += glob.glob(str(p / '**' / '*.*'), recursive=True) + # f = list(p.rglob('*.*')) # pathlib + elif p.is_file(): # file + with open(p) as t: + t = t.read().strip().splitlines() + parent = str(p.parent) + os.sep + f += [x.replace('./', parent) if x.startswith('./') else x for x in t] # local to global path + # f += [p.parent / x.lstrip(os.sep) for x in t] # local to global path (pathlib) + else: + raise FileNotFoundError(f'{prefix}{p} does not exist') + self.im_files = sorted(x.replace('/', os.sep) for x in f if x.split('.')[-1].lower() in IMG_FORMATS) + # self.img_files = sorted([x for x in f if x.suffix[1:].lower() in IMG_FORMATS]) # pathlib + assert self.im_files, f'{prefix}No images found' + except Exception as e: + raise Exception(f'{prefix}Error loading data from {path}: {e}\nSee {HELP_URL}') + + # Check cache + self.label_files = img2label_paths(self.im_files) # labels + cache_path = (p if p.is_file() else Path(self.label_files[0]).parent).with_suffix('.cache') + try: + cache, exists = np.load(cache_path, allow_pickle=True).item(), True # load dict + assert cache['version'] == self.cache_version # matches current version + assert cache['hash'] == get_hash(self.label_files + self.im_files) # identical hash + except Exception: + cache, exists = self.cache_labels(cache_path, prefix), False # run cache ops + + # Display cache + nf, nm, ne, nc, n = cache.pop('results') # found, missing, empty, corrupt, total + if exists and LOCAL_RANK in {-1, 0}: + d = f"Scanning '{cache_path}' images and labels... {nf} found, {nm} missing, {ne} empty, {nc} corrupt" + tqdm(None, desc=prefix + d, total=n, initial=n, bar_format=BAR_FORMAT) # display cache results + if cache['msgs']: + LOGGER.info('\n'.join(cache['msgs'])) # display warnings + assert nf > 0 or not augment, f'{prefix}No labels in {cache_path}. Can not train without labels. See {HELP_URL}' + + # Read cache + [cache.pop(k) for k in ('hash', 'version', 'msgs')] # remove items + labels, shapes, self.segments = zip(*cache.values()) + self.labels = list(labels) + self.shapes = np.array(shapes) + self.im_files = list(cache.keys()) # update + self.label_files = img2label_paths(cache.keys()) # update + n = len(shapes) # number of images + bi = np.floor(np.arange(n) / batch_size).astype(np.int) # batch index + nb = bi[-1] + 1 # number of batches + self.batch = bi # batch index of image + self.n = n + self.indices = range(n) + + # Update labels + include_class = [] # filter labels to include only these classes (optional) + include_class_array = np.array(include_class).reshape(1, -1) + for i, (label, segment) in enumerate(zip(self.labels, self.segments)): + if include_class: + j = (label[:, 0:1] == include_class_array).any(1) + self.labels[i] = label[j] + if segment: + self.segments[i] = segment[j] + if single_cls: # single-class training, merge all classes into 0 + self.labels[i][:, 0] = 0 + if segment: + self.segments[i][:, 0] = 0 + + # Rectangular Training + if self.rect: + # Sort by aspect ratio + s = self.shapes # wh + ar = s[:, 1] / s[:, 0] # aspect ratio + irect = ar.argsort() + self.im_files = [self.im_files[i] for i in irect] + self.label_files = [self.label_files[i] for i in irect] + self.labels = [self.labels[i] for i in irect] + self.shapes = s[irect] # wh + ar = ar[irect] + + # Set training image shapes + shapes = [[1, 1]] * nb + for i in range(nb): + ari = ar[bi == i] + mini, maxi = ari.min(), ari.max() + if maxi < 1: + shapes[i] = [maxi, 1] + elif mini > 1: + shapes[i] = [1, 1 / mini] + + self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int) * stride + + # Cache images into RAM/disk for faster training (WARNING: large datasets may exceed system resources) + self.ims = [None] * n + self.npy_files = [Path(f).with_suffix('.npy') for f in self.im_files] + if cache_images: + gb = 0 # Gigabytes of cached images + self.im_hw0, self.im_hw = [None] * n, [None] * n + fcn = self.cache_images_to_disk if cache_images == 'disk' else self.load_image + results = ThreadPool(NUM_THREADS).imap(fcn, range(n)) + pbar = tqdm(enumerate(results), total=n, bar_format=BAR_FORMAT, disable=LOCAL_RANK > 0) + for i, x in pbar: + if cache_images == 'disk': + gb += self.npy_files[i].stat().st_size + else: # 'ram' + self.ims[i], self.im_hw0[i], self.im_hw[i] = x # im, hw_orig, hw_resized = load_image(self, i) + gb += self.ims[i].nbytes + pbar.desc = f'{prefix}Caching images ({gb / 1E9:.1f}GB {cache_images})' + pbar.close() + + def cache_labels(self, path=Path('./labels.cache'), prefix=''): + # Cache dataset labels, check images and read shapes + x = {} # dict + nm, nf, ne, nc, msgs = 0, 0, 0, 0, [] # number missing, found, empty, corrupt, messages + desc = f"{prefix}Scanning '{path.parent / path.stem}' images and labels..." + with Pool(NUM_THREADS) as pool: + pbar = tqdm(pool.imap(verify_image_label, zip(self.im_files, self.label_files, repeat(prefix))), + desc=desc, + total=len(self.im_files), + bar_format=BAR_FORMAT) + for im_file, lb, shape, segments, nm_f, nf_f, ne_f, nc_f, msg in pbar: + nm += nm_f + nf += nf_f + ne += ne_f + nc += nc_f + if im_file: + x[im_file] = [lb, shape, segments] + if msg: + msgs.append(msg) + pbar.desc = f"{desc}{nf} found, {nm} missing, {ne} empty, {nc} corrupt" + + pbar.close() + if msgs: + LOGGER.info('\n'.join(msgs)) + if nf == 0: + LOGGER.warning(f'{prefix}WARNING: No labels found in {path}. See {HELP_URL}') + x['hash'] = get_hash(self.label_files + self.im_files) + x['results'] = nf, nm, ne, nc, len(self.im_files) + x['msgs'] = msgs # warnings + x['version'] = self.cache_version # cache version + try: + np.save(path, x) # save cache for next time + path.with_suffix('.cache.npy').rename(path) # remove .npy suffix + LOGGER.info(f'{prefix}New cache created: {path}') + except Exception as e: + LOGGER.warning(f'{prefix}WARNING: Cache directory {path.parent} is not writeable: {e}') # not writeable + return x + + def __len__(self): + return len(self.im_files) + + # def __iter__(self): + # self.count = -1 + # print('ran dataset iter') + # #self.shuffled_vector = np.random.permutation(self.nF) if self.augment else np.arange(self.nF) + # return self + + def __getitem__(self, index): + index = self.indices[index] # linear, shuffled, or image_weights + + hyp = self.hyp + mosaic = self.mosaic and random.random() < hyp['mosaic'] + if mosaic: + # Load mosaic + img, labels = self.load_mosaic(index) + shapes = None + + # MixUp augmentation + if random.random() < hyp['mixup']: + img, labels = mixup(img, labels, *self.load_mosaic(random.randint(0, self.n - 1))) + + else: + # Load image + img, (h0, w0), (h, w) = self.load_image(index) + + # Letterbox + shape = self.batch_shapes[self.batch[index]] if self.rect else self.img_size # final letterboxed shape + img, ratio, pad = letterbox(img, shape, auto=False, scaleup=self.augment) + shapes = (h0, w0), ((h / h0, w / w0), pad) # for COCO mAP rescaling + + labels = self.labels[index].copy() + if labels.size: # normalized xywh to pixel xyxy format + labels[:, 1:] = xywhn2xyxy(labels[:, 1:], ratio[0] * w, ratio[1] * h, padw=pad[0], padh=pad[1]) + + if self.augment: + img, labels = random_perspective(img, + labels, + degrees=hyp['degrees'], + translate=hyp['translate'], + scale=hyp['scale'], + shear=hyp['shear'], + perspective=hyp['perspective']) + + nl = len(labels) # number of labels + if nl: + labels[:, 1:5] = xyxy2xywhn(labels[:, 1:5], w=img.shape[1], h=img.shape[0], clip=True, eps=1E-3) + + if self.augment: + # Albumentations + img, labels = self.albumentations(img, labels) + nl = len(labels) # update after albumentations + + # HSV color-space + augment_hsv(img, hgain=hyp['hsv_h'], sgain=hyp['hsv_s'], vgain=hyp['hsv_v']) + + # Flip up-down + if random.random() < hyp['flipud']: + img = np.flipud(img) + if nl: + labels[:, 2] = 1 - labels[:, 2] + + # Flip left-right + if random.random() < hyp['fliplr']: + img = np.fliplr(img) + if nl: + labels[:, 1] = 1 - labels[:, 1] + + # Cutouts + # labels = cutout(img, labels, p=0.5) + # nl = len(labels) # update after cutout + + labels_out = torch.zeros((nl, 6)) + if nl: + labels_out[:, 1:] = torch.from_numpy(labels) + + # Convert + img = img.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB + img = np.ascontiguousarray(img) + + return torch.from_numpy(img), labels_out, self.im_files[index], shapes + + def load_image(self, i): + # Loads 1 image from dataset index 'i', returns (im, original hw, resized hw) + im, f, fn = self.ims[i], self.im_files[i], self.npy_files[i], + if im is None: # not cached in RAM + if fn.exists(): # load npy + im = np.load(fn) + else: # read image + # print('idx:{} file:{}'.format(i, f)) + im = cv2.imread(f) # BGR + assert im is not None, f'Image Not Found {f}' + h0, w0 = im.shape[:2] # orig hw + r = self.img_size / max(h0, w0) # ratio + if r != 1: # if sizes are not equal + interp = cv2.INTER_LINEAR if (self.augment or r > 1) else cv2.INTER_AREA + # print('idx:{} r:{} interp:{}'.format(i, r, interp)) + im = cv2.resize(im, (int(w0 * r), int(h0 * r)), interpolation=interp) + return im, (h0, w0), im.shape[:2] # im, hw_original, hw_resized + return self.ims[i], self.im_hw0[i], self.im_hw[i] # im, hw_original, hw_resized + + def cache_images_to_disk(self, i): + # Saves an image as an *.npy file for faster loading + f = self.npy_files[i] + if not f.exists(): + np.save(f.as_posix(), cv2.imread(self.im_files[i])) + + def load_mosaic(self, index): + # YOLOv5 4-mosaic loader. Loads 1 image + 3 random images into a 4-image mosaic + labels4, segments4 = [], [] + s = self.img_size + yc, xc = (int(random.uniform(-x, 2 * s + x)) for x in self.mosaic_border) # mosaic center x, y + indices = [index] + random.choices(self.indices, k=3) # 3 additional image indices + random.shuffle(indices) + for i, index in enumerate(indices): + # Load image + img, _, (h, w) = self.load_image(index) + + # place img in img4 + if i == 0: # top left + img4 = np.full((s * 2, s * 2, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles + x1a, y1a, x2a, y2a = max(xc - w, 0), max(yc - h, 0), xc, yc # xmin, ymin, xmax, ymax (large image) + x1b, y1b, x2b, y2b = w - (x2a - x1a), h - (y2a - y1a), w, h # xmin, ymin, xmax, ymax (small image) + elif i == 1: # top right + x1a, y1a, x2a, y2a = xc, max(yc - h, 0), min(xc + w, s * 2), yc + x1b, y1b, x2b, y2b = 0, h - (y2a - y1a), min(w, x2a - x1a), h + elif i == 2: # bottom left + x1a, y1a, x2a, y2a = max(xc - w, 0), yc, xc, min(s * 2, yc + h) + x1b, y1b, x2b, y2b = w - (x2a - x1a), 0, w, min(y2a - y1a, h) + elif i == 3: # bottom right + x1a, y1a, x2a, y2a = xc, yc, min(xc + w, s * 2), min(s * 2, yc + h) + x1b, y1b, x2b, y2b = 0, 0, min(w, x2a - x1a), min(y2a - y1a, h) + + img4[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax] + padw = x1a - x1b + padh = y1a - y1b + + # Labels + labels, segments = self.labels[index].copy(), self.segments[index].copy() + if labels.size: + labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padw, padh) # normalized xywh to pixel xyxy format + segments = [xyn2xy(x, w, h, padw, padh) for x in segments] + labels4.append(labels) + segments4.extend(segments) + + # Concat/clip labels + labels4 = np.concatenate(labels4, 0) + for x in (labels4[:, 1:], *segments4): + np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() + # img4, labels4 = replicate(img4, labels4) # replicate + + # Augment + img4, labels4, segments4 = copy_paste(img4, labels4, segments4, p=self.hyp['copy_paste']) + img4, labels4 = random_perspective(img4, + labels4, + segments4, + degrees=self.hyp['degrees'], + translate=self.hyp['translate'], + scale=self.hyp['scale'], + shear=self.hyp['shear'], + perspective=self.hyp['perspective'], + border=self.mosaic_border) # border to remove + + return img4, labels4 + + def load_mosaic9(self, index): + # YOLOv5 9-mosaic loader. Loads 1 image + 8 random images into a 9-image mosaic + labels9, segments9 = [], [] + s = self.img_size + indices = [index] + random.choices(self.indices, k=8) # 8 additional image indices + random.shuffle(indices) + hp, wp = -1, -1 # height, width previous + for i, index in enumerate(indices): + # Load image + img, _, (h, w) = self.load_image(index) + + # place img in img9 + if i == 0: # center + img9 = np.full((s * 3, s * 3, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles + h0, w0 = h, w + c = s, s, s + w, s + h # xmin, ymin, xmax, ymax (base) coordinates + elif i == 1: # top + c = s, s - h, s + w, s + elif i == 2: # top right + c = s + wp, s - h, s + wp + w, s + elif i == 3: # right + c = s + w0, s, s + w0 + w, s + h + elif i == 4: # bottom right + c = s + w0, s + hp, s + w0 + w, s + hp + h + elif i == 5: # bottom + c = s + w0 - w, s + h0, s + w0, s + h0 + h + elif i == 6: # bottom left + c = s + w0 - wp - w, s + h0, s + w0 - wp, s + h0 + h + elif i == 7: # left + c = s - w, s + h0 - h, s, s + h0 + elif i == 8: # top left + c = s - w, s + h0 - hp - h, s, s + h0 - hp + + padx, pady = c[:2] + x1, y1, x2, y2 = (max(x, 0) for x in c) # allocate coords + + # Labels + labels, segments = self.labels[index].copy(), self.segments[index].copy() + if labels.size: + labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padx, pady) # normalized xywh to pixel xyxy format + segments = [xyn2xy(x, w, h, padx, pady) for x in segments] + labels9.append(labels) + segments9.extend(segments) + + # Image + img9[y1:y2, x1:x2] = img[y1 - pady:, x1 - padx:] # img9[ymin:ymax, xmin:xmax] + hp, wp = h, w # height, width previous + + # Offset + yc, xc = (int(random.uniform(0, s)) for _ in self.mosaic_border) # mosaic center x, y + img9 = img9[yc:yc + 2 * s, xc:xc + 2 * s] + + # Concat/clip labels + labels9 = np.concatenate(labels9, 0) + labels9[:, [1, 3]] -= xc + labels9[:, [2, 4]] -= yc + c = np.array([xc, yc]) # centers + segments9 = [x - c for x in segments9] + + for x in (labels9[:, 1:], *segments9): + np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() + # img9, labels9 = replicate(img9, labels9) # replicate + + # Augment + img9, labels9 = random_perspective(img9, + labels9, + segments9, + degrees=self.hyp['degrees'], + translate=self.hyp['translate'], + scale=self.hyp['scale'], + shear=self.hyp['shear'], + perspective=self.hyp['perspective'], + border=self.mosaic_border) # border to remove + + return img9, labels9 + + @staticmethod + def collate_fn(batch): + im, label, path, shapes = zip(*batch) # transposed + for i, lb in enumerate(label): + lb[:, 0] = i # add target image index for build_targets() + return torch.stack(im, 0), torch.cat(label, 0), path, shapes + + @staticmethod + def collate_fn4(batch): + img, label, path, shapes = zip(*batch) # transposed + n = len(shapes) // 4 + im4, label4, path4, shapes4 = [], [], path[:n], shapes[:n] + + ho = torch.tensor([[0.0, 0, 0, 1, 0, 0]]) + wo = torch.tensor([[0.0, 0, 1, 0, 0, 0]]) + s = torch.tensor([[1, 1, 0.5, 0.5, 0.5, 0.5]]) # scale + for i in range(n): # zidane torch.zeros(16,3,720,1280) # BCHW + i *= 4 + if random.random() < 0.5: + im = F.interpolate(img[i].unsqueeze(0).float(), scale_factor=2.0, mode='bilinear', + align_corners=False)[0].type(img[i].type()) + lb = label[i] + else: + im = torch.cat((torch.cat((img[i], img[i + 1]), 1), torch.cat((img[i + 2], img[i + 3]), 1)), 2) + lb = torch.cat((label[i], label[i + 1] + ho, label[i + 2] + wo, label[i + 3] + ho + wo), 0) * s + im4.append(im) + label4.append(lb) + + for i, lb in enumerate(label4): + lb[:, 0] = i # add target image index for build_targets() + + return torch.stack(im4, 0), torch.cat(label4, 0), path4, shapes4 + + +# Ancillary functions -------------------------------------------------------------------------------------------------- +def flatten_recursive(path=DATASETS_DIR / 'coco128'): + # Flatten a recursive directory by bringing all files to top level + new_path = Path(f'{str(path)}_flat') + if os.path.exists(new_path): + shutil.rmtree(new_path) # delete output folder + os.makedirs(new_path) # make new output folder + for file in tqdm(glob.glob(f'{str(Path(path))}/**/*.*', recursive=True)): + shutil.copyfile(file, new_path / Path(file).name) + + +def extract_boxes(path=DATASETS_DIR / 'coco128'): # from utils.dataloaders import *; extract_boxes() + # Convert detection dataset into classification dataset, with one directory per class + path = Path(path) # images dir + shutil.rmtree(path / 'classification') if (path / 'classification').is_dir() else None # remove existing + files = list(path.rglob('*.*')) + n = len(files) # number of files + for im_file in tqdm(files, total=n): + if im_file.suffix[1:] in IMG_FORMATS: + # image + im = cv2.imread(str(im_file))[..., ::-1] # BGR to RGB + h, w = im.shape[:2] + + # labels + lb_file = Path(img2label_paths([str(im_file)])[0]) + if Path(lb_file).exists(): + with open(lb_file) as f: + lb = np.array([x.split() for x in f.read().strip().splitlines()], dtype=np.float32) # labels + + for j, x in enumerate(lb): + c = int(x[0]) # class + f = (path / 'classifier') / f'{c}' / f'{path.stem}_{im_file.stem}_{j}.jpg' # new filename + if not f.parent.is_dir(): + f.parent.mkdir(parents=True) + + b = x[1:] * [w, h, w, h] # box + # b[2:] = b[2:].max() # rectangle to square + b[2:] = b[2:] * 1.2 + 3 # pad + b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int) + + b[[0, 2]] = np.clip(b[[0, 2]], 0, w) # clip boxes outside of image + b[[1, 3]] = np.clip(b[[1, 3]], 0, h) + assert cv2.imwrite(str(f), im[b[1]:b[3], b[0]:b[2]]), f'box failure in {f}' + + +def autosplit(path=DATASETS_DIR / 'coco128/images', weights=(0.9, 0.1, 0.0), annotated_only=False): + """ Autosplit a dataset into train/val/test splits and save path/autosplit_*.txt files + Usage: from utils.dataloaders import *; autosplit() + Arguments + path: Path to images directory + weights: Train, val, test weights (list, tuple) + annotated_only: Only use images with an annotated txt file + """ + path = Path(path) # images dir + files = sorted(x for x in path.rglob('*.*') if x.suffix[1:].lower() in IMG_FORMATS) # image files only + n = len(files) # number of files + random.seed(0) # for reproducibility + indices = random.choices([0, 1, 2], weights=weights, k=n) # assign each image to a split + + txt = ['autosplit_train.txt', 'autosplit_val.txt', 'autosplit_test.txt'] # 3 txt files + [(path.parent / x).unlink(missing_ok=True) for x in txt] # remove existing + + print(f'Autosplitting images from {path}' + ', using *.txt labeled images only' * annotated_only) + for i, img in tqdm(zip(indices, files), total=n): + if not annotated_only or Path(img2label_paths([str(img)])[0]).exists(): # check label + with open(path.parent / txt[i], 'a') as f: + f.write(f'./{img.relative_to(path.parent).as_posix()}' + '\n') # add image to txt file + + +def verify_image_label(args): + # Verify one image-label pair + im_file, lb_file, prefix = args + nm, nf, ne, nc, msg, segments = 0, 0, 0, 0, '', [] # number (missing, found, empty, corrupt), message, segments + try: + # verify images + im = Image.open(im_file) + im.verify() # PIL verify + shape = exif_size(im) # image size + assert (shape[0] > 9) & (shape[1] > 9), f'image size {shape} <10 pixels' + assert im.format.lower() in IMG_FORMATS, f'invalid image format {im.format}' + if im.format.lower() in ('jpg', 'jpeg'): + with open(im_file, 'rb') as f: + f.seek(-2, 2) + if f.read() != b'\xff\xd9': # corrupt JPEG + ImageOps.exif_transpose(Image.open(im_file)).save(im_file, 'JPEG', subsampling=0, quality=100) + msg = f'{prefix}WARNING: {im_file}: corrupt JPEG restored and saved' + + # verify labels + if os.path.isfile(lb_file): + nf = 1 # label found + with open(lb_file) as f: + lb = [x.split() for x in f.read().strip().splitlines() if len(x)] + if any(len(x) > 6 for x in lb): # is segment + classes = np.array([x[0] for x in lb], dtype=np.float32) + segments = [np.array(x[1:], dtype=np.float32).reshape(-1, 2) for x in lb] # (cls, xy1...) + lb = np.concatenate((classes.reshape(-1, 1), segments2boxes(segments)), 1) # (cls, xywh) + lb = np.array(lb, dtype=np.float32) + nl = len(lb) + if nl: + assert lb.shape[1] == 5, f'labels require 5 columns, {lb.shape[1]} columns detected' + assert (lb >= 0).all(), f'negative label values {lb[lb < 0]}' + assert (lb[:, 1:] <= 1).all(), f'non-normalized or out of bounds coordinates {lb[:, 1:][lb[:, 1:] > 1]}' + _, i = np.unique(lb, axis=0, return_index=True) + if len(i) < nl: # duplicate row check + lb = lb[i] # remove duplicates + if segments: + segments = segments[i] + msg = f'{prefix}WARNING: {im_file}: {nl - len(i)} duplicate labels removed' + else: + ne = 1 # label empty + lb = np.zeros((0, 5), dtype=np.float32) + else: + nm = 1 # label missing + lb = np.zeros((0, 5), dtype=np.float32) + return im_file, lb, shape, segments, nm, nf, ne, nc, msg + except Exception as e: + nc = 1 + msg = f'{prefix}WARNING: {im_file}: ignoring corrupt image/label: {e}' + return [None, None, None, None, nm, nf, ne, nc, msg] + + +class HUBDatasetStats(): + """ Return dataset statistics dictionary with images and instances counts per split per class + To run in parent directory: export PYTHONPATH="$PWD/yolov5" + Usage1: from utils.dataloaders import *; HUBDatasetStats('coco128.yaml', autodownload=True) + Usage2: from utils.dataloaders import *; HUBDatasetStats('path/to/coco128_with_yaml.zip') + Arguments + path: Path to data.yaml or data.zip (with data.yaml inside data.zip) + autodownload: Attempt to download dataset if not found locally + """ + + def __init__(self, path='coco128.yaml', autodownload=False): + # Initialize class + zipped, data_dir, yaml_path = self._unzip(Path(path)) + try: + with open(check_yaml(yaml_path), errors='ignore') as f: + data = yaml.safe_load(f) # data dict + if zipped: + data['path'] = data_dir + except Exception as e: + raise Exception("error/HUB/dataset_stats/yaml_load") from e + + check_dataset(data, autodownload) # download dataset if missing + self.hub_dir = Path(data['path'] + '-hub') + self.im_dir = self.hub_dir / 'images' + self.im_dir.mkdir(parents=True, exist_ok=True) # makes /images + self.stats = {'nc': data['nc'], 'names': data['names']} # statistics dictionary + self.data = data + + @staticmethod + def _find_yaml(dir): + # Return data.yaml file + files = list(dir.glob('*.yaml')) or list(dir.rglob('*.yaml')) # try root level first and then recursive + assert files, f'No *.yaml file found in {dir}' + if len(files) > 1: + files = [f for f in files if f.stem == dir.stem] # prefer *.yaml files that match dir name + assert files, f'Multiple *.yaml files found in {dir}, only 1 *.yaml file allowed' + assert len(files) == 1, f'Multiple *.yaml files found: {files}, only 1 *.yaml file allowed in {dir}' + return files[0] + + def _unzip(self, path): + # Unzip data.zip + if not str(path).endswith('.zip'): # path is data.yaml + return False, None, path + assert Path(path).is_file(), f'Error unzipping {path}, file not found' + ZipFile(path).extractall(path=path.parent) # unzip + dir = path.with_suffix('') # dataset directory == zip name + assert dir.is_dir(), f'Error unzipping {path}, {dir} not found. path/to/abc.zip MUST unzip to path/to/abc/' + return True, str(dir), self._find_yaml(dir) # zipped, data_dir, yaml_path + + def _hub_ops(self, f, max_dim=1920): + # HUB ops for 1 image 'f': resize and save at reduced quality in /dataset-hub for web/app viewing + f_new = self.im_dir / Path(f).name # dataset-hub image filename + try: # use PIL + im = Image.open(f) + r = max_dim / max(im.height, im.width) # ratio + if r < 1.0: # image too large + im = im.resize((int(im.width * r), int(im.height * r))) + im.save(f_new, 'JPEG', quality=50, optimize=True) # save + except Exception as e: # use OpenCV + print(f'WARNING: HUB ops PIL failure {f}: {e}') + im = cv2.imread(f) + im_height, im_width = im.shape[:2] + r = max_dim / max(im_height, im_width) # ratio + if r < 1.0: # image too large + im = cv2.resize(im, (int(im_width * r), int(im_height * r)), interpolation=cv2.INTER_AREA) + cv2.imwrite(str(f_new), im) + + def get_json(self, save=False, verbose=False): + # Return dataset JSON for Ultralytics HUB + def _round(labels): + # Update labels to integer class and 6 decimal place floats + return [[int(c), *(round(x, 4) for x in points)] for c, *points in labels] + + for split in 'train', 'val', 'test': + if self.data.get(split) is None: + self.stats[split] = None # i.e. no test set + continue + dataset = LoadImagesAndLabels(self.data[split]) # load dataset + x = np.array([ + np.bincount(label[:, 0].astype(int), minlength=self.data['nc']) + for label in tqdm(dataset.labels, total=dataset.n, desc='Statistics')]) # shape(128x80) + self.stats[split] = { + 'instance_stats': { + 'total': int(x.sum()), + 'per_class': x.sum(0).tolist()}, + 'image_stats': { + 'total': dataset.n, + 'unlabelled': int(np.all(x == 0, 1).sum()), + 'per_class': (x > 0).sum(0).tolist()}, + 'labels': [{ + str(Path(k).name): _round(v.tolist())} for k, v in zip(dataset.im_files, dataset.labels)]} + + # Save, print and return + if save: + stats_path = self.hub_dir / 'stats.json' + print(f'Saving {stats_path.resolve()}...') + with open(stats_path, 'w') as f: + json.dump(self.stats, f) # save stats.json + if verbose: + print(json.dumps(self.stats, indent=2, sort_keys=False)) + return self.stats + + def process_images(self): + # Compress images for Ultralytics HUB + for split in 'train', 'val', 'test': + if self.data.get(split) is None: + continue + dataset = LoadImagesAndLabels(self.data[split]) # load dataset + desc = f'{split} images' + for _ in tqdm(ThreadPool(NUM_THREADS).imap(self._hub_ops, dataset.im_files), total=dataset.n, desc=desc): + pass + print(f'Done. All images saved to {self.im_dir}') + return self.im_dir + + +# Classification dataloaders ------------------------------------------------------------------------------------------- +class ClassificationDataset(torchvision.datasets.ImageFolder): + """ + YOLOv5 Classification Dataset. + Arguments + root: Dataset path + transform: torchvision transforms, used by default + album_transform: Albumentations transforms, used if installed + """ + + def __init__(self, root, augment, imgsz, cache=False): + super().__init__(root=root) + self.torch_transforms = classify_transforms(imgsz) + self.album_transforms = classify_albumentations(augment, imgsz) if augment else None + self.cache_ram = cache is True or cache == 'ram' + self.cache_disk = cache == 'disk' + self.samples = [list(x) + [Path(x[0]).with_suffix('.npy'), None] for x in self.samples] # file, index, npy, im + + def __getitem__(self, i): + f, j, fn, im = self.samples[i] # filename, index, filename.with_suffix('.npy'), image + if self.album_transforms: + if self.cache_ram and im is None: + im = self.samples[i][3] = cv2.imread(f) + elif self.cache_disk: + if not fn.exists(): # load npy + np.save(fn.as_posix(), cv2.imread(f)) + im = np.load(fn) + else: # read image + im = cv2.imread(f) # BGR + sample = self.album_transforms(image=cv2.cvtColor(im, cv2.COLOR_BGR2RGB))["image"] + else: + sample = self.torch_transforms(self.loader(f)) + return sample, j + + +def create_classification_dataloader(path, + imgsz=224, + batch_size=16, + augment=True, + cache=False, + rank=-1, + workers=8, + shuffle=True): + # Returns Dataloader object to be used with YOLOv5 Classifier + with torch_distributed_zero_first(rank): # init dataset *.cache only once if DDP + dataset = ClassificationDataset(root=path, imgsz=imgsz, augment=augment, cache=cache) + batch_size = min(batch_size, len(dataset)) + nd = torch.cuda.device_count() + nw = min([os.cpu_count() // max(nd, 1), batch_size if batch_size > 1 else 0, workers]) + sampler = None if rank == -1 else distributed.DistributedSampler(dataset, shuffle=shuffle) + generator = torch.Generator() + generator.manual_seed(0) + return InfiniteDataLoader(dataset, + batch_size=batch_size, + shuffle=shuffle and sampler is None, + num_workers=nw, + sampler=sampler, + pin_memory=True, + worker_init_fn=seed_worker, + generator=generator) # or DataLoader(persistent_workers=True) diff --git a/application/yolov5_example/utils/docker/Dockerfile b/application/yolov5_example/utils/docker/Dockerfile new file mode 100644 index 00000000..2280f209 --- /dev/null +++ b/application/yolov5_example/utils/docker/Dockerfile @@ -0,0 +1,68 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Builds ultralytics/yolov5:latest image on DockerHub https://hub.docker.com/r/ultralytics/yolov5 +# Image is CUDA-optimized for YOLOv5 single/multi-GPU training and inference + +# Start FROM NVIDIA PyTorch image https://ngc.nvidia.com/catalog/containers/nvidia:pytorch +FROM nvcr.io/nvidia/pytorch:22.07-py3 +RUN rm -rf /opt/pytorch # remove 1.2GB dir + +# Downloads to user config dir +ADD https://ultralytics.com/assets/Arial.ttf https://ultralytics.com/assets/Arial.Unicode.ttf /root/.config/Ultralytics/ + +# Install linux packages +RUN apt update && apt install --no-install-recommends -y zip htop screen libgl1-mesa-glx + +# Install pip packages +COPY requirements.txt . +RUN python -m pip install --upgrade pip wheel +RUN pip uninstall -y Pillow torchtext # torch torchvision +RUN pip install --no-cache -r requirements.txt albumentations wandb gsutil notebook Pillow>=9.1.0 \ + 'opencv-python<4.6.0.66' \ + --extra-index-url https://download.pytorch.org/whl/cu113 + +# Create working directory +RUN mkdir -p /usr/src/app +WORKDIR /usr/src/app + +# Copy contents +# COPY . /usr/src/app (issues as not a .git directory) +RUN git clone https://github.com/ultralytics/yolov5 /usr/src/app + +# Set environment variables +ENV OMP_NUM_THREADS=8 + + +# Usage Examples ------------------------------------------------------------------------------------------------------- + +# Build and Push +# t=ultralytics/yolov5:latest && sudo docker build -f utils/docker/Dockerfile -t $t . && sudo docker push $t + +# Pull and Run +# t=ultralytics/yolov5:latest && sudo docker pull $t && sudo docker run -it --ipc=host --gpus all $t + +# Pull and Run with local directory access +# t=ultralytics/yolov5:latest && sudo docker pull $t && sudo docker run -it --ipc=host --gpus all -v "$(pwd)"/datasets:/usr/src/datasets $t + +# Kill all +# sudo docker kill $(sudo docker ps -q) + +# Kill all image-based +# sudo docker kill $(sudo docker ps -qa --filter ancestor=ultralytics/yolov5:latest) + +# Bash into running container +# sudo docker exec -it 5a9b5863d93d bash + +# Bash into stopped container +# id=$(sudo docker ps -qa) && sudo docker start $id && sudo docker exec -it $id bash + +# Clean up +# docker system prune -a --volumes + +# Update Ubuntu drivers +# https://www.maketecheasier.com/install-nvidia-drivers-ubuntu/ + +# DDP test +# python -m torch.distributed.run --nproc_per_node 2 --master_port 1 train.py --epochs 3 + +# GCP VM from Image +# docker.io/ultralytics/yolov5:latest diff --git a/application/yolov5_example/utils/docker/Dockerfile-arm64 b/application/yolov5_example/utils/docker/Dockerfile-arm64 new file mode 100644 index 00000000..fe92c8d5 --- /dev/null +++ b/application/yolov5_example/utils/docker/Dockerfile-arm64 @@ -0,0 +1,42 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Builds ultralytics/yolov5:latest-arm64 image on DockerHub https://hub.docker.com/r/ultralytics/yolov5 +# Image is aarch64-compatible for Apple M1 and other ARM architectures i.e. Jetson Nano and Raspberry Pi + +# Start FROM Ubuntu image https://hub.docker.com/_/ubuntu +FROM arm64v8/ubuntu:20.04 + +# Downloads to user config dir +ADD https://ultralytics.com/assets/Arial.ttf https://ultralytics.com/assets/Arial.Unicode.ttf /root/.config/Ultralytics/ + +# Install linux packages +RUN apt update +RUN DEBIAN_FRONTEND=noninteractive TZ=Etc/UTC apt install -y tzdata +RUN apt install --no-install-recommends -y python3-pip git zip curl htop gcc \ + libgl1-mesa-glx libglib2.0-0 libpython3.8-dev +# RUN alias python=python3 + +# Install pip packages +COPY requirements.txt . +RUN python3 -m pip install --upgrade pip wheel +RUN pip install --no-cache -r requirements.txt gsutil notebook \ + tensorflow-aarch64 + # tensorflowjs \ + # onnx onnx-simplifier onnxruntime \ + # coremltools openvino-dev \ + +# Create working directory +RUN mkdir -p /usr/src/app +WORKDIR /usr/src/app + +# Copy contents +# COPY . /usr/src/app (issues as not a .git directory) +RUN git clone https://github.com/ultralytics/yolov5 /usr/src/app + + +# Usage Examples ------------------------------------------------------------------------------------------------------- + +# Build and Push +# t=ultralytics/yolov5:latest-M1 && sudo docker build --platform linux/arm64 -f utils/docker/Dockerfile-arm64 -t $t . && sudo docker push $t + +# Pull and Run +# t=ultralytics/yolov5:latest-M1 && sudo docker pull $t && sudo docker run -it --ipc=host -v "$(pwd)"/datasets:/usr/src/datasets $t diff --git a/application/yolov5_example/utils/docker/Dockerfile-cpu b/application/yolov5_example/utils/docker/Dockerfile-cpu new file mode 100644 index 00000000..d61dfeff --- /dev/null +++ b/application/yolov5_example/utils/docker/Dockerfile-cpu @@ -0,0 +1,39 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +# Builds ultralytics/yolov5:latest-cpu image on DockerHub https://hub.docker.com/r/ultralytics/yolov5 +# Image is CPU-optimized for ONNX, OpenVINO and PyTorch YOLOv5 deployments + +# Start FROM Ubuntu image https://hub.docker.com/_/ubuntu +FROM ubuntu:20.04 + +# Downloads to user config dir +ADD https://ultralytics.com/assets/Arial.ttf https://ultralytics.com/assets/Arial.Unicode.ttf /root/.config/Ultralytics/ + +# Install linux packages +RUN apt update +RUN DEBIAN_FRONTEND=noninteractive TZ=Etc/UTC apt install -y tzdata +RUN apt install --no-install-recommends -y python3-pip git zip curl htop libgl1-mesa-glx libglib2.0-0 libpython3.8-dev +# RUN alias python=python3 + +# Install pip packages +COPY requirements.txt . +RUN python3 -m pip install --upgrade pip wheel +RUN pip install --no-cache -r requirements.txt albumentations gsutil notebook \ + coremltools onnx onnx-simplifier onnxruntime openvino-dev tensorflow-cpu tensorflowjs \ + --extra-index-url https://download.pytorch.org/whl/cpu + +# Create working directory +RUN mkdir -p /usr/src/app +WORKDIR /usr/src/app + +# Copy contents +# COPY . /usr/src/app (issues as not a .git directory) +RUN git clone https://github.com/ultralytics/yolov5 /usr/src/app + + +# Usage Examples ------------------------------------------------------------------------------------------------------- + +# Build and Push +# t=ultralytics/yolov5:latest-cpu && sudo docker build -f utils/docker/Dockerfile-cpu -t $t . && sudo docker push $t + +# Pull and Run +# t=ultralytics/yolov5:latest-cpu && sudo docker pull $t && sudo docker run -it --ipc=host -v "$(pwd)"/datasets:/usr/src/datasets $t diff --git a/application/yolov5_example/utils/downloads.py b/application/yolov5_example/utils/downloads.py new file mode 100644 index 00000000..9d4780ad --- /dev/null +++ b/application/yolov5_example/utils/downloads.py @@ -0,0 +1,180 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Download utils +""" + +import logging +import os +import platform +import subprocess +import time +import urllib +from pathlib import Path +from zipfile import ZipFile + +import requests +import torch + + +def is_url(url, check_online=True): + # Check if online file exists + try: + url = str(url) + result = urllib.parse.urlparse(url) + assert all([result.scheme, result.netloc, result.path]) # check if is url + return (urllib.request.urlopen(url).getcode() == 200) if check_online else True # check if exists online + except (AssertionError, urllib.request.HTTPError): + return False + + +def gsutil_getsize(url=''): + # gs://bucket/file size https://cloud.google.com/storage/docs/gsutil/commands/du + s = subprocess.check_output(f'gsutil du {url}', shell=True).decode('utf-8') + return eval(s.split(' ')[0]) if len(s) else 0 # bytes + + +def safe_download(file, url, url2=None, min_bytes=1E0, error_msg=''): + # Attempts to download file from url or url2, checks and removes incomplete downloads < min_bytes + from utils.general import LOGGER + + file = Path(file) + assert_msg = f"Downloaded file '{file}' does not exist or size is < min_bytes={min_bytes}" + try: # url1 + LOGGER.info(f'Downloading {url} to {file}...') + torch.hub.download_url_to_file(url, str(file), progress=LOGGER.level <= logging.INFO) + assert file.exists() and file.stat().st_size > min_bytes, assert_msg # check + except Exception as e: # url2 + file.unlink(missing_ok=True) # remove partial downloads + LOGGER.info(f'ERROR: {e}\nRe-attempting {url2 or url} to {file}...') + os.system(f"curl -L '{url2 or url}' -o '{file}' --retry 3 -C -") # curl download, retry and resume on fail + finally: + if not file.exists() or file.stat().st_size < min_bytes: # check + file.unlink(missing_ok=True) # remove partial downloads + LOGGER.info(f"ERROR: {assert_msg}\n{error_msg}") + LOGGER.info('') + + +def attempt_download(file, repo='ultralytics/yolov5', release='v6.1'): + # Attempt file download from GitHub release assets if not found locally. release = 'latest', 'v6.1', etc. + from utils.general import LOGGER + + def github_assets(repository, version='latest'): + # Return GitHub repo tag (i.e. 'v6.1') and assets (i.e. ['yolov5s.pt', 'yolov5m.pt', ...]) + if version != 'latest': + version = f'tags/{version}' # i.e. tags/v6.1 + response = requests.get(f'https://api.github.com/repos/{repository}/releases/{version}').json() # github api + return response['tag_name'], [x['name'] for x in response['assets']] # tag, assets + + file = Path(str(file).strip().replace("'", '')) + if not file.exists(): + # URL specified + name = Path(urllib.parse.unquote(str(file))).name # decode '%2F' to '/' etc. + if str(file).startswith(('http:/', 'https:/')): # download + url = str(file).replace(':/', '://') # Pathlib turns :// -> :/ + file = name.split('?')[0] # parse authentication https://url.com/file.txt?auth... + if Path(file).is_file(): + LOGGER.info(f'Found {url} locally at {file}') # file already exists + else: + safe_download(file=file, url=url, min_bytes=1E5) + return file + + # GitHub assets + assets = [ + 'yolov5n.pt', 'yolov5s.pt', 'yolov5m.pt', 'yolov5l.pt', 'yolov5x.pt', 'yolov5n6.pt', 'yolov5s6.pt', + 'yolov5m6.pt', 'yolov5l6.pt', 'yolov5x6.pt'] + try: + tag, assets = github_assets(repo, release) + except Exception: + try: + tag, assets = github_assets(repo) # latest release + except Exception: + try: + tag = subprocess.check_output('git tag', shell=True, stderr=subprocess.STDOUT).decode().split()[-1] + except Exception: + tag = release + + file.parent.mkdir(parents=True, exist_ok=True) # make parent dir (if required) + if name in assets: + url3 = 'https://drive.google.com/drive/folders/1EFQTEUeXWSFww0luse2jB9M1QNZQGwNl' # backup gdrive mirror + safe_download( + file, + url=f'https://github.com/{repo}/releases/download/{tag}/{name}', + url2=f'https://storage.googleapis.com/{repo}/{tag}/{name}', # backup url (optional) + min_bytes=1E5, + error_msg=f'{file} missing, try downloading from https://github.com/{repo}/releases/{tag} or {url3}') + + return str(file) + + +def gdrive_download(id='16TiPfZj7htmTyhntwcZyEEAejOUxuT6m', file='tmp.zip'): + # Downloads a file from Google Drive. from yolov5.utils.downloads import *; gdrive_download() + t = time.time() + file = Path(file) + cookie = Path('cookie') # gdrive cookie + print(f'Downloading https://drive.google.com/uc?export=download&id={id} as {file}... ', end='') + file.unlink(missing_ok=True) # remove existing file + cookie.unlink(missing_ok=True) # remove existing cookie + + # Attempt file download + out = "NUL" if platform.system() == "Windows" else "/dev/null" + os.system(f'curl -c ./cookie -s -L "drive.google.com/uc?export=download&id={id}" > {out}') + if os.path.exists('cookie'): # large file + s = f'curl -Lb ./cookie "drive.google.com/uc?export=download&confirm={get_token()}&id={id}" -o {file}' + else: # small file + s = f'curl -s -L -o {file} "drive.google.com/uc?export=download&id={id}"' + r = os.system(s) # execute, capture return + cookie.unlink(missing_ok=True) # remove existing cookie + + # Error check + if r != 0: + file.unlink(missing_ok=True) # remove partial + print('Download error ') # raise Exception('Download error') + return r + + # Unzip if archive + if file.suffix == '.zip': + print('unzipping... ', end='') + ZipFile(file).extractall(path=file.parent) # unzip + file.unlink() # remove zip + + print(f'Done ({time.time() - t:.1f}s)') + return r + + +def get_token(cookie="./cookie"): + with open(cookie) as f: + for line in f: + if "download" in line: + return line.split()[-1] + return "" + + +# Google utils: https://cloud.google.com/storage/docs/reference/libraries ---------------------------------------------- +# +# +# def upload_blob(bucket_name, source_file_name, destination_blob_name): +# # Uploads a file to a bucket +# # https://cloud.google.com/storage/docs/uploading-objects#storage-upload-object-python +# +# storage_client = storage.Client() +# bucket = storage_client.get_bucket(bucket_name) +# blob = bucket.blob(destination_blob_name) +# +# blob.upload_from_filename(source_file_name) +# +# print('File {} uploaded to {}.'.format( +# source_file_name, +# destination_blob_name)) +# +# +# def download_blob(bucket_name, source_blob_name, destination_file_name): +# # Uploads a blob from a bucket +# storage_client = storage.Client() +# bucket = storage_client.get_bucket(bucket_name) +# blob = bucket.blob(source_blob_name) +# +# blob.download_to_filename(destination_file_name) +# +# print('Blob {} downloaded to {}.'.format( +# source_blob_name, +# destination_file_name)) diff --git a/application/yolov5_example/utils/flask_rest_api/README.md b/application/yolov5_example/utils/flask_rest_api/README.md new file mode 100644 index 00000000..a726acbd --- /dev/null +++ b/application/yolov5_example/utils/flask_rest_api/README.md @@ -0,0 +1,73 @@ +# Flask REST API + +[REST](https://en.wikipedia.org/wiki/Representational_state_transfer) [API](https://en.wikipedia.org/wiki/API)s are +commonly used to expose Machine Learning (ML) models to other services. This folder contains an example REST API +created using Flask to expose the YOLOv5s model from [PyTorch Hub](https://pytorch.org/hub/ultralytics_yolov5/). + +## Requirements + +[Flask](https://palletsprojects.com/p/flask/) is required. Install with: + +```shell +$ pip install Flask +``` + +## Run + +After Flask installation run: + +```shell +$ python3 restapi.py --port 5000 +``` + +Then use [curl](https://curl.se/) to perform a request: + +```shell +$ curl -X POST -F image=@zidane.jpg 'http://localhost:5000/v1/object-detection/yolov5s' +``` + +The model inference results are returned as a JSON response: + +```json +[ + { + "class": 0, + "confidence": 0.8900438547, + "height": 0.9318675399, + "name": "person", + "width": 0.3264600933, + "xcenter": 0.7438579798, + "ycenter": 0.5207948685 + }, + { + "class": 0, + "confidence": 0.8440024257, + "height": 0.7155083418, + "name": "person", + "width": 0.6546785235, + "xcenter": 0.427829951, + "ycenter": 0.6334488392 + }, + { + "class": 27, + "confidence": 0.3771208823, + "height": 0.3902671337, + "name": "tie", + "width": 0.0696444362, + "xcenter": 0.3675483763, + "ycenter": 0.7991207838 + }, + { + "class": 27, + "confidence": 0.3527112305, + "height": 0.1540903747, + "name": "tie", + "width": 0.0336618312, + "xcenter": 0.7814827561, + "ycenter": 0.5065554976 + } +] +``` + +An example python script to perform inference using [requests](https://docs.python-requests.org/en/master/) is given +in `example_request.py` diff --git a/application/yolov5_example/utils/flask_rest_api/example_request.py b/application/yolov5_example/utils/flask_rest_api/example_request.py new file mode 100644 index 00000000..773ad893 --- /dev/null +++ b/application/yolov5_example/utils/flask_rest_api/example_request.py @@ -0,0 +1,19 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Perform test request +""" + +import pprint + +import requests + +DETECTION_URL = "http://localhost:5000/v1/object-detection/yolov5s" +IMAGE = "zidane.jpg" + +# Read image +with open(IMAGE, "rb") as f: + image_data = f.read() + +response = requests.post(DETECTION_URL, files={"image": image_data}).json() + +pprint.pprint(response) diff --git a/application/yolov5_example/utils/flask_rest_api/restapi.py b/application/yolov5_example/utils/flask_rest_api/restapi.py new file mode 100644 index 00000000..8482435c --- /dev/null +++ b/application/yolov5_example/utils/flask_rest_api/restapi.py @@ -0,0 +1,48 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Run a Flask REST API exposing one or more YOLOv5s models +""" + +import argparse +import io + +import torch +from flask import Flask, request +from PIL import Image + +app = Flask(__name__) +models = {} + +DETECTION_URL = "/v1/object-detection/" + + +@app.route(DETECTION_URL, methods=["POST"]) +def predict(model): + if request.method != "POST": + return + + if request.files.get("image"): + # Method 1 + # with request.files["image"] as f: + # im = Image.open(io.BytesIO(f.read())) + + # Method 2 + im_file = request.files["image"] + im_bytes = im_file.read() + im = Image.open(io.BytesIO(im_bytes)) + + if model in models: + results = models[model](im, size=640) # reduce size=320 for faster inference + return results.pandas().xyxy[0].to_json(orient="records") + + +if __name__ == "__main__": + parser = argparse.ArgumentParser(description="Flask API exposing YOLOv5 model") + parser.add_argument("--port", default=5000, type=int, help="port number") + parser.add_argument('--model', nargs='+', default=['yolov5s'], help='model(s) to run, i.e. --model yolov5n yolov5s') + opt = parser.parse_args() + + for m in opt.model: + models[m] = torch.hub.load("ultralytics/yolov5", m, force_reload=True, skip_validation=True) + + app.run(host="0.0.0.0", port=opt.port) # debug=True causes Restarting with stat diff --git a/application/yolov5_example/utils/general.py b/application/yolov5_example/utils/general.py new file mode 100644 index 00000000..a9463ddf --- /dev/null +++ b/application/yolov5_example/utils/general.py @@ -0,0 +1,1051 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +General utils +""" + +import contextlib +import glob +import inspect +import logging +import math +import os +import platform +import random +import re +import shutil +import signal +import sys +import threading +import time +import urllib +from datetime import datetime +from itertools import repeat +from multiprocessing.pool import ThreadPool +from pathlib import Path +from subprocess import check_output +from typing import Optional +from zipfile import ZipFile + +import cv2 +import numpy as np +import pandas as pd +import pkg_resources as pkg +import torch +import torchvision +import yaml + +from utils.downloads import gsutil_getsize +from utils.metrics import box_iou, fitness + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[1] # YOLOv5 root directory +RANK = int(os.getenv('RANK', -1)) + +# Settings +DATASETS_DIR = ROOT.parent / 'datasets' # YOLOv5 datasets directory +NUM_THREADS = min(8, max(1, os.cpu_count() - 1)) # number of YOLOv5 multiprocessing threads +AUTOINSTALL = str(os.getenv('YOLOv5_AUTOINSTALL', True)).lower() == 'true' # global auto-install mode +VERBOSE = str(os.getenv('YOLOv5_VERBOSE', True)).lower() == 'true' # global verbose mode +FONT = 'Arial.ttf' # https://ultralytics.com/assets/Arial.ttf + +torch.set_printoptions(linewidth=320, precision=5, profile='long') +np.set_printoptions(linewidth=320, formatter={'float_kind': '{:11.5g}'.format}) # format short g, %precision=5 +pd.options.display.max_columns = 10 +cv2.setNumThreads(0) # prevent OpenCV from multithreading (incompatible with PyTorch DataLoader) +os.environ['NUMEXPR_MAX_THREADS'] = str(NUM_THREADS) # NumExpr max threads +os.environ['OMP_NUM_THREADS'] = '1' if platform.system() == 'darwin' else str(NUM_THREADS) # OpenMP (PyTorch and SciPy) + + +def is_ascii(s=''): + # Is string composed of all ASCII (no UTF) characters? (note str().isascii() introduced in python 3.7) + s = str(s) # convert list, tuple, None, etc. to str + return len(s.encode().decode('ascii', 'ignore')) == len(s) + + +def is_chinese(s='浜哄伐鏅鸿兘'): + # Is string composed of any Chinese characters? + return bool(re.search('[\u4e00-\u9fff]', str(s))) + + +def is_colab(): + # Is environment a Google Colab instance? + return 'COLAB_GPU' in os.environ + + +def is_kaggle(): + # Is environment a Kaggle Notebook? + return os.environ.get('PWD') == '/kaggle/working' and os.environ.get('KAGGLE_URL_BASE') == 'https://www.kaggle.com' + + +def is_docker() -> bool: + """Check if the process runs inside a docker container.""" + if Path("/.dockerenv").exists(): + return True + try: # check if docker is in control groups + with open("/proc/self/cgroup") as file: + return any("docker" in line for line in file) + except OSError: + return False + + +def is_writeable(dir, test=False): + # Return True if directory has write permissions, test opening a file with write permissions if test=True + if not test: + return os.access(dir, os.W_OK) # possible issues on Windows + file = Path(dir) / 'tmp.txt' + try: + with open(file, 'w'): # open file with write permissions + pass + file.unlink() # remove file + return True + except OSError: + return False + + +def set_logging(name=None, verbose=VERBOSE): + # Sets level and returns logger + if is_kaggle() or is_colab(): + for h in logging.root.handlers: + logging.root.removeHandler(h) # remove all handlers associated with the root logger object + rank = int(os.getenv('RANK', -1)) # rank in world for Multi-GPU trainings + level = logging.INFO if verbose and rank in {-1, 0} else logging.ERROR + log = logging.getLogger(name) + log.setLevel(level) + handler = logging.StreamHandler() + handler.setFormatter(logging.Formatter("%(message)s")) + handler.setLevel(level) + log.addHandler(handler) + + +set_logging() # run before defining LOGGER +LOGGER = logging.getLogger("yolov5") # define globally (used in train.py, val.py, detect.py, etc.) +if platform.system() == 'Windows': + for fn in LOGGER.info, LOGGER.warning: + setattr(LOGGER, fn.__name__, lambda x: fn(emojis(x))) # emoji safe logging + + +def user_config_dir(dir='Ultralytics', env_var='YOLOV5_CONFIG_DIR'): + # Return path of user configuration directory. Prefer environment variable if exists. Make dir if required. + env = os.getenv(env_var) + if env: + path = Path(env) # use environment variable + else: + cfg = {'Windows': 'AppData/Roaming', 'Linux': '.config', 'Darwin': 'Library/Application Support'} # 3 OS dirs + path = Path.home() / cfg.get(platform.system(), '') # OS-specific config dir + path = (path if is_writeable(path) else Path('/tmp')) / dir # GCP and AWS lambda fix, only /tmp is writeable + path.mkdir(exist_ok=True) # make if required + return path + + +CONFIG_DIR = user_config_dir() # Ultralytics settings dir + + +class Profile(contextlib.ContextDecorator): + # Usage: @Profile() decorator or 'with Profile():' context manager + def __enter__(self): + self.start = time.time() + + def __exit__(self, type, value, traceback): + print(f'Profile results: {time.time() - self.start:.5f}s') + + +class Timeout(contextlib.ContextDecorator): + # Usage: @Timeout(seconds) decorator or 'with Timeout(seconds):' context manager + def __init__(self, seconds, *, timeout_msg='', suppress_timeout_errors=True): + self.seconds = int(seconds) + self.timeout_message = timeout_msg + self.suppress = bool(suppress_timeout_errors) + + def _timeout_handler(self, signum, frame): + raise TimeoutError(self.timeout_message) + + def __enter__(self): + if platform.system() != 'Windows': # not supported on Windows + signal.signal(signal.SIGALRM, self._timeout_handler) # Set handler for SIGALRM + signal.alarm(self.seconds) # start countdown for SIGALRM to be raised + + def __exit__(self, exc_type, exc_val, exc_tb): + if platform.system() != 'Windows': + signal.alarm(0) # Cancel SIGALRM if it's scheduled + if self.suppress and exc_type is TimeoutError: # Suppress TimeoutError + return True + + +class WorkingDirectory(contextlib.ContextDecorator): + # Usage: @WorkingDirectory(dir) decorator or 'with WorkingDirectory(dir):' context manager + def __init__(self, new_dir): + self.dir = new_dir # new dir + self.cwd = Path.cwd().resolve() # current dir + + def __enter__(self): + os.chdir(self.dir) + + def __exit__(self, exc_type, exc_val, exc_tb): + os.chdir(self.cwd) + + +def try_except(func): + # try-except function. Usage: @try_except decorator + def handler(*args, **kwargs): + try: + func(*args, **kwargs) + except Exception as e: + print(e) + + return handler + + +def threaded(func): + # Multi-threads a target function and returns thread. Usage: @threaded decorator + def wrapper(*args, **kwargs): + thread = threading.Thread(target=func, args=args, kwargs=kwargs, daemon=True) + thread.start() + return thread + + return wrapper + + +def methods(instance): + # Get class/instance methods + return [f for f in dir(instance) if callable(getattr(instance, f)) and not f.startswith("__")] + + +def print_args(args: Optional[dict] = None, show_file=True, show_fcn=False): + # Print function arguments (optional args dict) + x = inspect.currentframe().f_back # previous frame + file, _, fcn, _, _ = inspect.getframeinfo(x) + if args is None: # get args automatically + args, _, _, frm = inspect.getargvalues(x) + args = {k: v for k, v in frm.items() if k in args} + try: + file = Path(file).resolve().relative_to(ROOT).with_suffix('') + except ValueError: + file = Path(file).stem + s = (f'{file}: ' if show_file else '') + (f'{fcn}: ' if show_fcn else '') + LOGGER.info(colorstr(s) + ', '.join(f'{k}={v}' for k, v in args.items())) + + +def init_seeds(seed=0, deterministic=False): + # Initialize random number generator (RNG) seeds https://pytorch.org/docs/stable/notes/randomness.html + # cudnn seed 0 settings are slower and more reproducible, else faster and less reproducible + import torch.backends.cudnn as cudnn + + if deterministic and check_version(torch.__version__, '1.12.0'): # https://github.com/ultralytics/yolov5/pull/8213 + torch.use_deterministic_algorithms(True) + os.environ['CUBLAS_WORKSPACE_CONFIG'] = ':4096:8' + os.environ['PYTHONHASHSEED'] = str(seed) + + random.seed(seed) + np.random.seed(seed) + torch.manual_seed(seed) + cudnn.benchmark, cudnn.deterministic = (False, True) if seed == 0 else (True, False) + torch.cuda.manual_seed(seed) + torch.cuda.manual_seed_all(seed) # for Multi-GPU, exception safe + + +def intersect_dicts(da, db, exclude=()): + # Dictionary intersection of matching keys and shapes, omitting 'exclude' keys, using da values + return {k: v for k, v in da.items() if k in db and not any(x in k for x in exclude) and v.shape == db[k].shape} + + +def get_latest_run(search_dir='.'): + # Return path to most recent 'last.pt' in /runs (i.e. to --resume from) + last_list = glob.glob(f'{search_dir}/**/last*.pt', recursive=True) + return max(last_list, key=os.path.getctime) if last_list else '' + + +def emojis(str=''): + # Return platform-dependent emoji-safe version of string + return str.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else str + + +def file_age(path=__file__): + # Return days since last file update + dt = (datetime.now() - datetime.fromtimestamp(Path(path).stat().st_mtime)) # delta + return dt.days # + dt.seconds / 86400 # fractional days + + +def file_date(path=__file__): + # Return human-readable file modification date, i.e. '2021-3-26' + t = datetime.fromtimestamp(Path(path).stat().st_mtime) + return f'{t.year}-{t.month}-{t.day}' + + +def file_size(path): + # Return file/dir size (MB) + mb = 1 << 20 # bytes to MiB (1024 ** 2) + path = Path(path) + if path.is_file(): + return path.stat().st_size / mb + elif path.is_dir(): + return sum(f.stat().st_size for f in path.glob('**/*') if f.is_file()) / mb + else: + return 0.0 + + +def check_online(): + # Check internet connectivity + import socket + try: + socket.create_connection(("1.1.1.1", 443), 5) # check host accessibility + return True + except OSError: + return False + + +def git_describe(path=ROOT): # path must be a directory + # Return human-readable git description, i.e. v5.0-5-g3e25f1e https://git-scm.com/docs/git-describe + try: + assert (Path(path) / '.git').is_dir() + return check_output(f'git -C {path} describe --tags --long --always', shell=True).decode()[:-1] + except Exception: + return '' + + +@try_except +@WorkingDirectory(ROOT) +def check_git_status(repo='ultralytics/yolov5'): + # YOLOv5 status check, recommend 'git pull' if code is out of date + url = f'https://github.com/{repo}' + msg = f', for updates see {url}' + s = colorstr('github: ') # string + assert Path('.git').exists(), s + 'skipping check (not a git repository)' + msg + assert check_online(), s + 'skipping check (offline)' + msg + + splits = re.split(pattern=r'\s', string=check_output('git remote -v', shell=True).decode()) + matches = [repo in s for s in splits] + if any(matches): + remote = splits[matches.index(True) - 1] + else: + remote = 'ultralytics' + check_output(f'git remote add {remote} {url}', shell=True) + check_output(f'git fetch {remote}', shell=True, timeout=5) # git fetch + branch = check_output('git rev-parse --abbrev-ref HEAD', shell=True).decode().strip() # checked out + n = int(check_output(f'git rev-list {branch}..{remote}/master --count', shell=True)) # commits behind + if n > 0: + pull = 'git pull' if remote == 'origin' else f'git pull {remote} master' + s += f"鈿狅笍 YOLOv5 is out of date by {n} commit{'s' * (n > 1)}. Use `{pull}` or `git clone {url}` to update." + else: + s += f'up to date with {url} 鉁' + LOGGER.info(s) + + +def check_python(minimum='3.7.0'): + # Check current python version vs. required python version + check_version(platform.python_version(), minimum, name='Python ', hard=True) + + +def check_version(current='0.0.0', minimum='0.0.0', name='version ', pinned=False, hard=False, verbose=False): + # Check version vs. required version + current, minimum = (pkg.parse_version(x) for x in (current, minimum)) + result = (current == minimum) if pinned else (current >= minimum) # bool + s = f'{name}{minimum} required by YOLOv5, but {name}{current} is currently installed' # string + if hard: + assert result, s # assert min requirements met + if verbose and not result: + LOGGER.warning(s) + return result + + +@try_except +def check_requirements(requirements=ROOT / 'requirements.txt', exclude=(), install=True, cmds=()): + # Check installed dependencies meet YOLOv5 requirements (pass *.txt file or list of packages) + prefix = colorstr('red', 'bold', 'requirements:') + check_python() # check python version + if isinstance(requirements, (str, Path)): # requirements.txt file + file = Path(requirements) + assert file.exists(), f"{prefix} {file.resolve()} not found, check failed." + with file.open() as f: + requirements = [f'{x.name}{x.specifier}' for x in pkg.parse_requirements(f) if x.name not in exclude] + else: # list or tuple of packages + requirements = [x for x in requirements if x not in exclude] + + n = 0 # number of packages updates + for i, r in enumerate(requirements): + try: + pkg.require(r) + except Exception: # DistributionNotFound or VersionConflict if requirements not met + s = f"{prefix} {r} not found and is required by YOLOv5" + if install and AUTOINSTALL: # check environment variable + LOGGER.info(f"{s}, attempting auto-update...") + try: + assert check_online(), f"'pip install {r}' skipped (offline)" + LOGGER.info(check_output(f'pip install "{r}" {cmds[i] if cmds else ""}', shell=True).decode()) + n += 1 + except Exception as e: + LOGGER.warning(f'{prefix} {e}') + else: + LOGGER.info(f'{s}. Please install and rerun your command.') + + if n: # if packages updated + source = file.resolve() if 'file' in locals() else requirements + s = f"{prefix} {n} package{'s' * (n > 1)} updated per {source}\n" \ + f"{prefix} 鈿狅笍 {colorstr('bold', 'Restart runtime or rerun command for updates to take effect')}\n" + LOGGER.info(s) + + +def check_img_size(imgsz, s=32, floor=0): + # Verify image size is a multiple of stride s in each dimension + if isinstance(imgsz, int): # integer i.e. img_size=640 + new_size = max(make_divisible(imgsz, int(s)), floor) + else: # list i.e. img_size=[640, 480] + imgsz = list(imgsz) # convert to list if tuple + new_size = [max(make_divisible(x, int(s)), floor) for x in imgsz] + if new_size != imgsz: + LOGGER.warning(f'WARNING: --img-size {imgsz} must be multiple of max stride {s}, updating to {new_size}') + return new_size + + +def check_imshow(): + # Check if environment supports image displays + try: + assert not is_docker(), 'cv2.imshow() is disabled in Docker environments' + assert not is_colab(), 'cv2.imshow() is disabled in Google Colab environments' + cv2.imshow('test', np.zeros((1, 1, 3))) + cv2.waitKey(1) + cv2.destroyAllWindows() + cv2.waitKey(1) + return True + except Exception as e: + LOGGER.warning(f'WARNING: Environment does not support cv2.imshow() or PIL Image.show() image displays\n{e}') + return False + + +def check_suffix(file='yolov5s.pt', suffix=('.pt',), msg=''): + # Check file(s) for acceptable suffix + if file and suffix: + if isinstance(suffix, str): + suffix = [suffix] + for f in file if isinstance(file, (list, tuple)) else [file]: + s = Path(f).suffix.lower() # file suffix + if len(s): + assert s in suffix, f"{msg}{f} acceptable suffix is {suffix}" + + +def check_yaml(file, suffix=('.yaml', '.yml')): + # Search/download YAML file (if necessary) and return path, checking suffix + return check_file(file, suffix) + + +def check_file(file, suffix=''): + # Search/download file (if necessary) and return path + check_suffix(file, suffix) # optional + file = str(file) # convert to str() + if Path(file).is_file() or not file: # exists + return file + elif file.startswith(('http:/', 'https:/')): # download + url = file # warning: Pathlib turns :// -> :/ + file = Path(urllib.parse.unquote(file).split('?')[0]).name # '%2F' to '/', split https://url.com/file.txt?auth + if Path(file).is_file(): + LOGGER.info(f'Found {url} locally at {file}') # file already exists + else: + LOGGER.info(f'Downloading {url} to {file}...') + torch.hub.download_url_to_file(url, file) + assert Path(file).exists() and Path(file).stat().st_size > 0, f'File download failed: {url}' # check + return file + elif file.startswith('clearml://'): # ClearML Dataset ID + assert 'clearml' in sys.modules, "ClearML is not installed, so cannot use ClearML dataset. Try running 'pip install clearml'." + return file + else: # search + files = [] + for d in 'data', 'models', 'utils': # search directories + files.extend(glob.glob(str(ROOT / d / '**' / file), recursive=True)) # find file + assert len(files), f'File not found: {file}' # assert file was found + assert len(files) == 1, f"Multiple files match '{file}', specify exact path: {files}" # assert unique + return files[0] # return file + + +def check_font(font=FONT, progress=False): + # Download font to CONFIG_DIR if necessary + font = Path(font) + file = CONFIG_DIR / font.name + if not font.exists() and not file.exists(): + url = "https://ultralytics.com/assets/" + font.name + LOGGER.info(f'Downloading {url} to {file}...') + torch.hub.download_url_to_file(url, str(file), progress=progress) + + +def check_dataset(data, autodownload=True): + # Download, check and/or unzip dataset if not found locally + + # Download (optional) + extract_dir = '' + if isinstance(data, (str, Path)) and str(data).endswith('.zip'): # i.e. gs://bucket/dir/coco128.zip + download(data, dir=f'{DATASETS_DIR}/{Path(data).stem}', unzip=True, delete=False, curl=False, threads=1) + data = next((DATASETS_DIR / Path(data).stem).rglob('*.yaml')) + extract_dir, autodownload = data.parent, False + + # Read yaml (optional) + if isinstance(data, (str, Path)): + with open(data, errors='ignore') as f: + data = yaml.safe_load(f) # dictionary + + # Checks + for k in 'train', 'val', 'nc': + assert k in data, f"data.yaml '{k}:' field missing 鉂" + if 'names' not in data: + LOGGER.warning("data.yaml 'names:' field missing 鈿狅笍, assigning default names 'class0', 'class1', etc.") + data['names'] = [f'class{i}' for i in range(data['nc'])] # default names + + # Resolve paths + path = Path(extract_dir or data.get('path') or '') # optional 'path' default to '.' + if not path.is_absolute(): + path = (ROOT / path).resolve() + data['path'] = str(path) + for k in 'train', 'val', 'test': + if data.get(k): # prepend path + data[k] = str(path / data[k]) if isinstance(data[k], str) else [str(path / x) for x in data[k]] + + # Parse yaml + train, val, test, s = (data.get(x) for x in ('train', 'val', 'test', 'download')) + if val: + val = [Path(x).resolve() for x in (val if isinstance(val, list) else [val])] # val path + if not all(x.exists() for x in val): + LOGGER.info('\nDataset not found 鈿狅笍, missing paths %s' % [str(x) for x in val if not x.exists()]) + if not s or not autodownload: + raise Exception('Dataset not found 鉂') + t = time.time() + root = path.parent if 'path' in data else '..' # unzip directory i.e. '../' + if s.startswith('http') and s.endswith('.zip'): # URL + f = Path(s).name # filename + LOGGER.info(f'Downloading {s} to {f}...') + torch.hub.download_url_to_file(s, f) + Path(root).mkdir(parents=True, exist_ok=True) # create root + ZipFile(f).extractall(path=root) # unzip + Path(f).unlink() # remove zip + r = None # success + elif s.startswith('bash '): # bash script + LOGGER.info(f'Running {s} ...') + r = os.system(s) + else: # python script + r = exec(s, {'yaml': data}) # return None + dt = f'({round(time.time() - t, 1)}s)' + s = f"success 鉁 {dt}, saved to {colorstr('bold', root)}" if r in (0, None) else f"failure {dt} 鉂" + LOGGER.info(f"Dataset download {s}") + check_font('Arial.ttf' if is_ascii(data['names']) else 'Arial.Unicode.ttf', progress=True) # download fonts + return data # dictionary + + +def check_amp(model): + # Check PyTorch Automatic Mixed Precision (AMP) functionality. Return True on correct operation + from models.common import AutoShape, DetectMultiBackend + + def amp_allclose(model, im): + # All close FP32 vs AMP results + m = AutoShape(model, verbose=False) # model + a = m(im).xywhn[0] # FP32 inference + m.amp = True + b = m(im).xywhn[0] # AMP inference + return a.shape == b.shape and torch.allclose(a, b, atol=0.1) # close to 10% absolute tolerance + + prefix = colorstr('AMP: ') + device = next(model.parameters()).device # get model device + if device.type == 'cpu': + return False # AMP disabled on CPU + f = ROOT / 'data' / 'images' / 'bus.jpg' # image to check + im = f if f.exists() else 'https://ultralytics.com/images/bus.jpg' if check_online() else np.ones((640, 640, 3)) + try: + assert amp_allclose(model, im) or amp_allclose(DetectMultiBackend('yolov5n.pt', device), im) + LOGGER.info(f'{prefix}checks passed 鉁') + return True + except Exception: + help_url = 'https://github.com/ultralytics/yolov5/issues/7908' + LOGGER.warning(f'{prefix}checks failed 鉂, disabling Automatic Mixed Precision. See {help_url}') + return False + + +def yaml_load(file='data.yaml'): + # Single-line safe yaml loading + with open(file, errors='ignore') as f: + return yaml.safe_load(f) + + +def yaml_save(file='data.yaml', data={}): + # Single-line safe yaml saving + with open(file, 'w') as f: + yaml.safe_dump({k: str(v) if isinstance(v, Path) else v for k, v in data.items()}, f, sort_keys=False) + + +def url2file(url): + # Convert URL to filename, i.e. https://url.com/file.txt?auth -> file.txt + url = str(Path(url)).replace(':/', '://') # Pathlib turns :// -> :/ + return Path(urllib.parse.unquote(url)).name.split('?')[0] # '%2F' to '/', split https://url.com/file.txt?auth + + +def download(url, dir='.', unzip=True, delete=True, curl=False, threads=1, retry=3): + # Multi-threaded file download and unzip function, used in data.yaml for autodownload + def download_one(url, dir): + # Download 1 file + success = True + f = dir / Path(url).name # filename + if Path(url).is_file(): # exists in current path + Path(url).rename(f) # move to dir + elif not f.exists(): + LOGGER.info(f'Downloading {url} to {f}...') + for i in range(retry + 1): + if curl: + s = 'sS' if threads > 1 else '' # silent + r = os.system(f'curl -{s}L "{url}" -o "{f}" --retry 9 -C -') # curl download with retry, continue + success = r == 0 + else: + torch.hub.download_url_to_file(url, f, progress=threads == 1) # torch download + success = f.is_file() + if success: + break + elif i < retry: + LOGGER.warning(f'Download failure, retrying {i + 1}/{retry} {url}...') + else: + LOGGER.warning(f'Failed to download {url}...') + + if unzip and success and f.suffix in ('.zip', '.tar', '.gz'): + LOGGER.info(f'Unzipping {f}...') + if f.suffix == '.zip': + ZipFile(f).extractall(path=dir) # unzip + elif f.suffix == '.tar': + os.system(f'tar xf {f} --directory {f.parent}') # unzip + elif f.suffix == '.gz': + os.system(f'tar xfz {f} --directory {f.parent}') # unzip + if delete: + f.unlink() # remove zip + + dir = Path(dir) + dir.mkdir(parents=True, exist_ok=True) # make directory + if threads > 1: + pool = ThreadPool(threads) + pool.imap(lambda x: download_one(*x), zip(url, repeat(dir))) # multi-threaded + pool.close() + pool.join() + else: + for u in [url] if isinstance(url, (str, Path)) else url: + download_one(u, dir) + + +def make_divisible(x, divisor): + # Returns nearest x divisible by divisor + if isinstance(divisor, torch.Tensor): + divisor = int(divisor.max()) # to int + return math.ceil(x / divisor) * divisor + + +def clean_str(s): + # Cleans a string by replacing special characters with underscore _ + return re.sub(pattern="[|@#!隆路$鈧%&()=?驴^*;:,篓麓><+]", repl="_", string=s) + + +def one_cycle(y1=0.0, y2=1.0, steps=100): + # lambda function for sinusoidal ramp from y1 to y2 https://arxiv.org/pdf/1812.01187.pdf + return lambda x: ((1 - math.cos(x * math.pi / steps)) / 2) * (y2 - y1) + y1 + + +def colorstr(*input): + # Colors a string https://en.wikipedia.org/wiki/ANSI_escape_code, i.e. colorstr('blue', 'hello world') + *args, string = input if len(input) > 1 else ('blue', 'bold', input[0]) # color arguments, string + colors = { + 'black': '\033[30m', # basic colors + 'red': '\033[31m', + 'green': '\033[32m', + 'yellow': '\033[33m', + 'blue': '\033[34m', + 'magenta': '\033[35m', + 'cyan': '\033[36m', + 'white': '\033[37m', + 'bright_black': '\033[90m', # bright colors + 'bright_red': '\033[91m', + 'bright_green': '\033[92m', + 'bright_yellow': '\033[93m', + 'bright_blue': '\033[94m', + 'bright_magenta': '\033[95m', + 'bright_cyan': '\033[96m', + 'bright_white': '\033[97m', + 'end': '\033[0m', # misc + 'bold': '\033[1m', + 'underline': '\033[4m'} + return ''.join(colors[x] for x in args) + f'{string}' + colors['end'] + + +def labels_to_class_weights(labels, nc=80): + # Get class weights (inverse frequency) from training labels + if labels[0] is None: # no labels loaded + return torch.Tensor() + + labels = np.concatenate(labels, 0) # labels.shape = (866643, 5) for COCO + classes = labels[:, 0].astype(int) # labels = [class xywh] + weights = np.bincount(classes, minlength=nc) # occurrences per class + + # Prepend gridpoint count (for uCE training) + # gpi = ((320 / 32 * np.array([1, 2, 4])) ** 2 * 3).sum() # gridpoints per image + # weights = np.hstack([gpi * len(labels) - weights.sum() * 9, weights * 9]) ** 0.5 # prepend gridpoints to start + + weights[weights == 0] = 1 # replace empty bins with 1 + weights = 1 / weights # number of targets per class + weights /= weights.sum() # normalize + return torch.from_numpy(weights).float() + + +def labels_to_image_weights(labels, nc=80, class_weights=np.ones(80)): + # Produces image weights based on class_weights and image contents + # Usage: index = random.choices(range(n), weights=image_weights, k=1) # weighted image sample + class_counts = np.array([np.bincount(x[:, 0].astype(int), minlength=nc) for x in labels]) + return (class_weights.reshape(1, nc) * class_counts).sum(1) + + +def coco80_to_coco91_class(): # converts 80-index (val2014) to 91-index (paper) + # https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/ + # a = np.loadtxt('data/coco.names', dtype='str', delimiter='\n') + # b = np.loadtxt('data/coco_paper.names', dtype='str', delimiter='\n') + # x1 = [list(a[i] == b).index(True) + 1 for i in range(80)] # darknet to coco + # x2 = [list(b[i] == a).index(True) if any(b[i] == a) else None for i in range(91)] # coco to darknet + return [ + 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 31, 32, 33, 34, + 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, + 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85, 86, 87, 88, 89, 90] + + +def xyxy2xywh(x): + # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] where xy1=top-left, xy2=bottom-right + y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) + y[:, 0] = (x[:, 0] + x[:, 2]) / 2 # x center + y[:, 1] = (x[:, 1] + x[:, 3]) / 2 # y center + y[:, 2] = x[:, 2] - x[:, 0] # width + y[:, 3] = x[:, 3] - x[:, 1] # height + return y + + +def xywh2xyxy(x): + # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right + y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) + y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x + y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y + y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x + y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y + return y + + +def xywhn2xyxy(x, w=640, h=640, padw=0, padh=0): + # Convert nx4 boxes from [x, y, w, h] normalized to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right + y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) + y[:, 0] = w * (x[:, 0] - x[:, 2] / 2) + padw # top left x + y[:, 1] = h * (x[:, 1] - x[:, 3] / 2) + padh # top left y + y[:, 2] = w * (x[:, 0] + x[:, 2] / 2) + padw # bottom right x + y[:, 3] = h * (x[:, 1] + x[:, 3] / 2) + padh # bottom right y + return y + + +def xyxy2xywhn(x, w=640, h=640, clip=False, eps=0.0): + # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] normalized where xy1=top-left, xy2=bottom-right + if clip: + clip_coords(x, (h - eps, w - eps)) # warning: inplace clip + y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) + y[:, 0] = ((x[:, 0] + x[:, 2]) / 2) / w # x center + y[:, 1] = ((x[:, 1] + x[:, 3]) / 2) / h # y center + y[:, 2] = (x[:, 2] - x[:, 0]) / w # width + y[:, 3] = (x[:, 3] - x[:, 1]) / h # height + return y + + +def xyn2xy(x, w=640, h=640, padw=0, padh=0): + # Convert normalized segments into pixel segments, shape (n,2) + y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) + y[:, 0] = w * x[:, 0] + padw # top left x + y[:, 1] = h * x[:, 1] + padh # top left y + return y + + +def segment2box(segment, width=640, height=640): + # Convert 1 segment label to 1 box label, applying inside-image constraint, i.e. (xy1, xy2, ...) to (xyxy) + x, y = segment.T # segment xy + inside = (x >= 0) & (y >= 0) & (x <= width) & (y <= height) + x, y, = x[inside], y[inside] + return np.array([x.min(), y.min(), x.max(), y.max()]) if any(x) else np.zeros((1, 4)) # xyxy + + +def segments2boxes(segments): + # Convert segment labels to box labels, i.e. (cls, xy1, xy2, ...) to (cls, xywh) + boxes = [] + for s in segments: + x, y = s.T # segment xy + boxes.append([x.min(), y.min(), x.max(), y.max()]) # cls, xyxy + return xyxy2xywh(np.array(boxes)) # cls, xywh + + +def resample_segments(segments, n=1000): + # Up-sample an (n,2) segment + for i, s in enumerate(segments): + s = np.concatenate((s, s[0:1, :]), axis=0) + x = np.linspace(0, len(s) - 1, n) + xp = np.arange(len(s)) + segments[i] = np.concatenate([np.interp(x, xp, s[:, i]) for i in range(2)]).reshape(2, -1).T # segment xy + return segments + + +def scale_coords(img1_shape, coords, img0_shape, ratio_pad=None): + # Rescale coords (xyxy) from img1_shape to img0_shape + if ratio_pad is None: # calculate from img0_shape + gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new + pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding + else: + gain = ratio_pad[0][0] + pad = ratio_pad[1] + + coords[:, [0, 2]] -= pad[0] # x padding + coords[:, [1, 3]] -= pad[1] # y padding + coords[:, :4] /= gain + clip_coords(coords, img0_shape) + return coords + + +def clip_coords(boxes, shape): + # Clip bounding xyxy bounding boxes to image shape (height, width) + if isinstance(boxes, torch.Tensor): # faster individually + boxes[:, 0].clamp_(0, shape[1]) # x1 + boxes[:, 1].clamp_(0, shape[0]) # y1 + boxes[:, 2].clamp_(0, shape[1]) # x2 + boxes[:, 3].clamp_(0, shape[0]) # y2 + else: # np.array (faster grouped) + boxes[:, [0, 2]] = boxes[:, [0, 2]].clip(0, shape[1]) # x1, x2 + boxes[:, [1, 3]] = boxes[:, [1, 3]].clip(0, shape[0]) # y1, y2 + + +def non_max_suppression(prediction, + conf_thres=0.25, + iou_thres=0.45, + classes=None, + agnostic=False, + multi_label=False, + labels=(), + max_det=300): + """Non-Maximum Suppression (NMS) on inference results to reject overlapping bounding boxes + + Returns: + list of detections, on (n,6) tensor per image [xyxy, conf, cls] + """ + + bs = prediction.shape[0] # batch size + nc = prediction.shape[2] - 5 # number of classes + xc = prediction[..., 4] > conf_thres # candidates + + # Checks + assert 0 <= conf_thres <= 1, f'Invalid Confidence threshold {conf_thres}, valid values are between 0.0 and 1.0' + assert 0 <= iou_thres <= 1, f'Invalid IoU {iou_thres}, valid values are between 0.0 and 1.0' + + # Settings + # min_wh = 2 # (pixels) minimum box width and height + max_wh = 7680 # (pixels) maximum box width and height + max_nms = 30000 # maximum number of boxes into torchvision.ops.nms() + time_limit = 0.3 + 0.03 * bs # seconds to quit after + redundant = True # require redundant detections + multi_label &= nc > 1 # multiple labels per box (adds 0.5ms/img) + merge = False # use merge-NMS + + t = time.time() + output = [torch.zeros((0, 6), device=prediction.device)] * bs + for xi, x in enumerate(prediction): # image index, image inference + # Apply constraints + # x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height + x = x[xc[xi]] # confidence + + # Cat apriori labels if autolabelling + if labels and len(labels[xi]): + lb = labels[xi] + v = torch.zeros((len(lb), nc + 5), device=x.device) + v[:, :4] = lb[:, 1:5] # box + v[:, 4] = 1.0 # conf + v[range(len(lb)), lb[:, 0].long() + 5] = 1.0 # cls + x = torch.cat((x, v), 0) + + # If none remain process next image + if not x.shape[0]: + continue + + # Compute conf + x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf + + # Box (center x, center y, width, height) to (x1, y1, x2, y2) + box = xywh2xyxy(x[:, :4]) + + # Detections matrix nx6 (xyxy, conf, cls) + if multi_label: + i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T + x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1) + else: # best class only + conf, j = x[:, 5:].max(1, keepdim=True) + x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres] + + # Filter by class + if classes is not None: + x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)] + + # Apply finite constraint + # if not torch.isfinite(x).all(): + # x = x[torch.isfinite(x).all(1)] + + # Check shape + n = x.shape[0] # number of boxes + if not n: # no boxes + continue + elif n > max_nms: # excess boxes + x = x[x[:, 4].argsort(descending=True)[:max_nms]] # sort by confidence + + # Batched NMS + c = x[:, 5:6] * (0 if agnostic else max_wh) # classes + boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores + i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS + if i.shape[0] > max_det: # limit detections + i = i[:max_det] + if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean) + # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4) + iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix + weights = iou * scores[None] # box weights + x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes + if redundant: + i = i[iou.sum(1) > 1] # require redundancy + + output[xi] = x[i] + if (time.time() - t) > time_limit: + LOGGER.warning(f'WARNING: NMS time limit {time_limit:.3f}s exceeded') + break # time limit exceeded + + return output + + +def strip_optimizer(f='best.pt', s=''): # from utils.general import *; strip_optimizer() + # Strip optimizer from 'f' to finalize training, optionally save as 's' + x = torch.load(f, map_location=torch.device('cpu')) + if x.get('ema'): + x['model'] = x['ema'] # replace model with ema + for k in 'optimizer', 'best_fitness', 'wandb_id', 'ema', 'updates': # keys + x[k] = None + x['epoch'] = -1 + x['model'].half() # to FP16 + for p in x['model'].parameters(): + p.requires_grad = False + torch.save(x, s or f) + mb = os.path.getsize(s or f) / 1E6 # filesize + LOGGER.info(f"Optimizer stripped from {f},{f' saved as {s},' if s else ''} {mb:.1f}MB") + + +def print_mutation(results, hyp, save_dir, bucket, prefix=colorstr('evolve: ')): + evolve_csv = save_dir / 'evolve.csv' + evolve_yaml = save_dir / 'hyp_evolve.yaml' + keys = ('metrics/precision', 'metrics/recall', 'metrics/mAP_0.5', 'metrics/mAP_0.5:0.95', 'val/box_loss', + 'val/obj_loss', 'val/cls_loss') + tuple(hyp.keys()) # [results + hyps] + keys = tuple(x.strip() for x in keys) + vals = results + tuple(hyp.values()) + n = len(keys) + + # Download (optional) + if bucket: + url = f'gs://{bucket}/evolve.csv' + if gsutil_getsize(url) > (evolve_csv.stat().st_size if evolve_csv.exists() else 0): + os.system(f'gsutil cp {url} {save_dir}') # download evolve.csv if larger than local + + # Log to evolve.csv + s = '' if evolve_csv.exists() else (('%20s,' * n % keys).rstrip(',') + '\n') # add header + with open(evolve_csv, 'a') as f: + f.write(s + ('%20.5g,' * n % vals).rstrip(',') + '\n') + + # Save yaml + with open(evolve_yaml, 'w') as f: + data = pd.read_csv(evolve_csv) + data = data.rename(columns=lambda x: x.strip()) # strip keys + i = np.argmax(fitness(data.values[:, :4])) # + generations = len(data) + f.write('# YOLOv5 Hyperparameter Evolution Results\n' + f'# Best generation: {i}\n' + + f'# Last generation: {generations - 1}\n' + '# ' + ', '.join(f'{x.strip():>20s}' for x in keys[:7]) + + '\n' + '# ' + ', '.join(f'{x:>20.5g}' for x in data.values[i, :7]) + '\n\n') + yaml.safe_dump(data.loc[i][7:].to_dict(), f, sort_keys=False) + + # Print to screen + LOGGER.info(prefix + f'{generations} generations finished, current result:\n' + prefix + + ', '.join(f'{x.strip():>20s}' for x in keys) + '\n' + prefix + ', '.join(f'{x:20.5g}' + for x in vals) + '\n\n') + + if bucket: + os.system(f'gsutil cp {evolve_csv} {evolve_yaml} gs://{bucket}') # upload + + +def apply_classifier(x, model, img, im0): + # Apply a second stage classifier to YOLO outputs + # Example model = torchvision.models.__dict__['efficientnet_b0'](pretrained=True).to(device).eval() + im0 = [im0] if isinstance(im0, np.ndarray) else im0 + for i, d in enumerate(x): # per image + if d is not None and len(d): + d = d.clone() + + # Reshape and pad cutouts + b = xyxy2xywh(d[:, :4]) # boxes + b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # rectangle to square + b[:, 2:] = b[:, 2:] * 1.3 + 30 # pad + d[:, :4] = xywh2xyxy(b).long() + + # Rescale boxes from img_size to im0 size + scale_coords(img.shape[2:], d[:, :4], im0[i].shape) + + # Classes + pred_cls1 = d[:, 5].long() + ims = [] + for a in d: + cutout = im0[i][int(a[1]):int(a[3]), int(a[0]):int(a[2])] + im = cv2.resize(cutout, (224, 224)) # BGR + + im = im[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 + im = np.ascontiguousarray(im, dtype=np.float32) # uint8 to float32 + im /= 255 # 0 - 255 to 0.0 - 1.0 + ims.append(im) + + pred_cls2 = model(torch.Tensor(ims).to(d.device)).argmax(1) # classifier prediction + x[i] = x[i][pred_cls1 == pred_cls2] # retain matching class detections + + return x + + +def increment_path(path, exist_ok=False, sep='', mkdir=False): + # Increment file or directory path, i.e. runs/exp --> runs/exp{sep}2, runs/exp{sep}3, ... etc. + path = Path(path) # os-agnostic + if path.exists() and not exist_ok: + path, suffix = (path.with_suffix(''), path.suffix) if path.is_file() else (path, '') + + # Method 1 + for n in range(2, 9999): + p = f'{path}{sep}{n}{suffix}' # increment path + if not os.path.exists(p): # + break + path = Path(p) + + # Method 2 (deprecated) + # dirs = glob.glob(f"{path}{sep}*") # similar paths + # matches = [re.search(rf"{path.stem}{sep}(\d+)", d) for d in dirs] + # i = [int(m.groups()[0]) for m in matches if m] # indices + # n = max(i) + 1 if i else 2 # increment number + # path = Path(f"{path}{sep}{n}{suffix}") # increment path + + if mkdir: + path.mkdir(parents=True, exist_ok=True) # make directory + + return path + + +# OpenCV Chinese-friendly functions ------------------------------------------------------------------------------------ +imshow_ = cv2.imshow # copy to avoid recursion errors + + +def imread(path, flags=cv2.IMREAD_COLOR): + return cv2.imdecode(np.fromfile(path, np.uint8), flags) + + +def imwrite(path, im): + try: + cv2.imencode(Path(path).suffix, im)[1].tofile(path) + return True + except Exception: + return False + + +def imshow(path, im): + imshow_(path.encode('unicode_escape').decode(), im) + + +cv2.imread, cv2.imwrite, cv2.imshow = imread, imwrite, imshow # redefine + +# Variables ------------------------------------------------------------------------------------------------------------ +NCOLS = 0 if is_docker() else shutil.get_terminal_size().columns # terminal window size for tqdm diff --git a/application/yolov5_example/utils/google_app_engine/Dockerfile b/application/yolov5_example/utils/google_app_engine/Dockerfile new file mode 100644 index 00000000..0155618f --- /dev/null +++ b/application/yolov5_example/utils/google_app_engine/Dockerfile @@ -0,0 +1,25 @@ +FROM gcr.io/google-appengine/python + +# Create a virtualenv for dependencies. This isolates these packages from +# system-level packages. +# Use -p python3 or -p python3.7 to select python version. Default is version 2. +RUN virtualenv /env -p python3 + +# Setting these environment variables are the same as running +# source /env/bin/activate. +ENV VIRTUAL_ENV /env +ENV PATH /env/bin:$PATH + +RUN apt-get update && apt-get install -y python-opencv + +# Copy the application's requirements.txt and run pip to install all +# dependencies into the virtualenv. +ADD requirements.txt /app/requirements.txt +RUN pip install -r /app/requirements.txt + +# Add the application source code. +ADD . /app + +# Run a WSGI server to serve the application. gunicorn must be declared as +# a dependency in requirements.txt. +CMD gunicorn -b :$PORT main:app diff --git a/application/yolov5_example/utils/google_app_engine/additional_requirements.txt b/application/yolov5_example/utils/google_app_engine/additional_requirements.txt new file mode 100644 index 00000000..42d7ffc0 --- /dev/null +++ b/application/yolov5_example/utils/google_app_engine/additional_requirements.txt @@ -0,0 +1,4 @@ +# add these requirements in your app on top of the existing ones +pip==21.1 +Flask==1.0.2 +gunicorn==19.9.0 diff --git a/application/yolov5_example/utils/google_app_engine/app.yaml b/application/yolov5_example/utils/google_app_engine/app.yaml new file mode 100644 index 00000000..5056b7c1 --- /dev/null +++ b/application/yolov5_example/utils/google_app_engine/app.yaml @@ -0,0 +1,14 @@ +runtime: custom +env: flex + +service: yolov5app + +liveness_check: + initial_delay_sec: 600 + +manual_scaling: + instances: 1 +resources: + cpu: 1 + memory_gb: 4 + disk_size_gb: 20 diff --git a/application/yolov5_example/utils/loggers/__init__.py b/application/yolov5_example/utils/loggers/__init__.py new file mode 100644 index 00000000..8ec846f8 --- /dev/null +++ b/application/yolov5_example/utils/loggers/__init__.py @@ -0,0 +1,308 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Logging utils +""" + +import os +import warnings +from pathlib import Path + +import pkg_resources as pkg +import torch +from torch.utils.tensorboard import SummaryWriter + +from utils.general import colorstr, cv2 +from utils.loggers.clearml.clearml_utils import ClearmlLogger +from utils.loggers.wandb.wandb_utils import WandbLogger +from utils.plots import plot_images, plot_results +from utils.torch_utils import de_parallel + +LOGGERS = ('csv', 'tb', 'wandb', 'clearml') # *.csv, TensorBoard, Weights & Biases, ClearML +RANK = int(os.getenv('RANK', -1)) + +try: + import wandb + + assert hasattr(wandb, '__version__') # verify package import not local dir + if pkg.parse_version(wandb.__version__) >= pkg.parse_version('0.12.2') and RANK in {0, -1}: + try: + wandb_login_success = wandb.login(timeout=30) + except wandb.errors.UsageError: # known non-TTY terminal issue + wandb_login_success = False + if not wandb_login_success: + wandb = None +except (ImportError, AssertionError): + wandb = None + +try: + import clearml + + assert hasattr(clearml, '__version__') # verify package import not local dir +except (ImportError, AssertionError): + clearml = None + + +class Loggers(): + # YOLOv5 Loggers class + def __init__(self, save_dir=None, weights=None, opt=None, hyp=None, logger=None, include=LOGGERS): + self.save_dir = save_dir + self.weights = weights + self.opt = opt + self.hyp = hyp + self.logger = logger # for printing results to console + self.include = include + self.keys = [ + 'train/box_loss', + 'train/obj_loss', + 'train/cls_loss', # train loss + 'metrics/precision', + 'metrics/recall', + 'metrics/mAP_0.5', + 'metrics/mAP_0.5:0.95', # metrics + 'val/box_loss', + 'val/obj_loss', + 'val/cls_loss', # val loss + 'x/lr0', + 'x/lr1', + 'x/lr2'] # params + self.best_keys = ['best/epoch', 'best/precision', 'best/recall', 'best/mAP_0.5', 'best/mAP_0.5:0.95'] + for k in LOGGERS: + setattr(self, k, None) # init empty logger dictionary + self.csv = True # always log to csv + + # Messages + if not wandb: + prefix = colorstr('Weights & Biases: ') + s = f"{prefix}run 'pip install wandb' to automatically track and visualize YOLOv5 馃殌 runs in Weights & Biases" + self.logger.info(s) + if not clearml: + prefix = colorstr('ClearML: ') + s = f"{prefix}run 'pip install clearml' to automatically track, visualize and remotely train YOLOv5 馃殌 in ClearML" + self.logger.info(s) + + # TensorBoard + s = self.save_dir + if 'tb' in self.include and not self.opt.evolve: + prefix = colorstr('TensorBoard: ') + self.logger.info(f"{prefix}Start with 'tensorboard --logdir {s.parent}', view at http://localhost:6006/") + self.tb = SummaryWriter(str(s)) + + # W&B + if wandb and 'wandb' in self.include: + wandb_artifact_resume = isinstance(self.opt.resume, str) and self.opt.resume.startswith('wandb-artifact://') + run_id = torch.load(self.weights).get('wandb_id') if self.opt.resume and not wandb_artifact_resume else None + self.opt.hyp = self.hyp # add hyperparameters + self.wandb = WandbLogger(self.opt, run_id) + # temp warn. because nested artifacts not supported after 0.12.10 + if pkg.parse_version(wandb.__version__) >= pkg.parse_version('0.12.11'): + s = "YOLOv5 temporarily requires wandb version 0.12.10 or below. Some features may not work as expected." + self.logger.warning(s) + else: + self.wandb = None + + # ClearML + if clearml and 'clearml' in self.include: + self.clearml = ClearmlLogger(self.opt, self.hyp) + else: + self.clearml = None + + def on_train_start(self): + # Callback runs on train start + pass + + def on_pretrain_routine_end(self): + # Callback runs on pre-train routine end + paths = self.save_dir.glob('*labels*.jpg') # training labels + if self.wandb: + self.wandb.log({"Labels": [wandb.Image(str(x), caption=x.name) for x in paths]}) + if self.clearml: + pass # ClearML saves these images automatically using hooks + + def on_train_batch_end(self, ni, model, imgs, targets, paths, plots): + # Callback runs on train batch end + # ni: number integrated batches (since train start) + if plots: + if ni == 0 and not self.opt.sync_bn and self.tb: + log_tensorboard_graph(self.tb, model, imgsz=list(imgs.shape[2:4])) + if ni < 3: + f = self.save_dir / f'train_batch{ni}.jpg' # filename + plot_images(imgs, targets, paths, f) + if (self.wandb or self.clearml) and ni == 10: + files = sorted(self.save_dir.glob('train*.jpg')) + if self.wandb: + self.wandb.log({'Mosaics': [wandb.Image(str(f), caption=f.name) for f in files if f.exists()]}) + if self.clearml: + self.clearml.log_debug_samples(files, title='Mosaics') + + def on_train_epoch_end(self, epoch): + # Callback runs on train epoch end + if self.wandb: + self.wandb.current_epoch = epoch + 1 + + def on_val_image_end(self, pred, predn, path, names, im): + # Callback runs on val image end + if self.wandb: + self.wandb.val_one_image(pred, predn, path, names, im) + if self.clearml: + self.clearml.log_image_with_boxes(path, pred, names, im) + + def on_val_end(self): + # Callback runs on val end + if self.wandb or self.clearml: + files = sorted(self.save_dir.glob('val*.jpg')) + if self.wandb: + self.wandb.log({"Validation": [wandb.Image(str(f), caption=f.name) for f in files]}) + if self.clearml: + self.clearml.log_debug_samples(files, title='Validation') + + def on_fit_epoch_end(self, vals, epoch, best_fitness, fi): + # Callback runs at the end of each fit (train+val) epoch + x = dict(zip(self.keys, vals)) + if self.csv: + file = self.save_dir / 'results.csv' + n = len(x) + 1 # number of cols + s = '' if file.exists() else (('%20s,' * n % tuple(['epoch'] + self.keys)).rstrip(',') + '\n') # add header + with open(file, 'a') as f: + f.write(s + ('%20.5g,' * n % tuple([epoch] + vals)).rstrip(',') + '\n') + + if self.tb: + for k, v in x.items(): + self.tb.add_scalar(k, v, epoch) + elif self.clearml: # log to ClearML if TensorBoard not used + for k, v in x.items(): + title, series = k.split('/') + self.clearml.task.get_logger().report_scalar(title, series, v, epoch) + + if self.wandb: + if best_fitness == fi: + best_results = [epoch] + vals[3:7] + for i, name in enumerate(self.best_keys): + self.wandb.wandb_run.summary[name] = best_results[i] # log best results in the summary + self.wandb.log(x) + self.wandb.end_epoch(best_result=best_fitness == fi) + + if self.clearml: + self.clearml.current_epoch_logged_images = set() # reset epoch image limit + self.clearml.current_epoch += 1 + + def on_model_save(self, last, epoch, final_epoch, best_fitness, fi): + # Callback runs on model save event + if self.wandb: + if ((epoch + 1) % self.opt.save_period == 0 and not final_epoch) and self.opt.save_period != -1: + self.wandb.log_model(last.parent, self.opt, epoch, fi, best_model=best_fitness == fi) + + if self.clearml: + if ((epoch + 1) % self.opt.save_period == 0 and not final_epoch) and self.opt.save_period != -1: + self.clearml.task.update_output_model(model_path=str(last), + model_name='Latest Model', + auto_delete_file=False) + + def on_train_end(self, last, best, plots, epoch, results): + # Callback runs on training end + if plots: + plot_results(file=self.save_dir / 'results.csv') # save results.png + files = ['results.png', 'confusion_matrix.png', *(f'{x}_curve.png' for x in ('F1', 'PR', 'P', 'R'))] + files = [(self.save_dir / f) for f in files if (self.save_dir / f).exists()] # filter + self.logger.info(f"Results saved to {colorstr('bold', self.save_dir)}") + + if self.tb and not self.clearml: # These images are already captured by ClearML by now, we don't want doubles + for f in files: + self.tb.add_image(f.stem, cv2.imread(str(f))[..., ::-1], epoch, dataformats='HWC') + + if self.wandb: + self.wandb.log(dict(zip(self.keys[3:10], results))) + self.wandb.log({"Results": [wandb.Image(str(f), caption=f.name) for f in files]}) + # Calling wandb.log. TODO: Refactor this into WandbLogger.log_model + if not self.opt.evolve: + wandb.log_artifact(str(best if best.exists() else last), + type='model', + name=f'run_{self.wandb.wandb_run.id}_model', + aliases=['latest', 'best', 'stripped']) + self.wandb.finish_run() + + if self.clearml: + # Save the best model here + if not self.opt.evolve: + self.clearml.task.update_output_model(model_path=str(best if best.exists() else last), + name='Best Model') + + def on_params_update(self, params): + # Update hyperparams or configs of the experiment + # params: A dict containing {param: value} pairs + if self.wandb: + self.wandb.wandb_run.config.update(params, allow_val_change=True) + + +class GenericLogger: + """ + YOLOv5 General purpose logger for non-task specific logging + Usage: from utils.loggers import GenericLogger; logger = GenericLogger(...) + Arguments + opt: Run arguments + console_logger: Console logger + include: loggers to include + """ + + def __init__(self, opt, console_logger, include=('tb', 'wandb')): + # init default loggers + self.save_dir = opt.save_dir + self.include = include + self.console_logger = console_logger + if 'tb' in self.include: + prefix = colorstr('TensorBoard: ') + self.console_logger.info( + f"{prefix}Start with 'tensorboard --logdir {self.save_dir.parent}', view at http://localhost:6006/") + self.tb = SummaryWriter(str(self.save_dir)) + + if wandb and 'wandb' in self.include: + self.wandb = wandb.init(project="YOLOv5-Classifier" if opt.project == "runs/train" else opt.project, + name=None if opt.name == "exp" else opt.name, + config=opt) + else: + self.wandb = None + + def log_metrics(self, metrics_dict, epoch): + # Log metrics dictionary to all loggers + if self.tb: + for k, v in metrics_dict.items(): + self.tb.add_scalar(k, v, epoch) + + if self.wandb: + self.wandb.log(metrics_dict, step=epoch) + + def log_images(self, files, name='Images', epoch=0): + # Log images to all loggers + files = [Path(f) for f in (files if isinstance(files, (tuple, list)) else [files])] # to Path + files = [f for f in files if f.exists()] # filter by exists + + if self.tb: + for f in files: + self.tb.add_image(f.stem, cv2.imread(str(f))[..., ::-1], epoch, dataformats='HWC') + + if self.wandb: + self.wandb.log({name: [wandb.Image(str(f), caption=f.name) for f in files]}, step=epoch) + + def log_graph(self, model, imgsz=(640, 640)): + # Log model graph to all loggers + if self.tb: + log_tensorboard_graph(self.tb, model, imgsz) + + def log_model(self, model_path, epoch=0, metadata={}): + # Log model to all loggers + if self.wandb: + art = wandb.Artifact(name=f"run_{wandb.run.id}_model", type="model", metadata=metadata) + art.add_file(str(model_path)) + wandb.log_artifact(art) + + +def log_tensorboard_graph(tb, model, imgsz=(640, 640)): + # Log model graph to TensorBoard + try: + p = next(model.parameters()) # for device, type + imgsz = (imgsz, imgsz) if isinstance(imgsz, int) else imgsz # expand + im = torch.zeros((1, 3, *imgsz)).to(p.device).type_as(p) # input image + with warnings.catch_warnings(): + warnings.simplefilter('ignore') # suppress jit trace warning + tb.add_graph(torch.jit.trace(de_parallel(model), im, strict=False), []) + except Exception: + print('WARNING: TensorBoard graph visualization failure') diff --git a/application/yolov5_example/utils/loggers/clearml/README.md b/application/yolov5_example/utils/loggers/clearml/README.md new file mode 100644 index 00000000..64eef6be --- /dev/null +++ b/application/yolov5_example/utils/loggers/clearml/README.md @@ -0,0 +1,222 @@ +# ClearML Integration + +Clear|MLClear|ML + +## About ClearML + +[ClearML](https://cutt.ly/yolov5-tutorial-clearml) is an [open-source](https://github.com/allegroai/clearml) toolbox designed to save you time 鈴憋笍. + +馃敤 Track every YOLOv5 training run in the experiment manager + +馃敡 Version and easily access your custom training data with the integrated ClearML Data Versioning Tool + +馃敠 Remotely train and monitor your YOLOv5 training runs using ClearML Agent + +馃敩 Get the very best mAP using ClearML Hyperparameter Optimization + +馃敪 Turn your newly trained YOLOv5 model into an API with just a few commands using ClearML Serving + +
+And so much more. It's up to you how many of these tools you want to use, you can stick to the experiment manager, or chain them all together into an impressive pipeline! +
+
+ +![ClearML scalars dashboard](https://github.com/thepycoder/clearml_screenshots/raw/main/experiment_manager_with_compare.gif) + + +
+
+ +## 馃 Setting Things Up + +To keep track of your experiments and/or data, ClearML needs to communicate to a server. You have 2 options to get one: + +Either sign up for free to the [ClearML Hosted Service](https://cutt.ly/yolov5-tutorial-clearml) or you can set up your own server, see [here](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server). Even the server is open-source, so even if you're dealing with sensitive data, you should be good to go! + +1. Install the `clearml` python package: + + ```bash + pip install clearml + ``` + +1. Connect the ClearML SDK to the server by [creating credentials](https://app.clear.ml/settings/workspace-configuration) (go right top to Settings -> Workspace -> Create new credentials), then execute the command below and follow the instructions: + + ```bash + clearml-init + ``` + +That's it! You're done 馃槑 + +
+ +## 馃殌 Training YOLOv5 With ClearML + +To enable ClearML experiment tracking, simply install the ClearML pip package. + +```bash +pip install clearml +``` + +This will enable integration with the YOLOv5 training script. Every training run from now on, will be captured and stored by the ClearML experiment manager. If you want to change the `project_name` or `task_name`, head over to our custom logger, where you can change it: `utils/loggers/clearml/clearml_utils.py` + +```bash +python train.py --img 640 --batch 16 --epochs 3 --data coco128.yaml --weights yolov5s.pt --cache +``` + +This will capture: +- Source code + uncommitted changes +- Installed packages +- (Hyper)parameters +- Model files (use `--save-period n` to save a checkpoint every n epochs) +- Console output +- Scalars (mAP_0.5, mAP_0.5:0.95, precision, recall, losses, learning rates, ...) +- General info such as machine details, runtime, creation date etc. +- All produced plots such as label correlogram and confusion matrix +- Images with bounding boxes per epoch +- Mosaic per epoch +- Validation images per epoch +- ... + +That's a lot right? 馃く +Now, we can visualize all of this information in the ClearML UI to get an overview of our training progress. Add custom columns to the table view (such as e.g. mAP_0.5) so you can easily sort on the best performing model. Or select multiple experiments and directly compare them! + +There even more we can do with all of this information, like hyperparameter optimization and remote execution, so keep reading if you want to see how that works! + +
+ +## 馃敆 Dataset Version Management + +Versioning your data separately from your code is generally a good idea and makes it easy to aqcuire the latest version too. This repository supports supplying a dataset version ID and it will make sure to get the data if it's not there yet. Next to that, this workflow also saves the used dataset ID as part of the task parameters, so you will always know for sure which data was used in which experiment! + +![ClearML Dataset Interface](https://github.com/thepycoder/clearml_screenshots/raw/main/clearml_data.gif) + +### Prepare Your Dataset + +The YOLOv5 repository supports a number of different datasets by using yaml files containing their information. By default datasets are downloaded to the `../datasets` folder in relation to the repository root folder. So if you downloaded the `coco128` dataset using the link in the yaml or with the scripts provided by yolov5, you get this folder structure: + +``` +.. +|_ yolov5 +|_ datasets + |_ coco128 + |_ images + |_ labels + |_ LICENSE + |_ README.txt +``` +But this can be any dataset you wish. Feel free to use your own, as long as you keep to this folder structure. + +Next, 鈿狅笍**copy the corresponding yaml file to the root of the dataset folder**鈿狅笍. This yaml files contains the information ClearML will need to properly use the dataset. You can make this yourself too, of course, just follow the structure of the example yamls. + +Basically we need the following keys: `path`, `train`, `test`, `val`, `nc`, `names`. + +``` +.. +|_ yolov5 +|_ datasets + |_ coco128 + |_ images + |_ labels + |_ coco128.yaml # <---- HERE! + |_ LICENSE + |_ README.txt +``` + +### Upload Your Dataset + +To get this dataset into ClearML as a versionned dataset, go to the dataset root folder and run the following command: +```bash +cd coco128 +clearml-data sync --project YOLOv5 --name coco128 --folder . +``` + +The command `clearml-data sync` is actually a shorthand command. You could also run these commands one after the other: +```bash +# Optionally add --parent if you want to base +# this version on another dataset version, so no duplicate files are uploaded! +clearml-data create --name coco128 --project YOLOv5 +clearml-data add --files . +clearml-data close +``` + +### Run Training Using A ClearML Dataset + +Now that you have a ClearML dataset, you can very simply use it to train custom YOLOv5 馃殌 models! + +```bash +python train.py --img 640 --batch 16 --epochs 3 --data clearml:// --weights yolov5s.pt --cache +``` + +
+ +## 馃憖 Hyperparameter Optimization + +Now that we have our experiments and data versioned, it's time to take a look at what we can build on top! + +Using the code information, installed packages and environment details, the experiment itself is now **completely reproducible**. In fact, ClearML allows you to clone an experiment and even change its parameters. We can then just rerun it with these new parameters automatically, this is basically what HPO does! + +To **run hyperparameter optimization locally**, we've included a pre-made script for you. Just make sure a training task has been run at least once, so it is in the ClearML experiment manager, we will essentially clone it and change its hyperparameters. + +You'll need to fill in the ID of this `template task` in the script found at `utils/loggers/clearml/hpo.py` and then just run it :) You can change `task.execute_locally()` to `task.execute()` to put it in a ClearML queue and have a remote agent work on it instead. + +```bash +# To use optuna, install it first, otherwise you can change the optimizer to just be RandomSearch +pip install optuna +python utils/loggers/clearml/hpo.py +``` + +![HPO](https://github.com/thepycoder/clearml_screenshots/raw/main/hpo.png) + +## 馃く Remote Execution (advanced) + +Running HPO locally is really handy, but what if we want to run our experiments on a remote machine instead? Maybe you have access to a very powerful GPU machine on-site or you have some budget to use cloud GPUs. +This is where the ClearML Agent comes into play. Check out what the agent can do here: + +- [YouTube video](https://youtu.be/MX3BrXnaULs) +- [Documentation](https://clear.ml/docs/latest/docs/clearml_agent) + +In short: every experiment tracked by the experiment manager contains enough information to reproduce it on a different machine (installed packages, uncommitted changes etc.). So a ClearML agent does just that: it listens to a queue for incoming tasks and when it finds one, it recreates the environment and runs it while still reporting scalars, plots etc. to the experiment manager. + +You can turn any machine (a cloud VM, a local GPU machine, your own laptop ... ) into a ClearML agent by simply running: +```bash +clearml-agent daemon --queue [--docker] +``` + +### Cloning, Editing And Enqueuing + +With our agent running, we can give it some work. Remember from the HPO section that we can clone a task and edit the hyperparameters? We can do that from the interface too! + +馃獎 Clone the experiment by right clicking it + +馃幆 Edit the hyperparameters to what you wish them to be + +鈴 Enqueue the task to any of the queues by right clicking it + +![Enqueue a task from the UI](https://github.com/thepycoder/clearml_screenshots/raw/main/enqueue.gif) + +### Executing A Task Remotely + +Now you can clone a task like we explained above, or simply mark your current script by adding `task.execute_remotely()` and on execution it will be put into a queue, for the agent to start working on! + +To run the YOLOv5 training script remotely, all you have to do is add this line to the training.py script after the clearml logger has been instatiated: +```python +# ... +# Loggers +data_dict = None +if RANK in {-1, 0}: + loggers = Loggers(save_dir, weights, opt, hyp, LOGGER) # loggers instance + if loggers.clearml: + loggers.clearml.task.execute_remotely(queue='my_queue') # <------ ADD THIS LINE + # Data_dict is either None is user did not choose for ClearML dataset or is filled in by ClearML + data_dict = loggers.clearml.data_dict +# ... +``` +When running the training script after this change, python will run the script up until that line, after which it will package the code and send it to the queue instead! + +### Autoscaling workers + +ClearML comes with autoscalers too! This tool will automatically spin up new remote machines in the cloud of your choice (AWS, GCP, Azure) and turn them into ClearML agents for you whenever there are experiments detected in the queue. Once the tasks are processed, the autoscaler will automatically shut down the remote machines and you stop paying! + +Check out the autoscalers getting started video below. + +[![Watch the video](https://img.youtube.com/vi/j4XVMAaUt3E/0.jpg)](https://youtu.be/j4XVMAaUt3E) diff --git a/application/yolov5_example/utils/loggers/clearml/__init__.py b/application/yolov5_example/utils/loggers/clearml/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/application/yolov5_example/utils/loggers/clearml/clearml_utils.py b/application/yolov5_example/utils/loggers/clearml/clearml_utils.py new file mode 100644 index 00000000..52320c09 --- /dev/null +++ b/application/yolov5_example/utils/loggers/clearml/clearml_utils.py @@ -0,0 +1,156 @@ +"""Main Logger class for ClearML experiment tracking.""" +import glob +import re +from pathlib import Path + +import numpy as np +import yaml + +from utils.plots import Annotator, colors + +try: + import clearml + from clearml import Dataset, Task + assert hasattr(clearml, '__version__') # verify package import not local dir +except (ImportError, AssertionError): + clearml = None + + +def construct_dataset(clearml_info_string): + """Load in a clearml dataset and fill the internal data_dict with its contents. + """ + dataset_id = clearml_info_string.replace('clearml://', '') + dataset = Dataset.get(dataset_id=dataset_id) + dataset_root_path = Path(dataset.get_local_copy()) + + # We'll search for the yaml file definition in the dataset + yaml_filenames = list(glob.glob(str(dataset_root_path / "*.yaml")) + glob.glob(str(dataset_root_path / "*.yml"))) + if len(yaml_filenames) > 1: + raise ValueError('More than one yaml file was found in the dataset root, cannot determine which one contains ' + 'the dataset definition this way.') + elif len(yaml_filenames) == 0: + raise ValueError('No yaml definition found in dataset root path, check that there is a correct yaml file ' + 'inside the dataset root path.') + with open(yaml_filenames[0]) as f: + dataset_definition = yaml.safe_load(f) + + assert set(dataset_definition.keys()).issuperset( + {'train', 'test', 'val', 'nc', 'names'} + ), "The right keys were not found in the yaml file, make sure it at least has the following keys: ('train', 'test', 'val', 'nc', 'names')" + + data_dict = dict() + data_dict['train'] = str( + (dataset_root_path / dataset_definition['train']).resolve()) if dataset_definition['train'] else None + data_dict['test'] = str( + (dataset_root_path / dataset_definition['test']).resolve()) if dataset_definition['test'] else None + data_dict['val'] = str( + (dataset_root_path / dataset_definition['val']).resolve()) if dataset_definition['val'] else None + data_dict['nc'] = dataset_definition['nc'] + data_dict['names'] = dataset_definition['names'] + + return data_dict + + +class ClearmlLogger: + """Log training runs, datasets, models, and predictions to ClearML. + + This logger sends information to ClearML at app.clear.ml or to your own hosted server. By default, + this information includes hyperparameters, system configuration and metrics, model metrics, code information and + basic data metrics and analyses. + + By providing additional command line arguments to train.py, datasets, + models and predictions can also be logged. + """ + + def __init__(self, opt, hyp): + """ + - Initialize ClearML Task, this object will capture the experiment + - Upload dataset version to ClearML Data if opt.upload_dataset is True + + arguments: + opt (namespace) -- Commandline arguments for this run + hyp (dict) -- Hyperparameters for this run + + """ + self.current_epoch = 0 + # Keep tracked of amount of logged images to enforce a limit + self.current_epoch_logged_images = set() + # Maximum number of images to log to clearML per epoch + self.max_imgs_to_log_per_epoch = 16 + # Get the interval of epochs when bounding box images should be logged + self.bbox_interval = opt.bbox_interval + self.clearml = clearml + self.task = None + self.data_dict = None + if self.clearml: + self.task = Task.init( + project_name='YOLOv5', + task_name='training', + tags=['YOLOv5'], + output_uri=True, + auto_connect_frameworks={'pytorch': False} + # We disconnect pytorch auto-detection, because we added manual model save points in the code + ) + # ClearML's hooks will already grab all general parameters + # Only the hyperparameters coming from the yaml config file + # will have to be added manually! + self.task.connect(hyp, name='Hyperparameters') + + # Get ClearML Dataset Version if requested + if opt.data.startswith('clearml://'): + # data_dict should have the following keys: + # names, nc (number of classes), test, train, val (all three relative paths to ../datasets) + self.data_dict = construct_dataset(opt.data) + # Set data to data_dict because wandb will crash without this information and opt is the best way + # to give it to them + opt.data = self.data_dict + + def log_debug_samples(self, files, title='Debug Samples'): + """ + Log files (images) as debug samples in the ClearML task. + + arguments: + files (List(PosixPath)) a list of file paths in PosixPath format + title (str) A title that groups together images with the same values + """ + for f in files: + if f.exists(): + it = re.search(r'_batch(\d+)', f.name) + iteration = int(it.groups()[0]) if it else 0 + self.task.get_logger().report_image(title=title, + series=f.name.replace(it.group(), ''), + local_path=str(f), + iteration=iteration) + + def log_image_with_boxes(self, image_path, boxes, class_names, image, conf_threshold=0.25): + """ + Draw the bounding boxes on a single image and report the result as a ClearML debug sample. + + arguments: + image_path (PosixPath) the path the original image file + boxes (list): list of scaled predictions in the format - [xmin, ymin, xmax, ymax, confidence, class] + class_names (dict): dict containing mapping of class int to class name + image (Tensor): A torch tensor containing the actual image data + """ + if len(self.current_epoch_logged_images) < self.max_imgs_to_log_per_epoch and self.current_epoch >= 0: + # Log every bbox_interval times and deduplicate for any intermittend extra eval runs + if self.current_epoch % self.bbox_interval == 0 and image_path not in self.current_epoch_logged_images: + im = np.ascontiguousarray(np.moveaxis(image.mul(255).clamp(0, 255).byte().cpu().numpy(), 0, 2)) + annotator = Annotator(im=im, pil=True) + for i, (conf, class_nr, box) in enumerate(zip(boxes[:, 4], boxes[:, 5], boxes[:, :4])): + color = colors(i) + + class_name = class_names[int(class_nr)] + confidence = round(float(conf) * 100, 2) + label = f"{class_name}: {confidence}%" + + if confidence > conf_threshold: + annotator.rectangle(box.cpu().numpy(), outline=color) + annotator.box_label(box.cpu().numpy(), label=label, color=color) + + annotated_image = annotator.result() + self.task.get_logger().report_image(title='Bounding Boxes', + series=image_path.name, + iteration=self.current_epoch, + image=annotated_image) + self.current_epoch_logged_images.add(image_path) diff --git a/application/yolov5_example/utils/loggers/clearml/hpo.py b/application/yolov5_example/utils/loggers/clearml/hpo.py new file mode 100644 index 00000000..96c2c544 --- /dev/null +++ b/application/yolov5_example/utils/loggers/clearml/hpo.py @@ -0,0 +1,84 @@ +from clearml import Task +# Connecting ClearML with the current process, +# from here on everything is logged automatically +from clearml.automation import HyperParameterOptimizer, UniformParameterRange +from clearml.automation.optuna import OptimizerOptuna + +task = Task.init(project_name='Hyper-Parameter Optimization', + task_name='YOLOv5', + task_type=Task.TaskTypes.optimizer, + reuse_last_task_id=False) + +# Example use case: +optimizer = HyperParameterOptimizer( + # This is the experiment we want to optimize + base_task_id='', + # here we define the hyper-parameters to optimize + # Notice: The parameter name should exactly match what you see in the UI: / + # For Example, here we see in the base experiment a section Named: "General" + # under it a parameter named "batch_size", this becomes "General/batch_size" + # If you have `argparse` for example, then arguments will appear under the "Args" section, + # and you should instead pass "Args/batch_size" + hyper_parameters=[ + UniformParameterRange('Hyperparameters/lr0', min_value=1e-5, max_value=1e-1), + UniformParameterRange('Hyperparameters/lrf', min_value=0.01, max_value=1.0), + UniformParameterRange('Hyperparameters/momentum', min_value=0.6, max_value=0.98), + UniformParameterRange('Hyperparameters/weight_decay', min_value=0.0, max_value=0.001), + UniformParameterRange('Hyperparameters/warmup_epochs', min_value=0.0, max_value=5.0), + UniformParameterRange('Hyperparameters/warmup_momentum', min_value=0.0, max_value=0.95), + UniformParameterRange('Hyperparameters/warmup_bias_lr', min_value=0.0, max_value=0.2), + UniformParameterRange('Hyperparameters/box', min_value=0.02, max_value=0.2), + UniformParameterRange('Hyperparameters/cls', min_value=0.2, max_value=4.0), + UniformParameterRange('Hyperparameters/cls_pw', min_value=0.5, max_value=2.0), + UniformParameterRange('Hyperparameters/obj', min_value=0.2, max_value=4.0), + UniformParameterRange('Hyperparameters/obj_pw', min_value=0.5, max_value=2.0), + UniformParameterRange('Hyperparameters/iou_t', min_value=0.1, max_value=0.7), + UniformParameterRange('Hyperparameters/anchor_t', min_value=2.0, max_value=8.0), + UniformParameterRange('Hyperparameters/fl_gamma', min_value=0.0, max_value=4.0), + UniformParameterRange('Hyperparameters/hsv_h', min_value=0.0, max_value=0.1), + UniformParameterRange('Hyperparameters/hsv_s', min_value=0.0, max_value=0.9), + UniformParameterRange('Hyperparameters/hsv_v', min_value=0.0, max_value=0.9), + UniformParameterRange('Hyperparameters/degrees', min_value=0.0, max_value=45.0), + UniformParameterRange('Hyperparameters/translate', min_value=0.0, max_value=0.9), + UniformParameterRange('Hyperparameters/scale', min_value=0.0, max_value=0.9), + UniformParameterRange('Hyperparameters/shear', min_value=0.0, max_value=10.0), + UniformParameterRange('Hyperparameters/perspective', min_value=0.0, max_value=0.001), + UniformParameterRange('Hyperparameters/flipud', min_value=0.0, max_value=1.0), + UniformParameterRange('Hyperparameters/fliplr', min_value=0.0, max_value=1.0), + UniformParameterRange('Hyperparameters/mosaic', min_value=0.0, max_value=1.0), + UniformParameterRange('Hyperparameters/mixup', min_value=0.0, max_value=1.0), + UniformParameterRange('Hyperparameters/copy_paste', min_value=0.0, max_value=1.0)], + # this is the objective metric we want to maximize/minimize + objective_metric_title='metrics', + objective_metric_series='mAP_0.5', + # now we decide if we want to maximize it or minimize it (accuracy we maximize) + objective_metric_sign='max', + # let us limit the number of concurrent experiments, + # this in turn will make sure we do dont bombard the scheduler with experiments. + # if we have an auto-scaler connected, this, by proxy, will limit the number of machine + max_number_of_concurrent_tasks=1, + # this is the optimizer class (actually doing the optimization) + # Currently, we can choose from GridSearch, RandomSearch or OptimizerBOHB (Bayesian optimization Hyper-Band) + optimizer_class=OptimizerOptuna, + # If specified only the top K performing Tasks will be kept, the others will be automatically archived + save_top_k_tasks_only=5, # 5, + compute_time_limit=None, + total_max_jobs=20, + min_iteration_per_job=None, + max_iteration_per_job=None, +) + +# report every 10 seconds, this is way too often, but we are testing here +optimizer.set_report_period(10) +# You can also use the line below instead to run all the optimizer tasks locally, without using queues or agent +# an_optimizer.start_locally(job_complete_callback=job_complete_callback) +# set the time limit for the optimization process (2 hours) +optimizer.set_time_limit(in_minutes=120.0) +# Start the optimization process in the local environment +optimizer.start_locally() +# wait until process is done (notice we are controlling the optimization process in the background) +optimizer.wait() +# make sure background optimization stopped +optimizer.stop() + +print('We are done, good bye') diff --git a/application/yolov5_example/utils/loggers/wandb/README.md b/application/yolov5_example/utils/loggers/wandb/README.md new file mode 100644 index 00000000..d78324b4 --- /dev/null +++ b/application/yolov5_example/utils/loggers/wandb/README.md @@ -0,0 +1,162 @@ +馃摎 This guide explains how to use **Weights & Biases** (W&B) with YOLOv5 馃殌. UPDATED 29 September 2021. + +- [About Weights & Biases](#about-weights-&-biases) +- [First-Time Setup](#first-time-setup) +- [Viewing runs](#viewing-runs) +- [Disabling wandb](#disabling-wandb) +- [Advanced Usage: Dataset Versioning and Evaluation](#advanced-usage) +- [Reports: Share your work with the world!](#reports) + +## About Weights & Biases + +Think of [W&B](https://wandb.ai/site?utm_campaign=repo_yolo_wandbtutorial) like GitHub for machine learning models. With a few lines of code, save everything you need to debug, compare and reproduce your models 鈥 architecture, hyperparameters, git commits, model weights, GPU usage, and even datasets and predictions. + +Used by top researchers including teams at OpenAI, Lyft, Github, and MILA, W&B is part of the new standard of best practices for machine learning. How W&B can help you optimize your machine learning workflows: + +- [Debug](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#Free-2) model performance in real time +- [GPU usage](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#System-4) visualized automatically +- [Custom charts](https://wandb.ai/wandb/customizable-charts/reports/Powerful-Custom-Charts-To-Debug-Model-Peformance--VmlldzoyNzY4ODI) for powerful, extensible visualization +- [Share insights](https://wandb.ai/wandb/getting-started/reports/Visualize-Debug-Machine-Learning-Models--VmlldzoyNzY5MDk#Share-8) interactively with collaborators +- [Optimize hyperparameters](https://docs.wandb.com/sweeps) efficiently +- [Track](https://docs.wandb.com/artifacts) datasets, pipelines, and production models + +## First-Time Setup + +
+ Toggle Details +When you first train, W&B will prompt you to create a new account and will generate an **API key** for you. If you are an existing user you can retrieve your key from https://wandb.ai/authorize. This key is used to tell W&B where to log your data. You only need to supply your key once, and then it is remembered on the same device. + +W&B will create a cloud **project** (default is 'YOLOv5') for your training runs, and each new training run will be provided a unique run **name** within that project as project/name. You can also manually set your project and run name as: + +```shell +$ python train.py --project ... --name ... +``` + +YOLOv5 notebook example: Open In Colab Open In Kaggle +Screen Shot 2021-09-29 at 10 23 13 PM + +
+ +## Viewing Runs + +
+ Toggle Details +Run information streams from your environment to the W&B cloud console as you train. This allows you to monitor and even cancel runs in realtime . All important information is logged: + +- Training & Validation losses +- Metrics: Precision, Recall, mAP@0.5, mAP@0.5:0.95 +- Learning Rate over time +- A bounding box debugging panel, showing the training progress over time +- GPU: Type, **GPU Utilization**, power, temperature, **CUDA memory usage** +- System: Disk I/0, CPU utilization, RAM memory usage +- Your trained model as W&B Artifact +- Environment: OS and Python types, Git repository and state, **training command** + +

Weights & Biases dashboard

+
+ +## Disabling wandb + +- training after running `wandb disabled` inside that directory creates no wandb run + ![Screenshot (84)](https://user-images.githubusercontent.com/15766192/143441777-c780bdd7-7cb4-4404-9559-b4316030a985.png) + +- To enable wandb again, run `wandb online` + ![Screenshot (85)](https://user-images.githubusercontent.com/15766192/143441866-7191b2cb-22f0-4e0f-ae64-2dc47dc13078.png) + +## Advanced Usage + +You can leverage W&B artifacts and Tables integration to easily visualize and manage your datasets, models and training evaluations. Here are some quick examples to get you started. + +
+

1: Train and Log Evaluation simultaneousy

+ This is an extension of the previous section, but it'll also training after uploading the dataset. This also evaluation Table + Evaluation table compares your predictions and ground truths across the validation set for each epoch. It uses the references to the already uploaded datasets, + so no images will be uploaded from your system more than once. +
+ Usage + Code $ python train.py --upload_data val + +![Screenshot from 2021-11-21 17-40-06](https://user-images.githubusercontent.com/15766192/142761183-c1696d8c-3f38-45ab-991a-bb0dfd98ae7d.png) + +
+ +

2. Visualize and Version Datasets

+ Log, visualize, dynamically query, and understand your data with W&B Tables. You can use the following command to log your dataset as a W&B Table. This will generate a {dataset}_wandb.yaml file which can be used to train from dataset artifact. +
+ Usage + Code $ python utils/logger/wandb/log_dataset.py --project ... --name ... --data .. + +![Screenshot (64)](https://user-images.githubusercontent.com/15766192/128486078-d8433890-98a3-4d12-8986-b6c0e3fc64b9.png) + +
+ +

3: Train using dataset artifact

+ When you upload a dataset as described in the first section, you get a new config file with an added `_wandb` to its name. This file contains the information that + can be used to train a model directly from the dataset artifact. This also logs evaluation +
+ Usage + Code $ python train.py --data {data}_wandb.yaml + +![Screenshot (72)](https://user-images.githubusercontent.com/15766192/128979739-4cf63aeb-a76f-483f-8861-1c0100b938a5.png) + +
+ +

4: Save model checkpoints as artifacts

+ To enable saving and versioning checkpoints of your experiment, pass `--save_period n` with the base cammand, where `n` represents checkpoint interval. + You can also log both the dataset and model checkpoints simultaneously. If not passed, only the final model will be logged + +
+ Usage + Code $ python train.py --save_period 1 + +![Screenshot (68)](https://user-images.githubusercontent.com/15766192/128726138-ec6c1f60-639d-437d-b4ee-3acd9de47ef3.png) + +
+ +
+ +

5: Resume runs from checkpoint artifacts.

+Any run can be resumed using artifacts if the --resume argument starts with聽wandb-artifact://聽prefix followed by the run path, i.e,聽wandb-artifact://username/project/runid . This doesn't require the model checkpoint to be present on the local system. + +
+ Usage + Code $ python train.py --resume wandb-artifact://{run_path} + +![Screenshot (70)](https://user-images.githubusercontent.com/15766192/128728988-4e84b355-6c87-41ae-a591-14aecf45343e.png) + +
+ +

6: Resume runs from dataset artifact & checkpoint artifacts.

+ Local dataset or model checkpoints are not required. This can be used to resume runs directly on a different device + The syntax is same as the previous section, but you'll need to lof both the dataset and model checkpoints as artifacts, i.e, set bot --upload_dataset or + train from _wandb.yaml file and set --save_period + +
+ Usage + Code $ python train.py --resume wandb-artifact://{run_path} + +![Screenshot (70)](https://user-images.githubusercontent.com/15766192/128728988-4e84b355-6c87-41ae-a591-14aecf45343e.png) + +
+ + + +

Reports

+W&B Reports can be created from your saved runs for sharing online. Once a report is created you will receive a link you can use to publically share your results. Here is an example report created from the COCO128 tutorial trainings of all four YOLOv5 models ([link](https://wandb.ai/glenn-jocher/yolov5_tutorial/reports/YOLOv5-COCO128-Tutorial-Results--VmlldzozMDI5OTY)). + +Weights & Biases Reports + +## Environments + +YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including [CUDA](https://developer.nvidia.com/cuda)/[CUDNN](https://developer.nvidia.com/cudnn), [Python](https://www.python.org/) and [PyTorch](https://pytorch.org/) preinstalled): + +- **Google Colab and Kaggle** notebooks with free GPU: Open In Colab Open In Kaggle +- **Google Cloud** Deep Learning VM. See [GCP Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/GCP-Quickstart) +- **Amazon** Deep Learning AMI. See [AWS Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/AWS-Quickstart) +- **Docker Image**. See [Docker Quickstart Guide](https://github.com/ultralytics/yolov5/wiki/Docker-Quickstart) Docker Pulls + +## Status + +![CI CPU testing](https://github.com/ultralytics/yolov5/workflows/CI%20CPU%20testing/badge.svg) + +If this badge is green, all [YOLOv5 GitHub Actions](https://github.com/ultralytics/yolov5/actions) Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training ([train.py](https://github.com/ultralytics/yolov5/blob/master/train.py)), validation ([val.py](https://github.com/ultralytics/yolov5/blob/master/val.py)), inference ([detect.py](https://github.com/ultralytics/yolov5/blob/master/detect.py)) and export ([export.py](https://github.com/ultralytics/yolov5/blob/master/export.py)) on macOS, Windows, and Ubuntu every 24 hours and on every commit. diff --git a/application/yolov5_example/utils/loggers/wandb/__init__.py b/application/yolov5_example/utils/loggers/wandb/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/application/yolov5_example/utils/loggers/wandb/log_dataset.py b/application/yolov5_example/utils/loggers/wandb/log_dataset.py new file mode 100644 index 00000000..06e81fb6 --- /dev/null +++ b/application/yolov5_example/utils/loggers/wandb/log_dataset.py @@ -0,0 +1,27 @@ +import argparse + +from wandb_utils import WandbLogger + +from utils.general import LOGGER + +WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' + + +def create_dataset_artifact(opt): + logger = WandbLogger(opt, None, job_type='Dataset Creation') # TODO: return value unused + if not logger.wandb: + LOGGER.info("install wandb using `pip install wandb` to log the dataset") + + +if __name__ == '__main__': + parser = argparse.ArgumentParser() + parser.add_argument('--data', type=str, default='data/coco128.yaml', help='data.yaml path') + parser.add_argument('--single-cls', action='store_true', help='train as single-class dataset') + parser.add_argument('--project', type=str, default='YOLOv5', help='name of W&B Project') + parser.add_argument('--entity', default=None, help='W&B entity') + parser.add_argument('--name', type=str, default='log dataset', help='name of W&B run') + + opt = parser.parse_args() + opt.resume = False # Explicitly disallow resume check for dataset upload job + + create_dataset_artifact(opt) diff --git a/application/yolov5_example/utils/loggers/wandb/sweep.py b/application/yolov5_example/utils/loggers/wandb/sweep.py new file mode 100644 index 00000000..d49ea6f2 --- /dev/null +++ b/application/yolov5_example/utils/loggers/wandb/sweep.py @@ -0,0 +1,41 @@ +import sys +from pathlib import Path + +import wandb + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[3] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH + +from train import parse_opt, train +from utils.callbacks import Callbacks +from utils.general import increment_path +from utils.torch_utils import select_device + + +def sweep(): + wandb.init() + # Get hyp dict from sweep agent. Copy because train() modifies parameters which confused wandb. + hyp_dict = vars(wandb.config).get("_items").copy() + + # Workaround: get necessary opt args + opt = parse_opt(known=True) + opt.batch_size = hyp_dict.get("batch_size") + opt.save_dir = str(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok or opt.evolve)) + opt.epochs = hyp_dict.get("epochs") + opt.nosave = True + opt.data = hyp_dict.get("data") + opt.weights = str(opt.weights) + opt.cfg = str(opt.cfg) + opt.data = str(opt.data) + opt.hyp = str(opt.hyp) + opt.project = str(opt.project) + device = select_device(opt.device, batch_size=opt.batch_size) + + # train + train(hyp_dict, opt, device, callbacks=Callbacks()) + + +if __name__ == "__main__": + sweep() diff --git a/application/yolov5_example/utils/loggers/wandb/sweep.yaml b/application/yolov5_example/utils/loggers/wandb/sweep.yaml new file mode 100644 index 00000000..688b1ea0 --- /dev/null +++ b/application/yolov5_example/utils/loggers/wandb/sweep.yaml @@ -0,0 +1,143 @@ +# Hyperparameters for training +# To set range- +# Provide min and max values as: +# parameter: +# +# min: scalar +# max: scalar +# OR +# +# Set a specific list of search space- +# parameter: +# values: [scalar1, scalar2, scalar3...] +# +# You can use grid, bayesian and hyperopt search strategy +# For more info on configuring sweeps visit - https://docs.wandb.ai/guides/sweeps/configuration + +program: utils/loggers/wandb/sweep.py +method: random +metric: + name: metrics/mAP_0.5 + goal: maximize + +parameters: + # hyperparameters: set either min, max range or values list + data: + value: "data/coco128.yaml" + batch_size: + values: [64] + epochs: + values: [10] + + lr0: + distribution: uniform + min: 1e-5 + max: 1e-1 + lrf: + distribution: uniform + min: 0.01 + max: 1.0 + momentum: + distribution: uniform + min: 0.6 + max: 0.98 + weight_decay: + distribution: uniform + min: 0.0 + max: 0.001 + warmup_epochs: + distribution: uniform + min: 0.0 + max: 5.0 + warmup_momentum: + distribution: uniform + min: 0.0 + max: 0.95 + warmup_bias_lr: + distribution: uniform + min: 0.0 + max: 0.2 + box: + distribution: uniform + min: 0.02 + max: 0.2 + cls: + distribution: uniform + min: 0.2 + max: 4.0 + cls_pw: + distribution: uniform + min: 0.5 + max: 2.0 + obj: + distribution: uniform + min: 0.2 + max: 4.0 + obj_pw: + distribution: uniform + min: 0.5 + max: 2.0 + iou_t: + distribution: uniform + min: 0.1 + max: 0.7 + anchor_t: + distribution: uniform + min: 2.0 + max: 8.0 + fl_gamma: + distribution: uniform + min: 0.0 + max: 4.0 + hsv_h: + distribution: uniform + min: 0.0 + max: 0.1 + hsv_s: + distribution: uniform + min: 0.0 + max: 0.9 + hsv_v: + distribution: uniform + min: 0.0 + max: 0.9 + degrees: + distribution: uniform + min: 0.0 + max: 45.0 + translate: + distribution: uniform + min: 0.0 + max: 0.9 + scale: + distribution: uniform + min: 0.0 + max: 0.9 + shear: + distribution: uniform + min: 0.0 + max: 10.0 + perspective: + distribution: uniform + min: 0.0 + max: 0.001 + flipud: + distribution: uniform + min: 0.0 + max: 1.0 + fliplr: + distribution: uniform + min: 0.0 + max: 1.0 + mosaic: + distribution: uniform + min: 0.0 + max: 1.0 + mixup: + distribution: uniform + min: 0.0 + max: 1.0 + copy_paste: + distribution: uniform + min: 0.0 + max: 1.0 diff --git a/application/yolov5_example/utils/loggers/wandb/wandb_utils.py b/application/yolov5_example/utils/loggers/wandb/wandb_utils.py new file mode 100644 index 00000000..e850d2ac --- /dev/null +++ b/application/yolov5_example/utils/loggers/wandb/wandb_utils.py @@ -0,0 +1,584 @@ +"""Utilities and tools for tracking runs with Weights & Biases.""" + +import logging +import os +import sys +from contextlib import contextmanager +from pathlib import Path +from typing import Dict + +import yaml +from tqdm import tqdm + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[3] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH + +from utils.dataloaders import LoadImagesAndLabels, img2label_paths +from utils.general import LOGGER, check_dataset, check_file + +try: + import wandb + + assert hasattr(wandb, '__version__') # verify package import not local dir +except (ImportError, AssertionError): + wandb = None + +RANK = int(os.getenv('RANK', -1)) +WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' + + +def remove_prefix(from_string, prefix=WANDB_ARTIFACT_PREFIX): + return from_string[len(prefix):] + + +def check_wandb_config_file(data_config_file): + wandb_config = '_wandb.'.join(data_config_file.rsplit('.', 1)) # updated data.yaml path + if Path(wandb_config).is_file(): + return wandb_config + return data_config_file + + +def check_wandb_dataset(data_file): + is_trainset_wandb_artifact = False + is_valset_wandb_artifact = False + if isinstance(data_file, dict): + # In that case another dataset manager has already processed it and we don't have to + return data_file + if check_file(data_file) and data_file.endswith('.yaml'): + with open(data_file, errors='ignore') as f: + data_dict = yaml.safe_load(f) + is_trainset_wandb_artifact = isinstance(data_dict['train'], + str) and data_dict['train'].startswith(WANDB_ARTIFACT_PREFIX) + is_valset_wandb_artifact = isinstance(data_dict['val'], + str) and data_dict['val'].startswith(WANDB_ARTIFACT_PREFIX) + if is_trainset_wandb_artifact or is_valset_wandb_artifact: + return data_dict + else: + return check_dataset(data_file) + + +def get_run_info(run_path): + run_path = Path(remove_prefix(run_path, WANDB_ARTIFACT_PREFIX)) + run_id = run_path.stem + project = run_path.parent.stem + entity = run_path.parent.parent.stem + model_artifact_name = 'run_' + run_id + '_model' + return entity, project, run_id, model_artifact_name + + +def check_wandb_resume(opt): + process_wandb_config_ddp_mode(opt) if RANK not in [-1, 0] else None + if isinstance(opt.resume, str): + if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): + if RANK not in [-1, 0]: # For resuming DDP runs + entity, project, run_id, model_artifact_name = get_run_info(opt.resume) + api = wandb.Api() + artifact = api.artifact(entity + '/' + project + '/' + model_artifact_name + ':latest') + modeldir = artifact.download() + opt.weights = str(Path(modeldir) / "last.pt") + return True + return None + + +def process_wandb_config_ddp_mode(opt): + with open(check_file(opt.data), errors='ignore') as f: + data_dict = yaml.safe_load(f) # data dict + train_dir, val_dir = None, None + if isinstance(data_dict['train'], str) and data_dict['train'].startswith(WANDB_ARTIFACT_PREFIX): + api = wandb.Api() + train_artifact = api.artifact(remove_prefix(data_dict['train']) + ':' + opt.artifact_alias) + train_dir = train_artifact.download() + train_path = Path(train_dir) / 'data/images/' + data_dict['train'] = str(train_path) + + if isinstance(data_dict['val'], str) and data_dict['val'].startswith(WANDB_ARTIFACT_PREFIX): + api = wandb.Api() + val_artifact = api.artifact(remove_prefix(data_dict['val']) + ':' + opt.artifact_alias) + val_dir = val_artifact.download() + val_path = Path(val_dir) / 'data/images/' + data_dict['val'] = str(val_path) + if train_dir or val_dir: + ddp_data_path = str(Path(val_dir) / 'wandb_local_data.yaml') + with open(ddp_data_path, 'w') as f: + yaml.safe_dump(data_dict, f) + opt.data = ddp_data_path + + +class WandbLogger(): + """Log training runs, datasets, models, and predictions to Weights & Biases. + + This logger sends information to W&B at wandb.ai. By default, this information + includes hyperparameters, system configuration and metrics, model metrics, + and basic data metrics and analyses. + + By providing additional command line arguments to train.py, datasets, + models and predictions can also be logged. + + For more on how this logger is used, see the Weights & Biases documentation: + https://docs.wandb.com/guides/integrations/yolov5 + """ + + def __init__(self, opt, run_id=None, job_type='Training'): + """ + - Initialize WandbLogger instance + - Upload dataset if opt.upload_dataset is True + - Setup training processes if job_type is 'Training' + + arguments: + opt (namespace) -- Commandline arguments for this run + run_id (str) -- Run ID of W&B run to be resumed + job_type (str) -- To set the job_type for this run + + """ + # Pre-training routine -- + self.job_type = job_type + self.wandb, self.wandb_run = wandb, None if not wandb else wandb.run + self.val_artifact, self.train_artifact = None, None + self.train_artifact_path, self.val_artifact_path = None, None + self.result_artifact = None + self.val_table, self.result_table = None, None + self.bbox_media_panel_images = [] + self.val_table_path_map = None + self.max_imgs_to_log = 16 + self.wandb_artifact_data_dict = None + self.data_dict = None + # It's more elegant to stick to 1 wandb.init call, + # but useful config data is overwritten in the WandbLogger's wandb.init call + if isinstance(opt.resume, str): # checks resume from artifact + if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): + entity, project, run_id, model_artifact_name = get_run_info(opt.resume) + model_artifact_name = WANDB_ARTIFACT_PREFIX + model_artifact_name + assert wandb, 'install wandb to resume wandb runs' + # Resume wandb-artifact:// runs here| workaround for not overwriting wandb.config + self.wandb_run = wandb.init(id=run_id, + project=project, + entity=entity, + resume='allow', + allow_val_change=True) + opt.resume = model_artifact_name + elif self.wandb: + self.wandb_run = wandb.init(config=opt, + resume="allow", + project='YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem, + entity=opt.entity, + name=opt.name if opt.name != 'exp' else None, + job_type=job_type, + id=run_id, + allow_val_change=True) if not wandb.run else wandb.run + if self.wandb_run: + if self.job_type == 'Training': + if opt.upload_dataset: + if not opt.resume: + self.wandb_artifact_data_dict = self.check_and_upload_dataset(opt) + + if isinstance(opt.data, dict): + # This means another dataset manager has already processed the dataset info (e.g. ClearML) + # and they will have stored the already processed dict in opt.data + self.data_dict = opt.data + elif opt.resume: + # resume from artifact + if isinstance(opt.resume, str) and opt.resume.startswith(WANDB_ARTIFACT_PREFIX): + self.data_dict = dict(self.wandb_run.config.data_dict) + else: # local resume + self.data_dict = check_wandb_dataset(opt.data) + else: + self.data_dict = check_wandb_dataset(opt.data) + self.wandb_artifact_data_dict = self.wandb_artifact_data_dict or self.data_dict + + # write data_dict to config. useful for resuming from artifacts. Do this only when not resuming. + self.wandb_run.config.update({'data_dict': self.wandb_artifact_data_dict}, allow_val_change=True) + self.setup_training(opt) + + if self.job_type == 'Dataset Creation': + self.wandb_run.config.update({"upload_dataset": True}) + self.data_dict = self.check_and_upload_dataset(opt) + + def check_and_upload_dataset(self, opt): + """ + Check if the dataset format is compatible and upload it as W&B artifact + + arguments: + opt (namespace)-- Commandline arguments for current run + + returns: + Updated dataset info dictionary where local dataset paths are replaced by WAND_ARFACT_PREFIX links. + """ + assert wandb, 'Install wandb to upload dataset' + config_path = self.log_dataset_artifact(opt.data, opt.single_cls, + 'YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem) + with open(config_path, errors='ignore') as f: + wandb_data_dict = yaml.safe_load(f) + return wandb_data_dict + + def setup_training(self, opt): + """ + Setup the necessary processes for training YOLO models: + - Attempt to download model checkpoint and dataset artifacts if opt.resume stats with WANDB_ARTIFACT_PREFIX + - Update data_dict, to contain info of previous run if resumed and the paths of dataset artifact if downloaded + - Setup log_dict, initialize bbox_interval + + arguments: + opt (namespace) -- commandline arguments for this run + + """ + self.log_dict, self.current_epoch = {}, 0 + self.bbox_interval = opt.bbox_interval + if isinstance(opt.resume, str): + modeldir, _ = self.download_model_artifact(opt) + if modeldir: + self.weights = Path(modeldir) / "last.pt" + config = self.wandb_run.config + opt.weights, opt.save_period, opt.batch_size, opt.bbox_interval, opt.epochs, opt.hyp, opt.imgsz = str( + self.weights), config.save_period, config.batch_size, config.bbox_interval, config.epochs,\ + config.hyp, config.imgsz + data_dict = self.data_dict + if self.val_artifact is None: # If --upload_dataset is set, use the existing artifact, don't download + self.train_artifact_path, self.train_artifact = self.download_dataset_artifact( + data_dict.get('train'), opt.artifact_alias) + self.val_artifact_path, self.val_artifact = self.download_dataset_artifact( + data_dict.get('val'), opt.artifact_alias) + + if self.train_artifact_path is not None: + train_path = Path(self.train_artifact_path) / 'data/images/' + data_dict['train'] = str(train_path) + if self.val_artifact_path is not None: + val_path = Path(self.val_artifact_path) / 'data/images/' + data_dict['val'] = str(val_path) + + if self.val_artifact is not None: + self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") + columns = ["epoch", "id", "ground truth", "prediction"] + columns.extend(self.data_dict['names']) + self.result_table = wandb.Table(columns) + self.val_table = self.val_artifact.get("val") + if self.val_table_path_map is None: + self.map_val_table_path() + if opt.bbox_interval == -1: + self.bbox_interval = opt.bbox_interval = (opt.epochs // 10) if opt.epochs > 10 else 1 + if opt.evolve or opt.noplots: + self.bbox_interval = opt.bbox_interval = opt.epochs + 1 # disable bbox_interval + train_from_artifact = self.train_artifact_path is not None and self.val_artifact_path is not None + # Update the the data_dict to point to local artifacts dir + if train_from_artifact: + self.data_dict = data_dict + + def download_dataset_artifact(self, path, alias): + """ + download the model checkpoint artifact if the path starts with WANDB_ARTIFACT_PREFIX + + arguments: + path -- path of the dataset to be used for training + alias (str)-- alias of the artifact to be download/used for training + + returns: + (str, wandb.Artifact) -- path of the downladed dataset and it's corresponding artifact object if dataset + is found otherwise returns (None, None) + """ + if isinstance(path, str) and path.startswith(WANDB_ARTIFACT_PREFIX): + artifact_path = Path(remove_prefix(path, WANDB_ARTIFACT_PREFIX) + ":" + alias) + dataset_artifact = wandb.use_artifact(artifact_path.as_posix().replace("\\", "/")) + assert dataset_artifact is not None, "'Error: W&B dataset artifact doesn\'t exist'" + datadir = dataset_artifact.download() + return datadir, dataset_artifact + return None, None + + def download_model_artifact(self, opt): + """ + download the model checkpoint artifact if the resume path starts with WANDB_ARTIFACT_PREFIX + + arguments: + opt (namespace) -- Commandline arguments for this run + """ + if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): + model_artifact = wandb.use_artifact(remove_prefix(opt.resume, WANDB_ARTIFACT_PREFIX) + ":latest") + assert model_artifact is not None, 'Error: W&B model artifact doesn\'t exist' + modeldir = model_artifact.download() + # epochs_trained = model_artifact.metadata.get('epochs_trained') + total_epochs = model_artifact.metadata.get('total_epochs') + is_finished = total_epochs is None + assert not is_finished, 'training is finished, can only resume incomplete runs.' + return modeldir, model_artifact + return None, None + + def log_model(self, path, opt, epoch, fitness_score, best_model=False): + """ + Log the model checkpoint as W&B artifact + + arguments: + path (Path) -- Path of directory containing the checkpoints + opt (namespace) -- Command line arguments for this run + epoch (int) -- Current epoch number + fitness_score (float) -- fitness score for current epoch + best_model (boolean) -- Boolean representing if the current checkpoint is the best yet. + """ + model_artifact = wandb.Artifact('run_' + wandb.run.id + '_model', + type='model', + metadata={ + 'original_url': str(path), + 'epochs_trained': epoch + 1, + 'save period': opt.save_period, + 'project': opt.project, + 'total_epochs': opt.epochs, + 'fitness_score': fitness_score}) + model_artifact.add_file(str(path / 'last.pt'), name='last.pt') + wandb.log_artifact(model_artifact, + aliases=['latest', 'last', 'epoch ' + str(self.current_epoch), 'best' if best_model else '']) + LOGGER.info(f"Saving model artifact on epoch {epoch + 1}") + + def log_dataset_artifact(self, data_file, single_cls, project, overwrite_config=False): + """ + Log the dataset as W&B artifact and return the new data file with W&B links + + arguments: + data_file (str) -- the .yaml file with information about the dataset like - path, classes etc. + single_class (boolean) -- train multi-class data as single-class + project (str) -- project name. Used to construct the artifact path + overwrite_config (boolean) -- overwrites the data.yaml file if set to true otherwise creates a new + file with _wandb postfix. Eg -> data_wandb.yaml + + returns: + the new .yaml file with artifact links. it can be used to start training directly from artifacts + """ + upload_dataset = self.wandb_run.config.upload_dataset + log_val_only = isinstance(upload_dataset, str) and upload_dataset == 'val' + self.data_dict = check_dataset(data_file) # parse and check + data = dict(self.data_dict) + nc, names = (1, ['item']) if single_cls else (int(data['nc']), data['names']) + names = {k: v for k, v in enumerate(names)} # to index dictionary + + # log train set + if not log_val_only: + self.train_artifact = self.create_dataset_table(LoadImagesAndLabels(data['train'], rect=True, batch_size=1), + names, + name='train') if data.get('train') else None + if data.get('train'): + data['train'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'train') + + self.val_artifact = self.create_dataset_table( + LoadImagesAndLabels(data['val'], rect=True, batch_size=1), names, name='val') if data.get('val') else None + if data.get('val'): + data['val'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'val') + + path = Path(data_file) + # create a _wandb.yaml file with artifacts links if both train and test set are logged + if not log_val_only: + path = (path.stem if overwrite_config else path.stem + '_wandb') + '.yaml' # updated data.yaml path + path = ROOT / 'data' / path + data.pop('download', None) + data.pop('path', None) + with open(path, 'w') as f: + yaml.safe_dump(data, f) + LOGGER.info(f"Created dataset config file {path}") + + if self.job_type == 'Training': # builds correct artifact pipeline graph + if not log_val_only: + self.wandb_run.log_artifact( + self.train_artifact) # calling use_artifact downloads the dataset. NOT NEEDED! + self.wandb_run.use_artifact(self.val_artifact) + self.val_artifact.wait() + self.val_table = self.val_artifact.get('val') + self.map_val_table_path() + else: + self.wandb_run.log_artifact(self.train_artifact) + self.wandb_run.log_artifact(self.val_artifact) + return path + + def map_val_table_path(self): + """ + Map the validation dataset Table like name of file -> it's id in the W&B Table. + Useful for - referencing artifacts for evaluation. + """ + self.val_table_path_map = {} + LOGGER.info("Mapping dataset") + for i, data in enumerate(tqdm(self.val_table.data)): + self.val_table_path_map[data[3]] = data[0] + + def create_dataset_table(self, dataset: LoadImagesAndLabels, class_to_id: Dict[int, str], name: str = 'dataset'): + """ + Create and return W&B artifact containing W&B Table of the dataset. + + arguments: + dataset -- instance of LoadImagesAndLabels class used to iterate over the data to build Table + class_to_id -- hash map that maps class ids to labels + name -- name of the artifact + + returns: + dataset artifact to be logged or used + """ + # TODO: Explore multiprocessing to slpit this loop parallely| This is essential for speeding up the the logging + artifact = wandb.Artifact(name=name, type="dataset") + img_files = tqdm([dataset.path]) if isinstance(dataset.path, str) and Path(dataset.path).is_dir() else None + img_files = tqdm(dataset.im_files) if not img_files else img_files + for img_file in img_files: + if Path(img_file).is_dir(): + artifact.add_dir(img_file, name='data/images') + labels_path = 'labels'.join(dataset.path.rsplit('images', 1)) + artifact.add_dir(labels_path, name='data/labels') + else: + artifact.add_file(img_file, name='data/images/' + Path(img_file).name) + label_file = Path(img2label_paths([img_file])[0]) + artifact.add_file(str(label_file), name='data/labels/' + + label_file.name) if label_file.exists() else None + table = wandb.Table(columns=["id", "train_image", "Classes", "name"]) + class_set = wandb.Classes([{'id': id, 'name': name} for id, name in class_to_id.items()]) + for si, (img, labels, paths, shapes) in enumerate(tqdm(dataset)): + box_data, img_classes = [], {} + for cls, *xywh in labels[:, 1:].tolist(): + cls = int(cls) + box_data.append({ + "position": { + "middle": [xywh[0], xywh[1]], + "width": xywh[2], + "height": xywh[3]}, + "class_id": cls, + "box_caption": "%s" % (class_to_id[cls])}) + img_classes[cls] = class_to_id[cls] + boxes = {"ground_truth": {"box_data": box_data, "class_labels": class_to_id}} # inference-space + table.add_data(si, wandb.Image(paths, classes=class_set, boxes=boxes), list(img_classes.values()), + Path(paths).name) + artifact.add(table, name) + return artifact + + def log_training_progress(self, predn, path, names): + """ + Build evaluation Table. Uses reference from validation dataset table. + + arguments: + predn (list): list of predictions in the native space in the format - [xmin, ymin, xmax, ymax, confidence, class] + path (str): local path of the current evaluation image + names (dict(int, str)): hash map that maps class ids to labels + """ + class_set = wandb.Classes([{'id': id, 'name': name} for id, name in names.items()]) + box_data = [] + avg_conf_per_class = [0] * len(self.data_dict['names']) + pred_class_count = {} + for *xyxy, conf, cls in predn.tolist(): + if conf >= 0.25: + cls = int(cls) + box_data.append({ + "position": { + "minX": xyxy[0], + "minY": xyxy[1], + "maxX": xyxy[2], + "maxY": xyxy[3]}, + "class_id": cls, + "box_caption": f"{names[cls]} {conf:.3f}", + "scores": { + "class_score": conf}, + "domain": "pixel"}) + avg_conf_per_class[cls] += conf + + if cls in pred_class_count: + pred_class_count[cls] += 1 + else: + pred_class_count[cls] = 1 + + for pred_class in pred_class_count.keys(): + avg_conf_per_class[pred_class] = avg_conf_per_class[pred_class] / pred_class_count[pred_class] + + boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space + id = self.val_table_path_map[Path(path).name] + self.result_table.add_data(self.current_epoch, id, self.val_table.data[id][1], + wandb.Image(self.val_table.data[id][1], boxes=boxes, classes=class_set), + *avg_conf_per_class) + + def val_one_image(self, pred, predn, path, names, im): + """ + Log validation data for one image. updates the result Table if validation dataset is uploaded and log bbox media panel + + arguments: + pred (list): list of scaled predictions in the format - [xmin, ymin, xmax, ymax, confidence, class] + predn (list): list of predictions in the native space - [xmin, ymin, xmax, ymax, confidence, class] + path (str): local path of the current evaluation image + """ + if self.val_table and self.result_table: # Log Table if Val dataset is uploaded as artifact + self.log_training_progress(predn, path, names) + + if len(self.bbox_media_panel_images) < self.max_imgs_to_log and self.current_epoch > 0: + if self.current_epoch % self.bbox_interval == 0: + box_data = [{ + "position": { + "minX": xyxy[0], + "minY": xyxy[1], + "maxX": xyxy[2], + "maxY": xyxy[3]}, + "class_id": int(cls), + "box_caption": f"{names[int(cls)]} {conf:.3f}", + "scores": { + "class_score": conf}, + "domain": "pixel"} for *xyxy, conf, cls in pred.tolist()] + boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space + self.bbox_media_panel_images.append(wandb.Image(im, boxes=boxes, caption=path.name)) + + def log(self, log_dict): + """ + save the metrics to the logging dictionary + + arguments: + log_dict (Dict) -- metrics/media to be logged in current step + """ + if self.wandb_run: + for key, value in log_dict.items(): + self.log_dict[key] = value + + def end_epoch(self, best_result=False): + """ + commit the log_dict, model artifacts and Tables to W&B and flush the log_dict. + + arguments: + best_result (boolean): Boolean representing if the result of this evaluation is best or not + """ + if self.wandb_run: + with all_logging_disabled(): + if self.bbox_media_panel_images: + self.log_dict["BoundingBoxDebugger"] = self.bbox_media_panel_images + try: + wandb.log(self.log_dict) + except BaseException as e: + LOGGER.info( + f"An error occurred in wandb logger. The training will proceed without interruption. More info\n{e}" + ) + self.wandb_run.finish() + self.wandb_run = None + + self.log_dict = {} + self.bbox_media_panel_images = [] + if self.result_artifact: + self.result_artifact.add(self.result_table, 'result') + wandb.log_artifact(self.result_artifact, + aliases=[ + 'latest', 'last', 'epoch ' + str(self.current_epoch), + ('best' if best_result else '')]) + + wandb.log({"evaluation": self.result_table}) + columns = ["epoch", "id", "ground truth", "prediction"] + columns.extend(self.data_dict['names']) + self.result_table = wandb.Table(columns) + self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") + + def finish_run(self): + """ + Log metrics if any and finish the current W&B run + """ + if self.wandb_run: + if self.log_dict: + with all_logging_disabled(): + wandb.log(self.log_dict) + wandb.run.finish() + + +@contextmanager +def all_logging_disabled(highest_level=logging.CRITICAL): + """ source - https://gist.github.com/simon-weber/7853144 + A context manager that will prevent any logging messages triggered during the body from being processed. + :param highest_level: the maximum logging level in use. + This would only need to be changed if a custom level greater than CRITICAL is defined. + """ + previous_level = logging.root.manager.disable + logging.disable(highest_level) + try: + yield + finally: + logging.disable(previous_level) diff --git a/application/yolov5_example/utils/loss.py b/application/yolov5_example/utils/loss.py new file mode 100644 index 00000000..9b9c3d9f --- /dev/null +++ b/application/yolov5_example/utils/loss.py @@ -0,0 +1,234 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Loss functions +""" + +import torch +import torch.nn as nn + +from utils.metrics import bbox_iou +from utils.torch_utils import de_parallel + + +def smooth_BCE(eps=0.1): # https://github.com/ultralytics/yolov3/issues/238#issuecomment-598028441 + # return positive, negative label smoothing BCE targets + return 1.0 - 0.5 * eps, 0.5 * eps + + +class BCEBlurWithLogitsLoss(nn.Module): + # BCEwithLogitLoss() with reduced missing label effects. + def __init__(self, alpha=0.05): + super().__init__() + self.loss_fcn = nn.BCEWithLogitsLoss(reduction='none') # must be nn.BCEWithLogitsLoss() + self.alpha = alpha + + def forward(self, pred, true): + loss = self.loss_fcn(pred, true) + pred = torch.sigmoid(pred) # prob from logits + dx = pred - true # reduce only missing label effects + # dx = (pred - true).abs() # reduce missing label and false label effects + alpha_factor = 1 - torch.exp((dx - 1) / (self.alpha + 1e-4)) + loss *= alpha_factor + return loss.mean() + + +class FocalLoss(nn.Module): + # Wraps focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) + def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): + super().__init__() + self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() + self.gamma = gamma + self.alpha = alpha + self.reduction = loss_fcn.reduction + self.loss_fcn.reduction = 'none' # required to apply FL to each element + + def forward(self, pred, true): + loss = self.loss_fcn(pred, true) + # p_t = torch.exp(-loss) + # loss *= self.alpha * (1.000001 - p_t) ** self.gamma # non-zero power for gradient stability + + # TF implementation https://github.com/tensorflow/addons/blob/v0.7.1/tensorflow_addons/losses/focal_loss.py + pred_prob = torch.sigmoid(pred) # prob from logits + p_t = true * pred_prob + (1 - true) * (1 - pred_prob) + alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) + modulating_factor = (1.0 - p_t) ** self.gamma + loss *= alpha_factor * modulating_factor + + if self.reduction == 'mean': + return loss.mean() + elif self.reduction == 'sum': + return loss.sum() + else: # 'none' + return loss + + +class QFocalLoss(nn.Module): + # Wraps Quality focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) + def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): + super().__init__() + self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() + self.gamma = gamma + self.alpha = alpha + self.reduction = loss_fcn.reduction + self.loss_fcn.reduction = 'none' # required to apply FL to each element + + def forward(self, pred, true): + loss = self.loss_fcn(pred, true) + + pred_prob = torch.sigmoid(pred) # prob from logits + alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) + modulating_factor = torch.abs(true - pred_prob) ** self.gamma + loss *= alpha_factor * modulating_factor + + if self.reduction == 'mean': + return loss.mean() + elif self.reduction == 'sum': + return loss.sum() + else: # 'none' + return loss + + +class ComputeLoss: + sort_obj_iou = False + + # Compute losses + def __init__(self, model, autobalance=False): + device = next(model.parameters()).device # get model device + h = model.hyp # hyperparameters + + # Define criteria + BCEcls = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['cls_pw']], device=device)) + BCEobj = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['obj_pw']], device=device)) + + # Class label smoothing https://arxiv.org/pdf/1902.04103.pdf eqn 3 + self.cp, self.cn = smooth_BCE(eps=h.get('label_smoothing', 0.0)) # positive, negative BCE targets + + # Focal loss + g = h['fl_gamma'] # focal loss gamma + if g > 0: + BCEcls, BCEobj = FocalLoss(BCEcls, g), FocalLoss(BCEobj, g) + + m = de_parallel(model).model[-1] # Detect() module + self.balance = {3: [4.0, 1.0, 0.4]}.get(m.nl, [4.0, 1.0, 0.25, 0.06, 0.02]) # P3-P7 + self.ssi = list(m.stride).index(16) if autobalance else 0 # stride 16 index + self.BCEcls, self.BCEobj, self.gr, self.hyp, self.autobalance = BCEcls, BCEobj, 1.0, h, autobalance + self.na = m.na # number of anchors + self.nc = m.nc # number of classes + self.nl = m.nl # number of layers + self.anchors = m.anchors + self.device = device + + def __call__(self, p, targets): # predictions, targets + lcls = torch.zeros(1, device=self.device) # class loss + lbox = torch.zeros(1, device=self.device) # box loss + lobj = torch.zeros(1, device=self.device) # object loss + tcls, tbox, indices, anchors = self.build_targets(p, targets) # targets + + # Losses + for i, pi in enumerate(p): # layer index, layer predictions + b, a, gj, gi = indices[i] # image, anchor, gridy, gridx + tobj = torch.zeros(pi.shape[:4], dtype=pi.dtype, device=self.device) # target obj + + n = b.shape[0] # number of targets + if n: + # pxy, pwh, _, pcls = pi[b, a, gj, gi].tensor_split((2, 4, 5), dim=1) # faster, requires torch 1.8.0 + pxy, pwh, _, pcls = pi[b, a, gj, gi].split((2, 2, 1, self.nc), 1) # target-subset of predictions + + # Regression + pxy = pxy.sigmoid() * 2 - 0.5 + pwh = (pwh.sigmoid() * 2) ** 2 * anchors[i] + pbox = torch.cat((pxy, pwh), 1) # predicted box + iou = bbox_iou(pbox, tbox[i], CIoU=True).squeeze() # iou(prediction, target) + lbox += (1.0 - iou).mean() # iou loss + + # Objectness + iou = iou.detach().clamp(0).type(tobj.dtype) + if self.sort_obj_iou: + j = iou.argsort() + b, a, gj, gi, iou = b[j], a[j], gj[j], gi[j], iou[j] + if self.gr < 1: + iou = (1.0 - self.gr) + self.gr * iou + tobj[b, a, gj, gi] = iou # iou ratio + + # Classification + if self.nc > 1: # cls loss (only if multiple classes) + t = torch.full_like(pcls, self.cn, device=self.device) # targets + t[range(n), tcls[i]] = self.cp + lcls += self.BCEcls(pcls, t) # BCE + + # Append targets to text file + # with open('targets.txt', 'a') as file: + # [file.write('%11.5g ' * 4 % tuple(x) + '\n') for x in torch.cat((txy[i], twh[i]), 1)] + + obji = self.BCEobj(pi[..., 4], tobj) + lobj += obji * self.balance[i] # obj loss + if self.autobalance: + self.balance[i] = self.balance[i] * 0.9999 + 0.0001 / obji.detach().item() + + if self.autobalance: + self.balance = [x / self.balance[self.ssi] for x in self.balance] + lbox *= self.hyp['box'] + lobj *= self.hyp['obj'] + lcls *= self.hyp['cls'] + bs = tobj.shape[0] # batch size + + return (lbox + lobj + lcls) * bs, torch.cat((lbox, lobj, lcls)).detach() + + def build_targets(self, p, targets): + # Build targets for compute_loss(), input targets(image,class,x,y,w,h) + na, nt = self.na, targets.shape[0] # number of anchors, targets + tcls, tbox, indices, anch = [], [], [], [] + gain = torch.ones(7, device=self.device) # normalized to gridspace gain + ai = torch.arange(na, device=self.device).float().view(na, 1).repeat(1, nt) # same as .repeat_interleave(nt) + targets = torch.cat((targets.repeat(na, 1, 1), ai[..., None]), 2) # append anchor indices + + g = 0.5 # bias + off = torch.tensor( + [ + [0, 0], + [1, 0], + [0, 1], + [-1, 0], + [0, -1], # j,k,l,m + # [1, 1], [1, -1], [-1, 1], [-1, -1], # jk,jm,lk,lm + ], + device=self.device).float() * g # offsets + + for i in range(self.nl): + anchors, shape = self.anchors[i], p[i].shape + gain[2:6] = torch.tensor(shape)[[3, 2, 3, 2]] # xyxy gain + + # Match targets to anchors + t = targets * gain # shape(3,n,7) + if nt: + # Matches + r = t[..., 4:6] / anchors[:, None] # wh ratio + j = torch.max(r, 1 / r).max(2)[0] < self.hyp['anchor_t'] # compare + # j = wh_iou(anchors, t[:, 4:6]) > model.hyp['iou_t'] # iou(3,n)=wh_iou(anchors(3,2), gwh(n,2)) + t = t[j] # filter + + # Offsets + gxy = t[:, 2:4] # grid xy + gxi = gain[[2, 3]] - gxy # inverse + j, k = ((gxy % 1 < g) & (gxy > 1)).T + l, m = ((gxi % 1 < g) & (gxi > 1)).T + j = torch.stack((torch.ones_like(j), j, k, l, m)) + t = t.repeat((5, 1, 1))[j] + offsets = (torch.zeros_like(gxy)[None] + off[:, None])[j] + else: + t = targets[0] + offsets = 0 + + # Define + bc, gxy, gwh, a = t.chunk(4, 1) # (image, class), grid xy, grid wh, anchors + a, (b, c) = a.long().view(-1), bc.long().T # anchors, image, class + gij = (gxy - offsets).long() + gi, gj = gij.T # grid indices + + # Append + indices.append((b, a, gj.clamp_(0, shape[2] - 1), gi.clamp_(0, shape[3] - 1))) # image, anchor, grid + tbox.append(torch.cat((gxy - gij, gwh), 1)) # box + anch.append(anchors[a]) # anchors + tcls.append(c) # class + + return tcls, tbox, indices, anch diff --git a/application/yolov5_example/utils/metrics.py b/application/yolov5_example/utils/metrics.py new file mode 100644 index 00000000..08880cd3 --- /dev/null +++ b/application/yolov5_example/utils/metrics.py @@ -0,0 +1,364 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Model validation metrics +""" + +import math +import warnings +from pathlib import Path + +import matplotlib.pyplot as plt +import numpy as np +import torch + + +def fitness(x): + # Model fitness as a weighted combination of metrics + w = [0.0, 0.0, 0.1, 0.9] # weights for [P, R, mAP@0.5, mAP@0.5:0.95] + return (x[:, :4] * w).sum(1) + + +def smooth(y, f=0.05): + # Box filter of fraction f + nf = round(len(y) * f * 2) // 2 + 1 # number of filter elements (must be odd) + p = np.ones(nf // 2) # ones padding + yp = np.concatenate((p * y[0], y, p * y[-1]), 0) # y padded + return np.convolve(yp, np.ones(nf) / nf, mode='valid') # y-smoothed + + +def ap_per_class(tp, conf, pred_cls, target_cls, plot=False, save_dir='.', names=(), eps=1e-16): + """ Compute the average precision, given the recall and precision curves. + Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. + # Arguments + tp: True positives (nparray, nx1 or nx10). + conf: Objectness value from 0-1 (nparray). + pred_cls: Predicted object classes (nparray). + target_cls: True object classes (nparray). + plot: Plot precision-recall curve at mAP@0.5 + save_dir: Plot save directory + # Returns + The average precision as computed in py-faster-rcnn. + """ + + # Sort by objectness + i = np.argsort(-conf) + tp, conf, pred_cls = tp[i], conf[i], pred_cls[i] + + # Find unique classes + unique_classes, nt = np.unique(target_cls, return_counts=True) + nc = unique_classes.shape[0] # number of classes, number of detections + + # Create Precision-Recall curve and compute AP for each class + px, py = np.linspace(0, 1, 1000), [] # for plotting + ap, p, r = np.zeros((nc, tp.shape[1])), np.zeros((nc, 1000)), np.zeros((nc, 1000)) + for ci, c in enumerate(unique_classes): + i = pred_cls == c + n_l = nt[ci] # number of labels + n_p = i.sum() # number of predictions + if n_p == 0 or n_l == 0: + continue + + # Accumulate FPs and TPs + fpc = (1 - tp[i]).cumsum(0) + tpc = tp[i].cumsum(0) + + # Recall + recall = tpc / (n_l + eps) # recall curve + r[ci] = np.interp(-px, -conf[i], recall[:, 0], left=0) # negative x, xp because xp decreases + + # Precision + precision = tpc / (tpc + fpc) # precision curve + p[ci] = np.interp(-px, -conf[i], precision[:, 0], left=1) # p at pr_score + + # AP from recall-precision curve + for j in range(tp.shape[1]): + ap[ci, j], mpre, mrec = compute_ap(recall[:, j], precision[:, j]) + if plot and j == 0: + py.append(np.interp(px, mrec, mpre)) # precision at mAP@0.5 + + # Compute F1 (harmonic mean of precision and recall) + f1 = 2 * p * r / (p + r + eps) + names = [v for k, v in names.items() if k in unique_classes] # list: only classes that have data + names = dict(enumerate(names)) # to dict + if plot: + plot_pr_curve(px, py, ap, Path(save_dir) / 'PR_curve.png', names) + plot_mc_curve(px, f1, Path(save_dir) / 'F1_curve.png', names, ylabel='F1') + plot_mc_curve(px, p, Path(save_dir) / 'P_curve.png', names, ylabel='Precision') + plot_mc_curve(px, r, Path(save_dir) / 'R_curve.png', names, ylabel='Recall') + + i = smooth(f1.mean(0), 0.1).argmax() # max F1 index + p, r, f1 = p[:, i], r[:, i], f1[:, i] + tp = (r * nt).round() # true positives + fp = (tp / (p + eps) - tp).round() # false positives + return tp, fp, p, r, f1, ap, unique_classes.astype(int) + + +def compute_ap(recall, precision): + """ Compute the average precision, given the recall and precision curves + # Arguments + recall: The recall curve (list) + precision: The precision curve (list) + # Returns + Average precision, precision curve, recall curve + """ + + # Append sentinel values to beginning and end + mrec = np.concatenate(([0.0], recall, [1.0])) + mpre = np.concatenate(([1.0], precision, [0.0])) + + # Compute the precision envelope + mpre = np.flip(np.maximum.accumulate(np.flip(mpre))) + + # Integrate area under curve + method = 'interp' # methods: 'continuous', 'interp' + if method == 'interp': + x = np.linspace(0, 1, 101) # 101-point interp (COCO) + ap = np.trapz(np.interp(x, mrec, mpre), x) # integrate + else: # 'continuous' + i = np.where(mrec[1:] != mrec[:-1])[0] # points where x axis (recall) changes + ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1]) # area under curve + + return ap, mpre, mrec + + +class ConfusionMatrix: + # Updated version of https://github.com/kaanakan/object_detection_confusion_matrix + def __init__(self, nc, conf=0.25, iou_thres=0.45): + self.matrix = np.zeros((nc + 1, nc + 1)) + self.nc = nc # number of classes + self.conf = conf + self.iou_thres = iou_thres + + def process_batch(self, detections, labels): + """ + Return intersection-over-union (Jaccard index) of boxes. + Both sets of boxes are expected to be in (x1, y1, x2, y2) format. + Arguments: + detections (Array[N, 6]), x1, y1, x2, y2, conf, class + labels (Array[M, 5]), class, x1, y1, x2, y2 + Returns: + None, updates confusion matrix accordingly + """ + if detections is None: + gt_classes = labels.int() + for i, gc in enumerate(gt_classes): + self.matrix[self.nc, gc] += 1 # background FN + return + + detections = detections[detections[:, 4] > self.conf] + gt_classes = labels[:, 0].int() + detection_classes = detections[:, 5].int() + iou = box_iou(labels[:, 1:], detections[:, :4]) + + x = torch.where(iou > self.iou_thres) + if x[0].shape[0]: + matches = torch.cat((torch.stack(x, 1), iou[x[0], x[1]][:, None]), 1).cpu().numpy() + if x[0].shape[0] > 1: + matches = matches[matches[:, 2].argsort()[::-1]] + matches = matches[np.unique(matches[:, 1], return_index=True)[1]] + matches = matches[matches[:, 2].argsort()[::-1]] + matches = matches[np.unique(matches[:, 0], return_index=True)[1]] + else: + matches = np.zeros((0, 3)) + + n = matches.shape[0] > 0 + m0, m1, _ = matches.transpose().astype(int) + for i, gc in enumerate(gt_classes): + j = m0 == i + if n and sum(j) == 1: + self.matrix[detection_classes[m1[j]], gc] += 1 # correct + else: + self.matrix[self.nc, gc] += 1 # background FP + + if n: + for i, dc in enumerate(detection_classes): + if not any(m1 == i): + self.matrix[dc, self.nc] += 1 # background FN + + def matrix(self): + return self.matrix + + def tp_fp(self): + tp = self.matrix.diagonal() # true positives + fp = self.matrix.sum(1) - tp # false positives + # fn = self.matrix.sum(0) - tp # false negatives (missed detections) + return tp[:-1], fp[:-1] # remove background class + + def plot(self, normalize=True, save_dir='', names=()): + try: + import seaborn as sn + + array = self.matrix / ((self.matrix.sum(0).reshape(1, -1) + 1E-9) if normalize else 1) # normalize columns + array[array < 0.005] = np.nan # don't annotate (would appear as 0.00) + + fig = plt.figure(figsize=(12, 9), tight_layout=True) + nc, nn = self.nc, len(names) # number of classes, names + sn.set(font_scale=1.0 if nc < 50 else 0.8) # for label size + labels = (0 < nn < 99) and (nn == nc) # apply names to ticklabels + with warnings.catch_warnings(): + warnings.simplefilter('ignore') # suppress empty matrix RuntimeWarning: All-NaN slice encountered + sn.heatmap(array, + annot=nc < 30, + annot_kws={ + "size": 8}, + cmap='Blues', + fmt='.2f', + square=True, + vmin=0.0, + xticklabels=names + ['background FP'] if labels else "auto", + yticklabels=names + ['background FN'] if labels else "auto").set_facecolor((1, 1, 1)) + fig.axes[0].set_xlabel('True') + fig.axes[0].set_ylabel('Predicted') + plt.title('Confusion Matrix') + fig.savefig(Path(save_dir) / 'confusion_matrix.png', dpi=250) + plt.close() + except Exception as e: + print(f'WARNING: ConfusionMatrix plot failure: {e}') + + def print(self): + for i in range(self.nc + 1): + print(' '.join(map(str, self.matrix[i]))) + + +def bbox_iou(box1, box2, xywh=True, GIoU=False, DIoU=False, CIoU=False, eps=1e-7): + # Returns Intersection over Union (IoU) of box1(1,4) to box2(n,4) + + # Get the coordinates of bounding boxes + if xywh: # transform from xywh to xyxy + (x1, y1, w1, h1), (x2, y2, w2, h2) = box1.chunk(4, 1), box2.chunk(4, 1) + w1_, h1_, w2_, h2_ = w1 / 2, h1 / 2, w2 / 2, h2 / 2 + b1_x1, b1_x2, b1_y1, b1_y2 = x1 - w1_, x1 + w1_, y1 - h1_, y1 + h1_ + b2_x1, b2_x2, b2_y1, b2_y2 = x2 - w2_, x2 + w2_, y2 - h2_, y2 + h2_ + else: # x1, y1, x2, y2 = box1 + b1_x1, b1_y1, b1_x2, b1_y2 = box1.chunk(4, 1) + b2_x1, b2_y1, b2_x2, b2_y2 = box2.chunk(4, 1) + w1, h1 = b1_x2 - b1_x1, b1_y2 - b1_y1 + w2, h2 = b2_x2 - b2_x1, b2_y2 - b2_y1 + + # Intersection area + inter = (torch.min(b1_x2, b2_x2) - torch.max(b1_x1, b2_x1)).clamp(0) * \ + (torch.min(b1_y2, b2_y2) - torch.max(b1_y1, b2_y1)).clamp(0) + + # Union Area + union = w1 * h1 + w2 * h2 - inter + eps + + # IoU + iou = inter / union + if CIoU or DIoU or GIoU: + cw = torch.max(b1_x2, b2_x2) - torch.min(b1_x1, b2_x1) # convex (smallest enclosing box) width + ch = torch.max(b1_y2, b2_y2) - torch.min(b1_y1, b2_y1) # convex height + if CIoU or DIoU: # Distance or Complete IoU https://arxiv.org/abs/1911.08287v1 + c2 = cw ** 2 + ch ** 2 + eps # convex diagonal squared + rho2 = ((b2_x1 + b2_x2 - b1_x1 - b1_x2) ** 2 + (b2_y1 + b2_y2 - b1_y1 - b1_y2) ** 2) / 4 # center dist ** 2 + if CIoU: # https://github.com/Zzh-tju/DIoU-SSD-pytorch/blob/master/utils/box/box_utils.py#L47 + v = (4 / math.pi ** 2) * torch.pow(torch.atan(w2 / (h2 + eps)) - torch.atan(w1 / (h1 + eps)), 2) + with torch.no_grad(): + alpha = v / (v - iou + (1 + eps)) + return iou - (rho2 / c2 + v * alpha) # CIoU + return iou - rho2 / c2 # DIoU + c_area = cw * ch + eps # convex area + return iou - (c_area - union) / c_area # GIoU https://arxiv.org/pdf/1902.09630.pdf + return iou # IoU + + +def box_area(box): + # box = xyxy(4,n) + return (box[2] - box[0]) * (box[3] - box[1]) + + +def box_iou(box1, box2, eps=1e-7): + # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py + """ + Return intersection-over-union (Jaccard index) of boxes. + Both sets of boxes are expected to be in (x1, y1, x2, y2) format. + Arguments: + box1 (Tensor[N, 4]) + box2 (Tensor[M, 4]) + Returns: + iou (Tensor[N, M]): the NxM matrix containing the pairwise + IoU values for every element in boxes1 and boxes2 + """ + + # inter(N,M) = (rb(N,M,2) - lt(N,M,2)).clamp(0).prod(2) + (a1, a2), (b1, b2) = box1[:, None].chunk(2, 2), box2.chunk(2, 1) + inter = (torch.min(a2, b2) - torch.max(a1, b1)).clamp(0).prod(2) + + # IoU = inter / (area1 + area2 - inter) + return inter / (box_area(box1.T)[:, None] + box_area(box2.T) - inter + eps) + + +def bbox_ioa(box1, box2, eps=1e-7): + """ Returns the intersection over box2 area given box1, box2. Boxes are x1y1x2y2 + box1: np.array of shape(4) + box2: np.array of shape(nx4) + returns: np.array of shape(n) + """ + + # Get the coordinates of bounding boxes + b1_x1, b1_y1, b1_x2, b1_y2 = box1 + b2_x1, b2_y1, b2_x2, b2_y2 = box2.T + + # Intersection area + inter_area = (np.minimum(b1_x2, b2_x2) - np.maximum(b1_x1, b2_x1)).clip(0) * \ + (np.minimum(b1_y2, b2_y2) - np.maximum(b1_y1, b2_y1)).clip(0) + + # box2 area + box2_area = (b2_x2 - b2_x1) * (b2_y2 - b2_y1) + eps + + # Intersection over box2 area + return inter_area / box2_area + + +def wh_iou(wh1, wh2, eps=1e-7): + # Returns the nxm IoU matrix. wh1 is nx2, wh2 is mx2 + wh1 = wh1[:, None] # [N,1,2] + wh2 = wh2[None] # [1,M,2] + inter = torch.min(wh1, wh2).prod(2) # [N,M] + return inter / (wh1.prod(2) + wh2.prod(2) - inter + eps) # iou = inter / (area1 + area2 - inter) + + +# Plots ---------------------------------------------------------------------------------------------------------------- + + +def plot_pr_curve(px, py, ap, save_dir=Path('pr_curve.png'), names=()): + # Precision-recall curve + fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) + py = np.stack(py, axis=1) + + if 0 < len(names) < 21: # display per-class legend if < 21 classes + for i, y in enumerate(py.T): + ax.plot(px, y, linewidth=1, label=f'{names[i]} {ap[i, 0]:.3f}') # plot(recall, precision) + else: + ax.plot(px, py, linewidth=1, color='grey') # plot(recall, precision) + + ax.plot(px, py.mean(1), linewidth=3, color='blue', label='all classes %.3f mAP@0.5' % ap[:, 0].mean()) + ax.set_xlabel('Recall') + ax.set_ylabel('Precision') + ax.set_xlim(0, 1) + ax.set_ylim(0, 1) + plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") + plt.title('Precision-Recall Curve') + fig.savefig(save_dir, dpi=250) + plt.close() + + +def plot_mc_curve(px, py, save_dir=Path('mc_curve.png'), names=(), xlabel='Confidence', ylabel='Metric'): + # Metric-confidence curve + fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) + + if 0 < len(names) < 21: # display per-class legend if < 21 classes + for i, y in enumerate(py): + ax.plot(px, y, linewidth=1, label=f'{names[i]}') # plot(confidence, metric) + else: + ax.plot(px, py.T, linewidth=1, color='grey') # plot(confidence, metric) + + y = smooth(py.mean(0), 0.05) + ax.plot(px, y, linewidth=3, color='blue', label=f'all classes {y.max():.2f} at {px[y.argmax()]:.3f}') + ax.set_xlabel(xlabel) + ax.set_ylabel(ylabel) + ax.set_xlim(0, 1) + ax.set_ylim(0, 1) + plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") + plt.title(f'{ylabel}-Confidence Curve') + fig.savefig(save_dir, dpi=250) + plt.close() diff --git a/application/yolov5_example/utils/plots.py b/application/yolov5_example/utils/plots.py new file mode 100644 index 00000000..7417308c --- /dev/null +++ b/application/yolov5_example/utils/plots.py @@ -0,0 +1,519 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Plotting utils +""" + +import math +import os +from copy import copy +from pathlib import Path +from urllib.error import URLError + +import cv2 +import matplotlib +import matplotlib.pyplot as plt +import numpy as np +import pandas as pd +import seaborn as sn +import torch +from PIL import Image, ImageDraw, ImageFont + +from utils.general import (CONFIG_DIR, FONT, LOGGER, Timeout, check_font, check_requirements, clip_coords, + increment_path, is_ascii, threaded, try_except, xywh2xyxy, xyxy2xywh) +from utils.metrics import fitness + +# Settings +RANK = int(os.getenv('RANK', -1)) +matplotlib.rc('font', **{'size': 11}) +matplotlib.use('Agg') # for writing to files only + + +class Colors: + # Ultralytics color palette https://ultralytics.com/ + def __init__(self): + # hex = matplotlib.colors.TABLEAU_COLORS.values() + hexs = ('FF3838', 'FF9D97', 'FF701F', 'FFB21D', 'CFD231', '48F90A', '92CC17', '3DDB86', '1A9334', '00D4BB', + '2C99A8', '00C2FF', '344593', '6473FF', '0018EC', '8438FF', '520085', 'CB38FF', 'FF95C8', 'FF37C7') + self.palette = [self.hex2rgb(f'#{c}') for c in hexs] + self.n = len(self.palette) + + def __call__(self, i, bgr=False): + c = self.palette[int(i) % self.n] + return (c[2], c[1], c[0]) if bgr else c + + @staticmethod + def hex2rgb(h): # rgb order (PIL) + return tuple(int(h[1 + i:1 + i + 2], 16) for i in (0, 2, 4)) + + +colors = Colors() # create instance for 'from utils.plots import colors' + + +def check_pil_font(font=FONT, size=10): + # Return a PIL TrueType Font, downloading to CONFIG_DIR if necessary + font = Path(font) + font = font if font.exists() else (CONFIG_DIR / font.name) + try: + return ImageFont.truetype(str(font) if font.exists() else font.name, size) + except Exception: # download if missing + try: + check_font(font) + return ImageFont.truetype(str(font), size) + except TypeError: + check_requirements('Pillow>=8.4.0') # known issue https://github.com/ultralytics/yolov5/issues/5374 + except URLError: # not online + return ImageFont.load_default() + + +class Annotator: + # YOLOv5 Annotator for train/val mosaics and jpgs and detect/hub inference annotations + def __init__(self, im, line_width=None, font_size=None, font='Arial.ttf', pil=False, example='abc'): + assert im.data.contiguous, 'Image not contiguous. Apply np.ascontiguousarray(im) to Annotator() input images.' + non_ascii = not is_ascii(example) # non-latin labels, i.e. asian, arabic, cyrillic + self.pil = pil or non_ascii + if self.pil: # use PIL + self.im = im if isinstance(im, Image.Image) else Image.fromarray(im) + self.draw = ImageDraw.Draw(self.im) + self.font = check_pil_font(font='Arial.Unicode.ttf' if non_ascii else font, + size=font_size or max(round(sum(self.im.size) / 2 * 0.035), 12)) + else: # use cv2 + self.im = im + self.lw = line_width or max(round(sum(im.shape) / 2 * 0.003), 2) # line width + + def box_label(self, box, label='', color=(128, 128, 128), txt_color=(255, 255, 255)): + # Add one xyxy box to image with label + if self.pil or not is_ascii(label): + self.draw.rectangle(box, width=self.lw, outline=color) # box + if label: + w, h = self.font.getsize(label) # text width, height + outside = box[1] - h >= 0 # label fits outside box + self.draw.rectangle( + (box[0], box[1] - h if outside else box[1], box[0] + w + 1, + box[1] + 1 if outside else box[1] + h + 1), + fill=color, + ) + # self.draw.text((box[0], box[1]), label, fill=txt_color, font=self.font, anchor='ls') # for PIL>8.0 + self.draw.text((box[0], box[1] - h if outside else box[1]), label, fill=txt_color, font=self.font) + else: # cv2 + p1, p2 = (int(box[0]), int(box[1])), (int(box[2]), int(box[3])) + cv2.rectangle(self.im, p1, p2, color, thickness=self.lw, lineType=cv2.LINE_AA) + if label: + tf = max(self.lw - 1, 1) # font thickness + w, h = cv2.getTextSize(label, 0, fontScale=self.lw / 3, thickness=tf)[0] # text width, height + outside = p1[1] - h >= 3 + p2 = p1[0] + w, p1[1] - h - 3 if outside else p1[1] + h + 3 + cv2.rectangle(self.im, p1, p2, color, -1, cv2.LINE_AA) # filled + cv2.putText(self.im, + label, (p1[0], p1[1] - 2 if outside else p1[1] + h + 2), + 0, + self.lw / 3, + txt_color, + thickness=tf, + lineType=cv2.LINE_AA) + + def rectangle(self, xy, fill=None, outline=None, width=1): + # Add rectangle to image (PIL-only) + self.draw.rectangle(xy, fill, outline, width) + + def text(self, xy, text, txt_color=(255, 255, 255)): + # Add text to image (PIL-only) + w, h = self.font.getsize(text) # text width, height + self.draw.text((xy[0], xy[1] - h + 1), text, fill=txt_color, font=self.font) + + def result(self): + # Return annotated image as array + return np.asarray(self.im) + + +def feature_visualization(x, module_type, stage, n=32, save_dir=Path('runs/detect/exp')): + """ + x: Features to be visualized + module_type: Module type + stage: Module stage within model + n: Maximum number of feature maps to plot + save_dir: Directory to save results + """ + if 'Detect' not in module_type: + batch, channels, height, width = x.shape # batch, channels, height, width + if height > 1 and width > 1: + f = save_dir / f"stage{stage}_{module_type.split('.')[-1]}_features.png" # filename + + blocks = torch.chunk(x[0].cpu(), channels, dim=0) # select batch index 0, block by channels + n = min(n, channels) # number of plots + fig, ax = plt.subplots(math.ceil(n / 8), 8, tight_layout=True) # 8 rows x n/8 cols + ax = ax.ravel() + plt.subplots_adjust(wspace=0.05, hspace=0.05) + for i in range(n): + ax[i].imshow(blocks[i].squeeze()) # cmap='gray' + ax[i].axis('off') + + LOGGER.info(f'Saving {f}... ({n}/{channels})') + plt.title('Features') + plt.savefig(f, dpi=300, bbox_inches='tight') + plt.close() + np.save(str(f.with_suffix('.npy')), x[0].cpu().numpy()) # npy save + + +def hist2d(x, y, n=100): + # 2d histogram used in labels.png and evolve.png + xedges, yedges = np.linspace(x.min(), x.max(), n), np.linspace(y.min(), y.max(), n) + hist, xedges, yedges = np.histogram2d(x, y, (xedges, yedges)) + xidx = np.clip(np.digitize(x, xedges) - 1, 0, hist.shape[0] - 1) + yidx = np.clip(np.digitize(y, yedges) - 1, 0, hist.shape[1] - 1) + return np.log(hist[xidx, yidx]) + + +def butter_lowpass_filtfilt(data, cutoff=1500, fs=50000, order=5): + from scipy.signal import butter, filtfilt + + # https://stackoverflow.com/questions/28536191/how-to-filter-smooth-with-scipy-numpy + def butter_lowpass(cutoff, fs, order): + nyq = 0.5 * fs + normal_cutoff = cutoff / nyq + return butter(order, normal_cutoff, btype='low', analog=False) + + b, a = butter_lowpass(cutoff, fs, order=order) + return filtfilt(b, a, data) # forward-backward filter + + +def output_to_target(output): + # Convert model output to target format [batch_id, class_id, x, y, w, h, conf] + targets = [] + for i, o in enumerate(output): + for *box, conf, cls in o.cpu().numpy(): + targets.append([i, cls, *list(*xyxy2xywh(np.array(box)[None])), conf]) + return np.array(targets) + + +@threaded +def plot_images(images, targets, paths=None, fname='images.jpg', names=None, max_size=1920, max_subplots=16): + # Plot image grid with labels + if isinstance(images, torch.Tensor): + images = images.cpu().float().numpy() + if isinstance(targets, torch.Tensor): + targets = targets.cpu().numpy() + if np.max(images[0]) <= 1: + images *= 255 # de-normalise (optional) + bs, _, h, w = images.shape # batch size, _, height, width + bs = min(bs, max_subplots) # limit plot images + ns = np.ceil(bs ** 0.5) # number of subplots (square) + + # Build Image + mosaic = np.full((int(ns * h), int(ns * w), 3), 255, dtype=np.uint8) # init + for i, im in enumerate(images): + if i == max_subplots: # if last batch has fewer images than we expect + break + x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin + im = im.transpose(1, 2, 0) + mosaic[y:y + h, x:x + w, :] = im + + # Resize (optional) + scale = max_size / ns / max(h, w) + if scale < 1: + h = math.ceil(scale * h) + w = math.ceil(scale * w) + mosaic = cv2.resize(mosaic, tuple(int(x * ns) for x in (w, h))) + + # Annotate + fs = int((h + w) * ns * 0.01) # font size + annotator = Annotator(mosaic, line_width=round(fs / 10), font_size=fs, pil=True, example=names) + for i in range(i + 1): + x, y = int(w * (i // ns)), int(h * (i % ns)) # block origin + annotator.rectangle([x, y, x + w, y + h], None, (255, 255, 255), width=2) # borders + if paths: + annotator.text((x + 5, y + 5 + h), text=Path(paths[i]).name[:40], txt_color=(220, 220, 220)) # filenames + if len(targets) > 0: + ti = targets[targets[:, 0] == i] # image targets + boxes = xywh2xyxy(ti[:, 2:6]).T + classes = ti[:, 1].astype('int') + labels = ti.shape[1] == 6 # labels if no conf column + conf = None if labels else ti[:, 6] # check for confidence presence (label vs pred) + + if boxes.shape[1]: + if boxes.max() <= 1.01: # if normalized with tolerance 0.01 + boxes[[0, 2]] *= w # scale to pixels + boxes[[1, 3]] *= h + elif scale < 1: # absolute coords need scale if image scales + boxes *= scale + boxes[[0, 2]] += x + boxes[[1, 3]] += y + for j, box in enumerate(boxes.T.tolist()): + cls = classes[j] + color = colors(cls) + cls = names[cls] if names else cls + if labels or conf[j] > 0.25: # 0.25 conf thresh + label = f'{cls}' if labels else f'{cls} {conf[j]:.1f}' + annotator.box_label(box, label, color=color) + annotator.im.save(fname) # save + + +def plot_lr_scheduler(optimizer, scheduler, epochs=300, save_dir=''): + # Plot LR simulating training for full epochs + optimizer, scheduler = copy(optimizer), copy(scheduler) # do not modify originals + y = [] + for _ in range(epochs): + scheduler.step() + y.append(optimizer.param_groups[0]['lr']) + plt.plot(y, '.-', label='LR') + plt.xlabel('epoch') + plt.ylabel('LR') + plt.grid() + plt.xlim(0, epochs) + plt.ylim(0) + plt.savefig(Path(save_dir) / 'LR.png', dpi=200) + plt.close() + + +def plot_val_txt(): # from utils.plots import *; plot_val() + # Plot val.txt histograms + x = np.loadtxt('val.txt', dtype=np.float32) + box = xyxy2xywh(x[:, :4]) + cx, cy = box[:, 0], box[:, 1] + + fig, ax = plt.subplots(1, 1, figsize=(6, 6), tight_layout=True) + ax.hist2d(cx, cy, bins=600, cmax=10, cmin=0) + ax.set_aspect('equal') + plt.savefig('hist2d.png', dpi=300) + + fig, ax = plt.subplots(1, 2, figsize=(12, 6), tight_layout=True) + ax[0].hist(cx, bins=600) + ax[1].hist(cy, bins=600) + plt.savefig('hist1d.png', dpi=200) + + +def plot_targets_txt(): # from utils.plots import *; plot_targets_txt() + # Plot targets.txt histograms + x = np.loadtxt('targets.txt', dtype=np.float32).T + s = ['x targets', 'y targets', 'width targets', 'height targets'] + fig, ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True) + ax = ax.ravel() + for i in range(4): + ax[i].hist(x[i], bins=100, label=f'{x[i].mean():.3g} +/- {x[i].std():.3g}') + ax[i].legend() + ax[i].set_title(s[i]) + plt.savefig('targets.jpg', dpi=200) + + +def plot_val_study(file='', dir='', x=None): # from utils.plots import *; plot_val_study() + # Plot file=study.txt generated by val.py (or plot all study*.txt in dir) + save_dir = Path(file).parent if file else Path(dir) + plot2 = False # plot additional results + if plot2: + ax = plt.subplots(2, 4, figsize=(10, 6), tight_layout=True)[1].ravel() + + fig2, ax2 = plt.subplots(1, 1, figsize=(8, 4), tight_layout=True) + # for f in [save_dir / f'study_coco_{x}.txt' for x in ['yolov5n6', 'yolov5s6', 'yolov5m6', 'yolov5l6', 'yolov5x6']]: + for f in sorted(save_dir.glob('study*.txt')): + y = np.loadtxt(f, dtype=np.float32, usecols=[0, 1, 2, 3, 7, 8, 9], ndmin=2).T + x = np.arange(y.shape[1]) if x is None else np.array(x) + if plot2: + s = ['P', 'R', 'mAP@.5', 'mAP@.5:.95', 't_preprocess (ms/img)', 't_inference (ms/img)', 't_NMS (ms/img)'] + for i in range(7): + ax[i].plot(x, y[i], '.-', linewidth=2, markersize=8) + ax[i].set_title(s[i]) + + j = y[3].argmax() + 1 + ax2.plot(y[5, 1:j], + y[3, 1:j] * 1E2, + '.-', + linewidth=2, + markersize=8, + label=f.stem.replace('study_coco_', '').replace('yolo', 'YOLO')) + + ax2.plot(1E3 / np.array([209, 140, 97, 58, 35, 18]), [34.6, 40.5, 43.0, 47.5, 49.7, 51.5], + 'k.-', + linewidth=2, + markersize=8, + alpha=.25, + label='EfficientDet') + + ax2.grid(alpha=0.2) + ax2.set_yticks(np.arange(20, 60, 5)) + ax2.set_xlim(0, 57) + ax2.set_ylim(25, 55) + ax2.set_xlabel('GPU Speed (ms/img)') + ax2.set_ylabel('COCO AP val') + ax2.legend(loc='lower right') + f = save_dir / 'study.png' + print(f'Saving {f}...') + plt.savefig(f, dpi=300) + + +@try_except # known issue https://github.com/ultralytics/yolov5/issues/5395 +@Timeout(30) # known issue https://github.com/ultralytics/yolov5/issues/5611 +def plot_labels(labels, names=(), save_dir=Path('')): + # plot dataset labels + LOGGER.info(f"Plotting labels to {save_dir / 'labels.jpg'}... ") + c, b = labels[:, 0], labels[:, 1:].transpose() # classes, boxes + nc = int(c.max() + 1) # number of classes + x = pd.DataFrame(b.transpose(), columns=['x', 'y', 'width', 'height']) + + # seaborn correlogram + sn.pairplot(x, corner=True, diag_kind='auto', kind='hist', diag_kws=dict(bins=50), plot_kws=dict(pmax=0.9)) + plt.savefig(save_dir / 'labels_correlogram.jpg', dpi=200) + plt.close() + + # matplotlib labels + matplotlib.use('svg') # faster + ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)[1].ravel() + y = ax[0].hist(c, bins=np.linspace(0, nc, nc + 1) - 0.5, rwidth=0.8) + try: # color histogram bars by class + [y[2].patches[i].set_color([x / 255 for x in colors(i)]) for i in range(nc)] # known issue #3195 + except Exception: + pass + ax[0].set_ylabel('instances') + if 0 < len(names) < 30: + ax[0].set_xticks(range(len(names))) + ax[0].set_xticklabels(names, rotation=90, fontsize=10) + else: + ax[0].set_xlabel('classes') + sn.histplot(x, x='x', y='y', ax=ax[2], bins=50, pmax=0.9) + sn.histplot(x, x='width', y='height', ax=ax[3], bins=50, pmax=0.9) + + # rectangles + labels[:, 1:3] = 0.5 # center + labels[:, 1:] = xywh2xyxy(labels[:, 1:]) * 2000 + img = Image.fromarray(np.ones((2000, 2000, 3), dtype=np.uint8) * 255) + for cls, *box in labels[:1000]: + ImageDraw.Draw(img).rectangle(box, width=1, outline=colors(cls)) # plot + ax[1].imshow(img) + ax[1].axis('off') + + for a in [0, 1, 2, 3]: + for s in ['top', 'right', 'left', 'bottom']: + ax[a].spines[s].set_visible(False) + + plt.savefig(save_dir / 'labels.jpg', dpi=200) + matplotlib.use('Agg') + plt.close() + + +def imshow_cls(im, labels=None, pred=None, names=None, nmax=25, verbose=False, f=Path('images.jpg')): + # Show classification image grid with labels (optional) and predictions (optional) + from utils.augmentations import denormalize + + names = names or [f'class{i}' for i in range(1000)] + blocks = torch.chunk(denormalize(im.clone()).cpu().float(), len(im), + dim=0) # select batch index 0, block by channels + n = min(len(blocks), nmax) # number of plots + m = min(8, round(n ** 0.5)) # 8 x 8 default + fig, ax = plt.subplots(math.ceil(n / m), m) # 8 rows x n/8 cols + ax = ax.ravel() if m > 1 else [ax] + # plt.subplots_adjust(wspace=0.05, hspace=0.05) + for i in range(n): + ax[i].imshow(blocks[i].squeeze().permute((1, 2, 0)).numpy().clip(0.0, 1.0)) + ax[i].axis('off') + if labels is not None: + s = names[labels[i]] + (f'鈥攞names[pred[i]]}' if pred is not None else '') + ax[i].set_title(s, fontsize=8, verticalalignment='top') + plt.savefig(f, dpi=300, bbox_inches='tight') + plt.close() + if verbose: + LOGGER.info(f"Saving {f}") + if labels is not None: + LOGGER.info('True: ' + ' '.join(f'{names[i]:3s}' for i in labels[:nmax])) + if pred is not None: + LOGGER.info('Predicted:' + ' '.join(f'{names[i]:3s}' for i in pred[:nmax])) + return f + + +def plot_evolve(evolve_csv='path/to/evolve.csv'): # from utils.plots import *; plot_evolve() + # Plot evolve.csv hyp evolution results + evolve_csv = Path(evolve_csv) + data = pd.read_csv(evolve_csv) + keys = [x.strip() for x in data.columns] + x = data.values + f = fitness(x) + j = np.argmax(f) # max fitness index + plt.figure(figsize=(10, 12), tight_layout=True) + matplotlib.rc('font', **{'size': 8}) + print(f'Best results from row {j} of {evolve_csv}:') + for i, k in enumerate(keys[7:]): + v = x[:, 7 + i] + mu = v[j] # best single result + plt.subplot(6, 5, i + 1) + plt.scatter(v, f, c=hist2d(v, f, 20), cmap='viridis', alpha=.8, edgecolors='none') + plt.plot(mu, f.max(), 'k+', markersize=15) + plt.title(f'{k} = {mu:.3g}', fontdict={'size': 9}) # limit to 40 characters + if i % 5 != 0: + plt.yticks([]) + print(f'{k:>15}: {mu:.3g}') + f = evolve_csv.with_suffix('.png') # filename + plt.savefig(f, dpi=200) + plt.close() + print(f'Saved {f}') + + +def plot_results(file='path/to/results.csv', dir=''): + # Plot training results.csv. Usage: from utils.plots import *; plot_results('path/to/results.csv') + save_dir = Path(file).parent if file else Path(dir) + fig, ax = plt.subplots(2, 5, figsize=(12, 6), tight_layout=True) + ax = ax.ravel() + files = list(save_dir.glob('results*.csv')) + assert len(files), f'No results.csv files found in {save_dir.resolve()}, nothing to plot.' + for f in files: + try: + data = pd.read_csv(f) + s = [x.strip() for x in data.columns] + x = data.values[:, 0] + for i, j in enumerate([1, 2, 3, 4, 5, 8, 9, 10, 6, 7]): + y = data.values[:, j].astype('float') + # y[y == 0] = np.nan # don't show zero values + ax[i].plot(x, y, marker='.', label=f.stem, linewidth=2, markersize=8) + ax[i].set_title(s[j], fontsize=12) + # if j in [8, 9, 10]: # share train and val loss y axes + # ax[i].get_shared_y_axes().join(ax[i], ax[i - 5]) + except Exception as e: + LOGGER.info(f'Warning: Plotting error for {f}: {e}') + ax[1].legend() + fig.savefig(save_dir / 'results.png', dpi=200) + plt.close() + + +def profile_idetection(start=0, stop=0, labels=(), save_dir=''): + # Plot iDetection '*.txt' per-image logs. from utils.plots import *; profile_idetection() + ax = plt.subplots(2, 4, figsize=(12, 6), tight_layout=True)[1].ravel() + s = ['Images', 'Free Storage (GB)', 'RAM Usage (GB)', 'Battery', 'dt_raw (ms)', 'dt_smooth (ms)', 'real-world FPS'] + files = list(Path(save_dir).glob('frames*.txt')) + for fi, f in enumerate(files): + try: + results = np.loadtxt(f, ndmin=2).T[:, 90:-30] # clip first and last rows + n = results.shape[1] # number of rows + x = np.arange(start, min(stop, n) if stop else n) + results = results[:, x] + t = (results[0] - results[0].min()) # set t0=0s + results[0] = x + for i, a in enumerate(ax): + if i < len(results): + label = labels[fi] if len(labels) else f.stem.replace('frames_', '') + a.plot(t, results[i], marker='.', label=label, linewidth=1, markersize=5) + a.set_title(s[i]) + a.set_xlabel('time (s)') + # if fi == len(files) - 1: + # a.set_ylim(bottom=0) + for side in ['top', 'right']: + a.spines[side].set_visible(False) + else: + a.remove() + except Exception as e: + print(f'Warning: Plotting error for {f}; {e}') + ax[1].legend() + plt.savefig(Path(save_dir) / 'idetection_profile.png', dpi=200) + + +def save_one_box(xyxy, im, file=Path('im.jpg'), gain=1.02, pad=10, square=False, BGR=False, save=True): + # Save image crop as {file} with crop size multiple {gain} and {pad} pixels. Save and/or return crop + xyxy = torch.tensor(xyxy).view(-1, 4) + b = xyxy2xywh(xyxy) # boxes + if square: + b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # attempt rectangle to square + b[:, 2:] = b[:, 2:] * gain + pad # box wh * gain + pad + xyxy = xywh2xyxy(b).long() + clip_coords(xyxy, im.shape) + crop = im[int(xyxy[0, 1]):int(xyxy[0, 3]), int(xyxy[0, 0]):int(xyxy[0, 2]), ::(1 if BGR else -1)] + if save: + file.parent.mkdir(parents=True, exist_ok=True) # make directory + f = str(increment_path(file).with_suffix('.jpg')) + # cv2.imwrite(f, crop) # save BGR, https://github.com/ultralytics/yolov5/issues/7007 chroma subsampling issue + Image.fromarray(crop[..., ::-1]).save(f, quality=95, subsampling=0) # save RGB + return crop diff --git a/application/yolov5_example/utils/torch_utils.py b/application/yolov5_example/utils/torch_utils.py new file mode 100644 index 00000000..350c506e --- /dev/null +++ b/application/yolov5_example/utils/torch_utils.py @@ -0,0 +1,433 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +PyTorch utils +""" + +import math +import os +import platform +import subprocess +import time +import warnings +from contextlib import contextmanager +from copy import deepcopy +from pathlib import Path + +import torch +import torch.distributed as dist +import torch.nn as nn +import torch.nn.functional as F +from torch.nn.parallel import DistributedDataParallel as DDP + +from utils.general import LOGGER, check_version, colorstr, file_date, git_describe + +LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html +RANK = int(os.getenv('RANK', -1)) +WORLD_SIZE = int(os.getenv('WORLD_SIZE', 1)) + +try: + import thop # for FLOPs computation +except ImportError: + thop = None + +# Suppress PyTorch warnings +warnings.filterwarnings('ignore', message='User provided device_type of \'cuda\', but CUDA is not available. Disabling') + + +def smart_inference_mode(torch_1_9=check_version(torch.__version__, '1.9.0')): + # Applies torch.inference_mode() decorator if torch>=1.9.0 else torch.no_grad() decorator + def decorate(fn): + return (torch.inference_mode if torch_1_9 else torch.no_grad)()(fn) + + return decorate + + +def smartCrossEntropyLoss(label_smoothing=0.0): + # Returns nn.CrossEntropyLoss with label smoothing enabled for torch>=1.10.0 + if check_version(torch.__version__, '1.10.0'): + return nn.CrossEntropyLoss(label_smoothing=label_smoothing) # loss function + else: + if label_smoothing > 0: + LOGGER.warning(f'WARNING: label smoothing {label_smoothing} requires torch>=1.10.0') + return nn.CrossEntropyLoss() # loss function + + +def smart_DDP(model): + # Model DDP creation with checks + assert not check_version(torch.__version__, '1.12.0', pinned=True), \ + 'torch==1.12.0 torchvision==0.13.0 DDP training is not supported due to a known issue. ' \ + 'Please upgrade or downgrade torch to use DDP. See https://github.com/ultralytics/yolov5/issues/8395' + if check_version(torch.__version__, '1.11.0'): + return DDP(model, device_ids=[LOCAL_RANK], output_device=LOCAL_RANK, static_graph=True) + else: + return DDP(model, device_ids=[LOCAL_RANK], output_device=LOCAL_RANK) + + +def reshape_classifier_output(model, n=1000): + # Update a TorchVision classification model to class count 'n' if required + from models.common import Classify + name, m = list((model.model if hasattr(model, 'model') else model).named_children())[-1] # last module + if isinstance(m, Classify): # YOLOv5 Classify() head + if m.linear.out_features != n: + m.linear = nn.Linear(m.linear.in_features, n) + elif isinstance(m, nn.Linear): # ResNet, EfficientNet + if m.out_features != n: + setattr(model, name, nn.Linear(m.in_features, n)) + elif isinstance(m, nn.Sequential): + types = [type(x) for x in m] + if nn.Linear in types: + i = types.index(nn.Linear) # nn.Linear index + if m[i].out_features != n: + m[i] = nn.Linear(m[i].in_features, n) + elif nn.Conv2d in types: + i = types.index(nn.Conv2d) # nn.Conv2d index + if m[i].out_channels != n: + m[i] = nn.Conv2d(m[i].in_channels, n, m[i].kernel_size, m[i].stride, bias=m[i].bias) + + +@contextmanager +def torch_distributed_zero_first(local_rank: int): + # Decorator to make all processes in distributed training wait for each local_master to do something + if local_rank not in [-1, 0]: + dist.barrier(device_ids=[local_rank]) + yield + if local_rank == 0: + dist.barrier(device_ids=[0]) + + +def device_count(): + # Returns number of CUDA devices available. Safe version of torch.cuda.device_count(). Supports Linux and Windows + assert platform.system() in ('Linux', 'Windows'), 'device_count() only supported on Linux or Windows' + try: + cmd = 'nvidia-smi -L | wc -l' if platform.system() == 'Linux' else 'nvidia-smi -L | find /c /v ""' # Windows + return int(subprocess.run(cmd, shell=True, capture_output=True, check=True).stdout.decode().split()[-1]) + except Exception: + return 0 + + +def select_device(device='', batch_size=0, newline=True): + # device = None or 'cpu' or 0 or '0' or '0,1,2,3' + s = f'YOLOv5 馃殌 {git_describe() or file_date()} Python-{platform.python_version()} torch-{torch.__version__} ' + device = str(device).strip().lower().replace('cuda:', '').replace('none', '') # to string, 'cuda:0' to '0' + cpu = device == 'cpu' + mps = device == 'mps' # Apple Metal Performance Shaders (MPS) + if cpu or mps: + os.environ['CUDA_VISIBLE_DEVICES'] = '-1' # force torch.cuda.is_available() = False + elif device: # non-cpu device requested + os.environ['CUDA_VISIBLE_DEVICES'] = device # set environment variable - must be before assert is_available() + assert torch.cuda.is_available() and torch.cuda.device_count() >= len(device.replace(',', '')), \ + f"Invalid CUDA '--device {device}' requested, use '--device cpu' or pass valid CUDA device(s)" + + if not (cpu or mps) and torch.cuda.is_available(): # prefer GPU if available + devices = device.split(',') if device else '0' # range(torch.cuda.device_count()) # i.e. 0,1,6,7 + n = len(devices) # device count + if n > 1 and batch_size > 0: # check batch_size is divisible by device_count + assert batch_size % n == 0, f'batch-size {batch_size} not multiple of GPU count {n}' + space = ' ' * (len(s) + 1) + for i, d in enumerate(devices): + p = torch.cuda.get_device_properties(i) + s += f"{'' if i == 0 else space}CUDA:{d} ({p.name}, {p.total_memory / (1 << 20):.0f}MiB)\n" # bytes to MB + arg = 'cuda:0' + elif mps and getattr(torch, 'has_mps', False) and torch.backends.mps.is_available(): # prefer MPS if available + s += 'MPS\n' + arg = 'mps' + else: # revert to CPU + s += 'CPU\n' + arg = 'cpu' + + if not newline: + s = s.rstrip() + LOGGER.info(s) + return torch.device(arg) + + +def time_sync(): + # PyTorch-accurate time + if torch.cuda.is_available(): + torch.cuda.synchronize() + return time.time() + + +def profile(input, ops, n=10, device=None): + """ YOLOv5 speed/memory/FLOPs profiler + Usage: + input = torch.randn(16, 3, 640, 640) + m1 = lambda x: x * torch.sigmoid(x) + m2 = nn.SiLU() + profile(input, [m1, m2], n=100) # profile over 100 iterations + """ + results = [] + if not isinstance(device, torch.device): + device = select_device(device) + print(f"{'Params':>12s}{'GFLOPs':>12s}{'GPU_mem (GB)':>14s}{'forward (ms)':>14s}{'backward (ms)':>14s}" + f"{'input':>24s}{'output':>24s}") + + for x in input if isinstance(input, list) else [input]: + x = x.to(device) + x.requires_grad = True + for m in ops if isinstance(ops, list) else [ops]: + m = m.to(device) if hasattr(m, 'to') else m # device + m = m.half() if hasattr(m, 'half') and isinstance(x, torch.Tensor) and x.dtype is torch.float16 else m + tf, tb, t = 0, 0, [0, 0, 0] # dt forward, backward + try: + flops = thop.profile(m, inputs=(x,), verbose=False)[0] / 1E9 * 2 # GFLOPs + except Exception: + flops = 0 + + try: + for _ in range(n): + t[0] = time_sync() + y = m(x) + t[1] = time_sync() + try: + _ = (sum(yi.sum() for yi in y) if isinstance(y, list) else y).sum().backward() + t[2] = time_sync() + except Exception: # no backward method + # print(e) # for debug + t[2] = float('nan') + tf += (t[1] - t[0]) * 1000 / n # ms per op forward + tb += (t[2] - t[1]) * 1000 / n # ms per op backward + mem = torch.cuda.memory_reserved() / 1E9 if torch.cuda.is_available() else 0 # (GB) + s_in, s_out = (tuple(x.shape) if isinstance(x, torch.Tensor) else 'list' for x in (x, y)) # shapes + p = sum(x.numel() for x in m.parameters()) if isinstance(m, nn.Module) else 0 # parameters + print(f'{p:12}{flops:12.4g}{mem:>14.3f}{tf:14.4g}{tb:14.4g}{str(s_in):>24s}{str(s_out):>24s}') + results.append([p, flops, mem, tf, tb, s_in, s_out]) + except Exception as e: + print(e) + results.append(None) + torch.cuda.empty_cache() + return results + + +def is_parallel(model): + # Returns True if model is of type DP or DDP + return type(model) in (nn.parallel.DataParallel, nn.parallel.DistributedDataParallel) + + +def de_parallel(model): + # De-parallelize a model: returns single-GPU model if model is of type DP or DDP + return model.module if is_parallel(model) else model + + +def initialize_weights(model): + for m in model.modules(): + t = type(m) + if t is nn.Conv2d: + pass # nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') + elif t is nn.BatchNorm2d: + m.eps = 1e-3 + m.momentum = 0.03 + elif t in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU]: + m.inplace = True + + +def find_modules(model, mclass=nn.Conv2d): + # Finds layer indices matching module class 'mclass' + return [i for i, m in enumerate(model.module_list) if isinstance(m, mclass)] + + +def sparsity(model): + # Return global model sparsity + a, b = 0, 0 + for p in model.parameters(): + a += p.numel() + b += (p == 0).sum() + return b / a + + +def prune(model, amount=0.3): + # Prune model to requested global sparsity + import torch.nn.utils.prune as prune + for name, m in model.named_modules(): + if isinstance(m, nn.Conv2d): + prune.l1_unstructured(m, name='weight', amount=amount) # prune + prune.remove(m, 'weight') # make permanent + LOGGER.info(f'Model pruned to {sparsity(model):.3g} global sparsity') + + +def fuse_conv_and_bn(conv, bn): + # Fuse Conv2d() and BatchNorm2d() layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/ + fusedconv = nn.Conv2d(conv.in_channels, + conv.out_channels, + kernel_size=conv.kernel_size, + stride=conv.stride, + padding=conv.padding, + groups=conv.groups, + bias=True).requires_grad_(False).to(conv.weight.device) + + # Prepare filters + w_conv = conv.weight.clone().view(conv.out_channels, -1) + w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps + bn.running_var))) + fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.shape)) + + # Prepare spatial bias + b_conv = torch.zeros(conv.weight.size(0), device=conv.weight.device) if conv.bias is None else conv.bias + b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps)) + fusedconv.bias.copy_(torch.mm(w_bn, b_conv.reshape(-1, 1)).reshape(-1) + b_bn) + + return fusedconv + + +def model_info(model, verbose=False, imgsz=640): + # Model information. img_size may be int or list, i.e. img_size=640 or img_size=[640, 320] + n_p = sum(x.numel() for x in model.parameters()) # number parameters + n_g = sum(x.numel() for x in model.parameters() if x.requires_grad) # number gradients + if verbose: + print(f"{'layer':>5} {'name':>40} {'gradient':>9} {'parameters':>12} {'shape':>20} {'mu':>10} {'sigma':>10}") + for i, (name, p) in enumerate(model.named_parameters()): + name = name.replace('module_list.', '') + print('%5g %40s %9s %12g %20s %10.3g %10.3g' % + (i, name, p.requires_grad, p.numel(), list(p.shape), p.mean(), p.std())) + + try: # FLOPs + p = next(model.parameters()) + stride = max(int(model.stride.max()), 32) if hasattr(model, 'stride') else 32 # max stride + im = torch.zeros((1, p.shape[1], stride, stride), device=p.device) # input image in BCHW format + flops = thop.profile(deepcopy(model), inputs=(im,), verbose=False)[0] / 1E9 * 2 # stride GFLOPs + imgsz = imgsz if isinstance(imgsz, list) else [imgsz, imgsz] # expand if int/float + fs = f', {flops * imgsz[0] / stride * imgsz[1] / stride:.1f} GFLOPs' # 640x640 GFLOPs + except Exception: + fs = '' + + name = Path(model.yaml_file).stem.replace('yolov5', 'YOLOv5') if hasattr(model, 'yaml_file') else 'Model' + LOGGER.info(f"{name} summary: {len(list(model.modules()))} layers, {n_p} parameters, {n_g} gradients{fs}") + + +def scale_img(img, ratio=1.0, same_shape=False, gs=32): # img(16,3,256,416) + # Scales img(bs,3,y,x) by ratio constrained to gs-multiple + if ratio == 1.0: + return img + h, w = img.shape[2:] + s = (int(h * ratio), int(w * ratio)) # new size + img = F.interpolate(img, size=s, mode='bilinear', align_corners=False) # resize + if not same_shape: # pad/crop img + h, w = (math.ceil(x * ratio / gs) * gs for x in (h, w)) + return F.pad(img, [0, w - s[1], 0, h - s[0]], value=0.447) # value = imagenet mean + + +def copy_attr(a, b, include=(), exclude=()): + # Copy attributes from b to a, options to only include [...] and to exclude [...] + for k, v in b.__dict__.items(): + if (len(include) and k not in include) or k.startswith('_') or k in exclude: + continue + else: + setattr(a, k, v) + + +def smart_optimizer(model, name='Adam', lr=0.001, momentum=0.9, decay=1e-5): + # YOLOv5 3-param group optimizer: 0) weights with decay, 1) weights no decay, 2) biases no decay + g = [], [], [] # optimizer parameter groups + bn = tuple(v for k, v in nn.__dict__.items() if 'Norm' in k) # normalization layers, i.e. BatchNorm2d() + for v in model.modules(): + if hasattr(v, 'bias') and isinstance(v.bias, nn.Parameter): # bias (no decay) + g[2].append(v.bias) + if isinstance(v, bn): # weight (no decay) + g[1].append(v.weight) + elif hasattr(v, 'weight') and isinstance(v.weight, nn.Parameter): # weight (with decay) + g[0].append(v.weight) + + if name == 'Adam': + optimizer = torch.optim.Adam(g[2], lr=lr, betas=(momentum, 0.999)) # adjust beta1 to momentum + elif name == 'AdamW': + optimizer = torch.optim.AdamW(g[2], lr=lr, betas=(momentum, 0.999), weight_decay=0.0) + elif name == 'RMSProp': + optimizer = torch.optim.RMSprop(g[2], lr=lr, momentum=momentum) + elif name == 'SGD': + optimizer = torch.optim.SGD(g[2], lr=lr, momentum=momentum, nesterov=True) + else: + raise NotImplementedError(f'Optimizer {name} not implemented.') + + optimizer.add_param_group({'params': g[0], 'weight_decay': decay}) # add g0 with weight_decay + optimizer.add_param_group({'params': g[1], 'weight_decay': 0.0}) # add g1 (BatchNorm2d weights) + LOGGER.info(f"{colorstr('optimizer:')} {type(optimizer).__name__}(lr={lr}) with parameter groups " + f"{len(g[1])} weight(decay=0.0), {len(g[0])} weight(decay={decay}), {len(g[2])} bias") + return optimizer + + +def smart_hub_load(repo='ultralytics/yolov5', model='yolov5s', **kwargs): + # YOLOv5 torch.hub.load() wrapper with smart error/issue handling + if check_version(torch.__version__, '1.9.1'): + kwargs['skip_validation'] = True # validation causes GitHub API rate limit errors + if check_version(torch.__version__, '1.12.0'): + kwargs['trust_repo'] = True # argument required starting in torch 0.12 + try: + return torch.hub.load(repo, model, **kwargs) + except Exception: + return torch.hub.load(repo, model, force_reload=True, **kwargs) + + +def smart_resume(ckpt, optimizer, ema=None, weights='yolov5s.pt', epochs=300, resume=True): + # Resume training from a partially trained checkpoint + best_fitness = 0.0 + start_epoch = ckpt['epoch'] + 1 + if ckpt['optimizer'] is not None: + optimizer.load_state_dict(ckpt['optimizer']) # optimizer + best_fitness = ckpt['best_fitness'] + if ema and ckpt.get('ema'): + ema.ema.load_state_dict(ckpt['ema'].float().state_dict()) # EMA + ema.updates = ckpt['updates'] + if resume: + assert start_epoch > 0, f'{weights} training to {epochs} epochs is finished, nothing to resume.\n' \ + f"Start a new training without --resume, i.e. 'python train.py --weights {weights}'" + LOGGER.info(f'Resuming training from {weights} from epoch {start_epoch} to {epochs} total epochs') + if epochs < start_epoch: + LOGGER.info(f"{weights} has been trained for {ckpt['epoch']} epochs. Fine-tuning for {epochs} more epochs.") + epochs += ckpt['epoch'] # finetune additional epochs + return best_fitness, start_epoch, epochs + + +class EarlyStopping: + # YOLOv5 simple early stopper + def __init__(self, patience=30): + self.best_fitness = 0.0 # i.e. mAP + self.best_epoch = 0 + self.patience = patience or float('inf') # epochs to wait after fitness stops improving to stop + self.possible_stop = False # possible stop may occur next epoch + + def __call__(self, epoch, fitness): + if fitness >= self.best_fitness: # >= 0 to allow for early zero-fitness stage of training + self.best_epoch = epoch + self.best_fitness = fitness + delta = epoch - self.best_epoch # epochs without improvement + self.possible_stop = delta >= (self.patience - 1) # possible stop may occur next epoch + stop = delta >= self.patience # stop training if patience exceeded + if stop: + LOGGER.info(f'Stopping training early as no improvement observed in last {self.patience} epochs. ' + f'Best results observed at epoch {self.best_epoch}, best model saved as best.pt.\n' + f'To update EarlyStopping(patience={self.patience}) pass a new patience value, ' + f'i.e. `python train.py --patience 300` or use `--patience 0` to disable EarlyStopping.') + return stop + + +class ModelEMA: + """ Updated Exponential Moving Average (EMA) from https://github.com/rwightman/pytorch-image-models + Keeps a moving average of everything in the model state_dict (parameters and buffers) + For EMA details see https://www.tensorflow.org/api_docs/python/tf/train/ExponentialMovingAverage + """ + + def __init__(self, model, decay=0.9999, tau=2000, updates=0): + # Create EMA + self.ema = deepcopy(de_parallel(model)).eval() # FP32 EMA + # if next(model.parameters()).device.type != 'cpu': + # self.ema.half() # FP16 EMA + self.updates = updates # number of EMA updates + self.decay = lambda x: decay * (1 - math.exp(-x / tau)) # decay exponential ramp (to help early epochs) + for p in self.ema.parameters(): + p.requires_grad_(False) + + @smart_inference_mode() + def update(self, model): + return + # Update EMA parameters + self.updates += 1 + d = self.decay(self.updates) + + msd = de_parallel(model).state_dict() # model state_dict + for k, v in self.ema.state_dict().items(): + if v.dtype.is_floating_point: + v *= d + v += (1 - d) * msd[k].detach() + + def update_attr(self, model, include=(), exclude=('process_group', 'reducer')): + # Update EMA attributes + copy_attr(self.ema, model, include, exclude) diff --git a/application/yolov5_example/val.py b/application/yolov5_example/val.py new file mode 100644 index 00000000..13049623 --- /dev/null +++ b/application/yolov5_example/val.py @@ -0,0 +1,396 @@ +# YOLOv5 馃殌 by Ultralytics, GPL-3.0 license +""" +Validate a trained YOLOv5 model accuracy on a custom dataset + +Usage: + $ python path/to/val.py --weights yolov5s.pt --data coco128.yaml --img 640 + +Usage - formats: + $ python path/to/val.py --weights yolov5s.pt # PyTorch + yolov5s.torchscript # TorchScript + yolov5s.onnx # ONNX Runtime or OpenCV DNN with --dnn + yolov5s.xml # OpenVINO + yolov5s.engine # TensorRT + yolov5s.mlmodel # CoreML (macOS-only) + yolov5s_saved_model # TensorFlow SavedModel + yolov5s.pb # TensorFlow GraphDef + yolov5s.tflite # TensorFlow Lite + yolov5s_edgetpu.tflite # TensorFlow Edge TPU +""" + +import argparse +import json +import os +import sys +from pathlib import Path + +import numpy as np +import torch +from tqdm import tqdm + +FILE = Path(__file__).resolve() +ROOT = FILE.parents[0] # YOLOv5 root directory +if str(ROOT) not in sys.path: + sys.path.append(str(ROOT)) # add ROOT to PATH +ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative + +from models.common import DetectMultiBackend +from utils.callbacks import Callbacks +from utils.dataloaders import create_dataloader +from utils.general import (LOGGER, check_dataset, check_img_size, check_requirements, check_yaml, + coco80_to_coco91_class, colorstr, increment_path, non_max_suppression, print_args, + scale_coords, xywh2xyxy, xyxy2xywh) +from utils.metrics import ConfusionMatrix, ap_per_class, box_iou +from utils.plots import output_to_target, plot_images, plot_val_study +from utils.torch_utils import select_device, smart_inference_mode, time_sync + + +def save_one_txt(predn, save_conf, shape, file): + # Save one txt result + gn = torch.tensor(shape)[[1, 0, 1, 0]] # normalization gain whwh + for *xyxy, conf, cls in predn.tolist(): + xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh + line = (cls, *xywh, conf) if save_conf else (cls, *xywh) # label format + with open(file, 'a') as f: + f.write(('%g ' * len(line)).rstrip() % line + '\n') + + +def save_one_json(predn, jdict, path, class_map): + # Save one JSON result {"image_id": 42, "category_id": 18, "bbox": [258.15, 41.29, 348.26, 243.78], "score": 0.236} + image_id = int(path.stem) if path.stem.isnumeric() else path.stem + box = xyxy2xywh(predn[:, :4]) # xywh + box[:, :2] -= box[:, 2:] / 2 # xy center to top-left corner + for p, b in zip(predn.tolist(), box.tolist()): + jdict.append({ + 'image_id': image_id, + 'category_id': class_map[int(p[5])], + 'bbox': [round(x, 3) for x in b], + 'score': round(p[4], 5)}) + + +def process_batch(detections, labels, iouv): + """ + Return correct predictions matrix. Both sets of boxes are in (x1, y1, x2, y2) format. + Arguments: + detections (Array[N, 6]), x1, y1, x2, y2, conf, class + labels (Array[M, 5]), class, x1, y1, x2, y2 + Returns: + correct (Array[N, 10]), for 10 IoU levels + """ + correct = np.zeros((detections.shape[0], iouv.shape[0])).astype(bool) + iou = box_iou(labels[:, 1:], detections[:, :4]) + correct_class = labels[:, 0:1] == detections[:, 5] + for i in range(len(iouv)): + x = torch.where((iou >= iouv[i]) & correct_class) # IoU > threshold and classes match + if x[0].shape[0]: + matches = torch.cat((torch.stack(x, 1), iou[x[0], x[1]][:, None]), 1).cpu().numpy() # [label, detect, iou] + if x[0].shape[0] > 1: + matches = matches[matches[:, 2].argsort()[::-1]] + matches = matches[np.unique(matches[:, 1], return_index=True)[1]] + # matches = matches[matches[:, 2].argsort()[::-1]] + matches = matches[np.unique(matches[:, 0], return_index=True)[1]] + correct[matches[:, 1].astype(int), i] = True + return torch.tensor(correct, dtype=torch.bool, device=iouv.device) + + +@smart_inference_mode() +def run( + data, + weights=None, # model.pt path(s) + batch_size=32, # batch size + imgsz=640, # inference size (pixels) + conf_thres=0.001, # confidence threshold + iou_thres=0.6, # NMS IoU threshold + task='val', # train, val, test, speed or study + device='', # cuda device, i.e. 0 or 0,1,2,3 or cpu + workers=8, # max dataloader workers (per RANK in DDP mode) + single_cls=False, # treat as single-class dataset + augment=False, # augmented inference + verbose=False, # verbose output + save_txt=False, # save results to *.txt + save_hybrid=False, # save label+prediction hybrid results to *.txt + save_conf=False, # save confidences in --save-txt labels + save_json=False, # save a COCO-JSON results file + project=ROOT / 'runs/val', # save to project/name + name='exp', # save to project/name + exist_ok=False, # existing project/name ok, do not increment + half=True, # use FP16 half-precision inference + dnn=False, # use OpenCV DNN for ONNX inference + model=None, + dataloader=None, + save_dir=Path(''), + plots=True, + callbacks=Callbacks(), + compute_loss=None, +): + # Initialize/load model and set device + training = model is not None + if training: # called by train.py + device, pt, jit, engine = next(model.parameters()).device, True, False, False # get model device, PyTorch model + half &= device.type != 'cpu' # half precision only supported on CUDA + model.half() if half else model.float() + else: # called directly + device = select_device(device, batch_size=batch_size) + + # Directories + save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run + (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir + + # Load model + model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half) + stride, pt, jit, engine = model.stride, model.pt, model.jit, model.engine + imgsz = check_img_size(imgsz, s=stride) # check image size + half = model.fp16 # FP16 supported on limited backends with CUDA + if engine: + batch_size = model.batch_size + else: + device = model.device + if not (pt or jit): + batch_size = 1 # export.py models default to batch-size 1 + LOGGER.info(f'Forcing --batch-size 1 square inference (1,3,{imgsz},{imgsz}) for non-PyTorch models') + + # Data + data = check_dataset(data) # check + + # Configure + model.eval() + cuda = device.type != 'cpu' + is_coco = isinstance(data.get('val'), str) and data['val'].endswith(f'coco{os.sep}val2017.txt') # COCO dataset + nc = 1 if single_cls else int(data['nc']) # number of classes + iouv = torch.linspace(0.5, 0.95, 10, device=device) # iou vector for mAP@0.5:0.95 + niou = iouv.numel() + + # Dataloader + if not training: + if pt and not single_cls: # check --weights are trained on --data + ncm = model.model.nc + assert ncm == nc, f'{weights} ({ncm} classes) trained on different --data than what you passed ({nc} ' \ + f'classes). Pass correct combination of --weights and --data that are trained together.' + model.warmup(imgsz=(1 if pt else batch_size, 3, imgsz, imgsz)) # warmup + pad = 0.0 if task in ('speed', 'benchmark') else 0.5 + rect = False if task == 'benchmark' else pt # square inference for benchmarks + task = task if task in ('train', 'val', 'test') else 'val' # path to train/val/test images + dataloader = create_dataloader(data[task], + imgsz, + batch_size, + stride, + single_cls, + pad=pad, + rect=rect, + workers=workers, + prefix=colorstr(f'{task}: '))[0] + + seen = 0 + confusion_matrix = ConfusionMatrix(nc=nc) + names = dict(enumerate(model.names if hasattr(model, 'names') else model.module.names)) + class_map = coco80_to_coco91_class() if is_coco else list(range(1000)) + s = ('%20s' + '%11s' * 6) % ('Class', 'Images', 'Labels', 'P', 'R', 'mAP@.5', 'mAP@.5:.95') + dt, p, r, f1, mp, mr, map50, map = [0.0, 0.0, 0.0], 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 + loss = torch.zeros(3, device=device) + jdict, stats, ap, ap_class = [], [], [], [] + callbacks.run('on_val_start') + pbar = tqdm(dataloader, desc=s, bar_format='{l_bar}{bar:10}{r_bar}{bar:-10b}') # progress bar + for batch_i, (im, targets, paths, shapes) in enumerate(pbar): + callbacks.run('on_val_batch_start') + t1 = time_sync() + if cuda: + im = im.to(device, non_blocking=True) + targets = targets.to(device) + im = im.half() if half else im.float() # uint8 to fp16/32 + im /= 255 # 0 - 255 to 0.0 - 1.0 + nb, _, height, width = im.shape # batch size, channels, height, width + t2 = time_sync() + dt[0] += t2 - t1 + + # Inference + out, train_out = model(im) if training else model(im, augment=augment, val=True) # inference, loss outputs + dt[1] += time_sync() - t2 + + # Loss + if compute_loss: + loss += compute_loss([x.float() for x in train_out], targets)[1] # box, obj, cls + + # NMS + targets[:, 2:] *= torch.tensor((width, height, width, height), device=device) # to pixels + lb = [targets[targets[:, 0] == i, 1:] for i in range(nb)] if save_hybrid else [] # for autolabelling + t3 = time_sync() + out = non_max_suppression(out, conf_thres, iou_thres, labels=lb, multi_label=True, agnostic=single_cls) + dt[2] += time_sync() - t3 + + # Metrics + for si, pred in enumerate(out): + labels = targets[targets[:, 0] == si, 1:] + nl, npr = labels.shape[0], pred.shape[0] # number of labels, predictions + path, shape = Path(paths[si]), shapes[si][0] + correct = torch.zeros(npr, niou, dtype=torch.bool, device=device) # init + seen += 1 + + if npr == 0: + if nl: + stats.append((correct, *torch.zeros((2, 0), device=device), labels[:, 0])) + if plots: + confusion_matrix.process_batch(detections=None, labels=labels[:, 0]) + continue + + # Predictions + if single_cls: + pred[:, 5] = 0 + predn = pred.clone() + scale_coords(im[si].shape[1:], predn[:, :4], shape, shapes[si][1]) # native-space pred + + # Evaluate + if nl: + tbox = xywh2xyxy(labels[:, 1:5]) # target boxes + scale_coords(im[si].shape[1:], tbox, shape, shapes[si][1]) # native-space labels + labelsn = torch.cat((labels[:, 0:1], tbox), 1) # native-space labels + correct = process_batch(predn, labelsn, iouv) + if plots: + confusion_matrix.process_batch(predn, labelsn) + stats.append((correct, pred[:, 4], pred[:, 5], labels[:, 0])) # (correct, conf, pcls, tcls) + + # Save/log + if save_txt: + save_one_txt(predn, save_conf, shape, file=save_dir / 'labels' / f'{path.stem}.txt') + if save_json: + save_one_json(predn, jdict, path, class_map) # append to COCO-JSON dictionary + callbacks.run('on_val_image_end', pred, predn, path, names, im[si]) + + # Plot images + if plots and batch_i < 3: + plot_images(im, targets, paths, save_dir / f'val_batch{batch_i}_labels.jpg', names) # labels + plot_images(im, output_to_target(out), paths, save_dir / f'val_batch{batch_i}_pred.jpg', names) # pred + + callbacks.run('on_val_batch_end') + + # Compute metrics + stats = [torch.cat(x, 0).cpu().numpy() for x in zip(*stats)] # to numpy + if len(stats) and stats[0].any(): + tp, fp, p, r, f1, ap, ap_class = ap_per_class(*stats, plot=plots, save_dir=save_dir, names=names) + ap50, ap = ap[:, 0], ap.mean(1) # AP@0.5, AP@0.5:0.95 + mp, mr, map50, map = p.mean(), r.mean(), ap50.mean(), ap.mean() + nt = np.bincount(stats[3].astype(int), minlength=nc) # number of targets per class + + # Print results + pf = '%20s' + '%11i' * 2 + '%11.3g' * 4 # print format + LOGGER.info(pf % ('all', seen, nt.sum(), mp, mr, map50, map)) + if nt.sum() == 0: + LOGGER.warning(f'WARNING: no labels found in {task} set, can not compute metrics without labels 鈿狅笍') + + # Print results per class + if (verbose or (nc < 50 and not training)) and nc > 1 and len(stats): + for i, c in enumerate(ap_class): + LOGGER.info(pf % (names[c], seen, nt[c], p[i], r[i], ap50[i], ap[i])) + + # Print speeds + t = tuple(x / seen * 1E3 for x in dt) # speeds per image + if not training: + shape = (batch_size, 3, imgsz, imgsz) + LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {shape}' % t) + + # Plots + if plots: + confusion_matrix.plot(save_dir=save_dir, names=list(names.values())) + callbacks.run('on_val_end') + + # Save JSON + if save_json and len(jdict): + w = Path(weights[0] if isinstance(weights, list) else weights).stem if weights is not None else '' # weights + anno_json = str(Path(data.get('path', '../coco')) / 'annotations/instances_val2017.json') # annotations json + pred_json = str(save_dir / f"{w}_predictions.json") # predictions json + LOGGER.info(f'\nEvaluating pycocotools mAP... saving {pred_json}...') + with open(pred_json, 'w') as f: + json.dump(jdict, f) + + try: # https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocoEvalDemo.ipynb + check_requirements(['pycocotools']) + from pycocotools.coco import COCO + from pycocotools.cocoeval import COCOeval + + anno = COCO(anno_json) # init annotations api + pred = anno.loadRes(pred_json) # init predictions api + eval = COCOeval(anno, pred, 'bbox') + if is_coco: + eval.params.imgIds = [int(Path(x).stem) for x in dataloader.dataset.im_files] # image IDs to evaluate + eval.evaluate() + eval.accumulate() + eval.summarize() + map, map50 = eval.stats[:2] # update results (mAP@0.5:0.95, mAP@0.5) + except Exception as e: + LOGGER.info(f'pycocotools unable to run: {e}') + + # Return results + model.float() # for training + if not training: + s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else '' + LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}") + maps = np.zeros(nc) + map + for i, c in enumerate(ap_class): + maps[c] = ap[i] + return (mp, mr, map50, map, *(loss.cpu() / len(dataloader)).tolist()), maps, t + + +def parse_opt(): + parser = argparse.ArgumentParser() + parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='dataset.yaml path') + parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'yolov5s.pt', help='model.pt path(s)') + parser.add_argument('--batch-size', type=int, default=32, help='batch size') + parser.add_argument('--imgsz', '--img', '--img-size', type=int, default=640, help='inference size (pixels)') + parser.add_argument('--conf-thres', type=float, default=0.001, help='confidence threshold') + parser.add_argument('--iou-thres', type=float, default=0.6, help='NMS IoU threshold') + parser.add_argument('--task', default='val', help='train, val, test, speed or study') + parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') + parser.add_argument('--workers', type=int, default=8, help='max dataloader workers (per RANK in DDP mode)') + parser.add_argument('--single-cls', action='store_true', help='treat as single-class dataset') + parser.add_argument('--augment', action='store_true', help='augmented inference') + parser.add_argument('--verbose', action='store_true', help='report mAP by class') + parser.add_argument('--save-txt', action='store_true', help='save results to *.txt') + parser.add_argument('--save-hybrid', action='store_true', help='save label+prediction hybrid results to *.txt') + parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels') + parser.add_argument('--save-json', action='store_true', help='save a COCO-JSON results file') + parser.add_argument('--project', default=ROOT / 'runs/val', help='save to project/name') + parser.add_argument('--name', default='exp', help='save to project/name') + parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') + parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference') + parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference') + opt = parser.parse_args() + opt.data = check_yaml(opt.data) # check YAML + opt.save_json |= opt.data.endswith('coco.yaml') + opt.save_txt |= opt.save_hybrid + print_args(vars(opt)) + return opt + + +def main(opt): + check_requirements(requirements=ROOT / 'requirements.txt', exclude=('tensorboard', 'thop')) + + if opt.task in ('train', 'val', 'test'): # run normally + if opt.conf_thres > 0.001: # https://github.com/ultralytics/yolov5/issues/1466 + LOGGER.info(f'WARNING: confidence threshold {opt.conf_thres} > 0.001 produces invalid results 鈿狅笍') + run(**vars(opt)) + + else: + weights = opt.weights if isinstance(opt.weights, list) else [opt.weights] + opt.half = True # FP16 for fastest results + if opt.task == 'speed': # speed benchmarks + # python val.py --task speed --data coco.yaml --batch 1 --weights yolov5n.pt yolov5s.pt... + opt.conf_thres, opt.iou_thres, opt.save_json = 0.25, 0.45, False + for opt.weights in weights: + run(**vars(opt), plots=False) + + elif opt.task == 'study': # speed vs mAP benchmarks + # python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n.pt yolov5s.pt... + for opt.weights in weights: + f = f'study_{Path(opt.data).stem}_{Path(opt.weights).stem}.txt' # filename to save to + x, y = list(range(256, 1536 + 128, 128)), [] # x axis (image sizes), y axis + for opt.imgsz in x: # img-size + LOGGER.info(f'\nRunning {f} --imgsz {opt.imgsz}...') + r, _, t = run(**vars(opt), plots=False) + y.append(r + t) # results and times + np.savetxt(f, y, fmt='%10.4g') # save + os.system('zip -r study.zip study_*.txt') + plot_val_study(x=x) # plot + + +if __name__ == "__main__": + opt = parse_opt() + main(opt) diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index 3315cd49..4c3ec80b 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -70,6 +70,10 @@ def _layers_need_scale_form_input_fake_quantizer(self): qnnqat.Conv2d_sophgo, qnniqat.LinearReLU_sophgo, qnniqat.Linear_sophgo, + qnniqat.LinearBn1d_sophgo, + qnniqat.ConvTransposeBnReLU2d_sophgo, + qnniqat.ConvTransposeReLU2d_sophgo, + qnniqat.ConvTransposeBn2d_sophgo, ) def prepare(self, model: GraphModule, qconfig): diff --git a/mqbench/deploy/deploy_linear.py b/mqbench/deploy/deploy_linear.py index 2795ca4c..e54c9eff 100644 --- a/mqbench/deploy/deploy_linear.py +++ b/mqbench/deploy/deploy_linear.py @@ -115,13 +115,16 @@ def clip_weight(self, node, name2data, inp2node, named_initializer): logger.info(f'Clip weights <{tensor_name}> to range [{clip_range_min}, {clip_range_max}].') new_data = numpy_helper.from_array(new_data) named_initializer[tensor_name].raw_data = new_data.raw_data + def get_correct_sophgo_tpu_input_tensor_name(self, node, out2node): + input_0 = node.input[0] + tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') + if tensor_name[-1] == '_': + tensor_name = tensor_name[:-1] + return tensor_name def post_process_clip_ranges(self, clip_ranges, graph, inp2node, out2node): def find_the_closest_clip_range(node): - input_0 = node.input[0] - tensor_name = '{}_{}'.format(input_0, out2node[input_0].op_type if input_0 in out2node else '') - if tensor_name[-1] == '_': - tensor_name = tensor_name[:-1] + tensor_name = self.get_correct_sophgo_tpu_input_tensor_name(node, out2node) if tensor_name in clip_ranges: return tensor_name @@ -134,15 +137,50 @@ def find_the_closest_clip_range(node): if node.op_type in ['Flatten', 'Resize']: tensor_name = find_the_closest_clip_range(node) if tensor_name: - old = clip_ranges[tensor_name] - new_name = node.input[0] - new_name = '{}_{}'.format(new_name, out2node[new_name].op_type if new_name in out2node else '') - if new_name[-1] == '_': - new_name = tensor_name[:-1] - clip_ranges[new_name] = copy.deepcopy(old) + new_name = self.get_correct_sophgo_tpu_input_tensor_name(node, out2node) + clip_ranges[new_name] = copy.deepcopy(clip_ranges[tensor_name]) clip_ranges[new_name]['ori_name'] = new_name logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') return clip_ranges + def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): + op_type_inAndOutShouldSameClipRange = ['Flatten', 'Resize', 'Reshape', 'Transpose'] + for node in graph.node: + tensor_name = f'{node.output[0]}_{node.op_type}' + if tensor_name not in clip_ranges: + pre_op = node + finded = False + while pre_op.op_type in op_type_inAndOutShouldSameClipRange: + tensor_name2 = self.get_correct_sophgo_tpu_input_tensor_name(pre_op, out2node) + if tensor_name2 in clip_ranges: + finded = True + clip_ranges[tensor_name] = clip_ranges[tensor_name2] + print(f'pre_op finded, transfer {tensor_name2} to {tensor_name}') + break + if pre_op.input[0] in out2node: + pre_op = out2node[pre_op.input[0]] + else: + print(f'{pre_op.name}\'s pre_node not exist') + break + if not finded: + if node.output[0] in inp2node: + next_op = inp2node[node.output[0]][0][0] + while next_op.op_type in op_type_inAndOutShouldSameClipRange: + tensor_name2 = f'{next_op.output[0]}_{next_op.op_type}' + if tensor_name2 in clip_ranges: + finded = True + clip_ranges[tensor_name] = clip_ranges[tensor_name2] + print(f'next_op finded, transfer {tensor_name2} to {tensor_name}') + break + if next_op.output[0] in inp2node: + next_op = inp2node[next_op.output[0]][0][0] + else: + print(f'{next_op.name}\'s next_op not exist') + break + else: + print(f'{node.name}\'s next_op not exist') + if not finded: + print(f'Waring:{node.name}\'s clip_ranges not exist, maybe have some error') + return clip_ranges def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend): model = onnx.load(onnx_path) @@ -278,7 +316,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend) elif backend == 'ppl-cuda': context = {'ppl-cuda': clip_ranges} elif backend == 'sophgo_tpu': - #clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) + clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node) context = {'sophgo_tpu': clip_ranges} context['w_qscheme'] = '' context['a_qscheme'] = '' diff --git a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py index 6d9f3f38..64466b38 100644 --- a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py @@ -142,12 +142,6 @@ def _forward(self, input): weight_shape[0] = -1 bias_shape = [1] * len(self.weight.shape) bias_shape[1] = -1 - if torch.isnan(self.weight).any(): - print('weight have nan') - if self.input_fake_quantizer is not None and torch.isnan(self.input_fake_quantizer.scale).any(): - print('input_fake_quantizer.scale have nan') - if self.bias is not None and torch.isnan(self.bias).any(): - print('weight have nan') scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) #bias浼噺鍖 if self.weight_fake_quant.fake_quant_enabled[0] == 1: diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py index 84550d0b..a9d7c66e 100644 --- a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -76,13 +76,14 @@ def freeze_bn_stats(self): return self def bias_fake_quant(self, bias, scale_w, in_scale): - scale = scale_w*in_scale - if torch.nonzero(scale).size()[0] != scale.numel(): - print('Linear error! scale has 0, scale:', scale) + if bias is not None: + scale = scale_w*in_scale + if torch.nonzero(scale).size()[0] != scale.numel(): + print('Linear error! scale has 0, scale:', scale) - bias_q = bias/scale - bias = (bias_q.round()-bias_q).detach() + bias_q - bias = bias*scale + bias_q = bias/scale + bias = (bias_q.round()-bias_q).detach() + bias_q + bias = bias*scale return bias def _forward(self, input): diff --git a/mqbench/train_all_model.py b/mqbench/train_all_model.py index adf6805f..80d0d488 100644 --- a/mqbench/train_all_model.py +++ b/mqbench/train_all_model.py @@ -14,12 +14,12 @@ "--arch=resnet18 --batch-size=128", "--arch=vgg11_bn --batch-size=32", "--arch=resnet50 --batch-size=32", - #"--arch=squeezenet1_1 --batch-size=128", + "--arch=squeezenet1_1 --batch-size=128", #"--arch=mobilenet_v3_small --batch-size=128" ] -cmd_str = "--epochs=10 --lr=1e-4 --gpu=0 --pretrained --evaluate --backend=sophgo_tpu --optim=sgd --pre_eval_and_export --train_data=/data/imagenet/for_train_val/ --val_data=/data/imagenet/for_train_val/ --output_path=/workspace/tmp_path_1024"# --fast_test" +cmd_str = "--epochs=1 --deploy_batch_size=10 --lr=1e-4 --gpu=0 --pretrained --evaluate --backend=sophgo_tpu --optim=sgd --pre_eval_and_export --train_data=/data/imagenet/for_train_val/ --val_data=/data/imagenet/for_train_val/ --output_path=/workspace/tmp_path_1119"# --fast_test" def worker(cmd_line): os.system(cmd_line) From 252ede69c4de4bd01af97845a0658297c4faad09 Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Mon, 5 Dec 2022 15:52:02 +0800 Subject: [PATCH 07/29] for qat int8 release --- mqbench/convert_deploy.py | 14 +++++++++++--- mqbench/deploy/deploy_linear.py | 4 ++-- mqbench/prepare_by_platform.py | 3 +-- 3 files changed, 14 insertions(+), 7 deletions(-) diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index 936a7015..cfde0d4f 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -133,16 +133,24 @@ def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, * file_h.close() cali_table = osp.join(output_path, '{}_cali_table_from_mqbench_sophgo_tpu'.format(model_name)) with open(cali_table, 'w') as f: - f.write("# work_mode:QAT_all_int8 //Automatically generated, do not modify, work_mode choice:[QAT_all_int8, QAT_mix_prec]\n") + f.write("# work_mode:QAT_all_int8 #Automatically generated, do not modify, work_mode choice:[QAT_all_int8]\n") f.write("# op_name threshold min max\n") ori_layer_names = '' + weight_scale = [] for name,value in blob_range.items(): if 'threshold' in value: f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) ori_layer_names += '{},'.format(value['ori_name']) else: - f.write("{} {} {} {} {}\n".format(name, len(value['step']), ' '.join([str(i) for i in value['step']]), - len(value['zero_point']), ' '.join([str(i) for i in value['zero_point']]))) + tmpstr = "{} {} {} {} {}\n".format(name, len(value['step']), ' '.join([str(i) for i in value['step']]), + len(value['zero_point']), ' '.join([str(i) for i in value['zero_point']])) + if name.endswith('_weight') or name.endswith('_bias'): + weight_scale.append(tmpstr) + else: + f.write(tmpstr) + f.write('#weight_scale\n') + for i in weight_scale: + f.write(i) f.write("#{}\n".format(ori_layer_names[0:-1])) diff --git a/mqbench/deploy/deploy_linear.py b/mqbench/deploy/deploy_linear.py index e54c9eff..3a4e024b 100644 --- a/mqbench/deploy/deploy_linear.py +++ b/mqbench/deploy/deploy_linear.py @@ -178,8 +178,8 @@ def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): break else: print(f'{node.name}\'s next_op not exist') - if not finded: - print(f'Waring:{node.name}\'s clip_ranges not exist, maybe have some error') + # if not finded: + # print(f'Waring:{node.name}\'s clip_ranges not exist, maybe have some error') return clip_ranges def remove_fakequantize_and_collect_params(self, onnx_path, model_name, backend): diff --git a/mqbench/prepare_by_platform.py b/mqbench/prepare_by_platform.py index 866475a5..89d3e4eb 100644 --- a/mqbench/prepare_by_platform.py +++ b/mqbench/prepare_by_platform.py @@ -124,8 +124,7 @@ class BackendType(Enum): default_act_observer=EMAMinMaxObserver), BackendType.Sophgo_TPU: dict(qtype='affine', # noqa: E241 w_qscheme=QuantizeScheme(symmetry=True, per_channel=True, pot_scale=False, bit=8), - a_qscheme=QuantizeScheme(symmetry=False, per_channel=False, pot_scale=False, bit=8), - # b_qscheme=QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8), + a_qscheme=QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8), default_weight_quantize=LearnableFakeQuantize, default_act_quantize=LearnableFakeQuantize, default_weight_observer=MinMaxObserver, From cfddb1d06e5cf20ca4a67683f355ba995be4b77f Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Mon, 27 Feb 2023 11:20:14 +0800 Subject: [PATCH 08/29] fix linear+bn bug --- mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py index a9d7c66e..760a3847 100644 --- a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -63,7 +63,7 @@ def reset_bn_parameters(self): init.uniform_(self.bias, -bound, bound) def reset_parameters(self): - super(LinearBn1d, self).reset_parameters() + super(LinearBn1d_sophgo, self).reset_parameters() def update_bn_stats(self): self.freeze_bn = False From 21f39c0415bcccb069af68e37d295218473fd29c Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Wed, 12 Apr 2023 14:27:13 +0800 Subject: [PATCH 09/29] add int4&int8 mix prec func and infer net output shape in xxx_mqmoble_deploy_model.onnx --- application/imagenet_example/main.py | 170 ++++++--- application/yolov5_example/train.py | 41 +- mqbench/convert_deploy.py | 34 +- mqbench/custom_quantizer/model_quantizer.py | 9 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 200 +++++++++- mqbench/deploy/__init__.py | 1 + mqbench/deploy/deploy_sophgo.py | 349 ++++++++++++++++++ mqbench/fake_quantize/lsq.py | 22 +- mqbench/fake_quantize/quantize_base.py | 6 + mqbench/fusion_method.py | 29 ++ mqbench/nn/intrinsic/qat/modules/__init__.py | 1 + .../qat/modules/conv_fused_sophgo_tpu.py | 61 ++- .../qat/modules/deconv_fused_sophgo_tpu.py | 59 ++- .../qat/modules/linear_fused_sophgo_tpu.py | 45 ++- mqbench/prepare_by_platform.py | 104 +++++- mqbench/train_all_model.py | 30 +- 16 files changed, 1003 insertions(+), 158 deletions(-) create mode 100644 mqbench/deploy/deploy_sophgo.py diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index 5bfde3ad..f14b3db5 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -1,3 +1,4 @@ +#-- coding: gb2312 -- import argparse import os import random @@ -27,7 +28,7 @@ if name.islower() and not name.startswith("__") and callable(models.__dict__[name])) -cali_batch_size = 16 +cali_batch_size = 10 parser = argparse.ArgumentParser(description='PyTorch ImageNet Training') parser.add_argument('--train_data', metavar='DIR', help='path to dataset', required=True) @@ -93,6 +94,7 @@ parser.add_argument('--pre_eval_and_export', action='store_true') parser.add_argument('--deploy_batch_size', default=1, type=int, help='deploy_batch_size.') + BackendMap = {'tensorrt': BackendType.Tensorrt, 'sophgo_tpu': BackendType.Sophgo_TPU, 'nnie': BackendType.NNIE, @@ -107,10 +109,12 @@ def main(): args = parser.parse_args() + if args.output_path is None: args.output_path = './' args.output_path=os.path.join(args.output_path, args.arch) os.system('mkdir -p {}'.format(args.output_path)) + args.quant = not args.not_quant args.backend = BackendMap[args.backend] @@ -145,18 +149,6 @@ def main(): # Simply call main_worker function main_worker(args.gpu, ngpus_per_node, args) -layer_names = [] -features_out_hook = {} -i = 0 -def hook(module, fea_in, fea_out): - global i - if i >= len(layer_names): - return None - name = layer_names[i] - i += 1 - global features_out_hook - features_out_hook[name] = fea_out.cpu().numpy() - return None def main_worker(gpu, ngpus_per_node, args): global best_acc1 @@ -192,15 +184,17 @@ def main_worker(gpu, ngpus_per_node, args): state_dict = torch.load(args.model_path) print(f'load pretrained checkpoint from: {args.model_path}') model.load_state_dict(state_dict) + train_loader, train_sampler, val_loader, cali_loader = prepare_dataloader(args) criterion = nn.CrossEntropyLoss().cuda(args.gpu) if args.gpu is not None: model = model.cuda(args.gpu) else: model = model.cpu() + if args.pre_eval_and_export: print('原始onnx模型精度') - validate(val_loader, model.eval(), criterion, args) #?????model.cuda(),??? + validate(val_loader, model.eval(), criterion, args) #这里未执行model.cuda()，会报错 kwargs = { 'input_shape_dict': {'data': [args.deploy_batch_size, 3, 224, 224]}, @@ -213,15 +207,40 @@ def main_worker(gpu, ngpus_per_node, args): module_tmp = module_tmp.cpu() convert_onnx(module_tmp.eval(), **kwargs) del module_tmp - model = model.train() #prepare?????train??!! + model = model.train() #prepare前一定要是train模式！！ # quantize model if args.quant: prepare_custom_config_dict= { - # 'extra_qconfig_dict':{'w_fakequantize':'PACTFakeQuantize'} + # 'extra_qconfig_dict':{'w_fakequantize':'PACTFakeQuantize'}, + # 'work_mode':'int4_and_int8_mix', + + # 'work_mode':'all_int4_qat', #int4_and_int8_mix + # 'extra_qconfig_dict': { + # 'w_qscheme': { + # 'bit': 4, # custom bitwidth for weight, + # 'symmetry': True, # custom whether quant is symmetric for weight, + # 'per_channel': True, # custom whether quant is per-channel or per-tensor for weight, + # 'pot_scale': False, # custom whether scale is power of two for weight. + # }, + # 'a_qscheme': { + # 'bit': 4, # custom bitwidth for activation, + # 'symmetry': True, # custom whether quant is symmetric for activation, + # 'per_channel': False, # custom whether quant is per-channel or per-tensor for activation, + # 'pot_scale': False, # custom whether scale is power of two for activation. + # } + # } } model = prepare_by_platform(model, args.backend, prepare_custom_config_dict) - # print('prepared module:', model) + print('>>>>>prepared module:', model) + + if args.fast_test: + convert_deploy(model.eval(), args.backend, input_shape_dict= + {'data': [args.deploy_batch_size, 3, 224, 224]}, + model_name='{}_mqmoble'.format(args.arch), + # work_mode ='int4_and_int8_mix', + output_path=args.output_path) + if not torch.cuda.is_available(): print('using CPU, this will be slow') elif args.distributed: @@ -254,7 +273,7 @@ def main_worker(gpu, ngpus_per_node, args): if args.cpu: model = model.cpu() else: - # model = torch.nn.DataParallel(model).cuda() #???gpu?????????resume??cpu?? + # model = torch.nn.DataParallel(model).cuda() #会导致gpu训练保存的模型无法resume后用cpu推理 model = model.cuda() # define loss function (criterion) and optimizer @@ -310,9 +329,6 @@ def main_worker(gpu, ngpus_per_node, args): .format(args.resume, checkpoint['epoch'], best_acc1)) else: print("=> no checkpoint found at '{}'".format(args.resume)) - - - exit(1) if args.evaluate: @@ -327,6 +343,8 @@ def main_worker(gpu, ngpus_per_node, args): model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) exit(0) + if args.fast_test: + args.epochs = 1 for epoch in range(args.start_epoch, args.epochs): if args.distributed: train_sampler.set_epoch(epoch) @@ -342,31 +360,33 @@ def main_worker(gpu, ngpus_per_node, args): print(f'epoch{epoch}训练后eval精度:') acc1 = validate(val_loader, model, criterion, args) - # remember best acc@1 and save checkpoint - is_best = acc1 > best_acc1 - best_acc1 = max(acc1, best_acc1) - - if not args.multiprocessing_distributed or (args.multiprocessing_distributed - and args.rank % ngpus_per_node == 0): - save_checkpoint({ - 'epoch': epoch + 1, - 'arch': args.arch, - 'state_dict': model.state_dict(), - 'best_acc1': best_acc1, - 'optimizer' : optimizer.state_dict(), - }, is_best, filename=os.path.join(args.output_path, 'checkpoint.pth.tar')) - + # # remember best acc@1 and save checkpoint + # is_best = acc1 > best_acc1 + # best_acc1 = max(acc1, best_acc1) + + # if not args.multiprocessing_distributed or (args.multiprocessing_distributed + # and args.rank % ngpus_per_node == 0): + # save_checkpoint({ + # 'epoch': epoch + 1, + # 'arch': args.arch, + # 'state_dict': model.state_dict(), + # 'best_acc1': best_acc1, + # 'optimizer' : optimizer.state_dict(), + # }, is_best, filename=os.path.join(args.output_path, 'checkpoint.pth.tar')) + print('disable_all后测试精度:') disable_all(model) validate(val_loader, model, criterion, args) - enable_quantization(model) + gen_test_ref_data(cali_loader, model, args) convert_deploy(model.eval(), args.backend, input_shape_dict= {'data': [args.deploy_batch_size, 3, 224, 224]}, - model_name='{}_mqmoble'.format(args.arch), output_path=args.output_path) + model_name='{}_mqmoble'.format(args.arch), + # work_mode ='int4_and_int8_mix', + output_path=args.output_path) - model_path = os.path.join(args.output_path, '{}.pt'.format('{}_mqmoble'.format(args.arch))) + '''model_path = os.path.join(args.output_path, '{}.pt'.format('{}_mqmoble'.format(args.arch))) model_pt = torch.load(model_path) if args.gpu is not None: model_pt = model_pt.cuda(args.gpu) @@ -375,7 +395,7 @@ def main_worker(gpu, ngpus_per_node, args): print('load fused bn pt后测试精度:') validate(val_loader, model_pt, criterion, args) - validate_onnx(criterion, args) + validate_onnx(criterion, args)''' def prepare_dataloader(args): traindir = os.path.join(args.train_data, 'train') @@ -401,7 +421,7 @@ def prepare_dataloader(args): train_dataset, batch_size=args.batch_size, shuffle=(train_sampler is None), num_workers=args.workers, pin_memory=True, sampler=train_sampler) - cali_batch = 10 + cali_batch = 20 cali_dataset = torch.utils.data.Subset(train_dataset, indices=torch.arange(cali_batch_size * cali_batch)) cali_loader = torch.utils.data.DataLoader(cali_dataset, batch_size=cali_batch_size, shuffle=False, num_workers=args.workers, pin_memory=True) @@ -435,12 +455,7 @@ def prepare_dataloader_batch(args, batch_size): return val_loader -def get_node_name_by_module_name(qname, model): - nodes = list(model.graph.nodes) - modules = dict(model.named_modules()) - for node in nodes: - if node.target in modules and qname == node.target: - return node.name + def calibrate(cali_loader, model, args): model.eval() print("Start calibration ...") @@ -453,22 +468,72 @@ def calibrate(cali_loader, model, args): print("Calibration ==> ", i+1) print("End calibration.") return + +def get_node_name_by_module_name(qname, model): + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + for node in nodes: + if node.target in modules and qname == node.target: + return node.name + +def get_node_input_by_module_name(qname, model): + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + post_str = '_post_act_fake_quantizer' + input_str = '_input_act_fake_quantizer' + scale_name = None + for node in nodes: + if node.target in modules and qname == node.target: + print(f'{qname} input:', node.args[0].name) + if post_str in node.args[0].name: + scale_name = node.args[0].name + break + elif input_str in node.args[0].name: + node2 = node.args[0] + print(f'{node.args[0].name}.input:', node2.args[0].name) + if post_str in node2.args[0].name: + scale_name = node2.args[0].name + break + elif 'x' == node2.args[0].name: + return 'data' + break + if scale_name is not None: + return scale_name[:len(scale_name)-len(post_str)] + else: + return '' + +layer_names = [] +features_out_hook = {} +i = 0 +def hook(module, fea_in, fea_out): + global i + if i >= len(layer_names): + return None + name = layer_names[i] + i += 1 + global features_out_hook + features_out_hook[name] = fea_out.cpu().numpy() + return None + def gen_test_ref_data(cali_loader, model, args): return model.eval() global layer_names hook_handles = [] input_data = {} - exclude_module = ['mqbench', 'torch.fx', 'batchnorm', 'torch.nn.modules.module.Module'] - nodes = list(model.graph.nodes) + # exclude_module = ['fake_quantize', 'observer', 'torch.fx', 'batchnorm', 'torch.nn.modules.module.Module'] for name, child in model.named_modules(): - if not any([i in str(type(child)) for i in exclude_module]): - print("add hook on", str(type(child)), name) + # if not any([i in str(type(child)) for i in exclude_module]): + if name.endswith('_act_fake_quantizer'): # if '_dup' in name: # name = name[:-5] # layer_names.append(name.replace('.','_')) - node_name = get_node_name_by_module_name(name, model) - layer_names.append(node_name) + # output = get_node_name_by_module_name(name, model) + # input = get_node_input_by_module_name(name, model) + # print(f'name:{name}, output:{output}, input:{input}') + # if input != '': + layer_names.append(name) + print(f"add hook on {name} for {str(type(child))}") hd = child.register_forward_hook(hook=hook) hook_handles.append(hd) print('layer_names:', layer_names) @@ -486,6 +551,7 @@ def gen_test_ref_data(cali_loader, model, args): input_data['data'] = images.cpu().numpy() np.savez(os.path.join(args.output_path, 'input_data.npz'), **input_data) global features_out_hook + features_out_hook['data'] = images.cpu().numpy() np.savez(os.path.join(args.output_path, 'layer_outputs.npz'), **features_out_hook) for hd in hook_handles: hd.remove() diff --git a/application/yolov5_example/train.py b/application/yolov5_example/train.py index 03aa2509..73405e54 100644 --- a/application/yolov5_example/train.py +++ b/application/yolov5_example/train.py @@ -300,6 +300,7 @@ def train(hyp, opt, device, callbacks): # hyp is path/to/hyp.yaml or hyp dictio backend = BackendType.Sophgo_TPU if opt.quantize: prepare_custom_config_dict= { + # 'work_mode':'int4_and_int8_mix', # 'extra_qconfig_dict':{'w_fakequantize':'PACTFakeQuantize'} # 'concrete_args':{'augment':False, 'profile':False, 'visualize':False} } @@ -438,28 +439,28 @@ def train(hyp, opt, device, callbacks): # hyp is path/to/hyp.yaml or hyp dictio best_fitness = fi log_vals = list(mloss) + list(results) + lr callbacks.run('on_fit_epoch_end', log_vals, epoch, best_fitness, fi) - + # Save model - if (not nosave) or (final_epoch and not evolve): # if save - ckpt = { - 'epoch': epoch, - 'best_fitness': best_fitness, - 'model': deepcopy(de_parallel(model)).half(), - 'ema': deepcopy(ema.ema).half(), - 'updates': ema.updates, - 'optimizer': optimizer.state_dict(), - 'wandb_id': loggers.wandb.wandb_run.id if loggers.wandb else None, - 'opt': vars(opt), - 'date': datetime.now().isoformat()} + # if (not nosave) or (final_epoch and not evolve): # if save + # ckpt = { + # 'epoch': epoch, + # 'best_fitness': best_fitness, + # 'model': deepcopy(de_parallel(model)).half(), + # 'ema': deepcopy(ema.ema).half(), + # 'updates': ema.updates, + # 'optimizer': optimizer.state_dict(), + # 'wandb_id': loggers.wandb.wandb_run.id if loggers.wandb else None, + # 'opt': vars(opt), + # 'date': datetime.now().isoformat()} - # Save last, best and delete - torch.save(ckpt, last) - if best_fitness == fi: - torch.save(ckpt, best) - if opt.save_period > 0 and epoch % opt.save_period == 0: - torch.save(ckpt, w / f'epoch{epoch}.pt') - del ckpt - callbacks.run('on_model_save', last, epoch, final_epoch, best_fitness, fi) + # # Save last, best and delete + # torch.save(ckpt, last) #wangxuechuan 23.04.12, Can't pickle local object 'add_module_to_qconfig_obs_ctr..get_factory_kwargs_based_on_module_device' + # if best_fitness == fi: + # torch.save(ckpt, best) + # if opt.save_period > 0 and epoch % opt.save_period == 0: + # torch.save(ckpt, w / f'epoch{epoch}.pt') + # del ckpt + # callbacks.run('on_model_save', last, epoch, final_epoch, best_fitness, fi) # EarlyStopping if RANK != -1: # if DDP training diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index cfde0d4f..c9ccb180 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -1,9 +1,9 @@ import json import os.path as osp - +import os import torch from torch.fx import GraphModule - +import onnx import mqbench.custom_symbolic_opset # noqa: F401 import mqbench.fusion_method # noqa: F401 from mqbench.prepare_by_platform import BackendType @@ -19,6 +19,7 @@ remove_fakequantize_and_collect_params, replace_fakequantize_and_collect_params_openvino, remove_fakequantize_and_collect_params_tengine, + remove_fakequantize_and_collect_params_sophgo, ONNXQLinearPass, ONNXQNNPass ) @@ -63,7 +64,7 @@ def convert_merge_bn(model: GraphModule, **kwargs): def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_path, **kwargs): pt_file_name = onnx_model_path.split('.') pt_file_name[-1] = 'pt' - torch.save(model, '.'.join(pt_file_name)) + #torch.save(model, '.'.join(pt_file_name)) logger.info("Export to onnx, onnx_model_path:{}".format(onnx_model_path)) model = model.cpu() output_names = kwargs.get('output_names', []) @@ -97,6 +98,10 @@ def convert_onnx(model: GraphModule, input_shape_dict, dummy_input, onnx_model_p do_constant_folding=True, custom_opsets={'' : opset_version}, enable_onnx_checker=False) + model_onnx = onnx.load(onnx_model_path) + model_onnx = onnx.shape_inference.infer_shapes(model_onnx) + os.system(f"rm -f {onnx_model_path}") + onnx.save(model_onnx, onnx_model_path) @register_deploy_function(BackendType.Tensorrt) def convert_onnx_qlinear(model: GraphModule, onnx_model_path, model_name, **kwargs): @@ -125,22 +130,31 @@ def deploy_qparams_tensorrt(model: GraphModule, onnx_model_path, model_name, **k @register_deploy_function(BackendType.Sophgo_TPU) def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, **kwargs): logger.info("Extract qparams for sophgo_tpu.") - remove_fakequantize_and_collect_params(onnx_model_path, model_name, backend='sophgo_tpu') + remove_fakequantize_and_collect_params_sophgo(onnx_model_path, model_name) output_path = osp.dirname(onnx_model_path) context_filename = osp.join(output_path, '{}_clip_ranges.json'.format(model_name)) file_h = open(context_filename, "r") blob_range = json.loads(file_h.read())["sophgo_tpu"] file_h.close() cali_table = osp.join(output_path, '{}_cali_table_from_mqbench_sophgo_tpu'.format(model_name)) + work_mode = kwargs.get('work_mode', 'QAT_all_int8') + if work_mode not in ['QAT_all_int8', 'int4_and_int8_mix', 'int4_and_int8_mix_no_fc']: + print('QAT_all_int8 not in [QAT_all_int8, int4_and_int8_mix, int4_and_int8_mix_no_fc],set to QAT_all_int8') + work_mode = 'QAT_all_int8' with open(cali_table, 'w') as f: - f.write("# work_mode:QAT_all_int8 #Automatically generated, do not modify, work_mode choice:[QAT_all_int8]\n") + f.write(f"# work_mode:{work_mode} #Automatically generated, do not modify, work_mode choice:[QAT_all_int8, int4_and_int8_mix, int4_and_int8_mix_no_fc]\n") f.write("# op_name threshold min max\n") - ori_layer_names = '' weight_scale = [] + int4_th = [] for name,value in blob_range.items(): if 'threshold' in value: - f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) - ori_layer_names += '{},'.format(value['ori_name']) + tmpstr = "{} {:.7f} {:.7f} {:.7f}\n".format(name[:-2], value['threshold'], value['min'], value['max']) + if name.endswith('_4'): + int4_th.append(tmpstr) + elif name.endswith('_8'): + f.write(tmpstr) + else: + f.write("{} {:.7f} {:.7f} {:.7f}\n".format(name, value['threshold'], value['min'], value['max'])) else: tmpstr = "{} {} {} {} {}\n".format(name, len(value['step']), ' '.join([str(i) for i in value['step']]), len(value['zero_point']), ' '.join([str(i) for i in value['zero_point']])) @@ -148,10 +162,12 @@ def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, * weight_scale.append(tmpstr) else: f.write(tmpstr) + f.write('#int4_th\n') + for i in int4_th: + f.write(i) f.write('#weight_scale\n') for i in weight_scale: f.write(i) - f.write("#{}\n".format(ori_layer_names[0:-1])) @register_deploy_function(BackendType.Vitis) diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index aa3d0789..9147493d 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -64,7 +64,7 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): def prepare(self, model: GraphModule, qconfig): model = _fuse_fx(model, self.extra_fuse_dict) model = self._weight_quant(model, qconfig) - model = self._insert_fake_quantize_for_act_quant(model, qconfig['']) + model = self._insert_fake_quantize_for_act_quant(model, qconfig) return model def _insert_fake_quantize_for_act_quant( @@ -77,7 +77,7 @@ def _insert_fake_quantize_for_act_quant( quantizer_prefix = "_post_act_fake_quantizer" node_to_quantize_output = self._find_act_quants(model) node_to_quantize_output = OrderedDict.fromkeys(node_to_quantize_output).keys() - + qconfig = qconfig[''] for node in node_to_quantize_output: fake_quantizer = qconfig.activation() quantizer_name = node.name + quantizer_prefix @@ -122,7 +122,8 @@ def _fix_succ_recursivly(self, args, target_node, inserted_node): def _weight_quant(self, model: GraphModule, qconfig): logger.info("Replace module to qat module.") - flattened_qconfig_dict = get_flattened_qconfig_dict(qconfig)#torch??? + flattened_qconfig_dict = get_flattened_qconfig_dict(qconfig)#torch锟接匡拷 + print('flattened_qconfig_dict:', flattened_qconfig_dict) propagate_qconfig_(model, flattened_qconfig_dict) self._qat_swap_modules(model, self.additional_qat_module_mapping) return model @@ -233,7 +234,7 @@ def _find_act_quants(self, model: GraphModule) -> List: if (node.op == "call_module" and isinstance(modules[node.target], self.module_type_to_quant_input)) or \ ((node.op == 'call_function' or node.op == 'call_method') and node.target in self.function_type_to_quant_input) or node.name in self.additional_node_name: - input_node_list = self._flatten_args(node.all_input_nodes) #wxc + input_node_list = self._flatten_args(node.all_input_nodes) # Means this is not Tensor + Tensor. if not all([isinstance(_node, torch.fx.node.Node) for _node in input_node_list]): continue diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index 4c3ec80b..dcbe1bc2 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -9,7 +9,16 @@ from mqbench.custom_quantizer import ModelQuantizer import torch.nn as nn import mqbench.nn.qat as qnnqat +from collections import OrderedDict +from mqbench.utils.logger import logger +from typing import ( + List, Dict, Any, Callable +) + +from torch.quantization.fx.qconfig_utils import ( + get_flattened_qconfig_dict +) @register_model_quantizer(BackendType.Sophgo_TPU) class SophgoTpuQuantizer(ModelQuantizer): @@ -35,15 +44,23 @@ def __init__(self, extra_quantizer_dict, extra_fuse_dict): qnni.ConvTransposeReLU2d:qnniqat.ConvTransposeReLU2d_sophgo, qnni.ConvTransposeBn2d:qnniqat.ConvTransposeBn2d_sophgo, } + self.exclude_module_name.append(nn.modules.dropout.Dropout) @property def module_type_to_quant_input(self) -> tuple: - return super().module_type_to_quant_input + self._layers_need_scale_form_input_fake_quantizer + return ( + torch.nn.Hardswish, + torch.nn.Sigmoid, + torch.nn.SiLU, + ) + super().module_type_to_quant_input + self._layers_need_scale_form_input_fake_quantizer @property def function_type_to_quant_input(self) -> tuple: return super().function_type_to_quant_input + [ - torch.cat + torch.cat, + torch.nn.functional.hardswish, + torch.nn.functional.sigmoid, + torch.nn.functional.silu ] @property @@ -76,6 +93,158 @@ def _layers_need_scale_form_input_fake_quantizer(self): qnniqat.ConvTransposeBn2d_sophgo, ) + @property + def _layers_need_check_is_dw(self): + return ( + qnniqat.ConvBnReLU2d_sophgo, + qnniqat.ConvBn2d_sophgo, + qnniqat.ConvReLU2d_sophgo, + qnnqat.Conv2d_sophgo, + ) + + def _insert_fake_quantizer(self, model, graph, modules, flattened_qconfig_dict, node, int84_layers): + if len(int84_layers) > 0: + layer = int84_layers[0] #澶氫釜int4鎴杋nt8鑺傜偣閫夊叾涓1涓氨鑳芥纭喅瀹氳妭鐐圭被鍨 + qconfig1 = flattened_qconfig_dict.get(layer.target, None) #棣栧厛鏍规嵁灞傚悕鍘诲彇锛屼紭鍏堢骇鏈楂 + if qconfig1 is None and layer.target in modules: + print(f'layer.target:{layer.target}, type:',type(modules[layer.target])) + qconfig1 = flattened_qconfig_dict.get(type(modules[layer.target]), None) #鍏舵鏍规嵁type鍘诲彇 + if isinstance(modules[layer.target], self._layers_need_check_is_dw): + if modules[layer.target].groups > 1: + qconfig1 = None #娣卞害鍗风Н浣跨敤int8璁＄畻 + if qconfig1 is None: + qconfig1 = flattened_qconfig_dict.get('', None) #鏈鍚庢壘鍏ㄥ眬qconfig锛屼紭鍏堢骇鏈浣 + fake_quantizer = qconfig1.activation() + quantizer_name = layer.name + self.quantizer_prefix + if hasattr(model, quantizer_name): + quantizer_name = layer.name +'_n2_br'+ self.quantizer_prefix + setattr(model, quantizer_name, fake_quantizer) + logger.info("Insert act quant {}".format(quantizer_name)) + with graph.inserting_after(node): + inserted_node = graph.create_node("call_module", quantizer_name, (node,), {}) + for int84_layer in int84_layers: + int84_layer.args = self._fix_succ_recursivly(int84_layer.args, node, inserted_node) + + def _insert_fake_quantize_for_act_quant( + self, + model: GraphModule, + qconfig: Any): + graph = model.graph + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + + self.quantizer_prefix = "_input_act_fake_quantizer" + node_to_quantize_output = self._find_act_quants(model) + node_to_quantize_output = OrderedDict.fromkeys(node_to_quantize_output).keys() + flattened_qconfig_dict = get_flattened_qconfig_dict(qconfig) + print('node_to_quantize_output:', node_to_quantize_output, 'flattened_qconfig_dict:', flattened_qconfig_dict) + int4_and_int8_mix = False + for m in flattened_qconfig_dict: + if not isinstance(flattened_qconfig_dict[m].activation(), torch.nn.Identity) and flattened_qconfig_dict[m].activation().bitwidth == 4: + int4_and_int8_mix = True + break + print('int4_and_int8_mix:', int4_and_int8_mix) + if int4_and_int8_mix: + def all_next_layers_is_trivial(node): + if len(node.users): + return True + for user in node.users: + if not all_next_layers_is_trivial(user): + return False + if node.op != 'call_method' or node.target not in ['view', 'permute', 'contiguous']: + return False + return True + + def find_next_int4_and_int8_layers(node, int8_layers, int4_layers): + for user in node.users: #鑻ュ悗缁ф湁1涓眰鎴栧涓笉鍚岀被鍨嬬殑灞傦紝鍒欐彃鍏ュ涓猧nput閲忓寲鑺傜偣 todo:澶氫釜鑺傜偣锛岄儴鍒嗙浉鍚岋紝閮ㄥ垎涓嶅悓 + if user.target in modules and type(modules[user.target]) in self.exclude_module_name: + print(f'user:{user.name} is excluded') + user.replace_all_uses_with(node) + graph.erase_node(user) + del modules[user.target] + find_next_int4_and_int8_layers(user, int8_layers, int4_layers) + continue #dropout绛夊眰鍓嶄笉瑕佹彃鍏ヤ吉閲忓寲鑺傜偣 + if user.op == "call_module" and isinstance(modules[user.target], self._layers_need_scale_form_input_fake_quantizer): + if isinstance(modules[user.target], self._layers_need_check_is_dw): + if modules[user.target].groups > 1: + int8_layers.append(user) + continue + int4_layers.append(user) + else: + if not all_next_layers_is_trivial(user): + int8_layers.append(user) + + for node in node_to_quantize_output: + int8_layers, int4_layers = [],[] #鎵惧埌node鍚庣殑澶氫釜int4鍚庣户鑺傜偣鍜屽涓猧nt8鍚庣户鑺傜偣锛岀劧鍚庤繖澶氫釜int8鎴杋nt4鍚庣户鑺傜偣鍏变韩1涓緭鍏ラ噺鍖栬妭鐐 + find_next_int4_and_int8_layers(node, int8_layers, int4_layers) + print(f'node:{node}, int4_layers:', int4_layers, 'int8_layers:', int8_layers) + self._insert_fake_quantizer(model, graph, modules, flattened_qconfig_dict, node, int4_layers) + self._insert_fake_quantizer(model, graph, modules, flattened_qconfig_dict, node, int8_layers) + model.recompile() + model.graph.lint() + graph = model.graph + nodes = list(model.graph.nodes) + modules = dict(model.named_modules()) + + for node in node_to_quantize_output: + if node.op == 'placeholder' and int4_and_int8_mix: + continue + qconfig2 = flattened_qconfig_dict.get(node.target, None) #棣栧厛鏍规嵁灞傚悕鍘诲彇锛屼紭鍏堢骇鏈楂 + if qconfig2 is None and node.target in modules: + qconfig2 = flattened_qconfig_dict.get(type(modules[node.target]), None) #鍏舵鏍规嵁type鍘诲彇 + if isinstance(modules[node.target], self._layers_need_check_is_dw): + if modules[node.target].groups > 1: + qconfig2 = None #娣卞害鍗风Н浣跨敤int8璁＄畻 + if qconfig2 is None: + qconfig2 = flattened_qconfig_dict.get('', None) #鏈鍚庢壘鍏ㄥ眬qconfig锛屼紭鍏堢骇鏈浣 + if node.target in modules and type(modules[node.target]) == torch.nn.ReLU6 and node.args[0].target in modules: + qconfig2 = flattened_qconfig_dict.get(type(modules[node.args[0].target]), None) + if isinstance(modules[node.args[0].target], self._layers_need_check_is_dw): + if modules[node.args[0].target].groups > 1: + qconfig2 = flattened_qconfig_dict.get('', None) ##娣卞害鍗风Н浣跨敤int8璁＄畻 + node_fake_quantizer = qconfig2.activation() + # node_fake_quantizer.enable_only_observer() + quantizer_name2 = node.name + "_post_act_fake_quantizer" + setattr(model, quantizer_name2, node_fake_quantizer) + with graph.inserting_after(node): + inserted_node = graph.create_node("call_module", quantizer_name2, (node,), {}) + for _node in nodes: + _node.args = self._fix_succ_recursivly(_node.args, node, inserted_node) + + if int4_and_int8_mix: + model.recompile() + model.graph.lint() + graph = model.graph + modules = dict(model.named_modules()) + nodes = list(model.graph.nodes) + for node in nodes: + if "_post_act_fake_quantizer" in node.name: + #post閲忓寲鑺傜偣鐨勪笅涓涓槸input閲忓寲鑺傜偣锛屼笖input閲忓寲鑺傜偣鐨勫悗闈㈠彧鏈1涓妭鐐癸紝涓旇鑺傜偣鍚庡啀鏃犲悗缁ц妭鐐癸紝姝ゆ椂鍒犻櫎杩欎釜input閲忓寲鑺傜偣锛岀敤浜庡垹闄ょ綉缁滆緭鍑虹殑鏈鍚1涓浣欑殑閲忓寲鑺傜偣 + for user in list(node.users.keys()): + if "_input_act_fake_quantizer" in user.name: + users = list(user.users.keys()) + if len(users) == 1 and len(users[0].users) == 0: + user.replace_all_uses_with(node) + graph.erase_node(user) + del modules[user.target] + if modules[node.target].bitwidth == 8: #2涓浉閭荤殑閲忓寲鑺傜偣閮戒负8bit閲忓寲锛屽垯鍒犻櫎鍐椾綑鐨勫悗1涓妭鐐 + for user in list(node.users.keys()): + if "_input_act_fake_quantizer" in user.name: + if modules[user.target].bitwidth == 8: + user.replace_all_uses_with(node) + graph.erase_node(user) + del modules[user.target] + elif modules[node.target].bitwidth == 4: #2涓浉閭荤殑閲忓寲鑺傜偣閮戒负4bit閲忓寲锛屽垯鍒犻櫎鍐椾綑鐨勫悗1涓妭鐐 + for user in list(node.users.keys()): + if "_input_act_fake_quantizer" in user.name: + if modules[user.target].bitwidth == 4: + user.replace_all_uses_with(node) + graph.erase_node(user) + del modules[user.target] + model.recompile() + model.graph.lint() + return model + def prepare(self, model: GraphModule, qconfig): model = super().prepare(model, qconfig) model = self._set_fake_quantizer_to_next_weight_layer(model) @@ -85,6 +254,7 @@ def _find_act_quants(self, model: GraphModule) -> list: nodes = list(model.graph.nodes) modules = dict(model.named_modules()) node_need_to_quantize_output = super()._find_act_quants(model) + self.only_enable_ob = [] for node in nodes: if (node.op == "call_module" and node.target in self.exclude_module_name) or \ ((node.op == 'call_function' or node.op == 'call_method') and @@ -95,21 +265,35 @@ def _find_act_quants(self, model: GraphModule) -> list: ((node.op == 'call_function' or node.op == 'call_method') and node.target in self.function_type_to_quant_input): for next_node in node.users: - if not ((next_node.op == 'call_function' and next_node.target in self._passed_func_type) or + if ((next_node.op == 'call_function' and next_node.target in self._passed_func_type) or (next_node.op == 'call_module' and isinstance(modules[next_node.target], self._passed_module_type))): - node_need_to_quantize_output.append(node) + if next_node not in node_need_to_quantize_output: + node_need_to_quantize_output.append(next_node) + self.only_enable_ob.append(next_node.name) else: - node_need_to_quantize_output.append(next_node) + if node not in node_need_to_quantize_output: + node_need_to_quantize_output.append(node) + self.only_enable_ob.append(node.name) + for node in nodes: + if node.target in modules and type(modules[node.target]) in self.exclude_module_name: + print(f'{type(modules[node.target])} is excluded') + node_need_to_quantize_output.remove(node) return node_need_to_quantize_output - + def _set_fake_quantizer_to_next_weight_layer(self, model: GraphModule): nodes = list(model.graph.nodes) modules = dict(model.named_modules()) for node in nodes: - if node.target in modules and "_post_act_fake_quantizer" in node.target: + if node.target in modules and (self.quantizer_prefix in node.target or "_post_act_fake_quantizer" in node.target): fake_quantizer = getattr(model, node.target) for user in node.users: if (user.op == "call_module" and isinstance(modules[user.target], self._layers_need_scale_form_input_fake_quantizer)): setattr(modules[user.target], "input_fake_quantizer", fake_quantizer) print('wlog:', user.target,'\'type is:', type(modules[user.target]), "add input_fake_quantizer") - return model \ No newline at end of file + if user.target in modules and type(modules[user.target]) in self.exclude_module_name: + for user2 in user.users: + if (user2.op == "call_module" and isinstance(modules[user2.target], self._layers_need_scale_form_input_fake_quantizer)): + setattr(modules[user2.target], "input_fake_quantizer", fake_quantizer) + print('wlog:', user2.target,'\'type is:', type(modules[user2.target]), "add input_fake_quantizer") + + return model diff --git a/mqbench/deploy/__init__.py b/mqbench/deploy/__init__.py index 9aed37ea..65e8c15e 100644 --- a/mqbench/deploy/__init__.py +++ b/mqbench/deploy/__init__.py @@ -4,3 +4,4 @@ from .deploy_onnx_qnn import ONNXQNNPass from .deploy_openvino import replace_fakequantize_and_collect_params_openvino from .deploy_tengine import remove_fakequantize_and_collect_params_tengine +from .deploy_sophgo import remove_fakequantize_and_collect_params_sophgo \ No newline at end of file diff --git a/mqbench/deploy/deploy_sophgo.py b/mqbench/deploy/deploy_sophgo.py new file mode 100644 index 00000000..80288e7c --- /dev/null +++ b/mqbench/deploy/deploy_sophgo.py @@ -0,0 +1,349 @@ +import json +import os +import copy + + +import onnx +import numpy as np +from onnx import numpy_helper + +from mqbench.utils.logger import logger +from mqbench.deploy.common import ( + update_inp2node_out2node, + prepare_initializer, + prepare_data, + OnnxPreprocess, + get_constant_inputs, + parse_attrs +) + + +PERCHANNEL_FAKEQUANTIZER = ['FakeQuantizeLearnablePerchannelAffine', + 'FixedPerChannelAffine', + 'FakeQuantizeDSQPerchannel'] +PERTENSOR_FAKEQUANTIZER = ['LearnablePerTensorAffine', + 'FixedPerTensorAffine', + 'FakeQuantizeDSQPertensor', + 'FakeQuantizeTqtAffine'] +ALL_FAKEQUANTIZER = PERCHANNEL_FAKEQUANTIZER + PERTENSOR_FAKEQUANTIZER + + +class LinearQuantizer_process(object): + # some method like dorefa need pre-compute weights + def weight_preprocess(self, target_tensor, out2node, inp2node, named_initializer): + def find_weight(tensor): + if tensor not in named_initializer: + _node = out2node[tensor] + for inp in _node.input: + return find_weight(inp) + return tensor + weight = find_weight(target_tensor) + + # TODO need more general method, like onnxruntime infer + data = numpy_helper.to_array(named_initializer[weight]) + data = np.tanh(data) + data = data / (np.max(np.abs(data)) + 1e-5) + data = numpy_helper.from_array(data) + named_initializer[weight].raw_data = data.raw_data + + redundant_nodes = [] + + def find_redundant_nodes(tensor): + if tensor == target_tensor: + return + nodes = inp2node[tensor] + for node, idx in nodes: + if node not in redundant_nodes: + redundant_nodes.append(node) + redundant_nodes.extend(get_constant_inputs(node, out2node)) + find_redundant_nodes(node.output[0]) + find_redundant_nodes(weight) + return weight, redundant_nodes + + def deal_with_weight_fakequant(self, node, out2node, inp2node, named_initializer): + next_nodes = inp2node[node.output[0]] + assert len(next_nodes) == 1 + next_node, idx = next_nodes[0] + assert next_node.op_type in ['Conv', 'Gemm', 'ConvTranspose'] + redundant_nodes = [] + if node.input[0] not in named_initializer: + node.input[0], redundant_nodes = \ + self.weight_preprocess(node.input[0], out2node, inp2node, named_initializer) + next_node.input[idx] = node.input[0] + return redundant_nodes + + def deal_with_activation_fakequant(self, node, inp2node): + next_nodes = inp2node[node.output[0]] + for next_node, idx in next_nodes: + # if next_node.op_type == 'Add' and next_node.output[0] in inp2node: #灏唎bserver qdq鐨勮緭鍏ュ啓鍒颁笅涓涓猯ayer鐨勮緭鍏 + # nextnode, i = inp2node[next_node.output[0]][0] + # nextnode.input[i] = node.input[0] + # else: + next_node.input[idx] = node.input[0] + return + + def parse_qparams(self, node, name2data): + tensor_name, scale, zero_point = node.input[:3] + scale, zero_point = name2data[scale], name2data[zero_point] + if len(node.input) > 3: + qmin, qmax = node.input[-2:] + qmin, qmax = name2data[qmin], name2data[qmax] + elif len(node.attribute) > 0: + qparams = parse_attrs(node.attribute) + qmin = qparams['quant_min'] + qmax = qparams['quant_max'] + else: + logger.info(f'qmin and qmax are not found for <{node.name}>!') + return tensor_name, scale, zero_point, qmin, qmax + + def clip_weight(self, node, name2data, inp2node, named_initializer): + tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) + data = name2data[tensor_name] + clip_range_min = ((qmin - zero_point) * scale).astype(data.dtype) + clip_range_max = ((qmax - zero_point) * scale).astype(data.dtype) + if len(scale.shape) > 0 and scale.shape[0] > 1: + new_data = [] + transposed = False + next_node = inp2node[node.output[0]] + if len(next_node) == 1 and next_node[0][0].op_type == 'ConvTranspose': + transposed = True + data = data.transpose(1, 0, 2, 3) + for c in range(data.shape[0]): + new_data.append(np.clip(data[c], clip_range_min[c], clip_range_max[c])) + new_data = np.array(new_data) + if transposed: + new_data = new_data.transpose(1, 0, 2, 3) + logger.info(f'Clip weights <{tensor_name}> to per-channel ranges.') + else: + new_data = np.clip(data, clip_range_min, clip_range_max) + logger.info(f'Clip weights <{tensor_name}> to range [{clip_range_min}, {clip_range_max}].') + new_data = numpy_helper.from_array(new_data) + named_initializer[tensor_name].raw_data = new_data.raw_data + + def get_correct_sophgo_tpu_input_tensor_name(self, node, out2node): #鍜宼pu-mlir鐨勫懡鍚嶉鏍间竴鑷 + input_0 = node.input[0] + op_type = out2node[input_0].op_type + tensor_name = '{}_{}'.format(input_0, op_type if input_0 in out2node else '') + if tensor_name[-1] == '_': + tensor_name = tensor_name[:-1] + tensor_name += '_8' + if op_type in ['Conv', "Gemm"]: + tensor_name += '_4' + return tensor_name + + def post_process_clip_ranges(self, clip_ranges, graph, inp2node, out2node): + def find_the_closest_clip_range(node): + tensor_name = self.get_correct_sophgo_tpu_input_tensor_name(node, out2node) + if tensor_name in clip_ranges: + return tensor_name + elif node.op_type in ['Flatten', 'Resize'] and node.output[0] in inp2node: + return find_the_closest_clip_range(inp2node[node.output[0]][0][0]) + else: + return None + + for node in graph.node: + if node.op_type in ['Flatten', 'Resize']: + tensor_name = find_the_closest_clip_range(node) + if tensor_name: + new_name = self.get_correct_sophgo_tpu_input_tensor_name(node, out2node) + clip_ranges[new_name] = copy.deepcopy(clip_ranges[tensor_name]) + logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') + return clip_ranges + + def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): + op_type_inAndOutShouldSameClipRange = ['Flatten', 'Resize', 'Reshape', 'Transpose'] + for node in graph.node: + tensor_name = f'{node.output[0]}_{node.op_type}' + tensor_name += '_4' if node.op_type in ['Conv', "Gemm"] else '_8' + if tensor_name not in clip_ranges: + pre_op = node + finded = False + while pre_op.op_type in op_type_inAndOutShouldSameClipRange: + tensor_name2 = self.get_correct_sophgo_tpu_input_tensor_name(pre_op, out2node) + if tensor_name2 in clip_ranges: + finded = True + clip_ranges[tensor_name] = clip_ranges[tensor_name2] + print(f'pre_op finded, transfer {tensor_name2} to {tensor_name}') + break + if pre_op.input[0] in out2node: + pre_op = out2node[pre_op.input[0]] + else: + print(f'{pre_op.name}\'s pre_node not exist') + break + if not finded: + if node.output[0] in inp2node: + next_op = inp2node[node.output[0]][0][0] + while next_op.op_type in op_type_inAndOutShouldSameClipRange: + tensor_name2 = f'{next_op.output[0]}_{next_op.op_type}' + tensor_name2 += '_4' if next_op.op_type in ['Conv', "Gemm"] else '_8' + if tensor_name2 in clip_ranges: + finded = True + clip_ranges[tensor_name] = clip_ranges[tensor_name2] + print(f'next_op finded, transfer {tensor_name2} to {tensor_name}') + break + if next_op.output[0] in inp2node: + next_op = inp2node[next_op.output[0]][0][0] + else: + print(f'{next_op.name}\'s next_op not exist') + break + else: + print(f'{node.name}\'s next_op not exist') + # if not finded: + # print(f'Waring:{node.name}\'s clip_ranges not exist, maybe have some error') + return clip_ranges + + def isQdqAdd(self, tensor_name, out2node): + pre_name = None + for i in out2node[tensor_name].input: + if out2node[i].op_type == 'LearnablePerTensorAffine': + pre_name = out2node[i].input[0] + if pre_name in out2node: + pre_name = '{}_{}'.format(pre_name, out2node[pre_name].op_type) + if out2node[i].op_type not in ['LearnablePerTensorAffine', 'Sub']: + return False, pre_name + return True, pre_name + + def remove_Qdq_add_sub(self, graph, inp2node, out2node): + nodes_to_be_removed = [] + for idx, node in enumerate(graph.node): + if node.op_type == 'Add': + isQdqAdd, _ = self.isQdqAdd(node.output[0], out2node) + if isQdqAdd: + for i in node.input: + if i.op_type == 'Sub': + nodes_to_be_removed.append(out2node[i]) + else: + fake_quant_output = i + next_nodes = inp2node[node.output[0]] + for next_node, idx in next_nodes: + next_node.input[idx] = fake_quant_output + nodes_to_be_removed.append(node) + nodes_to_be_removed.extend(get_constant_inputs(node, out2node)) + for node in nodes_to_be_removed: + graph.node.remove(node) + return + + def remove_fakequantize_and_collect_params(self, onnx_path, model_name): + model = onnx.load(onnx_path) + graph = model.graph + out2node, inp2node = update_inp2node_out2node(graph) + name2data = prepare_data(graph) + named_initializer = prepare_initializer(graph) + + preprocess = OnnxPreprocess() + preprocess.remove_fake_pad_op(graph, name2data, inp2node, out2node) + out2node, inp2node = update_inp2node_out2node(graph) + # self.remove_Qdq_add_sub(graph, inp2node, out2node) + # out2node, inp2node = update_inp2node_out2node(graph) + + clip_ranges = {} + nodes_to_be_removed = [] + output_path = os.path.dirname(onnx_path) + file_name = os.path.join(output_path, 'layer_outputs.npz') + layer_out_tensor = None + layer_out_tensor2 = {} + if os.path.exists(file_name): + layer_out_tensor = np.load(file_name) + for node in graph.node: + print(f'process node:{node.name}, type:{node.op_type}') + if node.op_type in ALL_FAKEQUANTIZER: + nodes_to_be_removed.append(node) + nodes_to_be_removed.extend(get_constant_inputs(node, out2node)) + if node.output[0] not in inp2node: + assert node.output[0] in [l.name for l in graph.output] + inp2node[node.output[0]] = [] + next_nodes = inp2node[node.output[0]] + if node.op_type in PERCHANNEL_FAKEQUANTIZER: + # fake quantize for weights, suppose per-channel quantize only for weight + redundant_nodes = self.deal_with_weight_fakequant(node, out2node, inp2node, named_initializer) + nodes_to_be_removed.extend(redundant_nodes) + self.clip_weight(node, name2data, inp2node, named_initializer) + tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) + #鍗风Н鏉冮噸per-channel閲忓寲鍙傛暟锛宐ias鐨刾er-chan閲忓寲鍙傛暟娌℃湁鍘昏皟浼 + if len(next_nodes) == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']:#褰撳墠浼噺鍖栬妭鐐瑰彧鏈1涓悗缁э紝涓旂1涓悗缁ц妭鐐逛负conv绫诲瀷 + next_node_output = next_nodes[0][0].output[0] + if inp2node[next_node_output][0][0].op_type == 'Relu':##浼噺鍖栬妭鐐圭殑绗1涓悗缁onv鑺傜偣鐨勭1涓悗缁ц妭鐐逛负Relu(fake->conv->relu) + #鑻ユ槸fake->conv->relu,鍥犱负relu浼氳瀺鍚堝埌鍓嶉潰conv锛屾晠鐢╮elu鐨勮緭鍑簍ensor鍚+Relu浣滀负閲忓寲鍙傛暟淇濆瓨tensor鍚 + tensor_name = '{}_{}'.format(inp2node[next_node_output][0][0].output[0], 'Relu') + else: + #鑻ユ槸fake->conv->not_relu_type,鐩存帴鐢╟onv鐨勮緭鍑簍ensor鍚+conv浣滀负閲忓寲鍙傛暟淇濆瓨tensor鍚 + tensor_name = '{}_{}'.format(next_node_output, next_nodes[0][0].op_type) + tensor_name += '_{}'.format('weight' if next_nodes[0][1] == 1 else 'bias' ) + clip_ranges[tensor_name] = {'step': [float(x) for x in scale], + 'zero_point': [int(x) for x in zero_point]} + elif node.op_type in PERTENSOR_FAKEQUANTIZER: + if len(next_nodes) == 1 and next_nodes[0][1] == 1 and next_nodes[0][0].op_type in ['Gemm', 'Conv']: + # fake quantize for weights + redundant_nodes = self.deal_with_weight_fakequant(node, out2node, inp2node, named_initializer) + tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) + nodes_to_be_removed.extend(redundant_nodes) + self.clip_weight(node, name2data, inp2node, named_initializer) + assert next_nodes[0][0].op_type == 'Gemm' + tensor_name_new = '{}_{}_weight'.format(next_nodes[0][0].output[0], next_nodes[0][0].op_type) + clip_ranges[tensor_name_new] = {'step': [float(x) for x in scale], + 'zero_point': [int(x) for x in zero_point]} + else: + # fake quantize for activations + self.deal_with_activation_fakequant(node, inp2node) + tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) + bits = 4 if qmax == 7 else 8 + for out in graph.output: + output_name = node.output[0] + # if inp2node[node.output[0]][0][0].op_type == 'Add': + # output_name = inp2node[node.output[0]][0][0].output[0] + if out.name == output_name: + out.name = tensor_name + + scale_name = node.input[1] + post_str = '_post_act_fake_quantizer.scale' + tensor_name_new = tensor_name + if tensor_name in out2node: + tensor_name_new += '_{}'.format(out2node[tensor_name].op_type) + # if out2node[tensor_name].op_type == 'Add': + # isQdqAdd, pre_name = self.isQdqAdd(tensor_name, out2node) + # if isQdqAdd: + # tensor_name_new = pre_name+'_qdq' + + # if out2node[tensor_name].op_type in ['Conv', 'Gemm']: + # if bits == 8: + # tensor_name_new += '_8' + # else: + # if bits == 4 and out2node[out2node[tensor_name].input[0]].op_type not in ['Conv', 'Gemm']: + # tensor_name_new += '_4' + if layer_out_tensor is not None and scale_name.endswith(post_str): + torch_name = scale_name[:len(scale_name)-len(post_str)] + if torch_name in layer_out_tensor.files: + layer_out_tensor2[tensor_name_new] = layer_out_tensor[torch_name] + clip_ranges[tensor_name_new+f'_{bits}'] = {'threshold':float(scale * max(-qmin, qmax)), #瀵圭О閲忓寲鏃惰繖涓弬鏁扮敓鏁 + 'min': float(scale * (qmin - zero_point)), + 'max': float(scale * (qmax - zero_point))} + if layer_out_tensor is not None and len(layer_out_tensor2) > 0: + if 'data' in layer_out_tensor.files: + layer_out_tensor2['data'] = layer_out_tensor['data'] + os.system(f'rm -f {file_name}') + np.savez(file_name, **layer_out_tensor2) + for node in nodes_to_be_removed: + graph.node.remove(node) + # delete initializer + out2node, inp2node = update_inp2node_out2node(graph) + named_initializer = prepare_initializer(graph) + for name, initial_data in named_initializer.items(): + if name in (out2node.keys() | inp2node.keys()): + continue + graph.initializer.remove(initial_data) + + # clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) + clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node) + context = {'sophgo_tpu': clip_ranges} + context['w_qscheme'] = '' + context['a_qscheme'] = '' + context_filename = os.path.join(output_path, '{}_clip_ranges.json'.format(model_name)) + with open(context_filename, 'w') as f: + json.dump(context, f, indent=4) + onnx_filename = os.path.join(output_path, '{}_deploy_model.onnx'.format(model_name)) + model_onnx = onnx.shape_inference.infer_shapes(model) + os.system(f"rm -f {onnx_filename}") + onnx.save(model_onnx, onnx_filename) + logger.info("Finish deploy process.") + +remove_fakequantize_and_collect_params_sophgo = LinearQuantizer_process().remove_fakequantize_and_collect_params diff --git a/mqbench/fake_quantize/lsq.py b/mqbench/fake_quantize/lsq.py index c133b0d6..7e879b41 100644 --- a/mqbench/fake_quantize/lsq.py +++ b/mqbench/fake_quantize/lsq.py @@ -36,8 +36,9 @@ def extra_repr(self): self.zero_point if self.ch_axis == -1 else 'List') def forward(self, X): + x_ori = X # Learnable fake quantize have to zero_point.float() to make it learnable. - if self.observer_enabled[0] == 1: + if self.observer_enabled[0] == 1:# or self.only_enable_observer: self.activation_post_process(X.detach()) _scale, _zero_point = self.activation_post_process.calculate_qparams() _scale = _scale.to(self.scale.device) @@ -53,7 +54,10 @@ def forward(self, X): self.scale.data.abs_() self.scale.data.clamp_(min=self.eps.item()) - if self.fake_quant_enabled[0] == 1: + if self.fake_quant_enabled[0] == 1:# and (not self.only_enable_observer or self.run_fquant_time > 0): + # if self.run_fquant_time > 0: + # print('wxc1 run_fquant_time') + # self.run_fquant_time -= 1 if is_symmetric_quant(self.qscheme): self.zero_point.data.zero_() else: @@ -72,15 +76,25 @@ def forward(self, X): X = _fake_quantize_learnable_per_channel_affine_training( X, self.scale, self.zero_point, self.ch_axis, self.quant_min, self.quant_max, grad_factor) + x_ori = X else: if self.use_grad_scaling: grad_factor = 1.0 / (X.numel() * self.quant_max) ** 0.5 else: grad_factor = 1.0 + scale, zero_point = self.scale, self.zero_point + # if self.only_enable_observer: + # scale, zero_point = 1, 0 X = torch._fake_quantize_learnable_per_tensor_affine( - X, self.scale, self.zero_point, + x_ori, scale, zero_point, self.quant_min, self.quant_max, grad_factor) - return X + diff = x_ori - X + if self.only_enable_observer: + x_ori = X + diff.detach() + else: + x_ori = X + + return x_ori def _fake_quantize_learnable_per_channel_affine_training(x, scale, zero_point, ch_axis, quant_min, quant_max, grad_factor): diff --git a/mqbench/fake_quantize/quantize_base.py b/mqbench/fake_quantize/quantize_base.py index 8fe8fc18..5794fb4f 100644 --- a/mqbench/fake_quantize/quantize_base.py +++ b/mqbench/fake_quantize/quantize_base.py @@ -35,6 +35,12 @@ def __init__(self, observer=MovingAverageMinMaxObserver, **observer_kwargs): bitrange = torch.tensor(self.quant_max - self.quant_min + 1).double() self.bitwidth = int(torch.log2(bitrange).item()) self.is_symmetric_quant = is_symmetric_quant(self.qscheme) + self.only_enable_observer = False + self.run_fquant_time = 0 + + def enable_only_observer(self, enable = True): + self.only_enable_observer = enable + self.run_fquant_time = 1 @torch.jit.export def calculate_qparams(self): diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index a06216b7..dcb623e5 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -65,6 +65,7 @@ def convert_qnnqat_deconv2d(model, fused_node): fused_deconv.weight_fake_quant = fused_module.weight_fake_quant deconv_parent_name, deconv_name = _parent_name(fused_node.target) setattr(modules[deconv_parent_name], deconv_name, fused_deconv) + @register_convert_function(qnniqat.LinearReLU_sophgo) def linearert_qnniqat_linearrelu(model, fused_node): convert_qnniqat_linear(model, fused_node) @@ -198,6 +199,34 @@ def convert_nniqat_convbnrelu(model, fused_node): model.recompile() model.graph.lint() +@register_convert_function(qnniqat.ConvReLU2d_sophgo) +def convert_nniqat_convrelu(model, fused_node): + convert_qnnqat_conv2d(model, fused_node) + modules = dict(model.named_modules()) + fused_module = modules[fused_node.target] + # We need to Insert Relu after Merged conv. + conv_parent_name, conv_name = _parent_name(fused_node.target) + relu_name = 'relu' + # Maybe has another name, but we cannot know for now. + if not hasattr(modules[conv_parent_name], relu_name): + setattr(modules[conv_parent_name], relu_name, + torch.nn.ReLU(inplace=True).train(fused_module.training)) + # Update modules. + modules = dict(model.named_modules()) + graph = model.graph + nodes = list(model.graph.nodes) + with graph.inserting_after(fused_node): + relu_node_name = relu_name if conv_parent_name == "" else "{}.{}".format(conv_parent_name, relu_name) + assert relu_node_name in modules and isinstance(modules[relu_node_name], torch.nn.ReLU) + inserted_node = graph.create_node("call_module", relu_node_name, (fused_node,), {}) + for _node in nodes: + for i, _arg in enumerate(_node.args): + if _arg == fused_node: + _tmp = list(_node.args) + _tmp[i] = inserted_node + _node.args = tuple(_tmp) + model.recompile() + model.graph.lint() @register_convert_function(qnni.ConvTransposeFreezebn2d) @register_convert_function(qnni.ConvTransposeBn2d) diff --git a/mqbench/nn/intrinsic/qat/modules/__init__.py b/mqbench/nn/intrinsic/qat/modules/__init__.py index cd4dd80a..7b13120a 100644 --- a/mqbench/nn/intrinsic/qat/modules/__init__.py +++ b/mqbench/nn/intrinsic/qat/modules/__init__.py @@ -2,6 +2,7 @@ from .deconv_fused import ConvTransposeBnReLU2d, ConvTransposeBn2d, ConvTransposeReLU2d from .conv_fused import ConvBnReLU2d, ConvBn2d, ConvReLU2d from .freezebn import ConvFreezebn2d, ConvFreezebnReLU2d, ConvTransposeFreezebn2d, ConvTransposeFreezebnReLU2d + from .conv_fused_sophgo_tpu import ConvBnReLU2d_sophgo, ConvBn2d_sophgo, ConvReLU2d_sophgo from .linear_fused_sophgo_tpu import LinearBn1d_sophgo, LinearReLU_sophgo, Linear_sophgo from .deconv_fused_sophgo_tpu import ConvTransposeBnReLU2d_sophgo, ConvTransposeBn2d_sophgo, ConvTransposeReLU2d_sophgo diff --git a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py index 64466b38..7d20fcd6 100644 --- a/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/conv_fused_sophgo_tpu.py @@ -134,6 +134,44 @@ def bias_fake_quant_proc(self, bias, scale_w, in_scale): bias = bias*scale return bias + # def _forward(self, input): + # # print('xxx2') + # assert self.bn.running_var is not None + # running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + # scale_factor = self.bn.weight / running_std + # weight_shape = [1] * len(self.weight.shape) + # weight_shape[0] = -1 + # bias_shape = [1] * len(self.weight.shape) + # bias_shape[1] = -1 + # if torch.isnan(self.weight).any(): + # print('weight have nan') + # if self.input_fake_quantizer is not None and torch.isnan(self.input_fake_quantizer.scale).any(): + # print('input_fake_quantizer.scale have nan') + # if self.bias is not None and torch.isnan(self.bias).any(): + # print('weight have nan') + # scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + # #bias浼噺鍖 + # if self.weight_fake_quant.fake_quant_enabled[0] == 1: + # _, fused_bias = nn.utils.fuse_conv_bn_weights(self.weight, self.bias, + # self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) + # in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + # scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) + # diff_fused_bias = fused_bias - scale_fused_bias + # # using zero bias here since the bias for original conv + # # will be added later + # if self.bias is not None: + # zero_bias = torch.zeros_like(self.bias) + # else: + # zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + # conv = self._conv_forward(input, scaled_weight, zero_bias) + # conv_orig = conv / scale_factor.reshape(bias_shape) + # if self.bias is not None: + # conv_orig = conv_orig + self.bias.reshape(bias_shape) + # conv = self.bn(conv_orig) + # if self.weight_fake_quant.fake_quant_enabled[0] == 1: + # conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 + # return conv + def _forward(self, input): assert self.bn.running_var is not None running_std = torch.sqrt(self.bn.running_var + self.bn.eps) @@ -143,26 +181,23 @@ def _forward(self, input): bias_shape = [1] * len(self.weight.shape) bias_shape[1] = -1 scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) - #bias浼噺鍖 - if self.weight_fake_quant.fake_quant_enabled[0] == 1: - _, fused_bias = nn.utils.fuse_conv_bn_weights(self.weight, self.bias, - self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) - in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale - scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) - diff_fused_bias = fused_bias - scale_fused_bias # using zero bias here since the bias for original conv # will be added later if self.bias is not None: zero_bias = torch.zeros_like(self.bias) + conv_bias = self.bias else: zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) - conv = self._conv_forward(input, scaled_weight, zero_bias) - conv_orig = conv / scale_factor.reshape(bias_shape) - if self.bias is not None: - conv_orig = conv_orig + self.bias.reshape(bias_shape) + conv_bias = torch.zeros_like(zero_bias, device=scaled_weight.device) + if self.bn.affine: + full_bias = (conv_bias - self.bn.running_mean) / running_std * self.bn.weight + self.bn.bias + else: + full_bias = (conv_bias - self.bn.running_mean) / running_std + # quant_bias = self.bias_fake_quant(full_bias) + quant_bias = self.bias_fake_quant_proc(full_bias, self.weight_fake_quant.scale, self.input_fake_quantizer.scale) + conv_with_bias = self._conv_forward(input, scaled_weight, quant_bias) + conv_orig = (conv_with_bias - full_bias.reshape(bias_shape)) / scale_factor.reshape(bias_shape) + conv_bias.reshape(bias_shape) conv = self.bn(conv_orig) - if self.weight_fake_quant.fake_quant_enabled[0] == 1: - conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 return conv def extra_repr(self): diff --git a/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py index 717ce582..50d77a59 100644 --- a/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/deconv_fused_sophgo_tpu.py @@ -144,6 +144,36 @@ def bias_fake_quant_proc(self, bias, scale_w, in_scale): bias = bias*scale return bias + # def _forward(self, input): + # assert self.bn.running_var is not None + # running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + # scale_factor = self.bn.weight / running_std + # weight_shape = [1] * len(self.weight.shape) + # weight_shape[1] = -1 + # bias_shape = [1] * len(self.weight.shape) + # bias_shape[1] = -1 + # scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + # if self.weight_fake_quant.fake_quant_enabled[0] == 1: + # _, fused_bias = fuse_deconv_bn_weights(self.weight, self.bias, + # self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) + # in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale + # scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) + # diff_fused_bias = fused_bias - scale_fused_bias + + # if self.bias is not None: + # zero_bias = torch.zeros_like(self.bias) + # else: + # zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + # conv = self._convtransposed_forward(input, scaled_weight, zero_bias) + # conv_orig = conv / scale_factor.reshape(bias_shape) + # if self.bias is not None: + # conv_orig = conv_orig + self.bias.reshape(bias_shape) + # conv = self.bn(conv_orig) + # if self.weight_fake_quant.fake_quant_enabled[0] == 1: + # conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 + # return conv + + def _forward(self, input): assert self.bn.running_var is not None running_std = torch.sqrt(self.bn.running_var + self.bn.eps) @@ -153,25 +183,24 @@ def _forward(self, input): bias_shape = [1] * len(self.weight.shape) bias_shape[1] = -1 scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) - if self.weight_fake_quant.fake_quant_enabled[0] == 1: - _, fused_bias = fuse_deconv_bn_weights(self.weight, self.bias, - self.bn.running_mean, self.bn.running_var, self.bn.eps, self.bn.weight, self.bn.bias) - in_scale = self.input_fake_quantizer.scale #浠庝笂涓涓猘ctivation_fake_quant鑺傜偣鑾峰彇scale - scale_fused_bias = self.bias_fake_quant_proc(fused_bias, self.weight_fake_quant.scale, in_scale) - diff_fused_bias = fused_bias - scale_fused_bias - + # using zero bias here since the bias for original conv + # will be added later if self.bias is not None: zero_bias = torch.zeros_like(self.bias) + conv_bias = self.bias else: zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) - conv = self._convtransposed_forward(input, scaled_weight, zero_bias) - conv_orig = conv / scale_factor.reshape(bias_shape) - if self.bias is not None: - conv_orig = conv_orig + self.bias.reshape(bias_shape) - conv = self.bn(conv_orig) - if self.weight_fake_quant.fake_quant_enabled[0] == 1: - conv -= diff_fused_bias.reshape(bias_shape) #杩欓噷浠庢帹瀵肩湅搴旇鏄噺 - return conv + conv_bias = torch.zeros_like(zero_bias, device=scaled_weight.device) + if self.bn.affine: + full_bias = (conv_bias - self.bn.running_mean) / running_std * self.bn.weight + self.bn.bias + else: + full_bias = (conv_bias - self.bn.running_mean) / running_std + # quant_bias = self.bias_fake_quant(full_bias) + quant_bias = self.bias_fake_quant_proc(full_bias, self.weight_fake_quant.scale, self.input_fake_quantizer.scale) + conv_with_bias = self._convtransposed_forward(input, scaled_weight, quant_bias) + deconv_orig = (conv_with_bias - full_bias.reshape(bias_shape)) / scale_factor.reshape(bias_shape) + conv_bias.reshape(bias_shape) + deconv = self.bn(deconv_orig) + return deconv def _convtransposed_forward(self, x, w, b): raise NotImplementedError( diff --git a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py index 760a3847..59bcdc35 100644 --- a/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py +++ b/mqbench/nn/intrinsic/qat/modules/linear_fused_sophgo_tpu.py @@ -63,7 +63,7 @@ def reset_bn_parameters(self): init.uniform_(self.bias, -bound, bound) def reset_parameters(self): - super(LinearBn1d_sophgo, self).reset_parameters() + super(LinearBn1d, self).reset_parameters() def update_bn_stats(self): self.freeze_bn = False @@ -86,11 +86,48 @@ def bias_fake_quant(self, bias, scale_w, in_scale): bias = bias*scale return bias + # def _forward(self, input): + # in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale + # conv = F.linear(input, self.weight_fake_quant(self.weight), + # self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) + # return conv + def _forward(self, input): + assert self.bn.running_var is not None + running_std = torch.sqrt(self.bn.running_var + self.bn.eps) + # input.shape = (batch_size, in_features, *) + # scale_factor.shape = (out_feature, ) + # self.weight.shape = (out_feature, in_feature, *) + # self.bias.shape = (out_feature, *) + # output.shape = (batch_size, out_feature, *) + if self.bn.affine: + scale_factor = self.bn.weight / running_std + else: + scale_factor = 1. / running_std + weight_shape = [1] * len(self.weight.shape) + weight_shape[0] = -1 + bias_shape = [1] * len(input.shape) + bias_shape[1] = -1 + scaled_weight = self.weight_fake_quant(self.weight * scale_factor.reshape(weight_shape)) + # using zero bias here since the bias for original Linear + # will be added later + # Linear layer takes permuted input since the format is (batch_size, *, in_features) + if self.bias is not None: + zero_bias = torch.zeros_like(self.bias) + fc_bias = self.bias + else: + zero_bias = torch.zeros(self.out_channels, device=scaled_weight.device) + fc_bias = torch.zeros_like(zero_bias, device=scaled_weight.device) + if self.bn.affine: + full_bias = (fc_bias - self.bn.running_mean) / running_std * self.bn.weight + self.bn.bias + else: + full_bias = (fc_bias - self.bn.running_mean) / running_std in_scale = self.input_fake_quantizer.scale #锟斤拷锟斤拷一锟斤拷activation_fake_quant锟节碉拷锟饺cale - conv = F.linear(input, self.weight_fake_quant(self.weight), - self.bias_fake_quant(self.bias, self.weight_fake_quant.scale, in_scale)) - return conv + fquant_bias = self.bias_fake_quant(full_bias, self.weight_fake_quant.scale, in_scale) + linear_out = F.linear(input, scaled_weight, fquant_bias) + linear_orig = (linear_out - full_bias) / scale_factor.reshape(bias_shape) + fc_bias.reshape(bias_shape) + linear_out = self.bn(linear_orig) + return linear_out def forward(self, input): # return F.linear(input, self.weight_fake_quant(self.weight), self.bias) diff --git a/mqbench/prepare_by_platform.py b/mqbench/prepare_by_platform.py index 89d3e4eb..3dd7a80d 100644 --- a/mqbench/prepare_by_platform.py +++ b/mqbench/prepare_by_platform.py @@ -9,7 +9,8 @@ from torch.fx.graph_module import GraphModule from torch.quantization.quantize_fx import _swap_ff_with_fxff from torch.quantization import QConfig - +import torch.nn.intrinsic as nni +import mqbench.nn.intrinsic as qnni from mqbench.fake_quantize import ( LearnableFakeQuantize, @@ -33,10 +34,12 @@ MSEObserver, EMAMSEObserver, ) +import mqbench from mqbench.fuser_method_mappings import fuse_custom_config_dict from mqbench.utils.logger import logger from mqbench.utils.registry import DEFAULT_MODEL_QUANTIZER from mqbench.scheme import QuantizeScheme +import mqbench.nn.intrinsic.qat as qnniqat __all__ = ['prepare_by_platform'] @@ -158,7 +161,7 @@ class BackendType(Enum): } -def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): +def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict, work_mode: str): """ Args: @@ -284,7 +287,49 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): qconfig = {'': QConfig(activation=a_qconfig, weight=w_qconfig)} if deploy_backend == BackendType.Sophgo_TPU: - qconfig["object_type"] = {torch.nn.Linear:createQConfig(deploy_backend)} + qconfig["object_type"] = {torch.nn.Linear:createQConfigForSophgoLiner()} #int8 qat, Sophgo_TPU use sym per-layer + if work_mode == 'all_int4_qat': + qconfig["object_type"][torch.nn.Linear] = createQConfigForSophgoLiner(bit_num = 4) + if work_mode in ['int4_and_int8_mix', 'int4_and_int8_mix_no_fc']: + w_qscheme = { + 'bit': 4, + 'symmetry': True, + 'per_channel': True, + 'pot_scale': False + } + a_qscheme = { + 'bit': 4, + 'symmetry': True, + 'per_channel': False, + 'pot_scale': False + } + int4_qconfig = createQConfig(w_qscheme = w_qscheme, a_qscheme = a_qscheme) + qconfig["object_type"][torch.nn.Conv2d] = int4_qconfig + from mqbench.custom_quantizer.sophgo_tpu_quantizer import SophgoTpuQuantizer + + import torch.nn as nn + additional_qat_module_mapping = [ + nni.ConvBn2d, + nni.ConvBnReLU2d, + nn.Conv2d, + nni.ConvReLU2d, + qnni.ConvTransposeBnReLU2d, + qnni.ConvTransposeReLU2d, + qnni.ConvTransposeBn2d + ] + for i in additional_qat_module_mapping: + qconfig["object_type"][i] = int4_qconfig + if work_mode == 'int4_and_int8_mix_no_fc': + for i in SophgoTpuQuantizer({}, {})._layers_need_check_is_dw: + qconfig["object_type"][i] = int4_qconfig + else: + for i in SophgoTpuQuantizer({}, {})._layers_need_scale_form_input_fake_quantizer: + qconfig["object_type"][i] = int4_qconfig + if work_mode != 'int4_and_int8_mix_no_fc': + liner_qconfig = createQConfigForInt4SophgoLiner() + qconfig["object_type"][nni.LinearReLU] = liner_qconfig + qconfig["object_type"][qnni.LinearBn1d] = liner_qconfig + qconfig["object_type"][torch.nn.Linear] = liner_qconfig object_type = extra_qparams.get('object_type', None) if object_type is not None: if "object_type" in qconfig: @@ -311,7 +356,15 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict): # 'pot_scale': whether scale is power of two. # } -def createQConfig(deploy_backend, onlyCreate_WQconfig = True, w_fakequantize = 'LearnableFakeQuantize', a_fakequantize = 'LearnableFakeQuantize', +def createQConfigForSophgoLiner(bit_num = 8, w_fakequantize = 'LearnableFakeQuantize', w_observer = 'MinMaxObserver', w_fakeq_params = {}, w_observer_extra_args = {}): + w_observer = ObserverDict[w_observer] + w_fakequantize = FakeQuantizeDict[w_fakequantize] + w_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=bit_num) #Sophgo_TPU use sym per-layer + w_qscheme.kwargs.update(w_observer_extra_args) + w_qconfig = w_fakequantize.with_args(observer=w_observer, **w_fakeq_params, **w_qscheme.to_observer_params()) + return QConfig(activation=torch.nn.Identity, weight=w_qconfig) #activation use global quant conifg + +def createQConfig(w_fakequantize = 'LearnableFakeQuantize', a_fakequantize = 'LearnableFakeQuantize', w_observer = 'MinMaxObserver', a_observer = 'EMAMinMaxObserver', w_qscheme = {}, a_qscheme = {}, w_fakeq_params = {}, a_fakeq_params = {}, w_observer_extra_args = {}, a_observer_extra_args = {}): w_observer = ObserverDict[w_observer] @@ -320,23 +373,35 @@ def createQConfig(deploy_backend, onlyCreate_WQconfig = True, w_fakequantize = ' w_qscheme = QuantizeScheme(**w_qscheme) else: w_qscheme = QuantizeScheme(symmetry=True, per_channel=True, pot_scale=False, bit=8) - if deploy_backend == BackendType.Sophgo_TPU: - w_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8) w_qscheme.kwargs.update(w_observer_extra_args) w_qconfig = w_fakequantize.with_args(observer=w_observer, **w_fakeq_params, **w_qscheme.to_observer_params()) - if not onlyCreate_WQconfig: - a_observer = ObserverDict[a_observer] - a_fakequantize = FakeQuantizeDict[a_fakequantize] - if a_qscheme is not None: - a_qscheme = QuantizeScheme(**a_qscheme) - else: - a_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8) - a_qscheme.kwargs.update(a_observer_extra_args) - a_qconfig = a_fakequantize.with_args(observer=a_observer, **a_fakeq_params, **a_qscheme.to_observer_params()) + a_observer = ObserverDict[a_observer] + a_fakequantize = FakeQuantizeDict[a_fakequantize] + if a_qscheme is not None: + a_qscheme = QuantizeScheme(**a_qscheme) else: - a_qconfig = torch.nn.Identity + a_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=8) + a_qscheme.kwargs.update(a_observer_extra_args) + a_qconfig = a_fakequantize.with_args(observer=a_observer, **a_fakeq_params, **a_qscheme.to_observer_params()) + return QConfig(activation=a_qconfig, weight=w_qconfig) + +def createQConfigForInt4SophgoLiner(w_fakequantize = 'LearnableFakeQuantize', a_fakequantize = 'LearnableFakeQuantize', + w_observer = 'MinMaxObserver', a_observer = 'EMAMinMaxObserver', w_qscheme = {}, a_qscheme = {}, + w_fakeq_params = {}, a_fakeq_params = {}, w_observer_extra_args = {}, a_observer_extra_args = {}): + w_observer = ObserverDict[w_observer] + w_fakequantize = FakeQuantizeDict[w_fakequantize] + w_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=4) + + w_qscheme.kwargs.update(w_observer_extra_args) + w_qconfig = w_fakequantize.with_args(observer=w_observer, **w_fakeq_params, **w_qscheme.to_observer_params()) + + a_observer = ObserverDict[a_observer] + a_fakequantize = FakeQuantizeDict[a_fakequantize] + a_qscheme = QuantizeScheme(symmetry=True, per_channel=False, pot_scale=False, bit=4) + a_qscheme.kwargs.update(a_observer_extra_args) + a_qconfig = a_fakequantize.with_args(observer=a_observer, **a_fakeq_params, **a_qscheme.to_observer_params()) return QConfig(activation=a_qconfig, weight=w_qconfig) class CustomedTracer(Tracer): @@ -433,7 +498,8 @@ def prepare_by_platform( # Get Qconfig extra_qconfig_dict = prepare_custom_config_dict.get('extra_qconfig_dict', {}) - qconfig = get_qconfig_by_platform(deploy_backend, extra_qconfig_dict) + work_mode = prepare_custom_config_dict.get('work_mode', '') + qconfig = get_qconfig_by_platform(deploy_backend, extra_qconfig_dict, work_mode) _swap_ff_with_fxff(model) # Preserve attr. @@ -454,10 +520,10 @@ def prepare_by_platform( if custom_tracer is not None: tracer = custom_tracer graph = tracer.trace(model, concrete_args) - # print('trace graph:',graph) + print('>>>>>trace graph:',graph) name = model.__class__.__name__ if isinstance(model, torch.nn.Module) else model.__name__ modules = dict(model.named_modules()) - # print('named_modules:',modules) + print('>>>>>named_modules:',modules['']) graph, duplicated_modules = duplicate_reused_nodes(graph, modules) constant_nodes = prepare_constant_dict(graph, model) modules.update(duplicated_modules) diff --git a/mqbench/train_all_model.py b/mqbench/train_all_model.py index 80d0d488..860c74b3 100644 --- a/mqbench/train_all_model.py +++ b/mqbench/train_all_model.py @@ -9,26 +9,36 @@ opt = parser.parse_args() model_list_all=[ - #"--arch=shufflenet_v2_x0_5 --batch-size=320", - "--arch=mobilenet_v2 --batch-size=64", - "--arch=resnet18 --batch-size=128", - "--arch=vgg11_bn --batch-size=32", - "--arch=resnet50 --batch-size=32", - "--arch=squeezenet1_1 --batch-size=128", - - #"--arch=mobilenet_v3_small --batch-size=128" + # "--arch=shufflenet_v2_x0_5 --batch-size=320 --lr=1e-2", + #"--arch=mobilenet_v2 --batch-size=128 --lr=1e-3", + "--arch=resnet18 --batch-size=256 --lr=1e-2", + # "--arch=vgg11_bn --batch-size=32 --lr=1e-3", + #"--arch=resnet50 --batch-size=32 --lr=1e-2", + # "--arch=squeezenet1_1 --batch-size=128 --lr=1e-3", + #"--arch=mobilenet_v3_small --batch-size=128 --lr=1e-2" ] -cmd_str = "--epochs=1 --deploy_batch_size=10 --lr=1e-4 --gpu=0 --pretrained --evaluate --backend=sophgo_tpu --optim=sgd --pre_eval_and_export --train_data=/data/imagenet/for_train_val/ --val_data=/data/imagenet/for_train_val/ --output_path=/workspace/tmp_path_1119"# --fast_test" +epochs = 3 +output_path='/workspace/tmp_path_0322_tmptest' +output_path='/workspace/tmp_path_0412_tmptest' +fast_test = '' +fast_test = '--fast_test' +pre_eval_and_export = '--pre_eval_and_export' +pre_eval_and_export = '' +cmd_str = f"--epochs={epochs} --deploy_batch_size=10 --gpu=0 --pretrained --evaluate --backend=sophgo_tpu --optim=sgd \ + --train_data=/workspace/for_train_val/ --val_data=/workspace/for_train_val/ --output_path={output_path} {fast_test} {pre_eval_and_export}" def worker(cmd_line): + print('cmd_line:', cmd_line) os.system(cmd_line) if __name__ == "__main__": wp = os.getcwd() po = Pool(1) for i,model in enumerate(model_list_all): - cmd_line = 'cd {};python3 imagenet_example/main.py {} {} >{}_log_train_{} 2>&1'.format(wp, model, cmd_str, i, model.split(' ')[0].split('=')[1]) + arch = model.split(' ')[0].split('=')[1].strip() + os.system(f'mkdir -p {output_path}/{arch}') + cmd_line = f'cd {wp};python3 application/imagenet_example/main.py {model} {cmd_str} > {output_path}/{arch}/{i}_log_train_{arch} 2>&1' print('start', model) po.apply_async(worker, (cmd_line,)) From 232ca1ff04138781baaefdd6ffc09fe4fb56a2ca Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Wed, 12 Apr 2023 17:39:24 +0800 Subject: [PATCH 10/29] fix int8 bug in int4 version --- mqbench/custom_quantizer/sophgo_tpu_quantizer.py | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index dcbe1bc2..694b9fa7 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -146,7 +146,7 @@ def _insert_fake_quantize_for_act_quant( print('int4_and_int8_mix:', int4_and_int8_mix) if int4_and_int8_mix: def all_next_layers_is_trivial(node): - if len(node.users): + if len(node.users) == 0: return True for user in node.users: if not all_next_layers_is_trivial(user): @@ -197,11 +197,12 @@ def find_next_int4_and_int8_layers(node, int8_layers, int4_layers): qconfig2 = None #娣卞害鍗风Н浣跨敤int8璁＄畻 if qconfig2 is None: qconfig2 = flattened_qconfig_dict.get('', None) #鏈鍚庢壘鍏ㄥ眬qconfig锛屼紭鍏堢骇鏈浣 - if node.target in modules and type(modules[node.target]) == torch.nn.ReLU6 and node.args[0].target in modules: - qconfig2 = flattened_qconfig_dict.get(type(modules[node.args[0].target]), None) - if isinstance(modules[node.args[0].target], self._layers_need_check_is_dw): - if modules[node.args[0].target].groups > 1: - qconfig2 = flattened_qconfig_dict.get('', None) ##娣卞害鍗风Н浣跨敤int8璁＄畻 + if int4_and_int8_mix: + if node.target in modules and type(modules[node.target]) == torch.nn.ReLU6 and node.args[0].target in modules: + qconfig2 = flattened_qconfig_dict.get(type(modules[node.args[0].target]), None) + if isinstance(modules[node.args[0].target], self._layers_need_check_is_dw): + if modules[node.args[0].target].groups > 1: + qconfig2 = flattened_qconfig_dict.get('', None) ##娣卞害鍗风Н浣跨敤int8璁＄畻 node_fake_quantizer = qconfig2.activation() # node_fake_quantizer.enable_only_observer() quantizer_name2 = node.name + "_post_act_fake_quantizer" From 1d09bc4fcd9aa1bceaef8da427253749eab1769b Mon Sep 17 00:00:00 2001 From: wangxc2006 Date: Wed, 21 Jun 2023 15:49:16 +0800 Subject: [PATCH 11/29] fix sub/abs op no fake quant node --- application/imagenet_example/main.py | 2 +- mqbench/custom_quantizer/model_quantizer.py | 3 ++ .../custom_quantizer/sophgo_tpu_quantizer.py | 40 ++++++++++++++++++- mqbench/deploy/deploy_sophgo.py | 30 ++++++++------ mqbench/prepare_by_platform.py | 9 +++++ 5 files changed, 69 insertions(+), 15 deletions(-) diff --git a/application/imagenet_example/main.py b/application/imagenet_example/main.py index f14b3db5..e706a403 100644 --- a/application/imagenet_example/main.py +++ b/application/imagenet_example/main.py @@ -231,7 +231,7 @@ def main_worker(gpu, ngpus_per_node, args): # } # } } - model = prepare_by_platform(model, args.backend, prepare_custom_config_dict) + model = prepare_by_platform(model, args.backend, input_shape_dict = {'data': [args.deploy_batch_size, 3, 224, 224]}) print('>>>>>prepared module:', model) if args.fast_test: diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index 9147493d..0e1299d7 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -242,6 +242,9 @@ def _find_act_quants(self, model: GraphModule) -> List: if self._is_implicit_merge(modules, (node, _node)): logger.info("Implicit merge: {} + {}".format(_node.name, node.name)) continue + if _node.op == "placeholder" and 'tensor_meta' in node.meta: + if len(_node.meta['tensor_meta'].shape) == 1: + continue if _node in node_need_to_quantize_output: continue if _node in g2node: diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index 694b9fa7..c7108072 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -52,22 +52,49 @@ def module_type_to_quant_input(self) -> tuple: torch.nn.Hardswish, torch.nn.Sigmoid, torch.nn.SiLU, + torch.nn.Tanh, + torch.nn.SELU, + torch.nn.LogSigmoid, + torch.nn.GELU, + torch.nn.GLU, + torch.nn.Mish, + torch.nn.Hardsigmoid, + torch.nn.Softshrink, + torch.nn.Softplus, + torch.nn.ELU, + torch.nn.CELU, ) + super().module_type_to_quant_input + self._layers_need_scale_form_input_fake_quantizer @property def function_type_to_quant_input(self) -> tuple: return super().function_type_to_quant_input + [ + operator.sub, + operator.abs, torch.cat, + torch.sub, + torch.clamp, torch.nn.functional.hardswish, torch.nn.functional.sigmoid, - torch.nn.functional.silu + torch.nn.functional.silu, + torch.nn.functional.tanh, + torch.nn.functional.selu, + torch.nn.functional.logsigmoid, + torch.nn.functional.gelu, + torch.nn.functional.glu, + torch.nn.functional.mish, + torch.nn.functional.hardsigmoid, + torch.nn.functional.gumbel_softmax, + torch.nn.functional.softshrink, + torch.nn.functional.softplus, + torch.nn.functional.elu, + torch.nn.functional.celu, ] @property def _passed_func_type(self): return ( torch.nn.functional.relu, - torch.nn.functional.relu6, + # torch.nn.functional.relu6, torch.flatten ) @@ -279,6 +306,15 @@ def _find_act_quants(self, model: GraphModule) -> list: if node.target in modules and type(modules[node.target]) in self.exclude_module_name: print(f'{type(modules[node.target])} is excluded') node_need_to_quantize_output.remove(node) + if node.op == "placeholder": + if 'tensor_meta' in node.meta: + if len(node.meta['tensor_meta'].shape) > 1: + print(f'add placeholder {node.target} to node_need_to_quantize_output by tensor_meta') + node_need_to_quantize_output.append(node) + else: + print(f'no tensor_meta, add placeholder {node.target} to node_need_to_quantize_output') + node_need_to_quantize_output.append(node) + return node_need_to_quantize_output def _set_fake_quantizer_to_next_weight_layer(self, model: GraphModule): diff --git a/mqbench/deploy/deploy_sophgo.py b/mqbench/deploy/deploy_sophgo.py index 80288e7c..6852393b 100644 --- a/mqbench/deploy/deploy_sophgo.py +++ b/mqbench/deploy/deploy_sophgo.py @@ -120,15 +120,18 @@ def clip_weight(self, node, name2data, inp2node, named_initializer): new_data = numpy_helper.from_array(new_data) named_initializer[tensor_name].raw_data = new_data.raw_data - def get_correct_sophgo_tpu_input_tensor_name(self, node, out2node): #鍜宼pu-mlir鐨勫懡鍚嶉鏍间竴鑷 + def get_correct_sophgo_tpu_input_tensor_name(self, node, out2node, have_int4=False): #鍜宼pu-mlir鐨勫懡鍚嶉鏍间竴鑷 input_0 = node.input[0] - op_type = out2node[input_0].op_type - tensor_name = '{}_{}'.format(input_0, op_type if input_0 in out2node else '') - if tensor_name[-1] == '_': - tensor_name = tensor_name[:-1] - tensor_name += '_8' - if op_type in ['Conv', "Gemm"]: + op_type = '' + if input_0 in out2node: + op_type = out2node[input_0].op_type + tensor_name = '{}_{}'.format(input_0, op_type) + else: + tensor_name = input_0 + if have_int4 and op_type in ['Conv', "Gemm"]: tensor_name += '_4' + else: + tensor_name += '_8' return tensor_name def post_process_clip_ranges(self, clip_ranges, graph, inp2node, out2node): @@ -150,16 +153,16 @@ def find_the_closest_clip_range(node): logger.info(f'Pass <{tensor_name}> clip range to <{node.name}> input <{node.input[0]}>.') return clip_ranges - def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): + def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node, have_int4 = False): op_type_inAndOutShouldSameClipRange = ['Flatten', 'Resize', 'Reshape', 'Transpose'] for node in graph.node: tensor_name = f'{node.output[0]}_{node.op_type}' - tensor_name += '_4' if node.op_type in ['Conv', "Gemm"] else '_8' + tensor_name += '_4' if have_int4 and node.op_type in ['Conv', "Gemm"] else '_8' if tensor_name not in clip_ranges: pre_op = node finded = False while pre_op.op_type in op_type_inAndOutShouldSameClipRange: - tensor_name2 = self.get_correct_sophgo_tpu_input_tensor_name(pre_op, out2node) + tensor_name2 = self.get_correct_sophgo_tpu_input_tensor_name(pre_op, out2node, have_int4) if tensor_name2 in clip_ranges: finded = True clip_ranges[tensor_name] = clip_ranges[tensor_name2] @@ -175,7 +178,7 @@ def post_process_clip_ranges2(self, clip_ranges, graph, inp2node, out2node): next_op = inp2node[node.output[0]][0][0] while next_op.op_type in op_type_inAndOutShouldSameClipRange: tensor_name2 = f'{next_op.output[0]}_{next_op.op_type}' - tensor_name2 += '_4' if next_op.op_type in ['Conv', "Gemm"] else '_8' + tensor_name2 += '_4' if have_int4 and next_op.op_type in ['Conv', "Gemm"] else '_8' if tensor_name2 in clip_ranges: finded = True clip_ranges[tensor_name] = clip_ranges[tensor_name2] @@ -241,6 +244,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name): output_path = os.path.dirname(onnx_path) file_name = os.path.join(output_path, 'layer_outputs.npz') layer_out_tensor = None + have_int4 = False layer_out_tensor2 = {} if os.path.exists(file_name): layer_out_tensor = np.load(file_name) @@ -287,6 +291,8 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name): self.deal_with_activation_fakequant(node, inp2node) tensor_name, scale, zero_point, qmin, qmax = self.parse_qparams(node, name2data) bits = 4 if qmax == 7 else 8 + if bits == 4: + have_int4 = True for out in graph.output: output_name = node.output[0] # if inp2node[node.output[0]][0][0].op_type == 'Add': @@ -333,7 +339,7 @@ def remove_fakequantize_and_collect_params(self, onnx_path, model_name): graph.initializer.remove(initial_data) # clip_ranges = self.post_process_clip_ranges(clip_ranges, graph, inp2node, out2node) - clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node) + clip_ranges = self.post_process_clip_ranges2(clip_ranges, graph, inp2node, out2node, have_int4) context = {'sophgo_tpu': clip_ranges} context['w_qscheme'] = '' context['a_qscheme'] = '' diff --git a/mqbench/prepare_by_platform.py b/mqbench/prepare_by_platform.py index 3dd7a80d..3fbc39c6 100644 --- a/mqbench/prepare_by_platform.py +++ b/mqbench/prepare_by_platform.py @@ -473,6 +473,7 @@ def _get_attrs(target, attrs): def prepare_by_platform( model: torch.nn.Module, deploy_backend: BackendType, + input_shape_dict: list = None, prepare_custom_config_dict: Dict[str, Any] = {}, custom_tracer: Tracer = None): """ @@ -529,6 +530,14 @@ def prepare_by_platform( modules.update(duplicated_modules) modules.update(constant_nodes) graph_module = GraphModule(modules, graph, name) + if input_shape_dict is not None: + try: + from torch.fx.passes import shape_prop + dev = next(model.parameters()).device + dummy_input = [torch.rand(shape).to(dev) for shape in input_shape_dict] + shape_prop.ShapeProp(graph_module).propagate(*dummy_input) + except: + print('waring, shape_prop fail') # Model fusion. extra_fuse_dict = prepare_custom_config_dict.get('extra_fuse_dict', {}) extra_fuse_dict.update(fuse_custom_config_dict) From 1f43b9dda74bd93c9ae1a75da7616355f7d341a2 Mon Sep 17 00:00:00 2001 From: "zhengjin.xu" Date: Wed, 2 Aug 2023 20:28:56 +0800 Subject: [PATCH 12/29] add some class and func to adapt to torch1.10_cpu and torch2.0.1_cpu --- .../custom_quantizer/academic_quantizer.py | 2 +- mqbench/custom_quantizer/model_quantizer.py | 4 +- .../custom_quantizer/openvino_quantizer.py | 2 +- .../custom_quantizer/sophgo_tpu_quantizer.py | 4 +- mqbench/fake_quantize/fixed.py | 3 +- mqbench/fake_quantize/quantize_base.py | 2 +- mqbench/fuser_method_mappings.py | 276 +++++++++++++++++- mqbench/fusion_method.py | 9 +- mqbench/prepare_by_platform.py | 3 + mqbench/utils/utils.py | 39 ++- 10 files changed, 326 insertions(+), 18 deletions(-) diff --git a/mqbench/custom_quantizer/academic_quantizer.py b/mqbench/custom_quantizer/academic_quantizer.py index 923d7834..30a50a52 100644 --- a/mqbench/custom_quantizer/academic_quantizer.py +++ b/mqbench/custom_quantizer/academic_quantizer.py @@ -6,7 +6,7 @@ import torch from torch.fx import GraphModule from torch.quantization import propagate_qconfig_ -from torch.quantization.fx.qconfig_utils import get_flattened_qconfig_dict +from mqbench.utils import get_flattened_qconfig_dict from mqbench.utils import is_symmetric_quant, getitem2node from mqbench.utils.logger import logger diff --git a/mqbench/custom_quantizer/model_quantizer.py b/mqbench/custom_quantizer/model_quantizer.py index 0e1299d7..d71403c8 100644 --- a/mqbench/custom_quantizer/model_quantizer.py +++ b/mqbench/custom_quantizer/model_quantizer.py @@ -23,13 +23,11 @@ from torch.quantization.utils import ( get_combined_dict ) -from torch.quantization.fx.qconfig_utils import ( - get_flattened_qconfig_dict -) from torch.quantization.quantize_fx import ( _fuse_fx ) +from mqbench.utils import get_flattened_qconfig_dict from mqbench.utils import getitem2node from mqbench.utils.logger import logger from mqbench.utils.registry import register_model_quantizer diff --git a/mqbench/custom_quantizer/openvino_quantizer.py b/mqbench/custom_quantizer/openvino_quantizer.py index 1509b832..46453d93 100644 --- a/mqbench/custom_quantizer/openvino_quantizer.py +++ b/mqbench/custom_quantizer/openvino_quantizer.py @@ -6,9 +6,9 @@ import torch from torch.fx import GraphModule from torch.quantization import propagate_qconfig_ -from torch.quantization.fx.qconfig_utils import get_flattened_qconfig_dict from torch.quantization.quantize_fx import _fuse_fx +from mqbench.utils import get_flattened_qconfig_dict from mqbench.utils import is_symmetric_quant from mqbench.utils.logger import logger from mqbench.utils.registry import register_model_quantizer diff --git a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py index c7108072..d60b8557 100644 --- a/mqbench/custom_quantizer/sophgo_tpu_quantizer.py +++ b/mqbench/custom_quantizer/sophgo_tpu_quantizer.py @@ -16,9 +16,7 @@ List, Dict, Any, Callable ) -from torch.quantization.fx.qconfig_utils import ( - get_flattened_qconfig_dict -) +from mqbench.utils import get_flattened_qconfig_dict @register_model_quantizer(BackendType.Sophgo_TPU) class SophgoTpuQuantizer(ModelQuantizer): diff --git a/mqbench/fake_quantize/fixed.py b/mqbench/fake_quantize/fixed.py index 1fd2ae2b..46c494fc 100644 --- a/mqbench/fake_quantize/fixed.py +++ b/mqbench/fake_quantize/fixed.py @@ -3,8 +3,7 @@ from mqbench.fake_quantize.quantize_base import QuantizeBase from mqbench.utils.hook import PerChannelLoadHook - -_version_under_1100 = int(torch.__version__.split('.')[1]) < 10 +from mqbench.fake_quantize.quantize_base import _version_under_1100 class FixedFakeQuantize(QuantizeBase): """This is actually torch.quantization.FakeQuantize. diff --git a/mqbench/fake_quantize/quantize_base.py b/mqbench/fake_quantize/quantize_base.py index 5794fb4f..36e34e26 100644 --- a/mqbench/fake_quantize/quantize_base.py +++ b/mqbench/fake_quantize/quantize_base.py @@ -5,7 +5,7 @@ from mqbench.utils import is_symmetric_quant -_version_under_1100 = int(torch.__version__.split('.')[1]) < 10 +_version_under_1100 = (int(torch.__version__.split('.')[1]) < 10) and (int(torch.__version__.split('.')[0]) == 1) class QuantizeBase(FakeQuantizeBase): r""" This is an extension of the FakeQuantize module in fake_quantize.py, which diff --git a/mqbench/fuser_method_mappings.py b/mqbench/fuser_method_mappings.py index eb1a56af..535d93f7 100644 --- a/mqbench/fuser_method_mappings.py +++ b/mqbench/fuser_method_mappings.py @@ -2,8 +2,6 @@ import torch import torch.nn as nn -from torch.quantization.fx.fusion_patterns import ConvBNReLUFusion, ModuleReLUFusion -from torch.quantization.fx.quantization_types import QuantizerCls from torch.fx.graph import Node import mqbench.nn as qnn @@ -12,6 +10,277 @@ from mqbench.utils.fusion import fuse_deconv_bn_eval from mqbench.nn.modules import FrozenBatchNorm2d +from collections import OrderedDict +from torch.ao.quantization.fuser_method_mappings import get_fuser_method +from abc import ABC, abstractmethod +from typing import Any, Callable, Dict +QuantizerCls = Any + +# pattern for conv bn fusion +DEFAULT_FUSION_PATTERNS = OrderedDict() +def register_fusion_pattern(pattern): + def insert(fn): + DEFAULT_FUSION_PATTERNS[pattern] = fn + return fn + return insert + +# turn foo.bar -> ['foo', 'bar'] +def _parent_name(target): + r = target.rsplit('.', 1) + if len(r) == 1: + return '', r[0] + else: + return r[0], r[1] + +# --------------------- +# Fusion Pattern Registrations +# --------------------- + +# Base Pattern Handler +class FuseHandler(ABC): + """ Base handler class for the fusion patterns + """ + def __init__(self, quantizer: QuantizerCls, node: Node): + pass + + @abstractmethod + def fuse(self, quantizer: QuantizerCls, load_arg: Callable, + fuse_custom_config_dict: Dict[str, Any] = None) -> Node: + pass + +@register_fusion_pattern((torch.nn.ReLU, torch.nn.Conv1d)) +@register_fusion_pattern((torch.nn.ReLU, torch.nn.Conv2d)) +@register_fusion_pattern((torch.nn.ReLU, torch.nn.Conv3d)) +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.Conv1d)) +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.Conv2d)) +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.Conv3d)) +@register_fusion_pattern((torch.nn.BatchNorm1d, torch.nn.Conv1d)) +@register_fusion_pattern((torch.nn.BatchNorm2d, torch.nn.Conv2d)) +@register_fusion_pattern((torch.nn.BatchNorm3d, torch.nn.Conv3d)) +@register_fusion_pattern((torch.nn.ReLU, (torch.nn.BatchNorm1d, torch.nn.Conv1d))) +@register_fusion_pattern((torch.nn.ReLU, (torch.nn.BatchNorm2d, torch.nn.Conv2d))) +@register_fusion_pattern((torch.nn.ReLU, (torch.nn.BatchNorm3d, torch.nn.Conv3d))) +@register_fusion_pattern((torch.nn.functional.relu, (torch.nn.BatchNorm1d, torch.nn.Conv1d))) +@register_fusion_pattern((torch.nn.functional.relu, (torch.nn.BatchNorm2d, torch.nn.Conv2d))) +@register_fusion_pattern((torch.nn.functional.relu, (torch.nn.BatchNorm3d, torch.nn.Conv3d))) +class ConvBNReLUFusion(FuseHandler): + def __init__(self, quantizer: QuantizerCls, node: Node): + super().__init__(quantizer, node) + self.relu_node = None + self.bn_node = None + if (node.op == 'call_function' and node.target is torch.nn.functional.relu) or \ + (node.op == 'call_module' and type(quantizer.modules[node.target]) == torch.nn.ReLU): + self.relu_node = node + assert isinstance(node.args[0], Node) + node = node.args[0] + assert node.op == 'call_module' + if type(quantizer.modules[node.target]) in [torch.nn.BatchNorm1d, torch.nn.BatchNorm2d, torch.nn.BatchNorm3d]: + self.bn_node = node + self.bn = quantizer.modules[self.bn_node.target] + assert isinstance(node.args[0], Node) + node = node.args[0] + assert node.op == 'call_module' + self.conv_node = node + self.conv = quantizer.modules[self.conv_node.target] + + def fuse(self, quantizer: QuantizerCls, load_arg: Callable, + fuse_custom_config_dict: Dict[str, Any] = None) -> Node: + if fuse_custom_config_dict is None: + fuse_custom_config_dict = {} + additional_fuser_method_mapping = fuse_custom_config_dict.get("additional_fuser_method_mapping", {}) + op_list = [] + if self.relu_node is not None: + # since relu can be used multiple times, we'll need to create a relu module for each match + if self.relu_node.op == 'call_module': + relu = torch.nn.ReLU(quantizer.modules[self.relu_node.target].inplace) + else: + # TODO: get inplace argument from functional + relu = torch.nn.ReLU() + op_list.append(relu) + relu.training = self.conv.training + if self.bn_node is not None: + op_list.append(self.bn) + op_list.append(self.conv) + else: + assert self.bn_node is not None + op_list.append(self.bn) + op_list.append(self.conv) + + # the modules are added in order of relu - bn - conv + # so we need to correct it + op_list.reverse() + op_type_list = tuple(type(m) for m in op_list) + conv_parent_name, conv_name = _parent_name(self.conv_node.target) + fuser_method = get_fuser_method(op_type_list, additional_fuser_method_mapping) + if fuser_method is None: + raise NotImplementedError("Cannot fuse modules: {}".format(op_type_list)) + fused = fuser_method(*op_list) + setattr(quantizer.modules[conv_parent_name], conv_name, fused) + + # TODO: do we need to make sure bn is only used once? + if self.bn_node is not None: + parent_name, name = _parent_name(self.bn_node.target) + setattr(quantizer.modules[parent_name], name, torch.nn.Identity()) + # relu may be used multiple times, so we don't set relu to identity + return quantizer.fused_graph.node_copy(self.conv_node, load_arg) + +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.Linear)) +@register_fusion_pattern((torch.nn.ReLU, torch.nn.Linear)) +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.BatchNorm2d)) +@register_fusion_pattern((torch.nn.ReLU, torch.nn.BatchNorm2d)) +@register_fusion_pattern((torch.nn.functional.relu, torch.nn.BatchNorm3d)) +@register_fusion_pattern((torch.nn.ReLU, torch.nn.BatchNorm3d)) +class ModuleReLUFusion(FuseHandler): + def __init__(self, quantizer: QuantizerCls, node: Node): + super().__init__(quantizer, node) + self.relu_node = node + assert isinstance(node.args[0], Node) + node = node.args[0] + assert node.op == 'call_module' + self.module_node = node + self.module = quantizer.modules[self.module_node.target] + + def fuse(self, quantizer: QuantizerCls, load_arg: Callable, + fuse_custom_config_dict: Dict[str, Any] = None) -> Node: + if fuse_custom_config_dict is None: + fuse_custom_config_dict = {} + additional_fuser_method_mapping = fuse_custom_config_dict.get("additional_fuser_method_mapping", {}) + op_list = [] + # since relu can be used multiple times, we'll need to create a relu module for each match + if self.relu_node.op == 'call_module': + relu = torch.nn.ReLU(quantizer.modules[self.relu_node.target].inplace) + else: + # TODO: get inplace argument from functional + relu = torch.nn.ReLU() + relu.training = self.module.training + op_list.append(relu) + op_list.append(self.module) + + op_list.reverse() + op_type_list = tuple(type(m) for m in op_list) + module_parent_name, module_name = _parent_name(self.module_node.target) + fuser_method = get_fuser_method(op_type_list, additional_fuser_method_mapping) + setattr(quantizer.modules[module_parent_name], module_name, fuser_method(*op_list)) + return quantizer.fused_graph.node_copy(self.module_node, load_arg) + +import torch.nn.intrinsic as nni +from typing import Union, Callable, Tuple, Dict, Optional, Type +from torch.ao.quantization.utils import get_combined_dict + +def fuse_conv_bn(conv, bn): + r"""Given the conv and bn modules, fuses them and returns the fused module + + Args: + conv: Module instance of type conv2d/conv3d + bn: Spatial BN instance that needs to be fused with the conv + + Examples:: + + >>> m1 = nn.Conv2d(10, 20, 3) + >>> b1 = nn.BatchNorm2d(20) + >>> m2 = fuse_conv_bn(m1, b1) + """ + assert(conv.training == bn.training),\ + "Conv and BN both must be in the same mode (train or eval)." + + fused_module_class_map = { + nn.Conv1d: nni.ConvBn1d, + nn.Conv2d: nni.ConvBn2d, + nn.Conv3d: nni.ConvBn3d, + } + + if conv.training: + assert bn.num_features == conv.out_channels, 'Output channel of Conv2d must match num_features of BatchNorm2d' + assert bn.affine, 'Only support fusing BatchNorm2d with affine set to True' + assert bn.track_running_stats, 'Only support fusing BatchNorm2d with tracking_running_stats set to True' + fused_module_class = fused_module_class_map.get((type(conv)), None) + if fused_module_class is not None: + return fused_module_class(conv, bn) + else: + raise NotImplementedError("Cannot fuse train modules: {}".format((conv, bn))) + else: + return nn.utils.fuse_conv_bn_eval(conv, bn) + +def fuse_conv_bn_relu(conv, bn, relu): + r"""Given the conv and bn modules, fuses them and returns the fused module + + Args: + conv: Module instance of type conv2d/conv3d + bn: Spatial BN instance that needs to be fused with the conv + + Examples:: + + >>> m1 = nn.Conv2d(10, 20, 3) + >>> b1 = nn.BatchNorm2d(20) + >>> r1 = nn.ReLU(inplace=False) + >>> m2 = fuse_conv_bn_relu(m1, b1, r1) + """ + assert(conv.training == bn.training == relu.training),\ + "Conv and BN both must be in the same mode (train or eval)." + fused_module : Optional[Type[nn.Sequential]] = None + if conv.training: + map_to_fused_module_train = { + nn.Conv1d: nni.ConvBnReLU1d, + nn.Conv2d: nni.ConvBnReLU2d, + nn.Conv3d: nni.ConvBnReLU3d, + } + assert bn.num_features == conv.out_channels, 'Output channel of Conv must match num_features of BatchNorm' + assert bn.affine, 'Only support fusing BatchNorm with affine set to True' + assert bn.track_running_stats, 'Only support fusing BatchNorm with tracking_running_stats set to True' + fused_module = map_to_fused_module_train.get(type(conv), None) + if fused_module is not None: + return fused_module(conv, bn, relu) + else: + raise NotImplementedError("Cannot fuse train modules: {}".format((conv, bn, relu))) + else: + map_to_fused_module_eval = { + nn.Conv1d: nni.ConvReLU1d, + nn.Conv2d: nni.ConvReLU2d, + nn.Conv3d: nni.ConvReLU3d, + } + fused_module = map_to_fused_module_eval.get(type(conv), None) + if fused_module is not None: + fused_conv = nn.utils.fusion.fuse_conv_bn_eval(conv, bn) + return fused_module(fused_conv, relu) + else: + raise NotImplementedError("Cannot fuse eval modules: {}".format((conv, bn, relu))) + +def fuse_linear_bn(linear, bn): + r"""Given the linear and bn modules, fuses them and returns the fused module + + Args: + linear: Module instance of type Linear + bn: BatchNorm1d instance that needs to be fused with the linear layer + + Examples:: + + >>> m1 = nn.Linear(20, 10) + >>> b1 = nn.BatchNorm1d(10) + >>> m2 = fuse_linear_bn(m1, b1) + """ + assert(linear.training == bn.training),\ + "Linear and BN both must be in the same mode (train or eval)." + + if linear.training: + raise Exception("Fusing Linear+BatchNorm not yet supported in training.") + else: + return nn.utils.fusion.fuse_linear_bn_eval(linear, bn) + +DEFAULT_OP_LIST_TO_FUSER_METHOD : Dict[Tuple, Union[nn.Sequential, Callable]] = { + (nn.Conv1d, nn.BatchNorm1d): fuse_conv_bn, + (nn.Conv1d, nn.BatchNorm1d, nn.ReLU): fuse_conv_bn_relu, + (nn.Conv2d, nn.BatchNorm2d): fuse_conv_bn, + (nn.Conv2d, nn.BatchNorm2d, nn.ReLU): fuse_conv_bn_relu, + (nn.Conv3d, nn.BatchNorm3d): fuse_conv_bn, + (nn.Conv3d, nn.BatchNorm3d, nn.ReLU): fuse_conv_bn_relu, + (nn.Conv1d, nn.ReLU): nni.ConvReLU1d, + (nn.Conv2d, nn.ReLU): nni.ConvReLU2d, + (nn.Conv3d, nn.ReLU): nni.ConvReLU3d, + (nn.Linear, nn.BatchNorm1d): fuse_linear_bn, + (nn.Linear, nn.ReLU): nni.LinearReLU, + (nn.BatchNorm2d, nn.ReLU): nni.BNReLU2d, + (nn.BatchNorm3d, nn.ReLU): nni.BNReLU3d, +} class ConvFreezebnReLUFusion(ConvBNReLUFusion): def __init__(self, quantizer: QuantizerCls, node: Node): @@ -202,11 +471,8 @@ def _sort_fusion_patterns(pats): for key in keys: pats.move_to_end(key) - # Sinse additional_fuser_method_mapping will not be set because fuser.py:54 # do not pass this dict. -from torch.quantization.fuser_method_mappings import DEFAULT_OP_LIST_TO_FUSER_METHOD -from torch.quantization.fx.pattern_utils import DEFAULT_FUSION_PATTERNS from torch.quantization.quantization_mappings import DEFAULT_QAT_MODULE_MAPPINGS DEFAULT_OP_LIST_TO_FUSER_METHOD.update( diff --git a/mqbench/fusion_method.py b/mqbench/fusion_method.py index dcb623e5..30326a1a 100644 --- a/mqbench/fusion_method.py +++ b/mqbench/fusion_method.py @@ -1,7 +1,14 @@ import torch import torch.nn.intrinsic.qat as nniqat from torch.nn.utils.fusion import fuse_conv_bn_eval, fuse_linear_bn_eval -from torch.quantization.fx.utils import _parent_name + +# turn foo.bar -> ['foo', 'bar'] +def _parent_name(target): + r = target.rsplit('.', 1) + if len(r) == 1: + return '', r[0] + else: + return r[0], r[1] import mqbench.nn.intrinsic as qnni import mqbench.nn.intrinsic.qat as qnniqat diff --git a/mqbench/prepare_by_platform.py b/mqbench/prepare_by_platform.py index 3fbc39c6..84fc2043 100644 --- a/mqbench/prepare_by_platform.py +++ b/mqbench/prepare_by_platform.py @@ -284,6 +284,9 @@ def get_qconfig_by_platform(deploy_backend: BackendType, extra_qparams: Dict, wo a_observer.__name__, str(a_qscheme))) if backend_params['qtype'] == 'vitis': logger.info('Bias Qconfig:\n TqtFakeQuantize with MinMaxObserver') + + if deploy_backend in [BackendType.Academic, BackendType.Academic_NLP]: + return QConfig(activation=a_qconfig, weight=w_qconfig) qconfig = {'': QConfig(activation=a_qconfig, weight=w_qconfig)} if deploy_backend == BackendType.Sophgo_TPU: diff --git a/mqbench/utils/utils.py b/mqbench/utils/utils.py index 51a0859a..8a0aa2ee 100644 --- a/mqbench/utils/utils.py +++ b/mqbench/utils/utils.py @@ -176,4 +176,41 @@ def topology_order(model): node2idx = {} for idx, node in enumerate(model.graph.nodes): node2idx[node] = idx - return node2idx \ No newline at end of file + return node2idx + +def get_flattened_qconfig_dict(qconfig_dict): + """ flatten the global, object_type and module_name qconfig + to the same qconfig_dict so that it can be used by + propagate_qconfig_ function. + "module_name_regex" is ignored for now since it's not supported + in propagate_qconfig_, but it can be fixed later. + + For example: + Input: { + "": qconfig, + "object_type": [ + (torch.add, qconfig) + ], + "module_name": [ + ("conv", qconfig) + ] + } + + Output: { + "": qconfig, + torch.add: qconfig, + "conv": qconfig + } + """ + flattened = dict() + if '' in qconfig_dict: + flattened[''] = qconfig_dict[''] + + def flatten_key(key): + if key in qconfig_dict: + for (obj, qconfig) in qconfig_dict[key].items(): + flattened[obj] = qconfig + + flatten_key('object_type') + flatten_key('module_name') + return flattened \ No newline at end of file From 47c3cde14fbddb09362f4d12753f9cd4dfa76dce Mon Sep 17 00:00:00 2001 From: gaoshe Date: Mon, 14 Aug 2023 20:58:57 +0800 Subject: [PATCH 13/29] commit message here --- .../qat_bertbase_questionanswer.py | 477 ++++++++++++++++++ mqbench/convert_deploy.py | 1 + mqbench/deploy/__init__.py | 3 +- 3 files changed, 480 insertions(+), 1 deletion(-) create mode 100644 application/nlp_example/qat_bertbase_questionanswer.py diff --git a/application/nlp_example/qat_bertbase_questionanswer.py b/application/nlp_example/qat_bertbase_questionanswer.py new file mode 100644 index 00000000..e1bebe5f --- /dev/null +++ b/application/nlp_example/qat_bertbase_questionanswer.py @@ -0,0 +1,477 @@ +#瀵煎叆鎵闇鐨勫簱 +import argparse +import transformers +import torch +import torch.nn as nn +import inspect +import unittest +from itertools import chain +from torch.utils.data import DataLoader +from tqdm.auto import tqdm +import numpy as np +from transformers import AdamW, get_scheduler +from transformers import AutoModel +from transformers import AutoTokenizer +from transformers import default_data_collator +from transformers.onnx.features import FeaturesManager +from datasets import load_dataset,load_metric +import torch.optim as optim +from mqbench.convert_deploy import convert_deploy, convert_onnx +from mqbench.prepare_by_platform import prepare_by_platform, BackendType +from mqbench.utils.state import enable_calibration, enable_quantization, disable_all +from transformers import logging +import matplotlib.pyplot as plt +import torch.onnx +import pandas as pd +import json +import logging +import os +import collections +import six +from transformers import DistilBertConfig +from transformers import AutoModelForQuestionAnswering, TrainingArguments, Trainer +from transformers import BertTokenizer, BertModel +from transformers.utils.fx import HFTracer +from transformers import Trainer, TrainingArguments, PreTrainedModel + +parser = argparse.ArgumentParser(description='MQBench bertbase Training') + +parser.add_argument('--epochs', default=1, type=int, metavar='N', + help='number of total epochs to run') +parser.add_argument('--b', '--batch-size', default=16, type=int, + metavar='N', + help='mini-batch size (default: 16), this is the total ') +parser.add_argument('--lr', '--learning-rate', default=2e-5, type=float, + metavar='LR', help='initial learning rate', dest='lr') +parser.add_argument('--wd', '--weight-decay', default=1e-2, type=float, + metavar='W', help='weight decay (default: 1e-4)', + dest='wd') +parser.add_argument('--wbit', default=4, type=int, + metavar='wbit', help='weight bit') +parser.add_argument('--abit', default=8, type=int, + metavar='abit', help='active bit') +parser.add_argument('--wob', default='LSQObserver', type=str, + metavar='wob', help='weight observer') +parser.add_argument('--aob', default='EMAQuantileObserver', type=str, + metavar='aob', help='active observer') +parser.add_argument('--wfq', default='AdaRoundFakeQuantize', type=str, + metavar='wfq', help='weight fakequantize') +parser.add_argument('--afq', default='LearnableFakeQuantize', type=str, + metavar='afq', help='active fakequantize') +parser.add_argument('--backend', type=str, choices=['Academic_NLP', 'Tensorrt_NLP'], default='Academic_NLP') + + +#鍓嶅鐞嗘暟鎹 +def prepare_train_features(examples): + # Some of the questions have lots of whitespace on the left, which is not useful and will make the + # truncation of the context fail (the tokenized question will take a lots of space). So we remove that + # left whitespace + examples["question"] = [q.lstrip() for q in examples["question"]] + + # Tokenize our examples with truncation and padding, but keep the overflows using a stride. This results + # in one example possible giving several features when a context is long, each of those features having a + # context that overlaps a bit the context of the previous feature. + tokenized_examples = tokenizer( + examples["question" if pad_on_right else "context"], + examples["context" if pad_on_right else "question"], + truncation="only_second" if pad_on_right else "only_first", + max_length=max_length, + stride=doc_stride, + return_overflowing_tokens=True, + return_offsets_mapping=True, + padding="max_length", + ) + + # Since one example might give us several features if it has a long context, we need a map from a feature to + # its corresponding example. This key gives us just that. + sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping") + # The offset mappings will give us a map from token to character position in the original context. This will + # help us compute the start_positions and end_positions. + offset_mapping = tokenized_examples.pop("offset_mapping") + + # Let's label those examples! + tokenized_examples["start_positions"] = [] + tokenized_examples["end_positions"] = [] + + for i, offsets in enumerate(offset_mapping): + # We will label impossible answers with the index of the CLS token. + input_ids = tokenized_examples["input_ids"][i] + cls_index = input_ids.index(tokenizer.cls_token_id) + + # Grab the sequence corresponding to that example (to know what is the context and what is the question). + sequence_ids = tokenized_examples.sequence_ids(i) + + # One example can give several spans, this is the index of the example containing this span of text. + sample_index = sample_mapping[i] + answers = examples["answers"][sample_index] + # If no answers are given, set the cls_index as answer. + if len(answers["answer_start"]) == 0: + tokenized_examples["start_positions"].append(cls_index) + tokenized_examples["end_positions"].append(cls_index) + else: + # Start/end character index of the answer in the text. + start_char = answers["answer_start"][0] + end_char = start_char + len(answers["text"][0]) + + # Start token index of the current span in the text. + token_start_index = 0 + while sequence_ids[token_start_index] != (1 if pad_on_right else 0): + token_start_index += 1 + + # End token index of the current span in the text. + token_end_index = len(input_ids) - 1 + while sequence_ids[token_end_index] != (1 if pad_on_right else 0): + token_end_index -= 1 + + # Detect if the answer is out of the span (in which case this feature is labeled with the CLS index). + if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char): + tokenized_examples["start_positions"].append(cls_index) + tokenized_examples["end_positions"].append(cls_index) + else: + # Otherwise move the token_start_index and token_end_index to the two ends of the answer. + # Note: we could go after the last offset if the answer is the last word (edge case). + while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char: + token_start_index += 1 + tokenized_examples["start_positions"].append(token_start_index - 1) + while offsets[token_end_index][1] >= end_char: + token_end_index -= 1 + tokenized_examples["end_positions"].append(token_end_index + 1) + + return tokenized_examples + +def prepare_validation_features(examples): + # Some of the questions have lots of whitespace on the left, which is not useful and will make the + # truncation of the context fail (the tokenized question will take a lots of space). So we remove that + # left whitespace + examples["question"] = [q.lstrip() for q in examples["question"]] + + # Tokenize our examples with truncation and maybe padding, but keep the overflows using a stride. This results + # in one example possible giving several features when a context is long, each of those features having a + # context that overlaps a bit the context of the previous feature. + tokenized_examples = tokenizer( + examples["question" if pad_on_right else "context"], + examples["context" if pad_on_right else "question"], + truncation="only_second" if pad_on_right else "only_first", + max_length=max_length, + stride=doc_stride, + return_overflowing_tokens=True, + return_offsets_mapping=True, + padding="max_length", + ) + + # Since one example might give us several features if it has a long context, we need a map from a feature to + # its corresponding example. This key gives us just that. + sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping") + + # We keep the example_id that gave us this feature and we will store the offset mappings. + tokenized_examples["example_id"] = [] + + for i in range(len(tokenized_examples["input_ids"])): + # Grab the sequence corresponding to that example (to know what is the context and what is the question). + sequence_ids = tokenized_examples.sequence_ids(i) + context_index = 1 if pad_on_right else 0 + + # One example can give several spans, this is the index of the example containing this span of text. + sample_index = sample_mapping[i] + tokenized_examples["example_id"].append(examples["id"][sample_index]) + + # Set to None the offset_mapping that are not part of the context so it's easy to determine if a token + # position is part of the context or not. + tokenized_examples["offset_mapping"][i] = [ + (o if sequence_ids[k] == context_index else None) + for k, o in enumerate(tokenized_examples["offset_mapping"][i]) + ] + + return tokenized_examples +def postprocess_qa_predictions(examples, features, raw_predictions, n_best_size = 20, max_answer_length = 30): + all_start_logits, all_end_logits = raw_predictions + # Build a map example to its corresponding features. + example_id_to_index = {k: i for i, k in enumerate(examples["id"])} + features_per_example = collections.defaultdict(list) + for i, feature in enumerate(features): + features_per_example[example_id_to_index[feature["example_id"]]].append(i) + + # The dictionaries we have to fill. + predictions = collections.OrderedDict() + + # Logging. + print(f"Post-processing {len(examples)} example predictions split into {len(features)} features.") + + # Let's loop over all the examples! + for example_index, example in enumerate(tqdm(examples)): + # Those are the indices of the features associated to the current example. + feature_indices = features_per_example[example_index] + + min_null_score = None # Only used if squad_v2 is True. + valid_answers = [] + + context = example["context"] + # Looping through all the features associated to the current example. + for feature_index in feature_indices: + # We grab the predictions of the model for this feature. + start_logits = all_start_logits[feature_index] + end_logits = all_end_logits[feature_index] + # This is what will allow us to map some the positions in our logits to span of texts in the original + # context. + offset_mapping = features[feature_index]["offset_mapping"] + + # Update minimum null prediction. + cls_index = features[feature_index]["input_ids"].index(tokenizer.cls_token_id) + feature_null_score = start_logits[cls_index] + end_logits[cls_index] + if min_null_score is None or min_null_score < feature_null_score: + min_null_score = feature_null_score + + # Go through all possibilities for the `n_best_size` greater start and end logits. + start_indexes = np.argsort(start_logits)[-1 : -n_best_size - 1 : -1].tolist() + end_indexes = np.argsort(end_logits)[-1 : -n_best_size - 1 : -1].tolist() + for start_index in start_indexes: + for end_index in end_indexes: + # Don't consider out-of-scope answers, either because the indices are out of bounds or correspond + # to part of the input_ids that are not in the context. + if ( + start_index >= len(offset_mapping) + or end_index >= len(offset_mapping) + or offset_mapping[start_index] is None + or offset_mapping[end_index] is None + ): + continue + # Don't consider answers with a length that is either < 0 or > max_answer_length. + if end_index < start_index or end_index - start_index + 1 > max_answer_length: + continue + + start_char = offset_mapping[start_index][0] + end_char = offset_mapping[end_index][1] + valid_answers.append( + { + "score": start_logits[start_index] + end_logits[end_index], + "text": context[start_char: end_char] + } + ) + + if len(valid_answers) > 0: + best_answer = sorted(valid_answers, key=lambda x: x["score"], reverse=True)[0] + else: + # In the very rare edge case we have not a single non-null prediction, we create a fake prediction to avoid + # failure. + best_answer = {"text": "", "score": 0.0} + + # Let's pick our final answer: the best one or the null answer (only for squad_v2) + if not squad_v2: + predictions[example["id"]] = best_answer["text"] + else: + answer = best_answer["text"] if best_answer["score"] > min_null_score else "" + predictions[example["id"]] = answer + + return predictions +def calibrate(cali_loader, model): + model.eval() + print("Start calibration ...") + print("Calibrate data number = ", len(cali_loader)) + with torch.no_grad(): + for i in range(len(cali_loader)): + X= next(iter(cali_loader)) + batch_input =X['input_ids'].to(device) + batch_seg = X['attention_mask'].to(device) + start_logits, end_logits = model(input_ids=batch_input, + attention_mask=batch_seg) + print("Calibration ==> ", i+1) + print("End calibration.") + return + +def prec(datasets,trainer): + validation_features1 = datasets["validation"].map( + prepare_validation_features, + batched=True, + remove_columns=datasets["validation"].column_names + ) + raw_predictions1 = trainer.predict(validation_features1) + + validation_features1.set_format(type=validation_features1.format["type"], columns=list(validation_features1.features.keys())) + examples1 = datasets["validation"] + features1 = validation_features1 + example_id_to_index1 = {k: i for i, k in enumerate(examples1["id"])} + features_per_example1 = collections.defaultdict(list) + for i, feature in enumerate(features1): + features_per_example1[example_id_to_index1[feature["example_id"]]].append(i) + + final_predictions1 = postprocess_qa_predictions(datasets["validation"], validation_features1, raw_predictions1.predictions) + metric = load_metric("squad_v2" if squad_v2 else "squad") + if squad_v2: + formatted_predictions = [{"id": k, "prediction_text": v, "no_answer_probability": 0.0} for k, v in final_predictions1.items()] + else: + formatted_predictions = [{"id": k, "prediction_text": v} for k, v in final_predictions1.items()] + references = [{"id": ex["id"], "answers": ex["answers"]} for ex in datasets["validation"]] + result=metric.compute(predictions=formatted_predictions, references=references) + print(result) + return +################################################################################################################### + +#杈撳叆鍙傛暟 +args = parser.parse_args() +squad_v2 = False +model_checkpoint = "distilbert-base-uncased" +batch_size = args.b + +#瀵煎叆鏁版嵁 +datasets = load_dataset("squad_v2" if squad_v2 else "squad") +#蹇熷垎璇 +tokenizer = AutoTokenizer.from_pretrained(model_checkpoint) +assert isinstance(tokenizer, transformers.PreTrainedTokenizerFast) +#棰勫鐞嗗弬鏁板鍏 +max_length = 384 # The maximum length of a feature (question and context) +doc_stride = 128 # The authorized overlap between two part of the context when splitting it is needed. +pad_on_right = tokenizer.padding_side == "right" +n_best_size = 20 +#瀵硅缁冩暟鎹繘琛屽鐞 +tokenized_datasets = datasets.map(prepare_train_features, batched=True, remove_columns=datasets["train"].column_names) +#璁粌鍙傛暟瀵煎叆 +model = AutoModelForQuestionAnswering.from_pretrained(model_checkpoint) +model_name = model_checkpoint.split("/")[-1] +args1 = TrainingArguments( + f"{model_name}-finetuned-squad", + evaluation_strategy = "epoch", + learning_rate=args.lr, + per_device_train_batch_size=batch_size, + per_device_eval_batch_size=batch_size, + num_train_epochs=1, + weight_decay=args.wd +) +data_collator = default_data_collator + +############################################################################################################### + +#閲忓寲妯″瀷鍙傛暟鍑嗗 +sig = inspect.signature(model.forward) +input_names =['input_ids','token_type_ids','attention_mask'] +#input_names =['input_ids','attention_mask'] +concrete_args = {p.name: p.default for p in sig.parameters.values() if p.name not in input_names} +extra_qconfig_dict={ + 'w_observer': args.wob,#'MinMaxObserver', + 'a_observer': args.aob,#'EMAMinMaxObserver', + 'w_fakequantize':args.wfq, #'FixedFakeQuantize', + 'a_fakequantize':args.afq, # 'LearnableFakeQuantize', + 'w_qscheme': { + 'bit':args.wbit, + 'symmetry':True, + 'per_channel':False, + 'pot_scale': False + }, + 'a_qscheme': { + 'bit':args.abit, + 'symmetry': True, + 'per_channel': False, + 'pot_scale': False + } + } +preserve_attr={'': ['config']} +prepare_custom_config_dict = { + 'concrete_args': concrete_args, + 'preserve_attr': preserve_attr, + #'work_mode':'all_int4_qat', + 'extra_qconfig_dict':extra_qconfig_dict} +#鎻掑叆閲忓寲鑺傜偣 +model_prepared= prepare_by_platform(model, BackendType.Academic_NLP,[], prepare_custom_config_dict, custom_tracer=HFTracer()) + +#鏍″噯 +device = 'cuda' if torch.cuda.is_available() else 'cpu' +cali =[] +for i in range(64): + text=tokenized_datasets["train"][i] + cali.append(text) +cali_loader = DataLoader(cali, batch_size=16, shuffle=True, collate_fn= default_data_collator) +enable_calibration(model_prepared) +model_prepared=model_prepared.to(device) +calibrate(cali_loader, model_prepared) + +#妯″瀷鍚庡鐞 +enable_quantization(model_prepared) +model_prepared.train() +class BertForQuestionAnswering(PreTrainedModel): + """ + 鐢ㄤ簬寤烘ā绫讳技SQuAD杩欐牱鐨勯棶绛旀暟鎹泦 + """ + def __init__(self,config): + super(BertForQuestionAnswering, self).__init__(config) + self.bert = model_prepared + + def forward(self, input_ids, + attention_mask=None, + start_positions=None, + end_positions=None): + bert_output= self.bert( + input_ids=input_ids, + attention_mask=attention_mask) + start_logits=bert_output['start_logits'] + end_logits=bert_output['end_logits'] + start_logits = start_logits.squeeze(-1) + end_logits = end_logits.squeeze(-1) + + if start_positions is not None and end_positions is not None: + # 鐢变簬閮ㄥ垎鎯呭喌涓媠tart/end 浣嶇疆浼氳秴杩囪緭鍏ョ殑闀垮害 + # 锛堜緥濡傝緭鍏ュ簭鍒楃殑鍙兘澶т簬512锛屽苟涓旀纭殑寮濮嬫垨鑰呯粨鏉熺灏卞湪512涔嬪悗锛 + # 閭ｄ箞姝ゆ椂灏辫杩涜鐗规畩澶勭悊 + ignored_index = start_logits.size(1) # 鍙栬緭鍏ュ簭鍒楃殑闀垮害 + start_positions.clamp_(0, ignored_index) + # 濡傛灉姝ｇ‘璧峰浣嶇疆start_positions涓紝瀛樺湪杈撳叆鏍锋湰鐨勫紑濮嬩綅缃ぇ浜庤緭鍏ラ暱搴︼紝 + # 閭ｄ箞鐩存帴鍙栬緭鍏ュ簭鍒楃殑闀垮害浣滀负寮濮嬩綅缃 + end_positions.clamp_(0, ignored_index) + + loss_fct = nn.CrossEntropyLoss(ignore_index=ignored_index) + # 杩欓噷鎸囧畾ignored_index鍏跺疄灏辨槸涓轰簡蹇界暐鎺夎秴杩囪緭鍏ュ簭鍒楅暱搴︾殑锛堣捣濮嬬粨鏉燂級浣嶇疆 + # 鍦ㄩ娴嬫椂鎵甯︽潵鐨勬崯澶憋紝鍥犱负杩欎簺浣嶇疆骞朵笉鑳界畻鏄ā鍨嬮娴嬮敊璇殑锛堝彧鑳界湅鍋氭槸娌℃湁棰勬祴锛夛紝 + # 鍚屾椂濡傛灉涓嶅姞ignore_index鐨勮瘽锛岄偅涔堝彲鑳戒細褰卞搷妯″瀷鍦ㄦ甯告儏鍐典笅鐨勮涔夌悊瑙ｈ兘鍔 + start_loss = loss_fct(start_logits, start_positions) + end_loss = loss_fct(end_logits, end_positions) + return (start_loss + end_loss) / 2, start_logits, end_logits + else: + return start_logits, end_logits +config1 = DistilBertConfig.from_pretrained('distilbert-base-uncased') +# 鍒涘缓鑷畾涔夐厤缃璞 +model_prepared2=BertForQuestionAnswering(config1) +# 鍘熷妯″瀷璁粌 +disable_all(model_prepared2) +model_prepared2=model_prepared2.train() +trainer1 = Trainer( + model_prepared2, + args1, + train_dataset=tokenized_datasets["train"], + eval_dataset=tokenized_datasets["validation"], + data_collator=data_collator, + tokenizer=tokenizer, +) +trainer1.train() +print("鍘熷妯″瀷绮惧害锛") +prec(datasets,trainer1) +print("**************************************************") +# 閲忓寲妯″瀷璁粌 +enable_quantization(model_prepared2) +model_prepared2.train() +trainer2 = Trainer( + model_prepared2, + args1, + train_dataset=tokenized_datasets["train"], + eval_dataset=tokenized_datasets["validation"], + data_collator=data_collator, + tokenizer=tokenizer, +) +trainer2.train() +print("閲忓寲妯″瀷绮惧害锛") +prec(datasets,trainer2) +print("**************************************************") + +#妯″瀷閮ㄧ讲 +keys_to_copy = ['input_ids', 'attention_mask'] +copied_cali=[] +for i in range(len(cali)): + text= {key: cali[i][key] for key in keys_to_copy} + copied_cali.append(text) +cali_loader1 = DataLoader(copied_cali, batch_size=1, shuffle=True, collate_fn= default_data_collator) +X=next(iter(cali_loader1)) +model_prepared.eval() +model_kind, model_onnx_config = FeaturesManager.check_supported_model_or_raise(model_prepared, feature='default') +onnx_config = model_onnx_config(model_prepared.config) +convert_deploy(model_prepared, + BackendType.Academic_NLP, + dummy_input=((dict(X)),), + model_name='bert-base-uncased-mqbench-squad' + ) diff --git a/mqbench/convert_deploy.py b/mqbench/convert_deploy.py index c9ccb180..fcfe3882 100644 --- a/mqbench/convert_deploy.py +++ b/mqbench/convert_deploy.py @@ -127,6 +127,7 @@ def deploy_qparams_tensorrt(model: GraphModule, onnx_model_path, model_name, **k logger.info("Extract qparams for TensorRT.") remove_fakequantize_and_collect_params(onnx_model_path, model_name, backend='tensorrt') + @register_deploy_function(BackendType.Sophgo_TPU) def deploy_qparams_sophgo_tpu(model: GraphModule, onnx_model_path, model_name, **kwargs): logger.info("Extract qparams for sophgo_tpu.") diff --git a/mqbench/deploy/__init__.py b/mqbench/deploy/__init__.py index 65e8c15e..85b29e82 100644 --- a/mqbench/deploy/__init__.py +++ b/mqbench/deploy/__init__.py @@ -4,4 +4,5 @@ from .deploy_onnx_qnn import ONNXQNNPass from .deploy_openvino import replace_fakequantize_and_collect_params_openvino from .deploy_tengine import remove_fakequantize_and_collect_params_tengine -from .deploy_sophgo import remove_fakequantize_and_collect_params_sophgo \ No newline at end of file +from .deploy_sophgo import remove_fakequantize_and_collect_params_sophgo +from .deploy_academicnlp import remove_fakequantize_and_collect_params_academic \ No newline at end of file From 261256252f4385c78dd85f94b05bf4814a610500 Mon Sep 17 00:00:00 2001 From: 18813187393 Date: Tue, 15 Aug 2023 16:58:17 +0800 Subject: [PATCH 14/29] commit message here --- .../nlp_example/qat_bertbase_classication.py | 255 ++++++++++++++++++ 1 file changed, 255 insertions(+) create mode 100644 application/nlp_example/qat_bertbase_classication.py diff --git a/application/nlp_example/qat_bertbase_classication.py b/application/nlp_example/qat_bertbase_classication.py new file mode 100644 index 00000000..221496fb --- /dev/null +++ b/application/nlp_example/qat_bertbase_classication.py @@ -0,0 +1,255 @@ +import torch +import torch.nn as nn +import inspect +import unittest +import argparse +from itertools import chain +from torch.utils.data import DataLoader +from tqdm.auto import tqdm +from transformers import AdamW, get_scheduler +from transformers import AutoModel +from transformers import AutoTokenizer +from transformers import BertTokenizer, BertModel +from transformers.utils.fx import HFTracer +from transformers.onnx.features import FeaturesManager +from datasets import load_dataset +import torch.optim as optim +from mqbench.convert_deploy import convert_deploy, convert_onnx +from mqbench.prepare_by_platform import prepare_by_platform, BackendType +from mqbench.utils.state import enable_calibration, enable_quantization, disable_all +from transformers import logging +import matplotlib.pyplot as plt +import torch.onnx + +parser = argparse.ArgumentParser(description='MQBench bertbase Training') + +parser.add_argument('--epochs', default=1, type=int, metavar='N', + help='number of total epochs to run') +parser.add_argument('--b', '--batch-size', default=4, type=int, + metavar='N', + help='mini-batch size (default: 16), this is the total ') +parser.add_argument('--lr', '--learning-rate', default=1e-5, type=float, + metavar='LR', help='initial learning rate', dest='lr') +parser.add_argument('--wd', '--weight-decay', default=1e-2, type=float, + metavar='W', help='weight decay (default: 1e-4)', + dest='wd') +parser.add_argument('--wbit', default=4, type=int, + metavar='wbit', help='weight bit') +parser.add_argument('--abit', default=8, type=int, + metavar='abit', help='active bit') +parser.add_argument('--wob', default='LSQObserver', type=str, + metavar='wob', help='weight observer') +parser.add_argument('--aob', default='EMAQuantileObserver', type=str, + metavar='aob', help='active observer') +parser.add_argument('--wfq', default='AdaRoundFakeQuantize', type=str, + metavar='wfq', help='weight fakequantize') +parser.add_argument('--afq', default='LearnableFakeQuantize', type=str, + metavar='afq', help='active fakequantize') + +class Dataset(torch.utils.data.Dataset): + def __init__(self, data_type): + self.data = self.load_data(data_type) + + def load_data(self, data_type): + tmp_dataset = load_dataset(path='laugustyniak/abusive-clauses-pl', split = data_type) + Data = {} + for idx, line in enumerate(tmp_dataset): + sample = line + Data[idx] = sample + return Data + + def __len__(self): + return len(self.data) + + def __getitem__(self, idx): + return self.data[idx] +def collote_fn(batch_samples): + batch_text= [] + batch_label = [] + for sample in batch_samples: + batch_text.append(sample['text']) + batch_label.append(int(sample['label'])) + X = tokenizer( + batch_text, + padding=True, + truncation=True, + return_tensors="pt" + ) + y = torch.tensor(batch_label) + return X, y +def train_loop(dataloader, model, loss_fn, optimizer, lr_scheduler, epoch, total_loss): + progress_bar = tqdm(range(len(dataloader))) + progress_bar.set_description(f'loss: {0:>7f}') + finish_batch_num = (epoch-1)*len(dataloader) + + model.train() + for batch, (X, y) in enumerate(dataloader, start=1): + X, y = X.to(device), y.to(device) + pred = model(X) + loss = loss_fn(pred, y) + + optimizer.zero_grad() + loss.backward() + optimizer.step() + lr_scheduler.step() + + total_loss += loss.item() + progress_bar.set_description(f'loss: {total_loss/(finish_batch_num + batch):>7f}') + progress_bar.update(1) + #losses.append(loss.item()) + #print(f"Epoch [{epoch+1}/{epochs}], Loss: {loss.item()}") + return total_loss #losses + +def test_loop(dataloader, model, mode='Test'): + assert mode in ['Valid', 'Test'] + size = len(dataloader.dataset) + correct = 0 + + model.eval() + with torch.no_grad(): + for X, y in dataloader: + X, y = X.to(device), y.to(device) + pred = model(X) + correct += (pred.argmax(1) == y).type(torch.float).sum().item() + + correct /= size + print(f"{mode} Accuracy: {(100*correct):>0.1f}%\n") + return correct +def calibrate(cali_loader, model): + model.eval() + print("Start calibration ...") + print("Calibrate data number = ", len(cali_loader.dataset)) + with torch.no_grad(): + for i, (X, y) in enumerate(cali_loader): + X, y = X.to(device), y.to(device) + pred = model(X) + print("Calibration ==> ", i+1) + print("End calibration.") + return + +################################################################################################################## + +args = parser.parse_args() +#load data +train_data = Dataset('train') +test_data = Dataset('test') + +#load parameters +checkpoint = "bert-base-uncased" +tokenizer = AutoTokenizer.from_pretrained(checkpoint) +device = 'cuda' if torch.cuda.is_available() else 'cpu' +print(f'Using {device} device') +learning_rate = args.lr +epoch_num = args.epochs + +#dataloader +train_dataloader = DataLoader(train_data, batch_size=args.b, shuffle=True, collate_fn=collote_fn) +test_dataloader = DataLoader(test_data, batch_size=args.b, shuffle=True, collate_fn=collote_fn) + +#quantize +model1=AutoModel.from_pretrained(checkpoint) +#閲忓寲妯″瀷鍙傛暟鍑嗗 +sig = inspect.signature(model1.forward) +input_names =['input_ids','token_type_ids','attention_mask'] +concrete_args = {p.name: p.default for p in sig.parameters.values() if p.name not in input_names} +extra_qconfig_dict={ + 'w_observer': args.wob,#'MinMaxObserver', + 'a_observer': args.aob,#'EMAMinMaxObserver', + 'w_fakequantize':args.wfq, #'FixedFakeQuantize', + 'a_fakequantize':args.afq, # 'LearnableFakeQuantize', + 'w_qscheme': { + 'bit':args.wbit, + 'symmetry':True, + 'per_channel': False, + 'pot_scale': False + }, + 'a_qscheme': { + 'bit':args.abit, + 'symmetry': True, + 'per_channel': False, + 'pot_scale': False + } + } +preserve_attr={'': ['config']} +prepare_custom_config_dict = { + 'concrete_args': concrete_args, + 'preserve_attr': preserve_attr, + #'work_mode':'all_int4_qat', + 'extra_qconfig_dict':extra_qconfig_dict} +#鎻掑叆閲忓寲鑺傜偣 +model_prepared= prepare_by_platform(model1, BackendType.Academic_NLP,[], prepare_custom_config_dict, custom_tracer=HFTracer()) +#鍚庡鐞 +class NeuralNetwork2(nn.Module): + def __init__(self): + super(NeuralNetwork2, self).__init__() + self.bert_encoder = model_prepared + self.classifier = nn.Linear(768, 2) + + def forward(self, x): + bert_output = self.bert_encoder(**x) + cls_vectors = bert_output['last_hidden_state'][:, 0] + logits = self.classifier(cls_vectors) + return logits +model_prepared1 = NeuralNetwork2().to(device) +#鏍″噯 +cali =[] +for i in range(20): + text=train_data[i] + cali.append(text) +cali_loader = DataLoader(cali, batch_size=args.b, shuffle=True, collate_fn=collote_fn) +enable_calibration(model_prepared1) +model_prepared1=model_prepared1.to(device) +calibrate(cali_loader, model_prepared1) +#鍘熷妯″瀷绮惧害 +disable_all(model_prepared1) +model_prepared1=model_prepared1.train() +optimizer = AdamW(model_prepared1.parameters(), lr=learning_rate) +lr_scheduler = get_scheduler( + "linear", + optimizer=optimizer, + num_warmup_steps=0, + num_training_steps=epoch_num*len(train_dataloader), +) +loss_fn = nn.CrossEntropyLoss() +total_loss = 0 +best_acc = 0 +for t in range(epoch_num): + print(f"Epoch {t+1}/{epoch_num}\n-------------------------------") + total_loss = train_loop(train_dataloader, model_prepared1, loss_fn, optimizer, lr_scheduler, t+1, total_loss) + Test_acc = test_loop(test_dataloader,model_prepared1, mode='Test') + if Test_acc > best_acc: + best_acc = Test_acc + print('saving new weights...\n') +print("Done!") + +#閲忓寲妯″瀷绮惧害 +enable_quantization(model_prepared1) +model_prepared1=model_prepared1.train() +total_loss = 0 +best_acc = 0 +lr_scheduler1 = get_scheduler( + "linear", + optimizer=optimizer, + num_warmup_steps=0, + num_training_steps=epoch_num*len(train_dataloader), +) +for t in range(epoch_num): + print(f"Epoch {t+1}/{epoch_num}\n-------------------------------") + total_loss = train_loop(train_dataloader, model_prepared1, loss_fn, optimizer, lr_scheduler1, t+1, total_loss) + Test_acc = test_loop(test_dataloader,model_prepared1, mode='Test') + if Test_acc > best_acc: + best_acc = Test_acc + print('saving new weights...\n') +print("Done!") + +#閲忓寲妯″瀷閮ㄧ讲 +train_dataloader1 = DataLoader(train_data, batch_size=1, shuffle=True, collate_fn=collote_fn) +batch_X1, batch_y1 = next(iter(train_dataloader1)) +model_prepared.eval() +model_kind, model_onnx_config = FeaturesManager.check_supported_model_or_raise(model_prepared, feature='default') +onnx_config = model_onnx_config(model_prepared.config) +convert_deploy(model_prepared, + BackendType.Academic_NLP, + dummy_input=(dict(batch_X1),), + model_name='bert-base-uncased-mqbench' + ) \ No newline at end of file From 4642917ad889aaff8976e41ee054044854828500 Mon Sep 17 00:00:00 2001 From: 18813187393 Date: Tue, 29 Aug 2023 19:17:47 +0800 Subject: [PATCH 15/29] qat gpt2 --- application/nlp_example/gpt2qat.py | 529 +++++++++++++++++++++++++++++ 1 file changed, 529 insertions(+) create mode 100644 application/nlp_example/gpt2qat.py diff --git a/application/nlp_example/gpt2qat.py b/application/nlp_example/gpt2qat.py new file mode 100644 index 00000000..b52cdcf9 --- /dev/null +++ b/application/nlp_example/gpt2qat.py @@ -0,0 +1,529 @@ +import torch +import torch.nn as nn +import numpy as np +import random +import inspect +import argparse +import unittest +import transformers +from cleantext import clean +from nltk.corpus import stopwords +from nltk.tokenize import word_tokenize +import pandas as pd +import numpy as np +import random +import datetime +import time +from torch.utils.data import Dataset, DataLoader, random_split, RandomSampler, SequentialSampler +torch.manual_seed(42) +from transformers import GPT2LMHeadModel, GPT2Tokenizer, GPT2Config, GPT2LMHeadModel +from transformers import GPT2PreTrainedModel, GPT2Tokenizer +from transformers import AdamW, get_linear_schedule_with_warmup +from tqdm import tqdm, trange +from itertools import chain +from tqdm.auto import tqdm +from transformers import AdamW, get_scheduler +from transformers import AutoModel +from transformers import AutoTokenizer +from transformers import default_data_collator +from transformers.onnx.features import FeaturesManager +from datasets import load_dataset,load_metric +import torch.optim as optim +from mqbench.convert_deploy import convert_deploy, convert_onnx +from mqbench.prepare_by_platform import prepare_by_platform, BackendType +from mqbench.utils.state import enable_calibration, enable_quantization, disable_all +from transformers import logging +import torch.onnx +import logging +import os +import collections +import torch.nn.functional as F +import csv +from torch.nn import CrossEntropyLoss +from torch.nn.parallel import DataParallel +from transformers import AutoModelForQuestionAnswering, TrainingArguments, Trainer +from transformers import GPT2LMHeadModel, GPT2Tokenizer +from transformers import GPT2Tokenizer, GPT2LMHeadModel, AdamW, get_linear_schedule_with_warmup +from transformers import BertTokenizer, BertModel +from transformers.utils.fx import HFTracer + +parser = argparse.ArgumentParser(description='MQBench gpt2 Training') + +parser.add_argument('--epochs', default=4, type=int, metavar='N', + help='number of total epochs to run') +parser.add_argument('--b', '--batch-size', default=1, type=int, + metavar='N', + help='mini-batch size (default: 16), this is the total ') +parser.add_argument('--lr', '--learning-rate', default=2e-4, type=float, + metavar='LR', help='initial learning rate', dest='lr') +parser.add_argument('--wd', '--weight-decay', default=1e-2, type=float, + metavar='W', help='weight decay (default: 1e-4)', + dest='wd') +parser.add_argument('--wbit', default=4, type=int, + metavar='wbit', help='weight bit') +parser.add_argument('--abit', default=8, type=int, + metavar='abit', help='active bit') +parser.add_argument('--wob', default='LSQObserver', type=str, + metavar='wob', help='weight observer') +parser.add_argument('--aob', default='EMAQuantileObserver', type=str, + metavar='aob', help='active observer') +parser.add_argument('--wfq', default='LearnableFakeQuantize', type=str, + metavar='wfq', help='weight fakequantize') +parser.add_argument('--afq', default='LearnableFakeQuantize', type=str, + metavar='afq', help='active fakequantize') +#clean data +def cleaning(text,punct): + cleaned_text = clean(text, + fix_unicode=False, # fix various unicode errors + to_ascii=False, # transliterate to closest ASCII representation + lower=True, # lowercase text + no_line_breaks=False, # fully strip line breaks as opposed to only normalizing them + no_urls=False, # replace all URLs with a special token + no_emails=False, # replace all email addresses with a special token + no_phone_numbers=False, # replace all phone numbers with a special token + no_numbers=False, # replace all numbers with a special token + no_digits=False, # replace all digits with a special token + no_currency_symbols=False, # replace all currency symbols with a special token + no_punct=punct, # remove punctuations + lang="en" # set to 'de' for German special handling + ) + + tokens = word_tokenize(cleaned_text) + filtered_sentence = [w for w in tokens if not w in stopwords.words('english')] + cleaned_text_0 = ' '.join(filtered_sentence) + return cleaned_text_0 +#train +def train(model,epochs,optimizer,scheduler,train_dataloader,validation_dataloader): + + total_t0 = time.time() + training_stats = [] + model = model.to(device) + for epoch_i in range(0, epochs): + print("") + print('======== Epoch {:} / {:} ========'.format(epoch_i + 1, epochs)) + print('Training...') + t0 = time.time() + total_train_loss = 0 + model.train() + for step, batch in enumerate(train_dataloader): + b_input_ids = batch[0].to(device) + b_labels = batch[0].to(device) + b_masks = batch[1].to(device) + model.zero_grad() + outputs = model(b_input_ids, + attention_mask = b_masks, + labels=b_labels + ) + loss = outputs[0] + batch_loss = loss.item() + total_train_loss += batch_loss + loss.backward() + optimizer.step() + scheduler.step() + # Calculate the average loss over all of the batches. + avg_train_loss = total_train_loss / len(train_dataloader) + # Measure how long this epoch took. + training_time = format_time(time.time() - t0) + print("") + print(" Average training loss: {0:.2f}".format(avg_train_loss)) + print(" Training epoch took: {:}".format(training_time)) + print("") + print("Running Validation...") + + t0 = time.time() + model.eval() + total_eval_loss = 0 + nb_eval_steps = 0 + + # Evaluate data for one epoch + for batch in validation_dataloader: + b_input_ids = batch[0].to(device) + b_labels = batch[0].to(device) + b_masks = batch[1].to(device) + with torch.no_grad(): + outputs = model(b_input_ids, + # token_type_ids=None, + attention_mask = b_masks, + labels=b_labels) + loss = outputs[0] + batch_loss = loss.item() + total_eval_loss += batch_loss + avg_val_loss = total_eval_loss / len(validation_dataloader) + validation_time = format_time(time.time() - t0) + print(" Validation Loss: {0:.2f}".format(avg_val_loss)) + print(" Validation took: {:}".format(validation_time)) + # Record all statistics from this epoch. + training_stats.append( + { + 'epoch': epoch_i + 1, + 'Training Loss': avg_train_loss, + 'Valid. Loss': avg_val_loss, + 'Training Time': training_time, + 'Validation Time': validation_time + } + ) + print("") + print("Training complete!") + print("Total training took {:} (h:mm:ss)".format(format_time(time.time()-total_t0))) + return model,training_stats +#quant train +def train1(model,epochs,optimizer,scheduler,train_dataloader,validation_dataloader): + + total_t0 = time.time() + training_stats = [] + model = model.to(device) + for epoch_i in range(0, epochs): + print("") + print('======== Epoch {:} / {:} ========'.format(epoch_i + 1, epochs)) + print('Training...') + t0 = time.time() + total_train_loss = 0 + model.train() + for step, batch in enumerate(train_dataloader): + b_input_ids = batch[0].to(device) + b_labels = batch[0].to(device) + b_masks = batch[1].to(device) + model.zero_grad() + outputs = model(b_input_ids, + attention_mask = b_masks, + labels=b_labels + ) + loss = outputs[0] + batch_loss = loss.item() + total_train_loss += batch_loss + loss.backward() + optimizer.step() + scheduler.step() + # Calculate the average loss over all of the batches. + avg_train_loss = total_train_loss / len(train_dataloader) + # Measure how long this epoch took. + training_time = format_time(time.time() - t0) + print("") + print(" Average training loss: {0:.2f}".format(avg_train_loss)) + print(" Training epoch took: {:}".format(training_time)) + print("") + print("Running Validation...") + + t0 = time.time() + model.eval() + total_eval_loss = 0 + nb_eval_steps = 0 + + # Evaluate data for one epoch + for batch in validation_dataloader: + b_input_ids = batch[0].to(device) + b_labels = batch[0].to(device) + b_masks = batch[1].to(device) + with torch.no_grad(): + outputs = model(b_input_ids, + # token_type_ids=None, + attention_mask = b_masks, + labels=b_labels) + loss = outputs[0] + batch_loss = loss.item() + total_eval_loss += batch_loss + avg_val_loss = total_eval_loss / len(validation_dataloader) + validation_time = format_time(time.time() - t0) + print(" Validation Loss: {0:.2f}".format(avg_val_loss)) + print(" Validation took: {:}".format(validation_time)) + # Record all statistics from this epoch. + training_stats.append( + { + 'epoch': epoch_i + 1, + 'Training Loss': avg_train_loss, + 'Valid. Loss': avg_val_loss, + 'Training Time': training_time, + 'Validation Time': validation_time + } + ) + print("") + print("Training complete!") + print("Total training took {:} (h:mm:ss)".format(format_time(time.time()-total_t0))) + return model,training_stats +def cal_ppl_bygpt2(model,test_dataloader): + total_ppl=0 + model.eval() + with torch.no_grad(): + for step, batch in enumerate(test_dataloader): + b_input_ids = batch[0].to(device) + b_masks = batch[1].to(device) + b_labels=b_input_ids + outputs = model(b_input_ids, + attention_mask = b_masks, + labels=b_labels) + bs, sl = b_input_ids.size() + logits = outputs[1] + # Shift so that tokens < n predict n + shift_logits = logits[:, :-1, :].contiguous() + shift_labels = b_input_ids[:, 1:].contiguous() + shift_attentions = b_masks[:, 1:].contiguous() + # Flatten the tokens + loss_fct = CrossEntropyLoss(ignore_index=0, reduction="none") + loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1)).detach().reshape(bs, -1) + meanloss = loss.sum(1) / shift_attentions.sum(1) + ppl = torch.exp(meanloss).cpu() + ppl=ppl.numpy().tolist() + total_ppl+=ppl[0] + avg_ppl=total_ppl/len(test_dataloader) + return avg_ppl +def cal_ppl_bygpt22(model,test_dataloader): + total_ppl=0 + model.eval() + with torch.no_grad(): + for step, batch in enumerate(test_dataloader): + b_input_ids = batch[0].to(device) + b_masks = batch[1].to(device) + b_labels=b_input_ids + outputs = model(b_input_ids, + attention_mask = b_masks, + labels=b_labels) + bs, sl = b_input_ids.size() + logits = outputs[1] + # Shift so that tokens < n predict n + shift_logits = logits[:, :-1, :].contiguous() + shift_labels = b_input_ids[:, 1:].contiguous() + shift_attentions = b_masks[:, 1:].contiguous() + # Flatten the tokens + loss_fct = CrossEntropyLoss(ignore_index=0, reduction="none") + loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1)).detach().reshape(bs, -1) + meanloss = loss.sum(1) / shift_attentions.sum(1) + ppl = torch.exp(meanloss).cpu() + ppl=ppl.numpy().tolist() + total_ppl+=ppl[0] + avg_ppl=total_ppl/len(test_dataloader) + return avg_ppl +def calibrate(cali_loader, model): + model.eval() + print("Start calibration ...") + print("Calibrate data number = ", len(cali_loader)) + with torch.no_grad(): + for step, batch in enumerate(cali_loader): + b_input_ids = batch[0].to(device) + b_masks = batch[1].to(device) + b_labels=b_input_ids + outputs = model(b_input_ids, + attention_mask = b_masks, + labels=b_labels) + print("Calibration ==> ", step+1) + print("End calibration.") + return +class GPT2Dataset(Dataset): + + def __init__(self, txt_list, tokenizer, gpt2_type="gpt2", max_length=1024): + + self.tokenizer = tokenizer + self.input_ids = [] + self.attn_masks = [] + + for txt in txt_list: + encodings_dict = tokenizer( '<|startoftext|>'+ txt + '<|endoftext|>', truncation=True, max_length=max_length,padding="max_length") #, padding="max_length" + self.input_ids.append(torch.tensor(encodings_dict['input_ids'])) + self.attn_masks.append(torch.tensor(encodings_dict['attention_mask'])) + + def __len__(self): + return len(self.input_ids) + + def __getitem__(self, idx): + return self.input_ids[idx], self.attn_masks[idx] + +###################################################################################################################### + +args = parser.parse_args() +#load parameters +batch_size =args.b +epochs = args.epochs +learning_rate = args.lr +warmup_steps = 1e2 +epsilon = 1e-8 +#load data +squad = load_dataset("squad", split="train[:10000]") +#processing data +que = [] +con = [] +ans = [] +for i in squad: + que.append(cleaning(i['question'],True)) + con.append(cleaning(i['context'],True)) + ans.append(cleaning(i['answers']['text'],True)) +main_data = pd.DataFrame() +main_data['Question'] = que +main_data['Context'] = con +main_data['Answer'] = ans +main_data['train_data'] = main_data['Question']+''+main_data['Context']+''+main_data['Answer'] +main_data = main_data.sample(frac=1,random_state=32).reset_index(drop=True) +train_data = main_data[0:9000].reset_index(drop=True) +cali_data=train_data['train_data'][0:100].reset_index(drop=True) +test_data = main_data[9000:10000].reset_index(drop=True) +validation = train_data['train_data'][8500:9000] +validation1 = validation.apply( lambda x: x.split('')[0]+'') +validation1.reset_index(drop=True,inplace=True) +#save data +train_data.to_csv('training_data.csv',index=False) +test_data.to_csv('testing_data.csv',index=False) +validation1.to_csv('validation_data.csv',index=False) +#load GPT tokenizer +tokenizer = GPT2Tokenizer.from_pretrained('gpt2',bos_token='<|startoftext|>', eos_token='<|endoftext|>', pad_token='<|pad|>') #gpt2-medium +#Building GPT dataset +traindataset = GPT2Dataset(train_data['train_data'][0:8500], tokenizer) +validdataset = GPT2Dataset(validation1, tokenizer) +cali_loader=GPT2Dataset(cali_data, tokenizer) +testdataset=GPT2Dataset(test_data['train_data'], tokenizer) +#Generate Text Collection +test_set = pd.DataFrame() +test_set['train_data']=test_data['train_data'][:500] +test_set['True_end_train_data1'] = test_set['train_data'].str.split().str[-20:].apply(' '.join) +test_set['train_data1'] = test_set['train_data'].str.split().str[:-20].apply(' '.join) +#dataloader +train_dataloader = DataLoader( + traindataset, # The training samples. + sampler = RandomSampler(traindataset), # Select batches randomly + batch_size = batch_size # Trains with this batch size. + ) + +# For validation the order doesn't matter, so we'll just read them sequentially. +validation_dataloader = DataLoader( + validdataset, # The validation samples. + sampler = SequentialSampler(validdataset), # Pull out batches sequentially. + batch_size = batch_size # Evaluate with this batch size. + ) +cali_loader = DataLoader( + cali_loader, + sampler = SequentialSampler(cali_loader), + batch_size = 2 + ) +test_dataloader = DataLoader( + testdataset, + sampler = RandomSampler(testdataset), + batch_size = batch_size + ) +#load model +configuration = GPT2Config.from_pretrained('gpt2',resid_pdrop = 0.3 , output_hidden_states=False) +model = GPT2LMHeadModel.from_pretrained("gpt2", config=configuration) +model.resize_token_embeddings(len(tokenizer)) +device = 'cuda' if torch.cuda.is_available() else 'cpu' +model=model.to(device) +seed_val = 32 +random.seed(seed_val) +np.random.seed(seed_val) +torch.manual_seed(seed_val) +torch.cuda.manual_seed_all(seed_val) +#quantize +sig = inspect.signature(model.forward) +input_names =['input_ids','attention_mask'] +concrete_args = {p.name: p.default for p in sig.parameters.values() if p.name not in input_names} +extra_qconfig_dict={ + 'w_observer': args.wob,#'MinMaxObserver', + 'a_observer': args.aob,#'EMAMinMaxObserver', + 'w_fakequantize':args.wfq, #'FixedFakeQuantize', + 'a_fakequantize':args.afq, # 'LearnableFakeQuantize', + 'w_qscheme': { + 'bit':args.wbit, + 'symmetry':True, + 'per_channel':False, + 'pot_scale': False + }, + 'a_qscheme': { + 'bit':args.abit, + 'symmetry': True, + 'per_channel': False, + 'pot_scale': False + } + } +preserve_attr={'': ['config']} +prepare_custom_config_dict = { + 'concrete_args': concrete_args, + 'preserve_attr': preserve_attr, + #'work_mode':'all_int4_qat', + 'extra_qconfig_dict':extra_qconfig_dict} +#Insert quantization node +model_prepared= prepare_by_platform(model, BackendType.Academic_NLP,[], prepare_custom_config_dict, custom_tracer=HFTracer()) + +#Post processing +class Quantizegpt2(GPT2PreTrainedModel): + """ + 鐢ㄤ簬寤烘ā绫讳技SQuAD杩欐牱鐨勯棶绛旀暟鎹泦 + """ + def __init__(self,config): + super(Quantizegpt2, self).__init__(config) + self.gpt2 = model_prepared + + def forward(self, input_ids,attention_mask,labels=None): + + gpt2_output= self.gpt2(input_ids=input_ids,attention_mask=attention_mask) + lm_logits = gpt2_output['logits'] + loss = None + if labels is not None: + shift_logits = lm_logits[..., :-1, :].contiguous() + shift_labels = labels[..., 1:].contiguous() + loss_fct =nn.CrossEntropyLoss() + loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1)) + return loss,lm_logits +configuration = GPT2Config.from_pretrained('gpt2',resid_pdrop = 0.3 , output_hidden_states=False) +model_prepared1=Quantizegpt2(config=configuration) + +#train parameters +optimizer = AdamW(model_prepared1.parameters(), + lr = learning_rate, + eps = epsilon + ) +total_steps = len(train_dataloader) * epochs + +scheduler = get_linear_schedule_with_warmup(optimizer, + num_warmup_steps = warmup_steps, + num_training_steps = total_steps) +scheduler1 = get_linear_schedule_with_warmup(optimizer, + num_warmup_steps = warmup_steps, + num_training_steps = total_steps) +def format_time(elapsed): + return str(datetime.timedelta(seconds=int(round((elapsed))))) + +#Original model training +disable_all(model_prepared1) +device = 'cuda' if torch.cuda.is_available() else 'cpu' +#model_prepared1=DataParallel(model_prepared1) +model_prepared1=model_prepared1.train() +model_prepared2,training_stats1=train(model_prepared1,epochs,optimizer,scheduler,train_dataloader,validation_dataloader) + +# Display floats with two decimal places. +pd.set_option('precision', 2) +# Create a DataFrame from our training statistics. +df_stats1 = pd.DataFrame(data=training_stats1) +# Use the 'epoch' as the row index. +df_stats1 = df_stats1.set_index('epoch') +# Display the table. +print(df_stats1) + +#Original model PPL +avg_ppl1=cal_ppl_bygpt2(model_prepared2,test_dataloader) +print("鍘熷妯″瀷PPL:{}".format(avg_ppl1)) + +#calibration +device = 'cuda' if torch.cuda.is_available() else 'cpu' +enable_calibration(model_prepared1) +model_prepared1=model_prepared1.to(device) +calibrate(cali_loader, model_prepared1) + +#quantize model train +enable_quantization(model_prepared1) +model_prepared1=model_prepared1.train() +model_prepared3,training_stats2=train1(model_prepared1,epochs,optimizer,scheduler1,train_dataloader,validation_dataloader) + +# Display floats with two decimal places. +pd.set_option('precision', 2) +# Create a DataFrame from our training statistics. +df_stats2 = pd.DataFrame(data=training_stats2) +# Use the 'epoch' as the row index. +df_stats2 = df_stats2.set_index('epoch') +# Display the table. +print(df_stats2) + +#quantize model PPL +avg_ppl2=cal_ppl_bygpt22(model_prepared3,test_dataloader) +print("閲忓寲妯″瀷PPL:{}".format(avg_ppl2)) + + + + + + + From 99f2a1641cf2b7f1588a5a6e628caf59ffcb7fb5 Mon Sep 17 00:00:00 2001 From: tjthereal Date: Wed, 30 Aug 2023 13:51:24 +0800 Subject: [PATCH 16/29] =?UTF-8?q?=E6=B7=BB=E5=8A=A0=E4=BA=86FP8=20fakequan?= =?UTF-8?q?t=E4=BB=A5=E5=8F=8A=E4=BF=AE=E6=94=B9=E4=BA=86config=E4=BB=A5?= =?UTF-8?q?=E5=8F=8Aprepare=5Fby=5Fplatform=E4=B8=AD=E7=9A=84=E4=B8=80?= =?UTF-8?q?=E4=BA=9B=E9=97=AE=E9=A2=98=E2=80=9D?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- FP8_Emulator/cmodel/__init__.py | 1 + FP8_Emulator/cmodel/simple.py | 219 ++ FP8_Emulator/cmodel/simple/simple_conv2d.cpp | 127 ++ .../cmodel/simple/simple_conv2d_impl.cpp | 977 ++++++++ FP8_Emulator/cmodel/simple/simple_gemm.cpp | 71 + .../cmodel/simple/simple_gemm_impl.cpp | 233 ++ .../cmodel/simple/simple_mm_engine.cpp | 45 + FP8_Emulator/cmodel/simple/vla.h | 47 + FP8_Emulator/cmodel/tests/conv_grad_test.py | 58 + FP8_Emulator/cmodel/tests/conv_test.py | 30 + FP8_Emulator/cmodel/tests/gemm_grad_test.py | 43 + .../cmodel/tests/gemm_irregular_test.py | 42 + FP8_Emulator/cmodel/tests/gemm_test.py | 41 + FP8_Emulator/cmodel/tests/linear_test.py | 65 + FP8_Emulator/cmodel/tests/net.py | 73 + FP8_Emulator/pytquant/__init__.py | 6 + FP8_Emulator/pytquant/cpp/__init__.py | 1 + FP8_Emulator/pytquant/cpp/avx-fpemu.cpp | 802 +++++++ FP8_Emulator/pytquant/cpp/fpemu.py | 73 + FP8_Emulator/pytquant/cpp/fpemu_impl.cpp | 2009 +++++++++++++++++ FP8_Emulator/pytquant/cuda/__init__.py | 3 + FP8_Emulator/pytquant/cuda/fpemu.py | 74 + FP8_Emulator/pytquant/cuda/fpemu_impl.cpp | 54 + FP8_Emulator/pytquant/cuda/fpemu_kernels.cu | 1381 +++++++++++ FP8_Emulator/pytquant/test.py | 151 ++ READMEFP8.md | 82 + application/nlp_example/config.yaml | 12 +- application/nlp_example/glue_utils.py | 3 + application/nlp_example/ptq.py | 15 +- fpemu_cpp.cpython-38-x86_64-linux-gnu.so | Bin 0 -> 9341480 bytes fpemu_cuda.cpython-38-x86_64-linux-gnu.so | Bin 0 -> 11085912 bytes mqbench/custom_quantizer/model_quantizer.py | 3 + mqbench/deploy/__init__.py | 2 +- mqbench/fake_quantize/__init__.py | 4 +- mqbench/fake_quantize/e4m3.py | 217 ++ mqbench/fake_quantize/e5m2.py | 219 ++ mqbench/prepare_by_platform.py | 18 +- setup.py | 31 +- 38 files changed, 7212 insertions(+), 20 deletions(-) create mode 100644 FP8_Emulator/cmodel/__init__.py create mode 100644 FP8_Emulator/cmodel/simple.py create mode 100644 FP8_Emulator/cmodel/simple/simple_conv2d.cpp create mode 100644 FP8_Emulator/cmodel/simple/simple_conv2d_impl.cpp create mode 100644 FP8_Emulator/cmodel/simple/simple_gemm.cpp create mode 100644 FP8_Emulator/cmodel/simple/simple_gemm_impl.cpp create mode 100644 FP8_Emulator/cmodel/simple/simple_mm_engine.cpp create mode 100644 FP8_Emulator/cmodel/simple/vla.h create mode 100644 FP8_Emulator/cmodel/tests/conv_grad_test.py create mode 100644 FP8_Emulator/cmodel/tests/conv_test.py create mode 100644 FP8_Emulator/cmodel/tests/gemm_grad_test.py create mode 100644 FP8_Emulator/cmodel/tests/gemm_irregular_test.py create mode 100644 FP8_Emulator/cmodel/tests/gemm_test.py create mode 100644 FP8_Emulator/cmodel/tests/linear_test.py create mode 100644 FP8_Emulator/cmodel/tests/net.py create mode 100644 FP8_Emulator/pytquant/__init__.py create mode 100644 FP8_Emulator/pytquant/cpp/__init__.py create mode 100644 FP8_Emulator/pytquant/cpp/avx-fpemu.cpp create mode 100644 FP8_Emulator/pytquant/cpp/fpemu.py create mode 100644 FP8_Emulator/pytquant/cpp/fpemu_impl.cpp create mode 100644 FP8_Emulator/pytquant/cuda/__init__.py create mode 100644 FP8_Emulator/pytquant/cuda/fpemu.py create mode 100644 FP8_Emulator/pytquant/cuda/fpemu_impl.cpp create mode 100644 FP8_Emulator/pytquant/cuda/fpemu_kernels.cu create mode 100644 FP8_Emulator/pytquant/test.py create mode 100644 READMEFP8.md create mode 100755 fpemu_cpp.cpython-38-x86_64-linux-gnu.so create mode 100755 fpemu_cuda.cpython-38-x86_64-linux-gnu.so create mode 100644 mqbench/fake_quantize/e4m3.py create mode 100644 mqbench/fake_quantize/e5m2.py diff --git a/FP8_Emulator/cmodel/__init__.py b/FP8_Emulator/cmodel/__init__.py new file mode 100644 index 00000000..9809e350 --- /dev/null +++ b/FP8_Emulator/cmodel/__init__.py @@ -0,0 +1 @@ +from . import simple diff --git a/FP8_Emulator/cmodel/simple.py b/FP8_Emulator/cmodel/simple.py new file mode 100644 index 00000000..608de9d4 --- /dev/null +++ b/FP8_Emulator/cmodel/simple.py @@ -0,0 +1,219 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +from torch import nn +from torch.autograd import Function +import sys + +import simple_gemm_dev +import simple_conv2d_dev +# backup the original torch functions +fallback_addmm = torch.addmm +fallback_matmul = torch.matmul +fallback_mm = torch.mm + +def is_transposed(input): + if input.is_contiguous(): + return input, False + elif input.t().is_contiguous(): + return input.t(), True + else: + return input.contiguous(), False + +def addmm(input, mat1, mat2, beta=1.0, alpha=1.0, out=None): + if input.dtype == torch.float32 and mat1.dtype == torch.float32 and \ + mat1.dim() == 2 and mat2.dim() == 2 and mat1.size(1) == mat2.size(0): + if out: + output = out + else: + output = torch.zeros([mat1.size(0), mat2.size(1)]) + a_mat, a_trans = is_transposed(mat1) + b_mat, b_trans = is_transposed(mat2) + output = SimpleAddmm.apply(output, input, a_mat, b_mat, alpha, beta, a_trans, b_trans) + ret = output + else: + warnings.warn('simple.addmm does not support the input dimensions - input :{}, mat1: {}, mat2: {}, falling back to torch.addmm'.format( + input.size(), mat1.size(), mat2.size())) + ret = fallback_addmm(input, mat1, mat2, beta=beta, alpha=alpha, out=out) + return ret + +def matmul(input, other, out=None): + if input.dtype == torch.float32 and other.dtype == torch.float32 and \ + input.dim() == 2 and other.dim() == 2 and input.size(1) == other.size(0): + if out: + output = out + else: + output = torch.zeros([input.size(0), other.size(1)]) + a_mat, a_trans = is_transposed(input) + b_mat, b_trans = is_transposed(other) + output = SimpleMatmul.apply(output, a_mat, b_mat, 1.0, a_trans, b_trans) + return output + # Batch MatMul implementation + elif input.dtype == torch.float32 and other.dtype == torch.float32 and \ + input.dim() == 3 and other.dim() == 2 and input.size(2) == other.size(0): + if out: + output = out + else: + output = torch.zeros([input.size(0), input.size(1), other.size(1)]) + a_mat, a_trans = is_transposed(input) + b_mat, b_trans = is_transposed(other) + output = torch.stack(tuple([SimpleMatmul.apply(out1, a_mat1, b_mat, 1.0, a_trans, b_trans) \ + for a_mat1, out1 in zip(a_mat, output)])) + return output + else: + warnings.warn('simple.matmul does not support the input dimensions - input :{}, other: {}, falling back to torch.matmul'.format( + input.size(), other.size())) + return fallback_matmul(input, other, out=out) + +def mm(input, mat2, out=None): + if input.dtype == torch.float32 and mat2.dtype == torch.float32 and \ + input.dim() == 2 and mat2.dim() == 2 and input.size(1) == mat2.size(0): + if out: + output = out + else: + output = torch.zeros([input.size(0), mat2.size(1)]) + a_mat, a_trans = is_transposed(input) + b_mat, b_trans = is_transposed(mat2) + output = SimpleMatmul.apply(output, a_mat, b_mat, 1.0, a_trans, b_trans) + return output + else: + warnings.warn('simple.mm does not support the input dimensions - input :{}, mat2: {}, falling back to torch.mm'.format( + input.size(), mat2.size())) + return fallback_mm(input, mat2, out=out) + +def conv2d(input, weight, bias=None, stride=1, padding=0, dilation=1, groups=1): + N = input.size()[0] + C = input.size()[1] + H = input.size()[2] + W = input.size()[3] + K = weight.size()[0] + C1 = weight.size()[1] + R = weight.size()[2] + S = weight.size()[3] + + if dilation[0] > 1: + sys.exit("ERROR: simple_conv2d does not support dilated convolutions.") + if padding[0] != padding[1]: + sys.exit("ERROR: simple_conv2d does not support non-uniform padding; pad_h must be equal to pad_w.") + if groups > 1: + sys.exit("ERROR: simple_conv2d does not support grouped convolutions; set groups to 1.") + + H_out = ((H + (2*padding[0]) - dilation[0] * (R-1) -1)/stride[0]) + 1 + W_out = ((W + (2*padding[1]) - dilation[1] * (S-1) -1)/stride[1]) + 1 + output = torch.empty([N, K, int(H_out), int(W_out)]) + output = SimpleConv2dFunction.apply(output, input, weight, bias, stride, padding, dilation, groups) + return output + +class SimpleAddmm(Function): + @staticmethod + def forward(ctx, output, input, mat1, mat2, alpha, beta, a_trans, b_trans): + ctx.save_for_backward(mat1, mat2) + ctx.a_trans = a_trans + ctx.b_trans = b_trans + ctx.alpha = alpha + + simple_gemm_dev.gemm(output, mat1, mat2, alpha, a_trans, b_trans) + output += beta * input; + ctx.mark_dirty(output) + return output + + @staticmethod + def backward(ctx, grad_output): + mat1, mat2 = ctx.saved_tensors + + alpha = ctx.alpha + a_trans = ctx.a_trans + b_trans = ctx.b_trans + + grad_mat1 = torch.zeros_like(mat1) + grad_mat2 = torch.zeros_like(mat2) + grad_out, out_trans = is_transposed(grad_output) + + if a_trans: + simple_gemm_dev.gemm(grad_mat1, mat2, grad_out, alpha, b_trans, not out_trans) + else: + simple_gemm_dev.gemm(grad_mat1, grad_out, mat2, alpha, out_trans, not b_trans) + + if b_trans: + simple_gemm_dev.gemm(grad_mat2, grad_out, mat1, alpha, not out_trans, a_trans) + else: + simple_gemm_dev.gemm(grad_mat2, mat1, grad_out, alpha, not a_trans, out_trans) + + return (grad_output, grad_output, grad_mat1, grad_mat2, None, None, None, None) + +class SimpleMatmul(Function): + @staticmethod + def forward(ctx, output, mat1, mat2, alpha, a_trans, b_trans): + ctx.save_for_backward(mat1, mat2) + ctx.a_trans = a_trans + ctx.b_trans = b_trans + ctx.alpha = alpha + + simple_gemm_dev.gemm(output, mat1, mat2, alpha, a_trans, b_trans) + ctx.mark_dirty(output) + return output + + @staticmethod + def backward(ctx, grad_output): + mat1, mat2 = ctx.saved_tensors + alpha = ctx.alpha + a_trans = ctx.a_trans + b_trans = ctx.b_trans + + grad_mat1 = torch.empty_like(mat1) + grad_mat2 = torch.empty_like(mat2) + grad_out, out_trans = is_transposed(grad_output) + + if a_trans: + simple_gemm_dev.gemm(grad_mat1, mat2, grad_out, alpha, b_trans, not out_trans) + else: + simple_gemm_dev.gemm(grad_mat1, grad_out, mat2, alpha, out_trans, not b_trans) + + if b_trans: + simple_gemm_dev.gemm(grad_mat2, grad_out, mat1, alpha, not out_trans, a_trans) + else: + simple_gemm_dev.gemm(grad_mat2, mat1, grad_out, alpha, not a_trans, out_trans) + return (grad_output, grad_mat1, grad_mat2, None, None, None) + + +class SimpleConv2dFunction(Function): + @staticmethod + def forward(ctx, output, inputs, weights, bias, stride, padding, dilation, groups): + #print("### conv2d fwd called input size: ", inputs.size(), weights.size(), stride, padding, dilation, groups) + ctx.save_for_backward(inputs, weights)#, bias) + ctx.stride = stride#[0] + ctx.padding = padding#[0] + ctx.dilation = dilation#[0] + ctx.groups = groups + + if bias is None: + bias_fw = torch.zeros(output.size()[1]) + else : + bias_fw = bias + + simple_conv2d_dev.conv2d_fp(output, inputs, weights, bias_fw, stride[0], padding[0], dilation[0], groups) + ctx.mark_dirty(output) + return output + + @staticmethod + def backward(ctx, grad_output): + #inputs, weights, bias = ctx.saved_tensors + inputs, weights = ctx.saved_tensors + stride = ctx.stride + padding = ctx.padding + dilation = ctx.dilation + groups = ctx.groups + #print("### conv2d bwd called input size: ", inputs.size(), weights.size(), stride, padding, dilation, groups) + grad_inp = torch.zeros_like(inputs) + grad_wts = torch.zeros_like(weights) + + simple_conv2d_dev.conv2d_bp(grad_inp, grad_output, weights, stride[0], padding[0], dilation[0], groups) + simple_conv2d_dev.conv2d_wu(grad_wts, grad_output, inputs, stride[0], padding[0], dilation[0], groups) + return (grad_output, grad_inp, grad_wts, None, None, None, None, None) diff --git a/FP8_Emulator/cmodel/simple/simple_conv2d.cpp b/FP8_Emulator/cmodel/simple/simple_conv2d.cpp new file mode 100644 index 00000000..8498efe7 --- /dev/null +++ b/FP8_Emulator/cmodel/simple/simple_conv2d.cpp @@ -0,0 +1,127 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include +#include +#include +#include + +extern int simple_conv2d_impl_fp(float* outputs, float *inputs, float *weights, float* bias, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups); +extern int simple_conv2d_impl_bp(float* inputs, float *outputs, float *weights, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups); +extern int simple_conv2d_impl_wu(float *weights, float *outputs, float *inputs, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups); + +#define gettid() ((int)syscall(SYS_gettid)) + +using namespace torch::autograd::profiler; + + +double get_time() { + static bool init_done = false; + static struct timespec stp = {0,0}; + struct timespec tp; + clock_gettime(CLOCK_REALTIME, &tp); + + if(!init_done) { + init_done = true; + stp = tp; + } + double ret = (tp.tv_sec - stp.tv_sec) * 1e3 + (tp.tv_nsec - stp.tv_nsec)*1e-6; + return ret; +} + +at::Tensor simple_conv2d_fp(torch::Tensor& output, torch::Tensor input, torch::Tensor weight, torch::Tensor bias, + int stride, int padding, int dilation, int groups) +{ + RECORD_FUNCTION("simple_conv2d_fp", std::vector({input, weight, bias})); + + auto N = input.size(0); + auto C = input.size(1); + auto H = input.size(2); + auto W = input.size(3); + + auto K = weight.size(0); + //auto C1 = weight.size(1); + auto R = weight.size(2); + auto S = weight.size(3); + + float *input_ptr = input.data_ptr(); + float *weight_ptr = weight.data_ptr(); + float *output_ptr = output.data_ptr(); + float *bias_ptr = bias.data_ptr(); + + simple_conv2d_impl_fp(output_ptr, input_ptr, weight_ptr, bias_ptr, N, C, H, W, + K, R, S, stride, padding, dilation, groups); + + //thnn_conv2d_out(output, input, weight, + return output; +} + +at::Tensor simple_conv2d_bp(torch::Tensor& input, torch::Tensor output, torch::Tensor weight, + int stride, int padding, int dilation, int groups) +{ + RECORD_FUNCTION("simple_conv2d_bp", std::vector({output, weight})); + + auto N = input.size(0); + auto C = input.size(1); + auto H = input.size(2); + auto W = input.size(3); + + auto K = weight.size(0); + //auto C1 = weight.size(1); + auto R = weight.size(2); + auto S = weight.size(3); + + float *input_ptr = input.data_ptr(); + float *weight_ptr = weight.data_ptr(); + float *output_ptr = output.data_ptr(); + + simple_conv2d_impl_bp(input_ptr, output_ptr, weight_ptr, N, C, H, W, + K, R, S, stride, padding, dilation, groups); + + //thnn_conv2d_out(output, input, weight, + return input; +} + +at::Tensor simple_conv2d_wu(torch::Tensor& weight, torch::Tensor output, torch::Tensor input, + int stride, int padding, int dilation, int groups) +{ + RECORD_FUNCTION("simple_conv2d_wu", std::vector({output, input})); + + auto N = input.size(0); + auto C = input.size(1); + auto H = input.size(2); + auto W = input.size(3); + + auto K = weight.size(0); + //auto C1 = weight.size(1); + auto R = weight.size(2); + auto S = weight.size(3); + + float *input_ptr = input.data_ptr(); + float *weight_ptr = weight.data_ptr(); + float *output_ptr = output.data_ptr(); + + simple_conv2d_impl_wu(weight_ptr, output_ptr, input_ptr, N, C, H, W, + K, R, S, stride, padding, dilation, groups); + + //thnn_conv2d_out(output, input, weight, + return weight; +} + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { + m.def("conv2d_fp", &simple_conv2d_fp, "simple conv_fp implementation"); + m.def("conv2d_bp", &simple_conv2d_bp, "simple conv_bp implementation"); + m.def("conv2d_wu", &simple_conv2d_wu, "simple conv_wu implementation"); +} diff --git a/FP8_Emulator/cmodel/simple/simple_conv2d_impl.cpp b/FP8_Emulator/cmodel/simple/simple_conv2d_impl.cpp new file mode 100644 index 00000000..b4923da1 --- /dev/null +++ b/FP8_Emulator/cmodel/simple/simple_conv2d_impl.cpp @@ -0,0 +1,977 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include +#include +#if defined(_OPENMP) + +#include +#endif +#include + +#define CHANNEL_BLOCK 64 + +extern void MMEngine_avx2_ps(int m, int n, int k, float alpha, float *A, int lda, + float *B, int ldb, float beta, float *C, int ldc); +extern void MMEngine_strideB_avx2_ps(int m, int n, int k, float alpha, float *A, int lda, + float *B, int ldb, float beta, float *C, int ldc, int strideB); + +typedef struct { + int nImg; + int ifm; + int nBIfm; + int nbIfm; + int nBOfm; + int nbOfm; + int ifhp; + int ifwp; + int ifh; + int ifw; + int ofhp; + int ofwp; + int ofh; + int ofw; + int pad_h; + int pad_w; + /* additional padding arams for feature maps -- used in WU kernel*/ + int pad_iw; + int pad_ow; + int nbofw; + int kh; + int kw; + int stride_h; + int stride_w; +} gemm_conv_t; + +INLINE void zero_buf(float* buf, long size) { + int i; +#if defined(_OPENMP) +#pragma omp parallel for private(i) +#endif + for (i = 0; i < size; ++i) { + buf[i] = 0.0f; + } +} + +INLINE void copy_buf(float* src, float* dst, long size) { + int i; +#if defined(_OPENMP) +#pragma omp parallel for private(i) +#endif + for (i = 0; i < size; ++i) { + dst[i] = src[i]; + } +} + +INLINE void init_buf(float* buf, long size, int initPos, int initOne) +{ + int i; + zero_buf(buf, size); +#if defined(_OPENMP) +#pragma omp parallel for private(i) +#endif + for (i = 0; i < size; ++i) { + buf[i] = (float)((initOne != 0) ? 1.0 : ((initPos != 0) ? drand48() : (0.05 - drand48()/10.0))); + } +} + +INLINE void set_zeropad_nchw(float* nchw, int N, int C, int H, int W, int pad_h, int pad_w) +{ + DECLARE_VLA(4, float, input, nchw, C, H, W); + int n, h, w, c; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c) +#endif + for ( n = 0; n < N; n++ ) { + for ( c = 0; c < C; c++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + if(h < pad_h || h >= H-pad_h || w < pad_w || w >= W-pad_w) + ACCESS_VLA(4, input, n, c, h, w, C, H, W) = 0.0; + } + } + } + } +} + +INLINE +void copy_NCHW_to_GEMM(const float* nchw, float* gemm, const int N, const int H, const int W, const int C, const int cblock) +{ + DECLARE_VLA(5, float, output, gemm, C/cblock, H, W, cblock); + DECLARE_VLA(4, const float, input, nchw, C, H, W); + int n, h, w, c1, c2; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c1, c2) +#endif + for ( n = 0; n < N; n++ ) { + for ( c1 = 0; c1 < C/cblock; c1++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + for ( c2 = 0; c2 < cblock; c2++ ) { + ACCESS_VLA(5, output, n, c1, h, w, c2, C/cblock, H, W, cblock) = + ACCESS_VLA(4, input, n, (c1*cblock)+c2, h, w, C, H, W); + } + } + } + } + } +} + +INLINE +void copy_pad_NCHW_to_GEMM(const float* nchw, float* gemm, const int N, const int H, const int W, const int C, const int cblock, + int pad_h, int pad_w, int pad_c) +{ + int HP, WP; + HP = H + 2*pad_h; + WP = W + 2*pad_w; + int CP = C + pad_c; + DECLARE_VLA(5, float, output, gemm, CP/cblock, HP, WP, cblock); + DECLARE_VLA(4, const float, input, nchw, C, H, W); + int n, h, w, c1, c2; + /* if channels are smaller than cblock, use channels as the cblock */ + int lcblock = cblock; + if (C < cblock) lcblock = C; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c1, c2) +#endif + for ( n = 0; n < N; n++ ) { + for ( c1 = 0; c1 < C/lcblock; c1++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + for ( c2 = 0; c2 < lcblock; c2++ ) { + ACCESS_VLA(5, output, n, c1, h+pad_h, w+pad_w, c2, C/cblock, HP, WP, cblock) = + ACCESS_VLA(4, input, n, (c1*lcblock)+c2, h, w, C, H, W); + } + } + } + } + } +} + +INLINE +void copy_pad_NCHW_to_GEMM_ex(const float* nchw, float* gemm, const int N, const int H, const int W, const int C, const int cblock, + int pad_h, int pad_w, int pad_wex, int pad_c) +{ + int HP, WP; + HP = H + 2*pad_h; + WP = W + 2*pad_w + pad_wex; + int CP = C + pad_c; + DECLARE_VLA(5, float, output, gemm, CP/cblock, HP, WP, cblock); + DECLARE_VLA(4, const float, input, nchw, C, H, W); + int n, h, w, c1, c2; + /* if channels are smaller than cblock, use channels as the cblock */ + int lcblock = cblock; + if (C < cblock) lcblock = C; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c1, c2) +#endif + for ( n = 0; n < N; n++ ) { + for ( c1 = 0; c1 < C/lcblock; c1++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + for ( c2 = 0; c2 < lcblock; c2++ ) { + ACCESS_VLA(5, output, n, c1, h+pad_h, w+pad_w, c2, C/cblock, HP, WP, cblock) = + ACCESS_VLA(4, input, n, (c1*lcblock)+c2, h, w, C, H, W); + } + } + } + } + } +} + + +INLINE +void copy_GEMM_to_NCHW(const float* gemm, float* nchw, const int N, const int H, const int W, const int C, const int cblock) +{ + DECLARE_VLA(5, const float, input, gemm, C/cblock, H, W, cblock); + DECLARE_VLA(4, float, output, nchw, C, H, W); + int n, h, w, c1, c2; +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c1, c2) +#endif + for ( n = 0; n < N; n++ ) { + for ( c1 = 0; c1 < C/cblock; c1++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + for ( c2 = 0; c2 < cblock; c2++ ) { + ACCESS_VLA(4, output, n, (c1*cblock)+c2, h, w, C, H, W) = + ACCESS_VLA(5, input, n, c1, h, w, c2, C/cblock, H, W, cblock); + } + } + } + } + } +} +INLINE +void copy_pad_GEMM_to_NCHW(const float* gemm, float* nchw, const int N, const int H, const int W, const int C, const int cblock, + int pad_h, int pad_w, int pad_c) +{ + int HP, WP; + HP = H + 2*pad_h; + WP = W + 2*pad_w; + int CP = C + pad_c; + + DECLARE_VLA(5, const float, input, gemm, CP/cblock, HP, WP, cblock); + DECLARE_VLA(4, float, output, nchw, C, H, W); + int n, h, w, c1, c2; + + /* if number of channles are smaller than cblock, use C as cblock */ + int lcblock = cblock; + if (cblock > C) lcblock = C; +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(n, h, w, c1, c2) +#endif + for ( n = 0; n < N; n++ ) { + for ( c1 = 0; c1 < C/lcblock; c1++ ) { + for ( h = 0; h < H; h++ ) { + for ( w = 0; w < W; w++ ) { + for ( c2 = 0; c2 < lcblock; c2++ ) { + ACCESS_VLA(4, output, n, (c1*lcblock)+c2, h, w, C, H, W) = + ACCESS_VLA(5, input, n, c1, h+pad_h, w+pad_w, c2, C/cblock, HP, WP, cblock); + } + } + } + } + } +} + +INLINE +void copy_KCRS_to_GEMM(const float* kcrs, float* gemm, const int R, const int S, const int C, const int K, const int cblock, const int kblock) +{ + DECLARE_VLA(6, float, output, gemm, C/cblock, R, S, cblock, kblock); + DECLARE_VLA(4, const float, input, kcrs, C, R, S); + int r, s, c1, c2, k1, k2; +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(r, s, c1, c2, k1, k2) +#endif + for ( k1 = 0; k1 < K/kblock; k1++ ) { + for ( c1 = 0; c1 < C/cblock; c1++ ) { + for ( r = 0; r < R; r++ ) { + for ( s = 0; s < S; s++ ) { + for ( c2 = 0; c2 < cblock; c2++ ) { + for ( k2 = 0; k2 < kblock; k2++ ) { + ACCESS_VLA(6, output, k1, c1, r, s, c2, k2, C/cblock, R, S, cblock, kblock) = + ACCESS_VLA(4, input, (k1*kblock)+k2, (c1*cblock)+c2, r, s, C, R, S); + } + } + } + } + } + } +} + +INLINE +void copy_pad_KCRS_to_GEMM(const float* kcrs, float* gemm, const int R, const int S, const int C, const int K, const int cblock, + const int kblock, int pad_c, int pad_k) +{ + int CP = C + pad_c; + int KP = K + pad_k; + int lcblock = cblock; + int lkblock = kblock; + if (C < cblock) lcblock = C; + if (K < kblock) lkblock = K; + + DECLARE_VLA(6, float, output, gemm, CP/cblock, R, S, cblock, kblock); + DECLARE_VLA(4, const float, input, kcrs, C, R, S); + int r, s, c1, c2, k1, k2; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(r, s, c1, c2, k1, k2) shared(CP, KP, lcblock, lkblock) +#endif + for ( k1 = 0; k1 < KP/kblock; k1++ ) { + for ( c1 = 0; c1 < CP/cblock; c1++ ) { + for ( r = 0; r < R; r++ ) { + for ( s = 0; s < S; s++ ) { + for ( c2 = 0; c2 < lcblock; c2++ ) { + for ( k2 = 0; k2 < lkblock; k2++ ) { + ACCESS_VLA(6, output, k1, c1, r, s, c2, k2, CP/cblock, R, S, cblock, kblock) = + ACCESS_VLA(4, input, (k1*lkblock)+k2, (c1*lcblock)+c2, r, s, C, R, S); + } + } + } + } + } + } +} + + +INLINE +void copy_GEMM_to_KCRS(const float* gemm, float* kcrs, const int R, const int S, const int C, const int K, const int cblock, const int kblock) +{ + DECLARE_VLA(6, const float, input, gemm, C/cblock, R, S, cblock, kblock); + DECLARE_VLA(4, float, output, kcrs, C, R, S); + int r, s, c1, c2, k1, k2; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(r, s, c1, c2, k1, k2) +#endif + for ( k1 = 0; k1 < K/kblock; k1++ ) { + for ( c1 = 0; c1 < C/cblock; c1++ ) { + for ( r = 0; r < R; r++ ) { + for ( s = 0; s < S; s++ ) { + for ( c2 = 0; c2 < cblock; c2++ ) { + for ( k2 = 0; k2 < kblock; k2++ ) { + ACCESS_VLA(4, output, (k1*kblock)+k2, (c1*cblock)+c2, r, s, C, R, S) = + ACCESS_VLA(6, input, k1, c1, r, s, c2, k2, C/cblock, R, S, cblock, kblock); + } + } + } + } + } + } +} + +INLINE +void copy_pad_GEMM_to_KCRS(const float* gemm, float* kcrs, const int R, const int S, const int C, const int K, const int cblock, + const int kblock, int pad_c, int pad_k) +{ + + int CP = C + pad_c; + int KP = K + pad_k; + int lcblock = cblock; + int lkblock = kblock; + if (C < cblock) lcblock = C; + if (K < kblock) lkblock = K; + + DECLARE_VLA(6, const float, input, gemm, CP/cblock, R, S, cblock, kblock); + DECLARE_VLA(4, float, output, kcrs, C, R, S); + int r, s, c1, c2, k1, k2; + +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(r, s, c1, c2, k1, k2) shared(CP, KP, lcblock, lkblock) +#endif + for ( k1 = 0; k1 < KP/kblock; k1++ ) { + for ( c1 = 0; c1 < CP/cblock; c1++ ) { + for ( r = 0; r < R; r++ ) { + for ( s = 0; s < S; s++ ) { + for ( c2 = 0; c2 < lcblock; c2++ ) { + for ( k2 = 0; k2 < lkblock; k2++ ) { + ACCESS_VLA(4, output, (k1*lkblock)+k2, (c1*lcblock)+c2, r, s, C, R, S) = + ACCESS_VLA(6, input, k1, c1, r, s, c2, k2, CP/cblock, R, S, cblock, kblock); + } + } + } + } + } + } +} + +INLINE +void gemm_kernel_conv_fp(gemm_conv_t* param, float* output, float* input, float* filter, const float* bias) +{ + int nImg = param->nImg; + int nBIfm = param->nBIfm; + int nbIfm = param->nbIfm; + int nBOfm = param->nBOfm; + int nbOfm = param->nbOfm; + int ifhp = param->ifhp; + int ifwp = param->ifwp; + int ofhp = param->ofhp; + int ofwp = param->ofwp; + //int ifh = param->ifh; + //int ifw = param->ifw; + int ofh = param->ofh; + int ofw = param->ofw; + //int nbofw = param->nbofw; + int pad_h = param->pad_h; + int pad_w = param->pad_w; + int kh = param->kh; + int kw = param->kw; + int stride_h = param->stride_h; + int stride_w = param->stride_w; + /* loop counters */ + int img, ofm1, ifm1, oj, oi, ij, kj, ki, ofm2; +#ifdef NAIVE_CODE + int ii, ifm2; +#endif + DECLARE_VLA(5, float, output_t, output + (pad_h * ofwp * nbOfm + pad_w * nbOfm), nBOfm, ofhp, ofwp, nbOfm); + DECLARE_VLA(5, float, input_t, input , nBIfm, ifhp, ifwp, nbIfm); + DECLARE_VLA(6, float, filter_t, filter, nBIfm, kh, kw, nbIfm, nbOfm); + + int lda = CHANNEL_BLOCK; + int ldb = stride_w*CHANNEL_BLOCK; + int ldc = CHANNEL_BLOCK; +#if 0 + /* compute new n-blocking to fit limitation of N<=16 of TMUL */ + for ( oii = 16; oii > 0; --oii ) { + if ( ofw % oii == 0 ) { + nbofw = oii; + break; + } + } +#endif +#if defined(_OPENMP) +#pragma omp parallel for collapse(2) private(img, ofm1, ofm2, oj, oi ) +#endif + /* pre-initializing with bias */ + for (img = 0; img < nImg; ++img) { + for (ofm1 = 0; ofm1 < nBOfm; ++ofm1) { + for (oj = 0; oj < ofh; ++oj) { + for (oi = 0; oi < ofw; ++oi) { + for (ofm2 = 0; ofm2 < nbOfm; ++ofm2) { + ACCESS_VLA(5, output_t, img, ofm1, oj, oi, ofm2, nBOfm, ofhp, ofwp, nbOfm) = bias[ofm1*nbOfm + ofm2]; + } + } + } + } + } + +#if defined(_OPENMP) +#ifdef NAIVE_CODE +#pragma omp parallel for collapse(4) private(img, ofm1, ofm2, ifm1, ifm2, oj, oi, ij, ii, kj, ki) +#else +#pragma omp parallel for collapse(4) private(img, ofm1, ofm2, ifm1, oj, oi, ij, kj, ki) +#endif +#endif + for (img = 0; img < nImg; ++img) { + for (ofm1 = 0; ofm1 < nBOfm; ++ofm1) { + for (ifm1 = 0; ifm1 < nBIfm; ++ifm1) { + for (oj = 0; oj < ofh; ++oj) { + ij = oj * stride_h; + for (kj = 0; kj < kh; ++kj) { + for (ki = 0; ki < kw; ++ki) { + /* let's do a 64 x ofw x 64 GEMM : M=nbOfm, N=ofw, K=nbIfm (col-major) */ +#ifdef NAIVE_CODE + for (oi = 0; oi < ofw; ++oi) { + ii = oi * stride_w; + for (ofm2 = 0; ofm2 < nbOfm; ++ofm2) { + for (ifm2 = 0; ifm2 < nbIfm; ++ifm2) { + ACCESS_VLA(5, output_t, img, ofm1, oj, oi, ofm2, nBOfm, ofhp, ofwp, nbOfm) += /* C */ + ACCESS_VLA(6, filter_t, ofm1, ifm1, kj, ki, ifm2, ofm2, nBIfm, kh, kw, nbIfm, nbOfm) /* A */ + * ACCESS_VLA(5, input_t, img, ifm1, ij + kj, ii + ki, ifm2, nBIfm, ifhp, ifwp, nbIfm); /* B */ + } + } + } +#else + MMEngine_avx2_ps (nbOfm, ofw, nbIfm, 1.0, + &ACCESS_VLA(6, filter_t, ofm1, ifm1, kj, ki, 0, 0, nBIfm, kh, kw, nbIfm, nbOfm), lda, + &ACCESS_VLA(5, input_t, img, ifm1, ij + kj, ki, 0, nBIfm, ifhp, ifwp, nbIfm), ldb, 1.0, + &ACCESS_VLA(5, output_t, img, ofm1, oj, 0, 0, nBOfm, ofhp, ofwp, nbOfm), ldc); +#endif + } + } + } + } + } + } +} + +INLINE void gemm_kernel_conv_bp(gemm_conv_t* param, float* input, float* output, float* filter, float* tr_filter) +{ + int nImg = param->nImg; + int nBIfm = param->nBIfm; + int nbIfm = param->nbIfm; + int nBOfm = param->nBOfm; + int nbOfm = param->nbOfm; + int ifhp = param->ifhp; + int ifwp = param->ifwp; + int ofhp = param->ofhp; + int ofwp = param->ofwp; + //int ifh = param->ifh; + //int ifw = param->ifw; + int ofh = param->ofh; + int ofw = param->ofw; + //int nbofw = param->nbofw; + int pad_h = param->pad_h; + int pad_w = param->pad_w; + int kh = param->kh; + int kw = param->kw; + int stride_h = param->stride_h; + int stride_w = param->stride_w; + + /* loop counters */ + int img, ofm1, ifm1, ofm2, ifm2, oj, ij, kj, ki;/*, oii;*/ +#ifdef NAIVE_CODE + int ii, oi; +#endif + DECLARE_VLA(5, float, output_t, output + (pad_h * ofwp * nbOfm + pad_w * nbOfm), nBOfm, ofhp, ofwp, nbOfm); + DECLARE_VLA(5, float, input_t, input , nBIfm, ifhp, ifwp, nbIfm); + DECLARE_VLA(6, float, filter_t, filter, nBIfm, kh, kw, nbIfm, nbOfm); + DECLARE_VLA(6, float, tr_filter_t, tr_filter, nBIfm, kh, kw, nbOfm, nbIfm); + +#if defined(_OPENMP) +# pragma omp parallel for collapse(4) private(ofm1, ifm1, kj, ki, ofm2, ifm2) +#endif + for (ofm1 = 0; ofm1 < nBOfm; ++ofm1) { + for (ifm1 = 0; ifm1 < nBIfm; ++ifm1) { + for (kj = 0; kj < kh; ++kj) { + for (ki = 0; ki < kw; ++ki) { + for (ofm2 = 0; ofm2 < nbOfm; ++ofm2) { + for (ifm2 = 0; ifm2 < nbIfm; ++ifm2) { + ACCESS_VLA(6, tr_filter_t, ofm1, ifm1, kj, ki, ofm2, ifm2, nBIfm, kh, kw, nbOfm, nbIfm) = + ACCESS_VLA( 6, filter_t, ofm1, ifm1, kj, ki, ifm2, ofm2, nBIfm, kh, kw, nbIfm, nbOfm); + } + } + } + } + } + } + + int lda = CHANNEL_BLOCK; + int ldb = CHANNEL_BLOCK; + int ldc = stride_w*CHANNEL_BLOCK; +#if 0 + /* compute new n-blocking */ + for ( oii = 16; oii > 0; --oii ) { + if ( ofw % oii == 0 ) { + nbofw = oii; + break; + } + } +#endif +#if defined(_OPENMP) +#ifdef NAIVE_CODE +# pragma omp parallel for collapse(4) private(img, ofm1, ifm1, ofm2, ifm2, oj, oi, ij, ii, kj, ki) +#else +# pragma omp parallel for collapse(4) private(img, ofm1, ifm1, ofm2, ifm2, oj, ij, kj, ki) +#endif +#endif + for (img = 0; img < nImg; ++img) { + for (ifm1 = 0; ifm1 < nBIfm; ++ifm1) { + for (ofm1 = 0; ofm1 < nBOfm; ++ofm1) { + for (oj = 0; oj < ofh; ++oj) { + ij = oj * stride_h; + for (kj = 0; kj < kh; ++kj) { + for (ki = 0; ki < kw; ++ki) { +#ifdef NAIVE_CODE + for (ifm2 = 0; ifm2 < nbIfm; ++ifm2) { + for (oi = 0; oi < ofw; ++oi) { + ii = oi * stride_w; + for (ofm2 = 0; ofm2 < nbOfm; ++ofm2) { + ACCESS_VLA(5, input_t, img, ifm1, ij + kj, ii + ki, ifm2, nBIfm, ifhp, ifwp, nbIfm) += /* C */ + ACCESS_VLA(6, tr_filter_t, ofm1, ifm1, kj, ki, ofm2, ifm2, nBIfm, kh, kw, nbOfm, nbIfm) /* A */ + * ACCESS_VLA(5, output_t, img, ofm1, oj, oi, ofm2, nBOfm, ofhp, ofwp, nbOfm); /* B */ + } + } + } +#else + MMEngine_avx2_ps ( nbIfm, ofw, nbOfm, 1.0, + &ACCESS_VLA(6, tr_filter_t, ofm1, ifm1, kj, ki, 0, 0, nBIfm, kh, kw, nbOfm, nbIfm), lda, + &ACCESS_VLA(5, output_t, img, ofm1, oj, 0, 0, nBOfm, ofhp, ofwp, nbOfm), ldb, 1.0, + &ACCESS_VLA(5, input_t, img, ifm1, ij + kj, ki, 0, nBIfm, ifhp, ifwp, nbIfm), ldc); +#endif + } + } + } + } + } + } +} + +INLINE void gemm_kernel_conv_wu(gemm_conv_t* param, float* filter, float* output, float* input, float* tr_input) +{ + int nImg = param->nImg; + int nBIfm = param->nBIfm; + int nbIfm = param->nbIfm; + int nBOfm = param->nBOfm; + int nbOfm = param->nbOfm; + int ifhp = param->ifhp; + int ifwp = param->ifwp; + int ofhp = param->ofhp; + int ofwp = param->ofwp; + int ifh = param->ifh; + int ifw = param->ifw; + int ofh = param->ofh; + int ofw = param->ofw; + //int nbofw = param->nbofw; + int pad_h = param->pad_h; + int pad_w = param->pad_w; + //int pad_iw = param->pad_iw; + int pad_ow = param->pad_ow; + int kh = param->kh; + int kw = param->kw; + int stride_h = param->stride_h; + int stride_w = param->stride_w; + + /* loop counters */ + int img, ofm1, ifm1, ifm2, oj, ij, ii, kj, ki;/*, oii;*/ +#ifdef NAIVE_CODE + int ofm2, oi; +#endif + + DECLARE_VLA(5, float, output_t, output + (pad_h * ofwp * nbOfm + pad_w * nbOfm), nBOfm, ofhp, ofwp, nbOfm); + DECLARE_VLA(5, float, input_t, (float*)input, nBIfm, ifhp, ifwp, nbIfm); + DECLARE_VLA(5, float, tr_input_t, tr_input, nBIfm, ifhp+(2*pad_h), nbIfm, ifwp+(2*pad_w)); + DECLARE_VLA(6, float, filter_t, filter, nBIfm, kh, kw, nbIfm, nbOfm); + +#if defined(_OPENMP) +# pragma omp parallel for collapse(2) private(img, ifm1, ij, ii, ifm2) +#endif + for (img = 0; img < nImg; ++img) { + for (ifm1 = 0; ifm1 < nBIfm; ++ifm1) { + for (ij = 0; ij < ifh; ++ij) { + for (ifm2 = 0; ifm2 < nbIfm; ++ifm2) { + for (ii = 0; ii < ifw; ++ii) { + ACCESS_VLA(5, tr_input_t, img, ifm1, ij+pad_h, ifm2, ii+pad_w, nBIfm, ifhp+(2*pad_h), nbIfm, ifwp+(2*pad_w)) = + ACCESS_VLA(5, input_t, img, ifm1, ij+pad_h, ii+pad_w, ifm2, nBIfm, ifhp, ifwp, nbIfm); + } + } + } + } + } + + int lda = CHANNEL_BLOCK; + int ldb = ifwp + (2*pad_w); + int ldc = CHANNEL_BLOCK; +#if 0 + int nbifm = nbIfm; + for ( oii = 16; oii > 0; --oii ) { + if ( nbIfm % oii == 0 ) { + nbifm = oii; + break; + } + } +#endif +#if defined(_OPENMP) +#ifdef NAIVE_CODE +# pragma omp parallel for collapse(2) private(img, ofm1, ifm1, ofm2, ifm2, oj, oi, ij, ii, kj, ki) +#else +# pragma omp parallel for collapse(2) private(img, ofm1, ifm1, ifm2, oj, ij, ii, kj, ki) +#endif +#endif + for (ofm1 = 0; ofm1 < nBOfm; ++ofm1) { + for (ifm1 = 0; ifm1 < nBIfm; ++ifm1) { + for (img = 0; img < nImg; ++img) { + for (oj = 0; oj < ofh; ++oj) { + ij = oj * stride_h; + for (kj = 0; kj < kh; ++kj) { + for (ki = 0; ki < kw; ++ki) { +#ifdef NAIVE_CODE + for (ifm2 = 0; ifm2 < nbIfm; ++ifm2) { + for (oi = 0; oi < ofw; ++oi) { + ii = oi * stride_w; + for (ofm2 = 0; ofm2 < nbOfm; ++ofm2) { + ACCESS_VLA( 6, filter_t, ofm1, ifm1, kj, ki, ifm2, ofm2, nBIfm, kh, kw, nbIfm, nbOfm) /* C */ += + ACCESS_VLA(5, output_t, img, ofm1, oj, oi, ofm2, nBOfm, ofhp, ofwp, nbOfm) /* A */ + * ACCESS_VLA(5, tr_input_t, img, ifm1, ij + kj, ifm2, ii + ki, nBIfm, ifhp+(2*pad_h), nbIfm, ifwp+(2*pad_w)) /* B */; + } + } + } +#else + MMEngine_strideB_avx2_ps (nbOfm, nbIfm, ofw+pad_ow, 1.0, + &ACCESS_VLA(5, output_t, img, ofm1, oj, 0, 0, nBOfm, ofhp, ofwp, nbOfm), lda, + &ACCESS_VLA(5, tr_input_t, img, ifm1, ij + kj, 0, ki, nBIfm, ifhp+(2*pad_h), nbIfm, ifwp+(2*pad_w)), ldb, 1.0, + &ACCESS_VLA(6, filter_t, ofm1, ifm1, kj, ki, 0, 0, nBIfm, kh, kw, nbIfm, nbOfm), ldc, stride_w); +#endif + } + } + } + } + } + } +} + +int simple_conv2d_impl_fp(float* outputs, float *inputs, float *weights, float* bias, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups) +{ + float *gemm_input, *gemm_output, *gemm_filter; + int ifhp, ifwp, ofhp, ofwp, ofh, ofw; + int stride_h, stride_w, pad_h, pad_w; + + int ifw = iW; /* input width, "W" */ + int ifh = iH; /* input height, "H" */ + int nImg = N; /* mini-batch size, "N" */ + int nIfm = C; /* number of input feature maps, "C" */ + int nOfm = K; /* number of output feature maps, "K" */ + int kh = R; /* filter height, "R" */ + int kw = S; /* filter width, "S" */ + pad_h = padding; /* padding in output */ + pad_w = padding; /* padding in output */ + + /* apply stride in both dimensions */ + stride_w = stride; + stride_h = stride; + + /* deriving some values image size */ + ofh = (ifh + 2 * pad_h - kh) / stride_h + 1; + ofw = (ifw + 2 * pad_w - kw) / stride_w + 1; + //ofh = (ifh + 2 * pad_h - dilation * (kh-1) - 1) / stride_h + 1; + //ofw = (ifw + 2 * pad_w - dilation * (kw-1) - 1) / stride_w + 1; + + ifhp = ifh + 2 * pad_h; + ifwp = ifw + 2 * pad_w; + ofhp = ofh + 2 * pad_h; + ofwp = ofw + 2 * pad_w; + + /*pad nIfm and nOfm to multiples of 64 */ + int ifm_pad = 0; + int ofm_pad = 0; + int nIfmp = nIfm; + int nOfmp = nOfm; + if (nIfm % 64 != 0){ + ifm_pad = (64-(nIfm%64)); + nIfmp += ifm_pad; + } + if (nOfm % 64 != 0) { + ofm_pad = (64-(nOfm%64)); + nOfmp += ofm_pad; + } + + gemm_conv_t gemm_param; + /* set struct for naive convolution */ + gemm_param.nImg = nImg; + gemm_param.nBIfm = nIfmp/CHANNEL_BLOCK; + gemm_param.nbIfm = CHANNEL_BLOCK; + gemm_param.nBOfm = nOfmp/CHANNEL_BLOCK; + gemm_param.nbOfm = CHANNEL_BLOCK; + gemm_param.ifhp = ifhp; + gemm_param.ifwp = ifwp; + gemm_param.ofhp = ofhp; + gemm_param.ofwp = ofwp; + gemm_param.ifh = ifh; + gemm_param.ifw = ifw; + gemm_param.ofh = ofh; + gemm_param.ofw = ofw; + gemm_param.pad_h = pad_h; + gemm_param.pad_w = pad_w; + + if ( ofw == 56 ) { + gemm_param.nbofw = 28; + } else { + gemm_param.nbofw = ofw; + } + gemm_param.kh = kh; + gemm_param.kw = kw; + gemm_param.stride_h = stride_h; + gemm_param.stride_w = stride_w; + + gemm_input = (float*)_mm_malloc( nImg*nIfmp*ifhp*ifwp*sizeof(float), 2097152); + gemm_filter = (float*)_mm_malloc( nOfmp*nIfmp*kh*kw* sizeof(float), 2097152); + gemm_output = (float*)_mm_malloc( nImg*nOfmp*ofhp*ofwp*sizeof(float), 2097152); + zero_buf(gemm_input, nImg*nIfmp*ifhp*ifwp); + if (nIfm % 64 != 0 || nOfm % 64 != 0){ + zero_buf(gemm_filter, nOfmp*nIfmp*kh*kw); + zero_buf(gemm_output, nImg*nOfmp*ofhp*ofwp); + } + + /* copy data into GEMM optimized format */ + copy_pad_NCHW_to_GEMM(inputs, gemm_input, nImg, ifh, ifw, nIfm, CHANNEL_BLOCK, pad_h, pad_w, ifm_pad); + copy_pad_KCRS_to_GEMM(weights, gemm_filter, kh, kw, nIfm, nOfm, CHANNEL_BLOCK, CHANNEL_BLOCK, ifm_pad, ofm_pad); + gemm_kernel_conv_fp(&gemm_param, gemm_output, gemm_input, gemm_filter, bias); + /* copy out data */ + copy_pad_GEMM_to_NCHW(gemm_output, outputs, nImg, ofh, ofw, nOfm, CHANNEL_BLOCK, pad_h, pad_w, ofm_pad); + + _mm_free(gemm_input); + _mm_free(gemm_filter); + _mm_free(gemm_output); + return 0; +} + +int simple_conv2d_impl_bp(float* inputs, float *outputs, float *weights, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups) +{ + float *gemm_input, *gemm_output, *gemm_filter, *gemm_filter_tr; + int ifhp, ifwp, ofhp, ofwp, ofh, ofw; + int stride_h, stride_w, pad_h, pad_w; + + int ifw = iW; /* input width, "W" */ + int ifh = iH; /* input height, "H" */ + int nImg = N; /* mini-batch size, "N" */ + int nIfm = C; /* number of input feature maps, "C" */ + int nOfm = K; /* number of output feature maps, "K" */ + int kh = R; /* filter height, "R" */ + int kw = S; /* filter width, "S" */ + pad_h = padding; /* padding in output */ + pad_w = padding; /* padding in output */ + + /* apply stride in both dimensions */ + stride_w = stride; + stride_h = stride; + /* deriving some values image size */ + ofh = (ifh + 2 * pad_h - kh) / stride_h + 1; + ofw = (ifw + 2 * pad_w - kw) / stride_w + 1; + //ofh = (ifh + 2 * pad_h - dilation * (kh-1) - 1) / stride_h + 1; + //ofw = (ifw + 2 * pad_w - dilation * (kw-1) - 1) / stride_w + 1; + ifhp = ifh + 2 * pad_h; + ifwp = ifw + 2 * pad_w; + ofhp = ofh + 2 * pad_h; + ofwp = ofw + 2 * pad_w; + /*pad nIfm and nOfm to multiples of 64 */ + int ifm_pad = 0; + int ofm_pad = 0; + int nIfmp = nIfm; + int nOfmp = nOfm; + if (nIfm % 64 != 0){ + ifm_pad = (64-(nIfm%64)); + nIfmp += ifm_pad; + } + if (nOfm % 64 != 0) { + ofm_pad = (64-(nOfm%64)); + nOfmp += ofm_pad; + } + + gemm_conv_t gemm_param; + /* set struct for naive convolution */ + gemm_param.nImg = nImg; + gemm_param.nBIfm = nIfmp/CHANNEL_BLOCK; + gemm_param.nbIfm = CHANNEL_BLOCK; + gemm_param.nBOfm = nOfmp/CHANNEL_BLOCK; + gemm_param.nbOfm = CHANNEL_BLOCK; + gemm_param.ifhp = ifhp; + gemm_param.ifwp = ifwp; + gemm_param.ofhp = ofhp; + gemm_param.ofwp = ofwp; + gemm_param.ifh = ifh; + gemm_param.ifw = ifw; + gemm_param.ofh = ofh; + gemm_param.ofw = ofw; + gemm_param.pad_h = pad_h; + gemm_param.pad_w = pad_w; + + if ( ofw == 56 ) { + gemm_param.nbofw = 28; + } else { + gemm_param.nbofw = ofw; + } + + gemm_param.kh = kh; + gemm_param.kw = kw; + gemm_param.stride_h = stride_h; + gemm_param.stride_w = stride_w; + gemm_input = (float*)_mm_malloc( nImg*nIfmp*ifhp*ifwp*sizeof(float), 2097152); + gemm_filter = (float*)_mm_malloc( nOfmp*nIfmp*kh*kw* sizeof(float), 2097152); + gemm_filter_tr = (float*)_mm_malloc( nOfmp*nIfmp*kh*kw* sizeof(float), 2097152); + gemm_output = (float*)_mm_malloc( nImg*nOfmp*ofhp*ofwp*sizeof(float), 2097152); + + zero_buf(gemm_input, nImg*nIfmp*ifhp*ifwp); + if (nIfm % 64 != 0 || nOfm % 64 != 0){ + zero_buf(gemm_filter, nOfmp*nIfmp*kh*kw); + zero_buf(gemm_output, nImg*nOfmp*ofhp*ofwp); + } + + /* copy data into GEMM optimized format */ + copy_pad_NCHW_to_GEMM(outputs, gemm_output, nImg, ofh, ofw, nOfm, CHANNEL_BLOCK, pad_h, pad_w, ofm_pad); + copy_pad_KCRS_to_GEMM(weights, gemm_filter, kh, kw, nIfm, nOfm, CHANNEL_BLOCK, CHANNEL_BLOCK, ifm_pad, ofm_pad); + gemm_kernel_conv_bp(&gemm_param, gemm_input, gemm_output, gemm_filter, gemm_filter_tr); + /* copy out data */ + copy_pad_GEMM_to_NCHW(gemm_input, inputs, nImg, ifh, ifw, nIfm, CHANNEL_BLOCK, pad_h, pad_w, ifm_pad); + + _mm_free(gemm_input); + _mm_free(gemm_filter); + _mm_free(gemm_filter_tr); + _mm_free(gemm_output); + return 0; +} + +int simple_conv2d_impl_wu(float *weights, float *outputs, float *inputs, int N, int C, int iH, int iW, + int K, int R, int S, int stride, int padding, int dilation, int groups) +{ + float *gemm_input, *gemm_input_tr, *gemm_output, *gemm_filter; + int ifhp, ifwp, ofhp, ofwp, ofh, ofw; + int stride_h, stride_w, pad_h, pad_w; + + int ifw = iW; /* input width, "W" */ + int ifh = iH; /* input height, "H" */ + int nImg = N; /* mini-batch size, "N" */ + int nIfm = C; /* number of input feature maps, "C" */ + int nOfm = K; /* number of output feature maps, "K" */ + int kh = R; /* filter height, "R" */ + int kw = S; /* filter width, "S" */ + pad_h = padding; /* padding in output */ + pad_w = padding; /* padding in output */ + + /* apply stride in both dimensions */ + stride_w = stride; + stride_h = stride; + + /* deriving some values image size */ + ofw = (ifw + 2 * pad_w - kw) / stride_w + 1; + ofh = (ifh + 2 * pad_h - kh) / stride_h + 1; + //ofw = (ifw + 2 * pad_w - dilation * (kw-1) - 1) / stride_w + 1; + //ofh = (ifh + 2 * pad_h - dilation * (kh-1) - 1) / stride_h + 1; + + /* padding feature map width to be a multiple of 4 to perform VNN4 operations */ + int pad_iw = 0; + int pad_ow = 0; + + if (ofw%4 != 0){ + pad_ow = 4-(ofw%4); + } + if (ifw%4 != 0){ + pad_iw = 4-(ifw%4); + } + pad_iw = stride_w * pad_ow; + + /*pad ofw and ifw to to be multiples of 4 (VNNI4) */ + ifhp = ifh + 2 * pad_h; + ifwp = ifw + 2 * pad_w + pad_iw; + ofhp = ofh + 2 * pad_h; + ofwp = ofw + 2 * pad_w + pad_ow; + + /*pad nIfm and nOfm to multiples of 64 */ + int ifm_pad = 0; + int ofm_pad = 0; + int nIfmp = nIfm; + int nOfmp = nOfm; + if (nIfm % 64 != 0){ + ifm_pad = (64-(nIfm%64)); + nIfmp += ifm_pad; + } + if (nOfm % 64 != 0) { + ofm_pad = (64-(nOfm%64)); + nOfmp += ofm_pad; + } + + gemm_conv_t gemm_param; + /* set struct for naive convolution */ + gemm_param.nImg = nImg; + gemm_param.nBIfm = nIfmp/CHANNEL_BLOCK; + gemm_param.nbIfm = CHANNEL_BLOCK; + gemm_param.nBOfm = nOfmp/CHANNEL_BLOCK; + gemm_param.nbOfm = CHANNEL_BLOCK; + gemm_param.ifhp = ifhp; + gemm_param.ifwp = ifwp; + gemm_param.ofhp = ofhp; + gemm_param.ofwp = ofwp; + gemm_param.ifh = ifh; + gemm_param.ifw = ifw; + gemm_param.ofh = ofh; + gemm_param.ofw = ofw; + gemm_param.pad_h = pad_h; + gemm_param.pad_w = pad_w; + gemm_param.pad_iw = pad_iw; + gemm_param.pad_ow = pad_ow; + + if ( ofw == 56 ) { + gemm_param.nbofw = 28; + } else { + gemm_param.nbofw = ofw; + } + + gemm_param.kh = kh; + gemm_param.kw = kw; + gemm_param.stride_h = stride_h; + gemm_param.stride_w = stride_w; + gemm_input = (float*)_mm_malloc( nImg*nIfmp*ifhp*ifwp*sizeof(float), 2097152); + gemm_input_tr = (float*)_mm_malloc( nImg*nIfmp*(ifhp+(2*pad_h))*(ifwp+(2*pad_w))*sizeof(float), 2097152); + gemm_filter = (float*)_mm_malloc( nOfmp*nIfmp*kh*kw* sizeof(float), 2097152); + gemm_output = (float*)_mm_malloc( nImg*nOfmp*ofhp*ofwp*sizeof(float), 2097152); + zero_buf(gemm_input, nImg*nIfmp*ifhp*ifwp); + zero_buf(gemm_input_tr, nImg*(ifhp+(2*pad_h))*nIfmp*(ifwp+(2*pad_w))); + zero_buf(gemm_output, nImg*nOfmp*ofhp*ofwp); + zero_buf(gemm_filter, nOfmp*nIfmp*kh*kw); + /* copy data into GEMM optimized format */ + /* compensate for the VNNI4 padding */ + copy_pad_NCHW_to_GEMM_ex(inputs, gemm_input, nImg, ifh, ifw, nIfm, CHANNEL_BLOCK, pad_h, pad_w, pad_iw, ifm_pad); + copy_pad_NCHW_to_GEMM_ex(outputs, gemm_output, nImg, ofh, ofw, nOfm, CHANNEL_BLOCK, pad_h, pad_w, pad_ow , ofm_pad); + gemm_kernel_conv_wu(&gemm_param, gemm_filter, gemm_output, gemm_input, gemm_input_tr); + copy_pad_GEMM_to_KCRS(gemm_filter, weights, kh, kw, nIfm, nOfm, CHANNEL_BLOCK, CHANNEL_BLOCK, ifm_pad, ofm_pad); + _mm_free(gemm_input); + _mm_free(gemm_input_tr); + _mm_free(gemm_filter); + _mm_free(gemm_output); + return 0; +} diff --git a/FP8_Emulator/cmodel/simple/simple_gemm.cpp b/FP8_Emulator/cmodel/simple/simple_gemm.cpp new file mode 100644 index 00000000..86e277e8 --- /dev/null +++ b/FP8_Emulator/cmodel/simple/simple_gemm.cpp @@ -0,0 +1,71 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include +#include +#include +#include +#if 10 +//extern "C" { +extern int simple_sgemm_impl( char* transa, char* transb, int m, int n, int k, + float alpha, float* a, int lda, float* b, int ldb, + float beta, float* c, int ldc ); +//} +#endif +#define gettid() ((int)syscall(SYS_gettid)) + +using namespace torch::autograd::profiler; +using namespace torch; +using namespace torch::autograd; +using at::Tensor; + +double get_time() { + static bool init_done = false; + static struct timespec stp = {0,0}; + struct timespec tp; + clock_gettime(CLOCK_REALTIME, &tp); + + if(!init_done) { + init_done = true; + stp = tp; + } + double ret = (tp.tv_sec - stp.tv_sec) * 1e3 + (tp.tv_nsec - stp.tv_nsec)*1e-6; + return ret; +} + +at::Tensor simple_gemm(torch::Tensor& C, torch::Tensor A, torch::Tensor B, float alpha, bool a_trans, bool b_trans) +{ + RECORD_FUNCTION("simple_gemm", std::vector({A, B, alpha})); + + const char *aT = a_trans ? "T" : "N"; + const char *bT = b_trans ? "T" : "N"; + + auto M = C.size(0); + auto N = C.size(1); + auto K = a_trans ? A.size(0) : A.size(1); + auto lda = A.size(1); + auto ldb = B.size(1); + auto ldc = C.size(1); + + float beta = 0.0; + + float *Aptr = A.data_ptr(); + float *Bptr = B.data_ptr(); + float *Cptr = C.data_ptr(); + + simple_sgemm_impl((char*)bT, (char*)aT, N, M, K, alpha, Bptr, ldb, Aptr, lda, beta, Cptr, ldc); + return C; +} + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { + m.def("gemm", &simple_gemm, "TMUL GEMM Implementation"); +} diff --git a/FP8_Emulator/cmodel/simple/simple_gemm_impl.cpp b/FP8_Emulator/cmodel/simple/simple_gemm_impl.cpp new file mode 100644 index 00000000..39f96f67 --- /dev/null +++ b/FP8_Emulator/cmodel/simple/simple_gemm_impl.cpp @@ -0,0 +1,233 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include +#include +#include +#include + +#define SCRATCH_SIZE 4294967296 +int lazy_init = 1; +float* myscratch = NULL; + +extern void MMEngine_avx2_ps(int m, int n, int k, float alpha, float *A, int lda, + float *B, int ldb, float beta, float *C, int ldc); + +__extern_always_inline +void copy_matrix_and_pad(const float* in, float* out, const int LD, const int OD, int ld_pad, int od_pad) +{ + int LDP, ODP; + LDP = LD + ld_pad; + ODP = OD + od_pad; + int ld, od; + int lp, op; + + for ( op = 0; op < OD; op++ ) { + for ( lp = LD-1; lp < LDP; lp++ ) { + out[op*LDP+lp] = 0.0; + } + } + for ( op = OD-1; op < ODP; op++ ) { + for ( lp = 0; lp < LDP; lp++ ) { + out[op*LDP+lp] = 0.0; + } + } + +#if defined(_OPENMP) +#pragma omp parallel for private(ld, od) +#endif + for ( od = 0; od < OD; od++ ) { + for ( ld = 0; ld < LD; ld++ ) { + out[od*LDP+ld] = in[od*LD+ld]; + } + } +} + +__extern_always_inline +void copy_matrix_and_strip_pading(const float* in, float* out, const int LD, const int OD, int ld_pad, int od_pad) +{ + int LDP; + LDP = LD + ld_pad; + int ld, od; +#if defined(_OPENMP) +#pragma omp parallel for private(ld, od) +#endif + for ( od = 0; od < OD; od++ ) { + for ( ld = 0; ld < LD; ld++ ) { + out[od*LD+ld] = in[od*LDP+ld]; + } + } +} + +int simple_sgemm_impl( char* transa, char* transb, int m, int n, int k, + float alpha, float* a, int lda, float* b, int ldb, + float beta, float* c, int ldc ) { + float* myA = NULL; + float* myB = NULL; + float* myC = NULL; + size_t ptlda = (size_t)(lda); + size_t ptldb = (size_t)(ldb); + size_t mylda = (size_t)(lda); + size_t myldb = (size_t)(ldb); + size_t myldc = (size_t)(ldc); + int mym = (size_t)(m); + int myn = (size_t)(n); + int myk = (size_t)(k); + int m_pad = 0; + int n_pad = 0; + int k_pad = 0; + + int o,p,q,pp,oo; + + /* check for size matching our TMUL emulation */ + if ( mym % 16 != 0 ) { + m_pad = (16-(mym%16)); + mym += m_pad; + } + if ( myk % 64 != 0 ) { + k_pad = (64-(myk%64)); + myk += k_pad; + } + if ( myn % 16 != 0 ) { + n_pad = (16-(myn%16)); + myn += n_pad; + } + /* update leading dimensions with padded values */ + mylda = mym; + myldb = myk; + myldc = mym; + + /* lazy init of fp8_gemm state */ + if (lazy_init != 0) { + lazy_init = 0; + myscratch = (float*) _mm_malloc( SCRATCH_SIZE*sizeof(float), 4096 ); + } + /* check for sufficient scratch size */ + if ( (*transa == 'N') && (*transb == 'N') ) { + if ( ((ptlda*myk)+(ptldb*myn)) > SCRATCH_SIZE ) { + return -1; + } + } else if ( (*transa == 'T') && (*transb == 'N') ) { + if ( ((ptlda*mym)+(ptldb*myn)) > SCRATCH_SIZE ) { + return -2; + } + mylda = mym; + } else if ( (*transa == 'N') && (*transb == 'T') ) { + if ( ((ptlda*myk)+(ptldb*myk)) > SCRATCH_SIZE ) { + return -3; + } + myldb = myk; + } else if ( (*transa == 'T') && (*transb == 'T') ) { + if ( ((ptlda*mym)+(ptldb*myk)) > SCRATCH_SIZE ) { + return -4; + } + mylda = mym; + myldb = myk; + } else { + assert((0 && "Error : Invalid parameters")); + return -5; + } + + /* set temp A and B pointers */ + myA = myscratch; + myB = myscratch + (mylda*myk); + myC = myscratch + (mylda*myk) + (myldb*myn); + + if ( *transa == 'T' ) { + /* fill the padding with zeros */ + for ( p = 0; p < k; p++ ) { + for ( o = m-1; o < mym; o++ ) { + myA[p*mylda+o] = 0.0; + } + } + for ( p = k-1; p < myk; p++ ) { + for ( o = 0; o < mym; o++ ) { + myA[p*mylda+o] = 0.0; + } + } + + /* let's transpose data first */ +#if defined(_OPENMP) +#pragma omp parallel for private(o,p) collapse(2) +#endif + for ( p = 0; p < k; p++ ) { + for ( o = 0; o < m; o++ ) { + myA[(p*mylda)+o] = a[(o*ptlda)+p]; + } + } + } else if ( m_pad > 0 || k_pad > 0 ) { + copy_matrix_and_pad(a, myA, m, k, m_pad, k_pad); + } else { + myA = a; + } + + if ( *transb == 'T' ) { + /* fill the padding with zeros */ + for ( p = 0; p < n; p++ ) { + for ( o = k-1; o < myk; o++ ) { + myB[p*myldb+o] = 0.0; + } + } + for ( p = n-1; p < myn; p++ ) { + for ( o = 0; o < myk; o++ ) { + myB[p*myldb+o] = 0.0; + } + } + + /* let's transpose data first */ +#if defined(_OPENMP) +#pragma omp parallel for private(o,p) collapse(2) +#endif + for ( p = 0; p < n; p++ ) { + for ( o = 0; o < k; o++ ) { + myB[(p*myldb)+o] = b[(o*ptldb)+p]; + } + } + } else if ( k_pad > 0 || n_pad > 0 ) { + copy_matrix_and_pad(b, myB, k, n, k_pad, n_pad); + } else { + myB = b; + } + + if ( m_pad > 0 || n_pad > 0 ) { + copy_matrix_and_pad(c, myC, m, n, m_pad, n_pad); + } else { + myC = c; + } + /* run gemm */ +#if defined(_OPENMP) +#pragma omp parallel for private(o,p,q,pp,oo) collapse(2) +#endif + for ( o = 0; o < mym; o += 16 ) { + for ( p = 0; p < myn; p += 16 ) { + float ctmp[256]; + for ( pp = 0; pp < 16; pp++ ) { + for ( oo = 0; oo < 16; oo++ ) { + ctmp[(pp*16)+oo] = 0.0f; + } + } + for ( q = 0; q < myk; q += 64 ) { + MMEngine_avx2_ps(mym, myn, myk, alpha, &(myA[(mylda*q)+o]), mylda, + &(myB[(myldb*p)+q]), myldb, beta, ctmp, myldc); + } + for ( pp = 0; pp < 16; pp++ ) { + for ( oo = 0; oo < 16; oo++ ) { + myC[((p+pp)*myldc)+(o+oo)] += ((alpha)*ctmp[(pp*16)+oo]) + ((beta)*myC[((p+pp)*myldc)+(o+oo)]); + } + } + } + } + if ( m_pad > 0 || n_pad > 0 ) { + copy_matrix_and_strip_pading(myC, c, m, n, m_pad, n_pad); + } + return 0; +} diff --git a/FP8_Emulator/cmodel/simple/simple_mm_engine.cpp b/FP8_Emulator/cmodel/simple/simple_mm_engine.cpp new file mode 100644 index 00000000..3ba40fb6 --- /dev/null +++ b/FP8_Emulator/cmodel/simple/simple_mm_engine.cpp @@ -0,0 +1,45 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include + +/* column major dat format */ +void MMEngine_avx2_ps(int m, int n, int k, float alpha, float *A, int lda, + float *B, int ldb, float beta, float *C, int ldc) +{ + for (int j = 0; j < n; j++) { + for (int i = 0; i < m; i+=8) { + __m256 cij = _mm256_loadu_ps(&C[ j*ldc + i]); + for (int l = 0; l < k; l++) { + __m256 aik = _mm256_loadu_ps(&A[ i + l * lda]); + __m256 bkj = _mm256_broadcast_ss(&B[ l + j * ldb]); + cij = _mm256_add_ps(cij, _mm256_mul_ps(aik, bkj)); + } + _mm256_storeu_ps(&C[ j * ldc + i ], cij); + } + } +} + +/* column major dat format */ +void MMEngine_strideB_avx2_ps(int m, int n, int k, float alpha, float *A, int lda, + float *B, int ldb, float beta, float *C, int ldc, int strideB) +{ + for (int j = 0; j < n; j++) { + for (int i = 0; i < m; i+=8) { + __m256 cij = _mm256_loadu_ps(&C[ j*ldc + i]); + for (int l = 0; l < k; l++) { + __m256 aik = _mm256_loadu_ps(&A[ i + l * lda]); + __m256 bkj = _mm256_broadcast_ss(&B[ l*strideB + j * ldb]); + cij = _mm256_add_ps(cij, _mm256_mul_ps(aik, bkj)); + } + _mm256_storeu_ps(&C[ j * ldc + i], cij); + } + } +} + diff --git a/FP8_Emulator/cmodel/simple/vla.h b/FP8_Emulator/cmodel/simple/vla.h new file mode 100644 index 00000000..f461878d --- /dev/null +++ b/FP8_Emulator/cmodel/simple/vla.h @@ -0,0 +1,47 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#ifndef VLA_H +#define VLA_H +#include +#include +#include +#include + +#define ALWAYS_INLINE __attribute__((always_inline)) +#define INLINE inline +#define RESTRICT __restrict__ +#define VLA_POSTFIX _ + +#define INDEX1_1(...) ((size_t)SELECT_HEAD(__VA_ARGS__)) +#define INDEX1_2(I0, I1, S1) (INDEX1_1(I0) * ((size_t)S1) + (size_t)I1) +#define INDEX1_3(I0, I1, I2, S1, S2) (INDEX1_2(I0, I1, S1) * ((size_t)S2) + (size_t)I2) +#define INDEX1_4(I0, I1, I2, I3, S1, S2, S3) (INDEX1_3(I0, I1, I2, S1, S2) * ((size_t)S3) + (size_t)I3) +#define INDEX1_5(I0, I1, I2, I3, I4, S1, S2, S3, S4) (INDEX1_4(I0, I1, I2, I3, S1, S2, S3) * ((size_t)S4) + (size_t)I4) +#define INDEX1_6(I0, I1, I2, I3, I4, I5, S1, S2, S3, S4, S5) (INDEX1_5(I0, I1, I2, I3, I4, S1, S2, S3, S4) * ((size_t)S5) + (size_t)I5) +#define INDEX1_7(I0, I1, I2, I3, I4, I5, I6, S1, S2, S3, S4, S5, S6) (INDEX1_6(I0, I1, I2, I3, I4, I5, S1, S2, S3, S4, S5) * ((size_t)S6) + (size_t)I6) +#define INDEX1_8(I0, I1, I2, I3, I4, I5, I6, I7, S1, S2, S3, S4, S5, S6, S7) (INDEX1_7(I0, I1, I2, I3, I4, I5, I6, S1, S2, S3, S4, S5, S6) * ((size_t)S7) + (size_t)I7) +#define INDEX1_9(I0, I1, I2, I3, I4, I5, I6, I7, I8, S1, S2, S3, S4, S5, S6, S7, S8) (INDEX1_8(I0, I1, I2, I3, I4, I5, I6, I7, S1, S2, S3, S4, S5, S6, S7) * ((size_t)S8) + (size_t)I8) +#define INDEX1_10(I0, I1, I2, I3, I4, I5, I6, I7, I8, I9, S1, S2, S3, S4, S5, S6, S7, S8, S9) (INDEX1_9(I0, I1, I2, I3, I4, I5, I6, I7, I8, S1, S2, S3, S4, S5, S6, S7, S8) * ((size_t)S9) + (size_t)I9) + +#define EXPAND(...) __VA_ARGS__ +#define CONCATENATE2(A, B) A##B +#define CONCATENATE(A, B) CONCATENATE2(A, B) +#define INDEX1(NDIMS, ...) CONCATENATE(INDEX1_, NDIMS)(__VA_ARGS__) + +#define SELECT_HEAD_AUX(A, ...) (A) +#define SELECT_HEAD(...) EXPAND(SELECT_HEAD_AUX(__VA_ARGS__, 0)) +#define SELECT_TAIL(A, ...) __VA_ARGS__ + +#define ACCESS_VLA(NDIMS, ARRAY, ...) CONCATENATE(ARRAY, VLA_POSTFIX)[INDEX1(NDIMS, __VA_ARGS__)] +#define DECLARE_VLA(NDIMS, ELEMENT_TYPE, ARRAY_VAR, ...) \ + ELEMENT_TYPE *RESTRICT CONCATENATE(ARRAY_VAR, VLA_POSTFIX) = SELECT_HEAD(__VA_ARGS__) \ + + 0 * INDEX1(NDIMS, SELECT_TAIL(__VA_ARGS__, SELECT_TAIL(__VA_ARGS__, 0))) + +#endif diff --git a/FP8_Emulator/cmodel/tests/conv_grad_test.py b/FP8_Emulator/cmodel/tests/conv_grad_test.py new file mode 100644 index 00000000..74b0c52e --- /dev/null +++ b/FP8_Emulator/cmodel/tests/conv_grad_test.py @@ -0,0 +1,58 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +import time +from mpemu.cmodel import simple + +n = 64 +c = 256 +h = 28 +w = 28 +k = 512 +r = 3 +s = 3 +stride = 2 +pad = 2 + +a = torch.rand((n,c,h,w), dtype=torch.float32) +b = torch.rand((k,c,r,s), dtype=torch.float32) +bias = torch.rand((k), dtype=torch.float32) + +a64 = a.to(dtype=torch.float64, copy=True) +b64 = b.to(dtype=torch.float64, copy=True) + +a.requires_grad=True +b.requires_grad=True +a64.requires_grad=True +b64.requires_grad=True + +ref_time = time.time() +z = torch.nn.functional.conv2d(a64, b64, bias, stride=(stride,stride), padding=(pad,pad), dilation=(1,1), groups=1) +ref_time = time.time()-ref_time + +simple_time = time.time() +z2 = simple.conv2d(a, b, bias, stride=(stride,stride), padding=(pad,pad), dilation=(1,1), groups=1) +simple_time = time.time()-simple_time + +#print("Forward Time : ref_time: {}, simple_time: {} ".format(ref_time, simple_time)) +print('Forward: L2 distance output : ', torch.dist(z2.to(dtype=torch.float64, copy=True), z, 2).item()) + +ref_time = time.time() +(z[0, 0] + z[0,1]).sum().backward() +ref_time = time.time()-ref_time + +simple_time = time.time() +(z2[0, 0] + z2[0,1]).sum().backward() +simple_time = time.time()-simple_time + +#print("BackProp Time : ref_time: {}, simple_time: {}".format(ref_time, simple_time)) +torch.set_printoptions(profile="full") +print('Backward: L2 distance input_grad: ', torch.dist(a.grad.to(dtype=torch.float64, copy=True), a64.grad, 2).item()) +print('Backward: L2 distance weight_grad: ', torch.dist(b.grad.to(dtype=torch.float64, copy=True), b64.grad, 2).item()) diff --git a/FP8_Emulator/cmodel/tests/conv_test.py b/FP8_Emulator/cmodel/tests/conv_test.py new file mode 100644 index 00000000..83babee4 --- /dev/null +++ b/FP8_Emulator/cmodel/tests/conv_test.py @@ -0,0 +1,30 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +from mpemu.cmodel import simple + +n = 2 +c = 64 +h = 28 +w = 28 +k = 64 +r = 3 +s = 3 + +input = torch.rand((n,c,h,w), dtype=torch.float32) +weight = torch.rand((k,c,r,s), dtype=torch.float32) +bias = torch.rand((k), dtype=torch.float32) + +output_ref = torch.nn.functional.conv2d(input, weight, bias, stride=(2,2), padding=(2,2), dilation=(1,1), groups=1) +output_simple = simple.conv2d(input, weight, bias, stride=(2,2), padding=(2,2), dilation=(1,1), groups=1) + +torch.set_printoptions(profile="full") + +print('Forward : L2 distance (simple) : ', torch.dist(output_ref, output_simple, 2).item()) diff --git a/FP8_Emulator/cmodel/tests/gemm_grad_test.py b/FP8_Emulator/cmodel/tests/gemm_grad_test.py new file mode 100644 index 00000000..4f6b2ec5 --- /dev/null +++ b/FP8_Emulator/cmodel/tests/gemm_grad_test.py @@ -0,0 +1,43 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +from mpemu.cmodel import simple + +m=1024 #356320 #356305 +n=1024 #120 #256 +k=1024 #576 #128 #2020 + +# Start here +a = torch.rand((m, k), dtype=torch.float32) +b = torch.rand((n, k), dtype=torch.float32) +c = torch.zeros((m, n), dtype=torch.float32) + +a64 = a.to(dtype=torch.float64, copy=True) +b64 = b.to(dtype=torch.float64, copy=True) +c64 = c.to(dtype=torch.float64, copy=True) + +a.requires_grad=True +b.requires_grad=True +c.requires_grad=True +a64.requires_grad=True +b64.requires_grad=True +c64.requires_grad=True + +z = torch.addmm(c64, a64, b64.t()) +z2 = simple.addmm(c, a, b.t()) + +print('Forward :L2 distance output: ', torch.dist(z2.to(dtype=torch.float64, copy=True), z, 2).item()) + +(z2[0, 0] + z2[0,1]).sum().backward() +(z[0, 0] + z[0,1]).sum().backward() + +print('Backward : L2 distance a_grad: ', torch.dist(a.grad.to(dtype=torch.float64, copy=True), a64.grad, 2).item()) +print('Backward : L2 distance b_grad: ', torch.dist(b.grad.to(dtype=torch.float64, copy=True), b64.grad, 2).item()) +print('Backward : L2 distance c_grad: ', torch.dist(c.grad.to(dtype=torch.float64, copy=True), c64.grad, 2).item()) diff --git a/FP8_Emulator/cmodel/tests/gemm_irregular_test.py b/FP8_Emulator/cmodel/tests/gemm_irregular_test.py new file mode 100644 index 00000000..50541b8d --- /dev/null +++ b/FP8_Emulator/cmodel/tests/gemm_irregular_test.py @@ -0,0 +1,42 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +import fuse + +m=1 #356320 #356305 +n=120 #256 +k=576 #128 #2020 + +# Start here +a = torch.rand((5, 107, 1024), dtype=torch.float32) +b = torch.rand((1024, 1024), dtype=torch.float32) +c = torch.zeros((m, n), dtype=torch.float32) + +a64 = a.to(dtype=torch.float64, copy=True) +b64 = b.to(dtype=torch.float64, copy=True) +c64 = c.to(dtype=torch.float64, copy=True) + +z = torch.matmul(a64, b64) +z3 = torch.matmul(a, b) +for a1 in a : + print("-->", a1.size()) + +#z2l = tuple([fuse.tmul.matmul(a1, b) for a1 in a]) +#z2 = torch.stack(z2l) +#z2 = torch.stack(tuple([fuse.tmul.matmul(a1, b) for a1 in a])) +z2 = fuse.tmul.matmul(a, b) + +#print (z3.size(), z2.size()) + +#print("torch :", z3.size(), z3) +#print("Ours : ", z2.size(), z2) +print('32b: L2 distance : ', torch.dist(z, z3.to(dtype=torch.float64, copy=True), 2)) +print('ours: L2 distance : ', torch.dist(z, z2.to(dtype=torch.float64, copy=True), 2)) + diff --git a/FP8_Emulator/cmodel/tests/gemm_test.py b/FP8_Emulator/cmodel/tests/gemm_test.py new file mode 100644 index 00000000..f42b8c7f --- /dev/null +++ b/FP8_Emulator/cmodel/tests/gemm_test.py @@ -0,0 +1,41 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +from mpemu.cmodel import simple + +def get_grads(variables): + return [var.grad.clone() for var in variables] + +m=128 #356320 #356305 +n=120 #256 +k=576 #128 #2020 + +a = torch.rand((m, k), dtype=torch.float32) +b = torch.rand((n, k), dtype=torch.float32) +c = torch.zeros((m, n), dtype=torch.float32) + + +a64 = a.to(dtype=torch.float64, copy=True) +b64 = b.to(dtype=torch.float64, copy=True) +c64 = c.to(dtype=torch.float64, copy=True) + +z = torch.matmul(a64, b64.t()) +z2 = simple.matmul(a, b.t()) +z3 = torch.matmul(a, b.t()) + +z2gb = torch.matmul(z2, b) +z2g = simple.matmul(z2, b) +z2gwb = torch.matmul(z2.t(), a) +z2gw = simple.matmul(z2.t(), a) + +print('32b: L2 distance : ', torch.dist(z, z3.to(dtype=torch.float64, copy=True), 2)) +print('output : L2 distance : ', torch.dist(z, z2.to(dtype=torch.float64, copy=True), 2)) +print('Grad a : L2 distance : ', torch.dist(z2g, z2gb, 2)) +print('Grad b : L2 distance : ', torch.dist(z2gw, z2gwb, 2)) diff --git a/FP8_Emulator/cmodel/tests/linear_test.py b/FP8_Emulator/cmodel/tests/linear_test.py new file mode 100644 index 00000000..c5306975 --- /dev/null +++ b/FP8_Emulator/cmodel/tests/linear_test.py @@ -0,0 +1,65 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +import torch.nn as nn +import torch.nn.functional as F +import torch.optim as optim +from mpemu.cmodel import simple + +class Net(nn.Module): + + def __init__(self): + super(Net, self).__init__() + self.fc = nn.Linear(576, 120) + + def forward(self, x): + x = self.fc(x) + return x + + def num_flat_features(self, x): + size = x.size()[1:] # all dimensions except the batch dimension + num_features = 1 + for s in size: + num_features *= s + return num_features + +net = Net() +net1 = Net() +print(net) +input = torch.randn((1, 576), dtype=torch.float32, device="cpu") +input_new = input.to(dtype=torch.float32, copy=True) + +output = net(input) +#print("fc output:", output.size(), output) + +target = torch.randn(120, dtype=torch.float32, device="cpu") # a dummy target, for example +target = target.view(1, -1) # make it the same shape as output +criterion = nn.MSELoss() +loss = criterion(output, target) +net.zero_grad() # zeroes the gradient buffers of all parameters +#loss.backward(retain_graph=True) +loss.backward() +print("fc weight grads:", net.fc.weight.grad) + + +torch.addmm_back = torch.addmm +torch.matmul_back = torch.matmul +torch.addmm = simple.addmm +torch.matmul = simple.matmul + +output1 = net1(input_new) +#print("fc output:", output1.size(), output1) + +loss1 = criterion(output1, target) +net1.zero_grad() # zeroes the gradient buffers of all parameters +loss1.backward() +print("fc weight grads:", net1.fc.weight.grad) + +print('Linear wtgrads L2 distance : ', torch.dist(net.fc.weight.grad, net1.fc.weight.grad, 2)) diff --git a/FP8_Emulator/cmodel/tests/net.py b/FP8_Emulator/cmodel/tests/net.py new file mode 100644 index 00000000..8f2e2c74 --- /dev/null +++ b/FP8_Emulator/cmodel/tests/net.py @@ -0,0 +1,73 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import torch +import torch.nn as nn +import torch.nn.functional as F +import torch.optim as optim +from mpemu.cmodel import simple + +torch_conv2d = torch.nn.functional.conv2d +torch_addmm = torch.addmm +torch.addmm = simple.addmm +torch.nn.functional.conv2d = simple.conv2d + +class Net(nn.Module): + + def __init__(self): + super(Net, self).__init__() + # 1 input image channel, 6 output channels, 3x3 square convolution + # kernel + self.conv1 = nn.Conv2d(1, 6, 3) + self.conv2 = nn.Conv2d(6, 16, 3) + # an affine operation: y = Wx + b + self.fc1 = nn.Linear(16 * 6 * 6, 120) # 6*6 from image dimension + self.fc2 = nn.Linear(120, 84) + self.fc3 = nn.Linear(84, 10) + + def forward(self, x): + # Max pooling over a (2, 2) window + x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2)) + # If the size is a square you can only specify a single number + x = F.max_pool2d(F.relu(self.conv2(x)), 2) + x = x.view(-1, self.num_flat_features(x)) + x = F.relu(self.fc1(x)) + x = F.relu(self.fc2(x)) + x = self.fc3(x) + return x + + def num_flat_features(self, x): + size = x.size()[1:] # all dimensions except the batch dimension + num_features = 1 + for s in size: + num_features *= s + return num_features + + +net = Net() +print(net) +params = list(net.parameters()) +input = torch.randn(1, 1, 32, 32, dtype=torch.float32, device="cpu") +# create your optimizer +optimizer = optim.SGD(net.parameters(), lr=0.1) +optimizer.zero_grad() # zero the gradient buffers + +output = net(input) + +print("fc3 output:", output) +target = torch.randn(10, dtype=torch.float32, device="cpu") # a dummy target, for example +target = target.view(1, -1) # make it the same shape as output +criterion = nn.MSELoss() + +loss = criterion(output, target) + +net.zero_grad() # zeroes the gradient buffers of all parameters + +loss.backward() +optimizer.step() # Does the update diff --git a/FP8_Emulator/pytquant/__init__.py b/FP8_Emulator/pytquant/__init__.py new file mode 100644 index 00000000..e04cab6a --- /dev/null +++ b/FP8_Emulator/pytquant/__init__.py @@ -0,0 +1,6 @@ +import warnings +import torch + +from . import cpp +if torch.cuda.is_available(): + from . import cuda diff --git a/FP8_Emulator/pytquant/cpp/__init__.py b/FP8_Emulator/pytquant/cpp/__init__.py new file mode 100644 index 00000000..60993b75 --- /dev/null +++ b/FP8_Emulator/pytquant/cpp/__init__.py @@ -0,0 +1 @@ +from . import fpemu as fpemu_cpp diff --git a/FP8_Emulator/pytquant/cpp/avx-fpemu.cpp b/FP8_Emulator/pytquant/cpp/avx-fpemu.cpp new file mode 100644 index 00000000..74f509ff --- /dev/null +++ b/FP8_Emulator/pytquant/cpp/avx-fpemu.cpp @@ -0,0 +1,802 @@ +#include +#include +#include +#include + +enum ROUNDING_MODES { + ROUND_RTZ = 0, + ROUND_RNE = 1, + ROUND_STOCHASTIC = 2, + ROUND_RNAZ = 3, + ROUND_RNTZ = 4, + ROUND_PINF = 5, + ROUND_NINF = 6 +}; // 鏋氫妇瀹氫箟浜嗕笉鍚岀殑鑸嶅叆妯″紡锛堝喅瀹氬浣曞皢涓涓诞鐐规暟鎴栨暣鏁拌繎浼间负鍙︿竴涓洿鎺ヨ繎鐨勫肩殑瑙勫垯锛 + +namespace { + + typedef union half_t { + unsigned short u; + at::Half f; + } __half_t; + + typedef union ufloat32 { + unsigned u; + float f; + } __float_t; + +/* Following implementation of xoroshiro128++ PRNG is borrowed from here: + http://prng.di.unimi.it/xoshiro128plusplus.c + main page: http://prng.di.unimi.it/ +*/ + static uint32_t s1_[4] = { 1387366120, 2798441831, 888998500 , 1099633400 }; + static uint32_t s2_[4] = { 2034269327, 2125325156, 1209715489, 1931656721 }; + static uint32_t s3_[4] = { 1555452618, 650181557 , 883695203 , 627677842 }; + static uint32_t s4_[4] = { 4195248041, 2146478152, 480059239 , 1468956197 }; + static uint32_t s5_[4] = { 1252084877, 500390994 , 977516591 , 1950666000 }; + static uint32_t s6_[4] = { 3936597502, 834151069 , 1477014702, 734008143 }; + static uint32_t s7_[4] = { 1983400973, 1164103095, 2110188261, 2019272068 }; + static uint32_t s8_[4] = { 1877096364, 2833629967, 4196320416, 1774181187 }; + static uint32_t s9_[4] = { 702309618 , 4077815558, 1512057936, 1868769368 }; + static uint32_t s10_[4] = + { 510001215 , 966559856 , 776583255 , 1475621065 }; + static uint32_t s11_[4] = + { 1271806057, 1881312534, 478635452 , 814821902 }; + static uint32_t s12_[4] = + { 733990058 , 1889991804, 1108257970, 1093480892 }; + static uint32_t s13_[4] = + { 4273743809, 4167473370, 558000409 , 1594848927 }; + static uint32_t s14_[4] = + { 444870959 , 1595722866, 1064124488, 3637102547 }; + static uint32_t s15_[4] = + { 703721499 , 3896407831, 1002360059, 1427395742 }; + static uint32_t s16_[4] = + { 1295231497, 1254972431, 1423497865, 861918264 }; + +/* seed pointer array */ + static uint32_t *sptr_[16] = { s1_, s2_, s3_, s4_, s5_, s6_, s7_, s8_, s9_, + s10_, s11_, s12_, s13_, s14_, s15_, s16_ + }; + + static inline uint32_t rotl (const uint32_t x, int k) { + return (x << k) | (x >> (32 - k)); + } + + uint32_t rand_xorshft128plus_scalar (uint32_t * ps) { + const uint32_t result_plus = ps[0] + ps[3]; + const uint32_t t = ps[1] << 9; + + ps[2] ^= ps[0]; + ps[3] ^= ps[1]; + ps[1] ^= ps[2]; + ps[0] ^= ps[3]; + + ps[2] ^= t; + + ps[3] = rotl (ps[3], 11); + + return result_plus; + } + + float __double2float_rn (double inval) { + float out[4] = { 0 }; + __m128 vout = _mm_cvtpd_ps (_mm_set1_pd (inval)); + + _mm_store_ps (&out[0], vout); + return out[0]; + } + + unsigned short __float2half_rn (float inval) { + return _cvtss_sh (inval, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + } + + float __half2float (unsigned short h_val) { + return _cvtsh_ss (h_val); + } + + template < typename scalar_t > float __anyfloat2float_rn (scalar_t a_) { + float f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = __double2float_rn (a_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = a_; + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = __half2float ((at::Half) a_); + } + return f_; + } + + template < typename scalar_t > + void __float2anyfloat_rn (float f_, scalar_t * out) { + scalar_t a_; + + if (std::is_same < scalar_t, double >::value) { + a_ = (scalar_t) (f_); + } else if (std::is_same < scalar_t, float >::value) { + a_ = f_; + } else if (std::is_same < scalar_t, at::Half >::value) { + a_ = (at::Half) __float2half_rn (f_); + } + *out = a_; + } + + template < typename scalar_t > + unsigned short __anyfloat2half_rn (scalar_t f_) { + unsigned short h_; + + if (std::is_same < scalar_t, double >::value) { + h_ = __float2half_rn (__double2float_rn (f_)); + } else if (std::is_same < scalar_t, float >::value) { + h_ = __float2half_rn (f_); + } else if (std::is_same < scalar_t, at::Half >::value) { + unsigned short *ptrh_ = (unsigned short *) &f_; + h_ = *ptrh_; + } + return h_; + } + + template < typename scalar_t > + void __half2anyfloat (unsigned short h_, scalar_t * out, scalar_t scale=1.0) { + scalar_t f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = (scalar_t) __half2float (h_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = __half2float (h_); + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = *((at::Half *) & h_); + } + *out = scale*f_; + } + + template < typename scalar_t > + inline void reduce0 (scalar_t * g_data, float *g_odata, unsigned int n) { + float sum = 0.0, sumsq = 0.0; + +#pragma omp parallel for reduction(+: sum), reduction(+: sumsq) + for (unsigned int i = 0; i < n; i++) { + sum += __anyfloat2float_rn (g_data[i]); + sumsq += __anyfloat2float_rn (g_data[i]) * __anyfloat2float_rn (g_data[i]); + } + g_odata[0] = sum; + g_odata[1] = sumsq; + } + + template < typename scalar_t > + inline void absmax0 (scalar_t * g_data, float *g_odata, unsigned int n) { + float absmax = 0.0; + +#pragma omp parallel for reduction(max: absmax) + for (unsigned int i = 0; i < n; i++) { + absmax = fmaxf (absmax, fabsf (__anyfloat2float_rn (g_data[i]))); + } + g_odata[0] = absmax; + } + + static inline __m256i _mm256_rand_xorshft128plus_epi32(uint32_t *vs0, + uint32_t *vs1, + uint32_t *vs2, + uint32_t *vs3) { + const __m256i vrplus = _mm256_add_epi32(_mm256_load_si256((__m256i *)vs0), + _mm256_load_si256((__m256i *)vs3)); + const __m256i vt = + _mm256_sll_epi32(_mm256_load_si256((__m256i *)vs1), _mm_cvtsi32_si128(9)); + + _mm256_store_si256((__m256i *)vs2, + _mm256_xor_si256(_mm256_load_si256((__m256i *)vs2), + _mm256_load_si256((__m256i *)vs0))); + _mm256_store_si256((__m256i *)vs3, + _mm256_xor_si256(_mm256_load_si256((__m256i *)vs3), + _mm256_load_si256((__m256i *)vs1))); + _mm256_store_si256((__m256i *)vs1, + _mm256_xor_si256(_mm256_load_si256((__m256i *)vs1), + _mm256_load_si256((__m256i *)vs2))); + _mm256_store_si256((__m256i *)vs0, + _mm256_xor_si256(_mm256_load_si256((__m256i *)vs0), + _mm256_load_si256((__m256i *)vs3))); + _mm256_store_si256((__m256i *)vs2, + _mm256_xor_si256(_mm256_load_si256((__m256i *)vs2), vt)); + + __m256i vl = _mm256_slli_epi32(_mm256_load_si256((__m256i *)vs3), 11); + __m256i vr = _mm256_srli_epi32(_mm256_load_si256((__m256i *)vs3), 32 - 11); + + _mm256_store_si256((__m256i *)vs3, _mm256_or_si256(vl, vr)); + + return vrplus; + } + + void cvt_fp32_e5m2_rne_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { + +#pragma omp parallel for + for (int i = 0; i < size; i += 16){ + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x007f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0100); + + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps(b, s_); + a = _mm256_mul_ps(a, s_); + + __m128i ah_ = _mm256_cvtps_ph(a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph(b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + const __m256i a_ = _mm256_insertf128_si256(_mm256_insertf128_si256(_mm256_setzero_si256(), bh_, 0), ah_, 1); + const __m256i maska1_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vnaninf), vnaninf); + const __m256i maska2_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vfixupmask), vfixupmask); + __m256i a_rne_ = _mm256_blendv_epi8(a_, _mm256_add_epi16(a_, vrneadd), _mm256_and_si256(maska1_, maska2_)); + a_rne_ = _mm256_slli_epi16(_mm256_srli_epi16(a_rne_, 8), 8); + + bh_ = _mm256_extracti128_si256(a_rne_, 0); + ah_ = _mm256_extracti128_si256(a_rne_, 1); + b = _mm256_cvtph_ps(bh_); + a = _mm256_cvtph_ps(ah_); + + _mm256_storeu_ps(&out[i], _mm256_mul_ps(b, sr_)); + _mm256_storeu_ps(&out[i + 8], _mm256_mul_ps(a, sr_)); + } + } + + void cvt_fp32_e5m2_stochastic_intrinsic(const float *__restrict__ in, float *out, + int size, float scale) { + uint32_t vs0[8] __attribute__((aligned(32))) = { + 1387366120, 279844183, 888998500, 1099633400, + 1252084877, 500390994, 977516591, 1950666000}; + uint32_t vs1[8] __attribute__((aligned(32))) = { + 2034269327, 2125325156, 1209715489, 193165672, + 187709636, 28336299, 419632041, 1774181187}; + uint32_t vs2[8] __attribute__((aligned(32))) = { + 1555452618, 650181557, 883695203, 62767784, + 127180605, 1881312534, 478635452, 814821902}; + uint32_t vs3[8] __attribute__((aligned(32))) = { + 419524804, 2146478152, 480059239, 1468956197, + 444870959, 1595722866, 1064124488, 363710254}; + +#pragma omp parallel for firstprivate(vs0, vs1, vs2, vs3) + for (int i = 0; i < size; i += 16) { + const __m256i vnaninf = _mm256_set1_epi16(0x7c00); + const __m256i vfixup = _mm256_set1_epi16(0x0001); + const __m256i vfixupmask = _mm256_set1_epi16(0x0100); + const __m256i vrneadd = _mm256_set1_epi16(0x007f); + const __m256i vdenorm = _mm256_set1_epi16(0x03ff); + const __m256i vexmant = _mm256_set1_epi16(0x7fff); + + __m256i rnd256 = _mm256_rand_xorshft128plus_epi32(vs0, vs1, vs2, vs3); + __m128i rnbits = _mm256_extractf128_si256(rnd256, 0); + + __m256 s_ = _mm256_set1_ps(scale); + __m256 sr_ = _mm256_set1_ps(1.0 / scale); + + __m256 b = _mm256_loadu_ps(&in[i]); + __m256 a = _mm256_loadu_ps(&in[i + 8]); + + b = _mm256_mul_ps(b, s_); + a = _mm256_mul_ps(a, s_); + + __m128i ah_ = _mm256_cvtps_ph(a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph(b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + + const __m256i a_ = _mm256_insertf128_si256(_mm256_castsi128_si256(bh_), ah_, 1); + const __m256i maska1_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vnaninf), vnaninf); + const __m256i maska2_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vfixupmask), vfixupmask); + const __m256i maska4_ = _mm256_cmpgt_epi16(vdenorm, _mm256_and_si256(a_, vexmant)); + + __m256i a_sr_ = _mm256_blendv_epi8(a_, _mm256_add_epi16(a_, _mm256_cvtepu8_epi16(rnbits)), _mm256_andnot_si256(maska4_, maska1_)); + a_sr_ = _mm256_blendv_epi8(a_sr_, _mm256_add_epi16(a_sr_, vrneadd), _mm256_and_si256(maska4_, maska2_)); + a_sr_ = _mm256_slli_epi16(_mm256_srli_epi16(a_sr_, 8), 8); + + bh_ = _mm256_extracti128_si256(a_sr_, 0); + ah_ = _mm256_extracti128_si256(a_sr_, 1); + + b = _mm256_cvtph_ps(bh_); + a = _mm256_cvtph_ps(ah_); + + _mm256_storeu_ps(&out[i], _mm256_mul_ps(b, sr_)); + _mm256_storeu_ps(&out[i + 8], _mm256_mul_ps(a, sr_)); + } + } + void cvt_fp32_e5m2_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 5 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x00FF; + unsigned short rne_tie = 0x0180; + + float scale_reciprocal = 1.0 / scale; + + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + + unsigned short can_round = ((h.u & 0x7F00) <= 0x7B00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + /* nearest rounding masks */ + unsigned short rnmask = (h.u & grs_bitmask); + unsigned short rnmask_tie = (h.u & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + h.u += can_round * is_normal * (rand & 0xFF); + /* stochastic round: denormals --> rne rounding */ + h.u += can_round * is_denorm * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + h.u += can_round * rne_mask * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + h.u += can_round * rnaz_mask * ((rnmask >= 0x0080) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + h.u += can_round * rntz_mask * ((rnmask > 0x0080) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + h.u += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0080) << lshift); + /* round to -INF, if rminf_mask is enabled */ + h.u += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0080) << lshift); + } + } + /* truncation */ + h.u = (h.u & mask_mant); + float f_; + __half2anyfloat (h.u, &f_); + out[gid] = f_ * scale_reciprocal; + } + } + void cvt_fp32_e4m3_rne_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x003f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0080); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x5F00);/* 2^8*1.110 a.k.a 448.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x1800);/* 2^-9, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x2400);/* 2^-6 smallest normal */ + + for (int i = 0; i < size; i += 16){ + __m256 s_ = _mm256_set1_ps(scale); + __m256 sr_ = _mm256_set1_ps(1.0 / scale); + __m256 b = _mm256_loadu_ps(&in[i]); + __m256 a = _mm256_loadu_ps(&in[i + 8]); + + b = _mm256_mul_ps(b, s_); + a = _mm256_mul_ps(a, s_); + + __m128i ah_ = _mm256_cvtps_ph(a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph(b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_insertf128_si256(_mm256_insertf128_si256(_mm256_setzero_si256(), bh_, 0), ah_, 1); + const __m256i maska1_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vnaninf), vnaninf); + const __m256i maska2_ = _mm256_cmpeq_epi16(_mm256_and_si256(a_, vfixupmask), vfixupmask); + const __m256i maska3_ = _mm256_cmpgt_epi16(vsatuval, _mm256_and_si256(a_, _mm256_set1_epi16(0x7FFF))); + const __m256i maska4_ = _mm256_cmpgt_epi16(vflush, _mm256_and_si256(a_, vnaninf)); + const __m256i maska5_ = _mm256_cmpgt_epi16(vxdnorm, _mm256_and_si256(a_, vnaninf)); + + __m256i v_shft = _mm256_sub_epi16(_mm256_srli_epi16(vxdnorm, 10), _mm256_srli_epi16(_mm256_and_si256(a_, vnaninf), 10)); + __m256i a_rne_ = _mm256_blendv_epi8(a_, _mm256_add_epi16(a_rne_, vrneadd), _mm256_and_si256(maska1_, maska2_)); + a_rne_ = _mm256_slli_epi16(_mm256_srli_epi16(a_rne_, 8), 8); + bh_ = _mm256_extracti128_si256(a_rne_, 0); + ah_ = _mm256_extracti128_si256(a_rne_, 1); + b = _mm256_cvtph_ps(bh_); + a = _mm256_cvtph_ps(ah_); + _mm256_storeu_ps(&out[i], _mm256_mul_ps(b, sr_)); + _mm256_storeu_ps(&out[i + 8], _mm256_mul_ps(a, sr_)); + } + } + + void cvt_fp32_e4m3_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16(0x7c00); + const __m256i vfixup = _mm256_set1_epi16(0x0001); + const __m256i vfixupmask = _mm256_set1_epi16(0x0100); + const __m256i vrneadd = _mm256_set1_epi16(0x003f); + const __m256i vdenorm = _mm256_set1_epi16(0x03ff); + const __m256i vexmant = _mm256_set1_epi16(0x7fff); + + for (int i = 0; i < size; i += 16) { + unsigned int rndbuf[16]; + /* generate 128 random bits */ + for (int r = 0; r < 8; r++) { + rndbuf[r] = (unsigned int) rand_xorshft128plus_scalar (sptr_[r]); + } + __m128i rnbits = _mm_load_si128 ((const __m128i *) &rndbuf[0]); + + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = + _mm256_cvtps_ph(a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = + _mm256_cvtps_ph(b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + + const __m256i a_ = + _mm256_insertf128_si256(_mm256_castsi128_si256(bh_), ah_, 1); + + const __m256i maska1_ = + _mm256_cmpeq_epi16(_mm256_and_si256(a_, vnaninf), vnaninf); + const __m256i maska2_ = + _mm256_cmpeq_epi16(_mm256_and_si256(a_, vfixupmask), vfixupmask); + const __m256i maska4_ = + _mm256_cmpgt_epi16(vdenorm, _mm256_and_si256(a_, vexmant)); + + __m256i a_sr_ = _mm256_blendv_epi8( + a_, _mm256_add_epi16(a_, _mm256_cvtepu8_epi16(rnbits)), + _mm256_andnot_si256(maska4_, maska1_)); + + a_sr_ = _mm256_blendv_epi8(a_sr_, _mm256_add_epi16(a_sr_, vrneadd), + _mm256_and_si256(maska4_, maska2_)); + + a_sr_ = _mm256_slli_epi16(_mm256_srli_epi16(a_sr_, 8), 8); + + bh_ = _mm256_extracti128_si256(a_sr_, 0); + ah_ = _mm256_extracti128_si256(a_sr_, 1); + + b = _mm256_cvtph_ps(bh_); + a = _mm256_cvtph_ps(ah_); + + _mm256_storeu_ps(&out[i], _mm256_mul_ps(b, sr_)); + _mm256_storeu_ps(&out[i + 8], _mm256_mul_ps(a, sr_)); + } + } + void cvt_fp32_e4m3_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 4 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + float scale_reciprocal = 1.0 / scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x5F00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 8 || (can_round == 0)) { + /* Software : saturate values above to +/-448.0 to +/-448.0 */ + mantissa_h = 0x0300; + exp_h = 8; + can_round = 0; + } else if (exp_h < -9) { + /* flush values below 1-4-3 subnormal range to zero */ + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -6) { + dshift = (-6 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + float f; + + __half2anyfloat (h.u, &f); + out[gid] = (f * scale_reciprocal); + } + } + + template < typename scalar_t > + void E4M3_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 8.0; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e4m3_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (in, out, vec_size, scale); + } + } + cvt_fp32_e4m3_scalar (&in[vec_size], &out[vec_size], size - vec_size, scale, rmode); + } + } + } + + template < typename scalar_t > + void E5M2_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 16384.0; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + //cvt_fp32_e5m2_flex_intrinsic(&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e5m2_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (in, out, size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (in, out, vec_size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (in, out, vec_size, scale); + //cvt_fp32_e5m2_flex_intrinsic(in, out, vec_size, scale); + } + } + cvt_fp32_e5m2_scalar (&in[vec_size], &out[vec_size], size - vec_size, scale, rmode); + } + } + } + + std::vector < torch::Tensor > fpemu_common_function (torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + + torch::Tensor output; + if (!inplace) + output = torch::zeros_like (input); + + if (!mode.compare ("E4M3_STOCHASTIC")) { + E4M3_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E4M3_RNE")) { + E4M3_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_RNE); + } else if (!mode.compare ("E5M2_STOCHASTIC")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E5M2_RNE")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_RNE); + } + + if (!inplace) { + return { + output,}; + } else { + return { + input,}; + } + } + +}//namespace + +std::vector < torch::Tensor > fpemu_forward (torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + if (block_norm == true && block_size != size) { + if (size % block_size) { + block_norm = false; + block_size = 1; + } + } + return fpemu_common_function (input, mode, size, inplace, scale, block_norm, + block_size); +} + +std::vector < torch::Tensor > fpemu_backward (torch::Tensor grad, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + if (block_norm == true && block_size != size) { + if (size % block_size) { + block_norm = false; + block_size = 1; + } + } + return fpemu_common_function (grad, mode, size, inplace, scale, block_norm, + block_size); +} + +PYBIND11_MODULE (TORCH_EXTENSION_NAME, m) { + m.def ("forward", &fpemu_forward, "FPEmu forward"); + m.def ("backward", &fpemu_backward, "FPEmu backward"); +} \ No newline at end of file diff --git a/FP8_Emulator/pytquant/cpp/fpemu.py b/FP8_Emulator/pytquant/cpp/fpemu.py new file mode 100644 index 00000000..9503c7c6 --- /dev/null +++ b/FP8_Emulator/pytquant/cpp/fpemu.py @@ -0,0 +1,73 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import math +from torch import nn +from torch.autograd import Function +import torch +import numpy +import fpemu_cpp + +from enum import Enum + +torch.manual_seed(42) + +""" + NONE + E5M2_RTZ + E5M2_STOCHASTIC + E5M2_RNE + E5M2_RNAZ + E5M2_RNTZ + E5M2_RPINF + E5M2_RNINF + E5M2_DAZ_STOCHASTIC + E5M2_DAZ_RNE + E5M2_DAZ_RNAZ + E5M2_DAZ_RNTZ + BFLOAT16_STOCHASTIC + BFLOAT16_RNE + FLOAT16_RNE + FLOAT16_STOCHASTIC + FLOAT16_DAZ_RNE + E4M3_RNE + E4M3_STOCHASTIC +""" + +class FPEmuOp(Function): + @staticmethod + def forward(ctx, input, mode='NONE', inplace=False, scale=1.0, blocknorm=False, blocksize=1): + if mode == 'NONE' : + ctx.mark_dirty(input) + return input + else : + if input.is_sparse : + input = input.coalesce() + size = input.values().nelement() + if inplace == True: + outputs = fpemu_cpp.forward(input._values().contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = input + else : + outputs = fpemu_cpp.forward(input._values().contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = torch.sparse.FloatTensor(input.indices(), outputs[0], input.size()) + else : + input = input.cpu() + size = input.nelement() + outputs = fpemu_cpp.forward(input.contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = outputs[0] + + if inplace == True: + ctx.mark_dirty(input) + + return output + + @staticmethod + def backward(ctx, output_grad): + # straight-through estimator + return output_grad, None, None, None, None diff --git a/FP8_Emulator/pytquant/cpp/fpemu_impl.cpp b/FP8_Emulator/pytquant/cpp/fpemu_impl.cpp new file mode 100644 index 00000000..262b07cd --- /dev/null +++ b/FP8_Emulator/pytquant/cpp/fpemu_impl.cpp @@ -0,0 +1,2009 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include + +enum ROUNDING_MODES { + ROUND_RTZ = 0, + ROUND_RNE = 1, + ROUND_STOCHASTIC = 2, + ROUND_RNAZ = 3, + ROUND_RNTZ = 4, + ROUND_PINF = 5, + ROUND_NINF = 6 +}; + +namespace { + + typedef union half_t { + unsigned short u; + at::Half f; + } __half_t; + + typedef union ufloat32 { + unsigned u; + float f; + } __float_t; + +/* Following implementation of xoroshiro128++ PRNG is borrowed from here: + http://prng.di.unimi.it/xoshiro128plusplus.c + main page: http://prng.di.unimi.it/ +*/ + static uint32_t s1_[4] = { 1387366120, 2798441831, 888998500 , 1099633400 }; + static uint32_t s2_[4] = { 2034269327, 2125325156, 1209715489, 1931656721 }; + static uint32_t s3_[4] = { 1555452618, 650181557 , 883695203 , 627677842 }; + static uint32_t s4_[4] = { 4195248041, 2146478152, 480059239 , 1468956197 }; + static uint32_t s5_[4] = { 1252084877, 500390994 , 977516591 , 1950666000 }; + static uint32_t s6_[4] = { 3936597502, 834151069 , 1477014702, 734008143 }; + static uint32_t s7_[4] = { 1983400973, 1164103095, 2110188261, 2019272068 }; + static uint32_t s8_[4] = { 1877096364, 2833629967, 4196320416, 1774181187 }; + static uint32_t s9_[4] = { 702309618 , 4077815558, 1512057936, 1868769368 }; + static uint32_t s10_[4] = + { 510001215 , 966559856 , 776583255 , 1475621065 }; + static uint32_t s11_[4] = + { 1271806057, 1881312534, 478635452 , 814821902 }; + static uint32_t s12_[4] = + { 733990058 , 1889991804, 1108257970, 1093480892 }; + static uint32_t s13_[4] = + { 4273743809, 4167473370, 558000409 , 1594848927 }; + static uint32_t s14_[4] = + { 444870959 , 1595722866, 1064124488, 3637102547 }; + static uint32_t s15_[4] = + { 703721499 , 3896407831, 1002360059, 1427395742 }; + static uint32_t s16_[4] = + { 1295231497, 1254972431, 1423497865, 861918264 }; + +/* seed pointer array */ + static uint32_t *sptr_[16] = { s1_, s2_, s3_, s4_, s5_, s6_, s7_, s8_, s9_, + s10_, s11_, s12_, s13_, s14_, s15_, s16_ + }; + + static inline uint32_t rotl (const uint32_t x, int k) { + return (x << k) | (x >> (32 - k)); + } + + uint32_t rand_xorshft128plus_scalar (uint32_t * ps) { + const uint32_t result_plus = ps[0] + ps[3]; + const uint32_t t = ps[1] << 9; + + ps[2] ^= ps[0]; + ps[3] ^= ps[1]; + ps[1] ^= ps[2]; + ps[0] ^= ps[3]; + + ps[2] ^= t; + + ps[3] = rotl (ps[3], 11); + + return result_plus; + } + + inline __m512i + _mm512_rndxorshft128plus_epi32 (uint32_t * vs0, uint32_t * vs1, + uint32_t * vs2, uint32_t * vs3) { + __m512i vrplus = _mm512_add_epi32 (_mm512_load_epi32 (vs0), _mm512_load_epi32 (vs3)); + __m512i vt = _mm512_sll_epi32 (_mm512_load_epi32 (vs1), _mm_set1_epi8 (9)); + _mm512_store_epi32 (vs2, _mm512_xor_epi32 (_mm512_load_epi32 (vs2), _mm512_load_epi32 (vs0))); + _mm512_store_epi32 (vs3, _mm512_xor_epi32 (_mm512_load_epi32 (vs3), _mm512_load_epi32 (vs1))); + _mm512_store_epi32 (vs1, _mm512_xor_epi32 (_mm512_load_epi32 (vs1), _mm512_load_epi32 (vs2))); + _mm512_store_epi32 (vs0, _mm512_xor_epi32 (_mm512_load_epi32 (vs0), _mm512_load_epi32 (vs3))); + _mm512_store_epi32 (vs2, _mm512_xor_epi32 (_mm512_load_epi32 (vs2), vt)); + + __m512i vl = _mm512_sll_epi32 (_mm512_load_epi32 (vs3), _mm_set1_epi8 (11)); + __m512i vr = _mm512_sra_epi32 (_mm512_load_epi32 (vs3), _mm_set1_epi8 (21)); + _mm512_store_epi32 (vs3, _mm512_or_epi32 (vl, vr)); + return vrplus; + } + + + float __double2float_rn (double inval) { + float out[4] = { 0 }; + __m128 vout = _mm_cvtpd_ps (_mm_set1_pd (inval)); + + _mm_store_ps (&out[0], vout); + return out[0]; + } + + unsigned short __float2half_rn (float inval) { + return _cvtss_sh (inval, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + } + + float __half2float (unsigned short h_val) { + return _cvtsh_ss (h_val); + } + + template < typename scalar_t > float __anyfloat2float_rn (scalar_t a_) { + float f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = __double2float_rn (a_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = a_; + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = __half2float ((at::Half) a_); + } + return f_; + } + + template < typename scalar_t > + void __float2anyfloat_rn (float f_, scalar_t * out) { + scalar_t a_; + + if (std::is_same < scalar_t, double >::value) { + a_ = (scalar_t) (f_); + } else if (std::is_same < scalar_t, float >::value) { + a_ = f_; + } else if (std::is_same < scalar_t, at::Half >::value) { + a_ = (at::Half) __float2half_rn (f_); + } + *out = a_; + } + + template < typename scalar_t > + unsigned short __anyfloat2half_rn (scalar_t f_) { + unsigned short h_; + + if (std::is_same < scalar_t, double >::value) { + h_ = __float2half_rn (__double2float_rn (f_)); + } else if (std::is_same < scalar_t, float >::value) { + h_ = __float2half_rn (f_); + } else if (std::is_same < scalar_t, at::Half >::value) { + unsigned short *ptrh_ = (unsigned short *) &f_; + h_ = *ptrh_; + } + return h_; + } + + template < typename scalar_t > + void __half2anyfloat (unsigned short h_, scalar_t * out, scalar_t scale=1.0) { + scalar_t f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = (scalar_t) __half2float (h_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = __half2float (h_); + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = *((at::Half *) & h_); + } + *out = scale*f_; + } + + template < typename scalar_t > + inline void reduce0 (scalar_t * g_data, float *g_odata, unsigned int n) { + float sum = 0.0, sumsq = 0.0; + +#pragma omp parallel for reduction(+: sum), reduction(+: sumsq) + for (unsigned int i = 0; i < n; i++) { + sum += __anyfloat2float_rn (g_data[i]); + sumsq += __anyfloat2float_rn (g_data[i]) * __anyfloat2float_rn (g_data[i]); + } + g_odata[0] = sum; + g_odata[1] = sumsq; + } + + template < typename scalar_t > + inline void absmax0 (scalar_t * g_data, float *g_odata, unsigned int n) { + float absmax = 0.0; + +#pragma omp parallel for reduction(max: absmax) + for (unsigned int i = 0; i < n; i++) { + absmax = fmaxf (absmax, fabsf (__anyfloat2float_rn (g_data[i]))); + } + g_odata[0] = absmax; + } + + + void cvt_fp32_bf16_rne_intrinsic (const float *__restrict__ in, float *out, + int size) { +#pragma omp parallel for + for (int i = 0; i < size; i += 16) { + const __m512i vnaninf = _mm512_set1_epi32 (0x7f800000), vrneadd = _mm512_set1_epi32 (0x00007fff); + const __m512i vfixup = _mm512_set1_epi32 (0x00000001), vfixupmask = _mm512_set1_epi32 (0x00010000); + const __m512i truncmask = _mm512_set1_epi32 (0xffff0000); + __m512 a = _mm512_loadu_ps (&in[i]); + const __m512i mm512_roundbf16rne_a_ = _mm512_castps_si512 (a); + const __mmask16 mm512_roundbf16rne_mask1_ = + _mm512_cmp_epi32_mask (_mm512_and_epi32 + (mm512_roundbf16rne_a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 mm512_roundbf16rne_mask2_ = + _mm512_cmp_epi32_mask (_mm512_and_epi32 + (mm512_roundbf16rne_a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + __m512i roundbf16rne = + _mm512_mask_add_epi32 (mm512_roundbf16rne_a_, + mm512_roundbf16rne_mask1_, + mm512_roundbf16rne_a_, + _mm512_mask_add_epi32 (vrneadd, + mm512_roundbf16rne_mask2_, vrneadd, vfixup)); + a = _mm512_castsi512_ps (_mm512_and_epi32 (roundbf16rne, truncmask)); + + _mm512_storeu_ps (&out[i], a); + } + } + + void cvt_fp32_bf16_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size) { + uint32_t vs0[16] __attribute__ ((aligned (64))) = { + 1387366120, 279844183, 888998500, 1099633400, 1252084877, 500390994, + 977516591, 1950666000, 393659750, 834151069, 1477014702, 734008143, + 1983400973, 116410309, 2110188261, 2019272068}; + uint32_t vs1[16] __attribute__ ((aligned (64))) = { + 2034269327, 2125325156, 1209715489, 193165672, 187709636, 28336299, + 419632041, 1774181187, 702309618, 407781555, 1512057936, 1868769368, + 510001215, 966559856, 776583255, 147562106}; + uint32_t vs2[16] __attribute__ ((aligned (64))) = { + 1555452618, 650181557, 883695203, 62767784, 127180605, 1881312534, + 478635452, 814821902, 733990058, 1889991804, 1108257970, 1093480892, + 427374380, 416747337, 558000409, 1594848927}; + uint32_t vs3[16] __attribute__ ((aligned (64))) = { + 419524804, 2146478152, 480059239, 1468956197, 444870959, 1595722866, + 1064124488, 363710254, 703721499, 389640783, 1002360059, 1427395742, + 1295231497, 1254972431, 1423497865, 861918264}; + +#pragma omp parallel for firstprivate (vs0, vs1, vs2, vs3) + for (int i = 0; i < size; i += 16) { + const __m512i vnaninf = _mm512_set1_epi32 (0x7f800000), vrneadd = _mm512_set1_epi32 (0x00007fff); + const __m512i vfixup = _mm512_set1_epi32 (0x00000001), vfixupmask = _mm512_set1_epi32 (0x00010000); + const __m512i truncmask = _mm512_set1_epi32 (0xffff0000); + __m512i rnd512 = _mm512_rndxorshft128plus_epi32 (vs0, vs1, vs2, vs3); + __m256i rnbits = _mm512_extracti32x8_epi32 (rnd512, 0); + + __m512 a = _mm512_loadu_ps (&in[i]); + const __m512i mm512_roundbf16sr_a_ = _mm512_castps_si512 (a); + const __mmask16 mm512_roundbf16rne_mask1_ = + _mm512_cmp_epi32_mask (_mm512_and_epi32(mm512_roundbf16sr_a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 mm512_roundbf16rne_mask2_ = + _mm512_cmp_epi32_mask (_mm512_and_epi32(mm512_roundbf16sr_a_, vfixupmask), vfixupmask,_MM_CMPINT_EQ); + __m512i roundbf16sr = + _mm512_mask_add_epi32 (mm512_roundbf16sr_a_, + mm512_roundbf16rne_mask1_, + mm512_roundbf16sr_a_, + _mm512_cvtepu16_epi32 (rnbits)); + roundbf16sr = + _mm512_mask_add_epi32 (roundbf16sr, mm512_roundbf16rne_mask1_, + roundbf16sr, _mm512_mask_add_epi32 (vrneadd, + mm512_roundbf16rne_mask2_, + vrneadd, + vfixup)); + a = _mm512_castsi512_ps (_mm512_and_epi32 (roundbf16sr, truncmask)); + + _mm512_storeu_ps (&out[i], a); + } + } + + void cvt_fp32_bf16_scalar (const float *in, float *out, const int size, + int rmode) { + int lshift = 16; + int rshift = lshift - 3; /* shift to preserve rounding bits */ + unsigned int mask_mant = (unsigned int) (0xFFFFFFFF << lshift); + unsigned int mask_mant_grs = (unsigned int) (0xFFFFFFFF << rshift); + + /* mask to extract G(gaurd), R (round), S (sticky) bits */ + unsigned int lsbGRS = 0xF << rshift; + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + + if (rmode == ROUND_RNE) + rne_mask = 1; + if (rmode == ROUND_STOCHASTIC) + sr_mask = 1; + + for (int gid = 0; gid < size; gid++) { + __float_t uf; + + uf.f = in[gid]; + unsigned int mant_grs = (uf.u & mask_mant_grs); + + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = + (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* stochastic rounding with 16-bit random number */ + uf.u += rand; + } + /* truncation */ + uf.u &= mask_mant; + + /* round to nearest even after truncation if rne_mask is enabled */ + unsigned int rmask_tie = ((mant_grs & lsbGRS) >> rshift); + unsigned int rmask = (rmask_tie & 0x7); + + uf.u += rne_mask * (((rmask > 0x4) || (rmask_tie == 0xC)) << lshift); + + //__float2anyfloat_rn(uf.f, &out[gid]); + out[gid] = uf.f; + } + } + + template < typename scalar_t > + void BFLOAT16_Kernel (const scalar_t * in, + scalar_t * out, const int size, int rmode) { + if ((size % 16) == 0) { + if (rmode == ROUND_STOCHASTIC) + cvt_fp32_bf16_stochastic_intrinsic (in, out, size); + else + cvt_fp32_bf16_rne_intrinsic (in, out, size); + } else { + int vec_size = ((int) (size / 16)) * 16; + + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) + cvt_fp32_bf16_stochastic_intrinsic (in, out, vec_size); + else + cvt_fp32_bf16_rne_intrinsic (in, out, vec_size); + } + cvt_fp32_bf16_scalar (&in[vec_size], &out[vec_size], size - vec_size, rmode); + } + } + + template < typename scalar_t > + void FLOAT16_Kernel (const scalar_t * in, + scalar_t * out, + const int size, int rmode, int no_denorm) { + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + + if (rmode == ROUND_RNE) + rne_mask = 1; + if (rmode == ROUND_STOCHASTIC) + sr_mask = 1; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + if (rne_mask) { + __half_t h; + + h.u = __anyfloat2half_rn (in[gid]); + unsigned short not_denorm = ((((h.u & 0x7FFF) >> 10) & 0x1F) > 0); + unsigned short is_denorm = (not_denorm == 0) ? 1 : 0; + + h.u *= !(is_denorm && no_denorm); + __half2anyfloat (h.u, &out[gid]); + } else if (sr_mask) { + unsigned int fval = ((unsigned int *) in)[gid]; + int exp_h = (int) ((fval & 0x7f800000) >> 23) - 127; + unsigned int mantissa_h = (fval & 0x7FFFFF); + unsigned int sign_h = (fval & 0x80000000); + __half_t h; + + if (exp_h == 128) { + /* handle incoming INF and NaN */ + exp_h = 0x1F; + /* handle signalling NaN */ + if (mantissa_h && ((mantissa_h & 0x400000) == 0x0)) + mantissa_h |= 0x400000; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + h.u = (unsigned short) (mantissa_h >> 13); + } else if (exp_h >= 16) { + /* saturate to INF */ + exp_h = 0x1F; + mantissa_h = 0; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + h.u = (unsigned short) (mantissa_h >> 13); + } else { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned int rand = + (unsigned int) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + if (exp_h < -14) { + mantissa_h += (rand & 0x00001FFF); + /* handle denormals */ + h.u = __anyfloat2half_rn (in[gid]); + } else { + exp_h += 15; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + mantissa_h += (rand & 0x00001FFF); + h.u = (unsigned short) (mantissa_h >> 13); + } + } + __half2anyfloat (h.u, &out[gid]); + } + } + } + + + __m256i _mm256_cvt2fp16_e5m2 (__m256i a, __m256i b) { + const __m512i vnaninf = _mm512_set1_epi16 (0x7c00), vrneadd = + _mm512_set1_epi16 (0x007f); + const __m512i vfixup = _mm512_set1_epi16 (0x0001), vfixupmask = + _mm512_set1_epi16 (0x0100); + /* b: lower half, a : upper half */ + const __m512i a_ = + _mm512_inserti64x4 (_mm512_inserti64x4 (_mm512_setzero_si512 (), b, 0), a, 1); + const __mmask32 maska1_ = + _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vnaninf), vnaninf,_MM_CMPINT_NE); + const __mmask32 maska2_ = + _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vfixupmask), vfixupmask,_MM_CMPINT_EQ); + __m512i a_rne_ = _mm512_mask_add_epi16 (a_, maska1_, a_, + _mm512_mask_add_epi16 (vrneadd, maska2_, vrneadd,vfixup)); + return _mm512_cvtepi16_epi8 (_mm512_srli_epi16 (a_rne_, 8)); + } + + static inline __m512i _mm512_cvte5m2_fp16 (__m256i a) { + return _mm512_slli_epi16 (_mm512_cvtepi8_epi16 (a), 8); + } + + static inline void cvt_fp16_e5m2_rne_intrinsic (const short *__restrict__ in, + unsigned char *out, int size) { +#pragma omp parallel for + for (int i = 0; i < size; i += 32) { + __m256i bh_ = _mm256_lddqu_si256 ((__m256i *) & in[i]); + __m256i ah_ = _mm256_lddqu_si256 ((__m256i *) & in[i + 16]); + + _mm256_storeu_si256 ((__m256i *) & out[i], + _mm256_cvt2fp16_e5m2 (ah_, bh_)); + } + } + + __m256i _mm256_cvt2fp16_e5m2_noINF (__m256i a, __m256i b) { + const __m512i vnaninf = _mm512_set1_epi16 (0x7c00); + const __m512i vrneadd = _mm512_set1_epi16 (0x007f); + const __m512i vfixup = _mm512_set1_epi16 (0x0001); + const __m512i vfixupmask = _mm512_set1_epi16 (0x0100); + /* use a non-standard exponent offset = 16, */ + const __m512i vExp_fp16 = _mm512_set1_epi16 (0x000F); + const __m512i vExp_e5m2 = _mm512_set1_epi16 (0x0010); + const __m512i vsMant = _mm512_set1_epi16 (0x83FF); + /* Exponent Offset = 16, reclaim inf/NaN */ + const __m512i vsatuval = _mm512_set1_epi16 (0x7F00);/* 2^15*1.11 a.k.a 57344.0, largest value */ + const __m512i vinfval = _mm512_set1_epi16 (0x8000); /* -0.0 as INF */ + const __m512i a_ = _mm512_inserti64x4 (_mm512_inserti64x4 (_mm512_setzero_si512 (), b, 0), a, 1); + const __mmask32 maska1_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask32 maska2_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask32 maska3_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, _mm512_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLE); + + __m512i vExp_ = _mm512_sub_epi16 (_mm512_srli_epi16 (_mm512_and_si512 (a_, vnaninf), 10), vExp_fp16); + vExp_ = _mm512_slli_epi16 (_mm512_add_epi16 (vExp_, vExp_e5m2), 10); + __m512i a_rne_ = _mm512_or_si512 (vExp_, _mm512_and_si512 (a_, vsMant)); + + a_rne_ = _mm512_mask_add_epi16 (a_rne_, maska1_, a_rne_, + _mm512_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + a_rne_ = _mm512_mask_mov_epi16 (a_rne_, maska3_, _mm512_or_si512(_mm512_and_si512(a_rne_, vinfval), vsatuval)); + a_rne_ = _mm512_mask_mov_epi16 (a_rne_, ~maska1_, vinfval); + return _mm512_cvtepi16_epi8 (_mm512_srli_epi16 (a_rne_, 8)); + } + + static inline __m256i _mm512_cvt2fp32_e5m2_noINF (__m512 a, __m512 b) { + __m256i ah_ = _mm512_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i bh_ = _mm512_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + return _mm256_cvt2fp16_e5m2_noINF (ah_, bh_); + } + + __m512i _mm512_cvte5m2_noinf_fp16 (__m256i a) { + const __m512i vExp_fp16 = _mm512_set1_epi16 (0x000F); + const __m512i vExp_e5m2 = _mm512_set1_epi16 (0x0010); + const __m512i vsMant = _mm512_set1_epi16 (0x83FF); + const __m512i vnaninf = _mm512_set1_epi16 (0x8000); /* -0.0 as INF */ + const __m512i vinfval = _mm512_set1_epi16 (0x7c00); + __m512i a_ = _mm512_slli_epi16 (_mm512_cvtepi8_epi16 (a), 8); + const __mmask32 mask1_ = _mm512_cmp_epi16_mask (a_, vnaninf, _MM_CMPINT_EQ); + __m512i vExp_ = _mm512_sub_epi16 (_mm512_srli_epi16 (_mm512_and_si512 (a_, vinfval), 10), vExp_e5m2); + vExp_ = _mm512_slli_epi16 (_mm512_add_epi16 (vExp_, vExp_fp16), 10); + a_ = _mm512_or_si512 (vExp_, _mm512_and_si512 (a_, vsMant)); + return _mm512_mask_mov_epi16 (a_, mask1_, vinfval); + } + + static inline void cvt_fp16_e5m2_noINF_rne_intrinsic (const short *__restrict__ in, + unsigned char *out, int size) { +#pragma omp parallel for + for (int i = 0; i < size; i += 32) { + __m256i bh_ = _mm256_lddqu_si256 ((__m256i *) & in[i]); + __m256i ah_ = _mm256_lddqu_si256 ((__m256i *) & in[i + 16]); + + _mm256_storeu_si256 ((__m256i *) & out[i], _mm256_cvt2fp16_e5m2 (ah_, bh_)); + } + } + + void cvt_fp32_e5m2_noinf_rne_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { +#pragma omp parallel for + for (int i = 0; i < size; i += 32) { + __m512 b = _mm512_loadu_ps (&in[i]); + __m512 a = _mm512_loadu_ps (&in[i + 16]); + __m256i ah_ = _mm512_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i bh_ = _mm512_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m512i a_rne_ = _mm512_cvte5m2_noinf_fp16 (_mm256_cvt2fp16_e5m2_noINF (ah_, bh_)); + bh_ = _mm512_extracti64x4_epi64 (a_rne_, 0); + ah_ = _mm512_extracti64x4_epi64 (a_rne_, 1); + b = _mm512_cvtph_ps (bh_); + a = _mm512_cvtph_ps (ah_); + _mm512_storeu_ps (&out[i], b); + _mm512_storeu_ps (&out[i + 16], a); + } + } + + static inline void cvt_fp32_e5m2_flex_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { +#pragma omp parallel for + for (int i = 0; i < size; i += 16) { + const __m512i vnaninf = _mm512_set1_epi32 (0x7f800000); + const __m512i vrneadd = _mm512_set1_epi32 (0x000fffff); + const __m512i vfixup = _mm512_set1_epi32 (0x00000001); + const __m512i vfixupmask = _mm512_set1_epi32 (0x00200000); + const __m512i vexpf32 = _mm512_set1_epi32 (0x0000007f); + const __m512i vmexp_e5m2 = _mm512_set1_epi32 (0x0000000f); + const __m512i vsign = _mm512_set1_epi32 (0x80000000); + const __m512i vmant = _mm512_set1_epi32 (0x007fffff); + const __m512i vmin_e5m2 = _mm512_set1_epi32 (0x37800000); + /*const __m512i vmax_e5m2 = _mm512_set1_epi32 (0x47600000);*/ + const __m512i vdnorm = _mm512_set1_epi32 (0x38800000); + + __m512 a = _mm512_loadu_ps (&in[i]); + const __m512i a_ = _mm512_castps_si512 (a); + const __mmask16 naninf_mask_ = _mm512_cmp_epi32_mask (_mm512_and_epi32 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 rneadd_mask_ = _mm512_cmp_epi32_mask (_mm512_and_epi32 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 zflush_mask_ = _mm512_cmp_epi32_mask (_mm512_and_epi32 (a_, vnaninf), vmin_e5m2, _MM_CMPINT_LT); + const __mmask16 denorm_mask_ = _mm512_cmp_epi32_mask (_mm512_and_epi32 (a_, vnaninf), vdnorm, _MM_CMPINT_LT); + + __m512i a_sign = _mm512_and_epi32 (a_, vsign); + __m512i a_rne = _mm512_mask_add_epi32 (a_, naninf_mask_, a_, + _mm512_mask_add_epi32 (vrneadd, rneadd_mask_, vrneadd, vfixup)); + __m512i a_exp = _mm512_srai_epi32 (_mm512_and_epi32 (a_rne, vnaninf), 23); + a_rne = _mm512_and_epi32 (a_rne, vmant); + a_exp = _mm512_sub_epi32 (a_exp, vexpf32); + + __m512i vminexp = _mm512_sub_epi32 (_mm512_setzero_epi32 (), vmexp_e5m2); + __m512i shft_ = _mm512_sub_epi32 (vminexp, a_exp); + + /*const __mmask16 ovflow_mask_ = + _mm512_cmp_epi32_mask (a_exp, vmexp_e5m2, _MM_CMPINT_NLT);*/ + a_exp = _mm512_add_epi32 (a_exp, vmexp_e5m2); + + __m512i vrshft = _mm512_set1_epi32 (21); + + vrshft = _mm512_mask_add_epi32 (vrshft, denorm_mask_, vrshft, shft_); + __m512i vlshft = _mm512_set1_epi32 (8); + + vlshft = _mm512_mask_add_epi32 (vlshft, denorm_mask_, vlshft, shft_); + + a_rne = _mm512_sllv_epi32 (_mm512_srav_epi32 (a_rne, vrshft), vlshft); + + a_exp = _mm512_slli_epi32 (a_exp, 10); + a_rne = _mm512_or_epi32 (a_rne, a_exp); + a_rne = _mm512_or_epi32 (a_rne, _mm512_srai_epi32 (a_sign, 16)); + a_rne = _mm512_mask_set1_epi32 (a_rne, zflush_mask_, 0); + + __m256i a_rne_16 = _mm512_cvtepi32_epi16 (a_rne); + + a = _mm512_cvtph_ps (a_rne_16); + _mm512_storeu_ps (&out[i], a); + } + } + + void cvt_fp32_e5m2_rne_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { + +#pragma omp parallel for + for (int i = 0; i < size; i += 32) { + const __m512i vnaninf = _mm512_set1_epi16 (0x7c00); + const __m512i vrneadd = _mm512_set1_epi16 (0x007f); + const __m512i vfixup = _mm512_set1_epi16 (0x0001); + const __m512i vfixupmask = _mm512_set1_epi16 (0x0100); + + __m512 s_ = _mm512_set1_ps (scale); + __m512 sr_ = _mm512_set1_ps (1.0 / scale); + __m512 b = _mm512_loadu_ps (&in[i]); + __m512 a = _mm512_loadu_ps (&in[i + 16]); + + b = _mm512_mul_ps (b, s_); + a = _mm512_mul_ps (a, s_); + + __m256i ah_ = _mm512_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i bh_ = _mm512_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + const __m512i a_ = _mm512_inserti64x4 (_mm512_inserti64x4 (_mm512_setzero_si512 (), bh_, 0), ah_, 1); + const __mmask32 maska1_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask32 maska2_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + __m512i a_rne_ = _mm512_mask_add_epi16 (a_, maska1_, a_,_mm512_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + a_rne_ = _mm512_slli_epi16 (_mm512_srli_epi16 (a_rne_, 8), 8); + + bh_ = _mm512_extracti64x4_epi64 (a_rne_, 0); + ah_ = _mm512_extracti64x4_epi64 (a_rne_, 1); + b = _mm512_cvtph_ps (bh_); + a = _mm512_cvtph_ps (ah_); + + _mm512_storeu_ps (&out[i], _mm512_mul_ps (b, sr_)); + _mm512_storeu_ps (&out[i + 16], _mm512_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e5m2_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { + uint32_t vs0[16] __attribute__ ((aligned (64))) = { + 1387366120, 279844183, 888998500, 1099633400, 1252084877, 500390994, + 977516591, 1950666000, 393659750, 834151069, 1477014702, 734008143, + 1983400973, 116410309, 2110188261, 2019272068}; + uint32_t vs1[16] __attribute__ ((aligned (64))) = { + 2034269327, 2125325156, 1209715489, 193165672, 187709636, 28336299, + 419632041, 1774181187, 702309618, 407781555, 1512057936, 1868769368, + 510001215, 966559856, 776583255, 147562106}; + uint32_t vs2[16] __attribute__ ((aligned (64))) = { + 1555452618, 650181557, 883695203, 62767784, 127180605, 1881312534, + 478635452, 814821902, 733990058, 1889991804, 1108257970, 1093480892, + 427374380, 416747337, 558000409, 1594848927}; + uint32_t vs3[16] __attribute__ ((aligned (64))) = { + 419524804, 2146478152, 480059239, 1468956197, 444870959, 1595722866, + 1064124488, 363710254, 703721499, 389640783, 1002360059, 1427395742, + 1295231497, 1254972431, 1423497865, 861918264}; + +#pragma omp parallel for firstprivate (vs0, vs1, vs2, vs3) + for (int i = 0; i < size; i += 32) { + const __m512i vnaninf = _mm512_set1_epi16 (0x7c00); + const __m512i vfixup = _mm512_set1_epi16 (0x0001); + const __m512i vfixupmask = _mm512_set1_epi16 (0x0100); + const __m512i vrneadd = _mm512_set1_epi16 (0x007f); + const __m512i vdenorm = _mm512_set1_epi16 (0x03ff); + const __m512i vexmant = _mm512_set1_epi16 (0x7fff); + + __m512i rnd512 = _mm512_rndxorshft128plus_epi32 (vs0, vs1, vs2, vs3); + __m256i rnbits = _mm512_extracti32x8_epi32 (rnd512, 0); + + __m512 s_ = _mm512_set1_ps (scale); + __m512 sr_ = _mm512_set1_ps (1.0 / scale); + + __m512 b = _mm512_loadu_ps (&in[i]); + __m512 a = _mm512_loadu_ps (&in[i + 16]); + + b = _mm512_mul_ps (b, s_); + a = _mm512_mul_ps (a, s_); + + __m256i ah_ = _mm512_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i bh_ = _mm512_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + const __m512i a_ = _mm512_inserti64x4 (_mm512_inserti64x4 (_mm512_setzero_si512 (), bh_, 0), ah_, 1); + const __mmask32 maska1_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask32 maska2_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask32 maska4_ = _mm512_cmp_epi16_mask (_mm512_and_si512 (a_, vexmant), vdenorm, _MM_CMPINT_LE); + __m512i a_sr_ = _mm512_mask_add_epi16 (a_, (maska1_ & ~maska4_), a_, _mm512_cvtepu8_epi16 (rnbits)); + a_sr_ = _mm512_mask_add_epi16 (a_sr_, maska4_, a_sr_, _mm512_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + a_sr_ = _mm512_slli_epi16 (_mm512_srli_epi16 (a_sr_, 8), 8); + + bh_ = _mm512_extracti64x4_epi64 (a_sr_, 0); + ah_ = _mm512_extracti64x4_epi64 (a_sr_, 1); + b = _mm512_cvtph_ps (bh_); + a = _mm512_cvtph_ps (ah_); + + _mm512_storeu_ps (&out[i], _mm512_mul_ps (b, sr_)); + _mm512_storeu_ps (&out[i + 16], _mm512_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e5m2_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 5 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x00FF; + unsigned short rne_tie = 0x0180; + + float scale_reciprocal = 1.0 / scale; + + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + + unsigned short can_round = ((h.u & 0x7F00) <= 0x7B00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + /* nearest rounding masks */ + unsigned short rnmask = (h.u & grs_bitmask); + unsigned short rnmask_tie = (h.u & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + h.u += can_round * is_normal * (rand & 0xFF); + /* stochastic round: denormals --> rne rounding */ + h.u += can_round * is_denorm * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + h.u += can_round * rne_mask * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + h.u += can_round * rnaz_mask * ((rnmask >= 0x0080) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + h.u += can_round * rntz_mask * ((rnmask > 0x0080) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + h.u += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0080) << lshift); + /* round to -INF, if rminf_mask is enabled */ + h.u += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0080) << lshift); + } + } + /* truncation */ + h.u = (h.u & mask_mant); + float f_; + __half2anyfloat (h.u, &f_); + out[gid] = f_ * scale_reciprocal; + } + } + + template < typename scalar_t > + void E5M2_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 16384.0; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + //cvt_fp32_e5m2_flex_intrinsic(&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e5m2_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (in, out, size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e5m2_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e5m2_rne_intrinsic (in, out, vec_size, scale); + //cvt_fp32_e5m2_noinf_rne_intrinsic (in, out, vec_size, scale); + //cvt_fp32_e5m2_flex_intrinsic(in, out, vec_size, scale); + } + } + cvt_fp32_e5m2_scalar (&in[vec_size], &out[vec_size], size - vec_size, scale, rmode); + } + } + } + + template < typename scalar_t > + void E5M2_DAZ_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, const scalar_t scale, int rmode) { + int non_mant_bits = 5 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x00FF; + unsigned short rne_tie = 0x0180; + + scalar_t scale_reciprocal = 1.0/scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = in[gid] * scale; + + h.u = __anyfloat2half_rn (inval); + /* values above 57344.0, saturate them to +- Infinity */ + + unsigned short can_round = ((h.u & 0x7F00) <= 0x7B00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + /* nearest rounding masks */ + unsigned short rnmask = (h.u & grs_bitmask); + unsigned short rnmask_tie = (h.u & rne_tie); + + if (is_naninf == 0 && is_normal) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + h.u += can_round * (rand & 0xFF); + } else { + /* round to nearest even, if rne_mask is enabled */ + h.u += can_round * rne_mask * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + h.u += can_round * rnaz_mask * ((rnmask >= 0x0080) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + h.u += can_round * rntz_mask * ((rnmask > 0x0080) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + h.u += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0080) << lshift); + /* round to -INF, if rminf_mask is enabled */ + h.u += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0080) << lshift); + } + } else if (is_denorm) { + /* Flush Denormal */ + h.u = 0; + } + /* truncation */ + h.u = (h.u & mask_mant); + __half2anyfloat (h.u, &out[gid], scale_reciprocal); + } + } + + void cvt_fp32_e4m3_rne_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x003f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0080); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x5F00);/* 2^8*1.110 a.k.a 448.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x1800);/* 2^-9, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x2400);/* 2^-6 smallest normal */ + + for (int i = 0; i < size; i += 16) { + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + __m256i a_rne_ = _mm256_mask_add_epi16 (a_, maska1_, a_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + /* saturating values beyond +/-MAX */ + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska3_, _mm256_or_si256(_mm256_and_si256(a_rne_, vsign), vsatuval)); + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska4_, _mm256_and_si256(a_rne_, vsign)); + a_rne_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_rne_, 7), 7); + + bh_ = _mm256_extracti32x4_epi32 (a_rne_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_rne_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e4m3_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x003f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0080); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x5F00);/* 2^8*1.110 a.k.a 448.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x1800);/* 2^-9, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x2400);/* 2^-6 smallest normal */ + + for (int i = 0; i < size; i += 16) { + unsigned int rndbuf[16]; + /* generate 128 random bits */ + for (int r = 0; r < 8; r++) { + rndbuf[r] = (unsigned int) rand_xorshft128plus_scalar (sptr_[r]); + } + __m128i rnbits = _mm_load_si128 ((const __m128i *) &rndbuf[0]); + + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); // might be avx-512 + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); // avx-512 + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); // avx-512 + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + + __m256i a_sr_ = _mm256_mask_add_epi16 (a_, (maska1_ & ~maska5_), a_, + _mm256_srli_epi16 (_mm256_cvtepu8_epi16 (rnbits), 1)); + a_sr_ = _mm256_mask_add_epi16 (a_sr_, maska5_, a_sr_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska3_, _mm256_or_si256(_mm256_and_si256(a_sr_, vsign), vsatuval)); + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska4_, _mm256_and_si256(a_sr_, vsign)); + a_sr_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_sr_, 7), 7); + + bh_ = _mm256_extracti32x4_epi32 (a_sr_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_sr_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e4m3_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 4 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + float scale_reciprocal = 1.0 / scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x5F00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 8 || (can_round == 0)) { + /* Software : saturate values above to +/-448.0 to +/-448.0 */ + mantissa_h = 0x0300; + exp_h = 8; + can_round = 0; + } else if (exp_h < -9) { + /* flush values below 1-4-3 subnormal range to zero */ + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -6) { + dshift = (-6 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + float f; + + __half2anyfloat (h.u, &f); + out[gid] = (f * scale_reciprocal); + } + } + + template < typename scalar_t > + void E4M3_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 8.0; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e4m3_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e4m3_rne_intrinsic (in, out, vec_size, scale); + } + } + cvt_fp32_e4m3_scalar (&in[vec_size], &out[vec_size], size - vec_size, scale, rmode); + } + } + } + + void cvt_fp32_e4m3_ieee_rne_intrinsic (const float *__restrict__ in, float *out, + int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x003f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0080); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x5B80);/* 2^7*1.111 a.k.a 240.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x1800);/* 2^-9, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x2400);/* 2^-6 smallest normal */ + + for (int i = 0; i < size; i += 16) { + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + __m256i a_rne_ = _mm256_mask_add_epi16 (a_, maska1_, a_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + /* saturating values beyond +/-MAX */ + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska3_, _mm256_or_si256(_mm256_and_si256(a_rne_, vsign), vsatuval)); + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska4_, _mm256_and_si256(a_rne_, vsign)); + + a_rne_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_rne_, 7), 7); + + bh_ = _mm256_extracti32x4_epi32 (a_rne_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_rne_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e4m3_ieee_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x003f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0080); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x5B80);/* 2^7*1.111 a.k.a 240.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x1800);/* 2^-9, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x2400);/* 2^-6 smallest normal */ + + for (int i = 0; i < size; i += 16) { + unsigned int rndbuf[16]; + /* generate 128 random bits */ + for (int r = 0; r < 8; r++) { + rndbuf[r] = (unsigned int) rand_xorshft128plus_scalar (sptr_[r]); + } + __m128i rnbits = _mm_load_si128 ((const __m128i *) &rndbuf[0]); + + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + + __m256i a_sr_ = _mm256_mask_add_epi16 (a_, (maska1_ & ~maska5_), a_, + _mm256_srli_epi16 (_mm256_cvtepu8_epi16 (rnbits), 1)); + a_sr_ = _mm256_mask_add_epi16 (a_sr_, maska5_, a_sr_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska3_, _mm256_or_si256(_mm256_and_si256(a_sr_, vsign), vsatuval)); + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska4_, _mm256_and_si256(a_sr_, vsign)); + + a_sr_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_sr_, 7), 7); + + bh_ = _mm256_extracti32x4_epi32 (a_sr_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_sr_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e4m3_ieee_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 4 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + float scale_reciprocal = 1.0 / scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x4B80) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 7 || (can_round == 0)) { + mantissa_h = 0x380; + exp_h = 7; + can_round = 0; + } else if (exp_h < -9) { + /* flush values below 1-4-3 subnormal range to zero */ + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -6) { + dshift = (-6 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + float f; + + __half2anyfloat (h.u, &f); + out[gid] = (f * scale_reciprocal); + } + } + + template < typename scalar_t > + void E4M3_IEEE_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 8.0; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_ieee_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e4m3_ieee_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e4m3_ieee_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_ieee_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e4m3_ieee_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e4m3_ieee_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e4m3_ieee_rne_intrinsic (in, out, vec_size, scale); + } + } + cvt_fp32_e4m3_ieee_scalar (&in[vec_size], &out[vec_size], size - vec_size, scale, rmode); + } + } + } + + void cvt_fp32_e3m4_rne_intrinsic (const float *__restrict__ in, + float *out, int size, float scale) { + + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x001f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0040); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x4F80);/* 2^4*1.1110 a.k.a 30.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x2400);/* 2^-6, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x3400);/* 2^-2 smallest normal */ + + for (int i = 0; i < size; i += 16) { + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + __m256i a_rne_ = _mm256_mask_add_epi16 (a_, maska1_, a_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + /* saturating values beyond +/-MAX */ + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska3_, _mm256_or_si256(_mm256_and_si256(a_rne_, vsign), vsatuval)); + a_rne_ = _mm256_mask_mov_epi16 (a_rne_, maska4_, _mm256_and_si256(a_rne_, vsign)); + + a_rne_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_rne_, 6), 6); + + bh_ = _mm256_extracti32x4_epi32 (a_rne_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_rne_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e3m4_stochastic_intrinsic (const float *__restrict__ in, + float *out, int size, + float scale) { + + const __m256i vnaninf = _mm256_set1_epi16 (0x7c00); + const __m256i vrneadd = _mm256_set1_epi16 (0x001f); + const __m256i vfixup = _mm256_set1_epi16 (0x0001); + const __m256i vfixupmask = _mm256_set1_epi16 (0x0040); + const __m256i vzero = _mm256_set1_epi16 (0x0000); + const __m256i vsign = _mm256_set1_epi16 (0x8000); + const __m256i vsatuval = _mm256_set1_epi16 (0x4F80);/* 2^4*1.1110 a.k.a 30.0, largest value */ + const __m256i vflush = _mm256_set1_epi16 (0x2400);/* 2^-6, smallest denormal */ + const __m256i vxdnorm = _mm256_set1_epi16 (0x3400);/* 2^-2 smallest normal */ + + for (int i = 0; i < size; i += 16) { + unsigned int rndbuf[16]; + /* generate 128 random bits */ + for (int r = 0; r < 8; r++) { + rndbuf[r] = (unsigned int) rand_xorshft128plus_scalar (sptr_[r]); + } + __m128i rnbits = _mm_load_si128 ((const __m128i *) &rndbuf[0]); + + __m256 s_ = _mm256_set1_ps (scale); + __m256 sr_ = _mm256_set1_ps (1.0 / scale); + __m256 b = _mm256_loadu_ps (&in[i]); + __m256 a = _mm256_loadu_ps (&in[i + 8]); + + b = _mm256_mul_ps (b, s_); + a = _mm256_mul_ps (a, s_); + + __m128i ah_ = _mm256_cvtps_ph (a, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m128i bh_ = _mm256_cvtps_ph (b, (_MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC)); + __m256i a_ = _mm256_inserti32x4 (_mm256_inserti32x4 (_mm256_setzero_si256 (), bh_, 0), ah_, 1); + const __mmask16 maska1_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vnaninf, _MM_CMPINT_NE); + const __mmask16 maska2_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vfixupmask), vfixupmask, _MM_CMPINT_EQ); + const __mmask16 maska3_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, _mm256_set1_epi16(0x7FFF)), vsatuval, _MM_CMPINT_NLT); + const __mmask16 maska4_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vflush, _MM_CMPINT_LT); + const __mmask16 maska5_ = _mm256_cmp_epi16_mask (_mm256_and_si256 (a_, vnaninf), vxdnorm, _MM_CMPINT_LT); + /* Handle denormals, shift the mantissa before rounding */ + __m256i v_shft_ = _mm256_sub_epi16 ( + _mm256_srli_epi16 (vxdnorm, 10), + _mm256_srli_epi16 (_mm256_and_si256 (a_, vnaninf), 10)); + v_shft_ = _mm256_mask_mov_epi16 (vzero, maska5_, v_shft_); + a_ = _mm256_mask_srlv_epi16 (a_, maska5_, a_, v_shft_); + a_ = _mm256_mask_sllv_epi16 (a_, maska5_, a_, v_shft_); + + __m256i a_sr_ = _mm256_mask_add_epi16 (a_, (maska1_ & ~maska5_), a_, + _mm256_srli_epi16 (_mm256_cvtepu8_epi16 (rnbits), 1)); + a_sr_ = _mm256_mask_add_epi16 (a_sr_, maska5_, a_sr_, + _mm256_mask_add_epi16 (vrneadd, maska2_, vrneadd, vfixup)); + + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska3_, _mm256_or_si256(_mm256_and_si256(a_sr_, vsign), vsatuval)); + a_sr_ = _mm256_mask_mov_epi16 (a_sr_, maska4_, _mm256_and_si256(a_sr_, vsign)); + a_sr_ = _mm256_slli_epi16 (_mm256_srli_epi16 (a_sr_, 6), 6); + + bh_ = _mm256_extracti32x4_epi32 (a_sr_, 0); + ah_ = _mm256_extracti32x4_epi32 (a_sr_, 1); + b = _mm256_cvtph_ps (bh_); + a = _mm256_cvtph_ps (ah_); + _mm256_storeu_ps (&out[i], _mm256_mul_ps (b, sr_)); + _mm256_storeu_ps (&out[i + 8], _mm256_mul_ps (a, sr_)); + } + } + + void cvt_fp32_e3m4_scalar (const float *__restrict__ in, float *out, + int size, float scale, int rmode) { + int non_mant_bits = 3 /*exp_bits */ + 1; /* exponent + sign */ + int lshift = 10 - (8 /*mbits */ - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x003F; + unsigned short rne_tie = 0x0060; + float scale_reciprocal = 1.0 / scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __half_t h; + float inval = scale * in[gid]; + + h.u = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x4F80) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + if (exp_h > 4 || (can_round == 0)) { + /* Software : saturate values above +/-30.0 to +/-30.0 */ + mantissa_h = 0x0380; + exp_h = 4; + can_round = 0; + } else if (exp_h < -6) { + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -2) { + dshift = (-2 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + rand_xorshft128plus_scalar (sptr_[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x3F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0020) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0020) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0020) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0020) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0020) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0020) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + float f; + + __half2anyfloat (h.u, &f); + out[gid] = (f * scale_reciprocal); + } + } + + template < typename scalar_t > + void E3M4_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int block_size, int rmode) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + __float_t f; + + f.f = maxval; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + + if ((block_size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e3m4_stochastic_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } else { + cvt_fp32_e3m4_rne_intrinsic (&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_e3m4_scalar (&in[start_index], &out[start_index], block_size, scale, rmode); + } + } + } else { + if ((size % 32) == 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e3m4_stochastic_intrinsic (in, out, size, scale); + } else { + cvt_fp32_e3m4_rne_intrinsic (in, out, size, scale); + } + } else { + int vec_size = ((int) (size / 32)) * 32; + if (vec_size > 0) { + if (rmode == ROUND_STOCHASTIC) { + cvt_fp32_e3m4_stochastic_intrinsic (in, out, vec_size, scale); + } else { + cvt_fp32_e3m4_rne_intrinsic (in, out, vec_size, scale); + } + } + cvt_fp32_e3m4_scalar (&in[vec_size], &out[vec_size], + size - vec_size, scale, rmode); + } + } + } + + void cvt_fp32_fp4_nearest_scalar (const float *__restrict__ in, float *out, + int size, float scale) { + + float scale_reciprocal = 1.0 / scale; + +#pragma omp parallel for + for (int gid = 0; gid < size; gid++) { + __float_t f; + float inval = scale * in[gid]; + + f.f = inval; + int exp_f = (int)(((f.u & 0x7F800000) >> 23) - 127); + int sign_f = (f.u & 0x80000000); + /* see if round up works! */ + if (exp_f < 0 && (exp_f%2)) f.f *= 1.6; + /* saturate */ + if (exp_f > 0) f.u = (sign_f | (127 << 23)); + f.u &= 0xFF800000; + /* extract the new exponent */ + exp_f = (int)(((f.u & 0x7F800000) >> 23) - 127); + /* round up did not work, round down */ + if (exp_f < 0 && (exp_f%2)) f.u = (sign_f | ((exp_f + 126) << 23)); + /* flush values smaller than 2^-12 to zero */ + //if (exp_f < -6) f.u = 0; + if (exp_f < -12) f.u = 0; + out[gid] = (f.f * scale_reciprocal); + } + } + + template < typename scalar_t > + void FP4_Nearest_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int block_size) { + float scale = in_scale; + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + if (block_norm == true) { + int nblocks = (size + (block_size - 1)) / block_size; + +#pragma omp parallel for + for (int b = 0; b < nblocks; b++) { + int start_index = (b * block_size); + + /* handle the last block */ + if (start_index + block_size > size) + block_size = (size - start_index); + + float maxval = 0.0; + +#pragma omp parallel for reduction (max:maxval) + for (int gid = start_index; gid < start_index + block_size; gid++) { + maxval = (maxval < fabs (in[gid])) ? fabs (in[gid]) : maxval; + } + /* FP4 max value is 1.0 */ + scale = 1.0/maxval; + cvt_fp32_fp4_nearest_scalar(&in[start_index], &out[start_index], block_size, scale); + } + } else { + cvt_fp32_fp4_nearest_scalar (&in[0], &out[0], size, scale); + } + } + + std::vector < torch::Tensor > fpemu_common_function (torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + + torch::Tensor output; + if (!inplace) + output = torch::zeros_like (input); + + if (!mode.compare ("E5M2_RTZ")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_RTZ); + } else if (!mode.compare ("E5M2_RNE")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_RNE); + } else if (!mode.compare ("E5M2_STOCHASTIC")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E5M2_RNAZ")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_RNAZ); + } else if (!mode.compare ("E5M2_RNTZ")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_RNTZ); + } else if (!mode.compare ("E5M2_RPINF")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_PINF); + } else if (!mode.compare ("E5M2_RNINF")) { + E5M2_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, scale, block_norm, block_size, ROUND_NINF); + } else if (!mode.compare ("E5M2_DAZ_RNE")) { + E5M2_DAZ_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, ROUND_RNE); + } else if (!mode.compare ("E5M2_DAZ_STOCHASTIC")) { + E5M2_DAZ_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, ROUND_STOCHASTIC); + } else if (!mode.compare ("E5M2_DAZ_RNAZ")) { + E5M2_DAZ_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, ROUND_RNAZ); + } else if (!mode.compare ("E5M2_DAZ_RNTZ")) { + E5M2_DAZ_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, ROUND_RNTZ); + } else if (!mode.compare ("FLOAT16_RNE")) { + FLOAT16_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, ROUND_RNE, 0); + } else if (!mode.compare ("FLOAT16_STOCHASTIC")) { + FLOAT16_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, ROUND_STOCHASTIC, 0); + } else if (!mode.compare ("FLOAT16_DAZ_RNE")) { + FLOAT16_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < float >(), + size, ROUND_RNE, 1); + } else if (!mode.compare ("BFLOAT16_RNE")) { + BFLOAT16_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, ROUND_RNE); + } else if (!mode.compare ("BFLOAT16_STOCHASTIC")) { + BFLOAT16_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E4M3_IEEE_RNE")) { + E4M3_IEEE_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_RNE); + } else if (!mode.compare ("E4M3_IEEE_STOCHASTIC")) { + E4M3_IEEE_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E4M3_RNE")) { + E4M3_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_RNE); + } else if (!mode.compare ("E4M3_STOCHASTIC")) { + E4M3_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("E3M4_RNE")) { + E3M4_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_RNE); + } else if (!mode.compare ("E3M4_STOCHASTIC")) { + E3M4_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size, ROUND_STOCHASTIC); + } else if (!mode.compare ("FP4_NEAREST")) { + FP4_Nearest_Kernel < float >(input.data_ptr < float >(), + (inplace) ? input.data_ptr < + float >() : output.data_ptr < + float >(), size, scale, block_norm, + block_size); + } + + if (!inplace) { + return { + output,}; + } else { + return { + input,}; + } + } + +}//namespace + +std::vector < torch::Tensor > fpemu_forward (torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + if (block_norm == true && block_size != size) { + if (size % block_size) { + block_norm = false; + block_size = 1; + } + } + return fpemu_common_function (input, mode, size, inplace, scale, block_norm, + block_size); +} + +std::vector < torch::Tensor > fpemu_backward (torch::Tensor grad, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + if (block_norm == true && block_size != size) { + if (size % block_size) { + block_norm = false; + block_size = 1; + } + } + return fpemu_common_function (grad, mode, size, inplace, scale, block_norm, + block_size); +} + +PYBIND11_MODULE (TORCH_EXTENSION_NAME, m) { + m.def ("forward", &fpemu_forward, "FPEmu forward"); + m.def ("backward", &fpemu_backward, "FPEmu backward"); +} \ No newline at end of file diff --git a/FP8_Emulator/pytquant/cuda/__init__.py b/FP8_Emulator/pytquant/cuda/__init__.py new file mode 100644 index 00000000..893c7619 --- /dev/null +++ b/FP8_Emulator/pytquant/cuda/__init__.py @@ -0,0 +1,3 @@ +import torch +if torch.cuda.is_available(): + from . import fpemu as fpemu_cuda diff --git a/FP8_Emulator/pytquant/cuda/fpemu.py b/FP8_Emulator/pytquant/cuda/fpemu.py new file mode 100644 index 00000000..31782cfc --- /dev/null +++ b/FP8_Emulator/pytquant/cuda/fpemu.py @@ -0,0 +1,74 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +import math +from torch import nn +import torch +import numpy +import fpemu_cuda + +from enum import Enum + +torch.manual_seed(42) + +""" + NONE + E5M2_RTZ + E5M2_STOCHASTIC + E5M2_RNE + E5M2_RNAZ + E5M2_RNTZ + E5M2_RPINF + E5M2_RNINF + E5M2_DAZ_STOCHASTIC + E5M2_DAZ_RNE + E5M2_DAZ_RNAZ + E5M2_DAZ_RNTZ + BFLOAT16_STOCHASTIC + BFLOAT16_RNE + FLOAT16_RNE + FLOAT16_STOCHASTIC + FLOAT16_DAZ_RNE + E4M3_RNE + E4M3_STOCHASTIC +""" + +class FPEmuOp(torch.autograd.Function): + @staticmethod + def forward(ctx, input, mode='NONE', inplace=False, scale=1.0, blocknorm=False, blocksize=1): + if mode == 'NONE' : + ctx.mark_dirty(input) + return input + else : + if input.is_sparse : + input = input.coalesce() + size = input.values().nelement() + if inplace == True: + outputs = fpemu_cuda.forward(input._values().contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = input + else : + outputs = fpemu_cuda.forward(input._values().contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = torch.sparse.FloatTensor(input.indices(), outputs[0], input.size()) + else : + size = input.nelement() + outputs = fpemu_cuda.forward(input.contiguous(), mode, size, inplace, scale, blocknorm, blocksize) + output = outputs[0] + + if inplace == True: + ctx.mark_dirty(input) + return output + + @staticmethod + def backward(ctx, output_grad): + # straight-through estimator + return output_grad, None, None, None, None + + '''@staticmethod + def symbolic(g, input, mode='NONE', inplace=False, scale=1.0, blocknorm=False, blocksize=1): + return g.op("::FPEmuOp", input, mode, inplace, scale, blocknorm, blocksize)''' \ No newline at end of file diff --git a/FP8_Emulator/pytquant/cuda/fpemu_impl.cpp b/FP8_Emulator/pytquant/cuda/fpemu_impl.cpp new file mode 100644 index 00000000..deeec337 --- /dev/null +++ b/FP8_Emulator/pytquant/cuda/fpemu_impl.cpp @@ -0,0 +1,54 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include + +std::vector fpemu_cuda_forward( + torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size); + +// NOTE: AT_ASSERT has become AT_CHECK on master after 0.4. +#define CHECK_CUDA(x) AT_ASSERTM(x.device().is_cuda(), #x " must be a CUDA tensor") +#define CHECK_CONTIGUOUS(x) AT_ASSERTM(x.is_contiguous(), #x " must be contiguous") +#define CHECK_INPUT(x) CHECK_CUDA(x); CHECK_CONTIGUOUS(x) + +std::vector fpemu_forward( + torch::Tensor input, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + CHECK_INPUT(input); + return fpemu_cuda_forward(input, mode, size, inplace, scale, block_norm, block_size); +} + +std::vector fpemu_backward( + torch::Tensor grad, + std::string mode, + int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + CHECK_INPUT(grad); + return fpemu_cuda_forward(grad, mode, size, inplace, scale, block_norm, block_size); +} + +PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { + m.def("forward", &fpemu_forward, "FPEmu forward (CUDA)"); + m.def("backward", &fpemu_backward, "FPEmu backward (CUDA)"); +} diff --git a/FP8_Emulator/pytquant/cuda/fpemu_kernels.cu b/FP8_Emulator/pytquant/cuda/fpemu_kernels.cu new file mode 100644 index 00000000..d006be43 --- /dev/null +++ b/FP8_Emulator/pytquant/cuda/fpemu_kernels.cu @@ -0,0 +1,1381 @@ +/*----------------------------------------------------------------------------* + * Copyright (c) 2023, Intel Corporation - All rights reserved. + * This file is part of FP8-Emulation-Toolkit + * + * SPDX-License-Identifier: BSD-3-Clause + *----------------------------------------------------------------------------* + * Naveen Mellempudi (Intel Corporation) + *----------------------------------------------------------------------------*/ + +#include +#include +#include +#include +#include + +#define CUBLOCK_SIZE 256 +enum ROUNDING_MODES { ROUND_RTZ = 0, ROUND_RNE = 1, ROUND_STOCHASTIC = + 2, ROUND_RNAZ = 3, ROUND_RNTZ = 4, ROUND_PINF = 5, ROUND_NINF = 6 }; + +namespace { + + typedef union half_t { + unsigned short u; + at::Half f; + } __half_t; + + typedef union ufloat32 { + unsigned u; + float f; + } __float_t; + +/* this implementation of xoroshiro128++ PRNG is borrowed from here: + http://prng.di.unimi.it/xoshiro128plusplus.c + main page: http://prng.di.unimi.it/ +*/ + __device__ static uint32_t s1[4] = + { 1387366120, 279844183, 888998500, 1099633400 }; + __device__ static uint32_t s2[4] = + { 2034269327, 2125325156, 1209715489, 193165672 }; + __device__ static uint32_t s3[4] = + { 1555452618, 650181557, 883695203, 62767784 }; + __device__ static uint32_t s4[4] = + { 419524804, 2146478152, 480059239, 1468956197 }; + __device__ static uint32_t s5[4] = + { 1252084877, 500390994, 977516591, 1950666000 }; + __device__ static uint32_t s6[4] = + { 393659750, 834151069, 1477014702, 734008143 }; + __device__ static uint32_t s7[4] = + { 1983400973, 116410309, 2110188261, 2019272068 }; + __device__ static uint32_t s8[4] = + { 187709636, 28336299, 419632041, 1774181187 }; + __device__ static uint32_t s9[4] = + { 702309618, 407781555, 1512057936, 1868769368 }; + __device__ static uint32_t s10[4] = + { 510001215, 966559856, 776583255, 147562106 }; + __device__ static uint32_t s11[4] = + { 127180605, 1881312534, 478635452, 814821902 }; + __device__ static uint32_t s12[4] = + { 733990058, 1889991804, 1108257970, 1093480892 }; + __device__ static uint32_t s13[4] = + { 427374380, 416747337, 558000409, 1594848927 }; + __device__ static uint32_t s14[4] = + { 444870959, 1595722866, 1064124488, 363710254 }; + __device__ static uint32_t s15[4] = + { 703721499, 389640783, 1002360059, 1427395742 }; + __device__ static uint32_t s16[4] = + { 1295231497, 1254972431, 1423497865, 861918264 }; + + __device__ static uint32_t *sptr[16] = + { s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, s13, s14, s15, s16 }; + + __device__ __forceinline__ uint32_t rotl_ (const uint32_t x, int k) { + return (x << k) | (x >> (32 - k)); + } + + __device__ __forceinline__ uint32_t _rand_xorshft128plus_with_seed (uint32_t + * ps) { + const uint32_t result_plus = ps[0] + ps[3]; + const uint32_t t = ps[1] << 9; + + ps[2] ^= ps[0]; + ps[3] ^= ps[1]; + ps[1] ^= ps[2]; + ps[0] ^= ps[3]; + + ps[2] ^= t; + + ps[3] = rotl_ (ps[3], 11); + + return result_plus; + } + + template < typename scalar_t > + __device__ __forceinline__ float __anyfloat2float_rn (scalar_t a_) { + float f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = __double2float_rn (a_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = a_; + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = __half2float ((at::Half) a_); + } + return f_; + } + + template < typename scalar_t > + __device__ __forceinline__ void __float2anyfloat_rn (float f_, + scalar_t * out) { + scalar_t a_; + + if (std::is_same < scalar_t, double >::value) { + a_ = (scalar_t) (f_); + } else if (std::is_same < scalar_t, float >::value) { + a_ = f_; + } else if (std::is_same < scalar_t, at::Half >::value) { + a_ = (at::Half) __float2half_rn (f_); + } + *out = a_; + } + + template < typename scalar_t > + __device__ __forceinline__ at::Half __anyfloat2half_rn (scalar_t f_) { + at::Half h_; + if (std::is_same < scalar_t, double >::value) { + h_ = __float2half_rn (__double2float_rn (f_)); + } else if (std::is_same < scalar_t, float >::value) { + h_ = __float2half_rn (f_); + } else if (std::is_same < scalar_t, at::Half >::value) { + h_ = (at::Half) f_; + } + return h_; + } + + template < typename scalar_t > + __device__ __forceinline__ void __half2anyfloat (at::Half h_, + scalar_t * out) { + scalar_t f_; + + if (std::is_same < scalar_t, double >::value) { + f_ = (scalar_t) __half2float ((at::Half) h_); + } else if (std::is_same < scalar_t, float >::value) { + f_ = __half2float (h_); + } else if (std::is_same < scalar_t, at::Half >::value) { + f_ = (at::Half) h_; + } + *out = f_; + } + + template < typename scalar_t > + __device__ void absmax_block (const scalar_t * in, + float *sdata, const int size) { + unsigned int tid = threadIdx.x; + unsigned int i = blockIdx.x * blockDim.x + threadIdx.x; + + sdata[tid] = 0.0f; + + if (i < size) { + sdata[tid] = + fmaxf (fabsf (sdata[tid]), fabsf (__anyfloat2float_rn (in[i]))); + } + __syncthreads (); + + for (unsigned int s = 1; s < blockDim.x; s *= 2) { + if ((tid % (2 * s)) == 0) { + sdata[tid] = fmaxf (fabsf (sdata[tid]), fabsf (sdata[tid + s])); + } + __syncthreads (); + } + } + + __device__ static inline float atomicMaxf (float *address, float val) { + int *address_as_int = (int *) address; + int old = *address_as_int, assumed; + + while (val > __int_as_float (old)) { + assumed = old; + old = atomicCAS (address_as_int, assumed, __float_as_int (val)); + } + return __int_as_float (old); + } + + template < typename scalar_t > + __global__ void absmax0 (scalar_t * g_data, float *g_odata, + unsigned int n) { + extern __shared__ float sdata[]; + + unsigned int tid = threadIdx.x; + unsigned int i = blockIdx.x * blockDim.x + threadIdx.x; + + sdata[tid] = 0.0f; + + if (i < n) { + sdata[tid] = + fmaxf (fabsf (sdata[tid]), fabsf (__anyfloat2float_rn (g_data[i]))); + } + __syncthreads (); + + for (unsigned int s = 1; s < blockDim.x; s *= 2) { + if ((tid % (2 * s)) == 0) { + sdata[tid] = fmaxf (fabsf (sdata[tid]), fabsf (sdata[tid + s])); + } + __syncthreads (); + } + __syncthreads (); + if (tid == 0) { + atomicMaxf (&g_odata[0], sdata[0]); + } + } + + template < typename scalar_t > + __global__ void reduce0 (scalar_t * g_data, float *g_odata, + unsigned int n) { + extern __shared__ float sdata[]; + unsigned int tid = threadIdx.x; + unsigned int i = blockIdx.x * blockDim.x + threadIdx.x; + + if (i < n) { + sdata[tid] = __anyfloat2float_rn (g_data[i]); + sdata[tid + blockDim.x] = + __anyfloat2float_rn (g_data[i]) * __anyfloat2float_rn (g_data[i]); + } else { + sdata[tid] = 0.0f; + sdata[tid + blockDim.x] = 0.0f; + } + __syncthreads (); + + for (unsigned int s = 1; s < blockDim.x; s *= 2) { + if ((tid % (2 * s)) == 0) { + sdata[tid] += sdata[tid + s]; + sdata[blockDim.x + tid] += sdata[blockDim.x + tid + s]; + } + __syncthreads (); + } + __syncthreads (); + if (tid == 0) { + atomicAdd (&g_odata[0], sdata[0]); + atomicAdd (&g_odata[1], sdata[blockDim.x]); + } + } + + + template < typename scalar_t > + __global__ void BFLOAT16_Kernel (const scalar_t * in, + scalar_t * out, + const int size, int rmode) { + int lshift = 16; + unsigned int mask_mant = (unsigned int) (0xFFFFFFFF << lshift); + unsigned int grs_bitmask = 0x0000FFFF; + unsigned int rne_tie = 0x00018000; + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + + if (rmode == ROUND_RNE) + rne_mask = 1; + if (rmode == ROUND_RNAZ) + rnaz_mask = 1; + if (rmode == ROUND_RNTZ) + rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) + sr_mask = 1; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __float_t uf; + + uf.f = __anyfloat2float_rn (in[gid]); + unsigned int is_normal = (((uf.u & 0x7F800000) <= 0x7F000000) + && ((uf.u & 0x7F800000) >= 0x00800000)) ? 1 : 0; + unsigned int is_denorm = ((uf.u & 0x7F800000) == 0x0) ? 1 : 0; + unsigned int is_naninf = ((uf.u & 0x7F800000) == 0x7F800000) ? 1 : 0; + + /* nearest rounding masks */ + unsigned int rnmask = (uf.u & grs_bitmask); + unsigned int rnmask_tie = (uf.u & rne_tie); + + if (is_naninf == 0 && is_normal) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned int rand = _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + uf.u += (rand & 0x0000FFFF); + } else { + /* round to nearest even, if rne_mask is enabled */ + uf.u += rne_mask * + (((rnmask > 0x00008000) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + uf.u += rnaz_mask * ((rnmask >= 0x00008000) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + uf.u += rntz_mask * ((rnmask > 0x00008000) << lshift); + } + } else if (is_denorm) { + /* Flush Denormal */ + uf.u = 0; + } + /* truncation */ + uf.u = (uf.u & mask_mant); + + __float2anyfloat_rn (uf.f, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void FLOAT16_Kernel (const scalar_t * in, + scalar_t * out, + const int size, + int rmode, int no_denorm) { + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + if (std::is_same < scalar_t, at::Half >::value || rne_mask) { + __half_t h; + + at::Half hval; + hval = __anyfloat2half_rn (in[gid]); + h.f = hval; + unsigned short not_denorm = ((((h.u & 0x7FFF) >> 10) & 0x1F) > 0); + unsigned short is_denorm = (not_denorm == 0) ? 1 : 0; + + h.u *= !(is_denorm * no_denorm); + __half2anyfloat (h.f, &out[gid]); + } else if (sr_mask) { + unsigned int fval = ((unsigned int *) in)[gid]; + int exp_h = (int) ((fval & 0x7f800000) >> 23) - 127; + unsigned int mantissa_h = (fval & 0x7FFFFF); + unsigned int sign_h = (fval & 0x80000000); + __half_t h; + + if (exp_h == 128) { + /* handle incoming INF and NaN */ + exp_h = 0x1F; + /* handle signalling NaN */ + if (mantissa_h && ((mantissa_h & 0x400000) == 0x0)) + mantissa_h |= 0x400000; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + h.u = (unsigned short) (mantissa_h >> 13); + } else if (exp_h >= 16) { + /* saturate to INF */ + exp_h = 0x1F; + mantissa_h = 0; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + h.u = (unsigned short) (mantissa_h >> 13); + } else { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned int rand = (unsigned int) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + if (exp_h < -14) { + /* handle denormal values */ + h.f = __anyfloat2half_rn (in[gid]); + } else { + exp_h += 15; + mantissa_h |= (exp_h << 23); + mantissa_h |= (sign_h >> 3); + mantissa_h += (rand & 0x00001FFF); + h.u = (unsigned short) (mantissa_h >> 13); + } + } + __half2anyfloat (h.f, &out[gid]); + } + } + } + + template < typename scalar_t > + __global__ void E5M2_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int mbits, + int exp_bits, int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x00FF; + unsigned short rne_tie = 0x0180; + + extern __shared__ float sdata[]; + scalar_t scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + __float_t f; + f.f = (float) sdata[0]; + f.u = (f.u & 0x7F800000); + scale = 2.0 * f.f; + scale /= 16384.0; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = in[gid] * scale; + __half hval = __anyfloat2half_rn (inval); + h.f = hval; + + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + unsigned short can_round = ((h.u & 0x7F00) <= 0x7B00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + + /* nearest rounding masks */ + unsigned short rnmask = (h.u & grs_bitmask); + unsigned short rnmask_tie = (h.u & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = 0; /* use a constant seed index to reuse the same set of random numbers */ + unsigned short rand = + (unsigned short) _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + h.u += can_round * is_normal * (rand & 0xFF); + /* stochastic round: denormals --> rne rounding */ + h.u += can_round * is_denorm * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + h.u += can_round * rne_mask * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + h.u += can_round * rnaz_mask * ((rnmask >= 0x0080) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + h.u += can_round * rntz_mask * ((rnmask > 0x0080) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + h.u += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0080) << lshift); + /* round to -INF, if rminf_mask is enabled */ + h.u += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0080) << lshift); + } + } + /* truncation */ + h.u = (h.u & mask_mant); + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void E5M2_DAZ_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t scale, + int mbits, int exp_bits, + int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x00FF; + unsigned short rne_tie = 0x0180; + + float scale_reciprocal = 1.0 / scale; + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = in[gid] * scale; + __half hval = __anyfloat2half_rn (inval); + h.f = hval; + /* values above 57344.0, saturate them to +- Infinity */ + unsigned short can_round = ((h.u & 0x7F00) <= 0x7B00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + /* nearest rounding masks */ + unsigned short rnmask = (h.u & grs_bitmask); + unsigned short rnmask_tie = (h.u & rne_tie); + + if (is_naninf == 0 && is_normal) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + h.u += can_round * (rand & 0xFF); + } else { + /* round to nearest even, if rne_mask is enabled */ + h.u += can_round * rne_mask * + (((rnmask > 0x0080) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + h.u += can_round * rnaz_mask * ((rnmask >= 0x0080) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + h.u += can_round * rntz_mask * ((rnmask > 0x0080) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + h.u += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0080) << lshift); + /* round to -INF, if rminf_mask is enabled */ + h.u += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0080) << lshift); + } + } else if (is_denorm) { + /* Flush Denormal */ + h.u = 0; + } + /* truncation */ + h.u = (h.u & mask_mant); + + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void E4M3_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int mbits, + int exp_bits, int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + + extern __shared__ float sdata[]; + float scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + __float_t f; + f.f = sdata[0]; + f.u = (f.u & 0x7F800000); + scale = 2 * f.f; + scale /= 8.0; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = in[gid] * scale; + + h.f = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x5F00) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 8 || (can_round == 0)) { + /* Software : saturate values above to +/-448.0 to +/-448.0 */ + mantissa_h = 0x0300; + exp_h = 8; + can_round = 0; + } else if (exp_h < -9) { + /* flush values below 1-4-3 subnormal range to zero */ + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -6) { + dshift = (-6 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void E4M3_IEEE_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int mbits, int exp_bits, + int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + + extern __shared__ float sdata[]; + float scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + __float_t f; + f.f = sdata[0]; + f.u = (f.u & 0x7F800000); + scale = 2*f.f; + scale = 2 * f.f; + scale /= 8.0; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = in[gid] * scale; + + h.f = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x4B80) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 7) { + /* Hardware : saturate +/-INF */ + mantissa_h = 0; + exp_h = 16; + is_naninf = 1; + } else if (exp_h < -9) { + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -6) { + dshift = (-6 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void E4M3v2_Kernel (const scalar_t * + __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, int mbits, + int exp_bits, int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x007F; + unsigned short rne_tie = 0x00C0; + + extern __shared__ float sdata[]; + float scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + __float_t f; + f.f = sdata[0]; + f.u = (f.u & 0x7F800000); + scale = 2 * f.f; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = in[gid] * scale; + + h.f = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x4B80) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > -1) { + /* handle saturation */ + mantissa_h = 0x0380; + exp_h = -1; + can_round = 0; + } + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x7F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0040) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0040) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0040) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0040) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0040) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h <<= dshift; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void E3M4_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm, + int mbits, + int exp_bits, int rmode) { + int non_mant_bits = exp_bits + 1; /* exponent + sign */ + int lshift = 10 - (mbits - non_mant_bits); + + unsigned short rne_mask = 0; /* round to nearest even mask */ + unsigned short rnaz_mask = 0; /* round to nearest away from zero mask */ + unsigned short rntz_mask = 0; /* round to nearest towards zero mask */ + unsigned short sr_mask = 0; /* stochastic rounding mask */ + unsigned short rpinf_mask = 0; /* round to +INF */ + unsigned short rminf_mask = 0; /* round to -INF */ + + if (rmode == ROUND_RNE) rne_mask = 1; + if (rmode == ROUND_RNAZ) rnaz_mask = 1; + if (rmode == ROUND_RNTZ) rntz_mask = 1; + if (rmode == ROUND_STOCHASTIC) sr_mask = 1; + if (rmode == ROUND_PINF) rpinf_mask = 1; + if (rmode == ROUND_NINF) rminf_mask = 1; + + unsigned short mask_mant = (unsigned short) (0xFFFF << lshift); + unsigned short grs_bitmask = 0x003F; + unsigned short rne_tie = 0x0060; + + extern __shared__ float sdata[]; + float scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + __float_t f; + + f.f = sdata[0]; + f.u = (f.u & 0x7F800000); + scale = 2 * f.f; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __half_t h; + float inval = scale * in[gid]; + + h.f = __anyfloat2half_rn (inval); + short exp_h = (short) ((h.u & 0x7C00) >> 10) - 15; + short sign_h = (h.u & 0x8000); + short mantissa_h = (h.u & 0x03FF); + + unsigned short can_round = ((h.u & 0x7FFF) < 0x4F80) ? 1 : 0; + unsigned short is_normal = (((h.u & 0x7C00) <= 0x7800) + && ((h.u & 0x7C00) >= 0x0400)) ? 1 : 0; + unsigned short is_denorm = ((h.u & 0x7C00) == 0x0) ? 1 : 0; + unsigned short is_naninf = ((h.u & 0x7C00) == 0x7C00) ? 1 : 0; + + int dshift = 0; + + if (exp_h > 4 || (can_round == 0)) { + /* Software : saturate values above +/-30.0 to +/-30.0 */ + mantissa_h = 0x0380; + exp_h = 4; + can_round = 0; + } else if (exp_h < -6) { + exp_h = -15; + mantissa_h = 0; + } else if (exp_h < -2) { + dshift = (-2 - exp_h); + /* handle denormals */ + mantissa_h = mantissa_h >> dshift; + mantissa_h <<= dshift; + } + + /* nearest rounding masks */ + unsigned short rnmask = (mantissa_h & grs_bitmask); + unsigned short rnmask_tie = (mantissa_h & rne_tie); + + if (is_naninf == 0) { + if (sr_mask) { + /* stochastic with 16 seeds */ + int seed_index = (gid / 16); + unsigned short rand = (unsigned short) + _rand_xorshft128plus_with_seed (sptr[(seed_index % 16)]); + /* apply stochastic rounding before truncation if sr_mask is enabled */ + mantissa_h += can_round * is_normal * (rand & 0x3F); + /* stochastic round: denormals --> rne rounding */ + mantissa_h += can_round * is_denorm * + (((rnmask > 0x0020) || (rnmask_tie == rne_tie)) << lshift); + } else { + /* round to nearest even, if rne_mask is enabled */ + mantissa_h += can_round * rne_mask * + (((rnmask > 0x0020) || (rnmask_tie == rne_tie)) << lshift); + /* round to nearest away from zero, if rnaz_mask is enabled */ + mantissa_h += can_round * rnaz_mask * ((rnmask >= 0x0020) << lshift); + /* round to nearest towards zero, if rntz_mask is enabled */ + mantissa_h += can_round * rntz_mask * ((rnmask > 0x0020) << lshift); + /* round to +INF, if rpinf_mask is enabled */ + mantissa_h += can_round * rpinf_mask * (h.f > 0) * ((rnmask >= 0x0020) << lshift); + /* round to -INF, if rminf_mask is enabled */ + mantissa_h += can_round * rminf_mask * (h.f < 0) * ((rnmask >= 0x0020) << lshift); + } + } + /* truncation */ + mantissa_h &= mask_mant; + mantissa_h += ((exp_h + 15) << 10); + mantissa_h |= sign_h; + h.u = mantissa_h; + __half2anyfloat (h.f * scale_reciprocal, &out[gid]); + } + } + + template < typename scalar_t > + __global__ void FP4_Nearest_Kernel (const scalar_t * __restrict__ in, + scalar_t * __restrict__ out, + const int size, + const scalar_t in_scale, + bool block_norm) { + extern __shared__ float sdata[]; + float scale = in_scale; + + if (block_norm == true) { + absmax_block (in, sdata, size); + /* FP4 max value is 1.0 */ + scale = 1.0/sdata[0]; + } + float scale_reciprocal = 1.0 / scale; + + for (int gid = (blockIdx.x * blockDim.x) + threadIdx.x; gid < size; + gid += blockDim.x * gridDim.x) { + __float_t f; + float inval = scale * in[gid]; + + f.f = inval; + int exp_f = (int)((f.u & 0x7F800000) >> 23) - 127; + int sign_f = (f.u & 0x80000000); + /* see if round up works! */ + if (exp_f < 0 && (exp_f%2)) f.f *= 1.6; + /* saturate */ + if (exp_f > 0) f.u = (sign_f | (127 << 23)); + f.u &= 0xFF800000; + /* extract the new exponent */ + exp_f = (int)(((f.u & 0x7F800000) >> 23) - 127); + /* round up did not work, round down */ + if (exp_f < 0 && (exp_f%2)) f.u = (sign_f | ((exp_f + 126) << 23)); + /* flush values smaller than 2^-12 to zero */ + if (exp_f < -12) f.u = 0; + + out[gid] = (f.f * scale_reciprocal); + } + } + +} + +std::vector fpemu_cuda_forward( + torch::Tensor input, + std::string mode, + const int size, + bool inplace, + float scale, + bool block_norm, + int block_size) { + + float fmax = std::numeric_limits::max(); + if (scale > fmax) { + fprintf(stderr,"Error: Invalid scale factor : %.2e, make sure the scale is not larger than : %.2e\n", scale, fmax); + exit(1); + } + + torch::Tensor output; + if (!inplace ) output = torch::zeros_like(input); + + int sdata_size = 0; + int threads = CUBLOCK_SIZE; + if (block_norm == true && block_size != size) { + if (size%block_size) { + block_norm = false; + } else { + threads = block_size; + sdata_size = threads * sizeof(float); + } + } + const dim3 blocks((size + (threads-1))/threads); + + if (!mode.compare("E5M2_RTZ")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RTZ", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_RTZ); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + } else if (!mode.compare("E5M2_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RNE", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + } else if (!mode.compare("E5M2_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_STOCHASTIC", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + } else if (!mode.compare("E5M2_RNAZ")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RNAZ", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_RNAZ); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_RNTZ")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RNTZ", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_RNTZ); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_RPINF")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RPINF", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_PINF); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_RNINF")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_RNINF", ([&] { + E5M2_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 5, + ROUND_NINF); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_DAZ_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_DAZ_RNE", ([&] { + E5M2_DAZ_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + 8, + 5, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_DAZ_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_DAZ_STOCHASTIC", ([&] { + E5M2_DAZ_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + 8, + 5, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_DAZ_RNAZ")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_DAZ_RNAZ", ([&] { + E5M2_DAZ_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + 8, + 5, + ROUND_RNAZ); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E5M2_DAZ_RNTZ")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E5M2_DAZ_RNTZ", ([&] { + E5M2_DAZ_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + 8, + 5, + ROUND_RNTZ); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("FLOAT16_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "FLOAT16_RNE", ([&] { + FLOAT16_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + ROUND_RNE, + 0); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("FLOAT16_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "FLOAT16_STOCHASTIC", ([&] { + FLOAT16_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + ROUND_STOCHASTIC, + 0); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("FLOAT16_DAZ_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "FLOAT16_DAZ_RNE", ([&] { + FLOAT16_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + ROUND_RNE, + 1); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("BFLOAT16_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "BFLOAT16_RNE", ([&] { + BFLOAT16_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("BFLOAT16_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "BFLOAT16_STOCHASTIC", ([&] { + BFLOAT16_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E4M3_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E4M3_RNE", ([&] { + E4M3_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 4, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E4M3_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E4M3_STOCHASTIC", ([&] { + E4M3_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 4, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E4M3_IEEE_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E4M3_IEEE_RNE", ([&] { + E4M3_IEEE_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 4, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E4M3_IEEE_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E4M3_IEEE_STOCHASTIC", ([&] { + E4M3_IEEE_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 4, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E3M4_RNE")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E3M4_RNE", ([&] { + E3M4_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 3, + ROUND_RNE); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("E3M4_STOCHASTIC")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "E3M4_STOCHASTIC", ([&] { + E3M4_Kernel<<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm, + 8, + 3, + ROUND_STOCHASTIC); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + + } else if (!mode.compare("FP4_NEAREST")) { + AT_DISPATCH_FLOATING_TYPES_AND_HALF(input.scalar_type(), "FP4_NEAREST", ([&] { + FP4_Nearest_Kernel <<>>( + input.data_ptr(), + (inplace) ? input.data_ptr() : output.data_ptr(), + size, + scale, + block_norm); + })); + cudaDeviceSynchronize(); + AT_CUDA_CHECK(cudaGetLastError()); + } + if (!inplace) { + return {output, }; + } else { + return {input, }; + } +} diff --git a/FP8_Emulator/pytquant/test.py b/FP8_Emulator/pytquant/test.py new file mode 100644 index 00000000..22f083c1 --- /dev/null +++ b/FP8_Emulator/pytquant/test.py @@ -0,0 +1,151 @@ +#------------------------------------------------------------------------------ +# Copyright (c) 2023, Intel Corporation - All rights reserved. +# This file is part of FP8-Emulation-Toolkit +# +# SPDX-License-Identifier: BSD-3-Clause +#------------------------------------------------------------------------------ +# Naveen Mellempudi (Intel Corporation) +#------------------------------------------------------------------------------ + +from __future__ import division +from __future__ import print_function + +import argparse +import numpy as np +import torch +import cpp.fpemu as fpemu_cpp +import mpemu +import sys + +sys.path.append('../../mpemu') +# importing +from mpemu.qutils import fpemu_device_fn + + +if torch.cuda.is_available(): + import cuda.fpemu as fpemu_cuda + +def check_equal(first, second, verbose): + if verbose: + print() + for i, (x, y) in enumerate(zip(first, second)): + x = x.cpu().detach().numpy() + y = y.cpu().detach().numpy() + + if verbose: + print("x = {}".format(x.flatten())) + print("y = {}".format(y.flatten())) + print('-' * 80) + + np.testing.assert_allclose(x, y, rtol=0.125, err_msg="Conflict : ", verbose=verbose) + +def print_tensor(first, second, verbose, sparse): + print('printing ...') + if verbose: + print() + for i, (x, y) in enumerate(zip(first, second)): + if sparse : + x = x.cpu().to_dense().detach().numpy() + y = y.cpu().to_dense().detach().numpy() + else : + x = x.cpu().detach().numpy() + y = y.cpu().detach().numpy() + if verbose: + print("x = {}".format(x.flatten())) + print("y = {}".format(y.flatten())) + print('-' * 80) + + +def zero_grad(variables): + for variable in variables: + variable.grad.zero_() + + +def get_grads(variables): + return [var.grad.clone() for var in variables] + +def check_forward(variables, with_cuda, verbose, sparse): + if with_cuda: + if not torch.cuda.is_available(): + print('CUDA is not supported on this platform ... ') + else : + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E5M2_RNE") + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E5M2_STOCHASTIC") + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E4M3_RNE") + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E4M3_STOCHASTIC") + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E3M4_RNE") + #cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "E3M4_STOCHASTIC") + cuda_values = fpemu_cuda.FPEmuOp.apply(variables.cuda(), "FP4_NEAREST")#, False, 1.0, True, 4) + print('Forward: CUDA ... ', end='') + print_tensor(variables, cuda_values, verbose, sparse) + else : + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E5M2_RNE") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E5M2_STOCHASTIC") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E4M3_RNE") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E4M3_STOCHASTIC") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E3M4_RNE") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "E3M4_STOCHASTIC") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "FP4_NEAREST")#, False, 1.0, True, 4) + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "INT8") + #cpp_values = fpemu_cpp.FPEmuOp.apply(variables, "INT4") + cpp_values = fpemu_device_fn(variables, "INT4", inplace=False, scale=1.0) + print('Forward: C++ ... ', end='') + print_tensor(variables, cpp_values, verbose, sparse) + + if not verbose : + print('Done.') + print('Use the option --verbose to see results') + else : + print('Done.') + + +parser = argparse.ArgumentParser() +parser.add_argument('-sp', '--sparse', action='store_true') +parser.add_argument('-c', '--cuda', action='store_true') +parser.add_argument('-v', '--verbose', action='store_true') + +options = parser.parse_args() +if options.cuda: + device = torch.device("cuda") +else: + device = torch.device("cpu") + +kwargs = {'dtype': torch.float32, + 'device': device, + 'requires_grad': True} + +''' +#FP4 values +input = torch.tensor(np.array([[ 0.0000000, + 0.000244140625, + 0.0009766, + 0.0039063, + 0.0156250, + 0.0625000, + 0.2500000, + 1.0000000 ]]), dtype=torch.float32) +''' +input = torch.tensor(np.array([[ 57344.00 , -61440.0, 65504.0, -500.0, + 448.0 , -480.0, 30.0, -31.0, + 26.0 , 15.0, -7.6505613e-00, 5.9832452e-00, + 1.5625032e-02, 3.1725775e-02, 4.3268750e-02, 6.2655000e-02, + 1.9545313e-03, 3.9045625e-03, 5.8845638e-03, 7.8089750e-03, + -6.0151856e-02, 6.9784373e-03, -1.6634936e+03, -7.6505613e-04, + 1.9545313e-03, 3.9045625e-03, 5.8845638e-03, 7.8089750e-03, + 1.5629950e-02, 1.5225650e-02, -3.1256500e-02, 9.3857500e-02, + 7.9284608e-01, 2.8815269e-01, -1.1787039e-01, 6.1035156e-05, + -1.1787039e-06, -1.0749481e-01, -1.3085605e+00, 6.5981364e-01, + -7.0325255e-02, 2.7448297e-01, 5.5694544e-01, -2.3220782e-01, + -5.9746221e-02, 15.23213444 , -0.00004323 , 1.9767435e-04, + -1.2203161e+00, 2.9099861e-01, -7.9642259e-02, 1.3200364e+00, + -1.5196867e+00, -1.2530587e+00, -2.0159689e-03, -1.9767643e+00, + 6.0834163e-04, 7.8943473e-05, 7.8247029e-04, -6.4658634e-05, + -2.3020705e-06, -1.5630834e-05, -7.4762434e-07, 2.1336775e-06]]), dtype=torch.float32) +#''' +#input = torch.randn(4, 16) + +if options.sparse : + check_forward(input.to_sparse(), options.cuda, options.verbose, options.sparse) +else : + check_forward(input, options.cuda, options.verbose, options.sparse) + diff --git a/READMEFP8.md b/READMEFP8.md new file mode 100644 index 00000000..968b97b8 --- /dev/null +++ b/READMEFP8.md @@ -0,0 +1,82 @@ +# FP8 Emulation Toolkit +## Introduction +This repository provides PyTorch tools to emulate the new `FP8` formats on top of existing floating point hardware from Intel (FP32) and NVIDIA (FP16). In addition to the two formats `E5M2` and `E4M3` defined in the joint specification from ARM-Intel-NVIDIA, the toolkit also suports a third variant named `E3M4` which follows the guidelines established for `E4M3` format. + +Following table shows the binary formats and the numeric range: + +| | E5M2 | E4M3 | E3M4 | +| -------------- | ---------------------------------------------------------------- | ---------------------------------------------------------------- | ---------------------------------------------------------------- | +| Exponent Bias | 15 | 7 | 3 | +| Infinities | S.11111.002 | N/A | N/A | +| NaNs | S.11111.{01, 10, 11}2 | S.1111.1112 | S.111.11112 | +| Zeros | S.00000.002 | S.0000.0002 | S.000.00002 | +| Max normal | S.11110.112=1.75 * 215=57344.0 | S.1111.1102=1.75 * 28=448.0 | S.111.11102=1.875 * 24=30.0 | +| Min normal | S.00001.002=2-14=6.1e-05 | S.0001.0002=2-6=1.5e-02 | S.001.00002=2-2=2.5e-01 | +| Max subnormal | S.00000.112=0.75 * 2-14=4.5e-05 | S.0000.1112=0.875 * 2-6=1.3e-02 | S.000.11112=0.9375 * 2-2=2.3e-01 | +| Min subnormal | S.00000.012=2-16=1.5e-05 | S.0000.0012=2-9=1.9e-03 | S.000.00012=2-6=1.5e-02 | + +![DataFormats](./docs/formats.png) + +## Installation + +Follow the instructions below to install FP8 Emulation Toolkit in a Python virtual environment. +Alternatively, this installation can also be performed in a docker environment. + +### Requirements +Install or upgrade the following packages on your linux machine. + +* Python >= 3.8.5 +* CUDA >= 11.1 +* gcc >= 8.4.0 + +Make sure these versions are reflected in the `$PATH` + +#### Target Hardware +* CPU >= Icelake Xeon +* GPU >= V100 + +### Create a Python virtual environment +``` +$ python3 -m ~/py-venv +$ cd ~/py-venv +$ source bin/activate +$ pip3 install --upgrade pip3 +``` +### Clone and install FP8 Emulation Toolkit +``` +$ git clone https://github.com/IntelLabs/FP8-Emulation-Toolkit.git +$ cd FP8-Emulation-Toolkit +$ pip3 install -r requirements.txt +$ python setup.py install +``` + +## Usage Examples +The emulated FP8 formats can be experimented with by integrated them into standard deep learning flows. Follow the links below for detailed instructions and code samples for exploring training and inference flows using FP8 data formats. + +* [Post-training quantization](./examples/inference) +* [Mixed precision training](./examples/training) + + +## Related Work +This implementation is based on the following research. Check out the source material for more details on the training and inference methods. + +``` +@misc{mellempudi2019mixed, + title={Mixed Precision Training With 8-bit Floating Point}, + author={Naveen Mellempudi and Sudarshan Srinivasan and Dipankar Das and Bharat Kaul}, + year={2019}, + eprint={1905.12334}, + archivePrefix={arXiv}, + primaryClass={cs.LG} +} +``` +``` +@misc{micikevicius2022fp8, + title={FP8 Formats for Deep Learning}, + author={Paulius Micikevicius and Dusan Stosic and Neil Burgess and Marius Cornea and Pradeep Dubey and Richard Grisenthwaite and Sangwon Ha and Alexander Heinecke and Patrick Judd and John Kamalu and Naveen Mellempudi and Stuart Oberman and Mohammad Shoeybi and Michael Siu and Hao Wu}, + year={2022}, + eprint={2209.05433}, + archivePrefix={arXiv}, + primaryClass={cs.LG} +} +``` diff --git a/application/nlp_example/config.yaml b/application/nlp_example/config.yaml index 00f64222..27c94274 100644 --- a/application/nlp_example/config.yaml +++ b/application/nlp_example/config.yaml @@ -6,11 +6,13 @@ quant: symmetric: False per_channel: False w_qconfig: - quantizer: FixedFakeQuantize + quantizer: E4M3FakeQuantize observer: MinMaxObserver bit: 8 symmetric: True - per_channel: True + per_channel: False # per-channel闇瑕佹洿闀跨殑璁粌鏃堕棿 + mode: STOCHASTIC # 鍦ㄨ繖涓湴鏂归夋嫨涓绉岶P8鐨勫叿浣揜OUNDING鏂规硶锛圫TOCHASTIC鍜孯NE杈冨父鐢級 + scaling_method: no_scaling # 鍦ㄨ繖涓湴鏂归夊彇涓绉峴cale鑾峰彇鐨勬墜娈碉紙max/mean/no_scaling锛屽叿浣撶殑鑾峰彇娴佺▼鍙互闃呰E4/E5鐨刦ake quant锛 calibrate: 64 pot_scale: False backend: academic @@ -31,13 +33,13 @@ data: model: type: bert - model_name_or_path: final_pretrain_models/bert-base-uncased-mrpc + model_name_or_path: Intel/bert-base-uncased-mrpc # 闇瑕佸皢姝ゅ鐨勬ā鍨嬫洿鏀逛负瑕佹祴璇曠殑缃戠粶妯″瀷 config_name: null # pretrained config name or path if not the same as model_name tokenizer_name: null cache_dir: ./cache_dir use_fast_tokenizer: True # whether to use one of the fast tokenizer (backed by the tokenizers library) or not model_revision: main # The specific model version to use (can be a branch name, tag name or commit id). - use_auth_token: Fasle # will use the token generated when running `transformers-cli login` (necessary to use this script " + use_auth_token: False # will use the token generated when running `transformers-cli login` (necessary to use this script " # with private models)" train: @@ -65,7 +67,7 @@ progress: # Deletes the older checkpoints in the output_dir. Default is unlimited checkpoints save_on_each_node: False #When doing multi-node distributed training, whether to save models and checkpoints on each node, or only on the main one - no_cuda: False # Do not use CUDA even when it is available + no_cuda: False # Do not use CUDA even when it is available (CUDA鐗堟湰鐨勪娇鐢ㄥ紑鍏) run_name: null # An optional descriptor for the run. Notably used for wandb logging. disable_tqdm: null # Whether or not to disable the tqdm progress bars. use False or True diff --git a/application/nlp_example/glue_utils.py b/application/nlp_example/glue_utils.py index 3580b2f4..25b0f8e8 100644 --- a/application/nlp_example/glue_utils.py +++ b/application/nlp_example/glue_utils.py @@ -45,6 +45,9 @@ def parse_config(config_file): def make_huggingface_training_args(config_train, config_progress): training_args = TrainingArguments( seed=config_train.seed, + label_names = [ + 'labels' + ], output_dir=config_train.output_dir, overwrite_output_dir=config_train.overwrite_output_dir, do_train=config_train.do_train, diff --git a/application/nlp_example/ptq.py b/application/nlp_example/ptq.py index 0ac959da..c37a7704 100644 --- a/application/nlp_example/ptq.py +++ b/application/nlp_example/ptq.py @@ -26,6 +26,7 @@ backends = { 'academic': BackendType.Academic_NLP, 'tensorrt': BackendType.Tensorrt_NLP, + 'sophgo_tpu': BackendType.Sophgo_TPU } logger = logging.getLogger("transformer") @@ -91,8 +92,9 @@ def quantize_model(model, config_quant): } } } + backend = backends[config_quant.backend] - model = prepare_by_platform(model, backend, prepare_custom_config_dict, custom_tracer=HFTracer()) + model = prepare_by_platform(model, backend, prepare_custom_config_dict=prepare_custom_config_dict, custom_tracer=HFTracer()) return model @@ -177,19 +179,18 @@ def compute_metrics(p: EvalPrediction): if training_args.do_eval: if hasattr(config, 'quant'): enable_quantization(trainer.model) - evaluate(trainer, eval_datasets) + evaluate(trainer, eval_datasets) #姝ゆ楠よ繘琛屼簡fake quant鎿嶄綔锛屾敞閲婁互鍚庡嵆鍙烦杩噁ake quant鐨勮绠楁搷浣 model_kind, model_onnx_config = FeaturesManager.check_supported_model_or_raise(model, feature='default') onnx_config = model_onnx_config(model.config) export_inputs = {} - export_inputs['input_ids'] = torch.tensor(eval_datasets[0]['input_ids']).unsqueeze(0).cuda() - export_inputs['token_type_ids'] = torch.tensor(eval_datasets[0]['token_type_ids']).unsqueeze(0).cuda() - export_inputs['attention_mask'] = torch.tensor(eval_datasets[0]['attention_mask']).unsqueeze(0).cuda() - + export_inputs['input_ids'] = torch.tensor(eval_datasets[0]['input_ids']).unsqueeze(0)#.cuda() + export_inputs['token_type_ids'] = torch.tensor(eval_datasets[0]['token_type_ids']).unsqueeze(0)#.cuda() + export_inputs['attention_mask'] = torch.tensor(eval_datasets[0]['attention_mask']).unsqueeze(0)#.cuda() convert_deploy(model, backends[config.quant.backend], dummy_input=(export_inputs,), - model_name='mqbench_model', + model_name='bert-base-uncased', input_names=list(onnx_config.inputs.keys()), output_names=list(onnx_config.outputs.keys()), dynamic_axes={name: axes for name, axes in chain(onnx_config.inputs.items(), onnx_config.outputs.items())} diff --git a/fpemu_cpp.cpython-38-x86_64-linux-gnu.so b/fpemu_cpp.cpython-38-x86_64-linux-gnu.so new file mode 100755 index 0000000000000000000000000000000000000000..b3cd17defa87a6156e5859599e55872bc60294ec GIT binary patch literal 9341480 zcmeFadwf$x`Ujjq3q^_vO0_CTk)R*~4Nz0XcH(3QZ}t} zj8SyG;Hta0t1jz$LDviF0tszt!CL_l5Yz$!CziVk1;xJKXU?V5L$kW?`~LQ?4WH(m z?|0@rGxN+d&sTr$gWHcJoPiMo;20}?bnMqdy3MwB_?hJ->L#m-WzOOS}%i_jp zAs60M-Y#?M)nJw}r4dWFv*`<;+3Z`pCRpV>saO6vmfk8Q&${a$ZO=Ngpf z^jC}WoO-1+=riS@H}C&VdRfYI$?wSJ(qH#>mgTEgHN8oAj`Z?dKZ-gHla=(M)x%0f zsCt!iCF_!8kj;#rm~DW1LX;OJdra586inf1{)w|vCpOJUPP+0cA?J?VWw-5cA=oeg ze;?tGo+X{XXmsA-e0Aux10V0=H#}TByVGUyS0-MxJT7LrA>p+bu3^bjeEsKR5TD|& z5r13pw+(;W@%IJ(cH(b0{^+wuzOe7T`2GriU*qo^{59ckAO60@pM<~f@OKPE}T?1D3tDg8=&78BlE`6=~`7H;Z zc*Oj{@yvTl7Ja>N$)gY6@XrG8o}vk#?;kkmsjE)jGHG8)M7r(qUn<_S{M>l?$(Q;* z^X!MNl$pLKyL5VJ`mXuYdcW7LblwLC9$&ooz?u7B+4e(m=9yq@?{!l%4_Dm(Ve0Wk ze_MI!rHixHAJ6WKv*)Mp9M7D7Pjl@vg%dt^ zy5A_6KD@42DaHktRj5GKHu%6{2_5tI!vH#ppK__ea0xF!xwpY6I*Pv+2Gmjf&hP^r z@vS{OmUA+S{M?H=mUC@ag8`y0_`GQ9SpFAL_ypv?qjGh+qGS2}qv$gYh3!brqw;o(qEG9U9mkJ}QeO>dp&jMtvMA-czgx%g&mf;2$vF~5pYF># z*4u(8`P>;rjvadLDBZtC;a`uUH*$s@$q}OD=iaik6-NfuQT*H9DEdr{(!b1$B4=Qfde4qxKXc$8Ix3eB{52Q!76wBv=&hspU8C6Z zTT%SLvr+WCI*Okh3_W*bKXan!|N1E9x&dWM>Y!Y=L(d(xm%F3r^97{aQT*$o*zLwB z{rhL|CmqT8JIIdsP4OM;e^?azDURY#o{W;u*Q1oTHHv=jk5cagqS)<2QS^3o6nij6 z>6b5wV$UWhtt0#Fi}_VYe0G%lKNO|kd=z@_NY1V(^))F8s?lPLb-1k&xO z9`{DcPhu21MDy)X&wHczv74ZOsvG@BjFSII6g&AkivHh?(k_-qDc7CICz8J4b5oS| zz8o<-%Kz~w{^VgOs3SeEixR&siXZzbN`LoP#P2Bo$D-8xktq5}j^bCPDE1Z;#lJlg zMQ^!Cw~MS(yPXh4&Xg$mv__FXKZ;-4 z5vAW86~+GdMUmer%6M;Vl=#P@j9WTIDc60d#E$Cs*(iK4N`59q;V+3|&z+*^ZF3a6 zwMUUZJW9LxFv@%?GfI9IN0GlJivO7##ctO^PDlFe7sWpuh|-R}jv{Ap6#djivFF#K zlxs~CKejeXxfVy^S4Oe(#whmodX#?IlGw37ycc{wc$(R7LQ(h(WZwqEAK~}<87?!F z{?wkIjPcSnhKmj9o8*XdB4;f0Nqp&-GVwO-l0FNt?nJ!dEg9b@ zUkop^{9IM~VL&7NIHKtbfWxj>6YqoMSdzWoYaI2grRM=q|^f9g@i;CG0aL zeyD5v_=gqwAw_<=65OiDsZ`{=sZ>~uVh_`F_Rvp})1=hPTqWIXT$%p=A`@!8UyAx7 zdlq!|oXhlfnITz`4|8RomG~w(Njm#6sP>?=!r0_kW)UUzqo?BE>>@N4_lzH85gWEYZ)1Buj^UN(Q_j<~vW@gMQDfUdwnKREL zhjkp{&aKFC=NA^|%qzSX5$~KhBG;NcJh`ygTjnb-T;Oq+ddu8MVy?$)y|K(Q&y!Q` zahG|@OMGRyo^rY`Fjgf->%oU%en{dP}Td0|Phb)<5WQQ|8` z2JOS~u#(=dN49d2)8^#lxl2mD5Sla3os)}vlq=Gv-p=xG?I?L768fuUz9~ zOhJb6w`FFzOLNL{=FRiWg9)JabMJBI7Tn{`&ncXT%5YTVx+g=yg+(3*m5`{iGWQgZ zcZ#>Huz0Q#g{n#pmz9#@c6%yvJxr-o<<=4HV20H(`PM1Pu(CU$LD&JZRI;$`5z_0g zh)~u4E74?yGpn2fQ!aMr=6G`pP*So}PjQ}t<$~U}z*Ck#uVkTCQWXC7oOwRAXxYUJ z3-QQgPr0wiqmJ}(u z&LXd;JnPm4MR=6eH>|t__3vpjv<}rT=cB_QrTtEI!a&>CnUVxEWaU1;%<~(PLV3## z?yN;_yNBfuwUb?3NTqVy%Co1AOVt|MSSk-lnXnb-O;LR86h(EK-sR%vd5Ut1=aRps zS{jv8US2r2SX1Sc@-$R%kk_TaBlg&idC49<% zA|>nADI;V#-cbvX;gT}v#Nj#KQBys|y{vQM6t^|GEqe~&QF$IO{Q8KIZ87D(7!5r?$2X7qi+S_h)WYOK zJ4T^zMmO^BQ~9;N`9Ds2IQ0G7lGYW_+EK-{7e&TZlyi^A?JF*v@AEh(rl7`i3sQ4@ z-jcaxIeFIMHsubEHpej`OCAF72~D2kK#{y?KahjQIv-}~pbkr}e|HHpe22aCHW(`` z6vK#rp9ono4V0Anv5sU}$|&&U-ZKvMMQwJ)LYO{E3YIzo)p^RM zFh#TNIb1BcHM?_s6%JWiq1RoAWO7iRIW#mWDp{a-lw{STz-!O-7RVhR%E=nEUU*um zkNi2tCXPAnBV^@L`%H$tSZ_o&(E*WpWkc$6i=oN~6A8J&VTsa?0}1zmsnq$?+J^GM5d%KN53+a%=Lvp0bj1_q@V; zJR@^qQy2k}|5W+}mPo2PXF#b^;A*EJUQQV`N*LTkbWe;Q<;q!vPO}|l_be#P)nYMc zqQtfimxqtz$Uf^vht);C!;v-B zO_L_;h#JkX>`IrHT(P&Rm1CW*-TR9`6um=iqR}UN{SFondL4( ze}Mvz^%c(ZqA+b~gLHa5MIh9v7Wn`UBEm%I8jP0KWO?NSl!N3tiwnI|m03E<$L{1z z;aNnI8Mn8*T-Hf@N}hJL=irG!s2OUfE7*aOU&<$@dO3NE5LeM8Lb*DH4Z~+BK}xZ> z%*;h9P%xk2?BZM+ws`X7oU`KDa_9RB%RF*4B&DeVai$g&mUGW*rI3p+XPgUns||za z?5w-Uphi+m#b$|4$yv~rqEZ59xl@@$D@HY!`U9-_u>7^-l;O3)!9FSqQC9MTSgyd} zVgcs4WVv~4Zm!73w3#N-h*auQ8gs@tUvVz2EvXZBwN9XDZa%KXLWvy}D1(|&4De{S zFDrt3R-LxX?n3S_C~3@)Q{5P@(KsKo?W{7a++bM5rpodR!b)wj*n>f$Ok=)I#(Re+ z=bkc8KBY9-lh1O1(VOh%b4zIfg3Mo8bOVB8abwzp;ikIQq0Ia;*ebwqWM--}Yh0Ua zZr80SlPWlTrH64qB+VkJC6_3pK(>O$u+qWNyt|F06mD-vXjaUugi3^>u-UazK`4Q} zFxQ*Kd?qu0XZaN09IxD9DToS5%|mPWjATAMYZ1H8D8Z73au3VnA~W?XGiX|`fXXJ8 zcxi5ku{cfcG^ZoGFtsJh&!v=cQ*UNO>gor1U0TE-!{Ua*)K{<^DXoOLMJ4+PQqrI` z0xcdEe`GhydX!>s+e!$nyi%1;S%l@@BDqQ_!Czg7Q#&=xbkM2HE6MSymhLJko{QxQ zx*$mIV%%&}(*~75`8iv;y~IZg%vx;58>R|b>gsRCJS<}%+=W$X)+jJ~8O4RW?Dw$+pwN$b|(Mtuu3?p!O?+ zS|A(7lwyqGUQk%(_2tlld#RUo_U?N;iG#iBZpd3 z4H?s>*~U8E)}bSY4(CA%4@U4$aRFKa){)w6+Z|Fc48t68SZ+yiUd}LNU)h4gg*Hi* z4P$mMho}r~8Ph&zEZz)Lwo@7Iif{Ty&v#|d6QZZ$v=B8Mq>=)~$L6ypi9+O9ZCq!= zMUWUvo1sdmy3ZqA8+;XWn26!8mMo`wo~7$k)&+xiCL#~aQl*vD|3A?7OiyjeNJ zT@5`I>>cztcG_bNS1Wk?8}~q2oekFm)5nM!drOf;_=(4!!p?JdLy8i=S79L(*u{{p z%DaNf6Jwa5V7YvqwDPI%K_XH1NnRvhOO^K{RX)8!<|}mk35BoJ@n&UTdzFsQQqD81 z*74~IzedL=EBsm=pQP|19dB0n1|4ru_(mOnB38E8T{^x=;hS`Pqryu%KBVv`bo?5H zZ_)9q6yBhmUr_bDLgC|de5Jyhb$qG9C+hgw3O`WKE9V&m9p9wzSvo#R!K-xq3gsNj zS{=Ve;X^uprGguEe4`S7myTECOFG`5oJTsLC+YZfMeZsc zugY)K@oSX$yL5c%-{khvq~lj8{0SYO{-PY;puAtK=D$HLr;blj6wu<1cc4(sg_~JP>`dR6CWUX}DBrXKMacy;bOVHQh=b-=yRxNx4#L*Wyn- zCsZ1RPv-dRx%jCZPiKJCk6ORniE9PNH*x%Oj{k_`S8)6>j$h63Q#d|Ov2)5N&nLAKBfneE z^Xhph3dNC5tzMEjIos$#<&)0w%_@m&7RMjv_-P!!i{oc={Er-8!0~j}P5o4G{8p93 zwUXmM=lJCu|2@aA;P?o~r*rz@`L$g9d%5^h6h2XDzobt-z7T~EadKRo{8hiS)t7uH z2d?IL40N>58je@b$Wh>0j>i!=?Gxg79JSRx4IGbQkM?Qg`0W~NFzn*^fgIn&@uxXn z;`om_{shO5$ER}q z9UL!kyn0rag0ncjkc&T!rG@e4V=lH;p5emTcK#PKUQ z{$Y+^$?=bH{3?!rg5%dJ{SnzX&#&g}V z$2V|1UfIw-jU10Dvi8}<@#@)r3T)!|tF_ql`z^ASQQh|C_!Au8$nh;4-;d)B#x^dw&d=kfB$MFL>eh|kebNoS$Pv!W*9G}kddpTa<_!~Gri{poI z{4|aq%JH)~ei+9WaC|bymvX$7<109RILB9V{0NR;&haT6zk=gOar{b-|BT~Tar})O zznbGyIerbtr*ZsRj{hCUhdBNwj&I=jF&y8>@#!4Di{ot^-^B4_IbP!U430m+@pg`H z;dlqf8#=YwzxoaZ1;=sxcrLz~;{}dSDaTLa_zI4n&heETFLC^Gj=z)RS8)7Y z9KVv|XK?%~j-Sc#t2ura$FJe|FFAfK$Is^Y5Xa|md;`bN;rK?5&*k`C9G}PWO&ssx zc!}flIsOF4&*k_Qj=!7Z4V~NU{~nHyI{9KVR;1&+Uu?r6 z9KW37t2urJ$3MXFD>?o_j$g&`%Q=2E$BP`lhT{Vqzn0^7aD0g4gB;($@sDzRBge1c z_+1?T7{@np{No%iar_@R{shNA$?+{5{}jg?=!d@W>59Ln*`53q$MMf_yqV+w$nl9B z{~X6Bas2ZfKak`9#PP`-|7VU*<@gskKAq$L!tnyf|CQsjIQ~VBpT_YobNp`Vxaq5NI~>X5fnYop+=jgv=* ztttCKpK2X&02fnV1H~_(ES`$&Yi(^^!LX6xts15&+^I-d!<`v^Q^PR~zo6k*h9A># z7lxNaT!?dAsDq_=ccZRJRrj3nLk$xKP!EjFvU(B#k!zPAL z{-T!G%Y&^Q%YreWIfLisgJ8ylx0 zIU2r#;Yk|4l3|;MX=CG5#H!)m4ENJ;ABKBs_-clY8t%*R$@6OYlNjEo;eHHn)$lb8 zhc(=v;WssGVfY0N4`BE)4PVRfG7VqH@O%vqWH?8|gBYHq;p-W;X?QThRt?hz1nOVI zLm2Lcetjp0ce{vE?M4UcBns^Oa$?x*1~4ENOV%?ukgd<(-Te^Sez z&hS1B+Zf)e;js*dHJriln;Nz={DOuZ3_qsfaSSih@OXyjYgk}7N5f8rCuw*B!!`}y z%CJ?#E{6MQIFsR?8m5JQ)W3#rWBBAbwftEO@6+%khPP^XGQ(jFPht2?4NqnG1r292 z{FsJsXLy;0?_hYohG`)m^{?US3{TSVoebMFd>6x34bNb>pN3~L+*8A|7&dCy&G5;y zYWZg~yidb93~$x&9EQUh&Sm&b4d*fZf`&Z|Kc?Y)hL>r0F2nOROdSx)ui-+5Cu#U@ zhHVqnga)#g3u$SQ% zH0)#eF%2(Zc$tP5GCW_y6%6NScoD;sG<+|^HVxm$uvNp08SbayB@Fk}@cj%MHN2GJ zlV{ZOS2Da$!^;@ns^Q-=9M-U(;Wss0#qbLnu4ec#4L`u}G7Ue-@O%v~XE;a0BEyq3 z{1C%74F?#uYWQJ>`)N4Ha8C_C!mv@pk1~Anv|9cZ4DZwMV+?QA@Z$`JHT(yL-_-CE z48Ne^CmDWB!%s21Ov6tzJYT~r8P3u0GYn7C@E;krY4}-&tr~ug;eHx^p5dMv{u9GS z4gZSev#o9H2f07k7@X2hL@=rqt8$9yL+a- z=r?G*2|V-3_BdOIGj-UZ!(((fMTdvz@Bkg|qr(Y0+(U2Q$_dvtiF4rl9drVcxFc#ICG z=99wKXX+ljC_UUku4tsQXrVeN8aHbABba;#or|9qy6=w_Lc_FaWEL698W2Ij(B@o2pGb2CZ z#u>QDLiF$IS{6#*Qb=$~NcsmxtuFD5^jBO3@f+z`%%XrP&I%UP>I(W+N>x}~Y~BgB zd9xB)Mh^rWAt6{H38TmK7mWsQoJ%|{?e2}^M6t#O7C(0>`#d&nVO>tg^}JHa{Q@_(3jNz|fjy7Q{_h&b}K; z5dx{w1T14nFJUc`GdIVRBX`AX4f8Jn* z=CZS-!N|T#Jfu|H5-MjP;WCzbiu*yi+Rnfn+9#2sB+&;+Tp)56^4uo!dRFHNH6$yL z60eAS@dA?wly6K$eMvrs-XuFIh+F&xN< zd*;mfw5^NpF}=}Hch1}+>35K}#n_m(amld2tw?VK*o5n4HZ12;$ld7QXapbgKwZn! zjdkbTo71)~jRdBch2X8`y0gXm#m2f*?!9$q@BjLQ=}l|eM&G5R=({MFm>>H?M(eib zN5qZI+ZoQbZbsn*|F=!TvLj7UWZn6a9`$KPL!JawI$+z$~~5bc(M6gr#n_%7*Dqwq;Wu*y=2=MnUYEzBKMah@~Bwo9JNn(R_GhoUhKwO8+r|WmK8gW0MTITvJQKD{B?b($ZuC+PN zhx}plgkTIUOd0B0%=WHHwYf_YU0sunwJ%hdV;-zMuq4sm)oiRAIel~8&(p)T8B6+X zjsx2#6jxibBsFZT%iPmYH*-&8tu5QPxz^_KZL77V?Ad}qS7=4u`T31;_&O0+J7>wI z4Un>>Zb?Z9&!>fq$oi!r2;ErQucBxC14#G^R~IDJ($n502~n=FcS$y;Z81HX1v7-g zYB-gwK!0`6c8BeDTefZL6jts2BePMb@mzh|Ez_(aQry2flitmJ0yjc4OG*+Rg#t-& zjZoZ@k{+u_pG!(&r{O9D`}BLrXmADle2I>!ZvU)Zr0JVwji;Q309P>K4GMy$AC&0< zOL27tdn^Y{V)vjO1unG*h8zG9h~0s2dmv>SE)MaC&H9B+thZTf!%f{B)-yKiCisLu zQ512^KX9>$JMG5roMH%tF$c^J>tS2CwtFCUDU&j!lEoZt7gK)Eh?INSGxc_HfDd;{ z0-W?fYzlj1%*`w!2}jNbQC#a`h#5?wfs|epii8|usw)`#3;Zr}i7fQ81*A|m74SbRnHIsjSr1_}+Ntfo*+_Twxr$xf7IKTq*5kPWHxSWJ?tXaSiE#Et`t zEOy5Q1-XeX;s7Tukl5Q1=pRV9fpIBTTx{YdhxMR!BUAPp|#eC+;AnBN3}8l6Z|H&`*qgksgbE2;XADb9A5Z z0EIFURk({HB9kJC{TzWLF?JF?7JDPU#RMnaCyby_Cc=t4CcX-FoLXPq57W0O4I0gwHEwFw zL-y>d8?&dWHk|U^NAMKBZ|es5cEh@pT^4iQkLMd|Z3Djf7SFjhhw66h3Dte&3)N+w z-&Y%6vZtoD#^x6?NxzSi)+ogP9{kADhWNA2Eai8np^E_zwCd?A3Q z+~y)pLeR78F53*-OxrA*duEN)_s9kQk_`qF|Dg6;snW~GS-<7R=z{t!GwZiZ4__C9 zm+;?cFmAC2Z!+7{&Y6M^cns5f8P~f4{p*_U*y9k-o8G(i2Ge`x$)@-28wy`CrFTBu zI+%=sK)3eFhu^;>f7u~J=X`OoWoZ7g!*pl2T!R?b=8I4oq253iif}a{3(z(6U=pEg zf!NK!Kt>{PFLaSYxT%xVSSxhSh;xdY#D@F8Z-?r?mVX>*YH!|#~ z+GeWyYbS#t-`*YJb!U%<(8oIcwJ}crrkL9BqS_Gr(;_(C@1}=p!}sc+hpB2B z9;pqNO^>lyPuCt;)Gu@~zTx2O&enwtcGG*i;IG7xO$^(`-6&Y=QY6}Bs0^RI$iJ_z zKVr&X*4jEN-SpHZo9XF|`TnzU>83x|o8J4{*6kc(?xXBw!-a3a0K`CrznEq85rtxdcns#xvMA|>>TUMm zWHf+2A1y$$H@&ynZgOt3EsNl%kKhNl+QR4h+QQ%U4xf#&8@KxRne2nM3b~tIfnFck z0v9;~_r^H_cgENPcgNeTpV?6tiDI+E`iRmI{wpxS1 zpFEys<$QQLBIOTg?QGg`62v*v!+%36{!Jz(Dtzc7;|WsF2rbrGinS#EeasbFR}MP6 zy=8lo>AiWDp;Wz>`0GvK!x#0d>lY>?gB_huCj*MCr}GV@$$6HGBgA{st>H$f3~x!`bE^N!+O-l8iO_Du%6DhBV)(#z1VVuYWur-yX6FF+Zytp?P7W& z1pkt-btt;I;N7hluwZ0IW10`qXc01G143{e@K0Rv^bh2&>RJY#$KMalH*5Mv`I)Ei z-Qh?M)tx-PSqR#Uwc+`-4a_A#ECzVzdd5HEO#8%CH4DN8FBGEEKws2Ei zzN7lIscH)zg1N?E_|sZ0+p^8S-=x^`#6Yh`TcDF_v5vqTSS&_4i!I-hy^493CWmz| znJAg(x3cxtqd|c_Oh!oq4oYv?StCr=#_Sh{Ycplkph?y_Li zM)tERmhIDjHpWzq4%P4>70!CpVMV&9$w3_>Lpp-`%CmeI$zqS#k$|`ro^qZrtdw%2~0A+r&nj=_%-Gzd790 z6R{g?%l0D+m1AUYe&OATi|sX>Q2r06qBR|d$&eAy2%s7twkW5!yDw1K7!8) zAqCV9B!`*Auimmb#149V^5ux4OBGnp_Za>V%6DEv5#2Q;*d=d>PwwoTF zg)~iX>@sdQ?m|N5v>H?JQc#GI=nM>V1?O3kT)`9p6oN6t9&g4q-WeE=siWPJ*gSp( z`n6u2OjWUnQoHYL@nohAax)~g$S%RK@-V%2>nQ& zE|o8_&VRvnae07y@58`=#=e<5(Tq4_WAsuQiPk-pk< zZ2liviRLHu+pp}`tK+;lY5hS~FAKku$9cN-iONA8txr58v;Rly6Kws2=B%{l#U`xM z6_m*uX&Y2*L!Bgx;o#V-{Todt=f-eL%rj?xO519BxRzNGhNiTfVha8R&duv<)gEx9&Nmw=RuX2lWB!72;hOS+d>v9oS0!Gd94`^! z1MZZ=`)rf139oSRa^!es^|?uyrk^v989k6X%(1gtYR4`~4AImV&0qGPb)Wq5W!?F* zAG_5ZJ-$Q8Jy-YD@k4c=96v=fh&}#|SmlXdN45t3DE^ud_f+?~qh;F?Xr^*8W-6GW zT%2sIJKwyGCNS$UO-Qs~Y(}5uo!(FzUW^f}I$fc)hy!dbA{J{A2hZ=S4fptNb6tvW zbKR35p2zi_Z)on}`p8`O8s;kJeX$=S>vb=k-%azCFcNdsTrgkxVftryMw_qvJiWd) zjD?D|SlYP46-PO#=xL8jB504pN=2jT(Jru0ny&;sCon&mS<|+LBirk6*sE;!9oR#+ z1I!*a!l#fum^-uwOM83h49HA@YF{yj@V#ct+fhF~p1l@@#PUV!b?ln5ldh*`ZK1V` zk7RB@F0NE_SPCoYbGrgIuRxXYwS=W1)Crb`>{KUwebn8ADI_i8U}cEagR(LdqLrZ# zR)!kt9Ov2UP>5ECutYQl3dicu#!7C2h}EGGB%vmFg=2}TqG!xnh=ztcR5;Bm+9_PM zKUv?~G=0-L5p&|0M`{a#c8g%eBy?Ep?+0L;>OpE;=z#6wCsmtG)ighW3-hJe(D#_C zasj0UaP%}8MNgot*O{v6DR`}Eve#1AUmexkf|L8&sWW$O3%B$QOIP{#H`#){JGpZA z`w#SWhQIAJsAgjDhHIV1pJ_bzJv9be8;qCJqneevAU#whNY+16oz}wwJ_sRlZTM{Q z=K`BnRJJa~aLr{&^&f=;_G0yaGpe~EAMQQfx(%JhF*r{q+)`T;xu(uYLzq!;_gMUw zd);J!fn5R6izcz&1Z};i%W{KM4y%mBX?}NX(B7$)dZcfxo2mO9_)P7Aeb;;{CrdqX zlI=_j^{y63p``*U(hV#VXJc4*mnk9O@n<4HqsYv!%k4zE zR8s+FK|kRkta|q?q1BCqQnS&(1_E_InR|>z71o`*GlZ=WCR6%uk22+yPN0L3egfpW zc&=MdZ)Xwc*2yAD`;$d1qPi8&JH;Pu;^z+WQ?d&T7MNYkl(6sh@ z1x6R7oK3QFWUWg5kr^_fwocDfO#=a#P`cXj)#R@T-rE;e;XEh)U(EfeMj2rJMZ7r_xrR-fU$#=j0`XnqWYa80Z={4 z^(%f!n?)?UQYwOEYmZDu`l#qpMXaJ9#C%w*=*g&=b`@>>QmyE&NJzQ_FkfCDX8TLn z`Gaz?X|efhh{u`;&3>%gDrevM3wfJ#_!YVo$d?MEP@kTJM+~^qwgFw)s0ddW{Dkvtk~q*?ZYq6s`4yDLmd>9lAe3 z^%ocd2D4hlJ3uaNj@-Prov}i1SnvLVh7WAPn++l4Z4Q@MP0QmhF*psEs*ramYQ1_k z?r{y4%Gc~lfSD`sv8qU4^)tUm3{ z6M|2gz*$ezR;~XVh!AUpppfYF|6=ssE{LDG#8Vj2N-Lp7%wmPneb(P?H27vX#dwPl zV8zt@w{IFc>n7BhvYv=Oo6SdHd?FqqvCZ|+zs>Zn!>~LfvH49U-0$kuVu&OlA1r=- z29!dZ$0NpWvX4kKG(RqhJDHKo8LS$LV@byZEkrV zQyPIhOoJ-?R_W!^Wbffia)e!js57Oe7LQ(Ddy4zTT zi|j%eo(30ds{RPTCa-0LB3tOGzD%AxPFjU&y|nvt&R*O0*PP-GWzr&k4?kKUW-hmj zj)(1HW+imssN#&+(E#@mh)tb~+7%ru9aSMy^&Vu}6< zD_b{%S&h=%jgp37TH@FFV&;=*+1AaQ98aUa*xuTTHtJY)CsPN_RZp`+0@W4Jij;>& zMsA=tqLJKQFh3!Xy$UBxkj@#T`bs|KCVQnI`c~m@4RW?xI=W4ceh^n7u#Sn^P4~f+ z?(xVa&WPL%$<-(TPPDOH+fDBlRZ>z$G}7i_ivIAxm)$9Wol=_eKtDWy0~>ODIs?;) z9tng~te5K8ZN2S%HtY#FFg`Lns_bIKCH@k~UM^+h&_U&rhq0plEc8Z&5JtyJ9;VNb zx628^pB`s)!V;Ocq26pm93s+GB*r2gMh;MTxNsp(7@e>N&r#Rpgbm}ch?zY<7S99Y z!~zcS9NH>t?`vF*j>6DMovRroDDMiUk zDuR)ZC6!2o1$<2QPBu^{g>loRL zgheu_C0t2$7)--a-F?5LxfRQRN z(h6m$_kB42a2rpL^gs|dy3S|s%c?%aNiFsZTeed?Z<}fpKc=-b41cw07`csl$Ezql zwy-vP2jiY}k(O!p_7xkRIkN3kHsi@8uW6f z^wFoRNuVvUV`_^b&3ApfHsTu~w-1TiiqzT%1WIuj05^XuE&HrZg)LAw8!wPf|3B(y z^l#J8+IIT!!cWTjc|500KbmW4(@&^F{Tw{2>gOMyF#V`U^JYMrUOyB6jeam5uEZk- z@6{i`_-H(w)l$7}>3Y&zIxM;QZMoeBvRZ-@T1BCyd7qSuDp7554;%_-aLnG?_FPcM zn4)V2hK?JHwbS8eDpN;bMCf&*ml}@bF^Qbmg)S+JJK++bXm24Jbz@W66s;FKJ7lUp zh2>WC-~8GntH;U;<51t5l@*;*eg81Mo8H+kCEwI-!76D+76uA%&+C7r2ZOQAv6Rdh zUjZ)lv3KJ_CXU;=xTOzn^V)86+HPm5w_>atw}>}GW$I(6K2PDuBds)J87?J$}*jjd*T#?2V;+7h$aN)A%=K}7A8KW!B`0D?F{L)2hLs-`;Z$& zwymWB6yhI>p`n1+l87mjBe}NbBo-xvV2@&~lB4wzSxOQ%Lc$zyBmoObpbgTtyODBW zNO#Dq%$y$M5{uH+gO}c}*3;6)c*+Vkdg;vSMS_1Aw%@*nj)_a+I2t6m=nN!Pbt3Cr zpp7Y%G5C?*7t2p1a!*u{6w(qTk>*kmSJv3g@H&E`v9~A|X^e(|X6Q)M7qM~DoRg%n zglC&6ff<(Mx^wqWrS3T85SDycp=B*W4^Y%cAfcF}7z{HtcAf<_;_)LOppT5?+5@zp zL$^$0o2UX*ZFQDj!YiYdx006bX|JW{y36I`wZvx1s+PKt=DY>%lw(I`NI9*FikWic zh{ITEnVI24?RqSIFwIxLO?Rx^zuSyEs3b133Y;_;uz8E-FQuypKh}o314oKs+Oi*z zerf=|AKj{gn+f04hW8LoY{T8jf0N071&YyZgo?V-6{{K#AFLP_8j zFm1C7xqs!Bmzn>AQ+sU&ndBEjpb~3i26ia4ZT=|)#>Jhk{0CZKrC_q3K=k_&y>(S0 zqEhxdkazShatsp&!%ghZQ#v`w7C)jh$EGNZE*k1)cRYJe?W&8|pwACsX|O zNnj(7gH*-`>UmNjP=bT!N4;0Il@LONK#@W6q5=@)1IoaUcZ{~47W_5ILaQZK&~g$<-q)X2{t zU#M>yIq>n-siZG8}oa?*m!em6M%HQ zj8DCpAj^^$wZVRIL=oL=m%Cl1e3HMCPFBOQSA~)I3d2=q`#h zh9dR6Fw!g~H=`(0B1JkDfvhAUP@1T%@&zi)aGvRA=@oJybTnNGA+AmVJrscl zp>H+?k)GuuB}U$+=azF6S_~b|-=XC8HG1w*ij;d{B(oA}IYqh~kw|@eP&=f)45iv5 z|3U!iG4fCK<`wnk1@(sZLs=Y3O0Cguj9CJPS7{N`yV}-2MCRL`P~{C#h5I@M?2Ftmx zP;{J~?}7IO*58h7r)9-?QMa+r+(Eq*PGG}lRIVVWfCi0*MQ7}BPa_?MId`YcSI*Pi zBu=_o&I+6jHfXKL2|Yq4A?~_sCX++@(DsxDwrk;uG1No5EmI)mAXn5)a(DdrxowBx?9C{+dpi(gY4n@vtH zcpnS8ks>Cxxe+UJI&dv4zP0^+6ci#=M9?;sakYSnW&z<1(!Pzdt>DFtW@alG$u##? z67Z%XCfUtsL?&PKPqJSc*(ejE+-$s)lL+w?=`utjqkROCfSYmhZDI2=;k& z56KydL&+1h^>$W2wEmBX$wJ`fb~Q8*HH4Ey>}97I!Qa*B3UqM>(v{B9>>cF7Qq7fU zATCx=s&^DP6UNrX=^tqHmLX53s)t!MV%LYZT@s~>U^AGW7tx3bW(Z0Mx~TFGOB{gjg*krQqHX`5=@W<9N*nn0d!X{RN!W0W8s!+nhR zAUF{kfo&k;@hIqJRC^A}gLn)p8`v+9&QLjnQ!N*`A4#d;t$-EX9SP+~OQV6EpHAuv&rZiL}esr)UizIS;e z{CYBGpEuAD0Me3aBpO;zwEp>^#It7p`xF*58KC z1HJ$+H>G$GL@D5Y3iUJ}z`zXBO)NDyb$Th1Hc^yN>1DJzX+4xd8mFG&3iW0%y9xBk zA>jxZMJ*D~VZ98aZ`K<6lJy%8P(f%y%m*b?(4qZu5K1CHl|XHX1${(8Fs=+_j9Se1 zZG#I{|4y8)m?{J=gO_&(kcBg@zyx>}L2?;43!^8TSb$|f;lu)q3*E{*xqMEsmj&a6 z*C!&&{*6?8+2^isiJM7x_0pxVyXGCto@m7qjgC4IaVMJI4(WF20qzbfh2-wy&~$_V zb~u^LJKM=*N1i#GAzf+X1X9io<{%t8<(1k$mh2@}CJNIp{@Ekt5bK-s^5)1+vW zcY~#;2>I&@D0AzafgA9i@lmh+F60OLi&P*~^AAd_r<-@l?X($#oaU8U`qCc4Pit6q z_1bnEt54=V^+AK-F0K7z zz#~sdsiRctwmWuPl&bxVJOOGJ(=$w50h%AlGy> z{b`3EHmrim*l8}a>D?<`fuZfjEf``-Ur|F1Bs_&IxdJI0CNK{r1SgoC!Eu=TRj(wa z<3sChVWHl^)-&aGGmNztBK{^v87;6m*eb4otp++5)YDOcuWIV}XmK}Rbd=UR=vU5I z)Fg@k=hf4ZPG4E{HM?q$cYG$Sy4YrVZ`~Ti%=fqUT(~qi{->K_7Ys_TY&CiNV}b8u z9C`lW2!i3E3viE-NRQ+2%X#ln8%RfXO0i^x1_DPK*EVdt2?{0`CO1{vESQnHzt&^f zQHbmT?+%d(vdGtPk$OU;6pv+h>F~SS_y%pV0Cr`n`a3JkLTNB!xdKI57)A)rw69D| zWoomImG&YSD4lE(zZU>OK*@xXu(bb@lO=mcymzs`ffn~W`6Z*T3#;2iL{ywXJv|m& zVwPsTL(e-fXQBzu87$kP8|W(4?*)xM;4^X)SXEkuyh)Q$wbbWbk5-6|A#w+X2Wn3q zD|tzzzsAho$Ow#Qjqr~6=1RH!q4CKRitDATr~t~Tjrt5$|4Lh|&xk`hw3L)YS8AH_ z{6*bgP-dYRZfu7)!32pq)4&~O+ExuLMW=$(7{ci6BvaKWriO)BSKsOs&pJhHZ=Y}t z+UGxXi*r56#-|c*baxTLXFJ>2YZR_v%toBF3GeG748pVnN9jHjj16!Z_m5@mfU1ao zCe44`m?<_x=Jo^hG6mVQ{}`y_;z{TB!!Ge`W*{}Q`bzH|u5L&db0g@#M%=xYI(s@R zDr|3}qcPo12xqnm{Z3Xcw!|5*C>W;wVjMQR#XHO94;!(`PcDsrgGU{woZ0H?r%Lv| zjBHgGq9UCB@8UAW?>F3nWRS1w0>(-BeTS3liHmqZkUr7-#wpDXoAdBL?Gk~%cpvj0 z`YaU&XXAAFxz0j(e;2&c_&Xu3!Bmw_+Y+=5PF00ontbX=VMB}j4Bn6G1%FL5ytKxJ zy@j+bf@%C#B#GDOzat?#@Dw7@ESA>trPpCg7?NVyU_W;0U4u?#rv2beP){0QdqBG3 zpJZjY!yayMD02oB0aAsR(w_UD&UplM1IZcj6LA=c0a*MCV=?|mEHpSUkby-79GRg4$ z%PF2j6jK$BQlRg@SP&ar-9pYYjm~}ztPWi!N600jrTNURL5GC!FP&j>0bimpm?os* zrY`j30oAAPcgwPy%}RaZML_`uvlXUqCq_AVOUnxbVAvI$h`dC0 z|Ds^%O)fFhWh4w9F7{s$IFzl(7#knB&=Cf1OLUDMpX41+ZQy#mBPEVcWVHZgI0G0d zZA7IB-5P{`Thgmeqs44=iUyq5bOzJo>c`7&L>zB!{z%a)ihhFZ<^lMpiigz&337;@MK<7tj3kel~)`!C(Tad6T5Q82B494qD<0dB(3Az&EG6x;GEqGHu zG>8NYinc8sj@jJOBCL1#CQt|YiAyXu(;i(t?K5?T{u9;y2FI9Wdq>gWp;=9{)P8=0 z%?T+%R>Mm%PpOw)eiMVVS6Tb7KJA-8FXSAJEPxg;{@%}CLc!TR}!-!lzF0Q)l>4=SXJ+w~cjG|LsD12`q$JxX>cc zoMxkzm`3u-@E-XBYB#VbuK8+aKSH3JP`%lU?r#y zeU~?1tM<20^MJaPo6Kiv?Q(M&9oGTcQDO!W+0US%0+Z-l@&P1PO-%qmeBNC zZZvQ(|390Yn|e(VfRigVIi;;%4j5?n`C97r=G z=?4a!fk|-9S!N8zu%A8}0Y7>Nx&~pEj6oYac?OR~CC8F;EF1vqM(Kqzv(vbf){QQq zH4DT{lpcW9uvg8phl)Y}+!fnQv=M8lp>Wm9V?lvAdI2jl@i!X{%k@87 zg4H7dXPDALrBs(@@dr;UXfrGE(AydjC6 zX5x@2UWDxG3RKd-mAU-Ea1Au4Vtv*o^fM?{Aa=+YqoIDB+^w>0=q}P*NS1s9PJE$b zI)xK?$eoSmr?mP6m0)L^t(4qEnMr0|9`oqYv1yPc=S!XOnyM3+dX+L@{!4Re)_$0Z zTo8bhy(f@IDtE2YOmk(*28 zvNT_$v=eOVku3hL69>ij*9zS9k)2AeWX06%!iAwO*5?$vc;fG>T`)apRw0|jWvE11 z>%2l#lh}i{6WPmJpkfEIS*d!JcE0{^l}g$ol_fHjF$L22;1O0UXQV6sChM&yuC$Js zD0RhMBo>Vr^9fniz2R}8_{bBQA0Qo)Zp~;xUs49&#}}-!zk^lQRpZJpK)tK|P)R{Tt;hkdQta%MpPT@ptlRnto8J z@^st&+qIe+yP+oG&A7(pLo|%0mm*Bt%%hv%Dooz3=m}9Y!QKq}Q}cBx z43m;cc|}gtlBQUnUdK)()9JV0rkjj@Pb)s{ecAH5bZ_zL*l^#_Ok{z~3@n z8w#tVM}KX+Ru&dp#Vng^*49_q=p+fFlLu0$!M0ysW!u6TSmo-9#d=fKM)*6nbQh>xjayni#E%cq))uBs z!IUfakP|pA#WVkewM#lhL0+5Qo0M+w1*mIG;^+ zT+BZQQ%a#V=CSLfchCTE3Ud?sR_vb@iR@- zUB5wPUP&3}3WSj*#tz4&HHfDaXa{a7t6G7oGT^4<)z8iKa77~lNxrm)#gq$**R?1X z9c`^1i#zG-KO<@??l9Pr*MFsth&oM!q0U1QC=WqK9C`YH?ADSf1!aTIT!rGO4=p)hBD zCh2xd0i6rR@eQwKHk-uY2O*`wN=?nyOz-0uITaEoxxT`|9t_|hFXi1`G>~hRB_34iA3>hVCQIjD}F(1FAx2xEdmSMjWEq#1gu5|CYur9`+=4I z1LThivv(&hfdb6lTBXZTAgUOqS2X@=n(#(M$XP*IPKA9-omrM)&xsu2x+swGPapjlxSDSx&rFjkUxaq;RVc5yGFU0l-x_fk=)N=S5;^ej^@ zOQ!(oR5Cv^70Tszwe#h4*vb;j^$A6hEGN=*CPAm%=O`620j-)@x?qq|r2^>p(kKqf zq%fltCRJm4DNM4)oKn2fK#gfsm~@Sybp@sajp@ahBFq8cYx@#F&EEpYV#7I@fFp@C zG7U}z)4)<{4b!CI=h|$BW?ShFvFQUdRc0XN^e}2SJxpJafnseC+D7YJRYOn6vfv;| z7FvYek|Z5E+a`Zd}4iR1? zW2*1VrPqnc2GiDl1F@TjQ01m{`kcy*0Kp)2Oxd)N3Da1d<=C^*MNa~1pDUl;-y#@!Ri~5YS)6~b03#e7tsx}_@?CT zj>1#GiI5zSrC7NZuu}<88zrB%KPX~uJ;(Bo3ovsBg&R5Y6&KD2?wqTB8} z$-_sakqAsLYbA6$Fo1-8gXxmeX&}wA4oe%` zDzfmfkupJ@sHQ#%YoAHon=Zn1wulBl|FM66n;S zntKcGdDrIc4tt|?_pfIKr~aCHQleV%0#Zc818tKq5~GbL*b4_olDLrs+Bw!hF;>8~ z(P3%n84^h+fF&>01Kfa_Ro>_^Z(oli4|ifn*B|P^wo?+4L_h4oPfi~XVS@?xIGGkK zLVt(ZWKRd=VTtLObkoza9PUVA5{)O@oJ89=%OqWlGBy8EdJqpO<(FsnSOjRkj?c~j zWXB6r^jb_*@GsUO736y2GgJUJP`+E9Ma$c7rgwLK`IZveim<5|}BnyFJ zXjdB?qHDo2D0O>D%ro~nOEsnp`jB@A)(811$6KgKHs$!+r^psG^b9mZ-55$GPdQdW zj&$HDu;>cF)=8&Hi5LJTDN_zC`?D#>#Z=p97^wYYl9d2VC`oDmrs_vohDb)+q~jjO z!O+;^6r#5s?ln~xQJ|k4@Kw7NW#5d=KTOrnBlGecq(3=~V9JqCs1QusRw@U6Hy&z+ z(x)_1y&q<&1XGUUjS8uWiX?t71QtLo_r*zTo+m87zlYP=aM4=LnyTpS0eR-|cic%Y z{sAy>4|>Ty&)~{9xk#g%nys*)%ko9BFuJL&Y}i2OP3R!Gfs8_=F3 zVHZnJk+3-wGyVz97@IFs+aP)g2&RVi{8l+fvRhDRHc1pT88WD3TaXdOjz2@4G&fQ! zJEV$wM{X28W@9+0Q77GrUK90`do2_mOsTpC%7hd zBhDRybfb|^JH}C4VA=P&W5=kUFX|w}SG95A4$o8HlQO z^=cAR{s>Ocw%}Zq(<&X^#?u(r%oBY+6Qs8d>R)X8ROfj0i(7~5XlaKyWLKr$>GZaB zOI*M#Aw7}W!|hz$7pVm>y)TT13h&di@Z&11_Comk$Ge$a+HHL+_cySdV;^=#WiD&u zVDK0GPh%!DDute(b1-;~?1efXizSz9a>^zNdy-ZuL4f@z=J2HIcsPT-;HOEKP(=LzaYrjh(3=D^AX>&+&MU2`a+{@rv|-nX$3B=8{!?S zDwtp;I|)AQxZ%5marIp#Urq+jT!~F@p>uW){IJ*)wB+{ zzk!Yg-H36d4w@S*b@4u1rjLRnevbpnC*zq>H3DdX*=b>&rP~66hn1z$fXb9@A7Ybz z{j{ya<|>2K<&|1ads?eR_zE7d24&v_H}np_@FgG{Rc8!;=~oEq=V;4KVixWw@oQ` zROXO_5?T>WPx}UtibK5T{lcW4DYiex_$!XhSS$HoPNWYqlcgoisShy(iub|vYjwPH zTTGVbq6X|(e^}d+c*i`BA-nPCqGD~2rFf3x@lDZ}$Q;Kw*vgFe#m2I*KVDOce#$w4ETT5F?h zCyG1K^GHeB`H=0$573}O8AisbC297d`;1{3H@hgzX`a3EV$*(y#!=rW@Po#Vg{64X zIjl8k>{tN%d#|+hsCfEN^B&RVZKd!NV#ZSIgFa?2C1a`n-d3u2BaX@29+ZKUoIg`f z+Z+g(z}u>-?H^*<-voM>)NuVrvqTsgV#%=3O3_TCCP4}#t4F2I*p%7|@MGqe*dv;4 zoSA*jWhG8&oB54*Wa0WR=WX^97AKp%m%PnhoCrD>WZDX2u%UPinoC;uw*fFF)>M+c zjuN3f%qG7Q%59*ZFu3nkW}VWe%b?77-X4q_Yl0@|&^0zYProidI~jg*Z$3jsg?nt7 zJ)Xqye6y+kgIMx%!a?^U&15SxYS4#bnoHjk^h@6j_V_3cDML-zJofoWW{)DF*eGoA zcFZ0sxblZSqOw67jDvLknT@s-5O9u#=>kn9?;yS>Y;V9l47T3c>q%rCYm_y`RKA05 zG^TmDm^bUl3n)uh#4$_SW1S>WXc*(NJsuCi!nCY+_DEmP=IzA$^v-^p#1PNmQ>fGw z$Vxa1@gh%w*;e5tGklK)l7GQ&Fefw@L*Dg#Bzq6iOQH8(e;eGt@aUBb^dFWpo4vMz z)<$M(;8W+jj}H6$cKQ!*#8|EIR1wb<;eeWlj$+Au#lv77sgDy3h?JjSXvP-V&y-<5 z(-oU~pQ63Z8sa(rYegbMTqW#UEQ}q3#c_=7ZAdU>R0dYB2?>_UI^5H46?Bh4tBYm# zTWxhC^{O^CR{C`2ph138x%(P3hXz1OhCDScpvU=vQALnXyQln+d`U; zOEtF;9#8!EczDq&JahS%WRLV!LD7!rB_CC-C_7Dj=+Om^?q4!#@`-CT6i>x*L}0hu|C^>6Lz7K!JuVWYENKs{L+phdKNm002538l?qLhi(DyeHcj(Wr5djyXL{}G zK=U#pKcP4xO|@`((;pFh6I&{*$cl#G2e6gwWX7Su5mfo)PNH`t8os9R1ci6@VI+#S z+Psm>BNKp2AtS-;X$3w_x$Jlb&vImAB*g3CeP3W>wK_Ow&4lddwv8HoDE9I0AbJz> zG0D?@W7N$UuH4M0g4+dP@Dbsabi;-}S%sblodze4c^z-taD=f+Z_m#qg_4~h!A<-R zJ*9o*mhKT6&drk^rP%+_9GZyeX*2mAc@f^L1u(tqYHkU|l5)+9RbL{_F75NUoN`QN zR!eO$k=s~nTX$}JRF|GT|876*+pT-K-Wj_?ytAB>bSz*m?ILwS+OVxh^Qpv#R^11s~IhFN`Ezrk7{g|4HwwgFs=GccNQn`Q!L4q`w!_pTS zNI%Byj~Q}IM@-3H$~Z;+jdX9i`w(tW=`kDB%^M%8kw~q|-{`)9FwN_VFmll(xZCsI z+SlEush(PBYM$0h6X~rrn@=Q#`B9 zGom&%a-i6n3`!BK5mwxhG-_f?zN527! zdmRE7Lg=ln=-46H2C|AF^IZPZ*P)sR{c42GI9&J`}F5 z-v5Z8`&ZyYw-XZSe;nqcRq&vEABx$%R6s^Sf8p`5nSX4?SE_b4186xtL9OtL)=c>s z1~ST0C+{ShAE-K5?mvguYDKSbcdI9%M)n$ZWjd^8XJE!dty{^zLM+KOwUUm*N@G9S zzz2`S?!Yl<<3m)zetxQtO5+wRmi$!J4jaD60?yhv0&dnYUIT_^s{w%o7a>^d-J%X- zCwp;|o6aD3W$SFiYVB35(;=Q0hzv93&kL=Y0q=TUpzTmNW8Ma3>m&+GuJ3r8S%76k zK?A&AfQ227Oi?%sC60A4zX-)TPZ5}0`^Fo$sGQTqQJRz8uIfr*?2sDu+@2DkXS>of z-}BaA5WHzy^%~&R>m?TWsWnmPHH)6_b^oGX$EF?yM{qZ{M?HT?+$&Stc^@@NRj|-K z4q?#qSLh@?UwJ{Pk_B5osq))O`O14TOnO(pR%f`?H+=MuhzlQOC`q}nr11tL}bSM;d=kQ&9E zOPq^c8J&(jZ(POt_rN>q-+=j*^f*)dHhvYv2X4D6%KDc=*6V%NTYT30!>pH*HJo9& zouiN5`pA76P9^TaX_9GH4SUefJ~Yk)72b!i=HH~S7R~x9Us(StxPA2Tcv#&DB2Qo4 zN+>aRccK5v+H+TS^~zhIgrWnKWX&`pWJ+E>8)--rp+1qWmS%IPjwF>@M3yZTAm-`? zs>EwUqv!REKHU?ZbKQvjIK#E8yY&jerqR=e?tI+I$!uhx`xq**FX`HcD8~?4vgm+0 zH6Mq-$ro&vwyNM7IY;QtJhGgvoT0)|CIbLN!>OdZL_?zLpTPl%wtSPtcSh^Hfqj*1 zT*%*|{9x9MN!$YH*48)CFT^!j`C6=!Mw@rI|3Pvhi)x4Dn{TCn z?1p}(g7{lWTLtCUk%klnow!Shp~hWZ7ek+YoC&G*mpLE(2liIKh*ecmFV;<_`6S%JQ;#z><3LDK>EQyol=n8@cc`@CkBelpkSE=7>KX1NaqJ_KzmZ5ltStA&5AnNAMw3 z9+!l*nPxhz*BW`s`4yJNAT?dKL_?o-?_4Hii{eoyY22&&P%*ssznE=tX0BR&Ms%m) zwRz?)n(aq>qMN{%jX%G(e#Xjz-u&A>2R2-%5A0A&r_*G{(W*Po5_jiKS zEqC?6$T49PEnkL;Z>ZipDsD7A_NJ)F^!pe0%i`jt+GceKF&+n*@~f>yZH&u*aG)CR zXcyIH%Kr>g4P72&%4h5+CVG#drU+rFkwc*74oxSwo^ca$UNx)_?@PVzCOWS{%p+$X zORO!mMqhW&w|V3ot7w=>rQ7RXkCa{bHKa7BYxG)*JO!+jcA2g)P=P>1UyT49&5QEK=#PLXCgz8jVro2jb z_EHF>7TPd|pPr~8xP&?e3OlbbhQiLzR4Fg+{GGT<=zS0Dyl7R`zD1&V?tQHuI};3g zVW<2aN-+0EyU#SBsTnn0q-D?85Xn_EsX3s(>;|+ak$h!0;M}_~j-cC8O};OG7_!c` zQd;wN!gJ`K_a_xG=o6TlsBM+oQwuqhE8XvwEy0ThNlVlCAaiV$O*72r+)Q|b$pOXf z&nT9^ir_krr9ump@LQ5Sc|F{33O$v`!u-w+2tOE!z11S*Q<2+Ev_P9!SPf%$?ajyn zaZlMVdHRnM@$@AWDe-w-2J=@y9_Lfbm7=$S;|jsOML=*rlwOy(OQ|Tg(4X5sEJXR8 z@9$pWGc(|J<9fAE;s;t{+I0+F1aN3I7456$s&b)8m5!m+tnR-s=69iub9-sI%M3sB zPQG{KJE~+#_>pZOa}>EFIWJSG_hIXy;Q0OG>ObY08H1#j-s3Y{SyXtq@5)C^%djgi zzO1$fEH_S6TN-v{F1XLbdz)&JWmi7tr^MuLji7^y1^#(|1E>~&Du#5)9Yl^SKsx6# z<;5;i;2lLZA*2mz<80c93d4Jb->z=|J2#-vxeYHV zKMtiuzqK``ew^N(WHolG_F2>zRWU9^loVRBvGLU{GDd3(kK289_j+djJd97nev_~6 zQ~@|HtWKhN->BYSLfeVdTiI**Rh$wviZ`mnwzC)$Lwv>{|KI7mJJIdI zSMI^6!b8J~m>K5cQw8c&U<1ojXaly&uWTsn+}nI+x1tPfHsH$F-NW>a>beD9;XC@O zaZc;#yk&;BCp=?bxzU7WcgIDAV!NqWB}j6Aq!?NuO-UGnQdXHu7i$brllIAUhPtN<^V%b$)V-s6kJ<#;=Dh7&>Wl1`Hobqk%bxj*tk?d;Qw&U8FAi=DZaOP$A zXLixdP7mvuQrJ(sVIl~tv+a%`{qPbXaeuNh51VjKQfg6myXw`S`-}+`n{7U+bJuzO zp+TNLtmDp^WWXa^vClxa^#!~>aCx7Sw%uIAYMm#C{c(3w(IQ+9`Ev<(25{v5mG@`p zeh(vSzr3QPty~#O+!@fzN!~xIX@7CO-QzJ0cI0gNGvy=y4Hb1Uvcc{h#n4`;TNJ~+}q0R$gp%f$ZJMw!m#H5P4WqeQFo3bY2~ zFN_s)&lmJlPbfUQ5PneMf`2%7JK^kJfD-DjZdG`{LO88(Jz+QZoX|haf5i9F&%OfR zdm6LuiSG}n+&Z~AzTf(4eA`}F&a*hgLR@*lRr#*MZhNPq*gmAG^m5ZJ*-yrf!J-q3NZXsxT5dWXRB~?xsf-eH@4Cw;FxB$rHQ{(l1>`g1IKQ%gTO z8Z;N_F^!|yqigUdP8O&S+s+)3H!A73HI0>#>~*T29ymbfK!;9fji=$nr}lz+A1 zu&{^r|9h<3?(FxNau}&#ql(!RgxQ>WqIuBL_}tI7K#57asHmkJt(bmHy6nw@%jx!5 zat5jK^py?R`s%Ef^TNZyMcI-7Cv)22eoVT#D7vsBw-@$P=AS5*}y&uw_akyl8we?WHe@kzN={!ZS@D5^i+lwI)@KV%M?&&kXgbrc?0NY2_TL~un7f>gV<#5cO$h*>eqtf(L=M~wT z`N!={rQ|FW%Vre41<~2x!@^ztS&^8!Ft>~N4;&>oIQ$5Y2^RiNqoBnGcD|d$s{=%{ z@U+kD*N!G;k}3BRiCol;_>ak7xI@wl7dlHqU-&-u;pyGoPghVuW-@x&Np)`QGlKm* zq?>PWW8j!|3?FZ)wC*iqn-24wnx9{*TyTbf*=4D7abxP|AhB^iTT0ZAS^tRXY;YhZ426%9hu>mQr1imYA>;` z7w`e(4txz5HNTz^DRQ{2??>$$r!flXUWs_47C+A$W1SBJy%s=E*pkbrzYn##GL|is zIt>a&3aeF1XoHaDd7|kve$;Pt@GE`Hhy3V|MCH+6x20igeHfA1Li)Smr<+R?Y!fc~ zTvgZpo@iRQ1LcOhkFI9lfogcq=9PEI#U+sR770UAV`w1~FI}i5?mC7cG?_3gtt2cS zLI5KTF$Y7>fL9GYT2h1>)G~56LkByvG}Dws)6WhH{A}SbU+FFlGKUsTtX!~}IMloP zU*v9w{v?fb{FWNiyU7wkSFceqtqO{7aw9QLbg<>aT#6D89H!bcwXgn~AHco|FvFW& zztlJzHZ!$1u)HhzVv^Z_hTG0+%qpb9vMoY>oy+d8X|?k7!eq)nda#Qa7^VC*KidS^ zwXDN9Ay*QN#V18upuNKcWNDv(pA_5F^iu)=bpntk+Q*t-p_V1R+y0*j+=aOXg#}UR zjVEY5#K&K_BT_}Wc*S(!M{bA+0 z&eqV0ZZ5Q`m2R@G%A*E^7vEdc&TlNaC?>#I8U3m_S;vIR5r?HONmt8C=Eu!gIY{lQ{I3|7&+Yv&%Tb_B+r-#$SV%Uo%GWnzowH1h zI+9fovCU7X^sdXy9&YA#B4w1Z(&o?b!_NU^uMHtu>w$;|P=wJ;`APIb^t)J4@+TyB z*)L}U{JLGbDHaX_Y+jSC?m>@yKNxVb8Qv2_k;?~xHJ<8E|D~^Trm9-92&Te5cmnm% zx?BB#>KR2CQEha@V_vW#Xr5f9ktmF+-`T6FhLSBt24gs`3YF$;t;f4%t2J~s~}ymg0HPobyZ;& z;mkcq76$VIq+`}D%sWeMG;bu0k-8kmA+SNqK=9kKWDZ3rE~-b>oO(8VzB1SX3JHe6 zVq(ehP*xm)7ME>ekK9C>yRKYm6|t!`QpXZq!5xb7AbnsjPW5BSZHUz>^GDcUM~!0z@tRt=(AZ>#PpaWy>^T6}I9`v1r^%xQ!>Oz?X8cn$0Fu z#KCykFfPiC^IY@5Sn{VJLi`9ye>E}a{Q!q~>J>MLrhQX*)FQrC6iGL(XCTD_t9Y{p zjs?BI&%3bPYG)fFW)!B`}d9si_88Gl`EDOzH173QcK~CPI2{vrNnI9YX?vbON$i3!qbaJc(oH?Np9)){h8Lc!aIh`SGjEkIxZ8*KQotjo^IaANJSSF?_ymv$tv)IHx-A#l&*k}u&G?X+)Gxl6 z;rnMW_nmSu{q2g9);$>eqeGT)8e}Y5Wok!iH4|gpYJAWEe_$wJA22Si`=MEv zi}21{^pUCECahx@N@SD7dA*64j|e=#l;7z<0!Nn@mL`TxTz$!w>4tUqTSvw-*4<0g z?sJZI4buiQekzvX)uLeo&SD441sAIAZdeB%7)Yd7G7MZas@F~>+3iTNeYanEyIdL0 zI6W=3p{b5_CZR2}fQ-%V$}eZ>+Ggs7o|<6S4^&evT!>q8_-grN3TL=BJ2M0JAysSX zDtD?$7^1A_vVuWaVh@PJ8xdDTP7@gRO~~mFWamCR3_}UiEwimC*C`>{ut6oL${j+x z*`WlZ{avQEAI)pIoC`yTve-YbaT3}7W-R^m9-j+ z)?_^mv$OCcQ5XG%>#OGKU)4LA=Zdqw?_a{IMU$eDT?T6Nn)6uap&Ib37f^9eH#gJ= ztDMN?H+MT^pZ(=0(R|6r|4i{u=i~2ZC8W^s&KoqeNS(2tckL?fdE?ydPd?QchWZ}% zjaq)SqQ#P@t`}Igs@=LmxIM^6RIV9OSshDml*ozHCgRoK(2ee2HYyK!dS+KY!v@Y9 z3;mtmh&@aU>PJqDCF?ZwNJ=X0?!5tUS20u(y+!?d;v^C>`F=p1OQ%D3hp+W?C#qpQ zhq*JXA)PhFlHWlPGVKRWxR9PS^UgeuYU&6?`F))_6EzLas}>%K(+~UQR{4l*s9hXCmCcAN@ohuqPv6 z_Cszo75HP=Mek5$xWoh*@3cnTA@2z*I+h(kVK;}`1i(FT6MlzwW&45+&o9GvYpV>% zweHDrn%cb)Nf^CVa5>NGVFTGu2u=6nVO(evDPFT23H-9_`H6)3j9Czc+od-i^xPgF zaIHG$4QFl-Dc-MYk*BxvpZ>&GHj}#DZz(UkPNeDqd(a9f)D(Z2so;CT5`=z($-dUr zj#XEU&mTEPSSw7%AHS`xl_M4_HWJZWE%r_0F8NdF;ip>cqaoqV5){L(_tlLLMf72o z<@zwoVr8i!OM#)}_hi{)7;5p5AQjTP4CrO-BV~8?L-D=oK_=quI*tB<*^#}0W5ptR zDo%e*>HB|m`XfqzhGrvT5gALXSYF;j zdi&%}DlL-UI$0Fh?>)N}zF zQfAY2G!gQTJcSk!v70$Z<3ry>EC}d^6O&5ASrU5nBYnI^b%J^t%{jLx9TG%4t79}{ zq|i?R$+g+NjA{CLuE{^YjiEVhU49-Cb(f-E{?PLS!ZQ_K?!)3QN4TFT=6A#t=m9=7 zS+TDWYgZ$AB&@w%fN@plLk)4AKM=sdh4*WLVXLJZXO$4o1#9Dad}c4y7aYZy?Cz>K zE*2ie^ixc;Vqk^6Ki2ODZe1&p56~6Q93m7Kp;qV)N8zMvg)_hDcl+H(4gcF1zboxA zK8xQx#&0E%{d)-6seXR^0H`G(>#;F_AKH ze!+m`gouWuh2K|AgGl5?ThsSlj!1-5;t{AEuS}x#^qDM>TEK}(=p_KC9rnY6ZygIO z+(4F=oJp5xCajuwU)RRZKvNw*r;cJ~)vaX7u}ZuwOoXFa*}vq=PEhhmN@hsJycE9t z$G)zJe_gKZuuc3Sg*ipI%6Tj~v%7Cp@!|KRKu0y)#2-t{&m9qdWYN|t!y&@gMw^Vo zmrUjRV|}M#Vtx~x6k6fdiKFv%WDln(tUztOlcKHY0c2>afNpF8{ajMEN62P7Kn4CD z23p*oRJzS5w~+UzMCP82z-*^e(f4_lbNCGcGvPrVtNu0c9rf?ZC^PyhMk>_&c?<`y zMWZubaIC*jJ?$Z7@2wm?$ly84R9J<|$Pr7<{kQs4AeO(8D=YT{o3!}CtIGW|C3rn_ zr!sa}X|}+E;`eB#vlkK*0e)8gyF178i^E4uql>4oSVmpkR>kg!L-R*S*crD<%}Pz_ z)sx?+%{-9}5cqzm4pu{QP(YoJ2yK0e>a?3L;tj`p96zi^-B{IJ=zE1~E>QRVp%p^3 z6lXd6%r`tEOpK`e$>PMTlvqUDzq!^o6w&r?l-X!|Y?w2m?KW~|M>72*Y`s+UEqAd| z{FczDH|Y94%)l(=7P=$;u675E{+E7gFGL8J$T;+N2ZnvL!t zV=G|qbZ>S*=-z$YPO;Jx8}NoBl^$1s(?=k0aE{JG-N*qt_7nEZ&+cBX4d3~}DPQqB zKhN;Gas9vb&JRzUq#LWG^Xynx(OQoLaxf4xJ{w1~Jz$?-w5ysc876-!Xvf|8^*52; zGhRQqrtJ}}eOzLm0J)(|)lFvq9Cmmv_sJ0`WBgTSJC_?1s7LX&_c`&G{L5un7pOcM z*=v^@s*(+Xrg{F50U&;k{vg;&_-TKwZ9QO{$a0YngpWx%?a0^f#!x)b}Qc)F}6s6U*_9x&8=mxQ+Wq&;l$ z3(j_GnqY$umJxj97VFyzw}P3P?E}&CzL$(9vljwCAqu|qYWV)FSjHmdMTX7l z^@L^AVHfa`n0K-|82)BFpV>8hc?RJAwD$}UOI}8s&8dGHcl2~B{h_08I_l79)lfyY zmVfg5TJWMw-|He_1?8!iDx&>&oY(D4HNVKva3yDPMpACOQR;u2JVEKwI?DTJlmkxz z0IJNMCbIQZ3Epd#gXG=& zhume$El%mB;ivl-O~5*DVhMnCK7#N;3*DEq2t5ljv!B8+B(tKCpBto0PwIi7x0RflEPQ$`d5k0^VQ@7yEt)P1TUH`hHoQsZs>M)&K@CcJ%oCsd;#|ER##(YQ zH^$SJy*}Hz8xG$Xfk7)dA`j+n#K9dd@7DFYxE#m**t2!yFT_XV>2i(2fWeXb1^Yqv zB=#4>5cZN|bkhD3q|8EuuZ5#CgR3REU8|V%gMvMgx?Ow3M zi`kFB2e?nSGw^?eSol8Z8om$COK9B^5YoWo_GZ+w#24G^g*b#!7Fl!X5rK61s^?2e zHcqEvhRM6DJWimc$5*&r7l!Mptovd%M=x$*4*6{(hv$jZcrKtfR8kniX{ZW+tM$wM zRpb1Jd(NMFF_AfhrpTxne5nQ`T@|E{s>ltB@EN3cF$Q|FB_3t=RCr^;pW2JD&LfD# zL7NO%R=GDv2#N9!H>r}{dlP!=$MHtCg!7{fTS&WG|6wdyZq;lf4ewuQH;#Sl&@QcG zYs$7P)uYQ@HAz@ch&7S^#id%~pAxjP(TOEz^NA-9<-t_ofHE`fi*@Q`m=PZMCH_U|(Uc)W zyBfMy_3Bu`ks+Fgvl-72;ETCZ-*=J4jxNMr@?Pt&hCY^i*6SntF3RpmTEv$)<7prI zSoR~>hte@?8y`;}KRTYdvL&85wayNywUkx3K8To@og}-kEkeP?Y;D#Qi`iH-KC6n+UP=J@n78W^owkL5OxIzWZhwzxhaWoWCzn!Y`l z2~&ZiL=?XaW+HulMPeAT%{~O&3tf3G((N8&;x?y`C~y&2OCp6J_LW@3$<#HHAH_xb z7P!dm)c+;+m^WH=P zy<(kDiZ1b^ChHR{vUndjmgL$Z_C2B8if0*w*6kPB%uM6sEOW(+9H7sQU90swmPATr zO%LjyZ+l5h{_%Kf*O~=y=?zb*;mv^KIH{sGW}j)-w7M)W_^VX z1*yqHv=~%FrGN(ya?ih{lClL0xzTfoL_ykQzg6y_!wAp~QUM0lH^Auew!AzNk7q81 zFRgM@o>gZ)%uazd+4&@74)f{VoVp~4dv`Z^?nG}>2g03{p)@{X4TgIzDN6sN0(Ad_VIIJXB}5T^jc|ch zH{EYW?=@~F!nlhc%JUcAFJew3au)c@e<`IjG0Do7;2X8TTwv`CT@^+VCZ+5mdvD1n ztrHYxTH_PgWRtcNB$x2M>?(I8BVi8E)0K=R(#LO7R*SQxoO^O%e=O-(|FUIu@ zx{-}h{|1A@o9JWExIgc2hl{e-=qXQYU1l}S%YTwMG&g^LuSmxvjy9&bACyp{83pN1 z`Yv?;5A6rT_58fe*Ymapzn(W~O4(Bz(>Y9Qr#aVpP(=CN(WTObS||r;z8pc_8p;b; z0o_)b%u!!MVt*I<4cGKn&_pZmUd*v*a(p}b=w_lla52!{#H5QN_apLGmk`-bq?(%y z!K#Yob@@!8n)OO558Ig(w(}sF3R5IppYyh=-Ys0m%7H-&l*3mmdaTeU?&B=ZZ266o zi6{L}mp2A;dE?dAy>tqD-sb{UdxW+oop+PawxWO?a}l4MW<%an)U}z^lfc7!5DhIs zdhB>+O6%X^F#HPL#c5zNbZPdq{xYbq;2s2bJa`clK_RHeS9CwyPJJP|tgG^JjGl~Rin=m+&D&{%g3gwJ&U-XILwg>Ntb+Ri*uKB0r znicMGx(RneqQS|+pvC3^HU%5n-2??$bZLCr&l>V?a*%6M6?UlHIKhm8c_mRwS5VS% zB$<=9W_PrxEr&~(jgM5QWjVymT(L!|+uo)KHWpgpmI&mGYy_LC+M9>i;bQSV=jP41 zTYkQ?Vs}ig?iIWCM_SOdjSKJMbu9m|V{xx(O+)k$9~+9V+^>u?epAE;d+tg&TBKLV z0%bT{gppN;yArHv#MPo*$@j(yZJrTc@{}0iibyKx*WIQ54Vc4{32X94y7f-6026l# zJFg->c+ODxVBvlxc!JsC#4U9;0t7m@p8Fdv!R7ZipNJ2h`<-Y~bmtC+F^!o7V_Ly3 z#5=Os4Q|7Op}*Ejkgk1QKPYt5<^A-fDVo~SWe@OdtUr*50P9)A-E#$Y2qtsGBIY+b z`p^-_Z!V;<%V;c$JUWldkkL-LNOc;Qyhyrqrs@MU7xKi@hze83gsC$~jcU`?8n+98 zs8|OH(2>#_#|&Mw@#jMr4`wO-aa+RGSZnYBx7mJ2w*Qj%H(>$E8tF*c`hR{;5f@m+ z)a$_oO!D}TJDacK0%vY5jOQhf&GaLqEPWF`us@N;2k3AjGpW)&!jLSOyZ-zqhrTc6 z(jt^ggF^dl&Heh3RodzdMA?94K_Ycq%@TSQhxEAN=Z%#Jmn=bJc83&>B^_^+ww23t zkb%Fk#t~zFDvi^4U(*s7W69UxHAA~{dq`dyj7_*N9n@RmToZw!;yatVr`@W$&f1U7 zK6c2OdA(pOM0UNmMaP?^>?%8#5t4bskkU5pJlz;(9Wi7UIrpZ@F4qgK59H3px}-rH zWFE3y_GGW`b|QVR9rVX$2Seg=mg#%BBumedkF{~}gsqS5;dnz|G_-~Dgv*qotj;)H#-kw*$d&OoytKITItxf;< zI~2&l;v5ymX2X8DtjcCz9tjp*DI&&;_mZ|9v*W0>u)nK+JeHJRjN=go zHRtcFHg9rgbvvFIEC>+NW2397U#EMkAH|Zl(d++wclAg6L*(Jz)is^d0hn#_8SoM; zj}C+Nqf?3Gi-)}3e?i1Jp(4l}Q5D_=jCJazZ7v6*Hal=hkdkvO_TIzAGFgD^pE~#1 z$Ku6i-`u%T*pZ!Sm<&A)Y}Ns7(3I$(84{&&MJ3U0bl9k*h2dejLj|+ z$z&JX<||3%Y}jX8mzFpZxQ=mHEctDXEqd4U;Z;vb__B3UjIQfEVFGs&99k+o9u}kD zSzt(fcPH6u)A|09wp0)3Bp(37M#t#&d-oOsJTocU)6d?D=&Kiq?-Xqa!u7eKh5a+~ z68`L0gLQjzW`{!myxbH^>UAE;P9LDYLv#{SCj|BLW64?c(ly;Fta2-jSkYK!%0$t% z$j*(sf|8!tw#iM2^oQJwzOW~at^6No%)`40eSNegUHgxBqqo|(KwomnTOg*HKvyKM z_8=yNSe~RQCZ+}v(+KsOX#1zm{p~|_O&fE-LWVt7)Kq2E#Jj4v^pmsPhibxx*t$a* z{@J5WzT8GO-Pz0eDw;pEnyoVeY6m<brco@54y5 zgZawOlV}|7lVS#NkFLn|iS`x0&yNFWj>chPVH|4i$d7|D;o=EWcj^DRarm(Of0`ff zL4*^G7}0McR-nuXk|PzY9ZM%jHAn~-{4%N#NF_L z$Ny-}jBoJ{09UThtr?G2gLPEZx>eGB%b`>e{cC;~Nd7UQ8i+u3PrAP=+{a;HWJI5= zX&0ajE0)49tMa@5B8+#59V*F)UkjjjkecUC53=?4lK?A6Y>jzr(5vh@L_Xi`=t_MR5s3wRjDOk%CCMWgu$=y=Iebqd>GXGLgvr`?*lO$ zIz&*qW{5C?DBdwkm9r$ud=|FSP6*{<{9!EOdl51uSYJRi}kBx$*Z+mZ^F9mDKDP2 z@@_4T`Ictn?UJm~y5<_&&#W1;DqjD7+qSs_qICqN0rv5xO%S_46Sajw!bEO+)BTF` zbmzxwVuBa?QR?54;|#=@N1g2(_T60KSh zjvj^jG*NsWXk`$%r2^j4L^jO1{l70QM7e=A#&N0eByAH(@2qtz)? z=RKHYSNO)fd}cKrWq8kzJ3_RX7DziTKPW#NsB{TPyy{t8Q%li0om!nWFW2ms|I72h z-F-)RCbUNLgst`1cy}-AU@&6Y!^k1H z7O8Wov@n-_+`d1`@d`yoWwK^p0?-de5zq$p`os(Edzv!l{^)jvJpBH!>Gcy_|x%->GzOT2<=#MX5U?|NbkVDl)97$V&mH z6qU3X+d6C`+p;m=UNook-7lWLp?>36r{^2^Y1qKac?>ShYVQ%%`~jnqe;>iWyvzN& zRg_Pkg%r7}apR3O4Wjps!8N6=F-{XIyi=}r0My}v_9(<;>vy-w>*ted4=qFf`}Y>s zf&(d;w8`Nh=)ZVxpd`2rHS*eJt7oEenI1>2)gLKg<*uWDeDOHnaamV?P^x5FMV&Jr$2Wc zcE7zGHINepeyB3D`X=l_N`1)H;{ zGF?8=&3_w6kURzk8E)W)<{k6g;N^N$7)w4!J+)nmVSI ztYo~ywHMZHy5KsRkGz50o(M0Rt3MJ-CHV9=5$B0^s6st{sAgHu=rh9T zyTa%S)%7>9q`bEjNB)Ay=Jahf?Eq*#)Ajr#DlzWWaJBX1nv?lAQFCf)Ntf#lB0Aoz zjwRnAi>|x038u_Qj)=7(Lfg(SvR;CJxI^DCX*-#jR>$|z{!`c5>la^iqm7{nV<_CF z3lfkgi1erLnML2TioVqq{gGu(QFK=!dT~*-(4;>J1deajygH1tAzGqHj8!`q(-{$C zgCGO&`9bXP`5X$+mlnGs2_f93ni$U(73!$dJyJrW6@p7Q@C$xT(Plf68P*9;>b}`v z*o+!=CYFqAtC{g6Z?ax$QDN*8udjUQ3NLbwa}DwtVbPl5+9kyQaYs! ziPX9W`6NQ;RXrYJ@Dl67sF6#UQ_dVm%-i(7ZW_hj?Ba z?6a5$2eIf1Vx#6+qw`CBUphI!GKlL0=}dJNwrh8tAWUX*Q2%o5+RoeLQ?w`2y*lDo^Q5QKM$?1N!<3Ge882N38cr&?UZ zn&y)MT_ubhbuVr@nUNM6H;&P!UIxS%glhh|r=O>Rj?YSColoqlI&=kDXb{w~&i)kUMBDgtmefq`mQj#+?`@I2?cwLP3=#|vX4@61doIsp<7%*gNyy77;pKHl#gSU)OF#pf!C{#q9$@0 zLR0tFEC&o_pS=s&*OT4W3rMlRTXEKiLpti;1g7j%pkRHO`xYJpUsolk@2auAF61j7 z9O*pk*JFFiTnsXe>w?S)y9a~TB?jR`WKl5pgDrxN4Tr|2epksI>#Ac@2iw=8QrUWN z!r^*p;i7#O&b~Y9ZOc(RR(kg#1N`fi19-i%1OFqa>fxYx@O$&~AiOS+$dqlFD0QZ1 zzXPri*Ha3pVo(A~-2a6|Nh8QfcV;xlGepWSK|D<$HQFp|R_=&6Q_6#JNn?yU+<7(F z2eprI4igpU(vk1NH_n~!S1DMU5K#1g9&i zkAv-y594DcG3eQ5*}5gXanP3F<`UP=C(_#rRYUnpr8#mf=mmOf>jM7-$;iVbmsz+ET^cT~5Cq{2)|5DxT zxHDBthWZgpeLVl*YmnLA_Q_M=mS%X^UO%d4sQ+?gdX zGn3ZEGh=s84Elga&-GO7+*RS!nHM9u;|r$F9Q*3Fsk1(qn`P>}QLd*_$7rgYS2v2O z0&$jA>d}``mC?L;M)PKU(D4TIW_#IX<8*S=fRdne9y7OfU>wy~YO8oAVA4F%x>G#O zN9sr>Ou&SBqLup_e58)#Oli!JO6@FnXRQ#aLa(4iXq>gZNJTNo0oAg~bRri`ods__ zv*$cP<)=*cTpLyJOu6h;FeK=sDx@b93r{XEyri*mji7SyUY~gZP^-9E^UIc2zsj(< z`>lcDr|!b_7kGMAagi_)=z8=a*}O|QmlqKt$l%G>(>m0=m{bx)=PEvtUx?Z=ks4MK zYL3XK2(nF9cJe8Ep_0XlCAFPj5-k&{ODr>ouJ)U>i=QpOH6E!lI(7}Yn_WSVOCZE6sm7+({QTrb*0OlyI959qJ{ zur-QZyDHqaD4)uf)wosL`l5FU92Gr8VkdAbPeeWWy6L(y>d(m5PjvUogkTC!STYK1 zHe1FB3TBIeSTtKK4Ck6@q*sP}YcEPMSw?N6ia6&s9H2su$TqA6U{qRvN>eshX^kIJ zVMB61!hq_Wqq>Ea9|*iKOH*=9+ev{H z^5fS}aNbxmlgz~!Q@D7qP>it&*rf}mF6BY@(V}n!f~qzhf>l4ful1yOn*%h-`{-k{5%{LMAP33XmUx7NpyvGKX7;3>-fYv#qmF)h z|Add8)BC^J`VHYG1(URRZi_7I3@_?{{^V}X-ze;qD~Nap-ubvHRAKwNCNMF!atXqK z?=jod>=!myBo~R*>#v90KfW*zE7WJ9vmrM7>4>z1JWBuzC59ADuHIsqSnMOO+t6lw zdn%+!=%;Gwsbn%ebJ1o7k)riPCnO?Ihd%E}?@F^W5b1AcZ~`sg+$3uUw`=bEiK?>bAT-2+YqCTz#Ua--b~A$OWiwTblbKg|#XlU<^I zW2|!r8CRPTF)@XM# z^lD$A?Ww~do=%IfJnij*neGiS=5-tT2radgt1_2@x-nT>3&f3x+?jqyjiC6u`}_K; z?nk}`mI03+$s)1F9c|Tu<=P+7N)Zn;XPnrgy2Ms0>A^A#ahxWkkkq)fk~oPh?%qo2 zXo&35=W;U!+yrHBSy^$(sb5-a$wiiC9eN+R;lIp+)8nK+zn8&uOdQwL%r-A zK1u(5kMxI$l=F2Mhq?`|=K?QpY^E2Ibt{T}=AGRiW67Zb_vCzW0p8#4`~m%scPxM( zyu~_Bz>WSfEyDOzv;KPOpW0>wnK>T!8{9ikb?#Rh^ahc!pBiN~!62pwl7ahF!&bb; z?JM0C%tJZzGmi{bk~SlLT|@(JCLA-Mr%H1_QvfOrcpV(N@}jp_!dB<@%Ma}R2_hwKf}%a_g(2f>Za26J+QdJb zA-mgOOj570Ep*$T&u3=9Zokk^?V&e`Bv-jFxcTMzlg9imjnJkW7pozcw!-oKBEcr)U^gU(RiA=|$3BZj{(5Q=Dd<3b(Z&APfl6Lq^eb4mqv-Ca7f8V9= zCHe;Yiql_yOGzcm2%JBMe=c5ZB2Jiv+>66cMXPE3@i`|SefV-ksext0QP=W>F1U<;w|o4=qx*wW`;g{ zKdYBWE5YoohSHWcFc)5g^Lb9%J&zR-c7<>H+RK|%H|JmO#WLd1H*qmTrHDgU5%R41 zz{RHi8m1l3T-eKuN}IWiIrU)nApSUP*hClW*oR;ug|!l^K|%}T2Z=#*T?__jQN|M8 zk}VYEa*31$1I3+{Js9+idzX0-wV3CPx1*E6y2AZp5WV}gDQFt#cZ3umbx$$f#5N2g zVv=W@Y~aP;09K#@P==Xc#b|?H{b}wf&7Ipo4b<6Xk=^cO!r2*&hledwdm=)Go2_=h zO`up#I^OmiIINe1%Gm6dajHl3X-MsOt2DbOfl#jJNdfFut-nnF!>IH7R%dvoka{hD zVgqTZ68%Xxr^(@NW`JzzN&gsvk4-*#Gx(PY6 zc9AI2&n_#ue`o-Nz+-J|)GC*>w#lhh^VeEg8`?E)gN@=QE9SkZRSg!mOPCk`#o^aL zx21iB*n-#khu?*!K*Sx0`J+g^=0#fo(Bql7R*;DbgP95Z>QADYzPT`;ey+&dicq0HY&+y+wuf3d^mufE}UnE4-nAbRKMt? zHni6Hc@^pw-TYO>lJ$zDrD(~Gs3(!R&EjMY@J-c}0og#~ zs;t^px$<1)jWevRCOPG+Dai!&z7B*zgHux{Q-n1YT-E2AipmIW?5$*hr5hc6$ zMhZ{Yx$bD#Ea%jWM(r65{9vlp2306Kb|RVDl13ldF1XX>r{AFF@Aa}f19GV*5V)Et z=vNL?o=C9wB71U6lmzO{Z)#vh8`!g|RRo#vD2SeGdTUI&s6&66tvpy`Cl&KqZ#@`4yDp*3~+V=v8?xQ9}z)FtD1(L3qi6M z4=Z~q$sRNP^b9^T1HSl)erk75t0ybm>p&Ojg^C;I)|D}9+UKgE ze2`#g-b@#*1AL=^FJc}%H}ObD(&n;N!onIiaSuBFgatluXDBcd$`B17ufQH*Ag;hh z4JP0lA3R7g%M>$5Bg!8Zy7PkA5&7He+5M2c-1Y4B)HB3@q61=`E_c50237QVU6l?$ z@jJHDa--Fc%<$hH08P9N8F_H`><0AwZGo;G( zgEm~%6=Y4{GFqvuAvV5p^5)G@P4-wJ$LsFt^-N;!jc$Biwe$y__M}b(02|vN^_hVY z(Q4FF_@RJ$jS6gfD{%F9wHMagy$b!F`HodvBK2{$jsX`!#JZH!LPoVQlWyna{9u@O#JswoD+PUd`>w%L)r25q#7 zrEyqD=Iq(@J8IxO&yuEyA9|Mb{nLbn0q6W!Keg9Q^+TDU;*cn_x+Wcd=O#xV-w$id znwy5q@JT} z9ve6)Nr}Hvn8XHmlYJdxZhBH-u9<+VT$d7-J>#aU?h?{r7Rs_W?Po6+Srh?^Hcf6h z-&KZkGUUM+NQJp;Lj+`Z2bl0%vsin8s1{Uned%f_`FIDZVnC|#o?U%wV#ZMhS~GBd z9;S7XH%dbZ=2@c(UmRvIu^`giP4rkXx3UogkKkV&yP- zTO(6kY!<^l7D84%)(Bz*TZ9!_4;9v*RyoaO7gz<{-}4o@7Jl7ZMti$txBWaEBQ^cO zcW8QyHEomf&}cbU;xi;_#+gKsDnp6t$Mb#_I+>z5|q^Ek>4T(qf9lMK9 zEkcsU(=c2crd1cPkH*awR7)Q~dNy)u$~?5D)ILL6e=VD1q7GH-MT7Fj&XoBsb$Dv9 z%?@U+k$a4}P)Ba>O~H)_>^2H(uN=sCa-ETy+)o7R$ef{38vfw8qT1n5&t1a^DDf3M zk}12Sktb>yjSb=pXEAo!UPyF)n%^w;;HUZJCkg5S$*KCO{R7o#;+^$rG|iC(mgHvi z>_SK=(~Th55C>`YZ1#F3{5YE|3{CcYa1x?vTR4c&bU+>>^CNx=CPFUm_1@WFLxi=W zJlM$;t!3E*!#t`}{=+tK4t(9uiFQBE@uiU38iB^^+GOmNQ}_& zLT=|mk=)*)MgGdaYTmCx#aq^PfFQV)OR(;{&_^~Vy;Tt`jRsp08qT3vdwtEYQNL&N$w*EHzcfsM4l1xv#i>lH6;EA_oI~3 z;;F?-L65}D66C)nCEl^JRAPV77RZYGsZ}1gnhcL1mAUt|(25^r&s`Nk98S#Yu<-}P zwN!;>UUlx)1m2tm_t+!&96J=Wwg(7VMj z5bH1u47tilRW zDty(xVRRz(hUUTV^3x{5LH07D3f5Nn`w{XMy7_$uJdIue0R$=Tv|U-jG0xE)ECcmx zTgRoAx_K-qoa-H(e5LK%++wfruNarM?S>Wecy>Z~-0hfCp|zv++l3Z&C8SEEqMgBa zf-{z|eZy_dLGB*tlKqr)ZwwCZV?vK@ZRc|pqjFSgxo&pPpYC#y(fzF4(s{J?Z%p;A z@!F=`VmtS6=TJ}bmDc;z&_)BX^>%%~MSPHAvwEaH@Hm3&Ul<{)7SSsjQwv<a^bNqu4XIPE|ZRK-L}Dn;+Tjs;=WMK>B4LwWPV4XrF+;;Daww8qqPD22GyZn)bP z)+^lgans%x=2}}-=OGZy!u9)=A97Gh$>rZ!_}PhFzB|}`^>0G3pxy#1rhtlSP+<>( z^Li_)%eE&Me#TR56>cfSweYh_K85?)?hK}el#uX1aw#kQ@^o(_I+C0L!ijcC(?I`qS@X{i*e0v{Q)FfiuL~4mJqg0VxUs6@q7;2DS%CYYZ+syYh-qdc@B?Bmd{`g3D&yS ztyc#f4SG3Cdc^eD>LArrEnw>+ts`r&TbR=zkE}@GnRwK+Giw@~H8h=8A8skqKb6q2 zb{_v!#*(-2!-fKdP^NazZ^A`(G+NP7AX{YtvIcdKJDu)pzHxrG(6vG(dNrd?=IEJ3 z&He}))ZwmErpy#%C*HfT0s@mc%x-oyPMt9MLD$~+u;ztTCLL{fGh?8sI!9LIb}Gyl ztnLj=4zA}~^7pV8ome?)+PO6&g7mi5EiEQI!L14USD~qXZFv6xzEI(IHdOG%d^x`3 z%`$vnwGG1UO@6}e8co=`{Dj>ve}1MuP578TF<)cJN2JR0b9PBAIiHaGY^AXBJ-9fJ znWneJsYOLeu25dlHUJt4LECvbGmc(m5pu8}wfwTFVyGIbrX^}dHVfgMRGk=B{(>-; zoZ|z>THqwGux1|U7|~0;#+orxyK+Av^U|vO`+^BQpEe{!PL{Y_=Q~r5gq|&|Z`EXufJ0R$gYEc)d|^SnIE4Z-7;j3*IlWL>;>3#?VqUEn{d*RQ(TpGq-{*l zaVEwoZ3x;U^Mk)*H2B}m5B^}Y%RuQl5tib|{ntQN828^3a$7)F#;?1ircag;*Q94x zKLa}M!Rjru)qG;V6icq~ak8}em&Jj1T>g>>dLQRQlR3!qxmf2C4NAv?mQks$9u7Sk zgu)xH5Y`R=g>f!*MevJ?F@M3G4T8Z&Ei1PGdueQH33a)#n7l<)+JrS)dfV(gt%TF5 zuhHsz!jS)8)Te=T9$jTzP(m5$}HgE`q{SNRIw|*)3%}qIHo+bLma0 zxy0XI$eu`7Lp`Iry+=NA#Hfi>bERe@oON4!q|VkCW>K~aIWX!E$jeSM5Pbhc{w3MI z#CSLsHIj13P9YS8uPd{+u^=)k5^V6A^3cpjW3nME$PtCrJbP9fkJ~pknZ`g225&ViT3jtKW zfVU(J{$20^FK-P{Wl8oiK%4G%p;57`W2o0{BM2C*DhRL_-P6&)FBu52iH|6*@MX%Q zmC7y0oxOIab-E0(%H4p7XA2d6a;n_Yy*6$frK?0$*#{}jF zMBrDt?Kf#(HtE#(6qJ%l5U$Us_4lLrc zdp=Rx-K5PWyURhD@o)n)h0Sx^a*b1ZY?r50%qfq<+_<$b^30DaFthtJGoIu*LFO_^ z3-c}(b@vq~OgAkqVvge!g=rkF15Ro@x$zV|*l`NdV7pp04?xY-F{K>|wF0AzUkQL9 z^AD8#NGuOeRw(swpr3RD;>%4+zz=46>`dsf`8HrZWloUFDC5Enl})pVZMq}feAjU@ zU{3vGH1QjkaQ3(2M0e_&qVP_u)6dDi2DLwDn=&of2QYXRz0m-1W;^dwY>m^#3no>^G@LcK7M!*3NKzIxi&-Fs!rb z9XB3_5t`-0q^pu^P^EIDW)Uh-iOR;~*cXtfQD+rUl0_@un@+S{MLV)0aqtc6rbiGa zTg^Bosmh%nj&p+BRos)2DOde+JQSpP_~qr+tcSJjjX37;2BQ)p_;}o82M=3YmogxE z&aqsawFshrfc?lv3gv0Ziv&eEqaAv_kw@)e>0oQ&=AZNxhF7l%xf zPo3Y>0_uErM?RzV#J%!5>lhZ$c{?d}_^%pk5e6V+xA2L6y$L#om$ZVI*1c`9TOFHw zMI)JJs=8$h+BJ z!UJ7;dBkbv=bAmsnV&zl$qkNTn$QnTt3eXTLD>+FxX@_2!pLfZxKDO7Vi5@_>6`E% z(5TP%#*xB0v8Q3waw~c}D_lI4=3YcMG3)mLPvQhR_fXh;*ow-fX()Ho|1y6|Q2FNk z?Py~E_xW4NN9yx`%irF-@c(}PcIB&^^0&bT^=ABSfEKO)FYvd|S%gD2^)-Jh;%^IV zr2h}_x6fYwU*&Jr7yKvw_R4Gj6@MGm>nr%%?XUdT{B73m@VBj@?2x}rq$?4B+kU4W z{H^JYJb&A2tmkhd45NAec7PQw=5KGDSIpmj#?kec@weNn#NYgh8v+fx!ixD~*kAvB zje)N*@HGa$#=zGY_!KkR*ZSXI^c z_dfUD!vPLlPzDiE?-d77nVba<FOB#90FbR0@F@94eF2)N;bo(zL=cD=R9sNt4RT z%8YWzuUT1Y*<@s8v-kb1z4tj>kotX}=l8yUywB6&xo7RY*JrK0_S)k)`O^c*mUL zyb5G7FQ?3rTRf+vpfEqrF%y|dflg)J9WyG*c}@jI<@u#0rTI9UijZ$X5eep$7tAQk zN1kQa636}qVT*DK9r;usU}Q#F@$tE(`8nnJS^4wJ^NYxsEMY=KB#U>9E`}9FW#y$6 zyu^9=GYg9H^WeepMLAIKC@+Rz%S+)82S5cyvj`47U_cM^;uzahW4x-t2E?<^#~`?>pe$-oS;5Wu5s^)}*h=T_$gAMC7Zeqg7oeVRhVzPxP!|=2&$RlEU=SnJL$e9x~D~WZbxv>oOgw zL()g4Bs+4-9UWqdOUh&5r@Wk)!h#twCA=y0jP4gxR#2WFRg#l?W6rGnvKTb6+}Sb6 zy09WIU&eCFN^@g!D$0vzmFDEd%*`n+plXPo-66h5pZ?t)uoXUWl;)Qdm!k2brOcu# z%2<#|x}qJ7{XfW?ip|Cp6)tdOq5&0`mN|-NI%XCY!-=T-l43Md^us*foD1{kJMySr z9l7WcrTI5i6rkHsVG5|xupzQ&qw>ph;LGwX2`S4W44a!z&2Yes;^M+V@$p%svI>gI z@=MFJaw_Ia9#2JsE-Wn0m0WJ0FfAPwBkQc9u&|^Y9YOTdvWgONS$-Z%={2fn*0{_u zNohmIWu_;I_;sUGSW53vJsL46P7Kr>n^}@Sry?u2q=cnrq|B+H9y>3mG*8B6ph|hP zqI_mlKWr&Kic_9bl*{{7Qnzjn?x_el4p67f%cm+u2gM+SZkz`<6%>}C zYmJwq93K-J`gw6~ZbfM+ij|c`tu!l(d3z5jkN0Ms@>m3mjCWk?(wPN?g$Ad_`neoo$kh)AcXcZ`TR7%H>opgHCi=9HB+o0-X_9C?1t z%KbxThMXvI7N!HVxqmL3ySgel znkXtmn-nuA+M3xoNlQaaratlVlrp?ylV(dgBBEpv{s%sJHN<5UMnr&Yfl3y9eQoQGYl6!5X=0E{46Y2 zvSt?-=H+AQ&&L7j%N&Ox@1ViQXdAN%<|0!=(LYUSTv@}c{a=g&Tw*q^2F%5%VM0`C z{wy?RxF?TWga4?>m}!JD%H!ff414j8qWpOcMv}^+VzWF=mGNRjqmh|0SmueP6*c0_ z(gNg+<)uucwLwLZn^A@>PKJJATthIUp)=CaXkpD9IG7`9kb`;Sf~)vOs!lAhu}VpB zFfkS(e9OXt42vIJFEKt2umM=_mO4sv3d$hB7xaV7@iBrHG>1gGyTfOf z78e!s#QgHyXver>jA3;30N)kldW*FgUGkN2b8%UMFs>14wQj7<`Q@EsHs*U=s*<(n zYSP?1Z{FF&@|0YFVwIXLN?u^{t55fJlE}5ZVSvR2Y(*(9g{kLJ206xRn3}$md%0om zz;&^=7krswT+KW3alu`Jil13pJO@pYuS6Oy_8qu5Eh;O_!DWM!6;;eBSrAQE4`LAs zJ1Abs?>tUkEqkXM+Wk|Gq0?wN~aaQ_)O zd09p^XXWsfryIXVEJo(iiZTpmS-3dLD#z@fU*4<*LSOF*@(OfiDZic26LFB#odW z(TJxusF>c^I>!EA>5sUjlQ3X(aZ$d^^sLK@P^AmM!3f}2iErFi;<3&}ix3|~Fg))^ zVe`1S^y%v+20iJY=lxf&Ut6vivTEd(;nz(&);8w(q2pSAC|LL!i{qrBY^`HM- zi~jlVs`a1$nDWnmi`YN^!O>@3uJ4Blhu0re+E^3Lpqg$*Lj7Y`UB4n+viGXXVM9%U z_5-ayc-7^P!B?-n>Nui=xor(K|NbDmW@|L!Z71K3z`V(|JqeoO&I7Q z;8n(IK$AhMUWXiL^&3}Rw}RHb4ZWa=?_G5T;%zT>?5b-TXbt|y>~7GE6IWfQK(jx* z>SAa?b*G@m69vH|Qk9^!XRf-w1p-(X&x4)?jeuSL zc&9HBG!b+@D80X0L+N-C>JX)ao(0VY{Ry-R)E{=&6Cc!p_r?mzZhTi!aXg1ylJp)<^S`S){H*{n1DX1Dy`oG-{yst1DwEB104a)FddM#)M=(nK$ zc!_fWUj3>Aoe!Gv7wiPB0jⅆ1%wA&?-=eFL3CkvqAmw$k`ImdYj8-v%?PDD-Qv! z!~OC+isKpK`4soWe+37P#j6bTVbu&!Px#%x1^(q9{9f78}G$!+WNa0w5Rba;*e);1#(V zP=7ok&hU}#>YgrF253FLM^OdJ`nX(&DZM}V_?~77o;HpJjZH`SKr2C4f>wjpfclSc zxz2)CjdZ#E@kw{bC_H2hnmroj0IeU32m3)wu7kc-pm>rm5wxnv zCgef054&8uK`Y-vKA?^dkPrGB`w)DhIH=fkC^uinl==Qj8z0S+K1$IMX#}a?i9Uoj z#O(g5_K@Lz`aI7{HaIlCYtN4SzBb7xhaWrPbg!t%>X*q)l>@g)7-9{PK_%3Cc#={g+xM+0li# zj-Np1gNLTak64`P#p-gT&7+hQ!Vj^BEVd4_J1kSZ?IA<#{zL4Zsd};zi~wu^Wa?kO z>e_Uz%q?y*%mMyvVDm6ezE4oInXb+v7ky3%8@X1!uHbx^tKP;SoPmFcnR^a#d8jiw>pfnPnw*fP?U zYWH7b8EOw%Z5?WNtg;QYN32wb*khM^46*lHtPQi5Sm#;2?fsB6#2x_%phNBcL+zf! zb*k@vc&nu2TUTAr&=d~FKj^Zyf}*BcQc>uU5Uqrs%}1`fo}w~yn^OMHreu4iVx6V@ zjZqM%I;n%cKi~bI>!c&FvB(t&y!WT7|J4lA5+5IP?#Zh`5(hoODm)(!*Y9y~LZBx{6RF={IdO?o# z)Ix3t?6W0dpn-o@QoCGg9b$JZwk6wRZ}7G|kbnk-nxTe7?dUl2>xcJlFx2sb+Qmhr zB_pj9rSSfcY4!)34drWs4?$X9GkEGNy}+9fUMA_{HlbTArM^-unmueu(bEjGSU`{1hU6Y@V8t4Veh zwcC{>yOi4PVvl5diRCWQ(4fL>u1Pxg<3;G`2%DpzLewE_UTjUaJ0{pehVkCb9ANol zP&$J3#d5;&ek$tPayyy>lE|06fH}4AstZG^#Edt6nE_rH)+IfNhlAQ=9@5f~_8i&I zb*B1yq^`EEvaM7{EcURDk6!?FghN^RsS-k<7_ z*JiAT9w9&fPyMM{vHsY6f1>Wm$Nyc>GZKD_LELsT?BVUdfS1>Sw8P*Jc>bzu9PzIg z_O7&~%Qh;;oNs}-(ZHl$i=jTU9P7702*2 z^0xb@(3s4quh56`d*RigO2SzT9L5saS4iS>_s%j*=Y54?opmc-Bc^GW4yuDdq%B8U zDAm^(v%RI5?F}7ov4`afZ>wfQ8}iy4N8UY3{Vx;pv#~Zb<`)Oj>S13V(q;7#M{WR#W45QsKPYs^wALpCr?`ui|@_Cog}q@j{%9HF)2uUKPdA%W zDyD_TeHVE`2lcxn(6^WLJtg$5wD9qQk71S>7&4Oi@HI-m$s97cKQ^PoW#H2hw?MXW zpUpC-*P(jDREw_8{V@%FU<2ePQ=Ws7C-*g`anCfJ{6*Ikr@)^@{P4znnsYaUAAz-b z4)Ha?Z&(jpZ{|(;B?EH=_}RcGqMh45LJNS;+{dVNw=pWwYD15U6}~dtPP$Ix%1W&1 zv$4jfHYjYK%lp}_i0%S!33#8P9g+TP=3BJMr5FRn+-*76>@1n4TgElZNYbtCEzNS! zO?W?vz#8s6zO6EcY{QX;-vFc~;u>QH^^=TdY-(6fB-=AA+x{*C>NCrsD-YKvs}QDx z^lwHQeWUSztDoxhB6Qhsy)y}QB-#>_90=2qf!`kV zqeP@Vjrz8E;DY6E>w&MK_8n-w~ZIY-)-veF}Y zv1X}=_RorD)?&HZ5NkuS%WeJe-4JvA;N#)HSX*jY>9I;(ZCgW4hQ?SLV-jlc&Jn2X zbul_o+Dh0k>7*QE_afa`<0O-R(pIT3A`M2YkM_4D(z0l{&h;&UY)|MT|JqhG=sN>_ z5g+1Olg6=I{(F7Lp~uhAH&*B)Tethr^?~bu-3~|>UAOMWb?X1NO;Nu;gg#k^_rgX~ zzkd`LzD@c)I=Q(v$*`Qbw%e!Zyj}UAZJxq4b1M91TPpNi)4v<~(lC3a#rlM;+1^B# z_k2vshR)+YE*E`H(RSdUbPhw;TypIK9UY(Qbrk8>ZnN#2&0kn(gBa zb%8~v^^fM4Q0cmLYV%_%vSG1MZ9`kE#_M=@*$}Qo1Z$0@mL}JCZ}ICpUJeIjYGB)+ zVJ@*g<@4GIyW?gog>ZR_=EQwB4XS42HRg+yM;}C*8>igMkin;gYey-ptwut}Zm96Y zy%+j2i0x}+(%khPpMT`FV}|8Ov#k^QXm0xx`qFR@hmQ{rLm#hSOop<4(Qg*3RKL*D z=yCw#4vkj>(5xom-jBk^7a{N7CSi)@IxLjXV$2I8k}HQ?tkdPXkJ@H-b8^|%iKxs5 z!Bo9Fl^=2&aBry&ZOFC|a>n?W(QqA(3^6dK>XZ-tyR?3|xAZc?a;`D%B@8nc4=FmP zVtHQy4Z;EFO32khZj6*$Y{|fyWrUuKGu16&`oV^8foYGhEe4L{Z2-N{Gg(i?YJu#X z2FyiZipf8I!0`6YuN$ee|890!C?)(#;3InDMmyopHS!DaRt+wZj!e^OP1g%c)A_hp zwgTa%Z7YTDBk;CRipq1pZ;+%@n=gUv5y4g752 zuh2M8H&h$iD^|er`W1_F%ho3C6}^_$4c|gOVvx(#5*Hb^kA(ahx<)4K&jyCY7Cz}SHWMkGp*(ldb+17ih-kKYTMnY9D^bbd!xTkKL^ZD zCS|!?rBu$pvy1l$xjx5ztkumI6tyAh*MWYrB-iB%qBfuVH)Z8I4c}Q#H(ONn7xEqT zyUdv`7kx9lL0^-5ZH@J9XjY%>leMI8mdo`q*%$sd_VGDJw!ds^+TZm_y7pO)`{d_u z5aXH{7sUNq@^5!s&R+!YA+#r(_BVMo*gni&Yq8F6etiV%#oYTV^6Pbr%QXyfk)IqN z9rD@|_ps9FcCWZL<#oRYc4RZ^@dM>WlBllxSKt9_E21i9V2U9K>a`^Cq7tm60Dl5roTLdc0* zZ14#AnD#%`;oG+N(;D=ckh^9L>fVN{6l+iNG18h`+wd`p=85yrd3LYM^&$0(6VS=a zjq9$VYpf}&ZR5lpY|FD$a9o+&rZK_t3DI?Vht+p!a(+Yb;!G+ujMO|%SzbNVh7Iv2f2`^dHcWvyfMg76X?zF0( z+St=pb+O7$TGf*__Kns0yN#{3skpK6tW901vconARjWAZ%zL5NLeGWTLXU;&Lfb;? zLd!yBA;ZT5VMZ^sWBfev3&YPJzmE7(r4c>Vld&`Sp>+lHfv({y{=L90>U#?NRq4u; ziRIZ&?az8)?r#$IAClox*d?U}&i7Fvzz^*SnyEfxVK=Dw6jrseM8PMbev({5w28@O zY@4Dkv9R|P^$&$zRMg8jyFm!%{_~55(qEM6|FN*ORBo_GKBcI;JlO||`miUfCfQFd z*3UHdo>kr8&E8bipFP_MA)mp?meQxDqNovQkdoo!Xs5AE!XM+aaY z*VMoK*z215y3W4V)R+9&8c+32KX%-cWKVmM)a`mZ#+KP(`Hyz>em{1XpSs$Q><;OU zw7%*K3R|SUs<7{<-ENSzACM^SeH;vd1jycQ*Epg%Um@__NkH z1UJ}JFt<>PeINCrvM*I3>b-&#FY&?&nerd{OlzgZ;186w7G;-3eZs;Hk|GQ)H~?zh zi*i<9SJ=af`klgR6~xhuICJa}^=A(@S-szb9aUO>>p{Kj3lFj!;X~G7oPV&YAA7KS zY?N}^Mo5@c?E#Z0{HLlu3a$r*Pk9KXFIY+ts8rPYV75U~UunhA{=N)m=h2Q^v4^bc zjy7zgt>yM$_NaQ<0`>=<>Y)&JmscNz>%1OP-fhM1($$NB?1)a%AL%v9i6FMqSA98v zo%f}rAAR>J&jheTe(L#_Y`s4vJ?pO`^`C zsE-D)?EwTl8le6Wz}5w-b%E@$K+=3JP(?1k2C6%P*xn!&M!XWF!pI+j)D6LGSuh!N zZ!lqMgVnlFb~2c({ydm6U)PH8%R|&vA?(o*^->#lFocTvXNdYn7+cp`-4(`OX{{az zV@F%74~MaXq3XS1?43~cKcVbc8}*kq?5j599HUYtZM#QN-}0aV@Hr2*UsY>8*zYQZ z>!qu0tq1dZ>{3T{e%6Z9MXUOSiu&q!R%Nf-)W2-(j7|N?#!gaea&0d<^?dRWEmEH69^<#mEG7mKlP6ow$5MuK8Bt3?}?PnEvqbm9tcz)i6!qo6w6)= zng_;}U{Y~4*uSP5t7_dAraT-P^hy_253fYA4Q)CCb9Y!QFz^VDuC!DC6UlCGufE-d9dLAl&c7U#%P&r~E|S%C z2#3_;;pz)r*@K#V7=_!x-ms|~HRT>v{ZnIyRrN=W2I6l#*iw(7jD6vu{^Y^dX~4atsTVYMmV)Z7W9yO(;Ohx(cq`m(ym zORXX$yF5v#){`KN@#zrPR8hDO*PY7IUx}nl!_*a4)&qg178M&#D=g|}3p+%O4u$#M zLIX3El8Zxj50@RK3{80sD~q+Nx?5Aus45r;m-3W_y+P#=`Xg6kwG2f^JlK;q^)nClj!ngbOTSWG zRiOZ?vcv;xfHytVD;OF*RBY^Qp#mJF(gg%kvL5P{vD$+@D15#TS^H<-=*wf+$Nk8t*ZY&~#(4GQboP3@`u23+ zoeAnMQ+*F5sNYQYeIMcJzS{;6_vnC+m2al9uLh}S$FLs;1${M!RVJzzCbLzE>hUq` zwM6x$iN2pDs_Q2BUP@G#Oz_cQdKsu9GzccglFG&?lXdUu9$`)GCJ zXts1T;E#<~4~}Lpj_!ibX;XaV7=pbpMtwDdy*EaMxp!u$Uyo%6XwQ3jCNXcHsD3?( zEuBdAR8Lgjoano6B9-FK$?C?*+S$qK50ka}De9`J+TN*Z&Gp*7)6@s2`#w62+_ZYS z`uTKr?{sp_`s>v#S-y{7PmZ&SevkDu8rQ;Fx|xfk8gw^)XlMTY_n;B|pakG6f{#;Y zqqL+|NIRhf;`x&MW!iPgcOvby(iK~DYpt}I^Q-q6q}2JUNcq?|lCc%`z+L|Ad3$HZ z&e?+=_GbtEpzkfeAl%A$#UH_wboz@<_q4zXId!;2-QkH&ZmIQTS1jstp6p?p`lKfw za#Ig_VcGFtqeQv>J^$;0e?9Q82mbZIzaIG41OIy9Ul082f&U*p5W7HFoj706dP)6p zoB5xP3=i6E!*f=W#!Bjt)L&92X+3|S76t&_A?(i%ytC9RUQQqmGhvn4fb!z*J( zuvvOm@*Q-`;Q!)Lg?)+UjvAT1p-Fyqf0NFh5qg^9V`X~NJThc@ViWyMasRL5P3`@k z@~e^dmOLVmRgxNViQA2kpob;iA#sUK@cvEkze;@FRUw~f+~AOEjoKQD_i*Si;v<{H zCpC%BZW6zxN&N06@pVn&Tbsn6k#Rg~z>f+1;dI8Zb0&^#L9mm^56`v2($+FQ`!^An zV*D9MVo9}ZzT}5VJhln>fmhCOM-un*M7L$Q+~vnEsfZq-qJy6QrvqC&{K%Jl+K;6J zTXy`wEgN|Vg;FisB?}fPxtgTL`$DJq!_ABt=`3CV3_DLrN8ph?ei-@TUsQ1WS$~m^ z_f2_Rn#<~C+}>C4f0OYz5&if4|Ct9e{xo`MW5Z*!MLbc`3`w&kEs?ZR(ke--C9RRP zR?<32&q`V^Da&E}@R!sfX{@A)l4eMnEoq6Qm6BFTS}kdfq_vXPNqSb&dP&&~nZKkC zNn<5Vlr%%qY)MNbt(3G%(rQU-B(0USPSUfI)=SEAW&V;nB#o6cQPK=avn4H&v{KS4 zNvkETk+fFQI!Vt;S}!Soz!N6;=LtI;lEzA!C~1bI*^-t>S}AFjq}7tvNLnjtoup?a zt(TPL%koR=kTh1(L`gFw&6X5ji{eM6q*anuOIjmot)z95o|Uv-QZ`fOFR4S)SVy`-!_<}ay3 z(pX6oCC!jDThbCqD{Noysolk}{l^^&q1Wd4#mB#o6cQPK=avn4H&v{KS4 zNi&+Po4p?EIm&O^yTk7~eJZox@vNS0>VJrAHE+_297cTAatF9z=J`+2e7YZrD^A8PkIeec0{WWTx3y*nf2!q;ys zT(V>8{HuGnef?mE=i{C%_nI^&ptf7n??*g3%l^Cj_aEH97je%61)bc#FVW)V4rTN` zidiR)>@>bd(Xb;c{A{dejdkSN|I6#^VM$5xj)?Iy@L81#NB`)a(Xml+eJgl0?zSGW z(XqWEMKbH|fgMzClqWwY$}6-bk;2l13PHyU5UblN;;JC*tc)8gF1AfD9Fp&laRVPK z;|4xM#$|S_QpWM{F+Zwge4dOOEAfspZmi^uc*%Cb*dgsTR`dY|{t*%AB;&>vg26Yg z6byc?#AiypW0weE!87JfCn#5})zMtgR%Q(I1 zNQeL9f>z3S)ou|tR_rx09xC~DPYAx@&$BWPmz)-H z{#FWtv7d=Jy*)xl?FB)X3&JwK5P^#_p7^B*&|4gI)JX=tg+YhoYe9_>EcP1_r?(~O zsJkSoY+%)YiMVtt%a&KjV?;8mk#Rg*#t-8Pn%+vF!?==OB;&O*;}wF4|JQGNfVjL^U1NeMEqsh^C`5Uc`;^SIam(#Y{({6dWd^tV+fW|5uL`^7Pa& z9oe^vI6VgK%<-pWIj8ZdPj(a& zqMk==nO7BNyKGBfOQ-s?>W$LHQsAm$vyC2@#I7oyw%#-Bpo-`G04wvNYw)XzMY|_( zFsM!I-EI`~qV@r#7E6(7h=kW?_G6iyf@pk&sK?C^4S&ig{5=_^|JySZ3E?$~)XY z(353$Z+#wN*rBo>e$0y^;iLL=WtmZJxJ`j;d|76%mIajl%ZIS9cK|)d8-8(!56kQm za2JLBbqCAr8@QFi&mWIrnf(GN=kSm37{)UD2OgpDp2+SjGd_U&cKFBJlUQa#;13kO zvST32Ol!&Wt?504Wu}XK4+o)08@O@deV#Bur);V5s>bU1rLC@;TPe@Ykw$4+R& z3+^?X5`y8~iQPhI1PT8d9-Y{|br2cxBD^>;sx|ez@O|*y#OT&36#fl0Ffp#R7lm8G z>l1r~Iw|}qJUy{zXn!7tcPI7=9ZO+-Vp}#bKA2im`1}o!9T331I0v68nmAB|_pJzH z69?%4xhli_-+XjMD-7B2EYClz-Bs!Fn4`hv-ySNt1@^R7FXX zl60z>q)Ev-RY=mLRQ(%Hhv~m?nxgdnJmLoV3_Y3S z#_C;&PL9_HEb?bjwu8P*nbP0qMH^n1Ret`yE1QxKH3^(>>&kX)%7Ca0guAW4|00+& zsKr-vTe9l`yaT`G_*tfd$+~sx<6u*g!mdOma&9#^J58=FE3<@=l9pLnb9n%+PX;?G zG}6Jd?v45U@3Y>tvIk|o*&~4c2U2*Tm02NR$#u_q=v!%=+XftZ#!Yaq&00Z!gMx6s z4_dG(qocSP+D~EJjQ#Frz88K)T^PCTKUdd9W#mynKU}PqD$Oqu$;#J@EXU zg`!j5^H#@K^k)ny9{5=vqP*8yN>{XIQ|i3cDZeT#4C!4Ao`ptML->7fHS(8HY#^9v z20+HxyFKRS|QU%Uppv>OU_wJTxgCbuN)jEv#zcvM^Vg@L_qW|+`B{ATEV8Z6f3 zDLmgf?Mbhxy#}-Sy#rWFtRA=xJ5vxK8;Y(EX3DQeaoN*lCDty`TXjxjE#ljQ8#LR^g;K<#-U+dxL>B^-{A`ihn1)X}@(P?9v9ESinhFw;-@Jcj5nyeEQAz*`&dj8V_NHwfVK zKA*XRzwr663HUs8u&BY82LQat$L{T>`F0;0qFgiH+M>~wFAz^!ZhsRk_xj1+CoL-z zyCXEsJL;=HV-U#l?gNQv(gLN9&1f#y=XwvPv<@O|(QDC2o8^5yr4bmos>=yTo9nX@ zQGM;F2`mJGY|8IF<)vezaJD5T;(T$vj+bJr7tVEMSv`D%R?Ne7L&7M{)z}jouIDV$K$AgczkN6cZ2a@_uf#J5N@&DStU%Idj^lu z496!Q=eT-9LE|SnznCvEe~%I9M45 zn{~c@8aNYt-YB=SJfz-Wa=JZsB!C4^@_lWv&Pd4};4CMu6KYv7&oEnSzt@{74>Fz~yW7@mi~+^kK*@`Pg&!5?8_xSrqxrC}`icCGOHZcNDr??!_! z=Fs3P1&J(piMDQB5>qO{TVe1OQpxsObFugguJVt=Hxqen?Ev6$0&y4pY`NJL$AUL% zcg%=ngUHZVOdj+43k@JCINkpfq#@8hloF2FPNffK!6}~3rxO@CL!nd_a4FiKO|>X( zhoem#Q2Za!Ss||XLnc6B#7du{XW|&GB$Q4_?1rC(a$_Oic{*9deDAuMql9_MV2v@@ z%KW~-{A@Y-WGsuwZP8_w$`+B0`33+}y9ME>@|d1X88FrRenie&C^q1FyOM^nYx#lPTWG`ge#tq2yf`gxF{= zCA@cS0`ndd_QW<{F}^>#h1}xxGAY@)&&j5Yh@v%I|Bz!{AyJ}N`{Cu?`o);?mEb*B zdNS|3HS3FLiCvH$g`ed$nRm|%WZ!W!?aJP;#2*_V#-hZ`<(M#>oTYOV}U`qb%x% zAY=mU({9+z`!s>wwVnbdw%|I`62Lj-|%SvAi)yRX!kWxZs9;#_lVy0XzV`%TA-= zM?D+dhwE+8vcV7Aud$)+XM}yf?NE*#=!PBblzN|XPyl+neL>iCwFk!*xM96INbC(< z2-yDL5cW(BIt|LU-VOVefgR2Rzz(Y?Y%VSk3Hyp0_O;$pZ$};g_85(5Q9Wi0d+~~< z+0LdWN$mci(%#$uB~Ssxy|sLt>a-IVCcvh+VaKIN>>e%z zYYo=mL_8Ohu#O>u(zg3?8jUP*yl0{`*bPh zabVwa!%j<=*aKV$*hy0e`{7gR9Q(5ycEE6neUJ+Q`&$-a?>*3tW1V==#_+>U@1WNSyEPgngpF z82fH_!?~k7ncH=6{NwD`nZrJbENbG1H z0CwF4!hX9>lugAqFpN6Bj477eJB|l{{m<8geR3z(me3pLhW%x^P?U~jG`>}}gI z7Xdrn4g29e65EdlfNgVyu(pCgj=jqbd*)t=9moT~X8%Fhzfv`9B67Au2H|$;268kk50(R^W!j8f$M|yv7!>-;UvEOqc zU@xZg(kYIceBQt^Y5Gwt*BupJ zx4kfm)gc%Cj|0){&c4XjJhADu*Lf4>>^g7ulYyD?2&NJ9#HR23p&en;-brL3p;FJf zbYRRAoBo6?iZIs|+1UUC(=P^jnI|@V!mb#?Onl3SWgD1*u~_$-CpP_}XAi=R?-Rk6 z7?@XwLXUZ3)3c7H^1MDvX1ff`&LO~u`rbUDne6`z1Ngu!pKrr(=^A)4f(^{|awdc^GS~}gdcnC7Y$-l* zC&Z~N6cZ-`u`X52N>b}rv|^-rupwn_X-H9~6cZ-`u`a*75=~OI_bObf!jPhy4-KZ8 zw<5UFtjnPrJxS`5zDZo_aYKr9Hb|KTlv4NZ3nHn*n+9^J&kU*IkNdkxQB2_{ZsEYX z`{dO$3)1jm24!HB#Rq+c1#Wshl?{xuc=h*7ODEhs4SkUf?2y!2 z-bP8vPSj7d4PZ&xtMn{R*XS$SQ@l#w#_3x96;AKb&vAOM&K$&Fr?=&Fy&li$efnfh z@7HhPbc22`ryKPxoIaqx#px!!p3}{`-AQ^L)Dt*;NT0-MwO+#MR(&0(+w^^$Zr6`- zx>Nsy(?@kj2a?~d599Pny^zxy{VqvV`liWVNJ*Us<37kH! zPvrCky>mz6zo-x7^d&uu)BSoSrw8;coW86d<@BKbHK(uWzMV+!HN7jRhx8GgzOK*V z^bLItr*G;toF3K>bNZJ4HK%XudS{Y90!q8K-qR z>q7E%dV5aa*9S>DozoBWTP3Xqjmg%P;6ps%qXd7(1HMYIHG)`e6g!nACcB+tCuhfG zI~6;5IVQWKVkh^;WJf4=As(sNg?LxRPU2AmR^UQDr-%1b?diS<#d3Od>RcYhlwC-% ztnY?(J@dK?;YTd9!rQW(UNHn!a2I)&qdi22c}Eo=PC+2uJH&x8VS=h4GQ!7($UO?( zn#dW^dT~6yWJLO^p-&k@w*y%8Ye-=IyvFwhqR(lHuprw)Tqg{MG>CQX zrf82HAHjw?kn2to-%U0JL|3M;&K(se--#yXH%k6ZtaFT_^-E~W`%5`V5LifjqZF32 z&OH?^VOB5BU)Pu)SlhJ|>ztrykH+ImM2xW)i2n@9^Ap_}VSyXQj%J^U zq5hRa4l^x~mm!#hz~5q+F69u#gbdyfVK# z@DAYBy^HvDB+pN7)xA&AUe-JD>VC}t$?C3EG)q)pMiLhdmZ)xSwy5r-iuTEQ2RjLn zFUqM##Lk|3~<5Y@fHqP?5km-9;-^F?*9v1pIA)_8T_Py9{;tnO_VZSe^!BZ&_UmZ)xSwy5si7VYE99oSC*{YfASxM#wvd#^=%#cvn~ zIpK_JK)kwNwrJns?gA-DH9@?($tAM7->_(J2B&im<`C$2pk;N>w>asE3NK0*)%~_b z`#s5zQNn5h3$uB3AGK)9FgQ`~e$vn*l*#J;z@q)oKA6h}^6EZm(H>33=nXWkU}es{ zcT=^ophw=cqq;w`_#E$IRQJ~w?R3y6=GPf;`u3*cK_B20gZal6=US?6Vv6d%WYHpS z6}SC|kOYB+gsASz7VYKOcy4k|W4@^FD;Dj=3n9F^mlEHH>B zs=cbH9Mszc@#-d*$m))=YBvRU;vUQ(PzbV<)%~f(Ne|9=QM#z^IIA`j|DB5x77|#P z&8xe&RXaafOq35AdW14r-TkfFuDZWRANi%xu?O;sO zj6Ql^6w$tVCDGhP`TAwd5$?0#w6XHnk@47%yv`|my zbdG)lr#ITU#+_ z8{e=GU=9nt^r#new)55ej>I#{zZ%$`q9c4vyVlI_TVF1uWpRR<7bJdt_pf*09n@EyLa(`{Du z@iSf6Sn#Gb=7%aLM=|H0K9jdBLJ0+K_shiWN7l;CWxz$&Da78@(1w5Y^S)UCc9%5lfUx;ST z9lqK!e>efCE(X#V9d`L@4}BHFM+bZ5{sGK+%2zw_8Lr#M06E@7kKnuuzFJN)?qT1E z^tlF4^c*4lt*@3a0_*UVVBT*6==nS*09SmqnlS82KLO^ejR3-Q-b)44<*V)Jiw&Ps zfPL2tl)FMeJ?+|Vd=65v#xm9xKQV?6O<@PSmNya`I7@)oXrjzE=B}{2UEA0TkNrFa*o!9gwgwjsvTN<xBnK&Hfc%`$+%}JyGvk@l(@#sS8z{8>NxYthHo&lrDCEPy>p*-VpNs!a z>MUs#!ak_l9v%fta z#VGL8#_qw`k_BeI0bm|?24XtA)lUm5Q-ukOz`K+9+{O0nDap*a)=!)2lfaZsNZ)D5 zijmp=xeK$-20!h!N<53YAIw(`0P`5)*Oob-^wX}#oXa1ZI|JV5#OKO8j$77`IWv7C z?xqgek)FqQj_`>c>Oi~lst3$t!O#v^j`?ZNO%LUAO55IW0)AY9u%M%Ew=m~8pAJ=| zMTi{r(@x}d=13(2@TnwX`i%|BDQC{}C^YtA$Q-2{{6&q`a?J03+M2bALeB>9w>8$| z(QPyeuls8$j${Ek0KjVohzWkMzgD3Kh>Si4?;P<(1D<^wrZcC%w#DWrZDI2@<0k z2aWXC?zqt^K+6EQ#{`K{je{opYhC+|7NFe#JZFN$fVF8L77@k%+CKXz0eT;RGbTuk zv>dd^U+eUnDnM5N(EA#-)OgBc9vr^hU)vGUUcfs6+S`PiGoD0N`D^DAItt_%pr$lJ zO3=gp+8*mr0V)Szi3w`xTO#9K{@PhRULYR=>QNKPOKfiuk45dL{#xq~M`39M=G!Jf z_>JydI4}8Y>2LT5E$0FF-UJDkaM17m+N^P{1<0!(Zc*ST8m^c!ILN<+_RPTH0u&8E zk_l?+Hx75S(E5BHC*acoonyk?{Kk>pTWDvWXd{s5n(UrNNan#o16pWN!EFU-Hvs!h zkelB)a$E~-(82Bkc@ik>gv!DQzg;)p&YX9)(C#b1Gep0D>C@i;@KtMrK0AkQX`y{M zJCZ3K0PE2hDw-Zg?QNk=niV5*9SPJl6Up@nXLIB`Ewo>rO&7?yKvi9fY*=?O4=(e0 z3vK2Lija8>_=AmQ+&s$RKef=N&Fd}T{{i%?YvEF{-cmcVJV79>@pz^ezlI@8*dbaI zSKPLxw*20KVn*))=ui{RT_=!hOK~+jrKNW6h(joz22`Pm$qhtwI9FnWXk(Md}yGU#}AjVM5u15 zoevuzjQRmQkA!PWD{6|v_qEhkchCgf0qEerfxA~17ksa!rs2UdzT%qBeH?))i0i|7YA?Tqa%tPcmTpTXyM zYur0Nhp!0K_8y5A@aq7bVZzNhl6i3C1A*F#A7TXZR-jgz$cCyA$j1Y0%wK+QHJO(5%lI%^``a_7j?f!e;{Q3ClpP=1LY0sZb;IP>7hp~2eXfwF^XNsP6^@0w=N{%b39 z<^^l#%aYh3>`*9iKn*dG4a1lRJ*?uK7p!d#Zpm*yDpP@)X(FjuOn{CDYr|Xh7NAN1 z)|((xoxGUhIxhuly>E01_)~x${TsOZsKf0(1Tm;PuJ-V$TDw@iBc?Z0BZ0=#6aV(dRYXQb}*&RW_`Xn7l)97c2rl z{+uLbAm$;S6`h1(&K7}QzwI8y6;B}a-{ivVF@uJ)PCG2>ul2ZbS^?EGXBXp`e`<{W5N><5qaBabKB2rX^xM8Zou$UNd>)xF z{{C2M?-0IS-LH0plcgpEd;pGT=VPTESn8k_1J+02<{idA{4Dp;)J!3TFsvQjolos6 z^agW*p)H<|#G~##fazyQun@W{l{!p&J2H@wpfVP$iNxMa4L*c!D5j3kp19bKXQPyW zHP>X{LiY<($7r|S7%$jsz*=vzZCNSeu_P~A2&F&$U^q(;KcB`1sMFhsjn0WV?i-!f zVc77aH_#e3I$K@p%+5pMPo#7!mGK2?G5q~EbEA`zupwdWS<)MBiWe!{XEpXLTOg*G zdzCLiTf<)E+Kv8?aK~-1aI*r^9p$O5?Uy zNqIKxRW70}{f52Dltz1%H@?)5y$H4DUge!x;z@$D(D4OjI=0bXWtXu2{2m$oM^rV< zV4HiDTV5Q%LXqqs)*dR2xmW2p9L#znIiZ1N>{Z5}4r3Weo@%fZ@?pbX<&R(caYvMc zf3tyU*sFYGcRS9$7p!dto7a@6E_1Ik`$Q@uqYeYavk)gn|qbY0o+iA zy@i055}daWbFcD-RqlJ0ofADcXd{3Rn_zRV^8P2HIs19AUN+h0Uge6Y1a=I`{~;E& zvxdFO*s=bMth-FC;nE7>IX=FddzH_A+Ml%opnGEw$u`-mTr;E}SD6Cj_-oM(dzG7W zl6iA00%(B=tfOvc?p0p5$b&rqp6H;&6V1omtGuIYDtoCpfU#Hk{-Cz(j0xcO@~7O) zy~-D#NM?VT0ICXwP7QmN$EI6ZyAg&BJVd94y~;<|xbIaKorqFrpidQo>{TXRL9t?iNHI~o zGN_HYE5u&qw$}J&##F%Yio8_#MuUs6S9#$Q_Al@t9b0JvreW}RSA)IEaaw<-Yz5*O z6GfhOM`5pWXRk!290uZ)iK5IT%Gj$69*eg&E&=hIfoif>>3t!A9U-}Z(U|`66OQ0j zYwlG}D(@#ixF^q&Ob~C>=3eE-lo0|n9e~*;i1$Tvuk!kpRsp&LfK?`lyUE_bU7L*Mxj`@OltmjNisyW#$azZSGZu;Qb^5DN}IDCNv+0&ArOW4QT9D?h6g# zAZ0m_s|n3@n|qZ{1+){|c7gY#smAuky^cIAQxm@Gdp*jlIfH ze0PQqbDkOK!1&4hjlIgxcXk*0BfyJp;2V3DZ5|2}{50@JH}H+U%1@WJx=esC})a zwxYEZr4((os47}oYNwW>MO(UAit>BjGiT=9OZxeK{`h^oUK8ie^UQnRvz~M2Ide`i ziH`?1#m1FZS#4`oiLV5lft1OMREPf4)n?5XCtIS!4 z=QDzuGgcTuHkj5b|H9Wrh0PRTnS_fb(pqJm`FQTbgZLIIt(=*)R{4D=462?0`zeC0 zRldACQfigsflVP?_ARAV9;N}QwMy{_zUwW}v{o4w>=xzs@=j+a>4!kQpy~RIXUUMzDm&nz zCSmvpmnyrSFbJ(O#V<pSqh%`@v-^VwNS5FjlXq9tw z3`rjg>g>nqwpRJvL#t8OgSFe2Y1f6+Orce7_}NdM(a(ZZk(fUacpxfL8hC^@cLdTi|cdX+-OZKeD3= zI<(4|zSdOr6{wdBpg&p)XqAr5wPh9$!H;-KmE6vP(paT{R{8C`U`bC0wP6ADM@s>% za!0&1A$JA;l>*Y(w~Jbext zRnjV^tyPZfohV3rJ(!y{TMC8HDu1bMU88;j?2^Lm@hj?vNjkL3qj-m@DAWT`Lp!L# z>T{&UgjV_0)Evny16DQ7w3|X}mACp>71#o-u9_*@gYvSq%0fe;rOj8snxvUtxht(Q zd{mZXz5~|BnyGWQiY4gKDqqSdF6l=>z5F=c)+#GiO_j`tU=`}9D!|KGrBxpOtFdI3 z1uIK4y_{8AWo(l;$!rN$PtAPPS*2AruUk^mM}a!yak{NlUTzT}neT!1iDr5^tF+4G zhKZ7ikE*jDHPg#krB&YC87rBr6Bf`BWOskmS*2ArG#!$j0c!2X>9$rm9BaU$Ydr;4 zFU|CFR%w+}?-iBI(O}KgOfP4ZR=FH|;Dyb#VC~Y(N1at#Wf-LbXa5Z#JAYG08lC{^ zyvOObRynA;#ry!Q&o$G_S*2C3{yTxKg+jr<2kTeO6m>#>nAR%4+AQ~CajBq4KXSRWGP?KzDyy(y>yA6a&$50hlh>B$+SwLU9oJMS|vIclss}vQcwMzQ+ zs8%TiJ8A|Y*wJeegkUFhC>OOu){wrXg<#hVYd=b`6CFYccD-mpt3bhSG!QM=eV9|9 zZ9v3M1mi5-(a;emjuz~MT_xCUnT!`6B7XR0f?1ySZ((O1ueH*V^b9ij93(4~m5#N=lSC$=K%xcB@K~Hz%Np|@366n4@k(A3K zKOcn_Dv=ixybXkOGRk7;75F8Z#+W6Cl@L~!BX&K4agV(6uYzV0UI_~>vYhnZ+PbWJ zvl0~J`jjFX-Ops*Tb7{i=lTW_HB-(vk&@|zSG^_ED&3In6hzsQY2J`T7KP~22u54v z#^dQv6UtXhrWEBRnI?`vPjZnlTQbeXFIzIb4kv8M^!t&e7?~IW&REK)6WP;}>8=h{ z1iT2)DuTNK7H)V+rX!z)>y^lLdwVD4x!O?MJ13zl+KTC-wsA@m!S2r9b*i%gu&Nc) zm*>mZ7!>J_QGj5q2H)P@xhb)hSgFedR8_-TF|E_FB5Q)^76kRC!f3_x*vsjx2crAh zkWx%<4Jgk>A$qccI8~vom>$QIIaE#W0Dn($Y{fLVdA5MR0CZHrqM>AKX~p!?dVfYv z-33WxA^(Ub)r#rf&kM0asM8n(Q3Pp3E2g)rXR-`n)qQcTn40B-g*PpMKczTcim5vy zTHtsuiI;s@V&a2DfWaWh$H%AE2gt%qpkGr z&Y`t)*$7Rit}REkR!rw~2w?P;MXxf6LTkk|wnvsYqtmAsWm||LqZQLB(Glz@uxki% z+7QLkf9&0zA*FNJL$I8^RLfg?zp=Zs`I%TTJ4p2ris@=RhR@T$tf<*qF|Aq&3P2;o zw^EoKIWjR6)8BUCRj<7PzMu&&QcFq#6w?N~a2A*daET8=mQRdQNrhtCaTA`w`4F_P z3ZROrkW?t9m73tCQ5QhF_b*g$H=&qDPQ{yU3-`wTFM_N)-Ko5(P)w&6MRn8!t%)zy z%U39-12eD%tScBVX_jaQ>SNvpp_uj`)qwHIpv~3PSL`Z+V*2EOAngAG_^BpTrjw3W zj!;Z%@&v|@fN@E)sHVMHP)tu}gLM~-z&@&=l$pg+im4G%m+=HJQWWbE#q{cs1U8D& z)d%4TMdG4qwPO0)Yr&G#8-y1$Npxzhm{#u{FG*8Dn6F7)VO+b)V1QzCx=g5Pth`^(***xmX4#tPP;29`{?n18imnlUX| zrmqDoSEtvC>Cd644;r7HfOXU9_3qBJk-)TKdZE4~@kwCL(AKnKTES*RF*S!)6(s%< zm^-v>t(f*(oGh~Br+}T;*0f?e6Au(pCj2h`u)eBn{z@?&R614q9s?}F#+70k^`}eX z)q&Nqaiy5vADb=hw*%J2#+72adSH~qhXEUD<4Q3dT0B>rA^3b?OKe;zrV-<-OZ+2X zJ8WDjrn9b8mG~)O=QXYs)7n*%u>TqH{{2+n*JILFOna`yc5^ovrF~eoVjA5GZ$Zum zSWgk`-JO4U5~K?qfjvXG=ssF84g4G%Oot+Vj7lrdTUs$4J`t0_e1J<7!B$Mizt5#& z`Vp`lgbU+(cjtH-kXkXNw=!|pLC~~fI=ga3k&fS^*#4FQ>48>Er;rt`m`)!aBbYoL z^ejpuu1Cdet9Ey8++dk)0lcHHS*4ghC|X+Dc@g+)zIL=?ntd}sk`{olRFQ1OwB9I} z^#5aEI|&yRAa`$Kcc))WRhj-YunTs2rI_|#nrM|D7$2b1E5-DsLT)*O!~!ctxG1cg z)s$j7WqODt)d8WgCdpY%DW(TDBuP?t5C&+HoYjrz3WZTp_o?cXGnSt zP#bHyzEZHnP$;I3x>1tZ9jpO9Oj|J>Qm(cny#c~}P4cQYkuen0iEltcqH(zitj{!4 z)|=22p_qO%-emkJz-yWyu3Bw-S}{HOH~K#tjOU*bWSIq3c0FMbis^HBLR#cn4y@{$ zDeTG07ECCnUvA8m%vNCadz|TAnIa7o(}eKGGR=7K7y730s!>6QVmfI{hNN!+^~=ZU zwqp8qhvJfX0jyiTOuH_$V!EzgoIIn4J&&^(XWey0@E(TNqM;p$-JX{NAAe{98c z-;fg0W;d|C`vC zE-BouaO;{(DW>0X+;&q>dH`zN3#zob41#W_w(@~u>i4xvCaj7?%?n8A)n-aD4Sd~{ zY5IUa@?X+eSwJxzot`J@^FiHI0Ntzglw#Tgnq`1R zphmr@if_{&Ed><$k;fUPdD&~7LZ0vG+Hq&yiA@5XMp*hX4`$zt}(5cu4y3$`4^y{ zD<8LVqKZojY6M#`y&9R$gu^pIn5XQa2`t65?KW&YKnnf=ux%RGiYYB( z+KQ>;n7<(L3t;}F*;+AeTF&yG4@D%jU{Gre+GifR7T%F^Zu zuzuA{ox8VTD5fuNwwyJGVJ1L$q!w(&G<>0TTagJ?j%Io}s}$44cx{qs|Mp<@(M&IA zm13G#BSJFAf;C?=A9YqKrp~!lBz+U8yC0|9ifIC#_!Tak1?#$IdO52U(_!OFOJ=~! zSoKAa72xHpQcMGu)|bo*V0kq2QD>E6`a{z=Nq-vDfsfN|#q`+-r({k9Yk_8ZIja=Y z=y#GOb2C_9Xr`C5N-@3tr74-`!1_xwA9YqKrVW=zNqXojjKw27)(r#`yE_Md(?~L_ zfz?nmy_{8wY5%va`_!&r4bn_6XO&`FqjZq8ISH(#n)#@+N-^DeC|}Y)2KB4Q>9%6} z?Esf#UIy!?W_mfR6w|&xlwju}Q1C*-@gx+2tN?o&llmiecaBbzuOjDJVC8Bi6^ls{ z6w?zUa-{0m353C#q|I5EBT6xyJt9-mCxiOlztF|VwT7=!Olvg7g)lC;_#PxWq!Uuj zS-nzV9O|w%px-{Q;|5Rt)vf_(z5sbWh8PG(PQr4WWtXRKn3AO};&Auc``-rSK{cwUXc@WIAkF#xA z@z_5l#p>y8B8{;M#kY`uT2`FAq$x`RCdVgXvioRRF}Gf0mRk@($%=oUt;D)(g783eONfcVkibKzo7N@Y)S3 zis}2~#Oxp?I>?F{qZ6?I5zIF}SHCvCB1@&|yoeF$EoR8k=;zWxVxXoS3uvmSyhAlZ^sDy3x=vf_a)Se>s1T8n?7 zdb60l70l?cTEywQOk-&ZkLgypz*kpLXfo}s!7X=+djc}>By(@Ak@$# zsw!Jn%x%NPl*n6y&|Z-&9I|5Ex@Dj!#!jzf#o-O{DQKF21_FD@#+9sC`ErEBrvrP_#+9sic1otS zzX8}58&|U8(d&&QehAoc8&|R-J&-&Z!KEW{wY^QvVL{wj$WFVv9F}rL0(ZBFc{- zjcZvkkUGDX6~)(5TqQx%vf{KmWkfpO0S^;KR~#nNVcpPuvM0u7ny{c1q7=D zsEwHYD<;bH6@XPETn+#wD-PL`W|bdUTRXjy72g=tRL&rSfDKpatyxXUiU-#hlcYC6 zcw3X?tfpke2bJne(kCE%sY!BHQ?ep^1s~t0+P(n7FPbDLEF~+h%FmJ{|H;_;iy#|7 z&a_HaH1VLUu#^Hq9Uqd{*-^`i_r9zs>Fq%6t?BxVXUmFx%i`e+a&Zh;Q+=4WthnKt zTas3RuuYTfv883ji(?x|=67J7)l5-hEh~Qe0-i+r3*bXdko6`+Hpq(GI(wv*=qb3` zLh!Bdq zx>vmk!;lpxgyu=+Ct!W)!?b0^SAwjw+654P*Celc6XqZ*7Vca~nhTnW^B;mNv8*>C zyg^p{^Z6!>rvl8;1aZ}B+tae*`oXx|X#-jhU#jeS!XRYD_pipuTweogx@HP{vi}Jt zWX0`cOH1Y&u)cVl>D>uM8pw*L-?hfkIq?7TP2*Lgf(}{nXmk^qMd&oFHz3F=5YEW4 zCzyC??B`V+l35L`d|#$r7g|cG&~)wrJ5;h(jQw^ z{A`|emGLoHdo)wrKWJHTLyusY@p%w_^C8JQ4J|7cc?~Bn|LJ%gFoGzzrTRctJn}<5 z=~O9TH56`FxHWqzS+QtoYwWZJ_4$9H+o`R5AS?E3Yz^HhNVKwmgkEi?WW}jl@}+OP zz(4*kX{;+SwxHT`g z0<~WO^hZkpS+PT2ap~50@E86|8mknL75#c!6VDb<4;4W7>Vv{9$cq1LPLo;u48C)w zD!E+>`*u-p#QfA_xm6tetOC-AqR62k@`0?le0Ge?rzQA(3rM4Lw8{fnF}<&KvKR;c z(gM=hcfMLyoQeH4V)FeM{BH|LBPJRxEAEQT7WAuNI%cWT+f8oQn3fgq_X`q9+@Mz} zAc@_MT2}n%_417Bu|4Rrhf`OQcRpHH{9`YkX9j6x0VFLew*0!jyvmyo)_TgFuJYts zg!bB&6;}){k54~?_AODx^%GT&Eh~;`QW7tXhlOh({HE-o2`pLh!=zdg51fs)Tm*UD zqh&=}#I$9_LKPbc60ZhkEzQ=l;`kcYWprC$eHAViSiA?mdTH#O`L$(X$AS8$rt5Pg zlZGKHcKa<;GB<#=Q#0+RkW9#m!}H2Z=4r5g)=beJRDqVJhnL0%ak*;5opW&iiy*7o zD|aO;{_tTHX)^IO~zsoWG!gAEh}#LppZ0N8Lax6 z>E*1F6>B6~HQO1i{+j9KtdbRfNyw8nCxEp`Gj;A(%_>=OE1o76qhKqjdmpFUvf}x! z8It(}Sbu1ym$OP%Odk^>nZa|hyo~TzH&C+T)>g@qSp}@dn)#@+N>&`VqL!q02la)= z>Gnp<;jfjF%&A~4)=V#Fm8@8Py47K}fwfOFy=qp;ig!kplQu7cbw@KFbymrW!|tU^ zdXafp1VebN8we&|8XL1dMlx%H)kHJBoK>=7ucMKyDux_?7OWRF)1Jn(tk}y>e)5)2 z18ae1Qn8qp6`y(+??S}@=9@v-uSwdRbvdGB#g9jaNctsE9hk;siG1navyze(=e~i@ zuu(p7NK{5Aq?)swfvkArP^cu;1)+^5+19=2kQLJgG?ny$ppv@ss581RkC6P_Tlleh_0j{d16zt;_=-cc7#qbyeSAz`I3~Z*aZtbydN;qw1oWvWkqr} z>koYFi)6(SSfWzq@y$vxj1uw+fT|H(hR(A}R@_)TO=QEH18SpTB`Y3&I$Xj70KKST zk`<}Jn2Ro28ojfxDr*$6zdk$b*pAmnmsk)JaT44ujt}YA!kCB;4Ep<)h>KsI&z}9r zEb@Utf8Q5z4NDrV$7lz5^mjMn7?Qxu6dcyvc?&NtjDOA%RAnKSv+1?-@%W$jyC4d)p($xykQhR78NL#wH%xPtPUAfKW)Nf1{saQiZ6owNqBEL^BmvckJp%a#+7D< zXWB0w=I~U`!G)&h!1y@kM7pxe<5jv=0Z$>_Nh79RFnjp$lrb@`C`{AQx8rK!h_h2h zeSEJO)rp;`lwn45lU!Y7xb{|OPCuGlJ}48KzXbU{;l%goJzp%2XI(v?T})%ocm}^? zvfiGHxh`gUKA4KncX%fKRhWf(HvCbCxjbu%7H4^${4WZ#1kXp?vse$$);T$BkZ1d` zlB~aH?I3^F(-Ye;g%$B6w|BCqJQG@aSeWNf)yk~2=g+a>EWlG^dIl@(=`pr2i}g&H zP@W}wBA#o=N_j@CjAR2nzh%^5DW2`yOxDWtd3IA4?J>$?Dao_@)o@nSv$#+si}N(@ zm&2T%1;6;QNY9oA@vNQasm_V4o2TmaBJ5euxjUS7_pF-Vl=bpFgU^-)dRpzvVD&st z4ajCCJws1~v20J1)hVo;=hV;`*1_}Y;_}SxY4ddrR@?K#5KOV2hfRvHaL=Tc zF=?!^r$&!B)}9^n>-=N|v_r(F2L4bppD=Fx+{kMAUReR5OVRSLm&3L>&#yHyP|B$G zRLB1Q)zLm5&dEoMl{-=#EnFocpVjf~|G6X^dD1xAq8>WeN#oBzOzp-=<3f#ESV22! zyf?imCcTq}>yI>C!k;vL!AnzF!b#)Ej#5~$^StmNn7!{=n^=$KoiqyLeeQnzlre5q zB03@U)B&rjqVGN&ao%z8ZdEoC|AsJEXW{dtTU`0vKcF}H{LP#A#Ed8N`vCZSqCp6J z4h>9%&q;Hg@cGGKOqOuUaQ*FX*=`>0wro#%&TZK~VHAD0s4_b*U^AKRHyjB(BPU>3 zdp5rt4C4VAg_++eW7FD-mZ{=V@ z($8L<4YS;nc+u!(vBtXDlAOSR9F_k0)0dQIUmRPlVPBdwjA^_pxrkE0`v zK}LKF<+k@Y-n2?Tu@k9xi=emROTFGLgJy~UTLpEF!>rr8L(qrfclV(0#Q!~meiQ%q z3JRzgjSnSH#Cf6J24`@>GVBEDQiHR08=aBlrPyXK+>N~c1Re`&_pCFxz9#i>M){Jq z^ZBQ8*xcnfyPn3TiawqUBJ*N@-c4YAMwo|pB(lN_6YwH<#EwKTqUcv5{;&xSR-_W^ z)6pFMF*I9|YylW+6ic*>acg88+K&_Sb#7j1)07!I!8lX^OJ@3DndvoPD(aUOs5J=@NUo#$*bc#pqkgGC<2Bq3Vv6bQQIF2R&&IXYmGx1w-@ zrM(Xrz582&xxm1BL28Rnj$*qX{pGGh_!Ehn2{ z_{IFUTy(R(H@La<*-{e!6xePXAGf<9>l0N|F=Z$=`eRas3pmZKv?BV(k?I; z+1bu7o5k2s92L+%w)2*4BUzt!QO{e~rLg`_WbG~s7hWFx{2jR`5&3mULumQ~quRuMF=85lCW&@rmOk-;_A5%7x4QL+x8a{YG znin3H&jz%J_rMp{Xy?8V*!g5cAndeu$F0Op<mw@bZte{2_x-%fOQEtpl3i+ zB=k5cl*ZrUxABmL=ggHXoVU*R^kWW|6W0-knUjrh$9leO$$3!nTP zVDi9rm&;1^_8Sn65~(K{82Asq7dIf7hm-#p$FJ{0nOr<#FhD*qcW)y$AcTis0zZs1 z7nP`YvA+|T%Y@fI@mGh1v4ICd9Bc3xCGVWc27Z^=reQV<28iDk*uW!+E%B^qfW#}Z zfkzYFEwfn>`MVMucr0;2mu%*izbmtW#}jAc60nr~U4;!ik+}1vY*t?WuF3|UOuV}^ zn^l*;tFeKn65rjH&2r`MEH?0T;!|&Cv!?QQW_dR7Owx`uN>7 z#j{CkMwMsJSrPE@T+*qYIqXF%0#2S!`gw2;8)-$r%L_?Suja5ftO&UILsHnd95%~} zfS(tWM$OJ)i>wGZdMRnyk{q_uih!q=lb&9g!#2u@3h?yDq=%uo>=P>jp8k}yzgRBY zV@1HzD@j$7a@n_51U$W(lw3ZSow6d}>9wRiy=$>cRs=l#Icbq6mtD6a;OQ?(^YJ~J zJ5~fd{Wa<9F1gHjPqr95y`J0H(I?6`bT~q1kwGddDM@NOG;Aj~gO-{hkS~|Ku zx;}Glp`*CTINC`^Rp;X9D>{mO8%Ia!=*d-0nd=-KHQ$7|t8|q98IJy>qe6Rd^bZ~N z{su=Ts-yS)6F4f2qlk=pv8=?6Xh+3{!K?x*ss(NAxK$7(gm|f zbNo@dJ<*OkEyEeTdZ`tWD$ylwgn+h3JB}@M2&g|nRD%|@DcbSwA_=`gP&pe~7wx$E zq%^pKAo9^lwKCdq(t>snL=|B{Z$~>`X)RNoCP-XWNjDZoJHFdhLAY^?AaQLZp*N!) zH(SY6h1UV1)d$Pq47jmVrb;D9wACg;JCe|EGh-SpTmO|91`AWBq>x z|F5*%NNo_XkG$yeLoBPq(k8dfW?`uf1H<0OI@K<`m@74}a9|_YL3tPO zMop1<{2EY#rS^1Q$E(;V&3?c~i9&f;ZY-sYo^y7{mk*i$3iysMWd?6DDTk#FaHf_C zVJ?(fR02-Vb&HSmZcp8r9RcGrf&WQwTXkDO7*6GZiv1zd?J?I`Om zRE!i+y`MT$uoy)}ZUVT?Bb1r9inG)&oNa0Z2}*y!!ych*SzMH*?sGo%r+hSf2H=H{ zP#)Y(V5tY3oy$}gRyG6P;Y&Gf1Vx}b9CB7_Qbcf$`r^z*O_r8is4y!2OyD;kpyDZp zm7=KBADjyYmSZ6saTkSPJWIus$R+3BpZl@Wh^?v+uZmBi&yS^^kNXk1rPTVlEK9xO zyzpvC!EF!z)0Ac)Sxcf1r>CBcqkfe#!-anNi?dKQQ!vJYHBmED&a^;je{)tDP=~#O zq|1P;B2;)K(c8{4rRs@Xb^`fYp>lpo8P)>^9y-73m@N!k0(?zTSd2jZx$|1BTw%e1 zx(G!O_N9dwdWIP_`y)Nz=C5FyctDlBvgYtJ1s2DiXpm; zg1j1Bl4@~h^5W7g8-#|MG?|RVh)z(KJHyXaVdTWqzQw`Z7nA!&3XT%l0OSn0AWmSp17P^!$mclHS4z$LT&o>L)<4@pA#gQE49f4P_Wum z0*F?do@!i*HA6&a1Y7^NRGH?npbN7Ki7w63TF`q{aT@H2q>rEag4C%mh)#Vp9)sxAM>8MG zx`ORYVJW?WE^WgAqpTSp(#sXbQU(^L;ezdd^j)Ep(PnCLCB_$numZu@NyQ(>1vbH4 zT?vacA0vK`!Z;nV_+f8UWGOF->6lAHQ_ZCLSjD{p_)ks#lB@~&GG%7;-{m6M0Yq>f zvK3W~Aadh^RP6q(vMgnp`9`@!kr~efUX?iCk&SrjTq#2DIX-b>oI$#Y&U0*4iFSR0lcZ2Ta1Q#%GYLJOv1mLeKU(n6M<-uNCa7UaiLp`hH1(J zbNZwJ$w>#EMI0(Try~|mS3xN+RegwV#{KEMkrKeocb=~rZEwlz48Ac%g>0v>JrXk65DNpHhySS3RM!#=imKl&>Yb+q$T# z(yrfvBk1CO;r!$b3k|t!8l2sOn3V?>tJ`lM#OJd7^frM?a7kv_J`+SjGO~} zO_R@&HP!9=wywo4Q=@RdkMRfIY8pxqNxS;eRd^PQBvd^ME8(-cf@E4tFo= z9?XJ9$2Mk_o(*PfEw9)d4`mQZ5jLF*k%YJlA!rKMd6YczyM9np1IF{y%2r^~)l z5y%a0xVvYvACSPWbL#=}bKRntR12(Z^L5EA;0JC*zFL9BqLMTzDx-Kd3qnkY_(-=b z8>Nw7;V%2a4;4j1elef5dMsfsB_x-uN)ZWPr5uCkyUpYuOZy%#xh&>x9$fBo{AlMY zj2~@7BX;1vw@*MZ=57}ljbm1A``UcwZZCcWR2z_&h~4wzM_4IJI1U48gf7B8nQHurM2Bgo(~WG6U=l`vUK9GQUth18e_Klu@=DEA{Y}z{sI|m zg#1&PJ%`wr6vF4&@Fb(b;RrSn&>RiFWy8}9k5Qbh2DDkjZ`<%Jqi;Z2whz!rAD9ah z3wRs&mi1RG`$?g4^s%Hx`ztYbtH@O-nTtvkQ+{R)a~J03_=*{9B{GQHiSi>D>nJw~ zM{+Z_L|sPS*926T;B5qpVdM_wMa3|p_?YHLsxfyo61>t>5c?u>uuhPBBdG>+$8qz6 zOL+8b0^+9GqF6ayr@F}FW)jq{Ss*M{B-vP1*0n~Vn)8T~XeYc}0W+Gr6*uo( ztjMAfvj^-$DusCeK6Af-Nyy!vn?-6DV?P4BrEy9jNujqFHDT^<+)TzAj1i2%8HFHg zK-8mt&%QmDT{OW)W)8-FX%dxNA%Rq$Au7`}pp&qgUo7GnE~9wR$!~B5p@X$>LL88{ z;(!V&4ydDv13z`p3I}B z>IJj;kO7>zx1oh!z*}u`>yu0uBks3&Sy2ZOm6jA|?p@pri1TAVAzAWnB&7qdKi^U# ziJZd4fO(T=&|q;9R~o>%!PT7$1Kt14fbb`x%Lpen%5(F<=DLDD^LUdz_A}gmmI_ufR(nP<`A*K&Z&fG|cKv>M)*xm|O(mh)nns)sM`UPEhUzTyz8O zBA!fL$}qoeh|@`5#0=4?erM*rUp0GevXBoI4n1<8IX;PJmYZ zajNBCKf^r3OELZg7+o~$Ec(6lFSJKXbl_!%S>vJ$GCCNqDV7}J#c0L5`X1-3nvL1t zcy)+-3D4e#a|jpj&vut_nE!OaTe869 zo$#ldN+Mrg;eH##ZMkM=I?R1#s|pgI1m+B4KcsdS+uq&P9p-~OInvsDz&2`Y60hel zm*0w%>AwbcSf@{>wF38YUjEp(@I-8jcQ2 z*KM4(#k*%Z%ucl;#YDgZ_MuZCSmmP&RQFtm`Tc#D#LECn*SOfw?taH%&Yy%w_Hq&5 zj4+y=#H1i!r0xF5VLm&(0pr~PzTiXPG;7Fh@9tBm=KcAMPXuGSV##gq?u!ny+D~<) z3u}OFAY3*<4xW8*|Kc!zTqYm!q%w_urhs_B0I^9b($g7V~1N=~!9BIamw|v4__^(`jaXYtfg3y55H_cLTfEIn7Sr zJEhHiAe{6eaVnO4&fooo)2#9)#@G!o?rIjD9hi7Mg!{PD4C<1I$CD1=<{v?}dg$qe z26I>OH;Z2gXGSXGYbwk>`6G*osmxu+-#oJ?kQpt&=;*^@p@j#evg&w{DQ+MLg8`1# z1aYMyXM;Ryq<^^6b+9%w=7F(JvqZJobrd=+JB(F-#&GUlk`rbo4Z76QgU37zz<@^Lf1a=f1!r5JEJYp778qd!#SFpz8 zL=vs>?EN%>RYXKR1f$rO+DBitY|Z3HccOd4u5R zVNh#4xr@?7HhdYNRT@?rPt~UW65a)9uZBtE5sNV{ntK`*|1*O%T9BQJ$N8e8-CHZM+n>a;!Qj2BxD+9|bi_^@hjOxrTz5ug zwj4RVrHJ%H66pxHUB9BJY4h@(@V{-~-tUmlA{N2S4(7CCk!$0o=lwcLXtEIRb z(iE4DSW5X1@n~oV^T@eItXD)L>kM9R#l3S7MG!vF5%iSw>cpLKE0}FAo5&`DH%D== zXDPdM#8M_yD^A>@iMecHy+pPJyj_ah|E{u2M=WL2)nur zz0^R|Zx8SWDefP-e(8v%+$mCBoldPOs=T3PZaS7 zU14;D{;?KKpH}?uOcr1s3yeR4Ma5#b{h4{9Xf2vNOA%t8a=vP&uzRqjr_q}7#@{D3m0zXHb&Q$K0r{{+-b3Yy> z&I}aaRJ;So0lNa0i$O6@y@&yOkUQ@!&J`Rt@Y0WPPG3OhJ;t5O)>LLRmDB;=+J_^I zNsHk-@ycrRq*YO`3_-1(;m)V=6h9?-8R0dR6!*s1#blo6&JpMR*c?Q^MGzIxf-Z6A zlyiP;1EN1s5GOnA3g0lGAv1sC@$`hXQDquc#Z?|GDyGVXiZz*eP5KbPTr*Jc@s5&j zLU=H9%)qo`e_?^e{4zNC7|w0YZ)Qt{Jlj>KhB)HcU*Z%@OGlhV4$Q=dd(8TN!K*+n zLo*R`&TPb z8&rSrWIK669ofs1MK>V5KLel_f-yp5r#$&5uq+#km@$;yL;)#JvX*7B>4;fO(3=El zPYmG+W#jEhH4r0LY`}&D^O9e1pHuLJs);$ukdOF}2+R>a`w}YsFTY|`l9-sz=oq*C z%p0xz*=68CC-4v)g0YZtOwD_@6f>FAyb@BG8N3XDbOKir7%D)+X%64&&x~BecTyOH zFpNp6Vos&w>xiFRJ|>8n_xy^}jA8h<+=;;$;4~|?@Rtc+N9yThfwJb}l<0ryI@}9$ znq#Zg7o*Ww1Huj;5~p0{P?Pis&`;Hs^i#mDY5FEIOu4h<%|}dT-uFuxfNC;aL*uxNh6080Q(wUDaA?3qDJ5NkwW^t!kHVhjxv%qNN%c4A3=%2mN=aZaf zZ*0UgI)l-hSX6C%2NC>38IXVNHGJgHOK463A81Zjhxt)ChWJO6irr zzD7(S((%IrQloIerhIk+F(nB)OOPIga%Q4zyhq`uF+6{~22rBWf7Tjzlmnk?BwY7B`)q2UVFd2l1N~R?vvzDDnfatUtrZcgV6Jo;|y!snK zc*gIEoaScYdVLip*E2yC+08kEYq3RsHvzl<0)XBl@TViTl07SV$^Yj zylnpIA4lUcoSrzt@_JAJ>MNWc`7`~^pps|tHqu)z&+`F zHJrw{d8>6g%Y>U%5sb<^Q0}$@f)Z8gVLoeu=r#)CbqGJrMK#ysAwfUkHp(F$#xwJI zNb`Z1f$4cEvn@C8pDxRKAmL<$nPi{=r4=H+na4va6PF$wF*|UxOoL*Avkkn@G*>+T zXm-Ip4IZCqg48Dfoh4YfDd8SC&sX*^s_5GUp@0IG502>JoZWmsK5A59o_%u0Z2EBL>v zJWAC&qo%=ncW9r^A}jt`5p%@*&cF+>5;18CO7(%$>9r8{5(D`5Wl|5TMM^3Pe= zfod*t1h4)9dDB4*G8YFI;k-ClgoAWyQ)^m$U)GtdvMmntyI-*)f>hTKZj$Tas4X5% zRVHWUHI5LfEIARs=Ln&yk*Pm}fRkx31DTgWsD>a@(~)ID)YV!?2vy(zt}5*N5f-^W z7Eddfv3|jWF2Yf9-^5ZaB@~(W?h_+!UiH{ARsK+*`6X$5-6tW?QQEbY|?(paa}CsO93 zi`vNSKO3RaE||f0a4}Jayho+2T7^rBzfAM($ilc$LfVOxK8x~}*iF-1at6y8s^0gghdIU%^_v`I8w!Rp%lwrq6wp=YeT{kQ>9E0Ot-OAE;Co zYy_Bh*Cenf0Y|~rI0T~u*>ku2Efv#mfOFArg;@neXA?xu@~(sr7Z*7{njuG!Ux51z z-Xdkz3~&YoS7cql?BmOJe~VZ0m<n5c)SKN@p^EmB78HIJ0DcqAPbjjg~wwbQFLkGXh$i+Yi90@ztQ zy;}2VI>Rm3JnjH{V5e7W9);3l<(fw^R8%}wmS}r%itT{<5OaI_VCyQG28Xz{q`vi^ z*E~K%M?1pj_DHwZJl4W?`rw-|sn-?cVPqonnzhz#Sc42dwO6lYR1SD#po^q zqp<2P==)<{aC*-Nqn6fT{Cm(YlBLPT&)!8$G)(ff z1nE&I*E}d2?@_psJn7I0`!ZTj|Dx1! z^a2QVo`M%ZT&O&~08;7(ey9bIXMV#EwE)tSey9bIvuKY0&^g5|95va0=%wP$jm~?_ zRTI?~neu%-xd2iEE~M|g4L~n|tog!^Re+n-5RACps06(L^1jo_@)6xuLA(s%_5#Q^ zu;jG>GJmSq0!YtJwOCIioPscm45U+9y#Vqman%AyuYq+1=OgfTXs%uW>9in@Wg+!R zK<5Y+Zt4Y){qy4(RrDPM2MQ=)uX}iE9=!m% z29vGH>_t&IY60W}d@QUtqF+!Dzv@*vSOAfgqZdHt^e(~XgS$fWZ-@#}3m|dme|iCA z{Kc|@_!Y>9D8Ws^RSO_d#1)l?1&}*mrU}k%@cz+Uy#UhTj32v))J0J`H-apNUI1x( z+K*AktPH4*g8!?^qg1``&rN2%FMd*tMb=q{FR_>tjNnhn7xSNddYhc<-?ew>;{}OzJZ%Z$<>=wTD^#} zjwU6&h|(dZo*>-^(S%V!=J!MjwTRM%Qn(1!izvfvBUnT^yT>1=F2-^YnrkC;WLZ*+ zDB~!ZUPQUlwmy3vq!BvF-qAR5{qO3NMGiUwInX@d2=4bMEt%k;Gcp$kWz@k%R7Ltc z{P`bxVqev>$b=YDQn7|n25DZRN0D29h%R~=*AJHEZWCD6* z4WmL6yg&)qX^q=!7`_)T|DXCQb5WH>_G*j=(`gOk5*0?RVSK(RN33CVMPy$DBZNE- zrxJy$H4N&-E&~5|?e9IFY9Ob`W#@{(TU<8v!_`-rO&GDvUc-3zC2Sgv01$&Iro zy@oNXe>BSl*31{zYZzx>yWlm9MRoGn4?ssDqWiky zQJlz%)8RL*VN}1PN`o~Fx>K;%FoNmIK(Ar!A5sbzLKu7qMiD?<;%W_JMqX8x0IZb4 z3%(hlOT@G@t7($`GPn=8`c8%GDbMFIl8lK_as>Vca+%o|4et?nyE%sE8~|@ou3yqa zZ+Zm!y!IH&SI2aE+4eUqr%^0vSNSZc{qez1miO0D3=d9I?dd}c{fktMGVBNJ{z>fL04vgYF9K)HPIoYTyE zhsouGvmpHGL*kUH_wvC&r}_7;8j>Dz7pr~<(&jBP?6rK*&RK35s>yH>MJ^w_gr^WH zp*Rf?^Q2CApAu3&*7Cv2PIF96Jox+^7_a!UC=YA-V1&~gw?2v)Gr(9#EUGqc>;pkA zAB=OF3G2$rM<%v_uw9w5@Hg-}t{q%HGI1Q(_sWR1d@$K*o=sHMW>VnPTv z4ZH(;D(db$h(D4eOQ|Bov`+ckLlkcm1RS}Y<4tCCRvdF5BcaF^pR;1QbjVQU0F*~C z0?3jcldFbS#m1Rf)))|v5Re{|{bei_Si#}R)^VFisZPCna<2|0gbSsaqL z5%v#bB@v8N;9^*D`pK&jT`wk3=hlk}ap*_#sUJK*ad{qI1>71bx{wj_44cC!C$*T+ z=Y@DtU&b&HMiPl~vKJF37^Ov}8S{ZHQAX_b18Tj3*AE(@zV!ORvk&n@tsg97++IIe zNk7#3!Ea#&` za|nWw_BG0_*AMPZk7kn*Jy$`zBH{M>!E{*iT0iJ<(QEx+Q{N(NB@!M)I6?-pD6L*U zXiQwSelT;ELvZeb$Kaym>h*(lgR@v|q>ctuoM7RmUO%`qD2q`=R|8Z}!Ctq(a&SHN z00o9HrHcIMou+dAU=dousvNa`5b&@un+lV&$!teaIcoi2`^WigHKI2wi1+lW9IPM6 z%F*iwP1n_Ar@+0e`TazNsPzLnUFh|LZ<}*L#9Jm9uJZ60zTm3$gMH1s)(?Kqa0(7y zn8C2}ZgKVc!N`k|Yy?t238(|Xq8NJpVD=A@j5_8EfJQ3#zp6Y+)w@ImJf891pga~? zB%dBESm6vzbEt}RYE%0G3aT^x07a`&3CxM)DF{^%jMk_v-kz#VKR`j1r5~W6s?iTn z^sHT-)kCUY2t##hIv%P?C~S0Z{}6i7JQ!Mly(i2;Vu?Q@4ATd6DmV)fiYrUi=N{;6aGMVl`IswuGcg1gQKJ>%hSQT^vX8YjxnHVu^MNB0<<4_zZs8RLAI zR3uVA&woH^erJjLvc$u5n;GW=hmoJ)L$tcPdEWmKatgF7QWT}dG|Pv%TYR1kA=9iy zDR5Vtfoi;lV1LfS$~3Fk6DibI#H;~i9nz4QW?dbgOMQx%eT2Rx)V2WjkAaXI3|8w-bC^F(E~^yqEQ+RIwrQS#wp^sIgbZ2+0_j?CfdK;RKE(f7|7eAD>GQ- z=kzylHzz*s;Dcjg45kP1EWO7u(fgZ%%K8anzfwOp^A^X%%jqzyyMuT?UisCSRoE4P zW1{x)TFQzASVRO$P(U0LG@YNII3|9%m`9a^Oj|@gYP4L(#LxBPRU+U)fJP9m$@IiA z@%3{>)ys&UYc#F;#KkeO9{IVBi4C=5)K*CD36&Pd#0$kUsuRF|4aLPV(G+QicTBV$ zRzbA}ngbRsh~O9})G=}Lm8>ceSYyGd;jqO_ziudpbuVA^)XySS+qbdNrO=_Osuv^& z8EL00>RlLm4ER4}TRuOwq~pQ34LR9`L3-kPFmiHkA|8yI{bJQ85Z=dx^8&IQ+;}kF z>Qhnu1ngIXhi5>Xq!zDP2ki|HMsYtLZ01E6*MkxAT()M%Im37jgyV;H$BoeSAwHS| zV0!?a5$r-=p};bXE1oTgw@7oS;ef0%8ZyIp{LDmE2{Eq{s!m9Ts^h_6Hr}Dif(Ej$ zL8a-z=n{pJ1UEnsxz~e1eeq!Y8fc=vrC4f9r9`mnGDq=XZ0uVTLpc|Kd<0t%6c0w- zuJM>PBfho4oaI_p?B6{Y@8&MxD2{~oM5e&3oujtIgK=|y8OJ5D2mNo%!5&O za`2^(uYmm^;*T)lP!GnFFq7@4h^6{ioKiHK)r|t;!PvYey|V#gI0T+GNx2@3wTZQz4G?pHEi{R^9*p&q zBXkF{Hv-%4>KhNn`ad%1USuB!_JgahH#oq9u{K{y$AfVb*j-m&cR4-%55$XPdBWfMDOczX#((bhbk{S}Rh;YY)a2q@D0r7MM81gApel zjErfksDEI)#Sna(1xfcc3@Q1$Lsa?MsDXo5aV`QMt@Wb$BuJ6cJ0@Rjov5sH5WB*Z zo??^8n4Afx&(nyBhMqM@LuO3&Yf(*&KukQLv4mtyIvxyW;~kS9vn0>^;SZ+=BU3um zkrh|*1Uw#$sxdx{uhB?$9D+TONs9;Lz$omI#DSuC^b!Fc%$mLKTBI0&U* zs5lk49#$4Ft=A|n9t?JF@nGoTV75bajEKQD$F!Epi=azp&DS{@4@T>0g|)rxDiEke zA?Bpy6_5`K6x2quI|1uv5^=Xfup7yC2zI6L?u=|Op13p0rN@PFXAH=I3$qw*gUNA<>d=9y#Dm6xJHs)K zxHHytYM@p?cAdyyqrCGn%@`{#1^S=3Gd}q_K?{Bh;Zsz&rDcsf;||_Im+cUL>@2Ni zQlj`#2#&0{Gd5I+RFN=V0YF6w>uQKQV{Z9K#g17AP;&!6knyOM)c?11Drv-`iYkyN zGY*rZJL57^aoz`kJ7a5YpB~X05vd2ea~I%24~{#7cgF|n?qS^-1Do=dZtTIxmI))s z?7=vN(c!;~9Ayti?NoWy6691Qqov9r*eh^ht!CcN4kzrv;3)SKlsyzg(ExHZv4of2Z7fi|cVP;cJ{4lp# zW>YsH6oJa%|r1skdI7MgjDncg>%k_>+(eKM%1a;s${) z_RTFU6Wk!10_eEL4KnOdYV8Kej>uR9`#9S~+#q`PZ5@y3&nV4hmgp=?{0D0!dob9G{RAIg z`+LXJcH|UzP7e73X%(vIkh8V$NSdaNy{c=|~{AcI}qpsS zDnIARf%UEMDOVp{9W^j@h-c|Nu8t2H;0x0Gp!*H=r!jAFb(AO!*Zyyazh*G&RaXG6 zj&^^*E0sDM@<7na)?5mRtAnQV6BJj+;v*H*BFNN0WIdzhx;pUPJoOgf#{oS__+6$a zu8uo6%wq(iCm2m@4RLXGw8m4&b#*-DPf%||a&4%zxH=|OuC4X}`z90@SI0+4JG`so z^@C;AHlQhC(F_Pq-9ud+>p!WY3IZ!HI5iwvsXZ9F`}!Bdwt+&&DmfmEZ4ed@2I-0G z!8pk2hVVc8P(v{`Xbe1*L&4agCFsA{;ax+grqeZBKk)aED+gP4esDO)HapM z4yZ7K%@i#8Q3apqP(TfS%y;bNXjK(>Z6l)!X3bbt4esK*(X@tYhuCfg(b_m_`un|6 zYH)YonqOnpK!ahL{VB_yId=t=y=PRO92kX+Q3V+_xUaAA__XRRh`&pfajcGm^!L4f zG)C<}>|TRdlQewi=h|xUbH4foa;YBx{hkc&vD;RIXZg;Z&!g@DN`V}mbpNz(UN5W$ zFUXcOx3OZ4BVq(ofxvWZR3g883z6|l2Zr3U20w@8^CkdVQ6v42D5c!963bLlSrF3~ zkd;$I=ALypXGWD5F)tE|CnWbQ$LGUry!R}cCou3OA4&qBPa7PnXm+g-x!31IeewC& zYpSUI5W9wO3&Bocj^guKk^}qTtqeIWl@-A*OHh11M@JXKQYhl<8_Zc;W5xd6=X0x2 zW=F9fv_~)nW^DydAq(;OyihC7@%g+8ffXS_7IRJJ^V#u6q$B<*ux~`XG7}E<`E=mf zlYWYb&u4H^yr^;;F;TfpVQNs3`M5rxs7~eKR)j?9P$}l&`h4E%iXAWwAkmsqG@DhA z0^;*IT{Wk(Yo#v)2AZUTICia+npn!&wek|MnI;j}=hGrZtnNVed%)JZ`o`zee`J*I zMfN^m-?;k5=kvmlXvgPs9@te^-}rpGoXg<&e4=5fOf;5mdp#f6T`aqE$I*X6cCD~) z-QVZ)4m#N(tH|M8!CfowBkhDz@i1|S&&S=hl5R~NysMl;&4%C`EXWZWQa=cpzt`u( zMh#5-3*Ud{n5-O&k#KE%$e4^7fcGv>LF{Ly)PyO?m>eI8gX$4uLC+?A{q@d&T*$t)dsQ3(UJ*+HVT0f-adgO6O#JK9@Qb(Dt%RLZCc_n3L=C$zH9LHk#c6*drzp*UiJ$3-9K+ z3t@5d&C_f`(9ILRcWTndR?=>s@ZHq2Yo$6Y6_qS|Mm09wTZ~a8$7&lL)QaM!@ zME5q3r98%go5wMZxOr+#3#jRkohR}swL#3T6+SM+&GXv3aa!YIS;w3A}em5X%`BqjL?ok?J^=b)etvNEqqCb9kVE)@&; zsZ?A&6=**RLy{Xvjv$pVcULmPY2$+43)w*OZyh?$x0v^XjSBxLm zaZ!GJta=`!*^bJO`Xksca%_v^Vkw8SI4(N8UtJ4Tf?ypZY)#V|#&Mx<0Dhw4xOm1* z1dfYGTBT6!0FObK<|cxF!IX^SVl>snaZ$2=GxY(4wunj`#>hi7CAxm;W-uT&kQ6Hz z+&A(DGS?eOo-}{r@F#(7>TV#R&13_~@n`C(da!>VgrNxbtFV?emnM@9Bs7+6Afahw z14-YKv1&ZDRw8T?ZT<@;p>2f|IB3X+*sJ73K={ND7Kc&bIk^U-i05Q&v0~bD(i)K+ z5bQ;)p?FUINmEljjp*To-l7FUH;}YGQ(R30Hb-#Ra}s)=|BuFr^0Q3?(>vxxW#Bnk zTnLph8%TQRiO`SMI7Ajju-|2g*XTAg8%Q{e`~)AO)jbTV#ZSp)2R6T zp);PT+wey4K@{j1OMH!ml^M?w+*p1^%vC_vEe)9&&$|;FsSAinizF?@TX?b=53}*k zc$OoLCDx2GiOwdGqtHv{IjPnT8H z5RKPp&~s9CVQF;^=n5dKUEQGPq+YQiijRfQfb0`pnJ_9pXPbenW%;@xJSQXYG>&KK zJ)V;lSU|CSg)J40U?(td@tpj+7AJ7zLwth4tg@~EJSVN+OT_vA0Go?IWeSMrgr@To z6wisTVw9=@nGuK_YqVU?$xAk0OgkO#45-mX(Kp(Z^s#T4Gfj&)E)Pj%_?cRoMmC-92DF76~G08oe0WV7KH-X)sUi?rE()^%%L5u z*gdrVp7T3jx>(mytOxCuOo3TjG0eiD^*Fz?*1X_Wj@}R$8zN*eS1+x{`JEs9+|;4< zbAhcC@or4mL+f#VXGxgJ_ESX8viNIy1U{mPnA4*86cw3|dzQu8&LCnp5#=v#imauW zN08Rz{LZwQaY|z>B#KaqX0x8AfShG9<=2J|t*--tMkc8sj`KTv;bzK{ZS5z3C7DEm zv>xYo*6)SQBuLR74{WllZ_cu~{C>3VMfOr)D_wm}+i`y9j-KTmTE7$67p}g>aeim) zTd@wUKLhM{(bo^F0q5{_PVUIme?n+I`__GEJsft=qaPizDqM->?H}iN4o5>ItSko; z%UKo|7AtiQb$cAplWeCuG@axqL>{TJ83P~JVc*ANEusYaer)NGQRttGV~$opY&BCF z#FS(d&RCYHMj+-3K-O3dnNc`;Tq89RF=q))B_yNJA@a<|I||1@Fi_?lIMPYvKP`z` zv8pPbfQQJBS(pzasxnGi1HpcUNehvmJHHBs0pbT6EZm5)SZ0^VADvghA@VOnX(<(7 z2d)Q_#Y<}e#f8YTI}4H51HcgZNl}%ReFz#qFcFr_TB>t0MESUKk>OkQau?ALWH1^25;r_vw$C$g`C+ zkq_T7HIaAD?@Wd2ByGE!MkOtOG?Pktv{E*eR68OLulzp8HU;r&UMe_Q3+?}$P;NxG z4?X(u-!cv%UMzuUnJbVhk_Qukcug8k_WNuLAzpo9 zDwlZGy;ENct%2ZHBW!KZ8isi32J;gY;#J8_1jK8~+WP8iz<(k5P%@n(POCx}b5b(I zs~**aczu|N^?wMJ6P3o*F+{^O-G5|CENG_Vf+mR9vNA!w6+q^ic%4My^`ANTlfX80 zi5G1q#OvVaN^OVMS%gb0`W9Hr`h+GE;zeT#@uF#jcvX2nxB3TK`A~o|2#z-Y1(Ptu zi{^WHrjR9T+?>9=&Yfi-UVY1=pk~Q>%469z@%j=)JII3WVGV_NB@U^penIpVLi=ff zP~!Fefzrx{^fDqixJ$eq$~aMewrSwk$#mX=c;&#OLx|T5Y;@;7wmB%xQkLitOZ=^F zGef*MjQj*2qSZabYXEWzjBbv-n2q-d6$im3*0ffY9pcrZJol78(r_uNN2iR9c{Ufy5Jabc4j}!?;{{ z+GG9?q@L)?177(#9|-gv&tv*QyslP2A}l>7*XG!O^HcGq@3GLGNc~5cw-B#+H8CrH z1M$lZW_5A}K)e=Q!MOrk0qzljZWIvWMbr5S3h`?FYf<$SWUeCerqObVSH(lM)YE`7 zqNv#r>_JRVh*z_5O;mA2S0u#3S|f-H@j8wCT;lZ>zA^kLBzuHP3-Nlim90hsn;41< z@fwS?!xOJZXO>nIfo_E6PSFk}UgcWURmXw-CO9=b#7lQye_I$KQ2t0Rr(ch*h$i}? zK}f&uHv-%^AArILc6)q8*X`H!a-~w85mN_{^@N5@zy9M@xI7W_IH6|l-L&;FTA6iWvD%B1&TT50HEL5A}urG}=%|EkjOw5e^{OLztuRpA?5nkpBRevbKeh?qM3ld> zDYBMg9xnemnN$?~2NFdnMYCDcC?Nc&#jvbF{sVzVCMlQyyf8K*$bWz(nM7Rv;~Sns zcOZK_u*t5z;Xez$%&mKoy%g9=SKsiTs8Ml2{sZg_SKsiT8EN59BL6xA?03=E&x~Ve zB7L*ce?s^V`_{kkAM~R`)_}(w8vcWZNVru8CKmovy-f-g^(RIEpeNZ*xpCp<7{Y%5 z2kuni_gKh(s-W-BY7{aG)9(d2S^=@uOzBlNf;?U3<6!K$i1`ALwMau|6ps8GdvFkQ zme9L|WE495huL^X;TQ-8N;gSFNyvYylK-Hp;6DI7{HNbye4oy$jFQ$su-7nY;Xfaj z#X*0MAbzmH!i^}4Wp??`8~L2?&e$(QX(<&y0ID9}*}z%IJcS}8Tpfs( zlZDKe{>-c{qnLlOUKRgAy=5UY|BUJ?3L04vG-OpH?k;3ngF+TEJFdeW*z8w78C^gX zf?`vI)=Z!db!8#58fDEwW_CQkC^H=4S&@~6%(r(GSB;?kI-muFbuncj^O0@E)gF}o zLqMMx*s~VL=8^sCA>S{cR-oL@hWO!qcfRTFaXK z>IW%n7BaWkHMGnI$ZZi>S;!nRtG*fs?Zbe+Bdn_-3zrJ9*8tr$@Bp&Y!`jdt)jhe7fiA0-8_%hSFSPiLSB4 zw{)8sKFwj|C-@Mp?%~s)Ag92l$AVvQ`xB*&--WvcsG(VRX;%Us-+2JuM6j!H?j$pw z1HEw$5n{FgvTA9_%y>S?nnIOF%uzxO2ql~GFdOfT=NSktv2GTy9X`FNF8Y5cpRV4= z%cnEZnf zD?S#c1DP$lGDTH>&MyOBKFN2K!KZ82htXJi51($@7hcY522Pcl;dU z|1y}>#}xpd{;ME*MRYw-Uj(ge4WNMVX`0SYQ26w?G1b*D$kaw;L!;&L>2qbcH@;)jN>f5GpNvI$u&+wHMftP+a)* zbfg`gPp2tfN6iN6heb0ZICT%@(<5g^tHQu42~G_UpVr;izYMkw+}?mCW6tFFK{$RN zPPO;+>l`1#bjKuqy{saD zY6Pbe6h8e=s{G1sgZQTn=FI+Cv47{&->F89;%sQY$rPBimF+UJ5I()?qx=p*-VA}S zLxe2mnv75HYZmK>{|@Yyh`+*wL;3Vrn9252MELZDw`(gqUqkqR5uC!jPDSS9^6Anu zv3Ii}BszpjF%OqdUw$b;+5I3ff>Jb_wSWS`r&qSC?C|MVAu!h@9y?%2AL?m+e-VBfm>hEI2IR#f*Q`x>xYuD;>ZCA(*I_;h9%B8JA$4KLpVtQ?Wx z@aeL^Dihajub&ywXd->H(tkquH2c>5`Sd|_wnJ8Y{n9}`{W;Q3xYZIS7CwDA9lpc< z7e)Y}^~^c}7jBMTKFwwf?B7+}8HKGIqwl}oJ!BNNT3=0BHz9VHDJ^9q$S53ltcY5P zm>B3;?`z16!mQI%sSSuJM`#Nn8HEm?W;WhYxSu7-w=pM5!k59DqE@V`iYMUV)9K&E z-g;{?lAVrVf5N1NPxnue4cnyt_-PP8ab#K=zmLd^#KKC44#;FN9C?LijW<44=M&7PwD;)O?z)r1^CCj_Kjk?O~~; z*6CBIr1)JqRMLWNHB{0JcoPG|x79?OlK&v{F!^!PUbZuzQRSm>!0c5q(kaY88GL(EY^?Fl7&LNOcf6Y3cVF0n>uDiq{Dt?uDJ zclvqx&$L&3sv26nJOryC*h$R9@Spv7jn#+yc^e=di9VyzApiNhQ>JMSw!hdev ziU$8d{9gvMM!5puKNqGo0{>|Z{)3>Et>-Bq{D-FV6BPb4Yhpe1B4lbKvZ2v(`A_<` z_?`*iZh)R5{3_EE{*(62#%eU8Uo@K5T;jriI^i+p@}D~q0rd_fH-t(H|LHd{mD&sJ zNGLA+XA#m4&wq|@tEAon>W4)$BRF*r<`w(x%==&X|+NY&>=@QYlfAW$zv$YQQu)`26nA8VAxVI5r{ z`izK=N3p%E14n9KhnZ|YMdang-q;~*&q2%zQJhRg=HtGcIQL0R47Wq#t57NC5o8@W zQai^4~jXgV=}PmuD-@`Ak~%; z=^fUw64?8qukX}ypu*YV!O7PD$AMH0V9i6;h7TI6IDH^h({{M|P3i&@hrFDa&EX(8 zQv2wOMOCVsc>f=Ql@RP0TsRzr`3ncZx+)tr@XCw%FcL8HkLiGvncY3dK<&O!9KCux z5}V8fXX~ajUjnbZC==eHK>T)th3nPHj)CeG2p(j897-3c_y*ft9~#3;YccbXnLoRf z%=~pHG&BFzv(jPJvOSjm5p>C{w{^!cGyfhRHFl1HsttjL6k<;9%zr@DEY2}d-GTKo ziTsaaplHDGG$0nn6B zSyronA|z~o4vi_xYK7}pPz_MbW(al>UrwwXna_gXNts)9NA%MMvPu$nm(}(mCC~eT zwG&gh%WAQCvZ^sqT#B%g36!UX;HcC2paaR3_1Nf__2(1t_j#*awByT14B$DjS zWNT^Tm}NEJuqZVa(Nhd$HSib*?*}@@k!7_vYgAG1Lw2*sH`4|&?+124|C434^3UYZ zf*TvSwMWBW2C9TH)m}S|*yE2*Ht+Wwi<0FrR^TF+imW>uSid+TP8W3?YZc zfF3dM0~wE6NvmxT%z;0Z!>8v@@Ki~h$h5K(%3)4qs)5ezpMiAcM5dE@Bh)ICe+wc% zV}|Q+VQs>Xdm zH?%4PuS=OPCHF8lW%9SfBRqHK2O4}KR5 zRSVw(0Uth#V*bIR9%F5V5AWnnN%-*BwGy>ZMkJUEK?_?awT9utL#W{=Dtvgln+W)D z+MNwmNxO9Xoglo}~NO&WOXfI?+JM{lt*=DouH zqf6@LxM7GKhhSe~iLbLpLPyz){RIDS##2k=6gcy27L*orbUbE2@x4ce(9x{Zak@zn z03{LZik$Wc9o?5ZwJUmsN12V6j^0H{mROI>&+2>? zt3+3HztH8ltpmMubT^1}1DI|XGT6(aJ8pv-%2Wd1&M?cs2ilCS29E?K2BnR6dYkjlw~YuK>Ob=r6*rGd-cB1v+F>QK&^$1YKBb0db+DcafhK4+E>yAUP5$u__U=E?9{VQd_#Jwtj+5{I66grx}7j~R> zK>R?1g`2qlFFM-t`zDU!JZQhg6qt1~Iy(Gtyu&YdK;UqQki}e+(b1MKCOG03f!!7H z6--#$HoyLYJLu>Hn9252MCfS#AG70~+$XU08Nn&cYAP}xmyXsdjh(d3AkjHgig~zn zv~o2ZE;|Sk&r*tJv({5U=;%+Qn>ciI4g?mOq=Go;=&Gs(9Xh%Z*mjeM5<1%DR4Ip! z9tZY=t8eJ&yg6wdI(ie>U02`G(Mk);I&?HU3=vCX=!TasJiXC4cvem&VAYB1whz5i ze>KxH;n8ai;m?|_+Ph~F{WQQ_?I=3iA#2d^xj&jzjLTh zAh?GGX^jijHpMqtRnudo)p11qz-Y=iES1IbvQ)Np;K)U^0a@yfp2(1;xQC@S%2@`T zy+{w-{}JrXx4(p-D2k)??B5`8kwVNVh*$DOCbg^Muv9;) zpANyPo_xERGX&ZENdtywsk@*lEcK_SaA8<#Y!WUEOU=&$8`ppAPly7}h zN^yRy{5RN`3ZDW_<*Q~zoj?&1M&-bl^@OE1eyXXufMQ-_C132q{f&mD{yo2#@h9XnE=#Rht*FWm#YPA%n7|zB3QJu{S;JDF$y`Fq41wGzkrkHu z+m1xF6xy=@%_FRf>9W)|SrrYv4$yW3dx)0Pzr(RqY6Uh%@D+1x{)|B@1dU@@YVqZT zRhnKn1Q5Yqsf}Y;YTNN~svx3E8OU1iF%DR&V;o_rXFd+7My7k??t#d#hZQ@wp%leS zE17hM@9tp_E8eoUA~FRv-qjl;$WY@QuVE7OE^-uVd^NFvItHU1MdVKi_DNh=XYk`v zV~%n^L7~Ru#{0BZG!oB-;As6$EuqG{%Gz3{6!0n{qf-`Y+@e7RbqQK+fp?+Ib;&)< z1ApkE@`Y1t-1jZ39Yqkcs6IYOBMC8Nf- z$762_(u+cH@G#W)zpRn+)1-l1y;GwypvD2rZ-g4J*-=g3LS~{g3s|BUmN?intGoJb&r@)iVuxXFf_zX%L|29S}Sf1Iby7)7kWLy_OV+4C73oF#vf3CLbhnOCK ztce;j)cD5t+0|ggJV$6Mp`3g*)YmwL?Q3YOg@|5>kfzwCCW<-gl&xbW@&rHb^lX(z zYBL0OBWNt_$DOF@<<~KgUSfUNu!6&{-+BgpIFw%(9_HoOmk!0M&bUhyfM8Jsdk0Hu z`1St3Q3s&4fixt#SEE6GT_v|t{eX4_($m!q^6SC*vnW0Yp9L~OblV=o`f&3jnbd4V zy&+1PcC#BQKPSY23!7@9L9hy(W*{1bmGaQ;$QuBUK=*s<|HL8*?Y?mq%g0v`|Bu0} z^R57B_x3zB@QL#NcpVTyD_d77Aheqn^%E4@oho-DbrUiT5!u{mxwJda@Qms%;HLmR zO*ry48@u(TVtD5|Rb;hivZ!&0o@O+y^u&dBSHT?7rQN6ZRZ|~8a!aVR(C&nmk?H`j z??Z8+-C2-!c-lQRK9|Y~G~EDH4Z*27n~Xnf0S=NRb1gyIiow2P`;Z! z`@LAa!3XKV2x)kG)?Jxx;&kPh9zO<@FE*9$pCK`d6<-X&H3(Tt#8%Z8(VQ$qcfwIP z!6J>}*AGs;1slbb*&LyjKYtz{D}(a4P&{jtc@fPX7Sz*!^~b344^OSG1GUh3bTvI6 zqhk4#{~h#>;?eb#O7H6#bpc`pS3h4^`TF}t7E4s6a9xwCWBOnoG>Jbi<3~+Np(*s! z(QL|85xG+yDXDzDqnTF399t(a&=)bkp&9_?(UQO{>K0tZ^auL}53j9e0D8l~PSJ*q z$&o?%X69sODq`4tyf^K8HHNvSiYWSL({#!=!Y;b5v?6{Ss%H?=96Fg*k1+e~S=eZ= ztDi>I$3fbN|$kaF+Cd6@8RG&efGnYyxqWt9-ZS*T&ek)LnZ*s|*OLNO+QKcbV z6~QhI$Z~LV=F-?%I93%{dxJZ~(|IfFPv;#!_<2{94DS8E47vCnyF>1%M>pb0_Zoot zOk~_$p=tZEA;Ibdpc}#E1jRkn zp($RU9ESK;4CXLqR_vZV()a9ENO~%xqqqs$yO;vAww}F%ESx>k_yG9nOTnX*e}KTX z5Fv}XCUXzvegR)n#-|AFv_o*~LvYf3j|qpmhnB)jwx1&69@=yXXD-!8Ogm9rOGW16 zx`zs8g`1!+Bu0lyF%Q>0GzVXx_YnKmefCI$uUto;JY@a#N_I6&?^gI5 zX(#l57A6jH4?QEh6~1~orE0EnsvjZv3k%W}7tU~G{$4*28#SIqX5U1msY=_+9tnZxD8!sxkJRRDRkhLVdBEN@iMabF*p0*^#ik7Jk!lCyiASmv zFN{a3D=&;kss}HON9qQe<33$g^HVn2efq54KJ9p8JS4#Z}+bl)oY6SalTGL`bmM;&{Kz@Qk z=F^_6thK&H5l}|-OD=+Zkl*xn(C-*Qm;_Rm4 zm!5I`K*qW+neUCl1mDWPHAQ*{6wXjt{YZx-WiW+{RLgx3liGa+X0{zVfxH>{}NreA`DH$?(HE%$IMIUF`>UTyU4n zhdz4#qrR&AG-=>M;oz5teqv?JI(NM^x%K_D3nF_X*cDjfs;rTadG=yI!H3uWo+H`( zm5@^)?;}lNdOYU(;Mp0!A0yU7=7+8?uB?mzvLV>dvas@)8?(EV8i<%OfUF4`GLN}0 zPSjSz5z~y&6hg@!bIiv3nEM4KSz?{beSP&f(7iwo5Z$ZMAP-*_TUT`e`Wujou5OTr-^iV)o&@Sc5uy4IxI#5x zf4&)>pz#1-7J(BK5W-E<`3VZ)&epV+It`gkh}>bcT*4ij)2Ds|d>qgZgfB8ZA>8FI zq*K=reb;DOe-anM-3^a7mvC1oTtVeUor@#rsyWg^xIcLzLDd4*G!z%Y{WsDMPq-)5 zMpv_Ps?!NwCK;c2fuOoek#9HCeHHxZFMG=Z?Xr)V1P7XwP5@%-TwPo%6vn7pf>9&d;4X zyP-qQ-hsf55Fv}oUM`38bDL!e9yj$Zu-`?z4HNcoIh>#S49sNvDI)K?PFvd)^PKUP z%7@?-rXv-ZkNduBxsEmP{bxut43%OY!HF==&yD*P2cbR=i6lzVY*rTvI9#r1q_5A% z4IC~v9s-k1Qb8Q&=eAiL<8Zm9z*d?>f?N*g=l1%er0ziWPGDcS`sRJtJs;Q5y~sWT z>~~k+yze@9eUQtgfFV-T7`ox*eb?%v;f>jfHIn?miV)XruOC*MaZ0zoOO9)rD%q!y z*|+Y)<#12^1f2~#&rg(BHT3zp8Bwb~QW_p`J}@9z>bPjv_3QA|FI2pJeNER=x=@b|C#_lZaO%DLGpWw-g z=-QcIz}620T{3I7&dIR13@OuVd)bvBP@O`|$z^Z9pUI|;W_JMgxJkriZ|pe2-q?A< zv$p_@C+w~CBwQHw)^9Q{40{{I3&Y-?L37-vb814uCe!RKd@t4PZOAY1w)z{uY=O#; z2k=zEeuCmCLPEw@(2TO5pxWw;sv3$}2f;oQ1tZ9Qf|uhes}6{M!a&yV#NGV_aY)It zpJ2&I&#w0x#ha=@P@IdfhzVSxuIwjxfwE>lL9f!bme~!tFGW`N6AXB|sJagAvw;2} ztcxl82?q5ps%YpGs9ZV(4SSZv**tQXl)LL4#-UbV{EZkG2m1+<&?d%HBl`&w7t~g7LG}ZYKd232_7g;*|H*!W zwzunP!J`m9L4{*l*6b%pNmuSh;g0e3b zSL~QI0W~r30~wE6Nj(l$QAssZRZxK*RU6=j0dK&fDd^0`8!#Om)&Bz06>q?a*K(^j zQ2sTD+{z3W0XSgrJ43FSUm82K1qZpQv~P zCc25h8!)U_MO6@RQ-sGP5&jFNWV``0sV3flRJ~%=SO`rKm41)m!S(R$sREO4HG~1- z4H)rK&gnh6(}*07V4q}(&#*?~4PY?;TI|kyGG{;@qQOH$~KJ%)a6aG!L0J zJv*{2wjTf}h+voF1Wcw)kKC!N3L~Z_AghLk%(SUR&YG$OVmcA3N2qOqcQdF4ha0QD zh#F=R3%g>^B6z1%f1;F2tWkI%W2$&Is^)Zb&CnJ5Pv?3)1=qfbR~J*|R7a7^3Fb6` z1u&k1Aqz4qJ_7y(@;A}x8V!01vIMf?(FVebOtK>cb%UOQzIkgZK59w#J)Cdx65?MsnDw4308ha` zucT1cDu5eBU@Zm2Q$W-C35uuSvo90WM#!8&4~SH z-m%OoH=+v@VqvYl#Klvv5&5~Ef?mJ)RTD_I3zZg6LEHU>R3BhNLvisG96;LPJq6$7 z&8Ch5T?x(gq8;if*j5${2-vrRQ^VnVvNXT>@l;B$fd4SOxViz!2(*!|nP&xDj{$!@ zSSzr+K}pPSusl2ka!$4LTG6B%u1jGPrx)T<7*Q^dY6;;^On3(%%fXFHVf&k9RDWPY z4IY-!1j@w6qZ46Kw(Kl)q6dG4JjGbQT7{$!A$-d${)$HFWKlM@7#C$P1Ne*JKLlk_ zwsi-56DE45r3xWv%%L5{ai6`1$ltRldpoV|C^m<7N2b86lkM4`ktKL=)gTB=3K6oH ztCz^*74mAYS9W;+5@2gZJR+@WSlP2*31+hW6p=;QQCG2$`aNRKiK3s1%*S1nZ8jD= z`|Ve8^gn`AWG%%!+(p@zKV`)Je@K*}6wPL(qkt^R#x|{z4|jq?_+`i z-E?Mi{*lV~(Bo;w&kkbYdUdkR`M-^HT=#ZPD3zjO4kYSq&gZ3-hj|E@W|tB&tvjJ1 z(_h@g3F2L$(T9n!WL6=n%I5r-LhuZ8bN)mKOr;QW3gXzD|Jj&a&gT5LfW2!HamjSZ z=6o73JelqW;|ZDW$qPfK`|!e$>HfShWcuXPAej!|LG*n7!CxVtg(gT^R1vG7*Dpq> zq*0A(s-%nk>f#NuuWF)!L4M}Hj?4}Dxr5C8<53(TKl$fkh`=_h5xEJ$E<6!O!lSwL%@MdT&Xs!lB-KXYq4GJX^wGlC~&AwP!(HdA$>RUCLl$~2bT!we!$ zC}*wljHU-NRtWh?{>twV6h6@USMCb{`8hquLw<6<dwD5c5I0u*#9g58~MBIM`BcX`xCMDHZj3s5rhb7iPc z9S3$saF_gqj_3cVuPQ%H8W>xwC@KT;bDotkGAA1;{PJO3Vn6SLGyaMn&puX;4b z4nSnEjYT`pObo&KHC;h94CoOc-xIy6(ICNToD`!50KEp}ma7{iIQ!qOs>T4#h$3V| zaCGI_to(ei2R>`TMJ5oOT?=3|mfk~fmLJ4kgI3UOPyL(BTL@0eM{qjX(}*8#Fzc=> z0D`l%HQqXS8Q@$Ih)jnRgy7J0eu6@9uJtUX(m-YxBKI0Cm*7-VQ7QxAGk|_4oR#Sb z!AUHSt;vW^jhg7fTDgb|!P$!ZT!Qm?uNs=v7=V{b&suSvDTVHWr3cRQ$wa-f7q~u%8Qs%gyIQFzjjt6nT@w! ze}Uu!pA^ACqg;{v=S`Q$5V?0nlKQeDxp+t^H556`M_7VjCoo4@k&OGj4yJsY0PG-G ziJ+`Ve)l8Z!##rdO9peMe5}|#MCP6q$@iW}?I>njgrSe1v$p1p@vcY?U)025CRHKO zDn!U)u3jR870HC`c^vUR5FI7r)lqCm(jhWfk+c^hLqA1iMRM|F)O8tR){A0YDl#8; zMKaa1I2~*+B)$ukVjk{_uKZA#H>R$~gtie&Y5*&HI11?`m+!7;pp^NXHY z^EpJOJg_Rnwe6jkkFg@zWn>}Uf$T?sb#nF1isY8|ak>}TLx7EP_05W8^Y|c zS6}1!qUYXjjU6Jh0oWGN*UyYr!y_Cb@m+^B$%qX5*8Nu`+rXNKtn=ma@h*%ndNxKw zBrJXxCYBY+jZNySOR!ccKo!_db}^Vv-yWHxP#3t6*o=X5se&7A4&y4m=cJHPnB!Rt zjFAAIV@fUA2r>#&R7k5nMrVBkkkw8@W)zkkmZ%;>%!h=!6OvJg3SvJgv+<6?77z>! zzlBXGBvFf(qE@V`iYMS%ku0`73r1Ab5-?!|yBCv|70HWfF~f{Ue0_t38&MX^?5;?* zndf{z-0ljcAyn*#MD;+jcxjDb9zvqnorOf{0boef;W=R6Z$aY&Cc=_gW2q`6YFB(V zZ7+K-1P)M$Ik_b2#Y=Uy(d<8fT{elhB#IqJRwUV!;YrkT7*9yl&SkhTB+6Qj3qzt( z^1_g)3Z2ju`O0Vb&Z&6;n@q1rhVP}G70GYen6F2dfKv)9lG{;)gso}3Z#efKkXL<; zVje-T>(+x2WJU7Mt=#GlL|-?M)q=RYBDohSc~&H|u8MNsa9%JL=0tYI5K1H1k5E@u zB>#r%QdT5?$`P$)+C#3h$jXXjm)Jzr71~1qjV7#%DJzn@awaModN!ap4eVJ|b4J=d zZ#ZMjQ7aI;EI*8c70KIgiE+$|WZx-O)OjR%mB|Jg(>P{Da?TsMRVrv?LeP+vgLu$5 zSdny$BP)`r^W|3+AzNGI^H5$sZ*!*zz743q4K&sA75%cf7VHJ#epD!=WzCA@4a%B?)4$6Y$*1paWn4GS{AfMDyo5Y?kmY ze!}(X|EMHhPr#?cK#=GOX!s7=z<2__YFa^i0^UYJS0UJ+u}#Diu&_p~`V`Uo3GD-v z%oFfgy<+M#unU5_o`BGY(tlJV<)=vlPnN@L!Sn?D#mX2@!254Iw~)bz9E)Io!xA57 zjl>hcUhF6Me>0w%A*Vpf9W_x}FsAvK^2K+36ur;&1RTv?1b$lp#S!cxob}0Et;@0; zDmP*p0J6$y$jsIDydJ0WBc?l{L_&HQ^h%kUY7nBHHHn2?24xYv^SIk6iS zuJEY@B#x+~8}t%no*%FHe5(kgn&`?jSot|?4je6kw^V)b5@cQtgR=ACa8hH#)RVHCD?(q!+QxH8=OJy2KoUsw}^JAm!Rj-^y&bx9|fm|Lo#I< zv~ce<&N67NUmB~skW7s>(lzregX%HhZwhM#qSEriX&4*}*p}r7hYXGqG5Anb9RQ7( z$hXB|85tatO2w#ChhJO(KmPZz2@b)INB}7AgjfG5 zsO$oWuWm4B+QW+dyF20U-0_ZLPiPNd3e4I%#QbH0^NJhA9RI*n2)r30WHHxd?u09U zR&vBQ1KTU&X;JKC?u3>wlkKO7xDyJF#P>n3AtuEKrZAbQ$b4LPLazpSF*krjkx(h- z;kpx+;~+r01|%9&ie|H7C?M{HoE@7v?t~{H(Ay*x#Nke;I4Za*^bK!Z$b#Y63Sfp9S`Zt8d&1&3DE*?u1AfDh-XL z+g?9-ruWF{ko%qJNon{&A30i^ee3@2geB-?hpaiZ%BpznPWS+6C!E^?6N@{c*Sjgbq&;yRZ zA0hS&Q;K6sGC13HFRltA#)h6%OhaaH9uDMIr4f^hP(?yAI2|tov+)kj4J^sTi#1Ua zco|ZyM`dXN6;Ht9WvG-Y9s_Lvk{yO%S7XxRWjHZEwX&xn{yl?*8)z2G?0Olxov-ig zd;Sbc-&3&;a6P^(URup5E?x%qa`7_gabdg+qfaF&J8B*FKO*RoS#5PrX5X`KK@P`W zr65p&Ld?naGVE-PPu?KLZV9ZNNyPOsus?~Hfqg2xmmwO)6E8!N4Y)8~hMphd!gv|_ z@WOZ*enoTKr_XB6!6wsQhVY%&<7KD;vjv8{Ud-__6hILYs)u+PZkBAQ%A=T75bT9@ zVFd9qeBUlXwM29$16eDHyIzK-pyrp(SNCQSM~< zR;9!VnUL9ruv28k%P=o{Cbb#bCjgx$tcxjLhE-WJDH{4Vf)5qcu!ktKdBX2|rdFU( z@p#9}&<(8+G>-8y^sL-iwMLR1nCwt(9OGqp{A5M-G@^$a$a>ym9C#TVk2 z0c2N*{EONkX5aG%c!-FX;fpKPwBT0|K1_vaTGn_O-l44VGQ4rUmX`S&au%xK$cmSt zZ2Ewj1ML_#LKWPZ9uVORsvMVzz<|RY9&QHRazxwI~Jh=&1Pdm8@&uaBNg#7 zj6Ljpu4WY?^$F9fur0{P`nIe8bjspo7`~&h8VIfVz~83KaLGN)Kqqta4fQ;w#02*n@%QyIut@@uXHvpe@G`XC zV%kaYfs%j?SO$c8|BK#Lj$#@y2P))oH`M*l9&O_*is9c_DybPf|=|8fk z3N#+iL;K)m_yq>k&V^&hT>p8~{ON>02^sMd?iT(ZmBj01xYLpczQD_{dNVq%@iLs7 zUR!$^(jqc5f<1$6B3_1uLkg+Fh%QU$H9*O{4C~kWR6}5G1b4j*568BCnlw=Cqu_pn z$1t}NFGFnYI(ky@9ZK^vOEjM)UcwrQmw~<5Pw@X{Jgq}cfh{j(hv_j_`vs+qpN|m> zQf&4c3`exaU(t~b1%gyymJ)T_Ap##0c%ZeJ3a7*E1Kk7HRHXgwfJ zh<4X#(36m}$x3+~3?FC0Uh3n}{ek);SR`>Uw7gRj>)(i1+k%)i!4-gmA;X0@ zWn}@FM+Bx&KpYITsGp!X7zPyukAO@QM7AbY4Gq;q zM9(mq))M04U?_#Do9ke>-a0|8gXH#5X>l+-cD}JX1nh@UTpSE5kal$epP;(iOdsma4Jy=GF zdOMpl$eQdx$C%MRWRUsGI=jq z5c-;s3^K=C!EC&Pj0FuW%gk3&;jOs!2}%<5RuH+@TS0yCR=iQZh8l#N<{~UYu#YfD z@mB1-Qx7ESLjYR|{zy=~6$6G(5jQJ};cObhpunw-i@m3^$SwZ(AdmymkuDaY<^%i-ci^T zf`O6^YCBB!;!e~mG0MAM-s*R3`9pL{1p9c%dii?0jLL!Nf(EkAgszu2L`t6Z@~htU z@-+T@syq}sA#`H`zlW@sAEm5evgfZ9)G{wZZkosnlfAw$kNOkZ%K^PdSQk^6?AP=2 zsH3RnZa@bN>>*?94&j+BwE|bKaYh24&FbAS4*P+Mr`0?@n=dQY#OizlM`#tJGP zvSmbm3*}|M!^>!2@kI34nNbUNfbio~_(aPZPsAk3iYH=u7A-RYa#KWBJP}1tmR5V9 zy%f+&!nzvbiRgN~w4&`l1+?G557haom2`eC4rk4ovxo|0dyads;fZ)=kH-_y?i1%V zkcx=ZUD)Z{d>T5Q2;LAMs71qiBG$HRf=r*aqfR6JF#UquFYDW$yZHlkkPueSp5mPw zMfh=wAqjo{FFb_)Ec%?gZ6HIjMyf0-W@rB#hd?0Mm03+8^c~opgwQYR?9)Q^AlT9f zTQ#(XA@q7+`H2dlFW@EuLLd8lDfI;4DG2i<5&jFNWC(p}stKW=al5?Q2BFVIW$6Yy zXC9s@RiOW7KMV*${}BwRdHf1wuKx@$e~RNz0^9T(M4P}mh4@1^N>uXg;FMa|*5`w*gg_XI^%%5th`G}bS$Xcf%!#Ku9 zr&LQ2vxv|pLdoVj%*H#{sRY3#uv{)@gFvr+FkdLGZ+P2F>*G@tQA2Sjy9>QY2-_E! ziJ|qy%4Sf!QRf1ZkLXQ}25J3^?Fy)IKofz~bajKYerVwgiVuViKpq!e&ENSvResJa z1NmN$fziOLm`TJwh*#7~^XG5ihzyI)APW39Vg`v3c_v%PHJb!QbN(ogL z=s{?n5N$nfg}+z0TR~kR2CM5@TGViem}cF&`}&`Tbpp%C?Afnpvxze8#j6pnyQ+9v zbowd)YB8l49J%54>*^O8s5rzt2FNO?A=9s)N}E;{L(DKjr3p#Dc9=A?@%HNs5Da9g z4}T8dhpuxFCCPsTcZG;}BFUIE4ymTT!=&3r`s%bPq_#oq48rdSc17kW??W%Cl>$>+ z>uW3mA=tGE%KOmW3*#A>3-J{UW;Mi8vc-zs!=&$dAG*qte2(Jd&`x3s%-WiZjWfz& z(m160Gt3OxJob19%n1>)n5&mb<9+D5gPS_yYk_?#;tiRwhe_ibW({E`+fNaBANuRp zaH`Zf#M~Ce=2T=p?)%Vr;1Rbo9Ex zE3i*pea)2dN$t&PYCBB&1hCVtzQ*xM?fiiz4wL=|SOkrw+g?92o@<9G*Qh;(l>WnY z|MOofXHL$(bsr{;#h`iUWQVMGx7xw?p%)?TggoEF#Q)=c==nd^QO97ri4dI1g8WWH zO8&yDkv1DOP;~0TSHxTfWcAdL z8Iz-5j#YgT6NMzLfrMmCqPpPK%*Hz=m$4)#Pvl2QKplTSggUa~DxQFcI_BAqBL%Fk zNVX?}J)B7kb=+4ErwWcn{2K-fH@+;E*`6@(Zx0qg4l|ghh16j?9yDR)PkdntQ za`A{~g}?R2?5Y_Q2O*4L0&QJt*vN|hhO)*l^6hfG`VN^z2ycn3_(c|9&ZD|Odpn?A zgmp2+FS7Jv9z{c+1oWGMJ=BoRBYqL9>iZSS=zfJ^My!t`HqhqRvtDK2lb57d!EC@y=n z%*fGl2Sx=cuJ5>q;(ocyOL6}`TusfuXtr;&a>o!}xOZJ;Il=jN z-HXtATA&E-zA9&zj&+o;bP8XsTJ_Z!V3P#bv$%&>Bb;r8tODyd(CLe<`LBG3%9ySB z=|?&rA4`wO90>MimiS}cW_#A9P`+utXbvMk!G~8y&sg^V#?t<9sxV568Bb13qvDt1 zZsB1er+&lk<4;haJuGp?axlNlc*f-_qf#Q~G$1RNhRlqo^vyM8{hj5-9}IY1V=xbGOw!mhX6hD9=+z@r~QV^%j;0PMO;AF%7P0IQ3@ zQxp((P1E@a3cJ4jd@c1fWRej1jL~ve#|vD_qy_<=1n6bLBbc7B>&#y^RZ9`Q+Gtv1 zi3_`~h-r_@u3u{1P#uKiiBM@_*Ml$DS66`j6N(GFo`|%=v+FAz6Vz0oC1KG-1jjgf z#0tA!5?NTa0Mg(;An!fI{Z)Tnt1;k@o9btVvbAu!Cs9&vkjk!rD7TURO26J zo$S^4#D+y2lG+{ueM5vSDtkFCUX4#tGKC}l0j+@a9X??p9^NP{S=Ya;qHe) zA$K6=Yf+p;MdstK4j=onv9f#Mx9-eMlY9odP5JL+y{xZ|GuIB={2G-@#smdtvWs>lNDBTuNDZ$(f6G)DEHA&FG z8~|v9hX1S1qf$dAEr<|9zI-iC#O|U`tk4bG=x_XP&~`bfW`jP1bh!_t(>Y1`s{cR3 zz1$nLpKq#(O0iTt?^_%}dZ~C3jv!Yu%rHyEKdBljo=2yZAhQRO^$PxxyYb^aq*qx{ zbdgB=BU5`h=Y9$kkA2~H5%{l;_<)Ez_k_Z`N64p)Rv;&)g0{`C}Pp8P*!ijXp3sG++I9h4LP6b50(^BzV#w##r zH(6XtCpi=KF5Dk33X$^}AbAr*I_iC6bg5{IoZkrQLr_rEOV#*Cy$e8GC+i`i;HBcb zU`kNbdp68J>K#)vS2Rc0bp!EcB`M2je2wTbQE+A+XCbTQD{D2Rm*6q!mQ4YEb+c$`%@5p3?NA=929>S-%w4(-}X-keUX%AF$o98->cD0 zmWwDR{+>S~9xX)94W?j6DoFMC;_o@}Sa&@TjIn_%s>2t5KR7Q(SPz1+f>_jTvMUix z{5@=FI~9L#2jOKc<;C$Vd1;G66@MQFc1(--;_pK(>+k@wegXEoue^@G&)?FSN0Akc zhE}1*^1f3a!^(Bq^%_@H@pl7YxrFnMHy_J-@uoNxf8PkKrzy|Zj9v8-RWzI$5gGI> zIgPD zZ`zHq$4`1+O*irvItrr#j>3(*TI2Z_03IZ(7wrUd6dt)DMBIj)TtIRFL;5J3^;nJ= zf}E}djU>n%h3YOcRpWmY{y>&wq|lQcm)>#4eUWJ;+qMe zfn!!9)2=LkCZGtNXJ!KW{)T__Ou%Co@sFMfSWExtnSh_MJN{*Ga$G>WjAx0;AEkZ| zjgLW3##+Bqu#1_1-Y_Baxqw^BBw`xen6Yg88tvbiU{5jIDFY1V+;C^BFNyIlxwa??yY-R$g{bMmP z1SW+eS*gs`x0KHjSI-1|_(>>pvcYR&xMn8c&#Y)s52d>R>On9!!^{LcwJBH7z#I-} zf`mCZI#>_TpjG)E$|9k*t+uz#6LCJTy*6 zxr;zE6Y%YimATXd$XsShZJ<(SCg8r4RclGqE+;Qhih6EOD=znOrPYLR|30plJl5HG_xYcl$C29mXl zT5DzkzNIsvnF&Z;*pf-lfcTOo%e`De&jhTb5-vi`Ou!315zGYaKGIR_2mBM#RU<-M zUL`#f@Fo>AGXd$Nl0*s`l7YlV$}j8SYWctGPZ7Ir7+yS1GXb8<-nc&smGcC^Coq>c z7Goo`9e)0YmiW&Elsd4ojPHNLCNmQdQR-IrzmeGi$vQy$#LNW9t!{A}vIi6NA)ug{ zfOo6b7E^%DGq`UiAkbs_KN_pJ|Bb3*4;9ee2fF|LCmN$?0!H4gUQ%BXnbnZ2kICW_ z)FLwzK%>}2;Qwy_2i*T=Ya7$?MEAep@}<3tBi1isIy<*1?tjCe{bX@{^7@Rg_H)DP zqB?TU0Ft>3X?2?%pj}VQiRAz+G=vpIFf#$vbuI$UOu*Ww+KJWR3_<2dUCK8T@KU36 zu@3M&Knn?en)J*}z@q86|BdWzx}@AjxS0v)4Ij-n6VN%kjW`14$w0Q53HV`n(hJdt;7=Ozk;-z>N(s3-C@>r$4{ac zS$_=R1cAp0G;6wtXW>ENUyvX1j~7#Z!hiRTD~H`g6>NmkEl7c?4*JH`=^Bl__rF0H z7C@3ztAG5BHQklZ<#_LZ16yI}CrQ{Z{>GYaTrF{hBZ^tm{c9ec1b-7bADMz@s36tj zThs0I;2G*~kaa1LMRoYrbWb?l5%<5ZBJ@S#Zj;{;!K~>n?vUoa{|!Pek+|a(jy2u! zWntd?-@tlm5#O5b1<7x!POPOp|wZyxo`(Esb%<*AZFMFLe z-H~hRi-CoSqBn^BXs?aOKPnoh`b{(rKah56?Aa@EI1&&8x3Zv2)m`(AtGz=I=E}`r zZzsW@c-QH>;W@A2)n)sT|Bc4V8`VMIxH|MarfsHHLcH}2;9_*1iGlBM;2#|W-$noE7Z+HOag7GthpQhB5`#L9J@7@IlqJVx8a%?_};~(q7H#;NgrO$>~YPNjyF>_`zqY7}%7L+1BtBw|O8q}Pli ztXdg@oub34Ke1C>(;#lbstNckr;RbS8JRn%)+qcVE7K@5VHFJ=7l9_MsufzyrM^Vw ze@v;WRLX=^{cAR6juQq{MDkNMVbyZn`b$Qsy1;XYQ_Iv|&dG&b)rB4X&NBbiAPNYp zg1`MW8pg+tSnEJvA*?#(v_vW|$c%PGC3N-=tFGE1A|1wAWiazNnc9F_Yr?7@=-@J8 z)vq5WGD$)(5{bz&mrLldYBiN`5o*G!mwh4#tHO(%A{lTo(oIH$w!BI@ta_V@nXu}H z=J8?zNHa~5&+rLEnJ`3 zcG}>+uo>p-l>_zN5zIuqw4#eZI~NLgr{Bs{>iwg<537DjLNu0{`C} zPc2bX?70wJ=qRkZ2Uqg6S~%zUg;i@(;G0$jkc?zKO2(S7Ds4bpu@E^0faFGobXb*r zCRHp$P7i{fB`7GYqH6rZs`Id9ojitIeyewq3&St)q1~uz{$bVV2b+ntxOl#Z3SK4C zKA}oB2=0PwXq5Jm@rpRd3;Hlt$M3g;hJ7Lu(lst(qii0ZZ3-xvp%iC#{yM_!aO>#q^G1 zU&RCIousDXN&5=K6p-hU@K8Wf;kx4dQBmS?VCyxmDpoINz1W;jD;Sy7(AdEX+pF4} zVAY_hoxa25t38Ka!eeURyRA@MK~Wn8tO7t%;kx$0i;{&0Sc=9~?ZKaVq7U%I)_6Hz z)&8P??erofU+wz?YJanDl9&kcEK)y2wJTiLK59m$SOx3}jjP(#>wJ7|r`H+9hV&~^ zwbMgr!hT0Usne&}M1_ya{9C}4dENGe>{e(&J6A|6Yk6Zw=BBHP zH~Uvi&#C$w!@a$f7!Yb0}R##tQ2OFm53hb(_p0f|)~Un^s-T zp^O1xl9uw~m_w@Qz=J%`d{RHB+giA6_Mqt4=e&u&?-({bu>xU*(d&>99BTmQ0#0SD|0 z7-(hkkBx4cKMs?TuRuF9vnskcU=F3aTG0?C?d0W=;vMwe;~=glLyl2LvOhxg`_G}! zPK`}ZYO9XPqDnXt-U>V>+n>!5@*HUANvSa@nPc+O_mxFc_w`mdIYz_@0 zGn+%Zvi#Yc7<8VQ&8bTN=-Hek`bW>^)TDp(Y|az4Ffi%8;N=fh4#a7f@w{L8W0qGm zQ2Z@i?a*yQ7Jr71mA4rdD~61yk}8Jm-BcZ~Hq4-V3z)U&5{t_9ti_wC+|>l;m|2UK z86Cyl=$qS-ITXnng@5EY>Mb*CK^^2G(9Bw_ZyCy^Rv`09Q))7mGP4$cObcPoYrx+% z99A~77ULV$7Bf)l6!7nfGtbmsP9IT}-6PfSxc;v?>tAOr(okva?eB%7uP|$|EY5G% z;xziow=Ddt^-*FWjI$mNp9p3x`n=Lud<^&sQY0+n8YOLcmGrE|5-Mh9EndCPDY8H+Fh$lhjPlME1dh4? zqyEIpy{bo{12Jn+w~BYNA_0~2ti>4^&l`_pBeOex${YRvN0#`{T9m%$hD}%&{tIj} zvlfs2(Sm0!Zb0TuNY+}~CuY_nXF{I18`)zC+5{+Q)?!$L4q`sAiy?iPd#-Dy_z5`?kd*Q^ zlnT1cQ8oUTx%JRkCx6~AM0I+Dcp1tC%~~wG(SO$Bo2Ns?0eFW)Kpcr=rBfw()?&rE zrs4$9xj-Hwv;m{uSqs;}s^Vjy8-YCQE9;%La6i~kdUPP;R zFl#Zc8aj=v_nWntnAcp$kOWJFBU!mrx0$t=v;hxFS4V!j#$da zu5`B9u3jtpur3l9-OBK!+Jb*X%)?k;N<|X>=*Wh>tBVBA1IR@wC_+CX=F@}d9~AlK z!xS!ZSr-YMxRB4+L=pNCF;}ZM;UWjms%5eD6TQt3oN19wYNH7Kh?uyc1zaR-P%78b zR~N~*Lj9ZHb2y67kBAwWpbYDqtlr|dL>IYkk3C|2{zU_><`2Adh;CN$PRYE->8nD> zxK*fveuT()>O`cdnEX-X)+8M0o0I84&l}LRsYw1Lf{Kg0cg~fFHZp`YL`c{ikuH)aO1n-;E3p|l z9f9{WocpW80L7UF{o~=x83}xn;XFDxo^n;v@K(8!C;|Gbruy$3oY?!-6bVzYq_}FUpg@xz9v-?^R&o{vQ%5$efMD|@)bhVqY zUUvYB4MDPYQMZ;dHpStbF%+L_n1=lM8k75dga;hXJ#=KQ1-LzcAV;}jx_cW{am=zu zjg1y^9~eh8OYL4|!%542Wj#jgIe-^5K^~%dLTUGAe{I>rA`(Oya%5Br4gw@LRFQbG zfA(*dy*&orKqkO^L!dGu^vQ(@BKwMEf49~t&cZO+4TPJCbe!=jqoPH&aM)`asmGP& zJ-{aT@X(zoAL6jEgC zA}D-F+u&AT?I1YnbW0E3SV=ol+IS^}eEL~L1Dxrua(gG%`sgED$O#}nq_rNsHd@ zO!vcW(^zX7K4K!HQ!P;&iQBk$oL8&rlU^-?qpG$uTKNLsSIi|X&U8m-w_~j)3$d2# zR**+)tx*%bT0cGE)gm}*y-zFNf>tzXai;slvN+b7y%o3$2J$8yn&AcxoTw(;@FUahYm_i7Ow^{P+nEFLzL?~oQ} zy8BNmVXehWQsCY~=O(Q+da76J^lGmb!BK5H7`sYBp|yszIMZFXcOGjkJqka20mx5k zt2s)C!+g zoC{jrNQ*PwtqRrnda)(4YSl(ikHll5_5EJE4zKcR5ge6LYV5i%99kKq#hLCWtE>I^ z0v;fdqd=amwcgfR2UmKv2#)&Br}b3?v~1GiO!te$5!}YLb)dBifL`QPBGG8EA2)d)z&(SZmf(IDNb7z(+vhHXfes)r!7JUB!CI9MB4Aw$ z@(!(a@By!uwcM*kaMY(=j9n9>p*4cEIMe;v+>WeOcLcP41ldaSwy|`HS1VxSGlzcYvnlbIB-66I%};%%e-1eOTAhIM}6eeDvp8Hi%&y~Gu?Zh z?Z8?)iZC%f3*=Q=EBA4)R`Z9wS_DV!>}KqGHWpeBkrrpVZ`!NI*SmZ0-R#34f1|Y; zt?_E*E%9m*9Mz_~(W+DhT6d8aXS%c@@jQkv-_}C6M3KTD5C?wK^~KY7rdOps~?vQw>^|o`M!Znia84t8(lNM*XS5(H;5_UCx6USpGko#+`VcA}-k011E5gb)9 z*Vt7t30nO~i!b$zHD<&Hoq+>M~68Bfv zVy{;HIbJP-qplVht)diYo!>-uQKq{O4twga*AL(+mdPM5(OR=Rd$l^?QpM*9f}^%J zHCj*Agw}r2;!Jn-@6}lSX*|a20g%7YTCd*Z)w<(;uNJ{kw>C3cH`apIlcdF&?(s*G z*se)yF}`B!TcREkw{cM)uh!>dyjlcD?Q3DQ+SY;A1l+umoavr)F`2cVU5eYUw}U)Z zYenAb)jD>sSBv1Nu7yTxa~ia|krrpV!%tOZtygc(5^@X3d$iU9t##xcuNJ{kRek-C~2)%OHPb(wf2?bRYUs!^8FdO06<9hwI%&UBBj9M5(QSsy0kVvsj# ztpnA(S}(ToY7rdOqoL7??G3wrBQ4H!7iVU&*0>??Cr*O=tJX?Q_G+!};O#Ghqio;# z>$h8=wW1^Xi!BTroaz3# zK_j+n>^lg$#(?~Q)>^8yPPX%E5ggUbH@P5 z2t3%L70&!9*;SVg7zO^N&CVC@o=24b5&9U;QPWUzkuDuD9{o$3VJJ#>K3GX)53{V? zgH?pwhLW!$MXc%AQlxLgd;d5Bp)FC``d9emU}lWLLi*3ZI;)wRwrVDAiPB!B@UFmo z{_n;@IDriu>t^F*;YP?_6^XlI+(z9%+7fQ=3J6CH#SqE^tG#9hc7?Zr zY%%EV^5Lxv0(G3G2U_foJif&!&NMx>T?|S<|6Ea-T?2^1T`h0W*nM2?8kG` z#C{Z!Q$ct@ll-S8+BxjUT=f-cJqX(jiEU9z-5mBeB@RV80797|DSPO)e)g>n`|27* zh2I0bXb4ohckbd&hux@Unwq<)fL@G6vJNrMGYW(8PUctJscDBSUtVi>yR ze3&A22cfSa@!W)(b{OxScBtgH7hlGn_}sehOIuXT9u+Ye#pX1|^zG!;G?lVjroU4~ zBuvNDzl>{u&FnGueZRL9@@4RLQ58ImG0)K(Z@|l*VDF#eg4=<7`6+(RXoBkflUBQw z7?C~I-q;oU&#GuKQRZL~Ad*r#X|1_GN@UNlH|>OyRnftmTENo{=c?o>n>qGl;~S_M z8>?an^E!dkOY_vsOseiN)i7P8ryCF@iMQPa{l+L<;xFps4S+VA(s6XZAaRZT^+#y)o5+4&my}ybJuhZq zJCV4l+CNQUbw66E6wEMbgCRs5eS61Ay*EvVNw? zL4{0?KKfN#Q3tsVH6qC;zoiKn&PuL~D=Ws_o2ac>vr4_QyUbh)#ON|84-?;)N@(af6%Y7 zt#^i@uZm^lfP9>mC`&kE*sRXd{%eVg?P`yXlvPPcK{rZ!NS2EYyMv%)f;_Kh<%^_l zGWv(224Xz2ryyCi2gzU)_m$CSLqagm0Pqb%VC@ZjuOjIV z8Iv025@g9|Ae;>(DFX&ddw93@;xe!hG>Hw6*<=OJbti3>cK(rCSaSogmLYKCdTrdA zBa-^a%2RVYiPpe6Bl)_MniAZd!*(ulVqDP?D_rrYGb8tK6CQkb75D~U|5i@%0JIiS zeND&~J@`7VtS{Chcbi6JGlKQt`_Ed;N4*W~LxUIj@afS5>4CS#z($2_}I7Bn4S40}+hdGq;UVHwA9IkUMDfjl(`L4b4N2Y&YlQcN_gui7&!bOXh2PZ;gxd^>O;p#1q|-6yFO&#s&N$%r2Xd4J zXY9}L2KnWH)*5&`nV|+}Gik3m6$_UR;2uNZ4l;+ZP-Bj|TMxWp=z-Wda}Y+-9W^>J zs;dg_woM`NW?^$uutO$0B91o0QBHIr>;F{P8sETiMAw{wMcuLNy%r9JPG0%LvAtX&U;1TU06W%M}39=4b0KPh!y`_UnIWF zV@A3tKh%YLbBV*G?V4kYkEjt@$Zbp|s4KkqXlYluTuFRFco$$j2xs$#-Z9@J635Dl z;R9xDfY(Y@R8}ha4(S6(87ITx}LI|MjIk4vTjI3X60(ws237Z9UHM9w<$)k zmz*i_OF-TV#BE1)gp1{77d@3!mt|h=;^`H_aG;;ZT#7dG>CW(Ffb< zul2!;L0d$px@(K4A@%{%Q6%eq)FaZewX{5XSatDI-hq-!e(79Qw4D$+t9BFHSTnpN6mpQHVDV?~*0> zxWg~tr@pc^U5v=Fno#3pi+k=JEkWuMGwGoGJa+tSsl(|0n1kT*ZOK=#2Cx| z))9&sA!4k-&o*KV1%r<)>(Vqh-0mwxZD-j#M!;Kr2aJ!)VMREH#9;^sam$~tG!XIC zu|aoq|DpTmF4B)Z*AI~)EmNr0O7B7}Q)mdBTSglq)_x3k*vHP2VLd@gpk5t2Plg}p zr@EjBnVpcVUNJZTGIn6^aarDSk#7FfFKY5al`ciTg(^=HTQ5`@3|9FIRb~J&3st_) zY$UcL<1HlXDH5P9NgT6Kg?05pl`c=f&p!kBJ6*yoRJpgzC8(gZU{n~AwH1P9p-QK{ zc!@v3H2~EyuwJNA)VZ~STLLOJFfCM3?e2rJ`O`O5#9FSA5nqFl)@~z~Aj!BFFEtY3 z57-^6VM=W>@@G)toLE$?paph~X1Itxg6wq~l7)odN$VBDmsFaZA>}R*_6L%t zr1r@X;VbP9!&5LBrZII+7Ow8dh>dNGGeOu~8CkCl_Blm0VT)vx4aJxjXXC?Xui=@Z z^bX<=aBHYbUqwr+kW!xhDGh2`i&lBlJ>W-K{ zV^P+^$_%W}B3nK%5Ioyb|VwF_0#4CXI8u(sTlr1K6sJ!%A71Q7SMD9ukgW3trK_G63YN>2!JI^JSuy_wkxlPix)KN9oaP> ze=kRPy2t*2mbs`!3E2-P2v4-M&shm#H7fWHnLkmfbyS;zW2ODjn7V?j4=cv>A(FL~ zU>@imw~XU~PWcHpmgNagRcVi1(wK?)$SgDk+Wb_rsqiF9d)G?3SAqOn2;;gujHO8X z=*o0464{e9BzIC}3a=yWTBB3MB4CdioQ^6lULP-ae7k~p4%nLkxDrtYz~8En1c&Ss zFupXbH+;1;miEJyOT{H%62q4bl-&Qyiacpgz1&R10jq6rD(`P3ctJoQD1w41ES7yUrETa(nzYO=Kl8g-+L`GtmC!giccf7-tNN`;o6obVl$U%Pu+v zHcOk5f+OL68L_el99`O&fRemHhVTZpfj4Lj@y17bqdGb>jf8|1Q~BOl;dxftJ*GBc zhwt|GFz;8o-YV!NY5#ageQfeRg7~JuGC6)E-nY9`+S}%L;9{wKpgtz+a`R$xp1{M% zd!${~AVxfdV((K?+VJ~%hx(sus?5j23YkU6fs43G04L^SlcQN6vO_>wmSO?6C#K(2 zr|=w-_MlF=OmA`>x~Si)f2i=BmG;sVn2VB~!0U!&y-y>ql)>}TK2ZlZg9ju3UX965 zh>%K`6VE1@Ku4K$5vt1kCGDrr;2gXfIa^E_+VU2pC9R$fGJ*E1bP?d@%52M?y8+Wt zpCIRlz=EGs|EOxogY-NqtMF<)7x7f-+LpbtPZ92om%#r;;%3ON$W+zPe9P|g73Pj* z6QHdO;~X)RL^I2t@C`zyo&X0K0!gU4J#1#NWxv-jPTU8=VnZSoCaJxsm?f57_x{Ra zBUrDM!&LW_ShJ61pNPe>(8Hj8c`emz-*C&mxjJqo{SJm*s+;&Y#=Wx7J*XQ7@H(uj z+=9D3iD0B@mO6Oj$q6ZV*ZD}BM&dkK+zNO{%~74jVV!4>W$#XJr$_@q7-dM@Uy6kF zn}2OcP^9@FEH@;wNx1Ln)k1i_x9lDRu)gP6fUg?@4+q+-URD){{VhH?BaeV_(y&;D zp9AweA)7}$4Tmh0)*^e-9QNNs8!%HkJ7O9ONsZf7TJr39Qrab(4ZDHEZe6n~lVl?> z^NIZ%Ro;S@E_(_bcIFMKO065Pn~j>nTRZGY&sJCE?*TT!luvya`@^$dW-q@9J0Wyj zX}Iuo#LbkaGMOaTg1O14r7j@$I@$P9AG?Rc{vbY+N%BK5KQ`K-0|yqs({R|42O21~ zpMYI7Y6>6Zu&WovtMU~(!9he)Z41349?I#g-l69ahrNE8RQQL$J~p`f!QW~N&qjy6p*5ZX`T_Zu2&1!;q~RkJ;Z=wIY#V$O zGqN+@dWGa8NHxCR>j5*5JM4uEFuodsk*8TAbXH8U@SJhj59}yVCUgbXi*Ozup{*(; z3eSHWcJ@wnMUbO`jW>84eGass$XTx<;(-OoeuN;Og{ky0V9(>S;k)o`Waxw<_>mPu z>~EI{X394~e?O>5w>fZC*2vr=z9K0hcFzh0%#;^E|A&f@f!=w@zik{_W3K1t+h6=J7Vh*b6F0dGwlpQ+607GfVblBO!^1AI`RPG~v;C(msm_9iSI zqo&H~Ak5K1Lg9l$>~(`%D3P_mHfa$Nde;=35bh1J_w-Ix<=+Cf&sU!DshCQ$E35Kf z0{hli{+Fj+!m|LqF*RPWcQ0?inGlH^B0?Eo9AbYNWs92BKQ(~WA)I%oJG!7mc%BHc z=Y?ZNraAK4>$2Xl!i48S?0Z{acyOE0dE3t0W zdyX~vB&nD=PJ3eSR*LBXt3e>schU`AJh~l@O{d*zeho!01@)!?x=^I8PP^6*wjzxL zVOju*o26#5JnuN|B5Lm(I~ox8(k_o&)x##?^>W`?)5*cjGx4YCk+SMKOMh!A3Id-;OSq6~zyhQxjqJI0EXY}+?A?;tLN z;Oed|ViJ2(sm-Ouf|MZsaq9z4;F8${sWfE4l9u1r3AsOIoq=aUKl zq=duWcb~kJDm;@yt?$NE#~Y4OXdV?Dd*+S!X89}% zzIAP3K5|X1!aXEgih^h0Iq&O)7rugj35jPl=pQw!;a6=Y9aITXt2+x%wrw>MZni5P zURO;dGcBUV1XUy4aRV@LYKK~frsAb3MPT(r;=Z6WrPAheCjW3Z#I-+LO%x2XKfI|( z$Vs5jMq;DrAAh6R9uuPRJS9&`xc)-qmA~bg4m^Xn7;cY#)QxM=Gl-iZSN;s*J3!0~ zVzZ9<;sP=p(6Q=ZoJ!h~#4$67tgB}byRA#c0|MZr>k?)Lv1Dnspn|dupb~;JXyEHd z-Bz_r=4#{}fCd{_&mgWnoUY*e0nIfq%^*@-B3yJ*?ohm{f#@&@-&f5a_YInnJ~+a2 zGko%%H({vA@q=9zMf#{{GFVt~4@D{`0PY5855SZ>MZK4d@$!xg!+(sXeT$ns@>YmsbIW z_tMSdXX4@~4uZtj_Ku%DN}?w4x|$;t{EOZ2yhlkC0qUe-^)l4>_m{_s^ndJ(vQWht z2z(TAsPU4vL=v56(#Pdq`2=Q!Ug&{3!=!y?s!LV+2srDg1lg@hgr1;oijnrG&u1#x zT_Efyl20~tJ@pT6%r%`Ku9Ut4;m05)awF60N&CUl6xHK4MpY#wKc$E-msVgGNVur0 z%H+Xyv}h*Fl95Ai$Er2Q+XD(jR=D)+!)SHL)QOf`WQECCdK}QPS^m=_S7b%Wv{q1+ zc;f#2Yxut+Cdg_e;${Le@1yGS=!j|aBVm}UE6c{}*(JzdtuYmJF1EOIC24o;Zi^QH z?$!j^niSOXhpfsnu547g=zy$GK{!(`DMrTK(8gxc-ynqbh8bN@LMeH;rYbVp^+H3D z2rP|oGE-7ZL_$w$CguS@eY={VoL0cw5vKUMOYGKHx`;4v%AgYj0LVOz7dk zLMrtG@JmKVacHb(HIerI*%g^n;TBAFA}M?PI0e!!OmQ=(5%3noVSD`QY9Z|xb|`y# z0>4%32xUqUc3+G#Wh|ikw1h}tD~hFE{r!$y!!kgR`bsm5`NX8!%77OEz3MAX$4(Y* zB5$pm&!vw7`a+lX?ltCgmG<2yDlq4F!u#kps6M45C+j9@-(RzUIn{vICXU~}VNP#p zTkn=Irx18akk0;8=-h_)-aMMfoI8LI57J>ye`%+m$zaYL;7fvZD)+4-vanjGdDVu@ z*$jL~kPdT(NPE)#Ma=mS`0*f}D!o%g)^KUJzcG_J7l8j0q{AHC$X+nH1#_z0iuG?u z{$p+io)yj-i)oQbq3o+R1>TA{YUDFK1y93&u9kuJ2GoyW9{HZ_$8C`{TiO@i$Q5Id zJ&PdIv-0N0aTZae#kl`8yrNhI!ZSoN!(NdR@|Puxtku$PTv;6-dqFxxM6$!5vq{>= ze^Gt+E%0B7L&kVHTLu-1tnJdSyQDR@IpQ|F^$aQ4sstutcD7}9RVLrIHNd_s-~gF^?QxrB8PA$#v(g9b1nk?8KZsuA}3JSEsK~1`l{>H;3%PD%%Ynb^XzKb)S#iUODNib&VA`pu+lk<}Ux$DgUp#JV8m6EV2eOl)u2f7DZ_ z#I7LqFGn!~ubJijhmW-2AXsD3uWJ;T>*U~JnB0r*O&S@J$#1FtO zX*2nv%|FZ)b3bx14DQ0F7Pr053x4&oTo6(^le+=U~Z?}n*0?UTTn zO+2#SpK`DBtp)0hFOLI$nkdu=FNMy=h2x!35$bI{Zvj3`lpu@JnRlNve6RY%^LOA} zCSI^rbb>6L?%Xl1Hup^Q00bsTO0Po&|7LnE;1aW)w;n3w60O0$l}fPk^-IimmVK&D zkz>Gq;5v1Pbe3P_Jn)Wc;FBP}d>wg2e>h(?E3vX#{qAD#1M-*j}$e>z#f7R440xAdb0?yjmy3W7~Bf4$L(c z{kmp?UM5#m#A-i|HjWSbmuRDaf)8J`u?=PYqm9=VreOUqGQTGqR?(qgqK&05;GGv@ zAf_;o7z>Uzs(uhAY5=US3GxY2@I@PS9&9N#BC8mL8_Ojj+F1Wi7?bV>VJwlJrNPWf zCfaD+KVQrTw%mwNOGLu*_8}Z?q#jKZl(P-^OT>Ae3{fzmjrc@BfDZvGBls1Db+j@0 zy#$7T1oQ{N)DcqEpregvYNRtKdJv{kk(8B+qoa*K8#ZN5W8f``!#aKu0HTdgj+Qd# zX5a&ij$b5!Xk&6tGv-VNJ~v2*oqa?bb@OcItOfpTkPh2}Xyc|_WzRdn-`6_6Xyb!2 zWy)DV-)jk9v=P#}fNOBvg+Yg;hKCA95N$luR2h&0D2-s+Z6ZNO8w;cBa_LrpI_lEi zKGo4i^WuEw+zEW7mhf_Pw9&9iJ?6{cf7~fo7FZTmFL@p7`LPgjRy!r>pH|!P3L>sq^OJQ3ZfYgkLWQRWo(Z)k1>Nx8GydQDM z7%xXh8*|q;Vx0-V=Mg8!Didv-y1~W7Cqa6ihj;9msM#;HC<+=2_BhYh`63%CMA8#kU# zXLb!Rn-H73Ke)X*+W2<0hj&&NF#8&9)BV0^W7tm)CXNSbu95X4>S&|&vqel?3)1%T z6m_&Qaaslw-vjA*d5SvP_-BhcA}@gCz%})nk*A}LMgLT0MGr{zi5NWcbhPpN**qq; z1*u1QiaOf(?4u4$91hZy@)UKn@zGPNA6J01sXRp;ZNxoSg%#ff>7#NKIocrS0E@bD z$gHvG*L4caGWpzqOo}!(S&^|Lu$YTB=(@`rbaCYkx=!*2T~B#~F7UiT?m#B#i8k!_ zKWo{Zm$B@x<8V9#hh)7?SBf?qn{2V{k-w+n9W2PdM`PO0;L=+y+)NG?^8hZ@1o<8* zC_kfcn^ocT8_mT>$l40R&T>i5Sry(bbTa8E2wxEC6FLR3l80-0-ijXhVwAW5>>ndS zTcUE*u|!e$w3SRf#y4S9*L{!vT4T{~nQXeIp0F&M{=IwLH4d3&rhnT7I%F-;!x(O6 z`nOF6&R=7YIf;PJxuf*-?{9Zt5yoQVKc+F=QF{9K>9NV;d4Rh#L4M`yD4PC_To5MC zA?p|jUzJOOL)La=F_W%{*&+M3Pc=a~ zErAyk=QlD$!EngF8PuNPzJTr|_)mtlLpB;u{!-~FfaVeGSA%xQN?#3U&Xd5m5QnW) zR%nMT>-z-eybJuW(Ft_O9w|y^&iBAC8l6CgEZOof$2|%!0zgvs_;KKnxjWQnPD9|$ ziNp5z)dh#FMWyDbgZ4iqa>#sHe6C43Iq7KgH8DWFxp(t7$grK2+71wcD} zrL{vgc4HXVeFV@+UD~_Xv_rP`=XmB^0xn1EHmE+;4%uZqi$pmIz-tl5Z{KK#>~^X8 zy#??NK|0zYt39PD>+}acEJ#N?WKHl;4%Iaq_~IZP?U41sjU3`U1N@~R9qo{P_FXh{ z4gvoxNJl$lcdt@){R;f=ARX-QSLMc|P2`>rY%SqS>mM*I3j;E?Sap(c9vfPU;cwfG#e zX~WcR`58>-7;SuD1wMzY)i4j6lLTfSv3XAgvb`cDrpwVlle}9=>g^9O-ba#1*+98|%Y8(@1g0z%~!HU`;`*B1? zCT;;~7ZHO;o_5G);&n~b#!o={raVRMkoEdpwfZVZ6>vGfMo~LtPs|Ku#Waxeh!||6 zcF0cS&AwFijUe4#jv_l`1720ox>@eg*{2a0x>5;Uq6kozXvXZ!?v_tmP?O4j5g8U4P>5kG4+2N)&MQeba zH9@BOI*J^!J?$e!J!IVl!ae1Z;E>JPT*{;cAUr~(Z0aahG7j0RS98P`V6PYvYKce) zzr8a%WYd4GEGXwN@MFYjLWU?94%sg+IT`*L&?SPKF{~Z3!r>tdSD1*401|hER5fUa zEa9yf<}?D{f;eoY;%JAg?H^T`(-U}qqvPi!VTNp9$HvT=1bnv9@pHB?LpIo1n>lNM zKNF;*XUHBbZNi+rzz+rKusv|dUiwSf^DXcnwT{mr`>nGwC3F(@KN2^|=a3ByRaVpj zluj@Y4>LpdLnmcG8$cz#(%K=5+*FC}y93a0UE15H+9BKaLke@|0AHddyd3S2_1&oU z>}KFQf^@V)HsFt%tn(r8<3T#wA*=L8TjpE<{!fsOcF4ANwV6}pKHUFC^54hWA#1iI zf;mlr7ZZni#cQ*6$nt)z#+<&uhXmas2Bflxps zGwi)rQtgnn+*E^Y?FQ0qL?k=>IdI4(OsLA7alr2<4jJR+Xou|FphV_83VbthO#k{5 zaX-FFL?#p82C0mQ!Il|^>_Ag?XMP09o}v|t^loEWueNr`#z(798&n6qfQqp8^^3qE z`}J%L*U}C2yRK7<&msHISapYeI+)9^Q-RMR+d46YwYPzJ;5ypg7HEeo>IrqT=_}B$ zT&Dv6B5=rt-rJgMiJgk2&q%800~_sg$X=|Q$L!W%_9AvUUF37f>`QsPvqphA-DuO< z(`$rw$f`b-%*0h7J!543h}t1rQ%#*0_JVY@JViZ2wl>VoivI!WN_mRfA?ta6dnU%> zdRhzVnvtg+vf;P2W?~DFIukK?GsW)47dUaS=#Ql&7d2vV<<` z=-3I;f$|i!LpGwNQv3>}U&~Qshm5Z}uI<m+Z`^^`a00?!-d4rG#^*CG1_es(>)^OtdBrAQ7-ACRbk&3$xONk zfM?f6RM$m9YK=v|WiqC2h}R)o z@W8(~WL*Lsvi^|wcgW&j%f@R;k-3^I=ts>ow=6F-#9QQEK>lux>5kG4+2^yGi!y+x zG(p}?3aX>XA=`Phwit!1%OHfz@YWDOf70B3~Pt1YMr_aKLKbn!PF5_ z)u0`+iLE*@=WXC05{Io+9PN;$AL+oHbHFbc9o~0lfn=qhlb91Z6Km6vl#b%CvjvCj ztrI28$p+q(IBbs}2Qy^*uOu_4JMg|iI@%$7@u;$AJn$)6$LElB|3#Uy0?=wL;d98= zjA_p`>;Uw-ue6>ayZx#%;A221e5JKR=E=+D(!T+^s!MzOR6AtxbJbpppM?hjko@~p zJ7ll4&10Qrz}peWZ{KK#>>IpniS|T4;CBb#K=O)RVJm9T~ z<2UBCL*~A$7E1I1-k&&Xh|CikRCCL^YkIk^?Qxb4%tgTsMTZJL4V(9 zU%v<(vXEChu!C|I^vl<&#pjTf-J8hlN^>#mkEDzbtib1xg=MRK+!D;*#O6K;WP4kn z9kQWk>Tm`3fWF{575Eo{LzWw+7L08K{k7}V;&aIUekYtOCr~)#$lm+N!#m5G zho%2W*X&j8kPS|4%EWY#ni4U1uWE?T}R++m4BUfE0$S^fj&44%w$Nk%=`xYCy!`R%?f> z`)}$9Dgmi?d5YR0o7_^c;yobEC`XYUGI9=ts|aV#H5UE4mVj9%bDv2khs@!K+&`au zENswqmpAC*${Tc@@&;Ytd4t@6Ow#i@WcZZO;Ocm=(3{=$XkLu@TsIBSKqVhs^0nq#k33Om*FAv}lb*zh&~h-H{?J zjAqD^mR{qKg_#+$K2Y$RA-e$iFuxhHMs_;B|Mnm{5XpL$J4(-xz0)7}ecK?vi^gH`h=SpeO>NSQ;YtgzCIrbUV^}+6+qUE| z+yGEBf~g~JSw}2lE($O7`wIxt_kP6a-P?4gQ{xdn&8{NXy< z-WF(w?8B{U;atcPJXDPIFD>9AaL7(oRktj2LGMUKxaR{K?Q_Tu{;Z~6?*wz=bt>>V zWQQJ6qy1qp*BNa(j=XkjhwQP3LwNVU4$?s*>qpcM+2M1=O#B+8i{&Y5hwSz`nM`y) zjGjQcW}Ik;Y|wxVCgy_Fj)=h{Pdj97x5hBBA4nt1Q`8RGyy8M8&I4&>d5YR0^K?-s zmFGd)Q=X!B$j029$ciUG`k_2U?U0oXsm#O>I0rFE!L4S8jGP1EI*Prp#-d+WO)$%3 z&r!*6EeabrB5z(sJ{C6Uy2~4MapeuVPVxp_PkDnb@Vr6pKql#V9kM14yU7l`!*vnV zmy<20={lnvvRsEP-)MpNA0hu`jcGrF9kM)!{p(&KJ^=WsCdeO1!RwIaJ1PwQAyxc> ztY1O+yIfL%qeA1;1x%`fMx`KGztbszm6Su)%n_XeJ9B{*5l&`G+VVPNjU36;W9*Qr zuFFJSYb^S0J+4_qleEhC+dQHc|7{U5vnKuDGU9LOv>w+gB0Qe{7Dd$NzpW!$^WQcR z{rGR&h;jV4IASUPZ5Oc_f4`N@e`zKLHH|t4JC29+y{VxX)YuXCfiC!%GW^@U@4gfb8sXmgm2i2Ju%T}2f z#%LQLB?jfDlL)_9mftZZTx40sed{`L)hEF<@3g(9>d7gcsp_bYvczN%*CK5q@%{c) zw;S$a^bH{I7}Qk#Un_879U@${K|Z zfw%iP2HZJZjIW^n6)4pFuO~aRf_&8>hbH^PV$V*)_bNzdG<0@4{B@|{)0V#!N^2oj z)lyQ#eo@7%^WwGk1c~^J`Q7qUI708qB}*{4*BnugtwiH_5gB=LCs!SVD?Hyw{Vfov7${&)q+<+ z585ZM1?-cGZeJ%{k-8Hg6g%}DyrP17;mS1{&4N0?NHh!Rf_gjgl4BebqgVc07d(l# z_@WCYzlirkf>?mm63Hs_?}GalHej?DklPGuy1*P0&4i20jU9nFl8t`_Hj;g`7XRrD<+aKRujckAEY;|!263aX zWHf#qZx6!`%zgCVI-$Y7I@AeuN=w8s)b$sVL;>rxeC2lX<1YO z7e+D~AI!tc0;vfzQ6Q)r0?#Z|$k!3RGYgBwc3;s%=^TVk;F*O8L7m|{v#_eKJMznJ z>m<^0@?G5?gF7?v^JtONtVU;ig2EbqbCnh%r!eL-=nHEw@p4*Km>5$iEEx`RMI`GR z>W*(DgPomMNd*f za4Tpxzyq2f!>U5zKm0!ToW7BpKaUV@dm)P02A-=|$_if)J52 z(OGOaWR4tYh{Dwi=46Cj#VgcuriVWQ;H*Xe#&*V2kKko$ zIWxoeGL`8v7Grv9=Ao1_;LDfXgLm5;F{d@Fmp~zWWpM<2ypZ}#(q)lk$`v3U4@&F%;|3CA5 zcK6xOUaZ~ET5Aui+pb-#V_9~whjpx5a!2lRr(8(~Dx%yHI!GcR$q|K+P*J%;p+ZWL zTl`+{_k3oa{p?r2KW07V^?JYOJ@d?Go;lt#vtlt^>m>beOs;?5zi_TsQ^XV?SPdxS zjVS^362|1kz4(45k%t&-00<*}NnA6>WLNECYUY^ck;H+MogXLA(*Jz9@~E0`YC)7o z89b*~84(I&y192OL8YP?F_5ChJmOzB4%W&a5cMMowIXWZq=^Hh6T!H9;7+DX5roJSuRAUDCU$_M~{?5R9XdHWt zP^rd@@L%+<@~V(e0-9i8u2f^j_}>|TDaqFmJ>NiFsmAn7c$5-Vm1@j0{@eFNFh_1N zMB!eIOl@`jR1rsAzG#&?j(_P2kgcxIN7NEyUb14g;NMBr(cK(dKX?PwH67b1co+Yh z9ejrW%?b98q4;LOmHFTN;1>ATQ>C=`ko>HrSR|p_rMlpd{LU|XucM*|T!!!-Zgr(Q ztNI!e?yM?8>UjV^S~_n3B}0Oym~^iudJvvRtpuCP-aT7S(LGv7?v^1Txi_BiYN7|>>0%dv1jFOs)VYJ^|?ga(**4&k&YB5b;d*vYubXX+rm@3F?IJrx?7mEQ`>a98Ii{Sjv7m(vTx}k zfK8?da+$PwJav*l$}LQyc_U7a_CX8Jy*f=mLSYUzBoyZQM%r*Zh>X|b>Z{9Uy1$L4 z^}AT=fSb!^`kkX6={$(Pg<$o40`>l3{1N7|nYgn!zh70r)1eMkdBALr(PG*;bBPPHULmEYjX!9#s5CNi>)Ps=?Tu@okB-?M6^)ZM=wZHo=%*hpMtPyH^h4>A zE>XDxjyNuK%=X03Jv`~N7X}nLX1hx92!_rQROFcL*0_a;UYt2*`zJa%*&MSiF4Pj& zAXf7oh|)u$KeI-mSbdM#-V6)f#5Kok*Nm^h3P%2y0^eh{-|Aa|>2Lf8-8p7E{YhL+ zEylAgSR zzQ=46V2;^-`X1bQ8X~5hui#25AC;^=W}E7ObQ8}Uv;A~im^uzIla2I;WU4u4yAO;+ z{|D#_!}ypO=9q13KsNe601p}hsZeQ~W3~gQq>Hm4+%zQ8!lWX{Y)2lhgB!{*>w%yO z!eROzvz=HGhmF(&t=a!lz4qal?K{nIi$@PIh8WgBKa@Y0jX7q!YcWJU55^45Qa>Sm z%yxQKn%GJfuK>PYb0im*IcEFwrZpAm00>77iOb6zv)wNwU6HPU@UJ0}P0lgf;{xCe z9`P<*KM)F42YRc{G23^k2~P(j+pt)TIc9s+Cv;Y^K4!ae!(_(?T*yE(CKav3F47=9ukgMg=Qvn}L01w3%bJ7c_`g^2dSwWaP~;+t<;;ILBjbQu(Poa>j=vG4 z|WOW42d3T1(*{1KaB0`k3uWP0A|$KLGpD!SylQeLoIU z_;p}+9b6x?z3g^8Hv}^3Jsbjo;LV>tX1jFh+6u1&408)!+#It#;mHQL*c|cQ38U7D zTNCG)?XS+@M>`VWb3O#A>X$xdd)II9fLj2@63uds*&g&*oHAh(uKqxGtkJ%pSh~SuQ z_ntb;lwCmYOCn^Tw;c2_+r^(r6xKL^lYI&LnC+yFaVp&hb`wi_L)CYBOD71%3A-W;?2-Z6}W-$DEu zE$gi-`k3wIi*aP+7XZI8gah1I>0`Fnokkz{SAdr^!5p*w`nn9&l=!d0BnE;iWon{S zD(0B&+#@-PnF3a(FO!oq$83L*7OI#JgEi2X$&I8rX1mn9D8+mZtogo7{s!}p!W^?b z>WMmvz7Et+edxj*v;E4f#)@`qtVbOP|aEkhv?d2_?@DOMx4V6k#m}9mVOj1jV)(sH;)g# z0!!tOdkf~6?fEa+itq^$4)goc$80}dy)tH_5r2%5;O2t5uI8BS>8^^(;48rYH5@8R zbIf*+?@B376nf|t5tJS-ZYG*zw(tE^OEdtXg&}co$@iFTe|LcB1;Q|2lIl%;co%mn z;4ta6`Ay+{1jaiEe3&$~2P$E18Tbz$Ce7n4>P_c^=xYk{wG%qqUiZKi&b=vtZ!LG%151>p@VA1;kKeC3H$Y z=^iPTkzfi4^$n?hd8DLBxYT{sgO>?ppz7?hm%NLa+9v=$V+f?!dwZuD zGwFfTR+yYP$K@}^_4d$#5+PNx3D9=LxCI9HUcFZo{y@}UaQ+4DA{+q5Qx9PMzgHBJ z{zSV(nixO5Y|h)r{VywuiQ)<0gv?xr_k40eApExKJ=&mwNcbUmc@FYe?~C)zMZ%F_ zx^FQsJGNgXk&q(8|9u>0UO?jikOBQ#;5|N#KSB{wWq74G8sMHj==2~c!ilh0xcSf& zZ^Bp{%iJJFjDSdW;OT}lV<8MsoEN0MyJbn{v;*GVaNZ|QEZyCjutM6)$DwlyY=Xf_ zD@`EO5R?Ht)D&dAa4&1PAcAIAlP71txb>sJ@PKyYY%Ou%k~o?WEtc~YY<5X}7- zot$HcKBpo165;v+!L7AhiJ6Fy@*W6o=w^9E(iaFm(+59_7>%h>iEz(>K=SqtRfWtS z9nt}4aN`#u*y7+p(GAtZluD`#Ci3g&;jZGD>!Q@Kf6aF9uzzV=@7R8qet1nkM78lk zH(Gli@(@$GZ2^8m&h1*AV#2wTw;cu)8TJ=cQ_j`L2`Vz|pYNwmrZvO<5G3LZ`y2e7 zB`!g%%102TheCg*k416$4*P$Fu$#DM*nc*%5i1zi$m_GYI=+3Zz$lF+7 zk?(`I7QvbYW1M0Ck-2D5_8|U{#^homm|=fhOXSi`s2TRZUK>Z>-9b$72CqIzPhLUa zVLt(8*x$Az)*90glkY3|4wa8eRuB7!^gv7HCY~AgpL-LgjzG)=BfXMLHN$=jH{3}% z8|V_l_>dT8*uS?7EZ+d|3qv3kDs40D-_)nJI1a*PLn19qDl+V^w4kQ&`xsmQ5x8Kb z!}J~YzcvIdUJ7WL|4Vg-{jCRJW}_n*eGMzG9m=1}#ti#E48q+ncm!xvU1dnX(V)1GR`V?4Py6uMYB5qm2oZifA@CIu>OYk_Sv+RU(j%<3{q{##&2jJz54XST;nFt=+R&Tq&h zc^^#wPrR8HX4s$YurchvUAY;PWHOkwh|RW{VSl6k8A@9#VC{`IGweT_?3I5U*hoiS z5BvYkF0O2!4s4c#>tX*7d&(+&HL&##u7~~QaA!SbLLR`M!w#;8{SoIw75)dXs}8P* z{R=Zfxgn6Dn=t*2;LV>N_V?+VrtoB7wG3{C{qL^B9D6?EI}k>#6SpSLu)plL7|;&_ zIM#jXI#nqhx2uDW{IUur-QGiCf{d=3x_i|Ap0X-5RZ{(5#jX3DmpcOwxp&|40A z*q?B@wy;J3oZw5)!~XiiVpY0xfxk%{$5cJ+|EW|JmC$D3U-;@V!~U;7Euj>h2H_X2 zsL-$wY%G5-uRLroyQ~fH6SrM!{zD!Qe4Et+ui&D%s zVD<84awBPm{q;_iQ_Qho&G2P9jjkE?clp`d?!5=_2fTLrBl3I1?B_y>)ZD*10z@#H@k%#u`(-2K6)6!~Q|z zP{|(xqqh&s8TO~lOn@f_z^R77!+xbk5Brz7tjP-`aS#-+C+^%(gPZV6KLHo{7 ziRBFYTR)k>{YC3f5dP65XW0K%l>jb3D{LE#j%$Q^&wAKDWUBh8S@pnb_8_Jn z_K&hvqiXdAYXmV(gHUAH|2BTnWUn;~tU~=TGweV6wMP-w5a9s7FFowPRSUOkeS!FI zC<$&ZsOxHm{fmP$Rn{&5|Hp8sD9y0HZ!#|PMT8ZK9(o*t(!<5gL^JHKaj(9p2|^P? z;@*<)us?D@J@F6-{e4NQH`TOJbH1gj$#1ZG8W@WZ_(CZ9Pv}lG3WbyB|MR$?x|2Na zr|zt9Ka5N!AI4T=8P^~r`7=yKZN3>O;u?jokE@2eOrh&X1S_j0^i`xQ>*HG3ncdvt zFT~yw5sFJ0W2AMt-YS@fBxKUoXKZ1tz*v`=+ zFz%_~#h=5J8NK~r$~gC6Qa~;OaT9__J1Zg~TND;FX9T@6Wk67h5wQrxuAmiUj1o_I4`#4RY%cV?t03u$=(v4^#Q zQ1HSM8Q1Ev8hHuO6@n{KQqhqIQ$^hCCAK|WAfBee2u41_5tJngUQ{BoSdw^*;HrR9 z31;ai8q&rsu`iE@_dpJ!+h|FdOxo*3&d3vS@0KpH2UbTt{Bw?on`{5j2q*ZD1My{q z;O_<}h())&>2daC?*Vaz%Gs}HwztUKHDPx0qs5wHN-{rW=MKz8b5T|-A6Qeww+z2= zAW`gt&Tq(wNBE^O(mK4M8~14;_OeDKWkxh-xLx=!_t6RUgT7)2EGb7gc>D05UcypW zd0;69f67t+u-#{9E725Cdjq2{h8KZ9qEcXTpokxCe{r)k7x3fY$mdB~A*1Xok(Ib? zrx3#FR$N}AjEL_PvAJfXcn7g-5Q10kictkcIa6iS@P1NFLH^Y;IOt})s#or}kw?$) z78s_)=gN@CU6?bVDB+4^OA4f&i$k^PUEibxUm{Z!)D|~5R}NrXB6b3rNl@(8lyni_ zUWT9jppK}F=;{d8T!K0?^pFhC$f_ZlBDz3Bs(7;*?>QGi1Yfpsgl^!Pq9 zGJj=&AWL2ZVYV+x8GsvbJfAfe?*aSB;BS%@VOLQH$?MAciNN@$LbATVOc+<2vEjwe;I2KC zZx({}F4>}x3E{KLHWwWb+fyTQ9l@%;Jrh2IZnzo+Y@)$Gaqvmue)TR0r%H9P3qmKsk^4wm)wj3pjUNj^=G-8J)2&x|Suqj6EL1gw zfB%91JRk9Izow!ra+`#}rA68}o06V{seFPfAIrK+L1*%l`7Kdan?nBHjoLz`Cn{6C za48FR+9!favTb9aT7Ki$w%cyjI+bmkMaU0?BtDbTNW^bXAw6!2t`z=po{0a_&e_va z{0PQ2M+qbQ?T~HN#Tg(+jgpF76Ay+rTi;wpZC+!pNcUO&wxI0PKxIV1(b8 zx0zthkKtc!#AO84K&WF#Y@)h9B>tz!L{FHY?6d~n#h0TjIBQSY5+#NJ8e`zAWQMB5 z8PYzPRtpp102dkp7toNm_LLCuM|rq@+*Kp7y>C4VYM+A8Nb&Raqc>PclQ8>K@Q*K8e9wUP*b(6 zY`;@(sud_)T2wxsiOblouYOS@Io zY*7FOvH}Rzh}4EmQGJjNq0}Lf0dyZ@TrX*N!z2lDWIO!nY{-rJ_Z=nT21q+2CRpJk zfIVgK_}sseMBGNMl+sP{0q#vPBJOc%|M^s?=tLQL2eB(jf{Me7&Z3H1G{b9>rEf8@pOS|~* zO@u6i_-Y!HltESN;^yHBz+NCU2H4UNsNlsnlHcZHX@~E_29-WQhZsh0VjN{B?YJc} zqQUE>)r)u$jM=`du&dV_i@0LaYS*I*M!Nv_AegQx_b9mo#nmKg7!;ej_KI{mPgEn5 z$HAq5`pOS~<@A*oRKfCsg@`}|Ya>~v`^vFV4Mi+sD-p7TO8J0-ccGV^=e}60naBpz z8i5V^;8paMNrP;>N87P~IrWu)g3alRUJ2s9^1l5rWx<37+*h6lVcl0I?W~A|IAslTpYzc z9MR*oq%MZML$ErkP4i^xD-YX+f%v;1Zbaa|^2_i0tMoW~GI@~hE2|j~j>E6(*p=f6`#-jQta4Ps9q@Q0BLigKpP$aGg!Us)B>*T}u+r8DT= zQ(rmjYkc7-s!(4!G$)fyDSwcv9=#{iSMI;4D%+AtkWG;3E3bX5zUYAHE(lgAL8h-9 zliolKMf6w=sp2(#`KDA4p1V`Y@y0r=Sc(OxUYPANs>qaRu#c1PRdHr;#6NrE>Ge92c@69<#;{r zD?j#)(^sB9D?;>xuE!|7c(O(JmHUZ?=fp^t@lr?31- zTR*mK8Cj6$*tXjqzNIwVww;hxgy_C9>2XuE?kguOsvz2fanVuY?JJLw&Ben&{Jzyn zbYHpq-UJbi*mwl1C#lzc<$>6uUK6oRG{SGp+e~=-%5~T9ea&&NGbBq}cS8mlbO+rjX!T5VjgpDxRFa z@@M6nDbf#!{@IXTflW?d+5Z+izitqSXk`p37GW!QBy(RmJseA{RRPvVz&E)D*Frqh z6x_11{ZzSCxrKdx-A}~Nb%p#)zRZt8W)K4U9}uK6q^FwvY2w5lP`Xr}o*KunTui0A z2-2o>xez>0=i)db>RW$qF!Zfg!UXQ`a6cpc`#;e3pl{vl2d8gMpF*c^{SU7AlYc|S zT?Fex%E1=Sx$aw^#3)Xd_zshx2uvu_w|*-pU3^Y*bwOyNNlJ_9Th~38uSnfM=x<2A zed}(m(#1-cAt!+_g-CnI6y3L`4vf>cE*{a4Ir0M#KQd&~w@$d0uJD7vzB9P)TT>;L zZi+X3>#kT5JU|%{-y;@*bL;f2mk$XPhY2qOtRmrTp6Oe^k>;=Z)>*)E4Q~3@Aw#9I zs28w)gi~<~)3?t3qO6eP5kEy^k_y4;TmMzIp^$F?Tww@Q@P+!;2ajODwH@d_!#G9^ zr*FMzfLpzYU%>dwmlbv+4s)mIThDpA9D0YSGl>YME6Y7luI*e+;`c&v)S)$zBJOrb z?u(ekYz%fwgswD;Nd>Ix6Tf6rDOD$&+YmQZUC%3TV)=!(2fYMQRmrhE=rqh`8F~<& z+kW27PiqVbyO9h@wgxG`&U>vy@@*@Loxv`H_#kI6+k@|y*BLAh&+H6Fks`U-+bG{2 zpN3RDqD4P1{U_@tZqT)=_gUDa;P_)SS`C>`s8Z`5)*EOhjwE(pvXtI9jpdXH6 zk%l)5ZPE|!ssW;M3g#^fxdtUaU6m{5aim?uHE91qb^hKpf{M5X$wYOQv2hK$2(^xD z(B_CNVIReY2Lz1!eExj&9eyUhu0bncp_{nIHE3wvhOFQPUs+bg3VdCI21S==`kMcs zJFY>W^~}VVV>fvF5vv3kmO3B8w}$RF^pHx&l7CS{{TE;2&6)#ZM=fS-?hbGAoydDp>jrAm{i0o=+oM@ zL=0Fd4`BLw1sy7ZPeu-C9sif=cm-v2#&Y#wFvc0y`gmQa#w#c+6dm4I!C0hOzFt9V zv1RQjS-b)GXPP6qu=wi}3$GwFONw+HgkKDa%S(~q6|`|q3G`MSub`=C;3u97Mq9&THO4DwS0ijOGn;`<-129p90U49E#i0uxf&Zb zyn>Rl>oZBd3+8HKpQ6N#SJ2A$n<;I3fE_T}j8{;%QPE2NH(-AldE*sy4vmZB71Srb z7L%msM+|}y)X&Ly1zm91@CvH-R&6H9reNk1n{6{*L48-1RNDFg8)&o{ub^|W0ZRTk zU@toI+AC;xU}I(b+rZv)aP1Y8^-*<&?*#UxgKMv#UC9wj|5;!c99(+^bqY;Txa%aw z3^$b$V6aO99(+^jhmdO@TS1>4Q{-GCT+*LDm@WDh%joHxFK=8f_l!z zK;apHGkpkB)i3Q8RCjHbkjueXrCE+wP`Nr@6LtgJPq=yo+AC;xdvyf9{1w6jp;}!JE->@)1dj+kp8pBMP26|=@5$zSU&=G-GP*7M!X3GAck022; z&|41LE9hqXXkkqSILnuyy@Ku@%~0vC0=|(rPGM@VpqV#=R6^eXKjEv#cm=(LQ<5mF z@*4i!ArgNs6h6?;uDrdxQsOy{`j6l(PQ4juR=pr4UmI8h!j(MZ*bEf)R-TgY2&}6k zuf2k1M7I)`$cz37VB;Kl?G;q}kp^NA_J7Mcz+N}<#w%#y2pkWv7V%rPthcUcub@jq z(H0#7c+wEgabu;uf(E6bNxKU0mL?dlprIRrR8vyoCu{;mP^C;wluE^T1uYm9rI;CD zwe)3ja>gs@?7Ojw*$=GIzD#Z;jaSfwb)kwm6Rc&vOsCN`UO}Cg2Ppb>Q1|)J9j~B= z4|_@HK={XpBq>{}vmN_|za3NwUFo2o(LqH}`KIrT;}sP2C(5lFz#5w1YzFFM!%-&> z@egTCbr{soSbGKSTvJ_GkAv})56kfi@^r)bEUy4uYzXWXq||7ypd+nvh4m2_Um6yD zcM7MdHv=6$(nwgRLAz+ERFaNYP;`zu@Y(V^gO4DBw>p=8JVw;-ApKwct~?jARUTNH z#<}REde7P`=!sdXq^$z5dOV1!y@J|i#jB3hQ(#RcrfCq0cm=)Kq`qRl2bRho_oR$h z&}y8vKpC>WAi_m{U)n2Z`%3s=ok0Arlms^y)YUXzLE|s0>So;s9)!-f;!sfvcHr@@40t-x;~6+>=36P~kzzFHS2sBUaU3_@&{%`4?%s`8z?5nV00RT=n5K=iG=G76xRxJ zL={Bf5?K(M`;znqiZKguJWW@Psh*s0|9}KN2PUIK8#q;ljQj;7By#BqoGC+pK`tUc znrw@}EtpjPJX(x}@Vf{yh0@xDClRq=GrSRAmZ7A_O<)9d$pe?jP%056{-v~s-^gR} z-+ja-{Z~XBACWBr7s$}o*dJ8osj8Jl;P<%RGrFZH1#?tll(b4QlD_wWx1{~tu;zkn zNF!2hDvrq89>`jjz4vAw8`|1H5!BxgoGwGJ6)Pnop2+ctzz<{~8O^Vl9&o7`)*GT@ z+lFLbz;tLAcR2pjIjqXBn7$bt7%gJE1XahgsQ2mWRwA}5|0l5Cu;%4NY&ZT-@roo9 z+rz35UR&%%(gzW&JhDt7eXLa#;17VUA)^Wwv)Tl+Tu8YY!g{NAuRAWOyqLTsuM?4Bp~(~t5kT7 z=npNAlcq;GeMVle8ZT}jCLwm3M&wWjo@PyZwt`p+=zRk};lMMk1%oPx?SS?fc$@>z zwtCb_5T^kB=>tnv@w$8nw%y*AChls~TR{^S?@JT09XwNzGdJZZCUsVfhz*zah50ZR z3DiVnJp^kCrKVtyw157!wjk@<0(ywx*9hkN8XG2~xV}<+Oxy43iP%!Ow!Cj16Q4!o zR3ngmIkAC=jg$7WbC@(;jQDp5JqnrxzBr5Oa9cb5@i6^(LYSUurvXTs)n`LNu{>5$IprqnZTMGoXcNX zQ4a+c+d}jNHq_uGU)alz69<(vTfS1rLg6tPd>*B;wd+V^|c5_fJv85Wl#H?s7@c7DrlYRNLbdBmgUb8cVr&gbkv8a!!3hVTRB zfgh+2@k1&7Fc1%wNl~HsoJTgPNt=~0p~!sB`77!$sV@jBGM_VWpE{@4%;#|S)O5@a zxE4%L5Jw;uc@d)YQ0UJ)RQ|s6IrKYl6W7e=v_2Kf3OfFm0^j+ZA-LI?l9~7)bZ0(i zSaVFJ$hqJxLanbi4dC=6H%@Iyl&6)J5rsnZlws$YO`+>l5MlZs609Qi9& zTn5YUlFpCA^qthX`whCgF`!lbU#c^yv&oIlNfsDw4C_AXy|T|t>U>N`*YyKqxMulI z>IC78=mTW&OTg!7j^x5JlRE2b#53gEOf!%OAa3Fl7P$bkHKsq|UGF4I7g> zfe|&CB*%mKEU^#MdpDChUCPx`+76) zlDr7!Uq+jm)Vb@hF{yLXlg}g>{U_Ex5tMCaQYT_^Ii;;0uttRQFVsxxbPcbr91cgUmM*b00{>`LLkN#MtOGA7nVU%}n6r4$&eotYI zts}s`J_Jdvh6vM>I$5RBEguKQB+YUrb?*IRD-&J^_9o$6Jp%NkPR$6ZmY%i%``qAq zQfJBg<={Vx=${F4EHslky}9b@Nu6+i$xP|~3(Y%1VG%v4GsF?Wq)x4m{Fx~;K+hu) zGSFKNdQxXq8N_!7IKY>nCv}#-9jDTr1biBC98>kAPVe6nRYJ>vf9R_xpjtsw^nLy8 z@qGf7qJ1EIqZK(@z2?o2RZ1=b`%5bkYAe?$KfC5nm6d$Z-#GsiferEEnAGVyuZ)tf z3M`dy^*Z#V&czF{f<`W~Ij}a4yq?sV+_SPcPWT{T!;HL{)YOibDo(AKhVbOQSnbff^Vx3W3#riLTdh2GZSHP4= zVZ|aoNn_5(A4$BNBx17y?5oqTA*=})t$bL{q|S;>n1|>CaHJt{Pg}JHdQzvsU!lT! z5sbx##pU3Xk=g3?>B$_-|A4mLP>JPC>g<`R*8i+yAe`1DXRBBHYBjk0tXsg`|2*J5 z>q#AX&7+u!VAXmMQ%~wN3rbYIt5#rjC#Go-icIP}wWzpaJ`I*CQSM2ZNuBFMn<>IV zBAny*r6+Y-ybyv_X2fryB)GYtKBbw|>3OoQGWa{-rwxaS(oE`1nj#hFCh$P?43r)& zZYG*ZozKsd5^*3T8xr@Hd?$5!CYKaVKxpSn;@(uu$#w-IHr2LHH4EhyViZUd5qOcE zx{RursPVwjQ>l`VgE@T|^Ms4RT93fT!BFF&tmI~d|L}1zVGsSBB4S^%tQRmrAP<56 z8_5hj+kzXMqB302X6V9O$#}5t&PE}ohxJ73mzxrN1>J%8Q_*!rZk-rTs@^+u@uJ>7svLd%ZL`5Kh6%UUnCvqEw4_#JGbU?(T2vz{LjY)cv#1Zko zWEB^=b(P*I7HASauQfJLPX;_qONeOtTIc53ZF8Ck5|qpE=Y4`hXdhZM9juqz!XEfb zCMP3z0orHaaeaq8Dso%fJ^D3L@L50?3=Gc(L4_~e)L7-;nc7t3?|;0m$ba#<8p4ya zF9xo;&0QhYZeVJu$tdjcYT-(t|8U9CjyRD{Zpq|nGw+t;X;bnxGHATm{eEwZxi3id zL$HG3nJn2oIZS)nbm@yWdpy7?nxJ~c+?gdW_D)?(iCB_*6NFWoq_ij!yx1?yz%&h! zc7yP(A^CdRJRKb*nn_Ib{0_q3L@Gy~(%REz8Vq+lZN_#FV2+HwiF$yb$i~yAOkP8U z*8|qb;M&uMys@R5;*FT-AwmgS1;u{0-c?9;GVdr(O9c+@R zV!ZK>H?+8^>Lt?#P4Zp;1A=;==i4+jybTe%5UdjP!_dC#x89KAC}PhN5>3CsM1qZa z;U?XlB|H3;GY-{V73BO`dHapdHsqHMeN=+b}Nq z#&P}0&uBc~4+4)NSZB!)?YRE(3K%d6(NqT+lJsO8*K6QE#&LaeY6bBp!qh zwbe7bxrSJr>Uk#q)sDIcF z-Q7&w1{4I7IAK}THF&t)w|Q|cvhv`_wj`}x(f2ni#iiAV5KgyQ0bxF|YGXh}ekePZ6v>1Q{3e z_{|C85TZ|NNEMrLF~9IR`h-^j-ZKQ&?YNk~a-^vsOQK4XnR7}>d5&{23HL*=CHXDciap1lzK2N%?=H))!?VOvs8~*F zMo|*lbNtQCF=8`fztD)BNU-)CZ+`?Gg5$t`G58A(J}G>`JnTxn4a^^@v1a+Qqkf7# z=$~4m6rd^wE?Pa7+EE>60KlV$Kr+I3XIIJ1$6hbU1R{`6b`3#gK~LU2 zkd%W5loFMTXL%_cr$VJvAJV3jI5VF2cff6pyt8{64BpxP=pmR-f=Eu84i^LXn=S-L zxvKj5n^Ip}xvK80RszEwc2%WFksSFc+;re?x*4Jm@Hb7ahbbI767iZ*?Rz954KTD9 zY=~CPqmJ~-n^Q&cn*D?0TzJ1Rjj>21quPfiikcI{QyRmiFy3;B@(5Okt*~SuLzC>4 z>*$oS`iRfbnDob?l|sqhvsdlJ{BajxeGT4|6oj#YS~7f20ve!k04Ey)cUrUPz`>fe zWyHcyYl*%Pk@8J2Rv8v2!_}u|D`}^?8Y$Lp5WY1ePAitv8j9th8&doV?7G1Rlj7kx zG*T4Y>t0n{6R606D1itB&Z+8)^V^GjnT85fwE4fpCJ9f46uK=7w;;460lo_Z&dEzE zN2!;+h5Ch_Up~j$9bE)5+>me)M3ug@v#-N2K@TE5XKE`RaXpF?$n}V;2R61yJ>q&A z!VeyC%?DyeTy5@^6uS}e1A_G(DWE4w95do#y?Vs;P#C7bZvwujCCrHHlQpFT3CfZ% zDh|Oq4uxjKm9POjj0vs-D8s;d#5MYTKLvLL)YZT=;%ZZsijj&V{nhPpobPV_PQLk! zQ=RY8kbH2yX96+#{*oKIMl#u@j80)lm%6ws3-j(KgYSHHHgnMs)c1iV^HnAi3SPLL!)xf_4>6Z{ht zn%Cy}CXthozX1B(!1}dC7t~g;AF>pLprq-w)kNEXtA-$wA6d5=K7?@5j;4z`(sAf6 zFWT;i^-~YO>W7c?L%q^osFZ#fh=rdMNZ~@H;5%A`81P%uQUguv_tzU7D2jv#@)mBnWOo z&BE@f>WJEon6HftJ$VIv7Iw{=?uR4rg}Vw?pxbMMWSI$6G%7{Crb|^ox`}7jbdO&_ ziZu|EVWgiUE6tkj{7UF|%l1Hf7{-glFl)N=mmr~00G~GmQlZi|Yr5oAO}4@Qn=`L5{>U7R6iki|WK57ZpVg~cCla?}s&US5%&1K|}z;_^}?IO>1$YK|hU z0Aal$kxkB;?wSd(@JoQl48gorXH9p)S?oc$490E4Vl`$>x640fl@`Z$hV8A-Oj#}% zeIo=_WzDS}3$7bB)^vCN9L6Ns2Fy;xo+;A$tMA;NjT>%K(Fb}If&K`YGq4c?HpXM=|0n;iNYTPHp0R6n(p$-EtLLg zz+Q83y{7wSsX7W@1#F#z>owi>^U8A@Aol}1yF}PmS{pt|r$X6nI13}bua^ ztmzI44`Ze*9_A9I5!gVLN|;{LtrLrth#CMhd)2=8b`&YJERkEN7s1GY;man^KKJlR0Wp8|Hyk=Ks;Z}csz z(FbuH78UMPf*t+9$0z8`EwPZ*K{8(87a7r(g;|#kvD6)u{eW3c0qh! zE$gi-dQCU`&tQyy08Tc9*SRw3HQk`?*q*ou;9Htt)^xvYn5UYOEr7n#u;gDL*GRLb zyJDk9F@FK;nlF=+Gi$o-rv@r!vEnX~fS}$S>*Pk#tm)3kPFBoxunK&cPNQqqbY;sJ zMIQj_NFTbhrdwuxZAF?6!W%v$&X%*L+w)U&;n#z)&9LY@hx0LBQz zqVJB=-4JF?Hv^tv)>P1D87j9w=GKm@&#H6Vt<@l`*Cc06xBP`%E+Y1VW5f@>4uIhCH@9Mq8G0c<9?Ozn(mMtQ6dV2ioPW7P4Re$ zdWP)zNu5&Rj)T3PGwoj&uqkPkqVd?jR4iPO4ZjkU zK*eD^_FvBlSBBEXhYT5yeHua=k9{(l-!(mmivk|*QBQk?wugCuG`E|v5K-&!5gka;b zAN8%j_}XC22*cfsh|nJUU2$2L@z|%E8>@59suRa}?Ath;wiK`g1gn74_j>GCot-9X zA-0i5%g`z0^+>|5Zv7fu8q!^3@&LNWS)3k%pCDIu)7B5@;4s) z@#iB&II4D81kQjY`NF2cW53LZFi{7DTtlL)FjK^8kNpN4(?nMg=sr`{q+XWs*uO;8 z&3Np;{#&S^-^X;Y=_*u}wkiSbv0wd84Y3;7rv@j9A|CtKZqGv+c)m+n3lr7~Fl@P3BRFrxF?)jBk{2YlZd{X#N<1rQ;2W+yzFFNX{*n#nxVj-ZF2L9g#xHz~0 zKZZ}nE^y>uB&`?V;x=asGUpT_oUUGgOGhR-?U?J zF*pbCvLUc;XTf)ITvb7qxM6N20w++|04(^vm=G?i082A?C|O|^d|P%(6IOG89Swmq z*KrF@5O!Zm75xxDTw_H`P%Q>?5ej!R*qJ;ly$RRFuW>c9u8XxkD=k(-#d=Z`LCNd7 z7%{4%*n`+Z8u6`*#U{cd`8Qx!3?5J7x-Le)NB%EJsRRORmgODwQ|w`>iJ~H)ItKpV zx>)x8y8K?-fFrAtw62TCW|!yJI*1TX_kne>PHuHE8O>ypwwkI>pg7VW}q~IJkfK?2E3yACVok#LSeZ=QztmvC?>a@Pejjs9j3_k+HX&H4A z1XT|h$L;d<-kyZlY0wPpjlvjT?Cn`#(cxI=IqxtSR^8}LaLdj=HqX{bjQyw&{+l6T zvE!2`ymyb&gYbml_&+9vBxnmj+N6vzXR&E+6*p;Akf=oMpCc3s5hUn1X}mQH!=cBCHcPF`I)9v5D23aL~cg-$W5VJ~hP_K1%HxzK-x zVmXu&)rJ0BSWT{L{Cj1@5X3&G5vS0-4TiUMFJBKUPxHZg+b}r^Q}PS{&JmQSt-$yA za&*ZD-)tdH0Q%Ly)5#W7@|{kW7uH>XHcF7~RHbA}-s=cM4FZ?EE`GXP)Km8+wxm!~ z-3-#E8RttyJrVap74~RhjIR^*{P=4+Ff`*Y8Vt=irNPZO^$dk)sjLH1asQCTAXVRv zDHgQI=NF>G++%deF~kMGdb)uK3-Wqg^05h94!08Lzzg=v!X1XlY4G_u!NT<<>1%u@ zRs?1~DnnXTgRz~qhKS5TGO;`UQzxurU6DChdVYgeku8^DKQqqVeyI<;j1`JmRluj^} zisa<@{?E*3q`mmd9OkqG-kmsXvf}*M4#oF8HgexmR*VKd$>`ZmuQ-M|)1|#6qZ#X2 z2z;5*qc~&xL(g1k*D9UKoNd6rETV@wOQaqDL}lij0sco3J)9QqBfWY@*%MI7B|;F? z+g7Hmm3Gj-39O<5plSqDmP8Dvuvyx-W0e)zfSNnf4DXco^c!VZx(}d%j`Y=z$iseV zhu{@c={yT)x|Wu7m#v6l&iB%;IVF@i%Ym=eD!iO`aPWNQDQUlQCW$%wfPY^^4|9H# z_N2GMnDZy_TSfFdx3`kW{0o~2unb6E2PG=IL^Oi8hA{k(w4eUA48t`5)g_o~NbK6< zw#d9I?Y)gNMN33?BgoVa`S4<7N0Fpu->Fks3fwqVW~;8%!K#Htu3Mp$<3e96S% zDp>kL_#exbj&3P3qb>Wn{w^+pG>{62$QnJAG`&bp-)jVm#98)?spsiTZrg)6Rmh}%Wn9p`fUW^ zix7jL-fB_eW0u{kdoF9N1yWNYa!EKu{)SPbE8O&%T4K>(cRMgWw@PM-z?3#t$dyzq zC{hER(%y<9u50*?>3Jfhqg8S+h{6@n>}EZY@~~ys+*?t)%ecf-rmFUci}Auy)D zXDt1QW$%b=B`N?+)dcw+X;9y@l-^d_eTD6?S!(2t`oUgTHVo zi!3|6eJ&eN5l|Ar^xD+tDdkN_ughUL2T&U=UFZ{*vfQ#iM6*sMI1u#l7fdlH@tOdTch#vYzrEIY524~b>vG0MOAdadb41a3bV{r;8 zec}EAbemwVA^dep`NFbGB{mhsQn2cQ;M5M)tSHhU%YOMjKuZ4EP4( zm^$lCkcqb~d$nDgi3dSCMZ}_(QPZ3vUG~BIWw{7$fE13lNNN04{}foQS2v3kciCf# zC9+6W(6dQ|bw5yq+TN5hE_-s%WR-u=A2+E`)8*C8#0oCE&!tc%P6lbN(Orm0jdV&? zm;LF9(o9?f()I^wWMYcT?%%rw6OV#){s9_;g@3S6jWi7bmRR)HeFMz&juws`%YJEhNkI|vE&N%5U{!w;g}}a+ zNuw;g+%U`t$?X8XATXUk_6|Ee*W?GPqf_8>nTes>pONG-3hq$kBH zPtK!B-K-e)>u?jm4xCBPS$5VgZ04(tnC8BMTZ-X*a2kp_6HRY57|{!~!G_x0$-;cg z4tNhWK|Tv~x?!{-hElQ6viC1URd^fVdLKgJgjQJgrgK^Fe+1*0VNpVbS!*o&+GAMV zzYNBG!@9izzer_t|Mnh{v=6=SZ$fbYZ5qCR2wbYFcCtri(gDj}+_tvpMB3|vkfTYG zbEZg_EZaXfN0EAhFvyTNbBg5evMb~kSET1bm|;j{l=9k43U}GBl)~ngWdJuA0>3CK zZ!fE&%MMP(ukb4{elRRn!@iwKy{$68u0>Ay5ay(6F8kgiwU{aIf$mpZS0DE6OzLA* z_-?&nqe+=m5<7W7l1cbelh|An*>^LkmdkF~I!0-039KFATp1N!*JV#UT2IM825f|p zXW!1GzE9U_4T8&9^FPI07 zZuZ?wYU;9E-!87Sod@=((WdY`mz~kSK*wpMslZ-waPsX;>W=T# ziUR(L$`!yqaB%YNOnSs+uNs}F@V&qeI`|8bxgzOtm;K&Ue}$h1_NRlB<7Ltam%Vb6 zI`>`%)aE>p5oF=#r=CXf&G*zWi4_G6@`M1EQ=_I}9vTMDlicxk3*29q}&&#CMIMDo1mXd!8*waQn zj65roK6cr0=y1x}h+k-A*{d>XlgsY=GIkJri1;lUlT_BqJ1}X7%U=Eiyehv1_@g0E zOC?@*UB^YKNe7X=1b9&01}f=Y^B=*kjY&sb_T0+eRv|JSgERzH%KX#k=0-8^xop`3 z`=HQt$VOnb_hmY*lcI+#qfS{{t5nkB+oPQ%wR4O=efG!QZi@wpCm=6+L=l;LOJU5L*}H^lcgSU_}M zO_B7FpZ!8HWurA7*sB_6^VAf<5!4g1+E zo~VZ2Gr+@!pfd4HST#|^@;i-E$r0Pr6}U=0C#kjwh3T>cN8tjJwlE35&W01#BI z4+|fbB$7VyvzvuCMBfkb^);pr%v8y+%V#~8e^9-O!fFFXU&G=`Qm9yW_R0|T$GZH_ zK8iHPf%dX5l@$h%4`xy^fBR0Jn;S1{8CV;LY1$5Tu1*-aQ6`o0w{M4q3+pQ|e(=#J z0$v$`h1$6;{{^?3aJ{hp0^!bsNN>9Q|85hhNFMa(Dj=vI0B3^VNyNQ&DDw7J@(y(0 z%4bY1i8k7bVMF~n;jV@LRr;FJ4S3&Ji-yj=vBeQ+0`+#kzDVf7`^F{$ zG5f}n4tT_BM0`qiRmU~ylAa`S%)T+!tM`qCmcbs=(|~``5@z4nh{_3q1m!*a@q=Mj z8Wfs+V~@l(=VW9YphN@fePiWf%PBYmP!j{wzOklL(Svq(hSvOD(}G3Dj*0nd-`L1} z^pI%-SwcDQ=CG#vAzVMqDdUAk>xbQVsC+4&y2_CeqdX#Zm9&44j_10%1qm0~H#VYG zX@-6vsK~yt+QpkPX7-J}n}9C2**6xSTVI@qScOIqrH4X)YAi!x`|ca-3&Y&RHT%X6 zyjhnO^!qObzWc^pTN^Na=6}$gePh|~RCJQx1#dNiRTsuM`^JvEi{9Z*#2?g{Y)AyN zZ>)fJ!ng@F`^Jv$!zcX)V(d(>K1okr!3XafE55if&VL6h%U7l;6%D6&Lh6lUR0oPn zH!;<=uZoy_nu#_3!H5}ca}l(Wp9Kx znJfUKhhfq0K-p(@j(H}aAVz^PQM1(WBATY)>hJddtxu}xNEXir{Q~@K+uvm@RIrhjM>hjUCRI%7Ow&2@J%#__g@2y3g(?FO0YuMO1*0@a! zljKWa&LnnMO5E%mOR7>^X@-R>|)J_KlG@JICHc^Wy9r`)6!*Cdr#% z-ZR?F&aveV8#~8_POZTtS+NP`{}7aIX6M)!@2MnZF0ht_^Uu@l9E+|OspR_sd(4s7 zJIA`7&QP{b1~$#X_0F+c_p2&=Ij~g@u6K?NZ?*_Kt!S&9u!C$pf_^-e&Ik?_A z7WQluw*oStDP{r?y!q2R$MQ}!QFujQ7*2U{vvVxsAdcB-g7{X1QQo=wI6KE~M}yD@ z;7A{WRP{@r20HfwJn&uwV}@oqJIB6_ucl0R57-*Qxq1ZXonxI|Daq}+{1VvL2G={s zKK{8J{2vj0nIOkPvvVw#tFGQT*5KP9X3CP;E>RAlu!!C{R@o83&awGzf|)6sgWi!u z$Utv7=$&H^EiNOh!2n15670l$C`cMroHz+lD92kT*9CO49bndWbIN%JyCf;H8b$sb_;QJ9@$?|2fF;<}}d6R8-%p zh_A1)!k-C!8tA|$!qEQ*qpJ_g**W&yerO&JaFQYL45Mld^vNs8o{9X`r)lY7lkGt=~ZSLzA4HV?Re#=kl}cW-d_@LH+KyMyU6! zcaF7tyrN>(0ITtXm^5mMJr-auiL9ZzR$ajwOia@t6gdqvHqu`)r-7yN$9*ZYb8N|( z;)<|>2>tne>78RQH3=5hHpG8PNpN#P{Z_McY}V2$%HT7=|1caXO0#q9k5pT60?<8= zLQs0RxS42nj(t$6uBZk=14H88lJCy3Co5)(HX!u!B{@6CsB6gMBkGn4cSqQpe+y@- zB$r*1!Vah7n>!Athhy@@VrY7Yw6&x0iFP>kRB0kULF`VANV=iYoc^`9CFZt|0y}H) z&LplKPDfUPgVjx7E~LhqWe-RF6nk!NmM96Rl7au%;dJMYI)ZN0Yz&Tkl%%!8X{!&@ z1)0;85KdP+oSwtHu5*?g=o)^IB6ZztF~kcUr*Vl%fhboc=q*kJDRAS_e^j z#^H3s?UtP2eu9P(WE@VPZlE)n6WG#kAux<;sN z0qob;2pvZWn@x#tgocUF!Tl3KMpIg%6)KjwMu_yd34HLT#F*M2e^4Fq;zJA6AkLoogVoDRD@TC}&e58C~VtQv6)b zNNeY!TyX&cdg3p%=7v)fJzJetX6S|Ws zsmd=4^8c$mo{Q7qE=%j?)6uxUypW*7RIMA?Cz7{}v3Rd!Tst`I>f4R%*3MqZ5Imc` zlJp<5SCal?_DbG)yrx)={vD><@B zrs#>-K^l?m9eA2mB`Zff4d_J!KkUFWt;yAJCLo{{2JY#=v#mb!ONp(3_WQt+Rj9p^ zQ@Yd=$2IEx-RQlNX~>zIa%A>OKL1<=(H996Z-Xx(f;E^@GkYa7wk8O&J_S&1f=3f< z_Db^4h~mv&$-lm>&ct4b>}LebUdd%svT^<~;-4jq(=vM{>(9#)^AP=(hU6ql%Z_@qGe24RFDQC7Su>AjKxmqNr85MI+H^|H)f z$!NSMXRlKBTJ~n~9OG^~F-Du|{e&QIg3kD~NB6}sn9*6V&eK3OX1Laqs2>cbH zrh|JbT;tu;z(uz@>D~?ye(-ef z5Flo{w_CSXVi6)%Ay_TZa7lWS#4*#otXEI>&T52{iw*+*PD_~S-j8YsL4xuR{JBbS zJ8B8_IpWtxr*bkf1U8pIP_UlvEq1b%f~x_lVPKl>+~OSJOMopic+ATE5hC@j%Wjlkg7IyDb{Uvo10__VXUkzCWeWQJ!krJ;bKo}S zFS{h!AyxXPk}L^U!Z<4}c`qIhaOzu*NAq1uxi21IQ`6GdtdzwjgLt6^smq-!Bq z7s=Wr#%B20!JRM*E1Lu8s6ly+7<`;#>P$b+dY8YrL-~=zz!>MtN@_!Cyy|Bkyx0Vc z*MPpE8S1B?teEd-&)$!{ROEqEAVmbkh7kvd+Qg?}+do*HEi)s?GP|y%IB!H` zG4r#LNd0B7CR*mKJeGNlK5pe<^nZZqkc<8fheLM+8sd?Cd1ZhAw*>IcG5A#qe8 zq5mtdDozRUeBYy*qN3{8L6}RVYm`(7KMB)AJk=MankV5CU>jZhxny*gnIWFTz0Crb z{sGug!TBD5ii<-$!`s1-?+X0?CXDOKJVkGZc)qKMRqni~*%&y+G?fZR`eU7-;_47j z(=km{d0;gKPoPqNgUgzIBSX~?GcuURWm8D(U5dawE}KJ&Y^a3C<)O{H7_F000?!eh z!+WDMMqa7~CFk(oX^qvFl}Pt}N;`>Swde5Ob70M(cfkLE#dK97*mHPaUW@iS1MITk zH3^qqZBR!a~%3W?K_ zb9i4E6sGzE8zp!)Ma%5c_3kC!v(}l#6ZH#D6F^q3D!N!uMiGhTe_qM}op(xs*EF7oK0|_&C z@<8%51`=Za^&V{d8iDnwW%5A6%X9##5_@{ zxYF~$T5g#&Q`hNANoaKiFDiQY&35vFdRW00s^+=&jpi}$`R`vlfM;2b>yu^X_Q&sAaK_-G@QA3j>Y0>)dy;?_2-#*U}e zF-*;v^)VQq35(<6F{E;9$1^Ugv7#dB!yud>QXxvKWgYB>+_?kie_w&$Z5#Oi)y5sv zQ%vdw4ALZuq@p%NUg@~jXa2gL-mWo4_SZF}*8U_)nk$!{7?oAIZ3gOfxD(c@D;}^M zFIst&6a7gVc3!@SU?bq11j?Vm#yFne8emHPI{X(|OeYb+JZP{!&38OnRe+DP0fZfa zq)2Q`P?aY+o)?>CD(5J`UoF9uG~`htOmjT5#%7ypx(`C0E;btmp8?bKS+!LgH4;_j)lLG{lZyNN4fK#Djx#=Al4=D~SGch8MVl#px+sKB#?xg6$+*5F z9AJn4OJh$H)GtD;`SZBztSR z`U%Y6Et^LKW+!4q@Ue%DZp!SY3#uv)O8E#J=bN8kX0LPO9`=1jX8+MAMWut;$g=s_ zWcEvV;!NCrPZo8IjKXcgd01cB4G-419{$7nrBRsOj6`JP;hbYszjjz}Te=aRhI#N` zZZTb#2zFRsI-sS}n}F>S{2bvjtiPQPv%+Hl&j}%LSbsY?Oz%P;)d*Ny!T+R)1BdmtFu2u2z!(kZ4(n8GGhpz5)no*|Qf7g% z(8lHWh=vV?nB}EznYt_thn?m03EvM|*pDY4QB*r&lgovx*;7DDb*A8zIVvOkFW_mb z@kfQVtJH|We%Y$Nis0H*S%ud`sp10%>9Ac-p?xSerPErzOXkOoaP{xN>Dp7l!5g%6 zci6wdIoU|ZhrOXa<)*@Tpf3EIS(IZIUzje#;{6zoTD(c9|!PN zA+#p~x4^(}w5MLDeCkQ~=v5$mA|#H*kbcpgR^?I+>1z;<2+1!dT%Eh)wl<_cLAWO* z?iM@`8dwUqYfVMJ$&bp2hvU*}O5J?U46N!zJ+CU5&z0&R)FM(3&M)0eKXA1Zd>>cB zQXA;@z@Dzn_j{yf&bQ0`ucf3HeXo_acRu zrq$EU_csj!=B0w~Z=y;VmGx_8ee?bOS1)u&3g`Fzy*Uc&5g#D*T1svK-iVwn$csU^7I-+hsJ_xroT z#%Q&X3icJ4-w=B%(`uQL|{OXZQR2OarX&^aA*T5cU!)@cTP^SULXwo&nN4B2p=G zeScFi|LgmERvBdDM+7_t=Z>$OAO04TZsy-LsWay6A6`azZy{LDx33h!KB1@4U7yI0 zNgu5qRl!Vt zJl=wHo*#t>=$9srD198SRr+Rq+(7A#fVbKZqV!FADN9j6eFT4gCU__!6s7OjgW8;o zz6t1_z*gzomZzD4V^E}`a3(YtO4#b6#o3tnW zH?yU?^zsQ=DsCapsYW8MVdtdEUiAfnJ)ofU!|zWDRd%3Z=XnJ75?2g6<9mc~1a*1` zgypC>0?tptZli-$9Nt(*UM$Y^KL0^?4LfJ7!P`-f0dEqVQ<)md7-`~lmFHdDx3dub zt1PB#5kU+)*P-6Emryb6Y&Hg8THnFv7YReZd;tRtJG}%L(@Kxy`KM|K;_8EL496MK z9Iye^k4e@Vcksy+l^myjqHCgrm295y9I{RGyPhcLOhOv>0m{z7%69P}*O8_pD6 zB6V#$oBj!*p*$Ehg+-5o$)DJE=G1@%=%ZkCwk-4Zu(q8`HfE@{l;a`5M_Z2O%Hqik z-3su_J|Afw2+M`U^<_w~?fm$xV`AD0!d@X!o?P3`DqZoxcM9NDAxNvbww=ck@n#P0 zi}gl0Qvn=}*mic@gSUs+b~eWWr^M7X@Tay7;@Wo3-79R^c78h%mt+$AY5aMP*qy1} z#kOyJK$U^fOwvDsc}!vx+s;In z4cpE$Gq58cB<(zh!5GfuO>8@FY)LnazVIgu`(Lef3~%JHJnApq$14 zTL%)XZKtZD+S~B~3)1g0R^}fF_1H0pfx3>Z`2u(KO3l6|SHk_|HtZnDwzHy2rwyMBt5^kOd zYuowGn#zhNDs6yulJH{N+4eDXg@N!NZo~TeinZ;W_9+On04@~5)7+V?ZD+MG=>4An z+-M16+nHy7Gc%NY59qXoHIF{rBgM8e`g3T_^x)qGD{P=`4>L|ta$?(=wO$!!8L-j= znLJ2}ZReNgVL3o?wgRhXAk!Um#kTWB+`7i}QJ_u@pesXK6YLqaJKd1p0by+biHqgh zc9u=Ym*HM8eiRnHXI$IP569p;`y#+=mf+qBkoO(zy$l%ys|Pr@YxWY= z7}Wz4?;9xMO2#f2%ts}V=VEvD&xCk7y z{|S{?u5IV4Iv$=cIy*twYe_1kXul$=W*Z&+^efR^f6lMK{<65~5!|gptZnC**-?fW z@jO2L;7m36^$2U*nL4PFnO#)}t1&TY#4@v`z*_;z{x8%pdxK>NJd+aJ&Mrxf4PgQi zp5tn=ww?X(CUfS)e+4DM!v)PZ#kTXC`B1FU-JGq!_X>x+#J2PCCN&M`4Djm#9L+?r z?Yz9)t2~%aJ_2VF<0(*}ZRefNWmF0XS%D;;O%*Bo5=>ul(|em;Av{8K0jUohn^w`3 z(Nq%+9x8J2^dfxI`=;-4^}!Ud7QyjN?=*Or4Dc|*fB2?%{(7T$_||rs`YLZ4DkP(N z>tMFa%N~Mv^bfoI?efYDf<%21)U$BT0z@r#c@@rvDSaE>K?uf>UP-vM%NzU}W*hV+ zrxQR(3M5&(yqEsQZrqv{Gc!HqrM$(wavtLt+2zp}j@ac<>BTPZ)6;oz_By8iaQYKU zYc77dc6k(!m%xYH<$ZA_o`dfT2(Ibh|FFw*W(TQT!ltB6iei^{E~L7mJQP5l^+u`= zvCC^WBa$bEmj%$5ni#;*KVkq! z|A+zHu9q^^DkS{@oU@N|W)0xJ7#yR%fbZ89(T5oJ8NjVBj7JR6C4r9* z1UpTb@^*!XHGm7nY(OpxpijdzE?F9J!~m|Yx24Jk)*Q|`&*}RN;MxtVqq@PjuSN8A z7oP5ftxQm(0ZkM5mJ82x>UgzU4CsFXKXBpM&ez%1)#rc?1i+f3Fa~fZ`^Bo`7WH*d zYXFyll6k2_VgOg{{WKMXQ79Up=Jdht<-UoB!C#e-P(@r1|#7=o-j^J4B+PXDyHVb`yC7E;*^*e!0p=RQ5%8n6grgx-c1eI%K|QUEU1nFyC^t?$YlW6zeQu^C_J2DaDK7KXzM=9UK+`A;=6>aoa?o4 zX>zSKkp5|rur-3pA|VaQ9GGVxsilvXRH03(=3VeI8TI3zCe?yBuFq0~5oQ!6Ii9M5 zjck%Hhn7%HY6edqVuYTz052oT45G}#AC>&Z&0?xa?cmh%Mkd+>cvmjUG8j2EsUIBs znK^&&1mN?zC=c!wS52~m=Y(hSS`mvkLIGZ`yxU$X&qeIX{e}edIRonDJuE!nu@APyWq`(TPm72Oad}p(4hzV zA#A7E+H}tlIw)0g>h?`g7n*brKL%Pdtv57z#&{X{Jx{$OQgEE9VnpkE(Maa z>i5Dbr|P4bxboq*YRdaN`CVwAs$zB(gV7l^W&CizrWC8jYh((Yb>?YIAZbnEeUM~L zp-v{&p~R-I#P~+)>oI7q$HD0c=lp>GT${eXW|dS&2_Fn>nBcOeFsgkM#$N|CS70s& zCFC#MQdVP{G|Dt<3WxEZtSJyx!lME5-Hw2?qAWPO1wynWuw0;`sAj>dMx2!3AbgCXK zVgbHx3Hk;lBx?%y8z;dK4B%Rq5P??}%Ryw*H}g_Glam7=9Cb+?i(OM#F%64?R{;KP z3GSN0Hq?u|rto5g(&{c1EN&z=LBctDEb=333a?#H;a;z^fYldV))eYZ#(l@M=I{it zr(E3H^qmW;ZPpY<02?Q`tSNYhq+oIa|91$ZX3K*|4TM%c_62sb{ZCzCwjEO*JT86KCSR1r!k0Sn+06(?_okfY5eh3@1 z-F@n)`tZ^EYcP%oixV<^1U6`&9tkq6KS8)BB#ugK(Dpo=S5Z-QEJ_#$=QN?D#0G8J z(j3!r)qrKWxV1rxx?Re&d^=!I2rf2gK^?G~{5<@J62^5UHfVJc{Rdk#omApOg2;1$Dy{U{`_pF=Byz*=1z#A8^zLw zT0v~4`n88_s9ypw&Vrl?9|Ja1HL(X7_8&l(3!^K=VXUEGGnIWbRP}-n&c6d=pRlMf zJ{D}IDpgAGvCe^UO;{Y4*i2pe8Lu`JN#{jLqTrmqlos1eRf|JTO2RLd{HYxOuQt*U zZPxT#Sb3q><9_hu^|;}T|B`#?sT`F}mHj+8Ln*%Ilv^=*U$C^OYEAg-z~%}ruSfN2 z3C2GLv_W8Q7fQ%~w8{b;)3O-UQ8IbIf&b+7NK^@6M@dO z4Bz<2#`Q3H-&rXW;~&6}EkO^Ugyi+OeRgpeLIV6j2!TuPJB~IsIr#&Ge}rUWv9HI> zqFBX>n1HH+Glew`36uAes2BJ3Sbs=)^#T#@|Ga_05e z7FcH&x39-Lt;|W^`UPOa1((<3?rO;9EcnkMjO$8Xk5}o?i4Wkv(PElP<-Q)j-l5b1 zU_T2!nHoLt^~gP0UXLG81P^&V=5F44sI-hL<74^t*klSu3aSP7^>}I|4sFOmtG9x4 zUZobaug3!gDx&6l!GEB|^jsp?*W*5DBzhvSnSw7STwagOa%$lD2l%lN0$-1{#|Gm? zfuicK!8jr;PDWmjXIs(h1gt+nxF;k|OJ0vxi37Fz)N|jWLNnRuD3o500~UMA>I~WA!=C{otlq=|iatHVo-I_%{2<#z`Xo>> zCyr!+*j0zG0W+;O{@~Dge9h+cfGNjK^Tju@pSV1_3OsoE;Ic%HY9-C0*tm3bL_eU1e zKT*;NZ8}s}qgy4MqEv6-biW`q`n}+w9wnHl4_Y>Q;>Gfv)(hkv5Q^fgYG&V3$J+js6Wlbd-?Kfp|< zDTti{$ohFc_nAqpJY(nLda?l^j1D9%a!HSSx&&eBI2(lJfh2Q+^lOywXFNq7(Xjso zVV5P@_Tr@ac|N*=uc~7J&sl=LNtrk8HKu>T8uud#{A$J zQ4~vY<50L0z|UBYDzxutj2d&?^WN`1lfS!wG`fQm3bI<4r^hRCYRt)~y}qYV7X?;A z@Ip^KQd^BV<2g{NhKao#a@O<)rh)kBz&#|zb~BglmWw;s68F} zIn0Sg-b%okgz_u`Cyr{r81BsJ_$*$U!0HK3v8uMmzQ+^xZcWu1{!dy=UJsRcvn)oi z%D!E*eUOEWlPj=u59dmRV8to1;Q{~ktBQj#aaxRcmQ!5gDG7V7M6Va{7TN^wT^7;_ zgxmG2@oQSBsqnz~55n((B)fjKw{{U61#2;rPvw099vRr>+3%mLDS4XMc^bin62a;^A3b(>!ex}82uEO9E zQ>hI`BwAIQF%z?Nvi1v|sUwykMp;#RV@WafA-E^tbmUt|>k59kt7;UFm%xXwswKD0 z&%ui$0XMj&fB$1uZRO1f)l%4$v`JA`)uL||QIv;1L@J0JxvOelJF%S62?DX+PI*+R zedZR!@S7Qa3SiUCw?b8X?dT86;t}`=9_!(pA{1F!dUeJ(^W0q8SnY%FAwr5%!Kq%% zTRMZDNH0}d#$S_h$8Y@5u#&IiIlNUH!bJRVZ~QF0J-cOU6+bdBh2lk93Iw~)fr`SZ zK79w)Qt_{QbUd0sS3!UbIA;%KA(pW>Jkv|zJGv#jJ6cH7FO@(o6+b=V0V2|cXftj6 zy!`Fk6;wY^Hb;Xp(FRZko}a(p?QD)lF9x)n;G>jOp@QGwOo9dZYb-6O8c^uZ0qt}{ zGrTbWvisFkC4x@?IwPTD=>d*k;<>v81I0h&jauL`=#vzCje@T=R`IKfo_-6t=Bb9M zgoeIK@Sa5pQa`?gd3Zujc(FGyec{S6rxG=f{^5MQYKbV@P_)Suqd{6lj~`rH^@i_Y zi)hN8?>wT`(Qkf(LyxBbdqeQ*F8+A*Ym1?Wz6)%v;I-W7J9>IQ2vU0h9T7P9c}0Uf zLo4OuD*F>0okO9`c|}7!Q!kg|+6qH~G^gtxzQcE7av{Ct5HRN2;WQK;M-(-ljgf)8 z{EjFpG^YY5(S{;wMM=mJMd?o$;WP#i)RrJQqG}=!<85)P81wxKzFBH$Pq=e zqwzK^3$T(9xNvSQnL8S}nv}ORl1Y94U?V=xs@a?FaaIQsJ*pp~7);STLrK`ezIrA(lGTKpqm2c z?jB1$1?OQY8iAsaVLg#T+vBWiAE>1$JLL)CbnS6gJiU2O?tIX z1C?+nPqb5%wo*p`DFo+iK?u#k4C6?i=y^EpTI&ixYYJl>Ma-Yr3G0G9m2X#8Ur@j{ zAaoVdL1e{8+7{&5(W{mr4Fh3__~bYco*s%Q2%%`((Tq-w&%jAI2&C8u+9H zjrJ;X__jhSUTaU0ZZMI@4Fakq&-9a zETnEwxaUC_X-Os)Lu#o#!*Q=LljeZ%wvddTNr$JzcfbPDRY#peY4m3x>>$!z%8Gf> zu8W|zu4Y*|zMJ+e(y7eRzk_&H$b5iDd~fY(TPn-o-gyuK;Czu~UrSN(8@O|6FZqXf zaaKk=t_N-30t<;WyUc`dJqkh9;lrQDosv8ADi7hEfpsUGv&Z^q{EIrO%rLWILk|Zw zR`3w|dWj#YJr6b)G&x!TY#HGkKdWT%uV~LRUt!L?0scEIrl}H4?}?wIJ->d7>FZH| zzY2lsRDGwpH5Y47?<{=U-2obm1~;u4K@;ese8W)u5?x?l`=+KLii43H$a>h^fO=QP znA<75BdFSfJSyR2q3SBTT9GN}X|f~eQAG6c9YN0mksU$xL#n8`@OYQf?MhWizcg`V zM-azrcLdeDjkU9VfWNaLWJge)$ij*O>fiC_D#5*|r1tEeh{>6pj1EPfBjF5ecLcrD zrGSCU0ZJ8^b_CJ)i}E%=jL$BqQ(HazLic!;di>=u6@_oRiukHZjHb^f`LA5IyA^dT_?NT4$Tjf1~>x7U%1o^h>-Xu^Rh^W|oa?~v3+Gh8x81qZUX>qtkB2iEp;&1)SV>U<`1_58EN>%3k#Q6k=;~ry z(|!cMX~L#hRcuG_;uqUrK^_!No%3%~exBtItc?@S(Q<0ns* zI%W$5;4W{3ua=+-)BCN!y%Lm#(B=3TPd%KsfznTQ4c`ySe(ecr1!IsWyN5UaD-woA zz+2;#EIoEXXz&uBsnFPkd9S>ez@|di5auqNvk+CSGkv{#pxW{qiPsck%$cd!uSF%S zfbdenHA{TAq^+eeG3O6Mo}ev~ZG?m^lD($+&d#JCD&yjp=w1*K&ViG-{~LcHVPs`| zg^h&&csQrXJj7{*g#6*i*<5(PV^~PvPLPD!$D4!J)X5;d05fa{vuB&c=TV^TxNhM0FI)1rA0>$Gc@ZmzjngtFA z?-LMQ)4%^AB$TbD;$*sbi(E|dBc`h}BnCDWa zlLogg!iQKUoNl4QQGWi~Cz>fH)q+Y|(7Gufky-FSUDe2+#Aj=!!zY3p*bSqvQ?FnF zgOX9Hf81}x#=k{Z;KgB2zYKocHU77I46h7ZH<4| zC6rWK;k(x&x(&lVD@JlW{(@D=) zUo8N%Qs9v;Jli=nv4q+TXkP%VISOO^`{}9_b=0E1?r)8M=b~g@Dv=ofet9KIjYk5J z?_d!G&Y4Q7iScjGh%iODPX$zk;MoL=@h?9O+S4zG)@r5>0Xr9fn@Gg?SIxkT z<1QE>%Y7NpE8J9I{5y1Fwkis&tl(V#V*ESN8>BLUH4~h|`)3u#zpJ9lscs+)6cQE1 zmy$L9{r+~6nh3&7OEQfm#=o&>C)fCQcXS0s&&LL^ci0fdt~bsa|GtFd2#)|eFF1wB zW&FEhOonpqgAs}cl$*d1cydb*xBD!6X`IX|6j9#n`2OKz-zM(G2+eo;P)7I%m4Say zAL1W3kTU;h1fN8T$ZyB@qb<#88k3NPTsyu~J~o?smJ!68OxzvcLmC(>GuiRY#WNeW zcWBSt*pg~3f?cDa^uzDZYO4Rh9pCgEc!?`JzT1y3#t~Fn84#AE;s`iDvg3RA_)w;I z{tvpl?T&ALH+YFBJHFRp>#ELxPeTd) z6=hX+eE(P~8U^hP^l4#yLk!vR{n9$D#*P3uT?iC~Nn3V&S5GahmVvNdNE8c`a_#t@ zH#%1B2kV!IFuA#8$9Kp-$ox&vLf*ASmq^_m-%A!^MKcbJ6k#1fzc=}l9p4Yqut--Q zj24z<9$vfSdv$}R>PO1))4&H=j^@hZZ+_bG9j9s-(kmdmB_ysdL&A>l`F zad&+Gb5PjW@xATU2qx(cV0Iz)F=}_&@!c_~j)`q3uu&45?D#JE77pZux1NhXiwWoO zvg7;PPk_me?|N(UGD&|4=GVmLXG(T_|LC%@3O^Tv0D#tq8jInf5kaUxpX23RK-w>!QYy@w+}2_FLNWf!+Q zzPpx{Hs8hVj_>a6QVsqou+1)RcYN=!T#5$*{UfkrE^c>xpE_I3;C}}t0>8RDPe}dz(^Pu4F_+FcV;ieM6dI1DYqlUZVd)_^)^S1}%Ny~C~d_R+2%w*vu zU?T}PRcLSXIn^ZDEQ-$own%WhrclqQ=bF< zTCN~=$M5_oUo zxLLJ3zU$~@lh9b;(*xs?9p66>sca%z3Bvm}B6r94={!|TBwqsi+D78;_#R%YkqQ4B zus_`JcE@*(-P(i?`2a?MaK7fSJHC(IZm#HyK$iuE>r3quVRwADo!d|?Av_yca|tgy zzMmTasqQKG_p@PreZ}th-tYkqoE#5uh7itiXRq3?(}O z9kj6KCyaZf?D!stD}!mU)0e@zAIRk7WXE^%`f$U1WDV9B;7rSLoIFU%j_;jSOB!Z9 zu-XMO-9cA&d>`sq&d>*fIy```WXJbyZ+%006ND83BrcY_<2(P#2&K1x@ujfnJ>%~9 z-q{ff`U!w%EWzFJy>}zdN4y39&<||^-5f{S9p9dw4vuF8BPoF8?)biaCa-ep0Bj`$ znu+^b*xu$-cr}(XdV%q>u(&DwQna`EJQs~-nhDwhp%Tm8@tx@y%UNd~2wN;k$&T+) zUM$toxPA!OIg6VfVcOa5_+A)_Gna|^FIf3Lvem%tY?yY(_m~4Qm(r$B)_zR&e2suDm*3ncMu zsz{x^`FY3pl$+^1LOcpmXE>1)Of|8b5E)gZBHowU)2U!R<%|JqCLBu%^pBAeOg1Q+ zMoNf1Uc0oa+0Qdz5ERExLEC1-&_DiRSb31o7TzJwAL`!XsKAUS9g|p---oGy)#Q1| zvZr*NMRX6>=IKwk(;XXG%x2+%tM zb0#Pu|Gf^B)tLUx@HshUWZY$U}vC{093W04#Ih=?p}xBM^jW@I$>3}0;5wPOZGZ+S<(WGK|n`X zhWT8Ot$?cw;(hr#z(tmz&6)7L83M=q=NplO&sl(5g%G&cVcG0@CMQQh_*FyU+LP^SR z;9iF%E0TGy!)1_e5s^xfYp(+pGuK{+xR$WIsn|2vS-h#P%0#cphU3!(scUtfyUwfV zhshV6u&bK!G5!ZdAe(V})P4CQjnT380XSqqO|{_eqqyE1<<)-x{l_wVUC(Ya+wT1n?4#IxyQ2CIKCu4+qqeX(A^uLKqwYuJM1E#<0HKGFI4apk(Pu(!MMc#kK^RY@ z!<3Zlqd0sv&Gd>zz}|6jyN{yzkgBFveGY7=;Ifb6t8rM~{R#f32;;hveH1a}%EDe6 z{tC5g%1QZk_fbp>e+27Fkp2aC&rUYO9ge!^r$qKqoS+CEvX3J7STz-OpF!uh(^2=E zH(?Z_T5$JKggj9k_AW?w4y9d%TF~yJXtoA6V)_I4Z?u@MO$57-qQ^8?ejEUHRB&?& zJa1vFqN9JF4vqf`z}rFy+((gAF)VCIY}L7ekM$xLV}-?W$&QP6 z`W07HB)tHHWkmX$(()_tHWVTzb0!(K;EjF=6j`OCfXt+pR9UaKs%}nre)}`j%Bv8q zRqhj>JM_d$KV%EO|KOZwD403G87Dl~nT9pplJHNpSgsSECtk+;r~$y1LYV5tfD@k2 zl!iLg6W{r-CqBNF0kH{h~d?DpWJ1H6Uyfl3z?XGBcrD zEkim8!bu@4(fd2}CQa5ijI^lWW3Ev6Ng;8V0;G9uhAbY}dlcCtY znhyWE7Sqc|Q-SR1&La}5n$w*-fzZv8+|!-^Lh25{4hoF8fDuqsWzv* z7nk6D(?WVa;dV)N^4c0|DLk}}LcB%cIJ_a*CDn`Zm^P$X%w$h_S0Tn2oS0#jR2yzZ zXS7SIR0>&Ay}l|~HACQ`aC#-BH4VSqB~^;YOW?zoR9`7tf`fk+5M0y0|FNXHpjD8% zC~QjFq$o?O7YEl;l!svCS-(%!Axo+eTVgq*#RZZjRes!FXcVJD*XU5nG%u9U50yG0 zC!7~OqjvZ$dfq`C_Qa!q@mm%>=^t72q<>`5^Y@ur&4b4hIA<&6%r1I%x)!I_!gs4h z^j?O2i=G=JYN#Io{Vede1k0jl+_$)4QLsaVDeq(Ou#29R5Bn{8{e=s zjwkY|8o+A9ImbDD-=gQ$Dos^u_&#Y7{hJFbHRNX=XdA6hH!8s2o zHCgm*u{~K)?n?qnB-pzUuq=9Veynj})=(!_54zLBGmyoC^+N5L`JvXJ~ zsFy*QXi28AWYIGW?c^?c*2n4X^nAPz_68dw*P`c{I;GUtz)lEGA#yExPTf#jIoH9s zFD$M{eq+*u?LNz18p*O67k|VodRF?zw~qM+Lh^zpWrTlF8TbeFA^vd@x#u4>;gd)a z`7L^G*;<+>ufs7$n+L9TLk4=^nC1X&6q5Da`9x*^D7Sw^$_eF1*IQ;e~ME5 z2QGTjbKoVeEP6)aWP1u%c4t6Xj*24)T=d*iFO}(!{s-M%^sJjjmc`G2*ALDqK|M|u zJwH2}4*eJYGc2Z4h#-re^pt5Yp|a@t$0>a5Z-CDp2}8ep0Rt92Wxcb?4D1-Z1eV(6 z%Y)`Hm8oh>igvw|x`Fl*Pu4qk{()R3!KbQ(u0gq!_0FSrA;;=wK-&qUHZk1w&b~=F z6@tZZ=l`$|5SIq zbHXev3Wo2-sV{KcNV*!jyvd)ecYf19L+Q$3)Uqt|+}icdlaa+#L&|YG;N2}pb7k>l zj@CODzn)}B!$FuNB(5(*!g^K-N3|tzOxLp8{-#gqQWsOlkyK?;JKDl}UO%n4d{( zvff$OWn;aw(l=F@q|btRQR0^M&T6kjnAn2$U=u8y$(yWqw)TXY@TGy3C!EVB>zy0Y zgG}BV0&D8xcD?g`RmI>vfj#TucD?iYl`3$grgqtd~>z%#3`_?-r1AA3)yWV-L5uV+*;r#(Y z)N@P@WxcaDcU`;Qc`iF2GxZ_Rf5{cZu6H(dgJ8WgbU++4b?9Eqyx};5zIw3hogGG1 zR!$PYs(}Q%-nrPDWzuZ{yaRFEtlIU?eS32u6JH6 zT*XAP9@u9#5_i3GYU64q{9#}x-0*h2v)tWgCj3oc_uTMyz4N#3o4(uVc*6}#TKH4#F@V1T29@F;gCyWV*; z3CEet1~}gmWWDovOW#nk7SK)$YaV^LN6LEVptsY_pz||WmjjucoUC^im{Q*`J^QgD z17})}1H0ZC->{Z)+Jf<{u;{(xpQ2sw z46a%Q%VD6sB2=nLcfIq*ttLF@b5?+`+LGM$&MA3Qxc;2ozz$j5^a#_=cD=L1+9<=k z3f8|5W7_r3X@x4AiB;hPSo(!?2Lv6s-gz`%QNwHsmMI_4q-4D_tY?BDJWYgl++KFQ zv%{AalrsYUlPL)vE@-ML>z&b8i^K-=b}z3t8$7qj!+ri z=2*aiI&&CtB<{gX!BuCz9RTB*{_r0{@eifiuddyrwbWF2 zXuSr6O@Sn<&U`xuCui)lm{n&Q!6WAH{8;=WgZH}M`KdGeu;wS~j9Q(deg`-9TQt>c zlvW~sx#|qX<0bIn>P(GS(l~ghfZ&?`{SS3!;qDr0im)kZlcK0Izm3SJC=c%t=}oE* zQD^D|RpyLt6-d;X(-&f0gLlew(w5F%WAOe*sJf2;yh2ShQ%tG_mGoF&%?9ryuh&-9 zP)S1KRqC?viJ%~UVf1M+c&B7k>adVPto?-l=%@YML`bXs+{JIve&`?3e&`?3e&*$` zq~3(b0yt+CmD_4RALlevAHsLDMf4{O`?R0L3z6y@KtBn*kzmn&qIQOOnFMpsl-fz^a_c5(VX?PtPQ392Q0J6c44>%!BW9%VCC zKR`nTKIFnPojaa7YC51f0v~tb+0M&flvb+&Z3%!iM`5&|gV!3Xy%x=-{me$myi_95 ze#Y(fsMAQ`Uw8)pfTEnI)I|HqHz`h0?h^o&CiptRqW$pGK>nirRG3VfG+E1xvn8aQIF|?S~$2)_%Chdug1^YM7^-(SF|g#V6;yhLEi5P)7I% zm4SayAL1XsA@}^F8hjEdB0ueCPGoI0qo;=^o(qfphXvj}z@@@U;dw{{#NhET)SPL9`!g3mnS@RJ5O2d$2ifEqr!L82aT4 z7@+-l2@nCO{06N1{6W;izKm!NSe)v|Bx?mAem8iDCj!v@xA6ik51(`iU6!&c0??U1 zN}-_5fwmV$3Nb_g`sFmVihck`3W1_9X^Q~#Y*LDP9fVavqF9)eO91MzEnj|$lwN>-Z}KMs(3XY8m97FtZOaN2fYyDe zRb|R?d*Dx5j^@e|0qFN^TtG~u5g<$!64#dqK&e}C12U18f$+YNC{L~clz9T{lRE(Z zAOvYuR{+|$6Cdao!1zm89E}J-pW^l43P7`Gwq&L*dK4!?z?pt40?=PS3mXDZgJ#)G z(m7zZB6bbr+7*Dx;0|F*SN8!nKw=XC=;fa)nebDAy(ZyB09s0oAOcY5uBlAY8^HWr zViN&qh0BHjbo`9NBz+FdOA@yTK&St$X<`fh2_gubDX9oRyZ<(c=rX{P3Foqj094>! z36u9mz?!+Z6@W5s2OGQ>u)Z#C1)%R<4>I|i2<#OXw*pYeALR|c0@!L7w*t_H+I4v# z(7S={cX2BKU1_Neeh%0r!9@Vdn;*7?`WTkJ;Y^>W?&AtTR~NvFwj{vH0R&B>hARLq zI1I~*CSbI*ELQ;PmtDhTp*OJpgqtd~0?>hLi9EFHDZpk3ZUvxWnPIShg7+GNX#C`1 zONjt9fV-|0fbxG)nVI@9=%;f9u>#O=HwXlv6UWLiQ+to&jR40PG^s>d0jTv$@GlQA zJ&<4pplqDiL^;$gfp;X1n^h|S{aH5DB=kJ+k%94u0Q726$3!#-ghe(YR{*LtvxJFc z1F+9+B(4CouUDuEe+1aiZg?vIgjWYvi*RmE5rDeh!xH@C@b6;7`ud6$fMy=X?xP_9M+>0=cP1+UZ4SW~Lrc3D`{5M@#k0cd#XDu#IytSf;`PEG`%K7Tug8TvCU5#UVAahyCz ziU3r(OqO9*2CH!((;ak00E*Uy4ZS<4&jrw30jT7AwGC+!2yX?DxLB?LG;|eC4|pGp z4Z@JN$23%zgR00?^I=by510nAF4hbzl*IvY*Et zXcYn06#@%DW;Cz@Q1jCzl+zB3zQW@1->oCL;;b5uLU+c3HeIMZ{)FslTT|8i&f)xt0tKM8SL4h)tQlB_z%wZkfd06gVhGO=A&1+`3PAJf<;UrY@Sj3S@Nhx% zO%Z@T{ivqN@G{`<3x}#w1R!tae1@|d_@MxfW}*l{$(vfJiy+()63?pw1)u}@o2h(w zffj=^IbxETO(oXXY{NbxpH@yoFxtSeIzayzb-)DT)@F4e_RA``imsn$Z5Pb_hJrTU zhM|A_!!VsiCYpV+$Yj1mqERO~)4X|APQN~3I3zc?sLFVCKB9xjlzAFo7`HGRaYd#l z3St&_5Rv}`=R8d{Zbhc>rI;&Sf%k0-=|O~Bkttzgef1Kxv@U>n6W}y?1ej zZ3T;2k?B3e7&Gv1SO<~F)cA~_$duI7Ph@%`Zz`64_c_W&Uxt{k%8zRk%9g=Qd2EN zu(#oyHz{XU2I?^(O|6IT7Z%Y=8TQFQ^>5cyhXDN|@JfP32I@4ug!)sk_X$(pHt?`A zP}(nkGSCU+DUC~(MjVlWl0GS`GJ(~Db3WnpeKJtaON~`q_;$01{@jJ9JGbL9)DS?U z1>WhxGo4?9Gt?V^77M)Bg=ag0J(k_rBWV3C2izmmVmKwrmHVq!aZJ|O{&3IiFa z=6(ph&%=KxVVsu8K##>WQLn;#u7&g^N=#&+{QH8{2f#K89w-BOKKH5vzo^L`BgiB`X81DiyCrf-uFBa>+oEXeU<&`gA+4{DIGhU~je|a>+ouUv|{Dz)lHH zA#%w;gX&gP&J8dW9#B(4R|cX7+kKY3G?Hay&i9!BRlV+0cHTm0mUt*5{DaECKd2A! zkIP7zf7F6cB1PmU1GOpT<=6TsWFeOfbia8BL-PpAB?G;_)R+f}2@n@gWS{}nAdw>2 z1qw<({Qmqy^&cn$(R1J>uE;^P0gYZ8gQ7lZ#r2w@U8m6{_b?_m~Km};w85~*p zJ7~B6Kh;%${&xeFn(ueaf8e;0bOyS-$)6}deOh1zI~|O=mKCS~rLBV14CS~Z@Sc{V zxw1q7`ZqbW^RmP(3Q(D)%Eacmgb68}DXAzxi&JZw@a2G|63%541*qRk!6xsGfwgdPs{rl) zD$(H20PE-CRskwdqLhh$GO$-&+$um%-Kb^oRlq)QajO7*dAKT%0eTOx11@eApy#^B z8vH!4%YusnlspH!p7Q*G2OZAzdFno{0`y5^Y;Y(IuxbE7)2QJpKzj$mlBg*dk6MKjF~#@GS+|LID;k?QGhZ_V6%4$!0Lens{oBT{D?{S zQQ)14<7U+=Ku2(^C1ptu0X{k~9#MeW53Ok;ng_xX8DN}Re-isYlQPv0WPuxQGilXiyP`0oE4+D?ny6 zunJH_M_8G60HdF<=)L2gqBQ~9(IymkJAgJrs676N0(7HhmQjG-1!0XPxe8GKm(scZ zoUec#wYcdKrk$+G10p93Oxsph z&49;ZN_-Y}iD~+!i6bWraJ=?}fv(js726K@OB+H?82B}zgrb1@6#kqecnb)UbMuIQXLh!8QH+AA(fh3+ZZJg_|I?q^d7^*b-_{ zyK<`T{|i_VJ`pKk`v>IzlSaH-i|wo?+{fM@|A3}qDG6jN2Umih(atl)^Qm|JNt~@W zDE76r3#f?3Lj#k@qC0Qd>X#7O9Q?5vh&-5vlFg zQ+1RF9uaWP6smJ8wSAC3NR@zZibeDr4Evi%ZXJF0C|xv$E!bc;n|LU0$08ON)3QDM`5J4hGf{AEeGq_$z`1+LVVo!CIp^HBtkRT((0yj)USsp>eK z2Ut_VDMT)*ZNzi1xatN*UtzgY8;w5hv+SiYI;-E;@-mUyxOgJ9 z&AC)n#Uj|_6qJ7W{kcf>A1JlabKoVeNNv5#z=9PXvjf9&R2)H|)HXIh-n^i0{SUe; zwT&JNV;_A8yyI}rHR^F9wGAvB0u>hi_bjIG5<#RkddjqyP?6etW6Ogsa~Dhha2%he zU%r3=Qk$0m(c4Z-Qyr8iS8oe@t-R7z0X7r@ zcL!=!S8v<*Ct9x)7`=qW(TLvGAFl^jZ|gp#Iy3dFpub^*xJQg`xhQPtZ7=pJ!z8^0 z%rA&tm=c#Qceg$&V`4iF?6kxtdYdYyP58UO^aC4S^tPx_G=k`D4KKrB10B{ z6s_op(e+(hn!IlS_PL8&y={GDU4tJ1_Opvyz3tW77+B%0|G}U8E^hUfQ z{{rVr-|B6>|9XT60$m+gEf=?Xn>V|F!P^6SQgG4Rewl|2q=Vr246{ZJ&8sFjEf) zeG&zs4EpN9T3p3mY>X=!0ImuoSiSA5g9S~x+kx*Vj+<4hx9uKR!X$JS_|?F8L~na? zOj#3=R|lzlaK0Xy+_Hg+>l5S|7E;bck_4;*;ifuSPa6{C`QU1#37-S3l^fpbZNH$C zQN8Fszy`SCt={(IQ}q;oeoO`Snj7BgZH=N^s1o#f@;_i7N_f%Rj=u?utKIPb&W82% z6|1+M{h$=~KLh+z2&H+jvU=Otc?}fKH40J@aHjNyT4U>J&2horrRB^}QWj7(3u_vp zOe&(cz1^*%VLl30w?HN*Cwg0<9fb|^Ww53NGI@{`z3qz)jSX`-SepWw?w~7r+jDy} z4gDafCj;oN-gbCwjv?IwAvBNeH(Vn6z(^7ny=Pp#?MzdwlGOp&$P!$= zt;PUcLEjnveJy5Y4dyYnTkcwKhp0RjjOhU^_lVIKMx`j{ZGfK$f%Ov;jn&)o#bSTo zS77`sEPC(wr)c%I65xD~L(zV)0v5bi;fKtmhxbw0c{+4^|mr^;l%%|@V`Sz@NhwMO3~YzbT4Z%Tp$=lhcg_iO3~Xo z4XR@})qyt-;Akd_-nMjdhI#^o=Y+)bszAN%n}gNVSP*6fl6W@tb5PxN;VLRLZe>WM z^LR**imQYF<30*;1|6!dHXy_f3iWR9c;87$MXp`WRB@k%IHkWyQ)}UU9QbL=Q6Y5o zXxzpSXXEEM(-0#{p1Z&_BIY>Mpbn#hRov!~yi_wP&uV=-Pl$@!5|WiM4u?C0=2@Jl z>V3R~-wO$982N~*k8rG6&<~@Ca5S3hl@$>cIUyRO2+lqIsXET}9SY(U8;bt%h2j9* zUpZ9d(__;((9iix!LkOtJvI>i<0p@}&7)m;B(~JNXfin~v#lb|dO-k%=^JO1{`r5Q z-GD4_(nOr)O_m|NIS|?5Q=3&lf22An3sm(oL6e%Qu?1Nlu$8;M_>t#jzz5lt~qH9$D)$< z%aukyRLth3ndUgQoz5}NeeIj*P zmrVJ3*gaL$kL^8G+0gI!k|>tiQb)gpgY%wzI{X@*frrh29pk8s@J|4yQ4B!ZRcfUAWvlusf)89%F(o4L z>ZcMM0YOSY=eKYP_*}P``+M=0BGNs7()}>Z{JP@)#~N{T=hn((7^t9_)$=~Qm}C-e zi%Av3#pIo%shx)6g9ycJkDw1L=9t&8Y=x*_|Bt9?tHqKUN$ZG$-|@0 zc>Jp(EW0-d3sQ-9a9f*e2kI??jSI&S?lhd!Ybm0$vitP>c&u*1TO$}ldLZFec7J+s zN%bN;v@Qxl*+7z&-Fw$Z)mOKemE9?$F_#NsIz_U3eVm6aN5NIbW^vhU{9X5osvQE4 zhtne{tvB(@mE9>GFM$u2-Ru3E#liOk1lRQMf5`5gPo}6F!ltB6iXyw$SW-w)9-@$E zJ&vkFWcTA)ah%a)f#gChDj{oFlwp-|*TIVd%OLmQo!A`bQM&C><9qayzVIIe=PaiL zYp!^LSBWWWbaL!B5sx zqu@Q&Li!}*%WH5O+q}lAFuY4! zNM9wqA>+SkC+r4R-D(1CB!nA8C^G)T9F=lWJNp_JP)mBrNg_fg>*?u%!CbdJnskL)zuDQcOfJaA?^Z29_e^Kom4`- z2yCpyO&ttc(-!h(ekX>iLOCHHPx7dg7{_z+as@?w%6Sj`bvBGoYNso|QVKhsv7Imr z-VgA5OHh&AkIFfow(DA8M>qVhTTF9?dOlJS38pxn^ry1Ep@j%r>S4N~jW zgF=h@EIWW<<^h5CS<Pz`m@)=!#r}0j+qFaiptb3Q_gXr7a?{F$sB&!z=oY72SiM`ZDEhQyzmh7{^}O3Y6)r0O8l7n1q% zqx+eYqaDwC)xC!F1PD)Ak_tJ|y0}V?bv&h7#~6GBuyMKYqK;?yqcMtVLoWoj%Hnz( zil@Tr%&w?tR;j#1Nevxaifi{UXb-F1Y$ubaAJGMD!v$V+dTf}j-HplD@SAIwJXPw; zEt+z}G?$=(ZGKaN?o8fkgOaC8?OqzMBeeN*2>Os*Wt|&>T+tl@cR`Rl1pZWy;CqZ} zvEt=Cp6h2TtL(SY9&dy5E}YXG|G9SvRNPZdwI+NAu&)G{I|Obn3uXK?pz{Lrq>mEv zKX#q688c}n=74mEK$(IVin?WL)d_b95LLpctnb&PnmYuZi$#y4aDI0PT-}YS=A#JR z4$kR7xs^Kv%07$Pw(bXDs0DQoVwm&QX9mUWm{>;jq4MZAz*ro}k~;*>T~7gH9nkHT zVTKR6Lty%c#j(j9;Au?L-ck+t3ef{5JRZLFeKqx09 z6N|k=pwmzYwY2~?umtxGfk;H|-XRcstfqQ_3f33QLBt+T`H?#Wo@#4OH_)#Dnb1ZN0ii4}NyD4A=%O5z9B%x^!kPXHy+OhM@=mC{?UTM*%zAP$F-}hu7qEl7mbQA3>?+9?W1zl z3WgW81f;Ji$&X3Dxe~tB35@aOw}?MtFu8yT=1Tam?ikQr26mV6#e`>9x^X3Zm(`dg zkVrHDK`GZQL0?!ND+BBNomH%Wh?Egv#4t-GquL8y3IF18HO)!~;cX_VwD?N+ZEs^P z2-pb5*N|dW$LgZ@YaMYN{4}8R3_~|;u0R|OLpf5rU5OFS<=~_`M!*2BIFEMNdQiY4 z#Q4v+lqiaz8&ER8F2f4*j#Kqi12kXe=>tuy##4Z}%~lfX9*!?!;w zq{MxZfF&;J-~Skf&s>%yVwp`z>lAqyp7C`yL3wCMB!A?{8iog~#0~{O?=i%~@EE+c zJPcP4aMi7Xf_{j|PPL1u%{@t>c*y4a!L_ z+~a)gvAM^uAV%S;WKqQ7%{@IaXZb#~OeRgSDt-6no~Qc@iUo*WZV;Jh!BZXg=H?YU z0UctvngvgH_>Rpheg$-c;S_>DmP-39$Nun=;%^%!mAo+5WhWr_PRccJ;CYU74)vfU z29cE!98E|gZ{XR`w-&yLTTnCt^ftjQ2<8nus!dWnZ{X=L*{O(Q5jlwkcmvN}nm_s+ z@kaBBI0J1kV_WMZB52Rzs@k^RE97BYCkA@hR=?A@=Mvm7Fys+Rt2};Lww3fa z34F6{ZU0lM67OIWm-O#{*w!)C3X2)crlfU>Y+Gx04il7zRYW>L)xlF7^A_ri9%5)D zO&I%qD6`d=6ex1Caiv_Vheba`WcUb&GOqKBco^43VU}@ShTm*l=^r+(^bZ@?*4qO_ zHAJK!IIdFGjB(w)JVvxcY-fYW+X{9Y*FLUFVi=(D3_l>4jjL#pCgw2qgfQWJhEf{i zn)8K+aqSci+if7}RwRNB(UP9z`7GUGLyCBvWULal;HfgbO zZTF~z2u5^a1Igl)7#r8Af#pOTuxgCkjjL}vf6)wBXB)1SuyOT|z&48!U`%9I87mbS z*RlnD#e87P7*}=7#`W(^QDP^s9~mck&#c0@eu1m?u7L0-lc*?Cr(}$4*veYM2QNY{ z1Xb-`#&rYQiDg`mU#%+W9jFX;4I|-YTz}}FQ#1$Gm2r~rGOp8W;iA|KFeWq0GOqOY zTJNiq%vEal3V|YZ#cMwi**XOlG^0WVO{1h#nbm3DHq2@y8Q< zJ-j5qc>ym8BtcKGpAVWt#y0`^ivlEZPcUy~QoAW28lW0j>(mb7V~#T8yMPpXbh<=I zGtLI|Ayt%^h$5{bio`wSgJ$Qz5<*5U88D5Qq{Rbum7bX@{4$D4|3Ssjfsir`y8l?= z6d9#uK%-bJ+T{7E9rn_fm4S~w&nqbAN08(Yf-YuOSfmU*5uZg|qxhcGxXKpXd_%A%R=Obb*2K!twl49M3br52ABg>u^1~A=M|S7a z6DQYz-C>+3R?0RC*LY^sc~%51uRp_AGK=qk==|au8mTM>tTf|1v9hT&%uq7o>k>wf zMN&Om6DxmI&MqA70QO~q<14W26Du=@MX8CE4?vntL@I^X#0nMjO%p5L4UU>v>RlYm zaQ*;I@qcG?sfm@BCDEt|vnEy+?tuAu1yUwpjzDCKCsqbb#O^|wAMwQuCJPb4OssUd z3q|q3YBFAoaGqFceiAR-I{-T|!Tzniv8zgm(o_|4G#FEurIO)^l`{DwG;1jcYni0d z;)#{Mi!mk!c82k2Qfy7EG+c!p#5aLHG7R0Yt+&s48ugGmdSQ@o%2Jr!Dzzt{wj!V6 zKc1eA&+KVM7C-hCag#24IUuw34)^z<(92T z+9-5@GN3f}${;9EHX#OEk-P&ci565I83RTYJBzJISgCqoGzHqmFubhDYaEo80|AaU z1lfiXVk^?{cZ2t@1i2J3u(dB(23*k{FwHWi5k|J%Zys$`4zSZ24ogl_(7?nsK%wp~w6+o(8O;g`4Tw!|BmNS&{C*`Y_H` zB&ZV}>;%NmAdDW1qrWv4N<=D!mldI6dRdW> zSq;+FCti0o_YsBAYDyufk5N+7QlEIGA39%yNxp8KRss^49p^u+fNLK?Z{jK9!Q_VN z*_f&Mre-NvRycb=EItj&=xdbwa&Q!gtN#cOw zb#VVL7`u%w4_ihLN;n6hP^nMG!@)WVPSBO|>989ZTfM}8Jhu8EtJ`kS4-wyTYNDF0 z_%I4j;nKgJvlVfba89E*;>#d7HZP+}R^MCv(D%JrmC*kox~_rbUc$|6Mb19i#UVsU z*$IR`c9NN`IGq>6=rIPvDSEbIUO8la$vBtrn=Z4Bght*;^YiNxVj+Z& zBFLkZ)+PM1W-CaKlfXC6R=n(8O^N410+zU>fB$2)qF|0hk;H6DTBpdf71xhD1?Ax# zBAue@;Mt0ajr~+c2QtJn!Ld2>S~J0vY32OT4-t9xY^s{9=v2XDvZ6t>HCfRVzj?BP z{^7|A`iCbg3NWkRz9fq^X5`AtSR#l_$rDN)|;No~*dnJX(AREi*~eE0w-` zvSMbE6w461-XPL%Ipy9xSuyx!vN#Cn1jE@ac)BC!?PPHs&_jmvSnw>zou{!PGqRKi zK_x2*mP&;_<~eR+JyFD_+l^P3_T!l6yyEF13|KiEu9fgaM(=p+@2dqyLuOU7Qo%$< zm4c;3Ct!UTSM|>m8Lj_F661l*W}M_bvkFJE#hK#bYY?_GiHah1N@gPCJ#17y4#F>n zq#qW~61PPQuqHBU-YF{R{g8OA0uWT?c}-+s1EMGltUTjNLd{cTH?zcFb*?NNDPS~a zmNk(}({;1+lkoP*~+4NpvH9omZkk>KK#%J_ye?I1^ipR%1+pUNeuOOCC5u)_N<4+Z; zfBP{{dJmk$`m?ynu(n%;E^US&h$E5>evCAGo>67%!EoeLD$tOGye}T`VC4=>r zU8X5jjZV>=@T6uSoy6l4p4C?)ml=qe%+l{rF8PFK&apI>UI=t0Guje^Pk4^}7)!=` z0iIw2snBWj3C}jYan~ORFPKDH6v^v^r^I35oN<{&aRgll7Sn#hGk6>Zt%;!3{eP-; z!t?9MIQTzc^kvo*GSp7Sp7kK=_qOs=hP{A5|PzxV|Pm0YOwUg`_-LCp?erL$wD3EWrfT66mQ~Cp@3j ztSO`mj1*=mHGIOeGkON=glCf@l@(L=27RCru}*j{&CG0^@T`0wMv>$KFc%ZM2PMuY zJev&2qP6V;wvV;(3D2a*aa#U2U{_h5Pk6qcjEBG{JpXyCoFYm8%9sR3(61Ap@SJ3^ zal&)J`angJ@n9wqTjh;UcxL2E(%M=8Ys1?3glCQ=QCfZ|uu+!0IpO(iueNH84c$KM(AZg_{$egO{Xf{28!+E!>>&+_kW}vH>!GJl2*F z+~qSTJQsB-r}21TNsRLe&;EIE*Wg=-Z%Y^%CS^#h6Q1$39cK{02{wY%?aQ3->{%YO zo%6w1WLVY-&*hEl>n!X9_8sB+5ttL6jdtiy<;n}derMdA@Vwp#8#Mo>=mc91`GjX- z)pX4X&-yz`D5fk8daRd-IpG;;iQt6i#XY$dQ?>%VD~V7B-SuEjc;*hRkNJOqlkEg^ z!ZRKp$f6v|MfkItI96qv6P^oC$LfR*0zW|7) zHzzz>#07~-giirBgXQ^z=Zb6a`IaGmy^(dd6?4L~-*}7xzXy1V3H_9@GABGc9>j&& zw*fvd1in!%rM0g%C4MgK|3c8UOeRXF!Y4cj*UG7xF<@15GXo;(BDo{qbW5%Mk7OYe8_02jvQv7=QT$Tm4z@z4(CGK zn~xl>2f{}V6I|KFWkfur#2=9freBgceB@B+HAfEj??>mA8>!|)P!fFPaOQnKL4vY8 zph^V)Lv?SC9HxX;QOU@LfSNIEjvTtakI`^%Km!@3BZpK0!bx@^b^QlXBDK;oEcR`D zQC+B6tM%aWtQA#`sxN|jPz_oJdSYAX8u|Daob)X4B+ z@G&x1$0PuPV;VAJ`BAf5V5es|;>#FJ&LRT)QKwMv(n%=$QLX{-YMLUZBg@b)x8R%o zsNHuj?m{+*)b_X0a)$+I6dG`r6G6<{|l-1t%V zu4dD?tc^tw1YI^C<42vW5UceU1{PuA#*eyQBagm#Tj0}^c=41I$Z@m?Q^X~wU*$7g%FXKlI>7R((;J~F?(gauwqr|$I z`tp$kE#Dv5P)pwUQFCf1X!)7I=2`N_k6QItlo;I)<=+5on__=e#FA9*V1}rtc)L3?+9k`$^)!q2<%6-K2cJe zl7@iV8d#Ev(y6c?HE~{f%^U{S2X>}Pj{T??e8`ZJItJEN zJJT|{>_;8=E4QY<0@W|g6qYh;`BAqH7uKXAAjH^6s#umEl{p>mWgRe@FpK&b%a7Vn z1iP}j0qkuEmLC;5vORQq;68skTmngs`FB^aA+EX$9|`2nud`4Qj+CMZ8j zTLa@qWgms9fJb2X)-{Dz)n?TZf1B*_qd1(30IeibmHpwHsb?=Pr~RnvAk;A=;bZ)$ zB>^r~e~$LR`Wale2>qOmA9ertaLt?w*8DdyjUN^ND32akZ3gQgG1(w^`B78fiPp@U zVCfoFBPsTyk}g)zgiJJu+p3svf$F0m4X9d?rCN72Yo`B5|| z6;ARS;;Wa%1}$vD?^q9y@tRLA_|-1dC-;pMl*1T6l?fh1FfU!JHc9ckbp6Sc zXhrOR$ZjmaOV^!;;EdrY#E&OTrNv9valhpepCWpxf#e8EjF+xmgKCKFzz#63mRvQC zrR##@tB7BK-Lc_X2`^p0UjV*`zyyImg35p#Yo&sv>*U;VA{l&aN5kgu@->QXt#06mY7$*s@J^1?)aKD@bFJLwV4=rkvgz{!Iv>JQ)b%uHRGAFg31ij)IEOVi90 zRe*5PNGmn{a21hyW_OH;JnF)(H(b}y9@ID81VOnr*ELMQZ*`#nMT*Gxg0b*|TLH#G z^bNNHL=O83i`r1V>pxV}RGm>h6Tg*eiWHFvxzJ}p^>1%h9Y4KTP1O_jYKoq4jlXA4 z*rPjk!ucRR0Kt*vYp63l;d=pzLKa4JSp&(Ogqxl)Aa_X-j0h=fgV4xMGCkqD{=P!C zH<;-ODWgH>o1+t^8+V4i<)AYd!W?g- zCsZYiA`bV2O+zC@FK8J+n$lJJ?w)Y(@K`Yku`>-K-?reXj)xy*5z7FrXSkIGPj}q^ zxP~|g=mf(ZEO?fq&G!YxbwE#Suv9Aa2CushL?)3`Z%U1ONuU0Sf^y#+P-}vR5zM`$YOfT}y=3)wlNE6?B0pjQ?j6<6g2VWG@4|%ecLl+*78Q5Y5qo5R_`wJ!mD|OYSKH zt5pJw^30lGrGj3vet3eY4XhF4s{Xl`ymGFt=nQNS<0S8yRrHc^xx>X25I$oP6-DZl zOfT8QVKQ^kQ4@vXd;Cs{uZ%qBeMS>EvE(IE?hwE-uQ5`axyT`?~OO)Q@oU zqYr+RBt;$z6XRkNl()GKS@2qzSlFqOf_@}OIYhJNkUmm`Y?Hq4h3NGRUYJnD= z=$BjY%?lHbfq3@{x5UL~2r3U!$t=*yb0d7VFAnAn7?8e~^6`Wu#h zZ_Ng;@$KTpCd%|~OvFJ73)C{aq2&Vq1-Nt7pRVPZ<9 zaO{5p=-bNF0zXx2VPfKZtO+-N5#*19+IV5&V4-4K z{vBX#EqSvr@k=?O^F9pNXbU$B6QAC%tnp8QEwpg6Fj2pb)cUsr`_{tE!o=$DDr)=} zV3#f2EKKC-P)ykX`5c&N!{svz6BACj@dCiY8RvzGi5GFSBo6V_2&1wqqhKve44MS1 z@ixFNHi9IpVSO?)DGz3&M}qObVOa|kLvm%)SzG{YG2yCt_?U%>PmUMR3lqD5?PJ_5 zOpKme6Z>BgeU%_9hrBTHCx&a%$1F?~xSd}yrGMMZA{fC_#4Jp_vP7^jvGi$zV#=zZ zr;-R|&|ME^VIsLZzBJtiV0SygEKGd9C{U;SKJaP8v9f9wCffXyMJKcj_G`i_dYG6mD~gN$SAQECN|<&HC2@iXoryig35>+AC=kn_pQNNzAUg9 z!u8`Y3lmw>atd-E-@iUFAyRDdo30V}j zXa&HHOxUX$lUbNJ9S03R0Q|`icwu6}I=5lE4d|JHC7CFl3NK7F{HT&PI)UvmvPaO* zLoroyyfASvS8>g(0#-dcQyEEKn5bGZKr_35HNwucj4m%s{CKaHrq2QOOB>xicYhj|vqIe-49pF$Q=%=3j*98s{66fRFVQ*pK zD1`VZgIS$Fl9*Cfge3d;w)(|aIBJ5?z{au`Cg$U60!Jr+1DQaBNB0vp3lsCZ7ZZ*t zU@T-7^*f#^nhjo^_9Dg2pzUQU)ugq-tJ-^d+0yYV2v-crT9_EWv81X$#|vQo9pCVr z&BDa<)wl|l`itUV#lDGY7A7X7rRafG1F+f~oxsj=S2##-|kf+g_whvax$ntI#M9uG~>a6>@N7^Z178XDuYCsd~{s8&Iw zwi%otf-022>OCDxoCz*(@S@65l}0~=t3|0A+K*b(&+Q{U%y9q6Gs^wrcN5cjq&p_X z{P+YvO70pCP1Xjl&x>k5>KT&t@}ufyXg?}X7eHQq)T3!ps>fwNYRCwX*pHgkp{^(a zu?{3kKRo_4q}ffo9~Fm;If=`D)WtzHm4YR9S*1!Tu=`OX&s10RWB)<7{HTh#G085k zgLfCf(TrwB*^kO7hpfr2nEFFdn0%KA>_>$ULs6WBvLE%Va1J435mSq0=$BjY&3=^c zD(rIV23Ce$rae`SPSN;LbRJbYiN}7_j`lcVwhS@rS-KpxO!lL?Y{R;}JPhMvTD*Z5Hff7EU9}J7!JyOyem|b3V~3HNt7qckNS6hG3i-9wgt;V{*dlz+7SVDn{!AiVX=6VsK6A8|Uj%G9;i_yt#*ZpLvZl`acfh{4 zaN|c!>QP1GzXQ8v;l_^&*;ifbcl1Dyg5cI~{3ze5aT+fItfYk-Kgw@Fw6X!RIkDhVJ+Q8fvmbSII#%B@5I>nPD!ZiSWBF0#n&F_#7Xa7T2vWB%v%%{?7W8Nb zz&K)9mLD}d7xvvkr@RjAF5$WgjURP)y88l{EIlz2LeRMJqoOmxu7o4H3_(^7*^hdf z$J3Ac_{UJilubZy?ImLTD2Hml%JqUD)oV>9#gr33pFtv&L3ceEKk5L!NbFb!aJ`*i z{HTgG^6PYu06$9{E33wjI`~&{ozMf|&+U5HkNWdggjSTN7v`rC+%3|43axa2E|;tb zEP-%co#vp-*-m-2d{bcWSn_6r*N)>wbS4G>8)nHHKPsY)zX&5gX%?_gEqUWdUA!4A ziW9yO*mjm@KWg2NSR?!i@xK~bcUv)j)Y4rw@%>MLFPSh*87t#Qecv(@#y`C={z1^C zH=jbgHNmM(NqIok3@ph+=~UQ{8Zj`pX0`13(PcmC`-cIVei78$HoE0UCAO}uNtya!2!^2BjVclSvHYmgb+H@26c`nm zMg5HBN3}`89_%!L4GqEC;PrS3GCJF}*O@rjLP`B58!9Kx|0 z;4UU8KT2Bz<41jqZ>l>^fpLpj)bDtvX#A*%kMjw~YtWp1P0^M8;SF9jbL&n0j-nux zG9=575+(FAanu0T#NfI`=;v(ws4f}ts-_)1!5aA{rtzcd+)U8}t53jMMocycUVc>J zjZV${9xPoxHIiaK>NY;UO$BmXBf@C)yo?_;FH4|sJV$&M46OBIr=ceMQCAPd=?oVF zUWz$XmF!1dFBPjf)qyv*aik)$ANAsB9?=Da!Aw%)D!U&ww?u9+6@>YAlI2Iy$WSe` z(4bT}$!~DIe;Zu{UH4X_zsE&QgCj78~LQ^5yg-Aw?#HK6Q348Evo6h6q*jmjf` z1>=UD#TPZz%tLDruYmdua3`V%lYHGcDz?iM92B$ z`Xdcvoq$1@|3q-q?H-CYr?3Pc-zit%^2t&FDjHBy!SJ~sl6Lv{t_#IUAz2q_GsAGV zCZ(g9kMHbc%(V3YIM@(mBT7U!+qOQwKW9x5ZzD#^kHJ{LER~S*UZMy3_zqZ^Q?s^! z@EwzsDj&ZUrA724AK!zE$_Of|ya2-QL~2Dz>3jXA`uHyGo?AcYmxQ4f+<1>NK_Ys( zkMEqrDY~JA04u_HcGYA*_wh}_?Nzce;;Ry->Pk0~MLxcndL`og6XM$#Oj4;tNT23a zMf6G^-}7s8iUGjJF+P=^|1tWYi@s{EqE9x}@Z;!%jnUhDf;*6cxNn;07tz~&3T!Fw z5PolbytmLG;>9bcx=^e1P#6}f1?+ise@mflH8|1_k?vDU`xAPC=0dG`&7)BB0Yh;A z8-l{*A|jXzwXW^Kaz$BSl^9=6IA5q07mn4=dH|a`bT?>z$!0kQImH&%M3HgitqZt(sB^-#-Q|96?otZpmEZ!g&bQ=ZY?yfO-ye z_!mTXRFbM5^|K4_Lrl0M*ZhQ3l(Qu?v?DFu=mBaF-y@1Yii1&d2!QtuC@KGLMx?_x z@Gd$4`62p*T zvxTgOL=W8&(Y*~MFA{FHko}$4UtFWdASZ(`%T6*|$hu>JMlLm&ejdU}vE`!&SQwTkkJ5$H&ault%@ zi`YsAk$wtxZy`(ixxA!4)|X)2)7cW)v4vQ`#x1hHogBEv0ss-sF>ezq(Kb#o^H*c zO1|O+m3=Q19HJEEYCKeSYrZd7QUoAIUIBKKadvAq2gfK}APFA?Ev}3{CFD75oC4{f zJ_XBbx27h@?A8#KWrXYQD);BXg5O6cL2@2$&B1td!o49qkdn(yxn;MeKuwr&ITgSs z29%-1P<>f6J`Yyo*Ic3~l}D}zW0#%9Zq2Cim;#cgfnG8U_dv8!wxS|>Zf4(lXK@Se z6M$ljI~TGPCB$w`r7Ku52?ki033j(;=rgHvk_bX=CTT6kt*P)Qt`2z@V0%Nb+?q(F zZn-r-bgeAPQNca{b2_nODL?GiTw7dLk4C=+ww7^rYpzA)*Z2>>ezI`m)>N2SNRM{! z0DH(dyERXKLFer=7F|DrdMuLa*>Y<({DXZQ5dbSOA(`@PcWXKhtf}0Zh9I>hB9+3+ zt)XK6ms_*s0vy{p5c~?k8lq7%k;<)6!XeutgVlAlg$ia7N5MR2*whl$b+zM zSNES60qVNi*2^&L`Wtiy+NMgti^=WSv+YxBcXjEMQ@c!0#(B6*PdsBVjs~L0KYZ_`5_Miqv0}* zN+~L4szKlZFn(f|N{C&iS80)&bq9neOj4@YWqQ97F7u{lnFA#ZLU7EZq}XMeI4emv zqH@4uE!?Y{ByJrx`TM)1Q;!<1-$=mm+7Q0UgM(U zF$sd;XhcuYxJ*yRmBs^1LVTLRp#SVa3}3>&|$4!r5gy_cR$3a{#9>!R|6$ zSXN(jrJ~BOz*xsDl?=N~Z*7XvtRo4Oa57!sB*)MVJ8K`XyUAh(ghwSz&tvuc1*pIJ&5J6D;r zk zXt0D?);?cEqL8_!U=j?$@hzooto_1%fMg`%D;rE6A%d~?4P&s!r5>>6jGrK!t^JX$ zuzkG%4rPMf+Min#CVqy9l(WG2f>|mVw)XM!ifGn05cV@krN!3X;a^Cc2X>qB-$=1~ zeiJ3t08IAadboGh548}8zz;wbTdI0jCVRsPEQuBeBN{=09wwU}lyKfcwWf|dT1q&R zz;Vs{0ewB$vUB*4zn;-0($kjpEuBYPr7VpCXFO@XO_jyAY*^uZ;xXZ009(p9+p=c8 z$|`&}p#2Q1$4?1)KK@j|LAU1M>5wgReCT1zh{`g;bvZ6t+p-}ap=pquhb{Z}9wxyf zAzcB%@tktY-pk+{@Qq{|fVT`NvwQ;v+p=Ew3k!e5NZALB;dU0=GFLcEi2NAn0>kjK zW%-(+7B>LgZ3r?KCB(Mu(#b^Z{{eW333l)0_Q6s*C(l60Jk4Dew-#f|mMnrZUjSgZ zAy~HTF6zayWvQ9!iM&*>I$+i(b|~eCZCRl^^|Y<(3al67Y|GwnSYP9lfK9V-W6P%2 zPuI3=Ik0aSXIpl*QY7v_NBnWZ=&?wuXUmrT_;niYvY#9~P%a&!&+Z4_P7G;jC#>II}A=nW?y-WT>uqdPQ^VJ9>r-Zq!dA3h#k*&S5$!IIP0US1T~vd48uaTVD&lwl!p5~9_db@wBzXsnm#A<&dR9yFA=}WV6r9=OrO)} zD4y_MU_UaRMmYC5Hy@(`{SNRx6YPCXpQE@Oo{B2{khMSrl?9y)_c=|@*4C^@5GpfC zrNwz$!0kjeDUOg~-By(m-W z=XyeZ4caE7+@lks8W+xYQ9{=RtN^M-nYA-%G>WSC@EbONkOk#@2n|n3i!VJu_8WHh zF9R?e#PT61P*VOazoF|w42NYopb5;#Mmn@dfJK>I2Qh<-7%ATZqb;+P9?im{Ou2`J z+^peXOkkGM#ePG#-FXBRNiGE8Dg7%=U>DU!V zeQht0*>4~!%LrG6msmcd4G_tn;7O63M_;?95tg{;LHY|yuD^N=#&7tw2}bjBGl0DY z#FraU9`0+a_bepFQF-JAFmBpe+}EBON_t)cbu>s-mwA16bM+*v1qX| zY=}=Jj2?@mdbax7nCo@m)dOtDgfEF@?`wPhl3n$+!$2BOL@I?>UrWXG?rY5kh;JYG z;Z6n!{)C|HhS~t3@}p)7m2k-Gsg;zQQ6_f@@i!%dhNjrm5|x`lKXfscQF3*soyx7< zjKEc z`|+F)Lo_STTwHyRpt7n}aX&s{XK6u2mGK}X5ve>S#r^oOezkNXY5}Z`g`0l7U#nEz zNQVL&#W?rl6>x-C&PM!agsHmXe!Tn>tP-q4{7!>ODwWlbpRZb8oB(!(aeA^|{kVFm zxF4@X3f|CU$Xd&b~uKN^5i)z8@$G888`P_PV>P-+F>*)!ae*DA3ma8r)=Hvl!n;IQQe%YQTN@8sH`-*!%H=H-p4sDysYmj9;0hlHq>b zXHL3iJpnerN#aD8GLXk2v~8(kCS3sKTf8KX9`zBVZJ$9~ioN4;5X#hsy0-HUj5^=7X^eL1ojU$0q^dEQUh3x*mzq zJ^rH4@JQ$(dGz=bFJp_j{2dx@k(Mg-sJX{auTD_F^LrdDn6CJ$n4JQJ&s)&1ZGQ9jgA{pyn#6dPoK19{+HItl}x*uYhHO z9*uL4|7CDFg@*tt!m!E-CFE)KmO(?%4;wLqPd$EHkh#YvD$5Aht`tAr<7fH;Pm1I` zdi>;MEV@mB^bAVwrFsme$N%9o+FZ6AzyEBjqtL&f8hsW9{=foInJvpL_f}`A{K65FbexJr+szZ1wmrwwJ~w`T(0S zp*ZE&-s4XQE2nz=9v}@SB9+3c$ERX?_4px8ABCzeJZt-MVmssxASeSvO-ps*^h4*X z49V9mF%4Vz@NLS=2Qfyt2Koa}5f3IeOwYzl%{O&*^YX&!hc+5t63WQhcU*`DOPFQt zUrfSmP$Edx5FBxowz2lZoY*%j-$s0UgUM<{FxEcDp>QDw0~^bD3gK++w=_jVoeOX= z6YSPrj0zF;AOim%jKj=Q$*{E_em}ovT?OF}CaJX8+K*Z1D>9*cfe5M`vI!|x&u^ln z8i2{3Sp*F>28=2Q`~Wmv)w?p;Yc^sgz5y8T7+oGFn;w*K=0+}3bF9uJoZY|)^1B8j zPlrI~;y-OSx@5__<3_Ny8}viC+JBa$4PL~T9tQ7Ztf#@7^Ds^vq< z@FKrU5GM)$1K4B6+2CE7T3X@RkVGJY7FTVA67p;#l+~aw_rUEUgV%@z+9pczPNZ*M zqOy!|)ft>s8@#!T(SAtI!{9k5A~z!-J(iL?N4e$c#M2G17$oNbSYkkVl^AUB?(fEZ zUQ`~r2aF%=EH-$#XvJGz0ea6cybRvq8pVWkEW-VN2r3uy0VTu+Z|g-Yaux*`#RR*- z8?-G>=cEn@O_-#$7=w4N7Uo~N0qkuEmcdIv)+~cJI#U(#gbMZvm`j~kC?gQF&vy5o2j;amVs@efzw zL-8<5$Csc{5oQ^sko8!s{t2YBlw@vXi;Ys}Q)saA9^(Htm<%O?F-nU@$pNu|X`sraYa z#av*^881hQRUIo+^V-J>}UU8t%HSM&5+I!BCMRu75fmwk|Gd^L~zum zOmv`H(S&H3tMbB{Le>CSp9yPQVANJ%rGFU_Hca|%4iw^Th;#wok2xDD1DZ2l`aWsx zr#Mr9&tc9_#0h!YEm?#um%f7r7ZNLhZDyRbino3_QB{l$#11gTpEMX(MtMPnL$itz zR4}Z7p)TV84#8WGYmtY4g}SJPor`Ef>XW`7_}(aqe*zJoAHmU$k}x}Y8=+s6QHYK= zkQ_j`*~!~yK&Tjo2t0ofTH8rxC-0RGFx$Nirt>MBcaZU*Rs}KJNjrHbE`v>7;{Mt% z)j99v?RcuNmoxG&SN#L7z@`k*stHl4ciA(zTKX&pqZ--5($c|2o zlGZ8mPTr9Lfr9c-he#8sI(R4VyoEt3qa7IHoxJoEcqcDqTJ7YeA0jea?|8M7x5-M6 zoxD|+5=AFFs5AWoqe|RVFj-;&O0+Rk6!7-O|W_I#^T&$FMN|^#( zA@XwtyLa+lD&a4J0EIKWgkavu8?!#UD9_l}gbC*?=rcQchv)U!$s4i)eTOPp6mfVb zZ@|ogVmP#nAx&#l`tF^)|N1z@Y{V`!h}>bpQyrs*6c!r+?O}MY1y6UZYEn*|0(6n# zLl!*CG5M*FctjAgrIM8dOQk|z26=5Q97=<_8>qPqaurJEq!RHi-+n{0ilfj_8T7X;(3>En}iTWoP)^uEWo>b^Nz;umi36=N|;KE zclo{;jju~1`kaB}HA;+k`L@caF75;ShjFz#RpYoa`DI`Qk?rfuA{0TXR@+;(65i!I zF%xEBDu7XyS$C{du*>)K--)6Lu-1&L`sZD~Z6ka|e_-!3PV$~v#V+5Yy@SMD5SBBE ziXwGNW|!~#&$5bLAbf8~`eE@d-}-pW)-K=2m>{9|;{n*ujf8ujv)Sc)etUJ11Fu~e zg02CQ@Y>~D{tVW!V!=pembJ^5UTo`Kc9JD?waT5YFEJUg#eG4^S){LSe58!1A5;eQ zgW8b#5ovym#E+7s$m0?d(ZpAca&{vNUYD5sdLgTV&J*NyiOHnk5QXt2CaQRODlQBY zA>T>g7n|#fyAUh63ZnGGs;|LYi{N-lEsifS*}nk;lY@vqVK6DzfxwrT&|4;*gz_aOPZnU9_Z%@< zSG)B|`sEgU^Cc!#4q}2Y3alD-8DAvk7BrWbP&2^yq`>1#O!gf{E_)zmFiU5nT=FF* zy<5U#mD7OEV@4n`_!5&Mx$+CS2H+khkP4kPUt-eW-5_xqgxgFaEsErIiAgqmrSmmd zf!~-qu$cBsOfFqW7P1s*@&8Y?E-@KD7QR9QFxoI{D4M*^A75fJqAUif1Hl+=So*y+ zmqE7apF`xQ9M1#3#BijlEH#Rxul@G=v%Dtl2H`N1RDEd@E-@LuDO!`Rg762EC{NZU zCg1PCs3Y?lv| zB_<~rBE ze2K~PhOvqypMm)=YvW5y23c%eV$%Im7DbW~YcU)_(0SuaOrDKO*VV7%3d>3FnEqQZ^Nv^?(I`0#KeQ4q45|efJDrK}kH-LR}nOr8%3(OLK!*jmC>^YAg3m^?V? zc6omQ_7mgg5|b{YU{|gq`XND94*3$3iK;Q0%OKl2lN3|tTaOVPf~SbN#AK!=f=f*5 zF3GQ$vM%V&NQ5%zt_O38N%x8Pq)ZQhgY5)!iAlbF6?D4OfqzCEE34)blZ9p732g+v z$F7GjFvHkNgXNr`HhuI>kLEECivCfXX9OH5it#=|iK zV>Pp=-|t`LK*eE%q1o*a^VZidl3HvC85j(javBYFM((DRzhAeDtMmqAW|wl`k}xey3n2KiFt6vq&8krE$7CYXLn;_zjV zO0T&L@~v+$Eh~_!4}y~5%OID|C@)A*h5;&0@Cd4V^R?fnjf$#dWK}>l88(+e&MX(I z;Z}e;Fie+0QUwSn*@e_buhK+n=!9It^^q@Z9<6C+-R3dHU&i4%#ouwTuNX^(S_gew zNarNV5l`{^r1^`Pgr5NRGvhqPA9y@M;ST{lWmwI)QbL|HuEPNb-Kv0H%rwPc5qaV% zexkCBaFvQEsi*kA-T}Xf%y7-zry8WzpnzVBcD}nc@$t?>?&b3$V+K^A!Jsvn7Rm zig<}?Z zFad&d5Y(EQ+A&PcM5-ObN|@W4XZxfU>6#E;N$)h8yvt*!k?%1KBu_!(dCJmTYQuxs zX*6su?Ra|#;DrGt70lXc9QQ(6Ph8UXvi})<4L0wn8)2J_fPpKf{OBhU2sjQtw z!^`CsiNNYIew3b^eaEnB_`K7oCn-I=7);$@!b)s@94U>`C5gcPgy z%v|F_=7e_)?%USnmu~f!;A({HJ2>9m7xeVVdCrXg7^=-IDo*63%I;0#|=W z9ld|(_dTd+s(O$8L#qnWrr;ukAUOKd1LXZfL)PF0mPr883@9o8*8ZW(WiaBD?SS@R z#!%9sD-%10Lk`D^QHYUpA{ZYtOX<-p>=>5Ws=8UL!Pv?yrHl6ujl$J&R3v#4gmXk1 zPic7+_$jKwmFK78y00Ddtw&#*oQ%08zA|lN@dDxnRca_WMG+iwJqpHsZLW|)!WS{J z3a}cCb6-16R#f=AfZ8*xazY7t+OnI-Xi)1xm|CO0b|J{z*AkUwge&8RRNdF++K1;s zavptcUaW4(Bal8p$z`M5a$kG74fVCR0Q_Y@c(-71U)yAtLlmO&$n1Nu^o5{QXcqUi z4?iY7(LfUo!>h0DTO(P>MgZS61X+v{;=VSbTB4Bs0FGpWeL20KKgNKRlTScc#3Zf7 z^tI>r)D?0oz}<#m^|g~xM60hIu{KJSCUrN!yhrS4$`ALo-_EJ2O{C9v*#Co|aqeq# zR@GNx%M!rKTDa+JQ&VHaHQF&;8(2Ey+}ED!f&EhL5#N(AdMuJkXZ5w0BAhsn4{#b7*o&8lEN8`78k!K4^Fyb;l1B}G(W;Di|i-?{iz@|Kyai`PK-tN#dmn6 z?1cC}29u46U@Y?d(KuEx9@uoon-k6!c{bjD`4zx*Ot3G!ubP!#w1$Y3KZ5Zyvs5x{ zk+ZxNqgj7~@RCU?Ew;$B8kQ8fP`<(lsvNQ-DOOJ-`yZ&y7+>iadkBl~$zaq)uxgWv zElZ6-*3}aIr4i_*T7l8k=<+bY^lIY*C*&e^XzjYfIT#$5|5l7o$P37Mz{4_qp5W;P zY+sT`bf<*2Kw=kZ?t}j<%k(_A5Q7N+8Q5=(vt?Q^ELh=B0lj2cJ$_2a^YM=X9CTzZ zwqlWGN|NGvYMAAp`Q3@^)6@|47K4#1~|Ag576Y?(U# zi&1gFK{&w(Dw|$jz{xs!bWX~H;9`>2Vk}eaa$Mrk7+`Zluq;ziq-{x6#w{mc4coq`yP#oFw5T5{s`mrZ6NKTB-fFCV{dLXNkmgPgZRq^lUs;j z?9FG#uuJa=Faf5CJ+TiW`2IB*Da=yIu)XOW;%2o2 zp$C&xT5NBwwFnVofX!h1Fez4b%ylH3(~;fO=rNUrb0LVXa`oH?jN=b`7>cYePeXAj zBbzu*x-LWF2B|+orQ{(@k+wnNGT{O`d=NCwhGM{$yb2EkRGeXzCrZfkxqS`oLDlva z(}to0$ZRNx$}++={RFVyYUiFgY+y)?gr(S4MmSjm>`iW0c5r0#8%KuBG92I-L$C}*Ix!ir|$%$u@13~L&L*X2cw2LD? znlO4SlIqzq6ob0s8_Kl-He*5_%CFr}JSwjbJ@p1@7!j!yUWS5->18N*y5P|*4BIzB z@Bo74`l$S<=>jFpZOyZNQj2ue9UiWy3$`Bhm@c@RS{VC3f5aprf@6yEZ_ISTw^ONe z3IkBufRYMkO&4SzfTkd;0ZlgycWW}!1*;Tbp!Rsnj#3#xDngqs7 zW~qdDy5RP)6wO)4;@CIX&k(G4 zhN?hM7qmED%spD27LVb8KPoR6K~;puXq9SQICr7?T;Em>*3$*KenLf4)q6}A)Kk|$ zG=he=Ny`d)fIMB0cYQi4p%;Lm29%V4Yr3Erc0J)VKG4sYv4(W$i2zI&w7pbVY(fmq zzk;!sSxS#)VY=XC??gB20vI=$rF8Lh!RvBm1r-S=#xMv%aO|M8*n;50=a6)DXpc1l z8k8>ug)j5SVe4xzG%F+OP^G4W)0kA%r-E@`d#gk_(UkBW!1^)HeQlXZfeN1jXa>V7 zCzOzT7$I!GywS zfKv=X_Mn8gukE}CbN62YT*U-?UpwLJ(mE#xK{&xAt;O`UhuT&W@+QFhhG6xzTTusA zUwdy`UD1yU7IYHd1V(TSq5SZm{IlPRYZDm@ERk{UYu{e&o-TL`SW62xeQkQ?+S&vU z1U8&;?rRact3mm%Al)D$mBOp9 zrDA&ZwcL0-8yq#U)SkcP!FP|*j^gkAnWXG&^l3CI!Yn(RyC>RBeUO?WI6fl%#?HQO z9)YIN9q|JVh9!OwjGdhvnT-7(z-BZ48R2Yahs9!IcO}5hOt9P8FZNUtUqS@qKQJyZ zOC`g0_F&FBn)MU}i882sN|hGd*>@JXL@=-tj4va_s*bsiU>=|%waxb>g)h0uIj# zfEx`#o}h%-P$c@}gz*mm&oIGmD8!+%Iwuc6c+MoP#TbgqSTK<}&!Q7V(1q0x(v>;e z71WDmC{BJJB7UZVRRyyqv45ldu%W0PUrQT`*1$S4&W7U8m+nL08NkL_xG@x`rn=`9 zKL@siaW)jUreRuUJK_%zMvp~OJzIvN%dKSL_!Zz?CR`}ciqA1upB-#*4vmU1%TQeE39~;Kq>+^51JZ8{g>MrY zbI(Nl=LVBch+quGg%y~`*#K-e<1YwjL(u_WZj+}0USfjXQ1~wcVI#H4=U`+(_EZ*h zGHfW)3t>D&tS}HtGfAbzhT@ah$|4C^1IB$x#4;3AN5WYj>7*|2n^QPjgXsFX4F*Cq zqWByCv7uN&mu>I@L|j^^$VR$82WJVX55#|#p_mX}SmY;sH?aMTv!Q6(KdZuj19X*P z_0m&9o`#|;;Gp(9!nL6&eBQ%Q5S3+w>(j5?BZ{uS;+c}1hoLCh5epFMkZz3N2&3Gx zp=kRhx;5DizyJfv;>2J>@qAZqQHIJRKLX=ZJBtlPi0SpF5fz6-!^GTxGKHWZm_V7T%MpdS)e%DonPA2bgV?I0p$Q81#I zrIKMok*jKoX4L_q36oS>Y$!^;RaJBaHiYpmq}VbPR7Y3^KyFembi=WNi6FX0bb)V4 zh9dko4?}S-$~wk#a4o7)tm}VD}kkLot0~n!g6(Wxhva|S|sI3+iV za?6II1v@hGObitoObE z=yTDX3ps-lVnb2<0ycVt0E}RQ-B5h>jjzs0H4xI7q_r4B@&4DSuJ!=C8iHjgDj;>s zP<(c(nD~SWHXY14#9l!8VM9^&kbBq9T40+QXG3uzJc~9YKLI;y;l@ym`X^Bviig0S zGR}r#Aa$$(cwTuC)MJrU&z7O+_&8KJA^}!m!eYv=-B5IBSx*^?W+1gAB9+3+P*5@7 zWGK8F95u1joj!Om=0Hm(i#QvkXPin;3f^2I)8@xrX!`Loqjy2A9_n z|IlD^6A_G|I355CnFTs?A}Cq8gK#z!3l3qy%3Qy9gMe`rIKMo zv8Pa+X7vSO1d~)+Y$)Ye97Ne%C#sNWJ1= zD3&FA8j7p+qQp_s^$-$IN&P7*B^!#hlgo+ogy)2gU<8e`p%{O`PvPYN#WJj3dP>OC zP;7_xpn>c0NXSt12bm28QCS9;*F3DK4aN4Wc%~%hVJNzGfrmp-4N5DV>QBrx?I76z!0C$57nE_n!Zuf^7u3mFTCGA2t+~ zAC=OE;1smqSZ70V8qZ|vlPHg%J#%zpC^i+y(1s#Eu1O&rI({}3t&icdRK)po(&(~C zs%OVg^u84$tddH_Yy^Kahnqw&JHh72o9-xPmWOihW4aJL{X^OOP+UTuqNd_eSQ(1>fZ0$G z%3<*BBusp$If_fya7`(kpP`sG5m)X9gg!{g)u7z6p}2jTmS`?R_|u58E)i@f>fMPG z4XHdb+f`%-hl)Z|Y$&3laPehXsFe+o#ZYue!?JN>h^-AEn^8h+D2AqB$ngfmp$r5X ziq3~KbWWxLSjdo$#Tbh8W$>w6196i9978b{MRW{BN4yE%iVAiSk)c= zj3}O7M*&H548^CvVaBNd&d1@f`cV94L^1X_MuxI7&et-U97upM6zg8aBt|P}U05GZ zIva}oufk9af%pysfrjGe$YSDc1d(zfh_9GZ$*`e__^yzqb^`c`A(a*zijNkDic8S` zVtqVC>=+8FBjLG+`p8&ap_uSkH()51zl8}~G8FUh7tc}js^xDePWQ|yCQ)3qz%;;N zO`}pehGMZL=8)b6+Uu;dp}6^BeWkw*X*|p7rl*Ab4MiB_sCJd>X+yCaf3cw;l*0&b zY>e-Pgs|UnO(~q8p?GH$ZqYk9_cspfW6CWXiY4xNRCoAIEPdioqWqi)HWYIzg^RDL zJTe7D^+1XZ#XrmNg#y_EY9~WvF%)M`;+2I#5Z^X{TuBMBp}4vQ&0`kCPZ70BA;0K0uEXGiTH7Shm-$A@&0LM^#gYr9uq7Bx$)>FYkZ{Z~f9M%@f4;u>aKk?d7 zlz~>Bbv6_QmQ>VwLukz$-583}ZGEmtJ)!kuoejl0Z1*6?Y)sT5fZ1r_soh9YZ&qb8OyEi6TNBGHag56x<#3`L3C zXjG&*hT>V3CfK75P$mxR5XEl{MQS@5QFOxjo<@_$2r!1CeUJD z*4a?FVtuYi1E39abYm!5bt)$=(`4p!XmeR-L(u}SF3S}-znL_;ERyQkF%$*%7Z%oG zh!+?rO8E^m6tmmLDns!vAQ!H_Dn%AULB)KYp~%|csEK8K_C*=tsfw6Vzn&7L3`Nts zXjG&*hNAf%JZkm^^d=>lNbws(;l)P+u>UL0PdA#ZNPsaEU1y@;UqM^LdK&3$D58IZ zpURIAk1-HvC_MSgikb+5_n$yKW=bW)hN9z~L`_8@|Ald=d`guT8;TVvO+-a#by=@Z z5j%#0>IkbqNF(EXcunDH4$!-DAj(LF;x_(bL(!v-zoDqzI#V>JxTb@dOVPhXrDQ`f zt4tH|GU;oeZDO4b#ncySEBy$hQ!J~So)Ypm6dfQ(Jz9mOAey7dd(Y2M5XxbMcRwB= z)uTs;`?#hQ&d*SExeb?}?D5^avBzjVLZYfJ{0+r;AW;n#`wT@oU>;Er%3*|e;`wcN{9``BYX-==KKfmzv58Y%<>k*kL~K| zoKyx-iy==p;tFns|RIqU%KOp*7$`2cgmDMA)q4)yY zGS=Bp3~ZIC^*zuIIJz+u`G1bmhT zX^x?|J|s=Zt$=n@k~b)RV( zCqP^bq}WjW+5r;`@)6YF$G$|e7>dj7;TRnavA6+b8A^x^g*~V;zW)ue9s_|Rid^YJ z=cE&Wo($<&jG<`t8>;JVh~o|57>XB>J;zYA^%fH4sbJrNTuXFi$`2cg6>Ehy1c#s< zW1S7fwcmUniTV@T14lQ8VnzwycagK>nndBy@w1`$^dWAjGB{s}G`cL3>e(?Ay|Dgj zHG^PJrn-X9Q#rb{cxQanbWW5FHY$&$B zm5TK*h%Yb@IHGvcG(xmR5Gi|t7{ruHh7HBz`*E832*5msR9b8(g7C4r70|Y^-kBnH zMif*>SOr2cGv-?bgy#@IZ}uIyZD>UC`hR|gqDMXd5yh>qa)@peS9UN_IIP}ON*+=C zIy<*`lk`Mr6G+d8BxT zPj=!^QD}+{MTxdF-Yfz&!4O#uMLFcDHpC_dkP|5(HWb^pVQu;qh~D?LHWYd8`5dYqLi>+( zHWXIJa@vsOL!HOs(Bbu4H1RIGQ0_u{L{SA=b=KKXw3~uCndUg(kuhs6a#U_KYHGM6nRiR|HZivKR^~=JO0i)&@sSEF)V}ERUW;OsUhcZbOFR zFNsD)nqw#y{Dpft66f>cu)d=BjiCtp6LXa^0q3h2O@2#&F%%2-pgV31tu^awNoPZ` zuTToszaS1}Aka|kXy+E25d{0cg7}yzl?)q-*2OJNZ3M83A(a*ziXnGH#VKevSl><& zJBEVl2&+J-kBt8PFxGkq&^zH{->SzH%jIidH-o}c>-!suruB-7A1JOWV5;M=4pJ%E zP#mmPQXC<@HM9<_v!U25Ew>*930{{nHa0pu-8hz-Tv{g_Ff1aUS4frjGS()n~w zRsq<;kdDO|imy&%*m?xwDFZl$;xm-rF%%!Xlt{StQl7V<=W;!}91t zh~F^ql&C;MG4W&}Whj0C^b3JhiY$hLiupW4k+s256U*>mpHh#Fc9gnv5Ly)sMT{Gb ziZsViRE-IuuPeJmIu6UT0r49{@e{gxDO=)vXQRpN1Qdwfj~piW>GFt070bO0%9*yDj7But6#@EJVadtaGN2O78{BM-IUtf@gHwiWjis0D5m8z;iT>C^q3Qp3JP+z~4|*hKFiNimNl2ZWMh2{&EaO z+O2FNh4i{>#)E50;rtB6PTD~>yN$STSgDj-HWc1Dm;sSxAyhV^tU&}DihiHg6m_XQ@7&4)~z&aZW zySMMrqzAOVj&2OaOK|AAOsgIfpnb$T8;TF6z`^VboL@;AT^338>=+8~i%o>}1H=;y zv?nUiP(%!_sSL#(K#vKeQe-g{RLtiYimVNenpnoPBh`f`3GFB~{#(83Q8ybJ6={y4 z@P3aGMOQ%Gaadg`eq$*9*+YHCD4d^QG})a1V<-lEgo1wzZ7J)0NM}RQ>~E|oY=gL; zfj~o1=g-1oAc9DF6~tYpR5ENRejFLCsZiuU4-S=2snTLY@ni8yq71Y&)`wHXj-jAB z!YUBb$k-c!<+iUTh@!3+ZO3k!h`B#7R}GEdcfgLT0Ct{TFdJpE8=h$*>lK6H))HtD|=sxf=>o zVRjn%3^DRfBeHfw@epir5b_+QRT;mWokl1g4~fs;X{1JcV->u6KyXQa{*RqTT6rpq zSxi&XIz`@Tq)bk)pggQ1=rUCY?=-S7E>30iAWOW{$RvQg(+Fi+?M^^H@Jjd8M77h% zqUN!E_{}?w&>y_h2>ro3jjTMGP1M7QMmVg$DQ9M8mmLTS_7DAfv{D(porE3<`(i0Z#x6??E zaF-~jN>+{--f86SOX=bYV!1^zMNlLqd^?SVe4a-(H>fF)+t1m z-9$>=fV1#BASN@V>QRkZ=;n3qR}Wb$uP9nq?RgF&3>?T6ZKzfMd-9#pJz{ArfoY~Ex+fgog zH<60@@s%fe7V0%7x)8y;iBw&Tk?2#1;WW=%W~F95NesUOhfb^drakz(C2z}N}mCqwD` z*6b!Sut#q3I_3Bp^m~Sps?SflhH3035|TYyLGl#H--v#L66f7SW~{2NV|xVc8OO%EiNw{nbol(aaTDOs;dwWa ztueR=yqieTym=HP>w|1ew7Q*mH<4Wqjon1*h=K}|13?bwxOq2`yv5;ihSPEe{+UO* zij8*@DW6b9hhGbAGwG^qZnK+6_HT0QydQ&h#?j4gBAvr?YW)GU$Bu4x6RGfW9vy#F zluH!A;fvqwCepU8N9z@#r8&CUP2{!L8z~zgTR>~;=w>&ON6!+qJ`mb))_FIP8CUQ~ zFa_u5lSXBi)O?)XL|S^`Q?m-=r@RQvTAk{ z8D1n#CuHS?5yat_5bq`uy0x;7s3d>{f^>^4)1S&gT-@ap+napp2E-P2}U=m?$U?v5WzD zH<2HfH`b=44y5KrmSm!ID!iM>%jhDt zDe7mO-9*+`f%j^Ah+PdJcsG&rW$+oBp*a7((Nz2PT^h5S$iHrUf%0PzO9LopH<7qQ z4TZH0;?E4w;L#V2*-a$-9DEAnDu~BSQNQD#qS;NvWn(vy$Y_^{!QpqyxPNF-RaE`f za?OQHC%rspRRZ8Okh7ae<86tm{;ZbJUNgFG5&Alt-9(PRlUtLcL4EW*^4bNxn@Htz zA$nl76x1dn*&t-uO(bSmc}<=HrOT&AQoNhU;wg3U7f(?UgE-3KunL9ads*~L62rTRsCdn8BB?QOsc8bHxe39$iF_IxDkz}r4e3pi ziz7nbO(X)}^HPz20BI`AW;c;7GwW&jYe*|urrkuS0)&U`LdMLrI(R48N*3N@H_-jj z>~!mbezVh4E@5EL9~h~5tAH3y{mL`M=K^Pqqt?L7Sf{ZvHih&;(2BFpv(ry*Hd1Tt|LpWWq!4xeCF})4v(sY$^XxRC97cGnUdXFwrymx=QyPU6QcwGK2#zc? z123pAN9fg*+$_p1&rZMTre&-jAsjOzb`1f+v(t$;DvBjk9(f1E<3NgMr~mi~?qD)! ztV2-xMY;O*+BLji6)0sb2*Etyq;9Z7vEM|5( zdJ2-63-J>JILlZ)k$Go!I<{mbv4#q^7vxVwZ=(G0>~zk3zPI?UL;Hhuo}FHrt(YF~ zgrLqNaOm*7-o5{7sGgmUhgObto}I2zwLac|!TF}7(PfcT&(7?0mqYkG=&KM1GO(Md zz}e|%uUl$%dNQCn1X3xo%uZ7=v&>HO=Rwc+z+CZB1iXyH`P8AxkNP~Q3e0WIzkO1R z^gb*}-?k~`4m}ZvYEYl=tmKaL)M+SWTw$#L;jp%=?-`oUgGRlM`$<-ZkYPkg1#>8k#x)-~>Y|D*ilZqbo^*iYo5{_=lh)loWp+v`<`BeW4>!19@@ybn|)8>HC6o zLn#lfGVA<#&;k)~d}xgGFOjC|N;i^E-F8TQeDk9_&JQq}q*6Jb2i+#q#rx3au)d$J zf8ggqRm11cgPx)Yp7VLotmfw_8bs=kW%UbOHHyNpP%Sts6gk)98n}vJW*CQcoUWi* zp}4*aZCl3Sd|9K(3j~-Iifz*|hfot*Bi65x&MOpkN5gNW3&h?G1g=n2%TrL?rlQLC zKulvwCBrKew@c;L)N%kD7*c8R3Psepvf?1LUs?Z)B3AdA`8+6@6X)}wnJ_XTsF8d) zoO_0L}R4Xh##RY|%f^9?0DXL0Gg6>nC=l5s98G^rRWnyTLK z^Pt`DAwAg_F$|(uexnP>p9ihn83T1W3Bqh6O3J_UdC+z*V1ZM93w09{S11l;M;m6GODs7Rb2de(8J}T1rDi$y3R>q0A(1`v6yZ;Bqjs*Kg9Y5 zaJuQ7h}`L>Z(t(71r@9h$bm$+rTlO=y}F6-J76C|o54DF)2rK8)zUZG75pW_7$KqPMoQw$)V zvf-g_`uN8}1oa@p(*(rJOsRypn;x|!LQ`)77|D=|io59tonr+RRn7*mkRT5tcDm`O zyF>Ja-T>`;M>pN{fe~KaP)>ZCThl&Cz8u9k*jp;~ad=|8(- zO?(;BT}5dRrYmT=>4I>{lX5T4A2FI7LxAa~gC}AN=Q^~%SbvXn?xsC6(9wI4Xl@)T z@+{wi3;n)^m`p{Li6E*nrIO)pdef>5O*IG5ks*~9chmoFj1mK&z03M^ia4N~CUe5w z^zC&RSbYn6D-P$Lp(@bbbSv(rqZe1h{*S=Un#ldSX{vGIc?VVC-8m~=che`zprWbj z{krL}6s)btyif|_u%hS!ayNZ+J(=(d5Na4vQvRK8`s5P2omxWe%tQf-gS+WF$Ad&1 z&S3vX5aXFr@o+ai=y0fy`V_=+rc_+qO}9g8R3!NWfP(~;q_o(A2+u84g?Cd~bkp=m z+_S7-Uwbv6uif!iNwJbDH7A&SIIQ(lFn%Pyv9pZWLV7Z^Dy(x~JN9Ztr8k4rie;4( zO32@qokd2YKCVzh_q8(s^CK~#97cG5E|9AG+Oy?wr6`s?w48Hso!gV9cpNQbTcFN!R#8E1b3`&48#i63m6!*13S16w1P?HRirLVo$A{6^y zL2PCKd72X9zV_?yV4k}{?9V{pBXP5?sX8Yg0GP#)j>Yt~x$vr(Tn=%y0i3>eB#P+t zwRftPz`mC#*cp%)h`vhs;l4H~)z_Ckh9*(3TIaqt;b|jnLJL5Pb9B?!F2Q<}a@b0P zR+Dw^Yo{mUyU49@z6)t|StQl7)7Li0#PsJ7i0?7*J5hms?W0^()g$pDK+6cEQe^3C zshC;%T5df44UU?acTHPMcb=P*{5sE*A7KpRO2qRI4r|@Zbhn$%v*UMo{*^@_Bp6Xr zPMyy4!buDmWgVzZ4Uwhu{H2r!`#(YKYXJEjC8FCQI?s(Ks)-#qBjrR8GnrBeap$=u zB~eo=0BmMRMa7-xlatYciYku*I7850N{XLx=T9i8uiOJ@j~(4~o|PK=e0rl$Lj`c? z@Z5PW>j=|W9_K5Qrs|42&)^GqQKm7@w>FxjQaPPxMnSje4Q&ML_vnHKcAlywbLaUJ zMev-?GpjwhfaZ}}a6*#mJg=9BrJ!1HI?vbsz%6HbM+Kcy>ZI?vBwW~Fo?Ob#4M zlOdY{n9j2+W=3RjXi2O`kj|ZFkx^(s^&mE5An+Nt?uyzXH-bpn3&dciR5ILomTy*3 zQ&RxUXGo>Ro#%!+ImAk6+gXpMh}Dgwc3taS6k~xSP|q2{uLq=R6P|6T1Mdj+S|uJ- zPbH(asQ&zVz{4j>A+!0R6vAO0rK`X_;K1SVXq1&8)HI@`OgKH@$R=3%kgcJ1W#Tl& z!9AdyUqqb88GQc(#QRLCc(@1b8`;=LeFoxNrc_+q10FwQ3o4S_2jCDvS1Bzv1Hy9~ zRgt=@CT@OAiyuyLrp43yqwdVK_$>r|{6avir^Qvg zW?J065A$TNLhflo@U(dHnFK)r<-7Q20?F&BBF!e-olg~2$;ic!zGB%-i{H81Sj#&g z?PZy!#UndJ;Xw*44qDBQJwIWHaQ_PS<`brVAQQD7M=29B_3r+kh8(eKF;Far` z9n!v7u9)bKU|&#B`r-G_tBtYz5a=(IfsA>GWq+X~Zzrn=t_6lwQK<-2esXsjg=f5< zq}}aSZiPoy3`mM`$6sjQOL1ta@!-nguzDghj=#`HVR#m=gY!*|Ci@Y<{z55tQA!V~ z>@T!n3GS~UIP)HdpR%3vo&}p;3(C@hlix&ZOV~Ao= zP!@lo7Uym87N{xDA=Q=ADYvuifeR4Bm%!HlU)1pz8rdDL5eGs1$`suII)Cgh^wnrQ z3jGD*KSOEzX#9m{ZLcccryS?4>=H$A=&DearA#&X3&r7WJOz0H)MrT5mxkak6zMLa zq0Rt$F+_QC{DtCL;zh+V5T`PrS^`~F$6x5e=Xm#Q35eB9sc6_==&Nh&0Tdr(H#=BG zk@5`i3nqx;FLeC|)9@Es+uc@>^i+Wp01oRzN}TWK*k5QAMnIH^ z%z)OE!?VB8SXddyUuebn5CzG>AV+X)>@PIIq2Vv|;kXt$j3UGqhXn0`Ta>+JP)-d(8}|KZSOcN>49nEsWtUL0B(M>qaL zXHpvL`0GMz=;+2@sK|_}T7MN z=M~Udh=*2=bk#iE#$RaSQs1X+8bE8py73p{z9*)rfd9#8m%p@ zNI4VuqAWp-zffN%2>gZeH-?KaNVyOH9Hk(XL0>%>f1yb`;Mi~z;)6iI_zP{C>Cx#% zq`5?X9DZ3f{z7N0!aAX3=+%gE5@LU$L*$d~wF1N%49rl*%J>V7Sb^TB1;n-nV1JCa^|~D*MC!LiN)0_iL=z0Q54D<1h5`z4EI5 ztZ~q08eO*teVvWJ(3|V>XmSOpZO1 z%cn+C>@W1&`9uwrBw(JpUdCVO@aHwK4G7LRz@hIH8auMT(A;|kb%r}b@5KyNCHo7# z*E&cuW1-IoV59=sUufgq6tNV*CWh3wD$rl(=SLO9AppMxLNIy0=OG#ysuv1qP%1pf z@z9X4`Z#-<*m)|dIPpIO1%P^MRgZv~G|ii~RSXBQdrlPD22qYD<1 z+pKkkbw(_DE^nPHstdk=tDeKY8vxu}ZbeBw33&$)FZku2nBUJYXIzo8BIG;RuUTLg z;IJa_moo_&{5M{KA$>ixEv)k-WbJ>Gm3|!3S(a5MC?WqjAuysx%YvBP2hboIN7BdNXtcDx&zYwb!z?p;$Ld4D_WJ)NwaZxqUfT+Mp$j`)f1r z5;DE6-z4PV9q2SCApVai4Y?Z;@+4%JuQ8<~zl5;Lh>~*ZOhV4QOO6HmpdK|umPyFS zw=gAp6XF8{$S_KTCn5K4tRiyajFdU*VCoNt%7spdCn58k@@lFgfOLjbR6Ge8>8T;8 zsIm=!P6Xwnq<9kYrG}Y$5^@-{(T;8=A*ZwOe7~*_UrL! zAXY%#%tRu^!SiZUdxVOLI3wk85EqzI@$kG_t6U*I>In!N8C6-4DlVQ^d+=|Hcp#Cd zH~=LHN~5%z)OgrXw0H+2>{OhqV*>llGE%(=PhCVEwW%-Wi>X@}0*<={id9+HSE+az zdA5`i-g-4_>AoSXzF*%!!5p8>fu9x>@6pv-jnf+`<;fHQFHm`h#fj;pAAxp?b?zJ1 zZfUIadypQota3sL`S%SY@OMkH#8rL+&skjAY3zDUMii_H{{tBjgOo`EXBZ1%5Pxbkh5=! z>KkeSYD^%NB1_*u#e81hkhQ^46U(T-68jH~K}@OVF6LHi8EZ1ps7Q0xGUlJf#aat! zGbOo?;x}s~te+;G*D@Bi#pHfYBw7%Miad)W zz=?W=#d!pgvI>aWOsQmeEu+g%UQM+J@H#^(EndqAY8)m;K%2z+Rf^bI%Q*N9k4zs! zU1|tFD?=5CPeh^|86PjrDLfm&MBQkL@+BX!!dwk8S&7M2EYB9dtMgM`3)R*K=Z6_hCvUFK=sxJAKZG`y^()`| zU!6a8W8QcL#LWx@t~8bSuabyHny~*ME-zS5!q zv@)#ULSFRMX-E~2oJ)qL{f9wTCWsa|lxv3kAipma+?*^j3iOcUgQUKwh zdX3s$4)g47W@Zap3a;E%e3%Py)_X&ALy3CV6SYLRDu+njKE1g5szh32)c=`S|F24{ zc@G!k2*RJDv`SF@>km#$ve#Tks6TM}k&$Fs(#=;T##hV~$*S^lA>LRVD!hiwS0yGt zf;&~B(R4Y4hccYH@erCTzIAbta@C-)f37aRg{te0^L;2}PpTP{tD()(u>OP79~w#a zBi-cc?_ov6V5)if6@WE?P@ZD#Du`@z?2bFJh`A5iQKRde#nO^MwxQ8|Poq|ZCmm|S znorQmG6sd^XoALFb8IfvrbBz=tEPWYbEh|aMAU|e1b`i7SSyg_t|6?xnKHUc`_lIX z6;%yVV;oi$syMnl|L2aPbfOODgM(35DzBuai9*LiLw-dYsIvQdrYLkGlyVhPC+4-1qEMoYjQIeK>N`a60}gBMHl#cm z{|F5v$;is(^Wyu*5U(>Zoq*`CL(7Ol!=yd?$A;n)3Ws#VaOg-P`%^g;GgjIwKQE=2 z;?R?rNgyWX@9ycM&=P5XSg?_(3+)BgDONHnc7F4=)ZDpuHe6-o0_Ev_s`)IucAA_09&~0%e(`&BAsHi zUgWMU!+V_tU8%)S`GvcI%(Vfz2`QTKYa`)46h50m{ge?d+!-?WQ7Abo9oicr+?kro zSIHxC33o%8XX;TQ3RhaRtf+9mB=al;nG_vdELpf)N_)l52zAZNzkqQr4r>^amnsjM zX(R0|_p2+W1@yLt5k=j{lZ1O;IF%&&on%|MJ4n0n6-$go45NXKBXlB_Ci>lP>I-*g zX>a^8L{QuxLt1KNNu?4oJ-UMFE$u#~@{6s|4zW%l{Gmb8E?LDRE(5q}ATdR*J&HW+ z38&j4B>GWCm~bx*a!p!LO3Xy~$Yyx{#i7fmQjGp4ytr_Wl6JvzsfwuxJc197r-Y5 z!n;`g8wmFdX~ztyB-Y~eDI-ZTka8^j^9}uTfd0v#e?CR(B6iLueCGIA0Wwn3L!rv5 zs1{1Q!ZWqGDKMr5e2C2)N#40+eqpj#HR+(9fcYiyPLUOLt3F4ay3W) zheKBbWlyN3Mt4;iK9sDCdnMfu52gL(*}969^?)}rLG)6iJ57dqcovxe?Z7#>AmKxo`mfh}hb{&2Vv`rjdz5eeGvFt$uQnY>)+G$6p z^}~>g}UguTVRBQL+4Meb-8m}c1)D^V2~A{Rdw`c zN2>^TKFcobPEvQJY!0oBqtkMpJI1ozx8ijC1E39a^x`=}ggef%|4H+x<|n5^o9pPU z$)?9!_U)u%I{X@Fn;d=U)p+49YuQ1q!nJ-B+G$5WUZ8+*Ct3EBdfBvoAKEjcYkTSb znv6bPB1v9URYUn(;z=Bbwu77KO2xL%jS=IK{4`vX9ukZFvlYEhH5oq5xzbfE`^@RO zq7r1;0snNTNR%~H##CR<8kXG?S3p7U0GJ#I6}wCER+r(QIPuoD>>azRD^jilzMX=o zPRK9bOv}C}%W3EgfUAL!ZYxbKyH>vx4T)A52jTFGNv;37TUz!D?F(qAIDn)q(5zn~ zg}aSqr~eZtN>fjr3E%|-33t`<)rI?2yl>OIjM#?SlRcpIWj#Xa-7Q-_gVXXjoS#CP zswcvQ``l_I@UF`>I3~Bg$-V}#jv;kLG&J4iI^QErx9y(+oFPcnjD}iQDkR*Wxm@RV zCh2SY007->{*ZmZE8M$WuF4Y&i&L-R+C>8>PLRrzj!8kkxm;-#ONes>)c}y0C8o=N zAf{249bL~>$yu)e=)(|ovcj#PcPzW(q!10g4`2>Ml$@kLMbV__P~pB4e)$YSg*=EB z?k6tS3m;+YTOig}AV)crx_h0V?mh<#3-eKtH<_1o4TK!aX310?%;P(5}?OGicrETwQP^$t-st%P0U+e`whu!+o?P|NBv<`Cs`~?m}xm8uD^X;D8 zAH$~Zw(EME3|-@ofrhp*C8SY`)K^)NZ`&^G#|D}#4yq!PG*s~?8{4izjS_Vhnu2{cSK;KGx=D`&`aa{-g9x>_>V9U~1GXk>z78>IH6`Ao3vuoLK2|h^)-gc$4gVC>IK&m+t&tc2;N3u| zDhefHex{_tLtN3zi;Fn`z6yj6P*W*-2W$B53*jw};lY^z^_1s6T0mLl(BOn*{UrC+ z-je#qvg-u|4gNwGZLBYD|BR`OV>2UV^U;iptg@|P zoT!(Us|gf#q!b4|;w_x`0Eg9=p2O&uB!(R+RlLTLvePnnL45;xjS0bylnZCL1qGA` z@y`*G2hk&#aiknxEmS2Ve}{CBW#dTs(XB`=hoVT4ICN-oq*U`g9vVY=XPw7fHoZde z4G3Pnr0EN1hI)8ZY4IUFthWHumSUPg&+XIuf2^rJ@-(JIE zeL=b98R{F)P-0maLK!2Q zgY7qd#k}475T_eJZlQ#DhI;-cEI2HMxR!yy)yp@o`Eqgyz-fkbEM|s!!5Tbq+=2MF z0h}4??Wh-LhT6TggxE<1i|vSujlpVl&K8dn!MaoYTzwnE~+;*&mS&!C`$tHz}`kmYRh*BKaYNxki*Kk+Pex0G8x++dJ#wO)$9< z>UW0lH6-&vfdc!m)bR_%UkxCaQ6jp<;$@3`-IK&BGB)xdh@j5CTzvs)G!#@+Sqne|f;Lc6yv{k#3fHR;U7)@0=w_YsQiBxTM#eyUk9A(> zyqSs_u7x=N8ELAnc%8H5P|O5x!uj1slT<2aopaUtY~nPuo2;LvOCPw-soFfRbAC?| zJZGIVtL15eiu2M-^6EwR{9RE)bjoj?vuZE8k}u=Jcf?`sqKJ8&^YKuiau9^KjVRSz zu_mz2*{wbiv!O0#;zx=@uVCUCrTN{O;%A(Z@_P{bm{Rd*3NO018(-2#T?X+dQz|ZA z=RDUfk23SpgOX&&VI8Nm90Te}K**@vTgg#mk{vb}$X2UUniM%g+I(n-cfsEmWk^Vy zsdmyjDHqihZKl{c?xMCY$!AGP0_Q7u!GwrvMebKnO9?ru%^#gh{R$CL+iv1*iIDeX z*!!=*hMSKOH$Zs)7;%XX#l%=du?mMwpwZMW{Bk}=Xg0$4B63%QPMg^{uuEXxam?JY7mq{qZ0lY@t}#xDx+;!Dz*8u6cO^Lv_k$a zCr06PaK{#Xb)Bois8tf>mP7@?M?5>hSsfGU*Iq@_>efCdT7&s$bWV5_$yJDA0}kup z0KDj7CenYVZ^<6S>61p1CrLLG=?zlqigP$2%aj5vUfyr~_ zawCy#mes|Jw};c9)x#0`PJxD^?^vlpSt?r$a-GF0XHDI5|xB?@$HOoU_uWvEI zf5t$IALG>xswlZ-A0XkOW)@X6Z-j1i^q|L$ZagqZogz27r$@pB`iOT39 zmQ;J9Jc=I8@tD-(rRa(x^;oSUqDp4x&>MZ=>qb+9nO$rAf!|dYd{AFxzLN6}f{X64 z6(U9Ec)MyFOc}~T-S7~M!}^6<`FN?sDR#_>G&BU9Z(%e^zvQoUcV41@ZV<1A!I{Go z)pJnnjLVfpX0I|=PM}45x&T+RiJ862(LAE`44^YKS)2Jr3Qf~W&p4dX+4;!xR*`h*I%~&f33(VjZ~+fT0YjxzOj;k&gJX-TWDb(v!h2 zO#9%<(}Q${?dGRtZR(0~J}%b}@oX=kQ1@=;wuyup;ZCTggWSwPyMu1HP2 z2VfdQl)iA&#A0SqX;+__Pf%Ns%K@w=NKzlACMGjWqX1*{XPWU)jqnOQ;F;lF3odUy}q&p34a>hUgfa5;M7)^om@HKhI5{BXs{ssPgo zR<}m%mGm4Ub4ZGMs8$oinYE?8Zfv?@WH*4lIc7BnnAup`uZ*j$BO4EG5=W->W|-AJ zP*;cl654VOucq-bvEWo|m(GLy0op-FrzzjeF4Fd#%%Q_ygLd1|Uu+Q}GCz~{n`_Dm zT9uP_Pj~>~(3jt)>Q&FxnLQIX;Tc(0BXjqd<%`ly;ZT%&_c;#&9N@ zhx1<;P0}4F+|eDqBD0@mf0jL6kO`IB0PG=1-x4&Jo;ldEyKkx{$e_ye(5^ZrM2eY99c@2%?+D7n6e4TlU5G3hGJK=>R^?5)&-~WNx+WN>~6_c5x$s ztp*ZqS_H`4W!Z;*ucY-;(0+4tS_H^EXxT4%GqnB)S}^)tT_%cBs6~LxqhGM>q4UBOC|>~GmO`iks%T?rGCy-zD$Vix?8p_%{&-Bf z0_7OclQ<2X5qd)_^9Cl*BcnC+C4g0dkTwzza0SNJ)@dIAaDpHwZ4EuL?92P>>X_~V zcoG0~)P_+`#sZAY`St`g* z+iVND5#p{uAk@tHwWVo1n%Ham>0Er+uL*vbSwP{rgi%h>u+tceB!;Rfz0MW!tPX#s zBK_!T_aig}seO`0MJ~x+T*#t*(Fkx@6L#SiG~Q-q2Vx8OzyzhwLeicg2X>58yaM zDlPVBYWUO^x1c>{{Ywhsc$<|d=oK;?HINU7Dybg&s|UdflgLZyU(|Ngl560|m8CW~ zc4eWNN)h#GYMR!_jfFMkwUpZMCLSOQeP8ZBNSU!4@!9kpuImdi#^bQwBHNQ$ zkffcilD**aKOd*RG?LDa4!X;gEf(4M4%!b!mt!dWG^u=7cIP~ur@VL%XU;>v70BTE z%ebbrFs><-YVUYlEYi~aHRV#{t@I_iD-O3pi@k}-o(zu= z$s?`E!w)K8Z+XzK;;_zA?zA@AvfsR&DXif*Kh9|K8UgyEVp(X=!4z>DXRJ>Ed>sfy zG@um5S$5U=x$&t*hzAUyE6M!=0TV3y*Zdj!LGvnr-whl(sFG_3oM9#9IvGxKEY+325}(+6LtV#ucf>fYluh$Sr2^&GyTUR z1MIbwKU(tGEeGu7Vx#FREIhF|k-BCtZaCzsrshEGRZZE;B0KWpdL?IhcC7N8lcCRk{ ze{nhp!IUJMkZy8yeRX!x94DkKNbv?jW~!j^pSgvs2Cae7b-7B|XMule$l1X?r zqbNLb28F!~RQcSsLk%mRJN3iyxhq*e1|WRy_Jb_f7XLUtcc+GH?}L>%yB&vBkt*)( zNZ&uYnNobhL+M1_yny*bx?z9EUu=x%50Nn8WLedEPYgo&wok<=ZoQsA5QAyGfj)QM z!svFA5F`zUbwZh3{NuFV;TKWo%^peQXhS&&g7wZ(OZoO0h)q0g1qlXHdkb5Y;HPvo$m}fA$KAO90_gQXg z+&MdT?5-x_q%!l=(nacmOA*RJZ|6Ix9PjU_4jJG3j{5KPJ@wxkf2se{?$dweV*kc} z#zC*?KQJH0LGL=XDt6H8eH%18=)H`ru!CMZ3S}Jh210Qh^oaXk4th19ISzVr^c z%D>?3uQ;qfs12(;u!EjAK2$M(Lw{lzcF?1eI1YNFGsPps5H|#qKsc;tR2p{BE7P)u zpt#c^H8irM+&T_=apNkAPSEg=FL7EO966g%h*ZIKQu zv>2zpGLiWl&fuWewmB97w&MJCqp6zxf276^dg*a#LjH%Ko?)gkVkkT8pf@8sZudeE zOEW;VZ+kpcr(i*g>yj9=LVC24WCXs?t^2*+Fl^U|6dUK+Iyw4}gQ- z^iA~rGl)MhK$l*Zwg@L{9Q4QpN)NH@ptt=P3j8OX8Ey)^7q_@yr8&{4aR@o+^=@1Y z{T&Jwk3cCLP}P)%;GlQ#{rnng44^ebs!9v^Yh|Yrt=tAdZ9Hc!!3AUZ1kr6eOeG z!b^ZS)Gg*X=-vM}UdL7uS{mufCa7mDa?l$vva}B00$N)R&klO7nYpw+5ZZ7@Hx7E? zMborC1KK=CHx7EIHWwA@qOXOv+0p4^6y%^c_L%Q0vd5sEadhLLw+x>Mqx(xffcDtY zje}m$J&%q*Y6QxULzmxe9Q4}!URgCiSrJ;AqZh_7t)JD@`ao#I z9o;zSy*bRM&w%!c(Y3u~2fYE!OY2BB0oct@Icf#QL63$3(nBIU=uLO7G#vD*KbGPI zWcdvLWE*Mj2W3q4<%ENt_ffotiUCLrgxEoEkrOW*^zxL(7o3>@^TCRfqW9sq~4K*m9DeuG&3JbnwnT?07| zdi|Fc(GGf{qu~F6Ltk}v&|6}c6|yAGCzD1kUIseoHJX#DZ;M6%S}~-q2s`NQe8Z#L zc0T}b5u|EHL&iaG?~hVn+nE3sWr2)?-slmTVm=1*ax;KE1gSjfn2dv7!&BA7B7)8X zxRoU)1O&OJ(1utrm1BcENb)b%eUi>Se8Y&H-Dnpcy%zv_1o5mKtKCmI6LThyTJWn6R;l%RmE2^tDLigUUH3KVVwu@ z;Qvy-o60!oCrXZz`yl0!gY4NqO+aZXEQgHOi&K z^aekM!zhx&2zJoRmMgce@%ccPF|Hi+^i?(vdiAyiX>t#!<4jh?Rgr!q9Q0PF7t~p} z1Ijh#{}fPVThHQ>gWk_$g>KS?fR<*QdJrE32feu0>T9SDfTj#l&*X#Ppx0xFE{yCB zU;slpcg8_4VL@&k(_{b(8KUH*N|PP*Zs8llat(-0hH@PAQt){S`7_R+F`B*(?4b8` z@9N?XG=V^22yEBnLej!6*m4OI_%J=KtS8Z_qX5&O?tksXdmKQVx2X zF)3jj^zzbR(>UlILzbTJp!X*fcF>#lR%1~RaU|fdmgDhC(l1F2JLsu+je}l=({Ng9 z3Z{h#!47)gPx1;1DBplIfaKNmyk;EqTy`CmjGP2%8q3B(FI&y}T3!Ze70cwHr=Hk6 zG=@qoQanvfuyushGZSpp#=w=}B0^lJP`zNfd4g^H0=(Yx6sJSLD@l*E%mmw+QLy?2 zp~V|r_NVYX!S?KOMKJ_tYC>-m$fyZ6J&S?~Ha&|%b3{?|BtEayCNd;!Gy>>n@|+?0 zMpC>{@4SZ5G-+hpS$mhX9&KAhi5|V5R2Qjd^EMDwT(5<;`Ovq7?RvddNBmYx*skt5 zS0k|KKK8f}u1WTu@$jpY7jf!34lA8n=Xj~aS@!jT7;*oH^ZCa4G)cc?L$Y7Z=$}@^ zt0ipL+lk7)swHe!|1zhp)A*t|a5dg`y;+V%+R{^*&d?IJYe0$`Xn8VlMrUWOy|^!B zrytI|g+paWwx={Su)*HH1*<-CD#Qg0bS1!->8*C@0dV144e@&hR4((VLhZF%4uhk@ z5oniKrxbKiH1vyoY$QBM9|8y&Z?ZwD_(K=%;_;Y?D+s{Lka}8pULEO(Z`ud@VEU#R z_|6;xR{4)jtm>3ADdRJtGDhw8;$F0V6;UZw`7dD3xC+$(&qBYeB zKr4pSlZ{%#c9oL$H(_-JH3r!az+i&<(3Mqd*sdhJN>sL#c88k+ZI+|c*x!Y%pubIM zqFSn432nWj)7afrRoeG&VK;M%|1h)@tSbxX8c~iqCH+|Bsw3?=k8>zS{s%DleP7|V zS#ym{QT>j3hH}yR!oX0)$T9%S6RdMiYg8_*QJ;R0qVYLb>jIW18wdm}^5^M1GhB}xv0UyD^p@R)qS4G9$18>dO4$*lp37|4V zRAo8=9bI3`uDLZ;lg&VN4kT5aY7N`cUN?%X0PZsMwyGLh!*=yapr<71AyKVi zyV8T~VfX?%MI;MOME{3FSB)yQ%7q$^xH5z6(>*Y;SQ|u!q4XNIt96jQE8Tdm7hUd>gcpQ?V25AkA1m{ z&gL#?he%f!UX@Ngcev&U*#)O$=+<)szm)DS&)xRwUlxmTy@dW!e}uRh>VIq^aC zNa5NMWWRo_s($WB0Z^SFzvL9OC&*5p5~ZIzS_9}B7*p(=_wx!@zog-qOoh88L^pl- z(RDJ&UOK@m#-noNhd4Wzg89X*ptC`ClkD|0vlvZN$?tQcR=7one!@P|`qpo!#|skbOS6n$~MVtMBOa zEbj^mwtHk$(0XTRy^OB9SiR2f3JbRXd%L8D#sQeZkZzx}&h8qOq8`(=G0GEc|MYJT z1SHHh8L6M!7X$dpKzg0s)i~Jxs5!n_AkdU{Lfgl>T4#5)2)0Wt!&u-qoWDt$ z%Br)@?&=h5*S~`ypupfM6o+3~)H=JXXRsYyH%!Plh-Cu-9>c3u0~+6{1$P>#s|9x& zx~m0u8r9=7)fi+I9hrtvQBo)k(KFu6URXqZGa64aG~W9Fk@I+u)&xb=kVDw*g|sF9 znp&arj(&(-(~BsTKk|DKg;uDY7g08RSXtDdA^SOmzf760OXF>R5#_&$ct3&k|DXjU z9D3uFfT^M>nNJUvzrKN=Yz5OYNQQanBwVGZ;A@2cQ7JNpJP!1!!_h*;aFQP2L zf>S(&^Lr6x?Iw)WKS1cIlw3>7Ex(8|whw*c=qm_oj3_%3p~fjm9|cG4#djvVQF-Jc z5T^qvei7y50L-1ryHKAP!Z*8WUPOtRSzO4RQ{exLL*+vDriA!Kl&07gOQt}q&OqRc zC~y8-S?8oRfUXSbSj>wk*Q?`6X(+@|25??P$wbzi7f~8+sVUy1f_({cInhHYKl~y} zs|CK#==}igAnW`h%En(S>QU7-Xty2Ryol1JXqQT-36 zJ>ZoT_yCx52tIBT3$2vVK@_v6@|UzPpZ<>!J|LF%amY zlkAQWT@j^}86cW7rIJxk2;`!3!4;~hHvkM}NTtOtIxknoE7{N%vfdPV@pI8hXUrH(BR*I#%Aw4H&<8dJjGDgteO3V;EyPZfvPUn{GB4Q zO|ma^MpKVKko-85BuA5OyaJMv8;WtNq$vQZ2SUbIVc$B~ho(l;*%O{0kyzBG^Z0Ty zz399LpYL}TA>-yq2&R#Hz{8q{LzD2+u zR<(-7!~;kI>8OBmEy+`5lYFH`On>X-&U6vU+Gf&(X8-~erE3<`(z78xQY5tANI4%7 z@dRW!RkCu#h!P~kjJCq>7Zz=yb*7lMtMrvJ)>_`Rj2MWsqm3dDIPwS9rQmom9n!}v z|Ki9WS($&96ssY9&+0*E^RTT4NE9C9{JXORUqxu?X(S<)&5HZc8 zZlGiyDp7ROv}h3%Deau`A>teoC@~v@ARN|ZN=?gA($0UsoS@v-h18JbKS|C-Ge9vB zGLLGnbUvC^;9~OQ13ELxDsT`^zQqCR{8=5RfYZ z7R#3b`PfMX(-60E#)%YY)mc~duX9la4(mZx#7oe+vQFXsvsxEk-1r{S5CHEnL`CtX z^as^NrnK=4Ar=5wVjz97Dt*%eg=!{qm&0`u(JRNl8vm5&tY~Bw}8X_SWi!gcXIGvoH>I2{fRLEZx}7z3d@N=FM9- z5AK99GPL;@z8z-PA^t=Mz8k;&aX6@%t-+6Fhh5l^pMIs%q$&&x!uz^bCbi+FJHjt?=Vw&KPE3D z=EKh6Pr^sJ0rHo}=!96FYA$ve!HIe^iPd;{uGCW;Zk zJIrj)Y>elBh#xaRQRuXJhnZj21&cKR_Ax}UD9Go>)4GcrH;`qYa#`XWsK1^=sw>4i z%zRfHUB71mx_%tG=p3oD!_2ZncwkEeQG=lqYA0nQsQi z3washeFpfdIy=ntIgMd*wuKm_;LsJIqTwB8Jf-l^siqWSu*1ypN%-bFNLda4)HXr1 zS4qsL(vEOw>@d@_RI-9(PmukHPC~B#Kf=B|u*Tzif97Vnxw1$immrBXgd`-fWVJ^k z1hv#wwU*XeJ5_saMX9xvqNPf0MXjZ_mhw^5zSL4$OKWXywbbu9Gk4B=uj=>r$IHz; z&zxD`d1ubdoMU;IIXpF1OPdI6Doaxug@{^S_AZa(bo}MO)^L3GFw=$_!FOQ2JR&03 z%U-pduOi9QU|wKp>|v&(#fFEO*f;HpB<%|@3C5%I#vW!Wjr7;jN&|~0T)mv^VP>*D zLC4PqR^N(mJj@IlT1w~rO<+ANoDMKV4EC}oJ#lD!9Iy{8+<2IIw|b10{{^s>7H&Mu zlpa?|<9mVqXyK$akC^UdkBH4wV}QI0?52g2);t0##B7*sRQ)VU>_fs; z_fQ)H5jVW-JNA0(CG3~LzGB>Xn0f1!lCb}V;KKx2Ib;tr-(gZCm7*~s*xP>oN2g-S zN1%Is>KVm&m^o}kfrpun<5Lt<#)4j%qEH51^)UDhraH0S_LxGIg)J9gb5DZtFmtkS zb)9Z6;O`R0%Bt}&({N3+PG|=31)g#&4>KDZXhEAn*lq*~_Aul4$f<>#26n*+u{_Lt zA?oY+e*=4N#Wx;ielP5$;}`r4Cn4~-n!|XQ$vIM8(4s}A0;@*2ej|*Bnb0?4MHj+b z0&C0h*~84(u2`k_NB9U6*VR{yhnWgq=uxu(e#(T(sxui6GZSxNu-XW4n<21=nN{lp z^iXmF&=mtq8lrS6YP%vLkB@!xI@b^;7UAd%9^D>_sgl#o!anvdz3Xb3Wx-1GWU4_@ zGb{So3zUn}%%)&<@nl+qu67n6MLHBbONCVdnVv$-*`Wj1QQlz9v>31sM-BpM4cCY>Pl! z#Z+Pm_AoQ%4XpsP{Q$y2L$W-~^m~@A>d$r!*dv4M9--UWc$n!wNjHZr^mE(<#`CWp zarKu35pmkrp732UJ+aCFE0>r&Ajshn`Gk05OdOUwzxvvvj|ORG53uxfO3kF$!%Y1* zDrmy{M95I>WjxG0a~2i01qlC=l2F41&05*ROy8CHbZcsdxeEND z2S+L*dzeY9m`?=Y+ZcgIC#I%A`fN;%T3A68vD+_nw~0g$YIu^=Y>F2MO7pFj#Eul? zHz&&1A8ulDiiw=S&b37RyPKc_%jps@9x{|s?RGe}qs(lpgA(zdufg1PSZcsmz{|ab ze}sRt#MnB4n*wrGhE`34AYGf7F&q>IvC9F?a??}IRdNp6yENxNSum!wr7Y-W;D z#VK8FuM)O^%~!E_J_PU#6G*UD7Wb-X3qD&lUVlmNf$*3}ZB>5BM0&5~HlIU1N~vm= zVW`m}coZ&Gefu^&9W8oou=#YkhqY@mR9UGYR5K)z@8I?T(W`?kuq#AXP8gmjdUdop z2@h}3Zx5`DoyBF}?)Y;cMWLq9(_;UR`v}p8TGT-e?rc~DD!*2peFBz)rxm1c`V?q9 z*|n2EKZNt)PCH2OD{+`fa5qH#Ky^Y@!Ontz!dFeJH9i6NnY7l&Nr@;D{SeMNEwS>7 ztm_p}Xc>~L2^$-c^}LGEABwbUDw0cm7?~;jB3Jm2T@Ec(*pzG$xwe?^3PcZjcYC;q z#6gYI@7IO>N5py+k1gL1X!9uiBZ3AKAxhe>701D!ZU6@uf-b)529Z!a?AumMl%#kc zf-u*Rv=mKh0FwemvNdTv2s@ahcOa%YVnt*NSc^SYP$aLzXmSRGi$p3*S&1U;Rpk1> zcad?aHi9DCN&D3m!HOg8Ut<1;N0Y0**{Pq1dDWMT1GI1pS?i{0N4p zzek=Q0F=nQZY!#IaagL>gjwY5DBNrf2U>y|HHdLa-K~sVAoDfONYV{a6^tBD);M`& zb(n}eYFjp;j*xEw?2G5MW8I1fsxF-#5efe?dDheXFBtcXLd`lo*zx-bm_F9WvbM?Bl z9@T9869;a}M&$(5ScMOms?|~@{a{`(H>64e#z=@9Kw>LXel^lCx?+f;4E6Qa_s@;oj{-<$K#^C-&Ib#oxxS!Z@_ z6h2D^bXkqBG6e~*FqHzvufcOdRT2Y8P+LkuLye^ULDN8$##Dkj5frhte5#0QB11Y% z%oIxyyaJD{8$m4;)Le$t2n`TB5q!Wvy4o@o-d2X(c#O5gC4e`Xpk#lp-sGr`GOTp_ z2thfqBfFt^R08f9=q&ARD%BCCfh93MfN~M^7kW$7avTt=2$g9afK8a73TM?)kImTp zQ==umb6Lnsd`fg3;TJV1pvPCZXr@|ua7eLBAz}(d%p@^GC<%=W4OtRYPAo&{*9MW} z2-e+WL`YTv94Xuf>=@%8Sop}0$Jv;vTnBcK@#&WMarQPtQiUHf5{^eDEMM#%3++R% z78Y|5DG408fMV_jS6xBb$t6Ukn-cSRUJ>!b?luZBHsvLZ!KcB!6zDxyQcaJ1>>y4nmd_Qes3#RS~tlJjroT zaOsc@*&=G6z0J@}u?CDkEFlUxV6QT%w%7>72kC1erBuf`7?M7xswjxiB6w`uNPJI) z90}R}Td=5r&{_sjE#Z1w73b%WkX}$^wg90clT;S<&T-Vmumj7B3o6bK;Nv_wItQ2S zeclWca{(=3_y@{{?sFN^UgdUzux$mnmkFwZtlHLnj;h;qygsM_XOEE%dUXDF8r%ylw#&-GS55xz8KM5kf@ZZonVU^Pj9JR{XFu1Td zB*Re&YVhXBiw6d<(5(d!#xqA|jcMe6m<>+8`NBuBisz7b8KR@(_=FCJSq zeWLX81?!vT;pl|t5Pr>IFRX7q>RAupKY)&}U6EyDO9Iw6sYP+5s0hHaOlUy_y>!9) zX6oXiq76c1CI}6gq@-xlMQP6{RZWw+fY6UgZj!LR*>gQzlO}^Ohe@96n=wnUzF7@$ z3lk`HYkl(`Y!}d_Z+Q%aGemlW@F26ki3q_7`#T7KVlb)dn%6c%I^uW=t!LUCVin(^5Tdv;Eg<7f;f_oWATOu^M+T1-->_!Cae}OR5lVpr7WcgzC_$9FQ2G=<= z#ugSmK^3i(ulZpB$}_|vOH|sCF7gbGAOnEPOk-7;NkB6cL-hC`s=|~7tTBXc1V|o` z>TQ5qT=IaVepuERN_k^(NY)ryf?4i5;#$@iWOePIv>Kp>cx+p#W~u$uKYA!2dBQ>I zI4e7`EkwiaVEn}c3jHDCXLPEptTA-jgo>SW2|;X)A?F`t`{mXccBSL8#J3Qz509IyyP>%>~jbSkQV{?RfA>AN6TF85sshq|dLl2)=#VHLui8%|1 z!`2uI%`PCafi-5Fq*4_m3Mv487&9|e;M7KM*Hs13 zGO(}^m}OvLOs(2#e5tS*`~P@sZIQfGd0+zzXJ@A>PIKU|8x9*-ppvkGh3OxZ5?vu- z6lmj#+KWoV1{Qw(9y%X{dRElh1p|@8N#YqNU#c(VG3tO8P7mX3zfgmc697+FJ zT14`IRKHr37zn~RPf{96CfEZ~ttoaf4}{MQ$uh7o?@|G=5y58-BxwY7Gq4au((!b~FJVkiLViSdIWVkqJ~=mVt$hr}2sY6yO>rkOZkprNzJl(wvWR zh;J_#Cz+)xU9s4}!k$XdncW2A5wqL~FtG4hc}y6Bw&InR7bb_7lw?g4ym2>MIH1{OM`AQ|U2oPWWitASFs z3@mJ+QATZrlLw@)`<78mSqJn6CW>WXVXwu8fra6}Co7U10Ok;4(-2}ASUB3eq?R@d z*gTe|w!+E4!s~x{>G&IfZRPm-(mXsMU0PpA<0pXqY~jYh!knZmjXwbP)WVH{1^diI zp;{~dx7Z25qw}ZsV#&b5fHV4Fj;stU!@`Y$1#vx1y_K>Vu+|oC3@rTEGE~bS0Bneb z8v_d`mZquhCuaehXW_=c!i`(?bo>p#wpzF`urSvdtnm}TeztI9V4=}pg*5&EnB49v zpB^vSz{03LxtdfMgi=geKrLkqEYO!7{=WdSfrVpMOT)mzglu1t0$8@gKix=(8V%K8 zs+$uA79KP$sYzo%_|TKY1{Qv?DOW<2%DIsT7(TORJ>YI_w6GfoFkH|8BH<< z7RuE0(rx>Y2-ur+C(#`y@_?F8in z$5L;v6_pJvIG5H|-)-9+PeMo@ux?aH!je&*AG~;-w2c%bkuz>~tk9@^21Wd+b`{#cUK))n$*ua94 zYYZ$T!qK(-3h)LKgAFVUUxj-<6#EeVIZm)QEy0X|g`Y-dsbu6WK=&9n1{S6st)t;U z6e%Ab9UHqls{Vw7=1|V}=HRX?xxg3wfT66SdT}B((AMrzpR@Pw6J#&`0|?( zvtn{o^uIl#OZxLaZbm%)4LW}s-c@3hv`&$4MhwqQ6_kgXL<*;F$u}d;+^eE8`Z_~p z(gzh#E-=3SI8ls3u-ZXW-BroojqallR#QPliMX&N;aAVUbFdIyA2p8@U1Mq$6{(Mw z7f|jCzCVoZIG6i^{rGL%7p#V0c3-f0BX;}bc?A86$F_!Y$nFc4y@gruV}ytO;KC&R zl3S@}chNsT5Z|~jP=lCpUl9H`#y&?e(2e^7S_DW(1qyMwFQ7G+bks(OabNK8BPj1= z4}=WDqp~9pQX0m6!HwfkY{|(0=P=P!Zh&!F!*{2BqRo`q|#O7u^NIba6GQDhu+}is%7S9D`+~#A;l+0^XurIKYDw0P3E;k9 zd(&`nhRQ6TB7%LtsRz{@niVMRWv*6pv5JEc&n(p@YRyIN3*L#VrCHe^G-i@obFuq^ z*6q>-^#$1-gg!*NNDZr27399)M(u)HZ#xOtGz+Iy3Ar!ujSf^DRjvTG*1~C3LGBA0 z42sbB0boZMS7S5W7trTYFWAU^LHk|NiX;C9@i~#zt6|(1s4t>g-I4o()?@1`jx6;* zC=KxFoT~{uxi6?vBtZ+y0+vg-5~gwNY7VQJtmAhE)}7<4IXJm5SlPaU#>WDiWZ|@4 zBliXAUl-H(5@0JVT)6{SEbWb-r3q50$vwb+BAn8fe^B+Rk0RU`Y{dssItWw;70G?U zsdxN!v}XwQJ7A*SRjs0A@09lYk4o#Z6$c@SNmOMz0WI!(+{{^5NHcT6YVFBXGTD7W zowiU0_W@%lvvgOXc?!8N2!e$GYC-uC2%j;D64f&va$gWoHeQpqg0PoK6wgH}YO}}q z2Ix-l3kWxu^iWj|nSOx#0$RUG2Z8Lq;8u>%vcrFbCIpYJQB`V{3wB@7DWRy4Rl&$K zEWJBL?h9sJiomB5;hhX7>5a487d(uGeR+B%hJ!GMNcxqag)q4C}p-F5DN`FIUyQ=T{JJ8zEAYjQfH`??mZ(ava3x0FTOv zi$seZa$j)FKUA-X%7KtdB)8-g3GNF@_X^M}q9!1;c|nqKUy!)bUne&Rgkgqcxi4t4 zDNEzCfz7vY#hr)0qGk8ql?wW@hus)B;V}m8k`q#H$0R zG`|;)&oHMkz75o5T};KK0Gn2C{)E^0Nu2wp)y_6J-Six(_yDt2#sr+V(tPh?>!c8Z zqYWf$5Nr+~r@ynX=k{kztFQ4&#R6a{Z|HC7?q(A>-#@p{qSy2$#(TP#g zIz^uG2DI=Kl!uB$YC_e)Gv2_G`Bg?6GQ@X*ue2>7@|nB9lxcN;kba2xV_oX0yTH{> zV1{At0+%^z-32a(-+ULC{@}a7^atMsF1kNO{D7nn;<0t2oSD18zK`>WO9;Jb5ZO(^ zuDihHq2`c~6o5xXl)VY&yTHBnmBC3qgAF81I7UF0xeJ_F-t8{1-!W_xs*;r>4&Md7 z{aaq~CZzNrNkdiouDih5kG#c5gibPu9B;vsY_@@U#X>;K8J=vxAKDtU$t$)4+RyL| z3!Y)?xFB1c2XxB=mP&-a3*4+=Y4Oyct`2JM0(U^k98@B{3mkPWR(y;Ek`S1J$F`7C z$^-$8!X}3;fT#8t{LK@Wq6wwD>OY z(w=s)8NuHfNUov8_%3kGIW@#-V81dB#gV~r7r1VXg5qyrj^i%T>W-@x!gqngXZhp& z0~lqQwb4oicY%E;`ipA7>N2kCpYH;f@8d7p0(+ZritnD)+R|R^VKp%tgc(etqDY;R zxeI)wR35Pmgs%-rHx}OoPD3MFcY#+ottaUHI1Ba_6GN3yM3}q4G1zv00?Z$;o~m3) zF*J$q0+)pK3|lcU;+UoCQJr(47u$N59W;_TpU%P}j!tiPK?J?ajnmu26okw!2~hdr3&VR>#VPvY|DanJhA(eN*OcFYw-b+TH+4&W`4@+29`30wzlJ@SUT1bo`%zUE=tBVR#BP0$&)OwKQCjWS+B_QsB|A6JHpfVX<*x zxN2OuBFPjms}ft~jV}y0Y45G2wE)(JrSXN~Pv4H!@dp7LX2mxbh8spF>%7kfHs8X{ zh2il73ut^3ux~BgTo^9&6C5)mSf0W^=Ple^7@juHmHs1O|5&)WFnp{|H8lpv@Sma1 z!{aKSxiIW|%c=1cU{x9C3&RckaTcyA!rKx?1Cts^tP8`3+Tdhme}H2=2vYYib76QD zS?8S##;1m5T^RNbtFN>04X~Yr>n0#C0CZuv`)HRX(X+sQVcc98zJC)3As!%Dp7Y2d zUl^{Ux-JvJB`IUx;EZ4Ch!;aA7$3S`EdNtwDd2qEH51^K?J)u}E7_w%;DvZc0;Ulv#b;X1y#F#Oe*={kO0V6Ry5&4uBZ zg{dNl%HI`OFDt&eFkEa+w1_5r0@RlL)h2ihr5!3cfJB;z5|Ee+}w(54u)h z(S_jwiRCouGzix{NRo=Bl~fVm``AmvppJYFMxKkN-1MEXE)0M5B@Rj!1z5@utP8_` zB}EHc2Eua zdW3Fgb78pNxnRw#1y+-nG0lbHLFK~q#Ht5aLy5`n=?fQzr*>YyXKW7eAC0`g$?d&RkI6iwZ zc(fc<-HON;hPz)cEUJJ|pGj)AWF7UR3&Y2oKn(;!A5T&mQnZ!?N`Ir)%??^OM1LHll;1$;P7<7l_BxGSzq$>eyE%d2Uks$@ zQ^v5n7>|3Q4}eL?Q&D9w#b>*IV{@vC{0J!oJld0^?E2{=t!&qi4(vKR_s3_Cj`y}i z1U}wNySKu5*c+2Y!uq>o>ZrFfvNrNT*i?5+o#S<;{*G&igspVP)DhnJhSxC1AJsh1 zRY0+HzI{}SfV+R##763W+SRBmb|65?!61zBBpKD&IS1y%vkd0aOgJdB&NicQScp_-tA0b)7voZ^--j}= zDmGhtsxy+~An@g?v$aoiRP<6vz=|&E&;L-Jg``&%O_@ze z>l9gacJ%L(g7WY-kq%IGuC z3_#fo-zJzS5G@$HPnd9ggF+hBS>?)ZsGp*m9~D@Pnwoi%$@MSKP+OG(lb zmA*@L_OwiWu@#|v4I+IHQtlNz$@cliTyYlA6@~*W_(NNnXLZD5Kt4!U$(5lNJi`|E zw4NvksEh|Jl?bgm>o?99T*rT?jiWO zfn-HWj8$hH@6{7w*KqX>kBaJ294Rw=P@N5!lp`tv%k;pt5LTV#3l7Bi0!AlhrCF&! zbymY3Cz{R-MINNf#dj`;u{r@19kt&bmykB({Tah)GlwsZ%nlv&^bt;#UxE z8g~Mpx-1{z#*E8f;slF&9>IaoU{h&Uiel#^d7UM?@33Ahl zjgQ2H9QUDqL;M%CVki6A6eNBJq`t7-vYuKoRXo;;?f3(JGZC#JMWr8Z|5T^? z_tc8fd*C20YsJPV=T`zg^o*-SDFG@!tQGs_^$jNGoPtq?}#FuNu@-{-(e-*5* zo^hH})#wzBR*brVbP$iVVy#QyEXQ<&%;(szQ!ZI6=Gz+QLgad&JDAat7_1e$I093H z69BI;fkf!ESu56~PL6m2Lhub!GLoW5FKERQg40AaSZOa|s+MA{Sjcy{rTz+N?fy-* zv|?{mLjDGTF_u{i(dBjiSSz+-71V?C!1%(j^m}WxVym|1h%S`l9l-x*I8s#>YsJP@ z!TBO0{RYAvCaL;jt=Rlp5t@|eCZ-p7bVE~~EUnnMZkUV30!(Lu>Iu}UmR4-ds|`>N zFj_N9iD9kSID8o_t=PqHvlUa027Q8wVrj+pq%a#=u^OuiD3V+Y<|bmlO^LHsENfVz zmUaZ#DVD}svD>%2b^JepJ>vMR6`TD4jlf#5+%M`Vk}P@)>pwiIRs?IsKC#%)isgiq zR3zB|%%;Rvd1I}Zo;ttYU4ERD5dKSX=!_>+Onu;Lr7*!B~Jb>3G2``W^dR?N3` zyv7d#J7(cVE7q-P0WJR)uzMD6v|>YY!!;gs8v_*{SN@Gw?9;>oY7CI&fK{|`qZRx7 zw;YW(0M?Xo){4coz}=dz2=7lA4NPhvv9w}aL!e8a2ym_kLF)cxv|`7PVeY#IjE#n6 zX~pU-DW|jW6R?wn>n1Q-u?fTUh%fH~d&s!aiZyp&s0+9QB?un39I{rdrs}#zEB42q zQHm+6fnN88C`K#Rz={H`*sf=_6;t*BeJDkt47%#UXvG>rQEZzIaK0zOXvI2BD5TTf z2z(cDtjaW6v7cE-Gu`^M*T1Zh~r3ly6X|!Tb z$_DEA)q!PM@r_oj!SGBSza6m7R(zus`}R(lphb%u0c@-l-)O~B`=*Hggf9fPgyXYT z?BUKt9M?qnUK7{VSBzF{X%nc;&H=o}gu!aCGFq{j6{B(g8=&`}ru5--_A70a&*_aZ zdMJqm6mMWjLzGU1wPNWo^hAT5%m%BuCsQTITCrxI>x zirv~@6z5+6W*CB{73=;%A=tk|cn5>&34?x(jaDo)JVw|Cf$^RP%hHN1-CRQ0<^WvA z1e!d$TG(jCmWPzVMRPEYF-tXtRYz<_FaMG(Vf!7l`%ER4r4=hUMqiz`<-LcGARgE2 zP`F1Sk=9iDRQ@23sy|zKVATw+dxUOhqZNBnxw2-q1gp!-m_{pBqf#+Fv3d`z>BQs# z;RUT&?Pd8j^J}o&reLfUn;M+035SU=LbaFCik-O$FQ8WtewUI^!v#$>Su0lJ4X4g< z9(?hgcr=HqlC@$72D`rL<$zc9;7CPetys$?0ip>A9hs!&Ri0Y0y}^0JAP~lTlGJQU zscL9usI+2gN$ePo34ChR;(_Yg!E}6i&9wtHLYQj@D%_t%rEJ5!Cr9rbiv0RT&D}@ zhsf~$JVl$$c=r)jMA=SXyf-hH1?(7t9g#4eVOj1%b31%}M zZVl3;g&-_vlA9##^w0B8)uiu1ILstZvl&5kF`!)mc$W#3x@9&aUmur&U4PVA2p(Iy z%C9k-k+R%N3~dP^vA`-DTvCu=(;2BJi)zyuuYl0ZkSw!+H6W5rXAmB)EdNzd)-k9( zQ*R&3Rh}&?+I;ox26(o7kPo*gKbZu#a9u(_KqO&2TP{0}{K%~c+C^y2NXRA_k?5w36~)gGD!&z2oV=ek((3C{oF(XFN$MZJaO z+44pqf6XcdLL!qWtAagShIUI9)EQ(w5E>CFom!SXTMo#srSvJX8?fFMPKyijYNp2+ImP}dva z+45J5jOz=t7DOwK^mz*9G9FjXY1u`dE#KZ4pz&y6v5c!FC3&{YdnrRFRvTD7!c}4# zhiA)a2SarHPQbcxeD-V^-Z)L;V}MPxaOHgxm!0o^TuB`CM)5ue_7&kIUuIGD8qb!E zsp~ojR1-+@Y$;&FMn(G>p;tLtu4)w}8=fuudf9c}&TaNFA!J`ovt(m2}it%iD{ZwHsY5)i$nM8@|xhi?KESXb6ljef3gh|x$ zE)qOjj+c3Kr`Q3)|CrQRRSjKlfM?6C{!}#ts>w2Ww#-=rd-hbshX}R*<56mr3-)Z8 z*#PH&BEcwOSbC03o-NzJGM`LEc&5Q5y>XUj%Nh>~3mQXZ8xYJ^Kud}eYOQa+ast}3Fq?0qAEuV+b*MmUz zY&oT6UD1MwT@c!jqPZ2tc(x2$9j!@|L72m&HdLv!tR>HuHGfLioVCEWF-H|wap2kV z@`PCN1|@a^_%Fn<^5|o3C0@8(SX6d=*La%Z%t;S`;ZSLyAl*gY%zAeLUsh22EwW523qjLV^t0 zf6E`|zX2{|LbJmlu=%Pr$BT+bh_?&)kIdiKmpNmI!{)0JchwiZ z$a@GLoe`2sJ2i=d3V==Sgc{e9=J6fEpvb1BP(Fh(F4JQ3?zdIH~ZojzHo^A2!Q`{5I*rFnSJs4({OurmBGxuIJHIU!L^W& zT(w6F#q`N=&y^U16A{-yxeBDP1!rrJty%f+sSItPSV^LP^xheDPT`<_=)7DUlSn!sK_7cQbU}ZJu9sS5i1-0pFsE+4 zC_rF_U!q;Q6s;GBQA38dZQDhguT`sUYE_ZiJT+8VCEJ1tMNGXwWq|B1#&eh>YZIpa z`&0k7-AR9y`>X)|GgirthPzoM`;Kamt&$B1Lh0Em87(&0D%nMfWvr5Q2Vz+zBi?^m zC94d~vP!nIcM(-<`w)5zk8MA-m&yZMC41|9F~zwF{4c{{t7KFXwo10?!-C>CB!v6= ziYPp`vs4*=^U6AZ#Ks604DYuqYGFy1KXb$X6#wmt730BE+PZkhEKp166mQ^z9 zFP2rZN!`4}FNnVk#IK30%BNCft7Nwu#VgK_z>gY^WtHsy>Vo1Lf*%n?<&jtEA4@4i zR>?a37%Bp>vWvvyMoPoy$FfTHC@E4TfskTImQ}Lw*OEki1k>tJ$B(D(J7ydHSw;W! zq(5uGzpuQB)Y&Q-y}i;wvDhkEg>o4BRhHY*L|I zAzufh2eVYAtFp6IvbM*uqwyXXlbPj4fK{^UTVZHvF+f^|>+Dj~>(c(s$r`IUofG1tC6I`)CPjsrfQ7YT-v#$!gz>6wgU|5kxA>5milT60DMK+m~OHGC^p_ zBvo6Q1gm6~FQ;l!7ZCa}Nv#Wl3iqiad^3G~>>Y~Z{42njhM-4Kwn`Q`1y>1{1EkfS zE(B%IvP$+ojWTL3MOMk8<5CqKaP(11t(>sg! z!@jvlJuNK`SQ6oCOi&XevPyROd3_x}7g%GCuPsf%D%qxj4vlvQ*2ltW&g2_lvp0yudzyY=WYSj{bU@lBnvlI$%34)`AD5I7g%EpH&)4h zcwK0`JFq?$Zmg0mObOHYBw!yIT#uJ*mF#}cAWd2W01EYz&~0XPzqY2B45^RDKGRkCeS^)#t42sGn%Be7MouPnJ(s7HSkqnNS* z=q)LVHAw4j0;^=BAJ^BUz90?m2*wq0(0AcnEBwBxxRkG+I*?N(>7KCj? zQhCyn6bV+z&X!Eji`0`KTzWy0u}W6y&pfIt*`9)64>Eb8kE~#;WQ#^s)1+b`Br=JT zlk_KBCA*5L5?duJvLagegJNqAN_UQ>-VQ4&TP3Sjxvu(d+s1-A_rK_Dm8?;|Jh*WV z+D@XX;w#B2=WLa%B^zmhC7XuRI?Ym~Ze?SYtj$gs zqQyIF+XmK;%%rJ`I}=vPoVP0KEL;WaDKk~AQ2|x9wR*@GR>__?z4b_+KN#PCJi0#U z3*sWdD%rUSu6~~mLKc(g%j6=#Dp?YY@2gbWfzXXfx*WzT*}{@3I=L|*%w&>MQmQoB zD%r7DaI|YV7;6m6vPxFKt01VCo-% z$E_FW>EGeMs|nvgKc8o%^N2(cYIu@DsH=EbB_o2dN=C~I$qQ?LvSUC&>Hr{3{*+a+ zl|KpLe|%P$pc#y^N=9q$|DRPdiQNj}pyf?!`-h=wdvGll;AVU9A6iwL?ZE@c(#y98 zuK?lg!Nny~L>Oe0#A92T3p*zCOA?2-2bEm2JvjLmR$*D-cP7#Ap3CXH+6cEHxj*uJ}m;O7R?Gc9vFAg%!HRE0rS5j-lYtV{9vcEIUX`NS&- zNdaEVlcR12XpI4G2WX7}=|h~QI^j5bs}R3{fry|Ns-b@2!*)Tg(>E5gX~L$vW1|)J z=PT&hWO0p@liK21xF1m2bm#KDpH@9cHo#w_3FMtd_~@7)1Q_kJF*n-4@URoDv!13y3y5 zV^?B`QpdX6?24tW;}ZTsHqmBx>{$Hv|7AvAb&^HI51osy@Os;TQHc6KY4>i3mDTI* z1G4`LqBWLq7HvbPhLQdk?&**1=em`!`yJF*{KOw&puWxYfc>2Qyh^cYAvzs8F(;gT zl95G{bK6}`Yb2q{;Wd!zy=G77OQ_!cu|>>nLV76S^#6e>0$6ExOlgJUN2W(Zfzc&^ zc9xZ62igZ!?(5WsX#CzaXT%xQV?F3J$pqadzXeEutM;-o$SA0LRTvBm7r`y|9U0ZZ(F} z#{hluyCTbFmIQ2K$JIs^76w>~2~&xnmNm4Aofc45tVW2e4njR9DJhzSP3%hNLp7-r z2)&r(CJCF^V{)Q2X#xl{ndG^Nz3E#hftCZ@zywO&+Qh!JNBjPkhe0?&r1h#mW)r*V z2rQ{@ApB3pc<*||D3?pS0MvFU9vzq6K70d7ymw8w-n$-G4mv3ZHCue@EL0a7y*c1Lwaa?I3rvaIb$JSbvz?@8&kO(c_Duiz~*b9xi zts?el4*@*GgbtPjG-|7p@S<=B;1ediMFeitLTkfBPlQNkxUVRRM`b}ODcq>dux3@H zG!SYt$xRX(b!OifO==55S0;Hj>WNAii$?&Q$OIB>HER7;5qe?wIS9*%)K3-2H0rz* zbmN@}KVUGa>Y7_N>AX{)O1J{T4MVa{B@Bi*Zdt;0%YHgZ|1w%T@|VK$~eN zE$>TujYa&C_E@vSN!v&u#!1`QIH)rkBcKHy+Yu6DoV2wXmn*s>bRZ$8fXMR%s}U(C z6K)0_ltgOTqZO3@weJcc@xd!%uOHe#v{I4DQ5>e!F z1f9iW8*>zC82@X%=A(DrLGTj;N&3b9*B0Ur_P-W;uz;9|*s>UMQW}qzW&E%8*%_e3 z$m)Qy2%btw8UJfR!%D)i4q~?j)X9qN^1n7eE?3;9Iv56MG{-jn*M5IBUC22IUTk8@ zk4U=lzxG}cg)lqo(U%MWG zL570d{I6|ogMC{_en5f(DGBz!wq<`wl|~4n$$SK{|26R_Qj|h)93ERig4q9B`NUvR z1Hril($&WP*D6Jq60!rp9!yZOE&po~CF%&u$v6Dxih|@EE6qLjekaSQG^-QRkPzd4jpR5e*!W)y52+_! z1*4lK#N~hOd2W<=4agW4V*Ib|K2=@JLFfWJw)Q06_+N{HC7E>y-E9!n60Wz^<$vvJ zOPqo_1;S+}sVuVpwV-@af{ODLxE+O8N$6r2|7)M$%@u_Km16i!$_D#id*fnl+-3JR&DD(N7WtZ>Ue*mft>ls^e$~$(ICc~Ry{~~#)j<G!= zR6vR*6$YUcle*!TJ_17i*KYL3`ST0}Wix>UyZouSmzv7y%sz}Wq*Us6K<84c%sDT_ z6$J)=YV#R`KebOFjfZ%V+GeIx`QZJfI5uBgK6q)gw|wyK$-<`iL5Mg_Ia)-eTt+!! zAH08@!xZo~z{iH5mwu{t$OmstY`?Fjcwxn`{=?%YMUxsx`?u!{YEmi)nM_hM8(-st zH!UYoj6*824G3=#X+32nimofe2QPhfEg!s->Qqu3IT6IEOs;Bt@XpEaukq!;)-Y~- z@KWcJ4hm-WjLFVv{c9{b=OKVNUe%1B@( z7-t{6TTU0%IjRDz7U4=h``}$us~G$uA^Z)4NvZ_hd&mcGS}yD#4hA@e2~_cJK6vNB zDXE+X^b2O}C5GjLH!~?tH^eS5e)ME<$AP;)xYpI*N9tY~h09=mlyKF{zkdnnIBwk+ zBH?$hzyh#9ELH0z;1tbAuLX)9sd^&7DUQc>l#*6XlN0WF*$*_qu{OdpfYoHY(3c1D ziG-(K_T?+fD*QD-?HN{iqJ%=*=Jpch#{~-;Ze1ha&Q+T*jcpZY4_E-QemUap1&GRV zM8*%PnIbqL#oPZ%G{y#s2X8%%0!xU5Hr^q%nqqmhAF&Tpa=%h;;}kZ<+y3#Vn5W9C z0RA+fyiE*s7Bb;OZ>NtVT->Mf$iR}A0^?C4G%M~EN@J?GeL+LCh%65@%`jY>R4S#} z-uC(LLR<6-z}E~xKB0tkHcxnm^gE9=Yj1!di@r^s$%ghsHjJg8c^OAH=rfFTK7TiiDv)5!H*?gdcYv_YOy0ET?0A zhj9D>NvUlvd5PepH~k|^qf-%vt=}eS@TB1(zn{YDEUFaV5YvPeS| zDL(~c1+!E#YA{a10oGP|D`?gp5Po8kN-IL8^|rJJHiZ8(U=J9tPJ&e(H~8oEvf>}F zvM?!u8pwx7l~i|Zu0Y|qkIKpk>gO*U#ldj~!xF#JFg2oBbh4xsG-X6maVX=$S*9s8 zL3CYV0Fz)0kxBGJl=y902{B|svFD40_(gW?(o;0a2n`8rASljBTR(@7Xi_EgZTvT9 zjST^JSEYz1RYPy!KOInSem&8oT4)yjqkt0oF4hx6)|S|`M1U=)l@(2D=6gD|tk?s2 zB^JTf#E|tRW+Snaw(g}Xh$gl2y@~%WtHk!hv{{qPeBa|g9nj+!M6aFiE&N9T7iv$iv6x93+pWLF;OpGYFI+^l%JVyZvjGbFgAGv7A8ddk84A+T z2$+D!)|o2K7@+Xmx-5Me#1a=DJ8~6CI+_+67K*T%{bj}bKx*D z(Z6z-m;-E?2d;&%k%{oFrQpL0jQz}7Wu=0H^+}Tw#06l#Gp_2GjZ9?U%oWdpaVSrwC<#JECQ(roQ$!de6U}-S7dap_H6%TNu#t%e)TIio8EY32O{INP zzJh|@fp@_kYht`$pS4Hdd}1!J<&0B|7mQ4dU7891PhcEima0cJ2$8;9?J=od?I@4v zImd3ETYQ>m9yB!0Z5}j^HXL{!)Yh3t+@Zpj1g9(>+XJd!o(Eld4D+Uh zX9BCsIM0K=Y@M#~j)1x_tTIOlxqp^!p#aVYe)!_hJm?FMc^*Vmjw3RP)hw;&LAT1G zTT(o?d64h)EZl!V>{FDS?KpDF^PoR(V3s9s19)se=|>En2R(coFM<#vr86G-PdrM5 zX7N1e@4wOSWirs3hVjBYC}Iwd&bI*A(GX-9CB*Zf`TNmx-vKz937+$yFYXu9IhhZ_ zGA3y$W*)SBCUk7u0e)`?);#D#6w#UoCH!1Y6rzIN1oJOq7p45D`995qmM_PF0Gj6n zqE7SS(Kye8?%ea&cs#HQ7H;N2wiCg69#kJ#6UKQSwD`}W(0?GjA7RuilIq!-2l@OO zCu|b{e$0eY#PXa6P1;&r&4bo~^eqvo6few!sF*LC2ff(gsEg%P?payZ}(LcjzCmKeEO((wcpnmf4pqJuWN|4-!~G!)qWoZiul7F z(9N;m*9TEwvfr1{+?k@sAerbk22Ju9gEsmlh<=pKr4X_Tk8LnjAdf-y{3?nOg#Q5S zAmcm+#k4G~@T-7sGOUVA3Av9!luc*TmN+d)V^A#e#A6UqIgZFEpRcGMgZflJccFN0 zV^Fh1_`+_6*sbx{#!_y144O+Pp=DnH!we{=5`)K};UjSbjLIYDfU($<#beOWR@l6h z-vHfX7%z-Lmosq^{bztz4MBcF3Go>8&tI7IJ_G2N2cwr2pV!mt)da=V%7t3iCgSFT|NJ=dmQ(lch%POK% z5oV1+(=nx%UxM@%CApsDn=z@MSb2DsbXAJ;+v?{Ed|?P5zeRwZv(F7f8r; zDy`Av$9``UeCxFmIL!KQY7 zRG4EE3gT?_Sw++Vc6}~XM&F?zw_Ts^i=$L}H%O|4B^_(mr)SwhDnUDvJwH zLqs6BI38OdLA>iT{o|q{8NoFTr0bA(ea^fGWp4|B9hsnHTf08ZZ{ec?nL|Ms?@8ia zpOJC3#V5d)GMa%SH;@w*dDsK^4xbC9Q#1)g&BSQTq6Ipl9-n8WeCj+3G8i zX_>+YfBWOH6{I9gU%A&eSrkENS%Y}?mCQ!ys?~tiWxNPQH+?1VF{s1Z0_)0nv?YF= zz4y~9Vi=&w4F9{YJgFV5+HNU0GL~YSzVc>pkZQVJgs601-dFSvq3$ccqNd`@t}iEo zqc#V$NX3emYDL6QxSyAnyxTtgdQFvPL<%G)T50aF&s<(crCFJfWJ1iY56N**u-Wzb z;d?2nfYII(;@b6DvMoZ?0`e{kF}psG=Ld@^2%U+?mP6vruFv`|v0@oQHycE?gsVAR zyFPh4#=xsIs|&it{J%XPz9h>yxK%wg^GJMBq_5kPRstyz8@|8??4b z0IM)TRS@?%Y4y24SI4`C28z$<35Sm8c&*c{j@MvR2{8g!ssBenS2ALSF_=y+$-(ebEO zL=1%obI0rYW}r%QCQ0sLrMbs`J2^w8xtfr^gqV&;avT(FI$q!H<-|K+oUw$sI$pp2 z6~u5L_gIMOctN??cSEj%@Yu$Xc+>H!ej6c*BQ)6{`Z^j7(w@EV=r*hpa=~iOOqEaW zeU1LAFQ`ntfWPa>F}?5B+H^4m(8mlCKk!rY;x-@pJ3h0M$Hv9@ROd@r5Clw#II9Kd- z#WUC(kXIn+1|FN&X=;|$U5X0Y$y)n!ec?ce^o5K7JR0ZK)7h-b3XcX9%dn~#O2~cn zL|JuSXo=+jt)9Ao%&RA&avYJdaCd@UJ$+RjosHtTt)9+1aDr+EVt-7@<)z&6>goQw zS?FeK0Bkj&bP|JCPdiQ)7KN!i@+26SJXySYD)=>29rACW-ql=*ys&yY(iW@0LI9%; zK^CKgc=gn828P&lfLToN)ZmnSP+I4t9SGf+q@|eE)8{q}8Y2OYGX!h()EL>bR!^_b zFC;V#(DL0uzjXpUEKzD-@?u6Y1722di9hSZQ{hE z<@4$(Ya>oD$0EEkVbm;=>e*U7IpU**Ef-*GCR8Anrv_*GjsUfK>JQQgB2p<{SUpiO zU$%OBvBOap%lNQW9QsljU;imJF?@g1t)xh(>>JW=IrK88A^!rU@!=T=j8j;eul=Xt zwb1wv0LU6HP*P4s+!l(R;cK7r42vxp4zz?}xVndu(ahJrydt`6D!@!bkh3Te-4EOO z+B?mN6Y~%vWm_=1GD{_-SJ?f1?dLa#YSw5FrZ7o~^7UI*RwRt~wNJVhEU2h*83?P1 z^cf}Pt6FZ7ul?1W>bm9j0Xt;j-D901;X_~h7hl)aEq@)@ZN}AJeZqWSdrn3PTpB^` z`Qy=brCa?oUwhsmP?;7(c$~rDk_U-L^lDL6BrNx}KkJbwY5;4*cq%pML0VGj&Gm}D zfvP9#jmpiw`InLay-~Tvx6sB!vr+kC^R7n=rQTSom&6NdVWgm1u$IK_zQ;;!8`9lP zX=hRknkDh%71V&I5&o;eWL+YdC2@5>jMRSv^M+g%SH40xFNp*G!Tx0-fYD6wRLe~% zSx+>hqRQ%E)MJ)PhL^;TzfISyP9XGRl1hu0#7B!qi7~)tGTxd5TWUF~8R3|L?B*0a z?!{lbshw|O+;UZE>`@PR_W%N9_*GX_^>W$s0?)f}S?iOf( zo&(I2=_2TEuCh<(VcO?&iar#tBnagVNlVcroQD~|DMXWUKxoP&JsO(xFmp;fME&~c z#63aiN2EcN6>}bDCo*oGhe?_qr#Nyhh@Uc<&%+$7UQ^@W0Ncs9IS=yyH6k4p&gWqk z<09!WD#%TQ{zWmUI;`_B<)Ryi(S!#gZazFZdwd?IUp1FUp?F{w80YgaB{~G^9MuQb zgm5LF&%>NNiE{{D5Z=#Vk}APE4>M^72CNAHXEK2*-t9b0%a5V#UjcL@GbRwjIuCPh zPM~gxBVe5KWbxNmpNF9;g7ySWAK&^oR^m9toQ)PBEqMXP#r3%ImLX$feG6N1XHh3kfdLHin%lX;8V=)OVtn!sAaDr^rnd* z%qiyDQ3aG3>5Gg8;ITEOq|7PiLY=(CcT~t|K(Pd?oM?Dn=&9XR#9D$g0oCQ$<`nbT z^7v-9MQ~RWQ??@M<`i>IE^^H+qfRmFhN1U3J~;~K1aXRaTpg^IRJdLp=7wDUuD;j+ zky}aZ3aWmMGz{rpKfgGP(6a`Sls(tFuyM%w<}mzr2iOzFH(7X-kSEFLrol+BFdijY zert(uX`i2yCdvb<#_)^l#@_bC^h8xN~H>eCu0N+$!H!vs}0tCsZYPE`}`ouC}?NyBO?wZgi(mRjLamoRZ0BF>VS zYm|ghE9{4(>Dvf>Y!G>iV53&Z+ZlsUUPvm0N5z!)EPP~0**kdg$^%Pd{IMl|oISlw zm}mf~4Z|;XkA?Q5`7(sSOE3r==}T{eQ7imbRtn0_G(uFmMy;Sm7_J4aR#0J%uThZr zIlc0r4mJnG;B_iTwIX6DT(DYU_V1XkA*mZAIlxMDkNxDHf+|hldO$)6F=_>p(KrF2cb`B&rr-^jXKMh$tZSScp+8th!THv_)tqJhqY~-l!F1-Zb$JLfe6S4>6q=z2VohLR2EsI(02w57eeYT;6HkDj7A|lrJnc&&<%#mQZ`tlkiS@d?8latn2t@|8Rcc81|Z9+9VJFdWIgI3fui1CU#frMvF*;)+^GStgI3Sayfq8nwx zajy`^xS|-@`y9F;n6D&b4JkhY`WZ7SlVCOFpi|6#OC6#nMcfL)UM8jE zmvxFcs<_moUl4qQNr}jlb&7eI6JI(Tf&%gA%#dK0GQqvnR8Hrik$H5ftAWm?R+)3= z4J- zo!#)RUR_M-jUCl%GyO&P9Aqp0#=0!^wZ4z2=4*YbF1)nuLBIh#wkc=9Ghgc>KSzj* z2>pYQnLwodE5Is4M=H^*lJelpUV{vgH>e^FOtIh%iXg?LADD37)fbaqZ-{s9m5L}V zE+(}=1oKHJ(dv^=mEfS7jZbI;`s_@ii9S zBxLO~>``0>c8l?imiU(T=Mxi!4cQ6BqZ0m?D!8}(z}o?;%HqJ0yD7H$q=#nLRkf8x zh)S0S4*f~jh3|m0or?=VF-4#BQH?MvP>|avz3bkpDv2p1XdflPpY+)w^;H^Q5_E_l z{-isyYKrX${vMC*I6?eL?*|)QXAu0GfpoR;C;jzj&o4t}CKuzHMNq+IYPed%if!qy&OcP6O9S+%5xS5-~I(H8lM{{g!%v{Nv) zvGqyscn;fO5V4BHT%jb)C;c(5weCXbj|P!<2{xbfWi2Cxyaenf;}0!-WQbQMbPpR+ z3dEx%%V(DOarW`OQ$;aAi46bylYafWzwp9KP!AmGptr$%((guA5R{!xgs61QCtVFg zTnqY>uEHGSkkRVYgg6N? zpLCMrpkVV!uivYzC=7--QgI1!ebNg&OcTX`L=mEdm`{4#{BV(k&=fqj(j?w|(*O6g zn8-zFYlEnkaJ{XrPx`KX&>HpzVF;6ybpE7|Ss5s(I5UAS^yHXNdauYh@in0B3@1@G z_>*4e?-X=sfESsdD#)sB-RG#f16>_&B5K9i{r3d@aic+uKW-@`Jmb-O(fX5K>{YzH zM0fk7_nYOw$X*Arn&7czB8F7a_>(^RIKG>*E71PTs7iwQlm6$1c+rp|P66R#CI#V_ z^+_*#GC`BRM(}nfRYabwPkQb-Y(5_cc##PtSOo2S;4PwNc?HOH=tc4%PB0#lk!yDY-kL>5D*|sgAh-o$&{5S*F`4!N?VGX z%MB|jj%*5ID<-qzX8UWQ8t(^eFylsXLv=456wZp9en-oSIaH7(2wg!js5&ggO@})5 z!~(+i06Rdq${s6jR?OC}%;XhdHyCHd&D6E79Q&d*0`O@0thiaX497i+BRt+xo9)nOv?sI65n z-p=&^ci?&LShpg&tE%3ivc$6vt>RH{wu3sNb2@DlkvWWi5LI_sTvpJtw5GiT^-q!d zV>mv~FdsUl1zzx)wcO2zo=w0DQySs%cx-1W^TvGW_SMK$O$66Bko<*kV?K0T2^>H} zfRt~7(8rTxjGDgv1=bSp8BBL&;h@YqKg{tF%Gu}KmdHAFI&${891_{tXSb@!;xj}( zgh$?_w0^-a%h@N%aS-@&XP+Az#Hi@`kbo6k(x3m~?6d2U08xY4l(bHfoqe7jP(V-~ zS`+Cnst$JcnLRL9Wpp4z?Cdk_%X}hV|Ju0axJvp_an;#v%DqTkSFf~coVZqKoNcw; z8>eUiY`uO5`hGmN^5@(dr)v(>ITsN8yMbgS!cF5GT~tn_BS1=T$jysK#n&X$IQO>V zOmj7e45*9)k(dhEf`F+yFA)ph1OeM4gp{heutq4aV0yDnATqU@FOX>ew8{7(% zc?Q_;2mQ}*+hWHjpHB}(={Y1J81PF#kH8>G87=~mD%Y0Ye%AuPy;ATcBx5?+r* zM@wr|HJEL?f^tu2S^in_pB8oVWN$)vLutC=9J_<>DDH$Mi>M>f?>6xkhR zu@7)TEf2M#>-DX+$nGZX{kg*VQ@++RWRd$P>84owl3yKGuPk-bEQ zempcn@sEIioC@kq^w3(-xkdI;nLD>g$M=7bf0wZSq>o<|2<<*OO=K^Vx$C~3Agtha z2!e)`lW@OdS&_XQj%63{szEZq%KilR-Z7}cDj8b&m13$2EkNi*B&veU@}dLjAK~6U zubIeRBSV*Y>nr9^utsa9aBr>@AhI{&wGF$7__-naWFD|Zgj0n@=!CkpMfO%{xBIb@ zur?!qx5gyVg?r__@*;buv=u(Z$r3)6H(rgF?bDwb<@hLyYxB}jF1O|$G1dqNfc2hh(VS!1Zy$qQe^Oz?qa zm--E+7I_!I0|MV7urY%x5XLj4j*wv;;s1ujn0$u_{EgJ?W>yR>N76&6lG$L{7kA1_YZagcKzfL$ zq<=*8&%&tXX5{QP(({})Ua{=Gm&sx26wr%?v4|K-#UGYkraHDn_W|bWq#L?}x{+I} zSbtge{7%KO_JL8>u&5x#@^WJyT6V7yMMZrO+8Pp-#Ux*dL1u~odksFE@CsOCpTgug zgR11<0Q=n7SRvv(9T_;9MUv(>+&9dOt5rqNlg+fs5X~C zxUET&yNgNLEduOKu~CW?+!^yflDhKTV~W%T9}d}+p-5#xsAfpiYwG)_+1Ua1>9d$P zTLA222s{?(y82l21MFw;9=99_#v6vkYWRaX*-HcLk_Afh2$b)Gu$V|ZhWRV1*^R7( z&AoA{IM#9Y+5o#|dPye9gJ2%jDuu!~1=vf*)f9Z&TmyE?XydT7**gO4Ji{Y!wo^@jwdZ8@+tMw`N~2iW~Wn<)8xzz#X` z^Z}pjdja6so@Y$A2rQxXjL8Su-gaP`P$`Y zl57ZOQ)2VraMNc%v%3e{$NCmj+Ij%%W3(y!)j+#=({v?20oY_CADYk%tM|A-yU|`a zqAfxGIxVXzuLu(Z?ZfSH%k2U9jUlY%yRFCdrr4JS+Q}Vh{RjA)Cg3}hRWn8Q=0N+} zQ7rAUJ1#wv>OXa>GU4+;J68c5`4t10VhBn*eMB_-bf8`1H`r0@0?g0^!5d z@trCv;yIA+_!FHesQ?9{9rK>6?_ZW;hx9=F3nbMSG;0X{bW-*jmo3it+LH=G9X}E` zN*M*Rcf0JWZ80xD2SyjeqDfeNLNxoN%kKITF4|y#BQ-&U_TQHwvah-9n{e!r?;?Mh z##EQ71&mq$x@_0)*oA%u#@GHVRSvU4+;*;4;7oY|jC=kpUZA*js@TeI`{rF3ohUp| zQ6#7Nl2?x$)J(ViTT{4KRR*nvKhuPP-zrbB=_-a&ASdKBTNY@Wx)nWevxLp%OXvEjm#h5pxYgum?D~==ooW`W;IU;1A za@#T0a7Wz*D0=FKs!r?id~uO&+4j+CcpWM~K(8i<(74KVFyBa5>%8ctRODB2upcqU zW>>WBHCI!F)eQOVHO7;X?|!9kzh_B;5M6AZZ+YpE5 zK)IfWd~JHjoo(CqrX>jLWq`ws5*{!74USM|Ewb$?C-C9h>0m7OXYuaHJX!Duh}bgX zp=XI+mnM2d+SZ~CRQzKsusZ%R85NxP$58nn@sEL+_{ZBvgoy~~h)1%%>4ozX`Xw*1 z^slw@iUyUVS+9b8T;VQJ`KBQ z?mqeS+=PgNA}ck#IC4sK z7QrH`n*C`|RnZw90W~yL=|09D;m!K+D{45;3-EML7d!2&_PJ~I!yU70S|yRyw+iVO z2`%uRUDoS%@{HPIG%_Y3Srw_!YWPRQ(e6L1V^PxRAuQou&srjDkp1F3d~gtde{T4- znSZuGWs%h_xgq{Y`Xr>d$Qog=Rv)s>4>+2E11>2WK+X@!|pI3kr}!AVE>0? zwIrS6Ww*XvMb>ybXiYpTDha3z!BkHTPqeRmRfDA)0cvKX<7ju7)w5!Kv^qt5L0^~8x@M2~H=Y>}iG^{ehICv)ScJ&x zT$%3;X8vwB8;rRhj0B`)V)DJgQ^^yzyN4!|jN7nJW?XDp7-zVDMDthSs`iIX*rwW=)9P?GUzNH%pFR&5@ zC#^Kol@rte{J{5~Xou*@f|`s6IQ;X}1Uc=`LU?$`QOarjk@z=RCZ#$rnP|J+)r5Q% z`GZL7JJd3Llrm{tDSZDI*|Rky-z8jI>u;UTBj!^C>Zrn~XjP)I`BkrxCa4ZxBAr2v z9GlXqi4&XpxSdM+; zryqB{h&(~gC`4M2lXP3J+4Mv7x-+PRNa;ToHq@~tgU*7OL{YHEmdbS}!eWcxSI9e# zWaT-JPK?Aq>T>K$rkCcI0!dpqyAYo}Ck$}To+}r`s5dVS97)GvTFlw=`^^wim`Ec)c*~Id&z`?? zg^Mb&=$a)UtRhk}HAt1@WXIo3rPdbz&8M3)nq_>$B$z1)$7B z`R43-R(6nBg0{^Xl7>}Wdx9yA2%(;R2dtMQRvc>(Bk z!$>8DbN0NYS}AoQYz&EfNa~t1i)Zn8rbZ}=UV@7q0g#~*t4c(YA? zrWGEDCjlf@Z+4F_8|EAK#$u7oXarU>V!lc`&F;~4Iu|SI0jQ5rg!#*B}_+GG^z-DBxg&75xMj!j_3X|T>4oqG4U>vVnA^EZ%3MvvY- z7LUYdj*y#w5avH5)tV&V?(xuyg51^$K&m>ZZ}(`eOl0&KAkR6dZ})hrVkB$s2V{VQ z`gV`*WsMk}0%WE^pSF8!9umQfEnsaoOuc*TwIG?%lR$oQP~Yw`ZDT6ml6QdIH>ln{ z4x+2-+dVdYl$RNW2P5Wqf{9d1>A3qpnAA{YcFi{xIz-RH z6~aX31Us-cRwB6rS$j$Kr$3`H6C{Jv>>0gb#Q72VcQq#Im)ruD+)n@OA%0UgZ7SOjD8*Akh1N*8$xuqT3Y=4=ZTnSINDH-T~zkfW;e zkzJ_@mS))uInN_;b>so+U`6=U9(4{oSNRISA%<`;N2RygGm@ZUD!_S$z_rZJjoNLG z>5O`R3~Z;tsRY#&MLJ|(I}4Z26CnI*NK{Hr(ogn_<;&yz3xtBhbX{Z7&rj_mrTC(K z`8_N*rNK`(GTfy(s(8h|KNb6jmLR<3PvU;h!F*u5K1s&+pTU}Lm`)X#8OqJZn~eXq z#Ve{SLEHTls-v0TzRmPVyI5ocag^FD&p_ac5#r0CSRput&y4f20*2#s5lLM%z9zg= z$&5ivoLQw5D-ne9hD23`@9^;|A~R0fxm#xn8Vj;12rY?pk}j;9rk%`E((X9Lw-x9E z>@^3c2ja{!c+s*&Bo9fvq<=fkf|1lT=MEl4RV0W5`Z zR;KV4((aE_1}Z}~1lH8Z^XD8h+e-WJkrajZ0M^IBY01d!EbW%t$}4;Vu*nYoOzTjQ zxm4O&Zv}{UbaSo%_9@|1zWj;0m%p!;*}Kg8X}~=M@;O1~N@;(2zqArNi`=V5>}S3z ztb3QVUmRXswJl&I_U1@RH#dh1Q0n$#t9Ly?F;l>*;m>59p{x^58;9PlhVc(Zd&5#g zg-#kWb6NJb4n>rzK_HAVBq~%LHDu<&@${VxMfw1Qm4-xD-baeH?43<3szI?Egu{k( zg}dg(2$#t0l~QODbq#@hjFFjY*=2((E8YJf*E7nm)m#hy=3ZuX%MPuK&214diffkI zduC=>_T$a)Evtt71{#xe#|d|Ihw37;zhzfG9xiAKl^sFoP9$|p&^J{x2U_;J(oupY zR5=RRcn3eb9+Uf9mOU9Z1ZuBb2<$_``NDJS_~V+HlP%ljs-i~EUJ#CIB~p>-ZC zEPHXKMAc8%K)Cw^iN2eex!$rDbQNL*&477E!~TPmvp6Pgw(O>MMKOj*6+oy-Bx;i` z?nptAxx=#O6so3*dme-ink3xxG1$yKmOX8Du)+reqiBD=j-ijiW*)ZezOUs~_#9wM zHO@0LJfYXbiOiGu;_m8rMcNL+SB9jn{a-tACDi?W749ZqrzlcG5T5ZTsY&9lWj{;{ zQf0pk!ayQ9Wh>G@mi;v>UsNwS4TSgoHMx1Z_}8-EhN~=-Hi59~2@-wyHPaPf|FEr^ zD1{fO^~BlVW>CK%?SIoALmQE*yp<{BcgQ}Y zA^C`KZA2<~sImyWKr7>4M1XFVS0rsj+8mBk;c$&FCqIi4aqBnn_9h@D1PT!O$K4MKPi8Fy9y%L9J}a5T+y|_ zwrE_{j8}nNgQK4>j9r`X!~;$Fd;>ZUwl3;ewsmozldX$?IJT~@kD=0J>#Bw|ru5gK zJGQQ_eM|7eLQz2FkgQx(MY5~>k8y>{iA>6igLZl)lGZEpW&r&mdUg7vfk+!)r?{x% zUK<>PDI~FG5qPM|#4d6CUsQ4aFFAq#`-lH4{Zk_Tirdzryl@{43Bl{Jl{_^u!#0#r zU1_R~r@gDYvJB}Gi?r zcc$VMsN@mh!rfTrpMG43*h-5qy|`P-{IkF;8yy&*D%@?P{X$Dv_3N5r{LmKE-=LZ~ItV^VG5ZZGYK&7qON?HJ!zA}e_?#ZB%AC@`D`<}J z(z&I3cA)Ft?xG?C+0PS1?U8TLKYX~UNLvD3+lm$ueL)!EPf}l)T>_^l_k=*#plij& z3=rmMlJL@lko#R}KOd1Q)*+joJyaq2X+)0u0{^U|f1aT~Gw9cwC|y0PxKG5;ohv;g zmCUN6zIA$5fx|y(N0Eu z3ROoDehzf4ZjGC6Aiy^afqKhFuw1S?A7Q?j4REC)kOtK%r6$Vds`nLMz1R-M5yRq6 z=e9TI#=x0)+a4TsUIgQ&VdWrflXkr(Ah_^=M-O?bTj|oP*3LA=@}r>pWDMQv(nDNd zgQeYPC-zAVk<;C1;3t`d|ii?`eXXJvr2myM@bjeLng;H=YgX zVM28SDHZH==dLP4`q3=I+X?sI()Qeu%#<}jr$-7FQBJAuYBFTF!+t33FJ4JulI#QK zYs980#P`K{fz95Myh__-U^9$1-W#|BEqm71+Dd*cu=Pe>eKbDEvN!BaRQO?F#~hrt z4es0)?%qI!-vIWfgVXbuJFjI=xR@^ZYK6RwsT)bvkDuS%5hx$8(bEhlQ-GCsaC-i8 z7qIMOkJMc$n*wX;;Pm|F_FDG-Mx~YhKEPgcaM}#Ji(2-+^rk%grMm(gjnCCm zqJR9<4 zsxD_u>_r#mRit4c(55p7DW3MC?sOS4%h6lcvLC&i!c6%Q=-Wty7oi+_GcEg@&9xQj zI0)zbNouTQS$2yr6BX$n5CSLrY2tmRyNzY7eRQ*kZ6$= zZYGU^8+Wivk=_7dx*<_MW95zzi zI7ig|%HdeyhWGrs^%Z&Dwl0JE@W1Gx&%_kPjqG*}I$2&=`QE|SAITY&l4~B?a*l-! zm)o_ja2{b*0i)^vvV1qyu?*~;vfQq|3HeoZdw@F7pH9sX?h6~R=UVJ`ZJQaS=H6*w zedy1m?%+E3w8t#Nymip+I(o9WlGzLXX(L0m+Y4|kgF>e?(o z4e5cPzG3LJf+!P(`>|z@Xf71#eGpa{60J-=Qb2%RuXik0QSJiakRhpdaB*90*FP_& zC{33@xMxUg1LHE;*x|lnyMk`PNijDpN%@fY%B$x+_g}Va+EKiTlZgEC8dKN7Tdg)m zW_*yV142?X0`{CA?%Vz`scDcaW^ozO2ZZ7NB(?oJ`cF;a4heEi$Vw8^L0IfhT0=u6 z?p6~l7Z*ZWpTI*R5%|%^eLRL|<>0`SRP}f>>_7?iLqAdmFWTBr+&GGmA}8L9;HS;N z4vot3^Cy0!io)3Z^AqVC^y9Bza68hE`&-J1nt$WL*V73bqKqZ$f<$D;a;5Q*gD=w4 zM4`S&FC~4Ez9;_vS>pF}yX+=>w5z^oJI&7PM!lUc($}TQ#rYz=G+(60x%rp1@LBqU zuF4`V_i0$+%LG%Ik?X!{pq{BdhNyX_`hy;6BZmdzwR3vjI+PRRbMLe<^`k`4xO7j@ zyI2SUdv?ViO@n3f3>+nl0yHiwntqF)0|1SC#$G!Hn;#Ey@*r7de}Ve*BsH$J-D@%c%{zo8+u7rP``c#7-wL#f#R0$idAvN|hZe5$M65Nuhu0Q=J5bqK#M z={tJkW`tC(T30kejywax6+>br1(@@0NR0sn6i3d4|2Gm>QnDfrcQS6iYvM4x6kHTo zvcXBMC@={DYR4^h?LQu^_C@u9HzrOquIH#+9x-mYt4?jWA9qA{cP%dRCzdTF#;tO- zwO~ISitNz_D$wg{9x-l>%ll_}L47Ib0$NBgD=l!fYM>bRsq39i+OXzHlQJmfs{3<*-Cl_0D)B!5qdT+t21n)ztLVGxcHsWMeX zdrHL42E;Lc3w6W;c6ejv$on9Q_cYmfN?dp=ufn5&#Ts0DO0-5VNDt*3Pl>eHGNJ}G zqzQ7LAqnaZ$5SHwT7FTN@RxzTN;p@~cuKVD6rzUFBw*7GZagLW?v|=XYk+-9IO{i_ z5?yRO2_8WHF^x&;1jkciPY>LUR{`EJ1nT%4o)VrTaGVN8)AA#!tC>Lz$5SG!XEk*p z%7IbcpT#Gd7K&7!64XWbdJg_hsqiekx)5G*IkI%XrsVhFL_QE1!%0gkx?^;}rmtVE zBc>tueL~vO9o&&%qhEMPciOOgu^e9UDlTlw@v9h!h{bdfh>pcXQXGqkD#fvwxK272 zQ@c8_aLW4-5YT{k_)vxpuh?=T7Q#qm)2Lxc(k~N>X)mNqET$3P)D$6fWgCFfL<^W$ zOuf9JtVVVN)SKXNDoV#<%Kdr)F@_pB7SKdT+82xI+?pz)Kfy}?tuoR&7L(N}7SI34 z{z^;Af}~x?Vp{nDYHfOsV=<`ajX|<{Q}-(* zD=hBA8lntxt7=42^-L_L^rI>8iUHQv-~%1JW!RAOc!GQd*bsvccho;`mwm;z|N zfu9Vv)88(dR)V{13pnyElGd@9T;JE=-ujLZF4shSQL&g*<9|nIMXrjhs$wyP&VwnE zvU0>?T9ep_i-?&EMek4%CKgkeExC+J1ied;iN#c4SY^=|*)5Q)c?6kQOydiNitfnn zuOZcKCKl5dC2;IL0pN5)VBJnErVnbT3aZH(5H|ahOf06?$|i_Iz`i&5BC3Un#nk;B zd}{Ruz`qTF8|UT#Kk(Mj@ar_6o!xmIahW;a5)fbB? z=x4m9|1StZXc&vDDwC6(Tn##z!-uwikPfZ`zLPbR)Y1+1~Q*e)}#Fs zL${}}>?R+V;o2Sta@s+C{)sEbRAclHAb&fk&p$EG@O0LkcM&#$NWNP6 z{1X=)l#G@JQo*24^G__Dw=^@J1FOAZYX8Jvip4TI0LU;0_4y~RoYjms!ZU%)HK_Ja ze4DPS&p$C}W(G61gY^|Ld4N&PjDI4HHtnC7|HlBP-2mZlqEnYkrfdJiPfn+?#=MKM z^+C#^QTr#>IU`w1dBEvJakOav#E$DLu$DG}+i5MvKe5Sjn-#ncWQbN^-fo%uM?sdI z4rI0?>wCN9aN}5({TRq5BOB*?yM^x!pMT=RpF@~&8m#k1kM>XGd&B3SIJS8NGXgO> zb0Mi~&J)TLx4eg@l#KR|lksWH>{O9<`ULiJIG*1Qsl@oEIS;~{RDs`i8|1emjh~^T z%|ZB#3D|V@NA@rc$z6o&gYd6L1&aN2W#oGxEb=GmhM7O%AmV(~fOP5 z`5&)alv&k8^fPQKT9s&Cw`klfMo=BzCektL4ijQ!>t`8U(d7m*;j%8Qi*mwcQFHl4 z45~dY&V9V6e0zY7?P2M>x)z8 zY&{W!3@OWlknT^?7pKBoSY5L;rY{cFZ-;Eqqp?^RMDen|2H}Z#Sp_q4#>;wpr&rKr z`vAlxr0q8Ki;kC7sc5432pOLN*+pmo)su%Ub&K$HL}vAdr!rN%thZPB$IDu@9Hmp0 z#DB!g$~_cTQ;-TFSw%^=iI>&%-8>={*)=pIOAv12WxZCmA^hI}wl{yEpzVtxqLCz|QR%Nke&D|`#|$=ATXC7dek z#LFu17{R|TBmWPLNuoRPvPwLgB3e-4xmUvejieN)cv^^;wP8@Epvx?ifu$18y=CHM zEm?)bWtjmW%LuD5R*08XdH`Gvx*@-x#?+88@v_Ey@am160Cc)xJj)Hx@v>eT1b?fQ z0Jr!NB#j&=Ue-GU@O*UyjI)NtLo0_a#LF5sGacdO!FcS)a^hueuYrwbByO2lBz5VP z1|2V}Q#dx6l>pY(1a+J1cv-a?;}ym>z}jitiI)|413mRB@&_5r+_ugH<~fSd8u{p7 z*mYH3(-{LaBJuIOt9;uiBnRDEN&SQRhp*Nwhd zSac_IP%IkJ{DX!*|DYktKWIQFQfCP#UY0$)sJdIuQ#s}zCtlVyl;Fh6D!Zn(q6d8l zmkuQBMXI%lm-YT8IIGCw0FnsoLZFG4mC_SSeQo47(U|N>1QRcd)+6a5)Wpl`|1Z32 zvXL{+$j~pJ;M3z}jU8ME-+uyYyT8mUD9k6Q<7LqTAU(u0@v{DU9kslSoIi|ow$nz$ z%ewb5>HTYsqg>iqsK= z-iAcI=ETb?QU)89(Ez6z0*?i{u1>tHQ%CT&-3l;1H7r(R;$;6Ja5QNz&>|y9WN{DrGg4S3G63>>v&nke#R-sZDco)9Go08wxJ&^W5_?NWmV-( zysSy(VQjtw@UbCG;JZ!7%jzc4E0LezAs9)OZsKL-8j9z)a>%c)F`srFFY9tLv_Avz zMMF^9b-b*3fjBXK4d7r+F!8e1{!mffSJMG4*05T^_yU-CSq)d!Q_Ro7I^@sf;!M1( z*7<~DUIOc$Ka=MW9WU#`&*c>{&!_P1LQ+?s?*(T{(($s=-^x(z8elf^V=D&@ikG!9 zIkzHp2H{md5;sbj2`OIIh0ECAy$Qw)!=g#piI){M01MPAfFEgs6EEwV+?e|JBL9TO zRF|m*OvlT*e3CW5nimw-2;@)E7*9sNJ4(mP8bH@;DHtF5v7C5W!w|rE z53q0iaT71=aa5A9eg$~fkKijt$ID6?5|8-DSSMqVR5#EKpW#oX}KNjbo$gQdoIS^m>5OFn{rHbs&;%S?al>S>uk$uQc zK311$?LqBI^n;{`Et%Q7qgLS|CW*ccnf+?I1e>5t0z)wUZ4e0>| zSL}ZfZ-Eqqp)aD4dpbAGFe_V_^LwKkgGHStzA|wBxG9u-L zY&`$sOyUa&=~xMiN4dldT<$T*lXK|e`^DN?fE;}y@#UU;;v@XMN>i1c=Um;GoF@{! zmOri{JXA!+d;6k=XFqD`>~!7SYa|2XgF9E2g29J0Sa?qMV=&n498{X@QgR3DK( z;h}OO?@K(B)3fwfTXQ^1??N-ov-HB?X5t#SQDDdtq^=}>InUBmc@Ke4f0iC|w-AfJ z;3qEW&;NLq4*E1&yk*!_v?|d&ODFFxC#VifiIjpGInUA;=9J)y?l2HPOH&<1M#&HH z#!-`|w&*plW4h=yrE7hWQg2KRQ48l)2fuJ)Y0&61r&XH98irqQF%s=j_5ZoL-f$Vlr-&s}WF4#kmdm8DyL7#S3Rj^AO zGXg)y`3I7(hQ3H_CtuBAv>1>?2lbs*4SKBJiKqpnfkE|I)f~F2zO$;KJ0qCU3#@En z^6;XXnX@VyYx=C}z$rvZ25k<~QleA0Nv7+wsxn`Nvc}H=?JS+w*&2mQ4pz&yn?=RUNu3S+*UJ z&PG$TyzN+IF~T`k!Lgiih>y*9Mo4d=60HKx~w8>m&}w6<`8 zp=r9`jvUkUm(XaY=^{g7#WV{Vk3aUe6A{9c7oN0Q~))HLN?grv%n(8Pr2HtKgdU={wN2IV{ z6GX3>_hUrLk|~vVvv3Updb4mpZ&UFV0-1hHO1e>t{Wc4|k)t;Y+X4dkGb<-GRV{=R zEz|^7qQIQ@=cb8;!B`GXed#O*q~9zDd*7`tqIbGQaU^Rf6*`()rk4ZK=pjrm2kqZW z6PftCwc$@-zHd30j6d{p5dEjkw85Z`)O52P%rMM1?DF*rGGhr?tB5&=beiQL9$8f0 zZlnVQQ$002(Qf!cGnT#p=rX}9t(SwdXt`bv7JpfZ8Ns^{1Q$su(yj}?80K`l#LzTm zR0OLUF-*T{~sEaj!4dhz~^(_a%-!x(LDv+B7ecE!6dvP=~!gu2k6v@{*-*WJGmc?i} zAe9}|w;cTRT_xU;WdUhzP`w;1psVUz4x*CkGGj1UV~NSbi)v<;0~%|3IcUE)k!dSI zU9S~Ort9Tkb?ZE=@oT^*{Wa?4VBO+UtmSvW|M+Xs%fYvIYq6FvT<<6(H3}5^qy6Wq z09H^2NJT=`mG>_FN}Os)?v2T$DY9-2%tDm*GCGMdBZKV?J_7&+jQoVLetNj~%i zj3~K~kr&BYM)#3+NnRdRQN$y+3?Ux^k)IIEC7o^oA9~Lf6rVOADTZB=hwa54zhHl? z6L3i$2$?tvdRS<9E|OxGWD;VRWRerwCAsR1NGuzZAuxjqdX63!wM%m4j(CJxiR|?n zlJv{CBtM5gj7##Hv#N{skd~*Adrk`&m*meMmt!^ZZ$OU-?n*^zm*hJi9WhU6Nm}gR{dH$Zn@4WiQgMU6QLEM6FHFu}d;{f+rmn zN%=Voz61C??BIQP-r;=hlDu+TuviI|>qzbA)cx8e`Sq)q|B!oFBa*6TT#};$a3OvM zcEjKY9K2=N$A$2WWud$fB-SjCIO?Cb53MXGiUKNU;3s`f``h01s@!Fb!I7s)TDv6I z48zB%kKW9iq2CJ#w9tp zb0se0D}pW&WL%Pm*RCbbBKsneb)6vNl6-7XdGQaj15p@5s@sf9a+7h0j};BDxFN7^ z$0hmq2QcD5XH5_?{Yl0pdFqrX(E(U5ga1LbFfPe*W}2`@0essKxN%M|DfdwBCgB-@ z3a9+IvNXFSFFSz6*XNSF2GV#{_~@`@3t`SK3M|>+xd>P4C)lfcHxjh~H8=2+!(*XsbqW-DA=C>TnV+P! zOY(yrB?VPyEFoO3c1h+L#`J=6N#;Dyax^4mZx{MR?$+GVxSdKY>&JYeH*$w+gs+6}ZuPk&=l(Pf@!vq0XGpBwxFjFgm0M7A zHUQt@&(SW)#ec3RjsZGr;Hp#&F=U@Lq&U>caY?>;97hRv06sPZQtWd{ z&e>{eXXJZPcp@Q}N}1p{NN==0AkzKP2l-s!Cz9#unNbLaVOxB3ZLyjhQ3s zf|y~*=0&m+I~pjw3$UIB*DsRM;F2E7H!qUSK%Ao~)R1?PJC7u&JDeBE-YHy8%p`mx zu+IqR>X{eGI!>yhj!#bj`_bU$MY61s)l`k{0((F>>o+fw4VaUF^B-I(FOs?x)Cp?z z6q+UNqOtM#{s+JshCm&k<3+NpHVuSq1+;@<%q52NB3bjMWz>Zj0>(IhmKiweMKZNM z((uaIUcIL9%m+JTYLSv6V@eONh^$ZrR!xc<{XO#4;6!lgQmcI5dY~U7vPmJ_Tts@9 zlN7T(N{v$bA<|xd2VO2P2TVSK`*%aDoaTVt^>JcZ9QjE|*4C?Ni#7+O!wM;DA-l1L z5T%v%tMedAaSo_*~ z!tlp)p;<>UpY#qYAiQmJwZd;v<%_jVh4%&iBdAKh%wodZj{nG;Kr-I;)}fbU#qTKk z4wCf?)l4B>tcQi`iXbQnM`A=?W%!SeAlS?537#&S6(}XVuUgMePY`tt zc9$^WS&mZp5y#ucbFEDnJ|4W4YCb$P-?vz`xn-q^BN7Pkj<_z~wVC z&Ps~1#R}wpsu3x#QSBL?WJUa*U+e?)t$}R^o@y=qKEL=C&@BUp5IkM7_)P0Y8;oMq ziOZCPNtPFJ^AA=K-d2$@&xQ#P)jGQDjA-EvllJ>~*C8Cl>d34^q6JB#f+MBfE-{I0Dv1+z1 zDJ7}`SYYByDKdbyGC23KDy53FTP;l&J%J52ILYU%YJJ4m zDjP1QfH2>Xs3}sF^gH!MCO&66&?DA^uw9eXj8Ta`jO%S7BRb+Li|9|r+qiNFpp?~2 z(B1JX*tfKVvY&agOC{odK@TS66E56Zkw~fsNJ5c9|C--Kc)LowZEC2nlEA2BSlo|1 zVbjg)+^-&*t*YOi8bkaePXBlfKN3liup+`= zj}+cD(hi;#!K>Q_R6%*lDF{)n+FT7|=o^Cg@b_IgsGk&Gd@i?o^BRoB(*){$UzX(N zMO0W=M0j^gJ1o0^xB#)j$015Ta{N<{`kzmn!}(sACCku#;32M3@RhXhd>Y9LI{cRc zu1`eAkuKpqBJDYCLYe;7f6ztpbDgmO!&NuDO+zdl3&C5CWL2gS*O)4mp`sQ*eOTA1W>s$?dDBn_?@Ebf5i zIjQ@=VJfFeX125J(|_P4#3G=T`CqE9etj+bt%u1%)(4}tVZDMOuj&`tqb0`f63hPZ zrvxE;gE2s})P2j*D5~yr-RMzDw5A%r1AMOLNbW4&!g_aGcKlc$X#)s542kxm0?uAqeX=gkHK2FxU6G7kDJlRrRr=0_=~&aTFd4Mv`H%8g{Jo zE|fK@VP#M;(Y%QPcKrEj%#weNf!D6i#U?18@o|@+XvXoPSjJ{Y5=QCIBQdQj{rL+ z+*a~00_)<)ldap^KfwO=^@ghM!-0))aI$uLhXmMT@EvICK=}c%B@RyZZts`?``KGj zO8*zYzI1T1czY)W*i*^{Df|quUmTp=D7`ZR?AC7dH0}ZUFR*}9rhR1T_RbEl$Cjz6 z@It`i3?4e^PJ-|*53v7k1G`BY^6L>sZRd$ZD8jY?`-@v}nraKMmmfi@@pTYS#9nx_ zcJ*M_JqnERnkC%hBJj`zPu2}#byXD>0$WZv4-XDW=Y=Qh>6c5ZN1dI(_8B}sg`llk z$JJvp=As{weTg8ahU~=Yoi9`GVM`|6WRmyh#aBf8Ml(}}o`&fkDW?d1Ez7$=R-%IgPDOQlPzgMMYtC2H3}+kd7~Ed20vS`+pIt-0{Gt z5yz=2b6TRHW|^v>)xbCU>tQccFZ@c|9m=N^eFMS?tw<=mXP})K?o)CT*j=qeuwSY7 z^+3C7W@ROx=SS=Wkhmf~d}N?qVqhaxi88<{60R->S>wHv1MMAslLS4O$Sh#39eKv5 z1=_u?)DUY4e+5{!k!R0R@4`U4)X%WAy^Z|$w5)HekY}m)<3Rgn0)k7d0l3)^I`L#h zo~7P>fp+X6m{|@3Jf;cCv$X0Pt|g0`s;T5Upobcke24LfWPed_E|f=}fvDpWHjcWpCai6@3n< zOa17|x754NWzWmbP^9f39PuM@v(yv2cdyI-w;Jp{7s0q`ShUV4-%{`QE_;3y+!tME z@Wh3ru6iiBD0(ZoZMz+u3yUJZoW|6KLEXmWqUf#Vw)>4LF06)NJmbg0XLj+)&Z<(n z?yZKW<4XVs83J!Il$uP0y07}LbahHW?71mmEHNxvcXF0U-w3VR#c~b)t{NW2LEB@f z)RV%!A2UnUk1W^bq4jxou+D;TQImu_rcbn}+Df{znnrQ|S&x8)o_)%7j=P54<1x1# z{!t#qOa!aS)0meploeilSoP54M76Pc2CUA+G!udx>4hnx>$qf0Bp2QGfpUnmg!)=z zz*6nwomAH!=T#Ek8*aM;oNJk|fCxSKdd!rcQcV=_Ie|1yY%ZGQ4EAsL*m^M z?}-(uh;3h8F^+_@7k(rScY7{UUR z=R7w4Be7F}4Q|7|@*>wn`;VUit$^;eq<;zZhPJ>ITm_@-Ze$G8UNrALOvWP^I~aj(&)gsgIKIFxMjY*B5hGV7^6{Xba3q{5BRC z`omaY=nrFonRFfRLL*}Ul69DBrY$hbaTYQWx$kO39%I;NfhpuJB~}62VBk{(8w-qm zqP*B^uyce7Pa-n31!m(5IV>JpjFG;J*kq7MS|H0tg$HoG)czkK`8?XlkC;6OJg#{+} z^ARF1#&|3ew~}$GO41gX%qzu2MG$If(i0Y#pJ_-L3(WZDjRf5TJ-~idOFUtLi7rw} zOaL~^;3V;c1*YUy_&lx!V~b&NKe7deZeHhp_0Viped?S>$^vu!me0D<43cbSp^ESi zY6JhEF~mQX=^uCTBasy4u)xgqMDem(=~q;b*8pRI*_@?}Gc5>u!UA(@)DspMZl1Bg zY-|)KxQ5tc^0yFnynvm(p&(N4GK?Gxgp}S0a2sIX%hn1q?`-hx2j12wq z2|nEdvtR~%QZfh}YSfRzn_RH(9z1!mCzDVl)L$&g43lb*1^9Qer=1HgLwDNKJ0 z%=J*bxU&efPyUzcSYX5A#r~h3rz9)k&5&J2;B{d>f~5p)*XzXb53}@8Ul|5 zx~h%^CTv;*e6t>mwT8uNj0MKs&e&97ff@E(2s7mg(0|k-js+&zVZ#Cwwcmr+YT2~+TwtfB%E&mV}TjkwvLjo3oOIP8w<=mx(LPsQ*e;QB-tO#fkvCL zz&vo+u)wrFU5iO_0hr5-ZexLYwQh#ewgcE+qs>@g($_~R`E$U2b>y`L=D@vts_u`0 zxvuH9^NX%zfr%L8>zg9LiaWTrz*K4xuk_ahR^P$31tz9zn8G^%>*nCv0<%>V;5k5! z1UAmWwFRcz@<4?z0JhBF#saf>J5Dt}L;gO(=(TsX7_dabd3YEL%onpWl?A2_uto;g7MLFoVP1I=*)I{~ z)X-R9c41TFSYX;7sKZP-74&&eh-eGUVMhcOm_{RWF;jjC`VkVL3i|p%TVM_rOcvHf zfH(aK+5!{yxQr?{=sLFONI9yiEih^4E2@H$fmbGuQ;@O1d_1#(Qq%&3XSE{70(1Vy zaHXU_uz^~MV}V&UteTRa25goiPd=1hSYV1)^R@O9V4EFzZGk!WK^Z|0Ch}Wg-#hZ! z0`qpErs4x^3FU2Ie;IjWf!U0hfimQVTNFT2ouI~ww!mz-jB~=$0MiViF;7<70@M2t zPRg?Yw$=p40@FiHVX`lv5gL{>MX3>KEHESR$xNE;0`=2J1V2 zCQp*a0y8XHD&}>tj{C73 z3(WFrxC^cUyl)7!dGx8#7MO0UgN2p%CMG;2H3axlIDHhVEik*j$&L6wpfxa5o_~x5 zCiq+tWr67kLU&DaEHEbz#&G{xqkv7-xEc}aI%^Beva-JAZ#7ukp2pM`n0fgds*Tlg zur3nQ+^0`iV9p)OR7@AGI@~_qNf`^wgWpV4=61(ESu9)xrY+?g!W$|06FT_E@|%B2TR=g%O;b6@i{!THk^Tv?<1fdG*4a4mxLTG0 zs6t8;8=lPYub+b~9rF8h5KL^nkkOB{458ae``2%1kSWF@_Z>n;0g;op05^*`alro8;1oPGlylk2Xhc%= zjDP*s78r(gfHgJvEC+8H)}=`~oc{ysWAG0g_0QYIw#JCjfTkJvNeAlwc6OBj?y?WT zk*i2r``25A3v+MnC4|d0UcAb`UN!y#Ix8}vE_N^EUoZc{ZkDog_}BNr9*eF@=shU< zh>9@&^-a>tav3EF+DMS`uU~bvp-4w|JtXThf{cHCj}l?xd1QChkm@$$U++!D-ggMV zafZOU9sl|T&sG#vlLa8G_9q$t`m3Ub*a7UI!FN$DjDP*hlgbI}Jiy-!fg9)alJa%t zZW5le=*E=kA2x+6^Pt?Wza9Vjw?_qu66m%RBx^qvq5bQ3zvL3Nk=sNgav$Bi+Q0rx ze=HcCfW2(+!z8Z#>j!Q?w3^YtCK>#gqke+D;roVSA)t>9{N(U}fBiq9Vj~JY2#!2O z(%QeiVCB4ms&kPLE?4{4^9*BpLHXBnp2rPt%G|FDpbxg@9)jDcH1&!|q9P(-Hc@YMe4PGeVLcy#n;IFQ6qdMe0QtQzkb-tBJlqK;ejD>6^(!W^q1=k zYEA?iUIdAYP|eW(^;sf9ln0b<;D4za#=l-v4HQ-@fE^5hJILv6HRh|-K+{|Gx6dOuXyM@j!Tqq}ClWn}fBpTUI69X5A@v;YPWK>}fk>WYlrW9o^tmuA)!c~g4?pMHQYTQ)3+4%AwP#EI0AVR`$* zCipj$KB7AI3Gw1JG=@go@8c6aQbs_g2ohuRH6ke4NA?E|F>_S_R@2~v2@hojEp+L3nARm8yeHckN=`ckq_|t5+6U#{THJ{JAl6ejpq0< zF0`~*1mPn{awnB_9>1L9N7CaV@ae~oC8jlG@q8%25tsDme;hydJDVnI88#KIN;Joh z!`{vX&-fmIevWOei~PFfPu{MBbnO6vtc<6S|@|3(tMmjKSWC7$VfhZg!`-G z0N(tObNtvEzs>O@{b7zD=?`=K_)7cg;wFl|i)8&kHPgqBa&IJ}KI2FMi4l30Vc+rN z#_nZBBB1gHzDTe+e!MX(P1G~kHNu2v1xnG!j}6-AIDQO&jAeyeR+>2G`0;RsykZKp z%p^^>xP0I7<9ny`ij~M+uMv6AfhSpIY7`Os0exrShYmc|`nFUNaS70E16wz__C5#j z$-m)b33cK!C1H}~&GF+v)ZRn2Hph?khL#aQAl5);T@no?jpq1q)#(gDHEai{GrpNW%^Il~B;_PT78j{gem^prI z8s11;1a{Nl{>P6Ey14Ms0vpSktyJ^5n^IzqAD69z4W|?s6%8xaDFw%me=G_YnZQ~b zocq`uKcEqVbPVsMhrSydcAZs;8;-T`5OAyHGLDoGzdzTY#i*Z{%~O?u+^u@kPc zbNu+?;c!9sz$LJ6Yl$b0A4hK~Biy*P@*{EQN|Jcu`0>h;{CJcBqq1Q+$B%ULI`^xG zW~=J;GD6kyWA6}K9W5S4`FzGm72zM$2L3@~h=0WEAFty_A}Pvo{J3R+ddD6P-%lPt z&fQm@pQn!!^u+PwpJ&Q2W*oq|dFJ>rs(b>z+-Hj-7DVYsj(-xU|NW02=|1of*Bn2- zFg%VGbo?&`{>P7Yp9)N$_#brV`0@KzvM$u#23@#BW6sOcBTKcF$0N(6KK zNOzg^5NeJelS|@}|2A@DfKQ*KUp~R7A3t7Qi$}cTU{&>(sX|?&O4P@Xd|c`wo;iMW zFM#?k$mwgO)2Wu`_;J_@oQTSCK&Ki;Jz|*S$I};Z^be;~vB?lfg(}+|KW2QEB)$gW zydjYmCOvWdxF)KkxC>UUK-~up)BpH!;~gx&v7nXxU#fHb*zF#!QUfsB7*;hT#z4*TKYlDwr>4lH8czj2Pje)97H^8k0ldLywG;_kR7u|+PNnLYp*?9#o!BZ01}bNrZa4`-%D!ALeNR%4DIXJKV<9Kh?A zufa^&3iP&G#5sOE+0n3Z{J3*Z0+Zw$V7^K0=2W;jeoQam)3zAcN~6sjKd$Vaq2v)k zQyehz=J+w2E`m9J+#j2dN%9t$_l!1k{5Zs6YDf#xmx;pau`0;$+LaOc~fQ@xBMoW zs*fM@G%2PE`Vjage?8{-v3tGy!^}5%v!$V8)&{GEKa(d(bNu++4r9U0$BTGZ!yRmk z+(_!Go8!mlGVz&@V#rU`nA$L?+gKkz=1wmmtVUq8^kX^4k9{9h6;@AxgAIX?AC($? z{CK;zC9HSASZY|b?&K^{A3y$f9bpN#fwtFBsV9Xwe(Zifg=Yur908Zp zpY<46p4?Bl&ieSV#DLn0SrV+Yr!g=49KcTwR+FaH609!7G!w!T$B)Mk!Z!l-wZ?*_ z+Q&O7bNpEU(q{; zpfsh5f>H!Qq=SG+6OgVH0kMDxDk?}5RNnh}W@pZBzWV;&>w5ntfP2s&|uQfLc3)2m^HXEA?2?i>%KD&>vthNGBTe{qm1%*YdTa^ zoE9k@!aV_U)PRvV8`d$=3ag4Tsk3uQR$)R-98qKFz8DdiQZ+bqLZpRD*AWk=5nJF) zTT0j9e9gB6Vg8HQ{jfPkP-zocdp+3d6a=G8PJ}RBi*hUxe5x#ER&dn0K4ry3s*hX= z;-f$+@hQp!_tuQwgFNJ6s9$J8%`H_a76e<%J7D>G6XIPB$d@Q11)K}cH>tE6r@!)| z%GkJ=QrWHavg{O`)_m?%8C8;s0IC^CrLuB%Gm+wQT49c282=%*)qrnD)(8$SiJT=H zt|`TwR_2-VVg^-gJjf|T&!O_LJ(^O&85unyNBy`jh4!A&UGzlnl+sS?!0XkOz6aU? zTd$E3BU0j>)^{}%#5J6RkyoHyGkV1Resx3&I(V~Xn4UaBxD)$K)tB1Y*C1Y}RkWW6 zP9Vez23SDlt{uo`sMGpobq%pX;`L8}niEL1kfSqjvq$=bhYdu^3(jb&rbu7k zzKBQ}=k(Mm>A;H#04z9oinF zvoW56+4B63SlE1t@Qb8zTdBRi+-04*g?q?;NBCo{Nvf6b3~gUaq^xyWHx@S*k!bl6 z*i<_#r5$<#&++p0w>(-pHV^}lxS6tsQjf!0VL=i~?D{?K@8 z1)OE}os<6~32%X#fqdss-tW;K)aT^8FG6RPs}a6QYjO<%`kZ{@p=i&;(7rJG2GY$r z`Mks6v)zPv*8qX%MWq3#Z6MCeoRhD4M2ObV zdKrBeC2XIQr;ab&JyDVxCkHk~dlYblb}NVmp(cW_vm5=kQVB=CX?N+&z}rY>1tnFC z8d{0v9MMog9M+YBQ6R2By=jOzN<$eDX%1`T!xAC^AyNh-`@GmV z6P1pl8ab@5I+gcP8Grb^b{WK~<8?0kk2gBIRX{bm1P3Qe?LL3P&<* zhI=65qk8(=3>~h+H<^j>xfJ^ss*AQ6^5sWmS0i|nmgF7MwaqYhT6yu18d)9(a3K)V zHbdCExRL*+*3<|f+`SMG)ra2kM0YvKLKp+I&2R!K8k=ETt?D8q3iZb(|Dn7Z;hSwU zP&#fB|J`P|3-gnrj|qq_>EHjb8Ll+V5~~bNaaE4SW@w7L52+0MNDBTP&0%bYpC;lC z4FrE}Bx5sNzfjb+8K}^F+LFGAq*se(vd!>O5kH$@8=6Gl<+Kamjm<#+7@L9qF*d{c z+NDKX1a!vcC`2XGHp7vtCB;yLKCcz&VcBOh^q$v1ybfuRkxP+mY=&c>JH%R}#giu7 z5eU#WLzCWqHp5svH71#BmQ0MX8CsnzBkm#KF{PBu`TJ~!ohK4RUc`#Q#)?d{kBr9p z7n{n<*bEPkQ;*v1E4)0psPfT&2Vc=mXNbS zEHPA9I~UjtQ&wk*kD%=`I=8X08P4KT($mnsGdjihFDh(?;E^~v1;B-V&b4G+s+_dV zP-1?%C<>sghSZ2*o-=cpx|F%g>92R91RVo8pxf&hxon1Mi(=`q=i(dU zN#h`v8p^gA=-{=FS2vATH8zP5Wivb}R-7uhKNKwoI3uZP zRF2wapl%@D#2K4mh8rgfk0Ru(iTwhV(%1|af2fW+{sQ%ZAtn=HY=#$C;#_yQ#}dV` zaU)1dLgj62hUXV1iAn$(8HiG0D3{G}b2NSdKt1~}WT4H^WoI<5{{}Yye^J|JC^H$W z#dRQd8!EW0Zd7A4oc;+jy)Qsq(UdxF^>b#Lo~|lhrV>9Oy_kPZ)rBr(Gt3)bKtUb= zY$8Dldn z!7qbtGmJct&ZPVp_@_FGZ8I#hY1j;&v|YHDw|bpj9ND;~SgdFWL#lW+*y3lp)y;WEYc~u^D7JL&IiR*`^jlaw^E_CT(Lg zJoROgN^J$SwI(%VGn8vtOvOJ4?Q=W6wi&z~(^T1ifcCSkYnx%!nTASt7RO8oo2r{j z+YA?m7gFh$fmXrRwat)(XH`%oWkYC9Y+c(7e=IM?V}R@pZGf$7n<3^=q|&ECn{ISt zGwj)fCBQO-e?S@yOp>~fZ8LN&i@TimK|B)xNY%fz&2Vlkytf-5e$|w1Gk83WRS`l; zU_8O*YXW+#J3VKn*G%(B87+Yn!27--3b`CUP6J-FAF!GZbEyPkc-Id1zmo_{L`V zpaRZ@|AFv-bX;Fw(KbWrOK_VDmcr!#*i5j5XB#T=45%LiNqf*W zHp7EUVG5UVIQ5TBRh9}X6olu@IPVo!P+0)!0T5Tqwi!y6#BY965bX>_zYey|urv(k zQU^gCt^xZlr=KHnd(tZiU#KJe%`Yn$PEe25}vfqLiPNNqE$zZIn>Ry#o*CDIHCxon1A-SR2&XHcqs zJd-juL*B`_XOZSZ7ADW~ZQftnW|+6qi3tM2^rnWPrZhIg z#Lv-CMCp$eeT4BOQYrCW$2)XR7E>jrrK73A4EhKQzL(X2*RzupOgBFZ)R*g*0bBq*Zk zea2Z$thAnuI2(Bc(n%vf-+OQ`k=5MVyQh?rZ$SFh$hbj9Py+}z4I;#;m7thrIKMAmbVtBYY>vm20DY6hm}cN~ zbu-*7h>6}D$P07~(+nYr(Sjn%kMOaLWa-}w=Rd^@!$>}bk8?)W&Cv7b0!sc3(mf+n zGhCo71w?|~0v7E@-2yT3|J4F%P)rNF|6*m)0|CRZ zIbtaR`X-4nEx_sO7HDu0bCr3J7wH(L1$u4G5fo8w!p9DhOZ&ILJO7mBY~&e87mTc1 z;IG{wO1=&0fsv^N!YLVdIKq-Yd(r z!bHc9oMB!7$`RY>jJj9GrH(23BBJiZ;jVImhWJC{Is_?*zK9xDDcqn&e&0WUjNg}M zxWC_bvR#50NR74v%vx-YVN_J(_jQgbB}S8e5ZdQPH-6vUXChg@2I(gw^GR9C$lvdy zqDC#fLRaI%?~5<5YX`rNP!mVgs1a6J`F%qYF*s5@KfiBT3f9uC5W535$9O8O@%x5+ zK}O6_2+wO#P9ehheFg7S6|Yi#D~Q)L zAZJlV#_y|=1AqAuL}vwGF?0ESLF3a^Ns0p~Zy=S5_WRt!G5f9qF-rrs-O0@ z#_v0F3f<}_gx@2Lc8jETw*9_=o$&(^iXB=Mn*x>-73lZ9{-g~1eQAJd6G*kl<@ZrF z|IP2q-QlQ<)wu4dqWr#Q$$oy{*aq0a3y}U&%A*P!AKLG;QgNC>Zh^2@i;_xe`+a@! z*bXVrL;YS8zV4y@z6V=j2S0=ul;lfaRyWQ_3jDtLi*Z{QJ0r0m5)8!|8Nct%ie(j5 zA3!q$aZ<+bYdx%(psLDe01PIm5oKlkzND66YR|n4?KNB1eqWw#h18y34Q;*Ajo9+{~S!)>3Tg%0@0uC>azwhUYxDrAxK=`{_ zlbZ?9eqRorDJ8c<+i&!pq#M7lqpKQ@e~8x%5a{={`_&=#QdQ*>5P48ME`rL&_;POS z`8tw$gOX}R4Q>3s@_R7U$PXZF(;{worf$!6A^g7M5e-3{fcljoI#3$M@4KAr6x|Ra z( z`iIQ}k-QKec)ezN2~lg1l{YF#(EM=%0$-*0zf-FXqg%x|5&q_G4JD~nD29VtmBM!5 zpHO-~aA*7yhkkjbuwW?jwaJJHQ_-7>MqP=M2k2$KM<*7~UlW(JBi;?lqQo8QL@dp; zjvp$I=l>!&1kqTMWl7hjVt83JE&}lUD*#D>kTw-3L}58yOKWPF5pF7I=Af=PFj1=8 zP~!{q^)J=7N4N}qG{VPI%shYil`0__HuziwFV&KaBwd&4VvAH!2m$#01F%03Dpb02 zn#i-zn%)^M>jh|6wXRAQOJ|VtjEPA}#<7QoGD1x%_;)#+i60qOyb>C>?$iR@ru;f) zYFut^x=&w3oi2%UQS>_IJ|L6tBd+~A=H$zDc#vuesWUc5C2F`B{o%hyDW9m(&*A5n z-l{VV|Czh~=pWJl`w>YZ^XMhFsOnl5npGQ(7g7s0tl84_1$*2W;)NEV(!tyh{BEUpY0o`^JLdYcnPgsZVEa2Xkf#ASco zRJcxt&!t!=vcrWdTNZo>B{Kcw;Sk~SDlR(REeZ%%6Ip1^mqL_CU-m(qaJ7|Tr%h!`bjRkHL;H%$pqP%*ditn(^uZYD6Eq{@T<21R>qt1&B<4jV z>gp=3pdX}IfE3;XwuaE9RGXM7>l+Ez)6$y!alD|k_d`0NWl6QdTRX5Uat*@c_DiOS ztI+;1I>qpZMoMezhGrrp4ZpzHR35YCmM>9;L*dj{LSp{R4im05LC&1Usp37vuL7_R z!CXJiF=jp1W3CAj*Q)1brUUeDn!(ea@l0{of?#KX4zLb_ChrQBuZ4Y*? zTu?(S1n_aa47)$?V)3*)CUBoWLyE$BEi_(!R-4rX4lUgdRIeqC$i?US^giWDNQ}0IhM${vWxi=*NPcuQ+kh?<7Q>VWMxP;wa$zVCR$5n5is>xXu97T0X$xbl!-^ zc(EVi7Y3jdRIOApg`Ligw{Vc$1o5Y#xY4=pO}H|+r>IC{+(?`^9ey!3wQKxY>@(7u2QfPuh+^^B>WcX*; z^;3|u!s9Hlk5XTYNLx)rZYc$Qg+0E)si0E;E*XehOF`GA^~(KZ1>FN6GITAe2=M-> z?=)~aQ!m1^DFm^k27J>;x_Hjj!s)z!wyuyF5bJ3`Qt<@aHLhwhJd;Kl7oDGQJ;b{? zVjD0i`vV`UqbR$^Rb7T>+4Nu1D!)9DA-MqLQljY)Q+5sR{j=7tETK}{4sDM~jqMs& zu)~V}u8N9(0ooN4-$j;)E5u<<|GR*N=>3OcM&qkKamw3RjB5+G)iqJvtNDm0pWxT0rVMXo@9C-z>t4|Yq9bHE86Pt#5grYBFjc#=Hkxxq2?ebJ zurUxSeVfv)CBwVh>DG5xOW#RkQeFi914Xe1X_YSC0`mElDhhf6AWyXbNcENG4y)~> z1_~+-pkgkFJ4{=LHTcU&1!V(hoeP?O2EDGM!}@K0wkVHTn;Z;agocExT52uf>g}+` zpUV<^KS6`N25p|vBUta}uzLNCdE^>|Zy}A_iE!b%vI!YXaXLqis;v%-lK?Ini1&zs z<~f~bcKMF2y8xaL#4V$s_UXli>piD)a2H<(EK(i24x2w@9jhT+2b@mt*EPlaG;q}d zkWCPmNu|WlH%{k2w+f2&1oZ?kI9E#7enmqGi#NIFMb=zK*KGQ4?4Wp{DCfE_p<&g$_)(T+WY6 zR1}V$AV&W$r6wV+FS0RrYVLA2{3J>hcMj0yfjCt|xW3zr!*99E+5L@lHTLcTbuN&k z=HN2;QZKp!pcM^HvlMO& zp~y*=v(?awip&Pp!H_gn@h5XE=d#3LRfOT7UN$7}52~QA2(zgOnU?eGr}C&Sy&UK| zgVQgFGN@cn99E6rD=O$S0A~zDzf3+TD9EbRDx8Zbe*y5oK&rQLb~`L*)C-9!rF^yU z(}zv%Q_4@A?IdynRF&x^90lIJc$52j)bKmh;u>ip8KuAshPEl6jA&d!{yYM5AcpgUSBTT(Ma4nh4tmxHwy}y z!31~1oR`+2_)-aP$%1tFVnH#PzC6Igv}rjI`e=Kqc!<1&dj_Ip51f!Aq8w$g_Ut&* z3ZsGI69>HpJ0zsz>sI(Y{Qteg_c>O|R@~p*P;{JaMJA)%SLKtG6oK%x!C*o}%~E&M zc(of+bIHU4>YM1_rD~4ozoAqGkvVxw5mD9qbm)%8_^qO8oVRD~J@{Vrtr?Akw^#nX zNJaFEP0J^|)2x{LI0qpHwX{0nMg)!)(ZCW#u&HQ~yjM8wPjT_>m+`7@ zl>yYZAtaVOym=kgh|l5_bR57115rF5RKj89pI=gSiaP-QHc%mM8k)3w`zO(K z9qWIHdg(ARCvcmCXrn20d5AaLVV!#e z6W9R=AE`A-N1SlQbgLn}&pNET-BSe(p>h_0HwjXQ1g(3#qcCB=9V}=-l^dXKwRM^R zdtY%_^?og*ig^m!CDM7vbM1Hn?49YbQcuC&r-I)DAhLX!NCnXZ*t^(aMaHG7c8UQ| zCKp6=Tkj@^B}0m<32v|#go;;??06t1#+1;FbX z5-wV>c@JPgx-3KK??e00)@i}!J?_BkJDVx}2(&M?E@^nc6`feHdCxhl#lkmW{0iVt z1F3ya%QNqo%Dfg+2l{so>#gS_8I&a&;nF{Bs)C$sEG@jeV>4+z<{L+Tz%4n$oD9mA zpu12E+JnA=(DK`R(_x*SS5!fx089>q)F5%sf!jPCD(@u#RuROF%Xus4PlvTWKQ4=a zCO^f;F@pTEBvE&{B;a!Fqup>s8DzWFv&PB#nzoxSAAtV}mUBZv2uz197+kxGoX&vN~@hG^U%5gvHsRic=}cVMMa)GihSKya>FDiPtjzj{ z<;ljG>b&?cu0!_%71(vShkOJChlh3%Ujf0cgSr;tDAc4OIbleoQdmV`fvy8XnOT?( z)A{|q(Bpge^FO~I8jOns+9SLxWl@I)80|V_#N#on!x21QOEQaeeSUw_>Vl#vl}RoD z@NOWaU58d1@ob09TGOrr6*M#M*=#OVUoT2^roVrw!WVjl{1wn$idp`iU#Y5n=)$mv zCxBvVM&rOFsmOQa(J_df#241~1naQGIi|7OrSXg!zffC(^Mhl9vHX%IbB2d={# zK-qPmHf7g=zmUnJT!&RvaLoX@4wpeD55$Me7uZ)4eq&@v|3=PF;X^1O1+S9#Hp2Da5C<8cEdeof zUvcrV()#R~6ww1AFF~Jen30>Qp3F>@R+Z(k%&dXF#V~V-@z6QR;`jmI)VQ!X3hk`X zDOGBMR6!NMmrANh-BjVsV>7C=!Lq77>Kjsny_Cc{{O4P&)L^&VtH*!#)u%tnD;naD zHdxMO`57!rsH4~h3-YoJmYPld43?`DOB*a5px6crasOqo#6hzSmN5lAT#c;={S=#H zIdw}egE3etwJOcbMd;sZ#uzMA6JxO4zvB?Ak%BV^{ScdDJ=Mk-EZe>+Ehz2MkSc0f zQfY02Wx(E=q5-tlMyDA55Db=&qML|*0ETEt*ai!&3yr~2^P_CB8S&o)xQt+~ALnQc zmT%XUU}h)uy_&HNmIuvq#90LYKoZqQ?xc@EgJtb7ASK-t%NQ(=4&cU2ISoNGbz(ll00zqs zyK!60y9hs~HE!AeBR6BPEK0!-%lC+S%S1m!#W4oU+Ufv;n!}>NX1A7Yu)I7Aza^z1 zrWt@zkX$Qcu-tw#Lr5=(j)vk!XUZ5X3of9hLqR-mC_iAQw2Cgl1Gi>FTx$T@^{TeV zOt#u!p;1k`i8Thx)E_bae}RxYCIz~(&#%$!WQ@U5YNilpDD|+XEKwMnY6xyAW3W88 z%AufC05uFm)l_%RxL~mCzT-=&Er6#DL`AR-mb@3M*Xp}Jq%k^zdnUtRcKdqzq87wvpgJo&&8Vt!lK|Udx4l&zcnYpl{O07r>Tm+9z zjR`#OB7>#Nw(2T=CbZh5b9`g4oSz<{^bXLv*}67Z+Pz&w>0_Wxuyt*)ELs#Jc&9If z_O`8SgJs#sP#)dn4rrg)x;9wWgvRqxDZhgDjjd~gW&gC|D*Zp9J+XCdu!Nlp;_fGl zv_$=}`Rb<)mb5(aDt;!k+P1C@mLXoD^bXLv*}67Zrhi>n>0_Wx(YhKhjlnV{6sv3M z`AY%3Z=egbr?kOBzwFXYqHVC)dm0AI$=1!p$B^Y!eEdp@@My?`soI^j@p|=9kqQcF zg-HQ6Kgc#%>~vwUThv#g0#U>dt`|C z0CN-hHh?t*@xD_jX@h0W*$nX!K?eYw%$1ThSo(S;XXm&H;I4t_N7go2I^1?BD54F{ zSz@ybF6m#}V8K+$7%c8qY2pgfaWnwZ!NlT@OC?oNjlmM_-EL-#8)fb2+hXI&uAj(d1p2lGL zYA+7P1t6Ac$~IWusfLx)W`uvDHMJj%!BY9NRPiOWYXN#LgC#M4Me!#9H;x{^t`%so zEcMnGu>g_-p=;Dt0t^-cw86roG8-(slA{GYe<6c~m(l?S3(a8IV4)@V|2Ko>7FIa@ zdseF<`bX{0BQl>ZSCcJ}T~FH<$fx+OEs*^vo3TI+e~rs&Lx5N($Pps9g)X8CVzq%Dv^9kJqm8vL^+ z4BykTvumnq~+02T*A-0ptlW9u1-$@&IR2MuW#fk|V5^ry2kmw?^-7t~JKSRmPN zUi=e$@>)Haj9#U25JE85hb!~x!Y|0dLjzeCA_5hOjWmZ09gj2CQ}(03*>Nb zha$U!8X8D)n#KZ2-G{5{UIH=OP^znF3uH(Wsgha?V5@;BQ=d%%3uKo_RM05^mkdOE z-Uq<~`C?*(>J;|?h_1ShWw~j576?tKrJF=!f&9IqmP$Jjp*2i2t~Hm!SRilxktk$K z5bZT(TOeyBE@2;x@G)AGbd=Z@$itt#f`(8z7r;V-)FGiQkQZx~5Hz65P0)7Oy0$>d z5BC-G3usqJ=N-?rGZsk0_P*Zr5I|5jUnWvP+5*{nv%YGlVgTZELD~XYHMq8#qSgbD zLy%u~+5$PWEmBQU`vMr2D zj6vB3ba#qDd(c-9ZGk-fcTEM21u!)bQiB96kc_KgD(|-etRaXSm-AK-ERdJpZ=_N> z0N`X`O8h2UFD#JXtf~sS3E*xnNLwJOt9=$n1f~uJvH7~5wm^#i=1|M4WN2wdHx|gs z_4pa~BD@W0T+~1dakTJuXum$oE`2<05B+@S8TOip2Q8z;# zhnnCE@oE;ZIW*PtkF%Es%WaIJ+|v z!IQKk+mf!$hV`uyL}x0KTnu1kAfzpjXBOgq;cZ$|=Z}P&3Ys}{Q#meG|DGt-?7{w} zYQ6^dH{J&HfMQmA;8&_S_3#jyFhnVgjU`!!bX}_JI60n$04Y-e)Ch#M1yb@l?!jpR zt+Uou$@pgnH(pG{S`-$@VdS3l_Z~cG0T##tplpFqo3aJMUu+A+iopDcERb(NCJ(`f zZGkKat-`Z}g9tr~&C!DzE}Cwop$|IkKh$@7E9Db)Jpm{4$pR6*{45aqNA%CW6)ZB( zypYPiz$Zh|ce=^0;q}c1v*5R5!;bMvEDeYusQZnZHzB)e`8HSY1e?%NXwE+Yx@E*WlD+8&;}Zv zV)#Sw1?uMs7Lx!>(~#{8P+zfqfnGhl;sD~W2e^Y^t{>-Ue1SPP%P?~W`UTC{zQDko zRB;Qzf0IP@kw@tx&=(l|L9)o-#}cKm`9Xocz>K3AA_G884cWfH*IyMCEfG9YOL8#v zNZS`!L;tcba084o57oa=Uas3y9R_MfRU!d1K926fT{F>IdW&e-dj4yCb zVio-eQJsBtJNLM&?lYAxFr$ZQrOWDST-4L~VKu9fix`n`)Q(0YIv zW+-lSri?FeI2#wiybNNlq5J^&0_!8N4tyWtegn|1SGB!tvemu-9qQ6etnmecJ`NS~ zXN2T=Mpu~bclT>FI~n5(+z3w*-&5+P5vihy$Sq}jfu-eQ6_gF2wSl;`6a-&jcC~Z` z4F)jAKvV?V7g#w3mkGQMaghep=*jo9(v25;zQpjl0piCRkW@U|7kGt68RH8$2PZKp zuL8fRqu9Q{n>Gz!pvuHz49R@`@FReY4>8*p*x9C}N-Y&y2I)K|@MMI1fmQdLs`xFS zwKMUJFK}#MgwmgdHpYpvghIXj0Ogy%nDQ)QwbS=$%b*P*_G$^u9?5M?LnU*ikh#Z<}o z0)yw&7QX^ZR$zw+P-$P_#?->9xS@k_>I<7+qEroGe1U13YOArg3aI9RBsGVmGWb&0 zzQ7+dTq;gK@Z(J!CQTgQ?G(82;;EHcs>K%oU2Smo1=L>FzQECabriV|)EPsvFW^VQ z7ie`(sv`UXN)Gv-3UaZ1XY+C6Md`mnRhKRbw5-AD7sLm_7ntWNte`poniz43g5Laa0z8IzQECOWii#v^DNdL*c^|jCdL<7zuhflS%lZonxt=1JVal2(MK3Q zj4!~)qV@&uHNu3%odaC^0yHa=?rs#~^95+qE8Wi`MEe4F_F`!$7b4_6Y+M|fpYqVY zz=pF}Ld%^H4;mnv0OJeP9*!g~LA-7NE~WMbmM_E82_8cW9p;~ds*3gn7KGq>l@b6_ z3`Dv3LzgYbE0_p31klSsJhS|_HqySpxQ!)*90UFp6NhoXBHpm37sut)O95;Ogt*=P z$j4R(ysb*z0go5Yk3UtqzbA}anI zXbVhy;|t6xiX6XBz>Sq91r4b3b7-e+UHbwV zcq9zfSN;s`KIy#Uxpu}E*!@kK>OGO8FbTrOnMeg`UtnBTxN4^=0P5s|v@a0cEKW^P zI|ArUkY9G%7g$uQzM7&=0Pt$Al(a9fq;G`E?mYl&G-Ue%Ul$2e`T=N1ZC(2UA4kV4 z{Tj62w9da?wlDDG8eb;v(Kz*u&9CCIK3{-WOU4%n8d{7&Ssip1#o!9k2yT1EN1Hr2hC04+YeaCFrV5$^;H^MN4HEDL%C2pq^4t@0VHZjxU?^@x=xJJv!FFKy72{a zUd2mTdLVodX`X0m&0|E90Xa>c;08NwF7odp~ z`vNql!i&!_o@rlzCP+!1j|E@nerhb8vVR4Pxz2rMLt$U%PS3`)uX8Wa2-mqMAW9lG zM`IkFvO683<~sNHOW^Fe7h)R&bjGu;?CacrOwSap5oIv+F@`xu1u)mSm;1RgGjpK7 zWtcm}nCsk+)lL%|q3t$0rAp_yQw3FkuX9%q_WS`2mib{Erp35RqtElG@71CHX292f z;OS`pAe~@njy{x(wkSTF-9*SD2rjK9IgoU12HYRtObkPSl(hh42SVDS`1w2By3kQ; z+M+m%3^RB2!izRgs_++3s;7qem+I7CczWGD#9Bl#%TgoiQeB#aD=I5 zj0QHI(0x=JV=Zh*iW8Lfd`Qc+EUC1%wb1%TwAcpiGh62(*$}J+@p-(s4B&eW+13K> zW7}HTu)m==jQH;H82_>P>PLwgYvD-4D$FE7udEr{T1a1;S2RX&dy=R=@&tVZS_|#J z3loC?j0=PUt%a2##l#!{3p8X~3vUFK5$h0qT1ztNBr>tBg~9Z1-B0oHC2}{`0v&PE zO|gu%5ZeUL2bGTyBqr!RpiY(JmV0HvU6rINujopSoZ0oe`WxS_bwx$eeV zm|B6Z{{r!=q5J??3q`hIW)?gV%Ta7<-%)2%wf)v)tF47T!E~n>vAzmR>yK1it z*1}sh4QnARPZ~pVAjn}v(-2}?3&RrYsMKaan`2Vr`3zYLDO>BQ_#Z&qXyO}dp_MzY z(vLwqW9!;l*f^+x(tm+=$JVvAumTU4C zaC=oSACjGO^&j;(8Jq5B`c_#Z&qXzSWq zsF>40rGE_C8C%!Z!uWlSl>Q5}2U=Hys<9SiP-6wJaFSMt$wQzv!b2o{` zT3BQ6X;=%z>XsGBkY!tZJWYx4Xvo92+MTc#tg-nN^gMu<0wH59Y_rpawUF1H#-v;i zdZ-YKJphe0WLpcN?ZcF{&;!~tMmN?%_NQ$G1 z#Cyb83u!l{>f46_d_fSmjDobakWw*C?b}-b9_E7BHNZ2Sn$OBF#H-kM`KRLjAJ|lx zR7wotnNF3S@(3qE835`Lbc7RFD^;p`lp0rWKx{m2UQOsDz3geYhdfY}D3>?Hkb ztc71ORWjDXCo)>x0pi#O=EgZj*qKexG=*U2vsv(TEux5NUHTKQ|wI-0H=HN2; zQrFf(zUE#P=K%PxOdKjLH=!!Gx#b|UOrmP>KY#{J(=}vkLG5L2Eqs$$RFTC%B^i>f z1wRtj!jm_PsvSPNS(`MUH-ppy(vzaZ+g3O&;)p*QCI2$72c ztTYe>`5;&e1vXXYBFeo0J~xmmowgP}{h^Xd=?4J!3`E&U&eK>61A1c86Z$gl|HsCA z-nJG#-ixcI$|1ax*3^D5*21hGLqsEJ?E>^%*21K*O~gO|F9bq?*21EqCB$q1%L1W? z)Kvnk1p>6SK+6ls)&h;syfCI9ep4Pz{zA!GSa%u{J=2M1Fl;T1>PgGR|MzD)NxWB0 zxM_Kl{n@(8A~R<%*7L>pB=5)U!W+AzkSM;-`DJ5F9kLM66r00KMfDaVX}|OLg*e;% zG=c|eNliOc)B{d;i+XteC$t$_mrqgrxsvbwKA8W5oNsL} zf-$}a!e7vuq;E3#5e|(!^bv^<<()VuRpi5ly%XoK3Ike`ck&8wo_sooC-WS^y_iDC zJ8_PvOtX6FUWX7>oGsRtQ*kKHS%iFxjf*1-Q637|Zk?Eia~bkJL^)jp9s+!Y{=_)|_{WVE+Sn z$3Q%F{5o}k#aRc zl>Q;KEk@_LpR=R1`gN|MGCK+FEa{w?(tE>OHgBgSOuCXw9^Cmwrrdr>tNtq($< zHPO;|uW;J^9o9ST@~g5>2e8;c)MP3HmE2G~?GR7Yq-K{}KphMuIZZZCoUh}7p^2a_ zfw*oc)m7-H(HR|NEoqafk`k}u7Xq8A5M`=X6#z57$X~;Zr4oRXVgLIR~=83aTkd^*>b(Qucgw8h6xYk?>{%v-8gRJY7aGTgF z5bHFh%oAt(AS*EfKJI>mpU|45qePh}&WS-*|FNlp#!z_!z^??ULxR>`&X=-%}WPSc>yqbjX0`OU2O0f%H zDJ-1Nrp(5fMd>EdMK2$8UI?;UKFKfYpmF3M2>pkm`K8U!-bO(~u3&<7Ah?XH^y&%Scr3$lVc$Ezlp z3Shd1l)>YC6lAr@NLKm^Xlreq7PiiyV5|Cu3Q9i+?UdH}Csr9e&ahzX`I5dt>n8yB z45a!e89dGjm3hso2BSj3)|PI07?hqlcoHl&)duu4qYNJB#7tVZs`DYvlEKzX$psjc z%|Lgc7+hP9p~kQ>!B(?sAqpA}V0<8?2C9@`YgMNjD(?jV-X(~45~rk~>cLj+@N|{Z zZUBb^Q&K}(?OA?!5G`Vzje@P~&qS$3tcCeaI5uDRCxgeC6KtJn*ED*E^;mAn^NclW~ zmjWSe9`#y@{K13FJJd(nO|rTPdf zmtPS4KudB6>AF;j-Aaq+5Flkfr0c=P@fD=aqdmoNDO4)78d_H+;~4<%Z<*Q~gYG)n ziLM*U^MYH(RoPozpOu zUq~t=>LDZw`If+@5_#_x6u@-es(tWWW-yu18yIFUF{bmbSyxzefY!_C6hnBn?`X=M zx6_by(W1kLu}M&zx>llJ!!x*3}z9@$GSgYn-A zEK!gAIu88{%^3fUYGVAism}$A%1GfMuzv{6q}mw&t%pYnO1tnvtT(W+EUC1%|Mq9c zB#{BFkd6w{Dx!gXAq zB`P7ft(N}V7JCZ08xxL3Ej+|{bM zaI_wXmWJX+XUdpx+bdvLeg?#FL-_$P;atz*rFO4CeAfW9>s4(#n`{G3IAVSl+)~DbYq`0Mf)W6v8HlQBKldxq zDyyJo06G|mieQ^?zpugtokJjw)qwi-F!%TDzJ}kZSr8X%KvMB+6E2;G8e_tJS1X7~ zxgYou9mO``>f1C-xbjb>FeGn-d_XiEVzvqQM~`HcTE4e%{|h#544)1YZs?JMDt;=o z4AMEiG2xCEk5_sNXzgrWn{Xd)4^sNG&_>z1HsOM2rU>5Yv!Ttmb$x%2Guihd_K%=# zvvqC4{joiPk4kwO+Id^oCfrlys;KmDLwjKB+JxJ2A)33N%(oQv$L6b_HsOBwt%i!9 z3N6FdwF%dGyHI)yXzgrWn{dVQ6jl1O&|c8G8Z?aw*LI#qL2m+BW}s(iPX(HABpMU$ zDSJ=DgnPewkk|xSUc$$9N`yy49!z-&;erXb>*-_#iMO%-$L0qa6RwM$F3y3pIaQfS zSs8dOieeAaDqWax`#&qFp!NWI213S!+c2|)g2n-OITxhw?|FJ#kb;&2SeFaZCfvo> zvef$JGXTdmWSel0dlXhC+>g+18Qqw0#pw>Fpm#8mU{g&HXu{>qNmqwOJb=mu;yq$a zxPc?0e{Q z&FkZ5!hO~|zZi^k95sNnFtPaPrIMZ?<%MGxh|~W|`3@>=!u>TRK^6CBppOG_s)jHo+@ub1YV3_#jvGF)`DIVd z!DaBJu1&bjUuUQ|+2FgHI8<702UTuk!d-bPRkiqNpf4MoO*plewF!53K!hTfgW70F zHsSn8m~cr~@~R>n1NEIDd4Es^eMQhFT=+NPs!Klx>Uvk#l72ya{u)fUKIeTu2PFVh zFcAGR`5>5Z^FAxZMU)KzJY^tNI&H#bMf$Sq4`8f;C_Bk{8WXP6yu$eX1M#M&Y!mKo zOUy?;K=?MTsr_I~xcRc0I0o%MJ?I)h^hYdcLxlh=5tx zLdRh_{A*fCUsvNfR(Pe7NaoPuMB&2AcrTSo5wzNI>;e7-w!#&Q*A(N@a zKmU`VUtY~C;Z;_pf~}%i5uy*+k=S4_ju2EO&U9Z79N2t3b1g$CQ*)MwZA)rIwk>J8 z{Qujwq*Y<&sFNZ5BKIRW1H?W>R(i~HboN2N$o(qHsSLQRD$(>tR#|m#laeB0RPxLZh6KpNGtIg+2o(v-eQV$6%&IbHoC_vsC=^SglQsbiQIx&M8Kwk!B33UM6{tHn1CF6c}nCEWkDx(NARIhMDK(;qODtGPgft1AgO}v?-*UEs=$|$mR|@nMjoCx*vdQ+) zd*HkF&^sY%GUz1cz!M+ zJait=wTJ!)7eGt*ap2lRr&XhLe?uWY51kg}(tR5t+C%^B4fH)(WDRE6*tj@yKINf3 z^q@C!eY~s&v7rIpBEWd)`KKd^P7wPTfJ>=8^p<~MN&Gyt=|-mHuq`jAC9DgP&U_gY;)#{G);t951cWoVhCtCpb!8+qtQr)Da>6|@evu08b2nQ9^MGhp?-)~FCDi~xa|2OQKA#&Ny8CdDg8Bg%VIbP`J_sKA zw?)gUPVpLmg$CNoP2=;>X~i$yBpMGr?rt|1gRabJ@i7aIs}cVvMIEdwyr(&^80+f zcp$V?aMJJ=YW38&Vvg=4^MQ$L;vc0qJp*o z*cS+?feIda9yARVQeMW#j|ACyD+nI?>F79>lH7>Ly9@&JoI3^gH;}Z_+=o#OFHL^($OWdf*L~|n1bcU#?wgp$8Q>`vbJZC&8FZA$D=JG zge(oPA~q(}!oV!t*q)~~=zrMrw3HL>W+<<==V@J*eWGX$;U0`MvlD+w5bn|RZP!Dm z$gcfCh{!DQ2p9Qb^-^K8y?Tj!jJ4T%aGR+lG4y&Ly?S}OD4xM^1i@cwDRA{td@9y! zH=*6tx-3EQ&FbaGt93BOHJrQ;xwRMn={Y=Ud^Cbz)RH_%y7uDV2+uD*M}U-X0azUf=_>?=jKIB@JG7?u zw{TMhqYe+o9Fx|Er?>j850@iFvp#G#JX~A=>jWWBQ(lGf&0Zf;I&KpGeSPTN>f-3x z0nsJ>`ycDW*B_*cQHG|hRgPwTIBH2VL1maj(0OVOvp%fYFpCSi&PWALq25x$)0zK?EK6}bT^-F<6%u8ECSr5kq>?G6r=#$Mx}pw3n`lM; z#`2vH5wXAHWTza_4N`w2-y?aBY!+Q!c!oJXxSb@PH`-&;g!>9A$z9sh*6n&BpGa!I zgGxRw!n++$;o_Pl6C=u!;F;*?g9Xq*r1CkXB>$n}u`<<>-baXU5c-o=;7S{LhT~rU z{Nf)-ZY0bRWwCC;mIehvE7S_B1mKe0>@I+;*_G2qog(M^H@P8oC|3o$z@39a~7TmS%~{2g~yQR z>cC&95Al~y`pa5;iKm3{1m^-HqJ-xIX_YRkE+F{@@zwcoC2f(`&Z`yp zC(Zg85}yTsLypGPBAy-6N-E-Ftu+070^^_l2QCuZ z_izc%H9S;yAMOm67rg!k=hO)+eAeoelD=k}-6YNjIq~ z&SSh!x(Q|;@w+f^V&n8Aee*@66NsJ_GKTi6jNp7MhxJ`3eyux$dNwdlHZ_gPk;ZV( zVp*8F0Um$_PQ@eP_pnhhEHbkf-@HXD7`)~us z*AQMP%1hc$UC%${)|W*TZ0y1Xi1MBi5EyYD)zPq$&+`Ywp|no`GYG~FOn z?U1wo9=GU3CH@Th8O=y;EI#k!+2^p_U!*GN7XS|o#OFo`rhGstqj_Gn7DrWVW+o~uFDvJuIu2p6DTAzd{gT}I>i1vxCPu1;S*g!Yfo3(>2H ztJQO!*^dX=6!;Wp$Fcd9urw_!JWFMz{M>cvOot~j*xDV|h)J0RyhW}kt)^j?^0v%~ zwWGiv9KEy|lX493DHMeY=xc|u7cmG|2)5iW30%4d`@%us>WXHe#zFT-^23u)|DhXPc$Z61K z+VNSR4Si*OF^lw7(AJsw5&5clF_HvZYprnH|AO!nIaFQ zu=1YD6rMxD*4IB`k308c<$z7qKAO&|R-5b`{j5_BC8Z&yX<5<`rE=kus-8SftLmr- zMK%M~HIU@&6j{`1b$Kpak)uJq97ysYsmL^^wYgiEBA0{O7D(EIE}ska%yn7^@UA5q zZ{=yAmjiHh!ql_gY4va1NI~}jxDM!k!!@E0byC!G&}rpd!O33_h;oLa-x2=`Jg)4<0_-0pb0$rX~#PFup%MQF!XOaEV2Pa7+O4Y5*l7 z@(oQC)ueQG=@Ts+??BvO0G?#1WW2PK{+7@R~7C0w6jM5*?n!%qJv^Ywbs)4?PC|WpCcVwmey5| zQ2ROVC(Itda9RE8S5stnP(%NX{PDXK;rYsCy%JYeO{`u8wTMVFAjk=@SVga?IMd>} z=CVGo7OKdPLHSL=dR>A^ zMKoP+Q*EvD`-Zhtf%D-f&4W!b)RaDEyR%vd)w(FFFF~ zZy=sm*%u>wid)u-G3CSr0Ivr^Jex8L1ir_Gm%Z+nFoDl5R3J-a|K7$Wve$p}|1fsm zfmIY=yr0?R-kVDbgd_w)Xi0!j0wDyFK!AiEn)EJRdhfl7Qbf8SQba%j5fl`p2?BzE z2qHzAh;#%|REh}rzGvpn*?aTD_ud~1H}m~W+1c6MDd)`A5n1jAvH1K65{OuHrVjQI zye9*m;;*Pu6$^{kf0H>nt*@>=cXAVfe^5R%=xIBFszzmzIVv9p-CsgU4{@VMmUz=I zaL(6Q`Mmc4reVMM1s58F-l=n*yOcBf`?p3***;`A~D|K zTJw5}z_vE<|AW?@b!F?ESlZ!=aAAZq7U5G#qstLepC#lQclpa9DRK z{a#9%D<&6LVUjc1bJQQiHO7)|{Q;4VWduV|Im0PKDRuP}-83&o+% zL;C+~>X_?DxN{+TmgC0&;Vuj?>t@;9!d;Sn&23i&bFTC&Kvod$s`!rCYd^)2e#IWyDM8_H; z0$O3#DOGyjloNCU_>m;2@WNde$)wGhh?^h&Xou;1mtd@hcJkT7@&X=-ML&h-a%Y71 zptSl>GG-1u?F@$JV-Y;nNOBPAW)A$1T!q941W5TFfRB73GY5X6I_xgKFq)YIPenjl zNO=qzQL2I8qErDBy-PJ|BfO>_0eVI;`%xvDQiTsKj{ScSr2q~k$swehQq38aON>H* zlqmqJ`$EP`t{%QGAzMM~Vsu?H<0ZFx1eR4{0*2fZYUIX(2fhNe9H{yV6upnsSD@4n z`zuh_zQKJ(UxB&~vdl32V}AuIF;0K{=o^Gyz+sJ}ilcYj|GelZp9p7MxL##(9y2eM z;DAW_L-d|LteQw@^Dt2Pm8Bxc_?6KW&*oP~mtUk=HzJi^*&k@?@iRu^VI<@Gj8^}h z9jpH9J5K$VHlF?~6Fm|C8NafIlf3-OToP5yeq}pHp_=%UQPWTq-q*1$#WH?ntD)F_ zWyJlLUs-=>wqIGUFEjKL{5aM>aah5~9(z2{J>?&lB%Ck_+0na@Vv0g9Wf=A=qcdT@ zvQl@$MHnJC2G)|$d@7|W8|w&XXX%>JHe68J10anyvZT`5er3Iu3o#qoyR1_TZwP*6 znMF&9&j9Q&knLAS4a?rwF;}Mwq7dT$4Dc?&>ikrW>{9kggQALYe2355{PvIog{N$v!3kSQ+!E@d@GH^TZq#D@&nXPd&= z8kaJ9E=V`A>{8axiB-yi-=mw0Ltp(zxW!d-)Rkr@qu#pCU;G@^|EMIYQ0lD^sS8I` zHKn0*(seYxzJ|sEn97ia zRFJF*vL4ZNi`gz^Y4a_eS~qCDIW;vEN&7mU{Jp4-KLy$hj?XS-bKk`>C4%K@{PUr$ zD-WR{%XQ;dOyZFK75+J5>&B%l?4`0oUG!_vezA4qQs#VAUOxf-PvOI0ICS~db~?_y zmTPsB;_9xHanMTIx^XF6upzHbzb3SLwr*U?4&RMa%};iN*4x&NOWE^oWpw;0&}P`W zaVeWPGpE*9L;KLyjZ4|4iFLL96}00<*Uw9KDa-%Ft)bfh{$i*GT`A*IM(RQibz0YPt;mNyDM<2lX)3mlG~!iGxdNs2P9`z7V^V{br{Nm$CzG^C(ij z4ty>}u^-YpU93F(ey^y8)&tn;3$aUCmw5#>^c{c;*&x+mu&?8y4>C3M0Dw5-lbzGJ zlobvss?Bwx02DTm?NXNDO3^N5m7ry?&Msx6KSA%JCBi$CrrsL9E@g$j2-CO47yz#` zq^=0Nl%+q*(rtSMfb|5an$eJPDSNUdm%g?K0DPAXGA?CtuY`$m7jf-=0q}qzRVJO1 zaVh)eK~+(apqxKo@&Jc-cE+XbPM!djos|rr8bkDw73@-0@k@PESZx6GVu-Sn^e4NN z6+=tleI56qW1R)Wnhj(%$5MB%9hF_mD$Pn!?`>;0(3Ag#vrE~6UGT5D4eW11)#0r0S4V z22bk7r7U=|pMG911;2^oD3asYpNtB+7_KkNH$X2iuAG7NRW>eVk^4h6`2dvToH zo*{ZOc_2SO*HbJ{sa)O$u!?7tyT7Xm zXK;Yy;DZ{X7l2W|&`Z=*d|b*1FfL^@ywLkPs<9A#GKzwf#Q=5wd>y2y2c!0Nq`AER zzkMBPc$4*xqk>55ybgUg*xN0^NZ5D_^(*L){tp7;a9DF`7-;P6?$kw(G!4NujikFB z##?CMq;f*Gf!59FauLO6d%GfKN{D3$83%ouFQe@3^f(*#c6yvmvj`C}9WpU$gSXHw zM9AhX^i?%vNZMI%O?wM{{7-pkFn*(l;2w*%>F=qG-=gvN{~EuoL$Pe*w}(j2%lNIQ zm+@Qg`s<=a?`q@mLaebiR5`o+b{*rz=)o|D?^*pX{5H-mb&(T0yF?sGM<;y*g2v&n zT2LGC_>C-b?JR^v)y^46x)#U)}3<59Vg^xp5f}N%!+1X{;yBMHU1XiC>y9{0^ zoL#<%#TDrcY|IN#bx9S^&MsFU=M$soq~$_Hc#k8g>%q=0b;c%ns9hkwW=dTYHGZHC zBRAj)q0;&pz+Hya_<=W!{Oy-0L5)E=f5IdH4r>BkS$1|gojafIAs2&I+SZM;%fEjV zP%Tx~hStE=jkC);^*wrbXnk2X&MxX1u1&__>=FoHePZN%{Ii%~b#D~(IJ>B)yM{5n znH3kR82Kf@g9Pi6D?@Q?7@2!hZmnO2c7t`*siX}f#okIfGjSR7-#Bz;T8Fbs!CK{X z{KC+Rk*?EMmf>)Asq}fK)-#~hwsqs|QmtN5K{HCS6SO|0)5*x!sCtdF%M3i$q?<(T zoq`P`zu8+wM_Yu@l^pF2bybYBOXiWvy6n3E9A${AOlP2z!-kQ+1*K{72B?R=q)L;W zT~ZIg2tDX$%nsntWz|i^IJ^ARBwQy|4nR6Xl&OAu(}t1tzlqRLO8}i2qO0nG;Ow%u zPchwS#sGMop}DGRJkBmO7Ljfe+1aJ^?kt`5W`yqJXzHw0DcITNPzmf?b{51XL)p$Q zQ|>un|Ap{>jV9?Xv7KGoHY_jb87lK#f&Ui{&xJS6E^F`B6ZAxtY0#?Kx^Z^tm|8&> zvkkOvq^k?B&W@d37W(JYt!F%d=_V1WA>-_l`}GK2PpbfI%mx`}mmZ%xbnpHUfRhAy zWoMjSdZA~ldiTEqc{7`ow1m3W+8{=6Qmt7I#HT?Vq9 zU7n`JX#EVdi?(i@U2fzrrS*r<|mohai4dn+=j3M1VJsU= zr2?pGAluo+uY5hNw}#e%b#`_+HWIzcp$MNqnkuTVv&(mb@olKLAg*9Qm4KaHszqQE zzO4}V`U1)$g?h?r;fgwr$|Hrk!)n7w>L9Ddbm}_`cV%32ug;q+p!V|tR7zh~d@ zTK=vv-h26b%Gq?Wj+$jfFzGm~kExyW^7r+d8Db0REupn%otMAg`!G`J!yt`eSrwQv z@?MP;kR!&OK);-pzqjKrUj8PO7Uj{2kL5%Um)GYI8$)rXkfrJC?t1 z#bXh^H^gBEkl#~Ay!@Ti8tXuBKwQXx@ACIs^D}fwJ^`?kA)Sg@{{Gtmqn#5Fe=vZ( z{Jj$AZ!dpGM_A%4o!Gx19k+1ym#93v{Jr2#PTlb-1g$9Ry!_o}ls1l*)u7d~b+i0^ zC_P*+e|LoT3hTW5y{kCRcMQT`Cyg$Pq zcLdT|WLy5G(|pnLclHKHO)P8E1SkAn(2mmHsgt3Wzf*oeqaw{-{*HKr9(5|9syM6% zl)hR1esC7ASJ@ol9gQZ11;8wSk7!XD_dm1=tUF2PS1{na{M~J46%m9e zQtkwCkSUdo>i*L5ccDAEHFX)l9fnk1y!>5YaikD9zg#%fdC1(9u&U!m&tEPnTCB$1 z69YBD5V~R8cb_{a($1;|4_S9QaSxWmS1+qqP*?mBcd z8PWx&HH?9=R?viVXfkUcZf1aT*Ej3HWGh!A{(fI>n+D6eDhb{ zLinFXld7(r&CeAT1Gid^o_QMyvKf)VsPZrzI)SibdxFHkE>`evBv&zULOC(8s}(_d z-bO17 zrMTMFWOXqij{`VwAYGgrI?FY#9ixk2{}+h=7(m{jEbpq%Z6vP^EA>_mahFOGfsz)+ zq06P9)U+LCMe_P^Rmi~e6&eaEp>VfFS`nZAgWJBXUqDc2MAGjm|A4UBNJ`x)^g~44 zXo~FHc6S85&QK3W19s)d;DMi_V~8!4GGXrF4xVEPtA{%V2_m7{OE}v z@v{O0y!~IEUDF>Yyo+r0bV3JMUC{a9=+ zPsA7dzntn?5GE9$2KmN$g{ns9X#8L3!o&Xs9Q(h_teFA-7lgdevHMUd+5e?r3br?r zyP$r}#2_Nr|K--Syzqa4_$vdHgwC7&Uv9+J5!UY*_2STHMyV*sGtX=MUy5|8DN2LN zd;zJh6#KuF*^Rns1+2&aqPG9bk@wPZ|ATmwsTw&^^{Q^z|D}3KT&`6hHX2Ib)5iZL zW^05PK_&hg`uB#Bsh?xHHfZEsbtvy<>x7wl~h4L{ej1yyzYzoGrh@!9{S zGCmAp`@iIkETJG-{13d`ap>EL{a-R{8vZXm(t{NwTYzjwv??3>zq}tFqf;9OZ49Tz z{x8jE!`~Ib@-6(cgmfL>_`i%BP+FIL3$&fKZv0>FK2Ov7X=oR0-T1%EI8aQd{{Y%Q zwr>1i2Dvk}9`PsqUvPNxH~ue==NDDa09gTAWm`A?FUdRVYrO@ucC54i%bix4=v^Uv zENS#$QV$Z_|E0+}tewt-xY7rZx_ue{my)nXm)k(>Hk9rEGId80U4*mHE|IRQ(D=W6 zbg!!JD?EnwoOR>>^4T#APs9Gg_!oy)3EBT;scO2$|D|fbAVtdBz?)=?V*Fp$+EL*D zQlM!aMaogYr%)6spr;;;|I3}9(%}CB@qJ&w_`hV1s-W}T4gD}Nc4Zp>muh~|I-@Jl zfAvj={a^m}Yp4@)J%AB34&A=B4*!>$_iO7!ibG2vU00{^f7$X;xQ<^3T0=X&@qZay z)L+N%0j-}M-}t}89|#alQBXM*+M9NK=xcvw}X5xC< zit&G$e*#-$T!Z)<1Fx!wmGOUhXE*F|9S_l4$Dz;O_`m$Vr-6Qy#6e0nvZP0p&PDsG z!2hM-C-94);6|W2_>w9+O~U`B!8a8&ITF+~Us63JH3|Qh`5)ESMnyBcAN3B(K^%JzRbc{5m8D*$LFZ7m>#QWkza=;@K{x1lBOj)SM1$8yq|7C!?wk~ieI^NMZG(%O%{x8GU z#cQSl^jbcQR3Q7m41OR*TL8ToQvE91a+3UCx)f|ACIFc23#s0e8aGhSP)*rUr&PFU z+>l;;XK_71*zfRk{v#qhKhn$(m-+EVm`Cbse(b`J_$#441BCJqV%0k#_z7xF6_E!B z2MX(Xpm>s0!E6V4G28MTAS}!6JwRA>!4j1aZG?*I`%iTT9%;Svl1mm)15W6y5#37a_*Luh@&@2<@*UvsQxqzoryk4zVea`TbORtB^aacSA>%?_o$pW3e<8*$su>{UdMQw6 z3$2k!L^#a=;ryKjI(|=R{q6W> zfKWLwQpcYLZI&J13=o!V$q;L3fV39cMmxS4AoPM$!EVwIK|99rd4SMoFdQ$hA^bNJ z*V9(a0HNwZ%$hnZ>=KSc*V171urdRLYiXr~EC#W(0fZSK%&J;jKT2vrYGGtak0_lB z4-gjg5SknSYJxARvf}|l=!r_2TnuWRFR31qJU|$hFH4jAL7nv_?T0Q85QdB>r}004 zKJ&rt0m6t@@fwQs!^aA7=o(RHME}?WgwIaYLH{2_U8d+gV-FBU%!N7O->xGbAe3n15LRyg08s#k*Dd2T4)QM@)~;q&3V7Kf`1nSGsz0k0v>ry+ zEka*sGeBrDuYe{eftvdw(hLypjB@La)p}4niR1@DwgEz&uL^1MA}C#>swc$*gb(T@ zY2YydBh~dX1B6)>B83%#PIW#U`rN6j$peJ3qOdM-67(v}P*w5(VS3{tnrR8WyALB3 z$OD8G4K%h>kaIe8YBCW{tOto-%(itG) z=eiKie!GL`U~%p~5_?1muE)FrFI=v>1x^MdPF@^JlbZ-I3zsWL7l;27v{cqVC!H5A zk5ob@xDmuQ4EQcw{#Y+fd_i*=awv!iOsQ;CH;Wc7n{IBXsl@^YBYUEVtHS$kAgwssVHwnFcvTNvpy0G2)0oZQ+M?Osv11N*NiBq62W#288H8Olg>x} z3nE8=CkailT|>*RnC+HDpq4g-?p(85e{9JNjKv@}G=Q8>8L?f%57;kQc7@oV0o5qf z3A%Ul2--8&*{-3%q*Sd(qRtE8(D9wduHo$iA=<7X1zHu>*{t;$8yD#kbt;IPR9e{tzz=W zEohFEH!qEOJ6*hm$op`}Un#Fs_+@WiO6j;seDUU`afhm^=uTu{N0;>He{5db^vi-G znQ6*e=g6CvP8yUbs0=R=^e0sZZ(jQ0<}_8%-YoIvr8K?6o0n3dm1~2V>q&U|`#4PV zGzqGT^(WrEv?f+y&E}=F8q1rP(jUBeDgD8lmp)5Q5m%A*Z5)<^X2i_qrI%_u#505j z09OipElBokURvX4>?R1QILo0V^X8@R_s6z1tVNP0+)I$2*}QZPru}*IQd<5kqt2`h zF}!){<@_;XGE$jFDHTxpdp0jE+dEt=Md%u%$XHvRX}vlyr`QSUAj>6ed5$%+ZBB6> z(sh>0lFXZz_PJR@JoM3|iZ9IOrF&6&HK59Ym-hRlmUs*;0EPrAX|-#oPJ}lvt#be~-~~aHV5*^= z3pOuxeOpCTg_gy-s$<@~v{;om(E(b2)+xSsQL%YxTw~f-)as4-*ivI8^x}^#^@hS9TUxTF zjCdOXYboE^R3G$962l)`Qt6tHEq!$W22n>Ke`jLw$CgH|iWC%4-o`(_k^B~2MgG`Q z%LtW?3_zJfaA?_lZ0X#Y0$MH(DS>7B*b+UxgqzlEvStr0C$i@G7Z+*HA#fVUJLL?} z_@Y?2M(5MB8OZlZc;@#<_)tpb2o=J-Qw|@;tl4w~FEEn)j&$=*>76r5oS|xv8v$(h zh0Hsp_<29nfzix6g*wI&UoOJFgY-_JiEn9|RJ@jE4Bay}IuK3HDS$tCN*-YF^5s)%V!Q`S01ey7|zl1oq-RuOcCs)OGtTT3OV zg6?LC-zhXA;CBiYTD4I%JV?mcQeM4N=3&Z`-zjfn(%!sNXab+#Df9=wQ|J$VryST2 zCsGhl1&8$)mCU?TrsMl^O%U4NDDsJtJ@1qU_z3nuNMl$QGz8;!O7{i{VkT=&(uDgr zoTPcDw8Kmhzf)+%po}`RGQ`;Llr{Cl8KiQNQVOC(%!b(ej~5ZYBlI7m$lSI((^|97 zDRLswC>$!H%xlYYtQ8+RMHxtGEEgo1-zh2OqC`C(O~>bV%1o5rO{L~{%8i9FA`XeX zhCo`Yu}V^k{7y-^TVGHKS3~-cwY;Dcp8BhI0C;@+)0@0xsUKi zq^Z34of43%ga`@`5Ro{PBr8&8{7xwe+m6!EQdswWr|kP9Ks12X-bdGo@H^$#ju?Or z1~HDQbUPQkQ%2y!WAmUbV_nrTzf-0ROcI-+?PHzddlwb&lp1UEiXQ;nWQa~h>YU6w zWlC}t@e}~9m8tX1_D)%cHe$b1>ifZA6Cq_l(=wZ`AzcIJol-VBN;HDjfpvU>cj{Hr)SCd_W=JLF3|LoEBu;a>E?)=~%W!gXGk|Rbt)#5<&~&EL zHS1YbJv99u+F4ugTPi{%&T+but9f?5{R7%#)7PS>d($)X;#wydwEvpyDzExq)dOR_U~4eDCzZRxf< z^K7I9^tN<6oKc^bGuzV9_4FRU(p68-__n5=3%M4BjzyEoXteeghZVoYa@rx}e2j7* z>_WOc7jo|%e1=0_hw!@*c*jUo(uW0d$QJJp_Q<8 zGZ!-bNlxA1YCy|ko##ScJAkV1i16N|(PfcT&-PqM-ZG)Wnh0?Y19V?xn+w_7J5bGq ztON8Jfpiwx=0fN+Uo;nzy)DuGlm74FKunN^S%IhV*_?Q4Uih-t7r5t>3R|HIuVKPU zMnK7r!}^RGsTPY{t`a+d$|ML?j3{^5M1tiCZHb9J*&J#|CcY#>HMgW>%k}k~RB@2b zQjP}k8dEAAP1UhnV-FYbP%A*JXG*1|9tlZJEm!XS5#l^Fc>w=>L(oymON~#|W6^yZ z6_)mce{MA+d_5ZHOI3hQxo0QGqmy7|?}u3&0Ws39ZkvT*q%#5@!)Cx8R?hq+g>_RQw_}UL>uz3ewf~ zzWTqQGO|A0oFo!1wu)D7InNMDx#e`q!WWXrD~G+lqjKDVb54_A;U_T|;p2>! z%{gaKW?}fhLtM_l?{*4s&M7?#bK9Rm+{3^_0@Rft=bT3aYKo@_k>>zhV@RclxK^Mamm5FMaLaP;w?Hd+_KYkjF>Gt&R`dP>3$iw*`i~43CQjN6yn*Uqe(o4 zOhSm+q9bH1);i=mgnWiW6-RcUGMOzp0#X729D#V6fp4i0yhX>GJ&?p5hz}W1r8HY~ z^!*dNZv+;>Qy+(&gFY3rMaRe2ajg>pWH3azctclQ?Yd){qXmGW485C#GkQ@SnJqdh zv_kz&2fvu(DBP=v_gww4^RV0qV81V<>Rpk%MaTQI@#Z}T>>i_b8N5)JbhY>hzN`K* zsD2#!#BJ10TD_PYIi+iSN^Q{;buVQaBGlwa>UwA@3}06|WvGrIdNZXiin)^&j@oz)> z&DPBp9lh7Vd5&~G{RsHHbrB{*&tNCEG*Cz_LlybloH>vkM2d z=*Y7%R!6IZ&{`br97WSmZ_D-5(R{k>9RUnrh^kDg3^a)?I{GdT)Z}zfi+xFzCU4R4 z<7P~9ehgv6wWd`Wt|MQFGRGkC2UIajCf+!QIA+tqCV%0{vo@xVVk_|FjbR^`- zFNWY%Bl`dtP7t+YPfBKsj!or@i;)D)0kFh3C9_3G#c+CikjPtfe0wUtxUQbT2>qU- zd8KW(=;)FXr=hz59x-%>t|l3SId0>-FtD*y3FRt|mjDj0gcO6#|BKv?756E#Wa!n1 z@vai{J=!8}T_tS*bTp8?MaP^>4F$^ILH|QBR7FuE=kL)v6F5c@e^=!X3Mf#<#ACV`hdx^s zLqEgtJ=#`XGBi{NKvQ2xKd7)p$I@ffbl!adj3CI)TSM5Q>N^z`1~OHQ?#Oz zT*_-jh3bk|RHl^7FDfH)PaJYM=hMC$4%mkS5#KauCAhgrQf!vjipqx480e@oD?<#gs62qv zbUbM|tPfQFo>ltHi8VxZgf=va{K%GPTIu`!MQ2ETS>9~RbF5>V{KZ5_Z?L?LWL{B8 zjL#>Q`e;(c=M|M!D7~9X%_}Oc3RV-lk;o4S{E?!5MJe)%%Cp5zK_z?$=_$!aNaht4 z)g~#NS5&HE=K%^Ro`5TiLr35hmCJ8sVe3$YXOX7z;uV#*@P*zE2<~Mh`5k3uuc$1@ z662vwXWe&2<>i^ac`|Du0i{abydf_qSB>9 zJ>ie$8;(PrfTZ}|Ma7Cr^1J0l2>@vf(Wxj>M4J_r%&VcI5r8%Z(oYOtrT>ST6n|p0 z=+_y7?tzJ*-!L)sVhOL(yEfJnE1-SCI>pGgqVlL#J@~(YIKh;?qCz)s;77Oe3)QOK z^t4LPbB#=E@MTd)Wsi4M55!d85v78tA9N1t2el#fW4HMcf{Ki%guFg6x;-jH4MJWj zi{kluPz@d&sHEN`W%HnV6TAM<7qxkd1$BD7#X{Kif?_P9ZKtU8!|R`mRR6vnRCFJ> ziDeI}Gv%XI0#AJ7s-#o`z8+NXEGnz;xI~|<7`HvB7WAhPd1Y{!IIJtw;`kGzb?Rf1 zrxn7x7){_OG6a}f4FLdX)1L%%!`U+h6O<|H<9*#qj7Z=8EnH9AM* zK}F3#x`|^Csy;1I%4Y}(DCao?`4^RvJ*c`C!|bIj1hpg+Pl#X-svB*vLY4usF$0u@ z&YL}`QjWUCD*#3_M5!n!n+MhD2>1emTKxjj*Mq9qNX!N70CwbmQQLzm^g}oYUk33z zQ#sJ&b@|wX>ggI78aa~!L?{k*DfNA8wpbXJvx*QjNm~+n1;gmdQbT##Vqs?e1P#3e zpe;kHzBGg_7EWf>)X-1>6Bwd0*&b98G3oIChxk4NswL1>wLPfDE`o8=P7nu~QpvCf z)d;*9>@61ZG^wOWc@OwsCW`GrwRH~D@SqwHTUS9cx;#38I4md4Ypbp*ZL#qE!1_A1 zbZFH{S2uvx;X#%7G+4)P53LKwXAi2JbP@Oyqj|R^D@e`&IfqkY52|pRh6mNGw__9} zw}ITvX|o5F|L?VRYG9>Z~!Pboj)r6;cwLTKscw09fR4oIms%Ll(R9X@JuID#jUWS5XVsFGCEH6BzO zK1f%jTnGHKY*CB{RaH9*Jg9~QMJrNX1b&mEPys#lU_7Wk_`AHYoU@hpFHMYHRO3N4p`O3as223bzUiidf9WfN1%OY$2T5S^+zTMHHiBK+V6IJ<3TlQa-2}( z_#7$dk>k*plRc;!SHO#+7{beurYb?V72`owE&_$C1Fxe3%>Us7eKNo=vOd~8KcUI2B+msAf) zO=63M)?2D+(vgZTFAjZrDtG&#%N|q_SMzB+8E7>hT+e~g77I#UgEb1k1OwR~RF7-aQ}t&phPKY=x<%;gY&@v;?JA?m{h&_1h%_El zar>im$Ld#5&xqs)LN*Vo&53SJ7OZ5>PxYkOgQ^|AZ%QX*RU#m_x?aYE>f@oQ!fK50 zb~topQs0q1s1hdV=4K6rK7kpkO7@_dzpbu5skfo8^Wl(Pe1tctq>aw&T&rs#QfNdN@x7^Ck|f zJ5?z=9<6U$UGyb=4YUocv*Xd93kxd!Af%%#vjKp&9SxN*?UTIdGh;Sscsd3#JJIX0 zXOnhq$51u|;o&%}`~fJ2nN6w}Umg=>2u?PVEJnJSP3qAmT9ihBl#Kwi@rBH6Qm5Tm zY3gk>Gn;fBrHZI_73tG#(&Q?r#nn^NMZg>xT=i~#dp2p@%y2Osk+=M6FP=?${5)Pof8rBe(x3k^o0N5~p@>IubY_&b&XH%6UX6_qRE8P^rBZe9Y*L3| z)m1?|u*9=T;=a?KO`<}p(JB292`BrdtJ$PRnO?I=sa5USq$>E$vq|&^&nD3yJe%}p zvrO>=vi=c=RgX$$W|MMmsV?p#^r=x~V;NmF_8Ow&t~N{T}WK4~PmkTT<$roF$ViCfVA zWZieB=|rDKA|L}+12|ODYKlcC!ZS^yrlA*821FWDOYK}R)AUWn6wwG;8`f3*^Gws< zo8m-&XyaL@_})duOw*|P^~HPus~DnFkvb|@1_p+OwWa;OyuZWsy z8qm-)i*y3{tLY>vi26b2pngyrQa|F&kCFHhPYHR=GR(b$UG0Bu+ZS4MdBrj;Qp*>z|cW|GqO#bRW2h<(a0i<5g7x-To^9 z-Han$gnC+U|b;)`dR3O&XYNeNI@edBDWs?j-`nI>um(oG!CH2ty+O@CTnJp-gw2OTLKp7S#^O;?X)X*~v7JnKBuwD$t~vsDqEMH-#EdXU&N zO}*a3@@z+l1AG9f+n1SXau3H1_8N#c3}w$W%?M7^MR*_DdeT+%;F+e9`|_)&x;y~w z8`jNC)1%*tV*VGwcSy2J$TLmP&|Q?`U=ze$4E&%TR%WK@<)d(<{2t<21Mp1K$1N-CN6CFivYx4B_4H9Kl4qLA zm8hx7{Gdwtk}5l%X}bJLIZf6A)xwul4@sVBI=H=}CI^6;;!D~OU7l&G|5FK#F9*8L z2e)UMTvL-Zv>(7pA4t+^*)vV6UWtMIABex0qW6qF({%I!tp0P<#~2ESzUmKUq;CN(5>H$!=| zHa{*SvR=9&L{`eAdLp6p`Z^-D(04&I8sUvI^ko#*FST3c%A(Mjz>Bx?VCa9~Ry|SZ zhrsq2cPG}2?p|CJDl0=Xza$%EWcxj3{Tn@igWPD{8c39(6KBG1`4+@K8R(D$K-Buc zQlijkSc>+Kz(77)1R>u@96FI=9Z&$xye3_hf5I0$z$8Mi%uID+qUlQ&h1N<}#Cm)c z23kATDTZiq{zO&r?1O007vUp}ru%-FswyYwH)EvM@7M)3#%~vy%wEWF(oFbTd=c^GdoXhyW=c1MoxID!zuy zOw}}eh);$a%`*|;o{b1;*RNp^j8fgAQY9?*E|vd>_{L8ogg2*{1*w8fsX{kl@1$M` z9%>|6f^<`=WwX*mSp-Nq3&3JuC{JAbWRYXB>&$hm8*PNP-RQbx>T7yAMnw&LO=5yT zwhN$LUBRBdcs$YVBw!NJSb0N%s^$n7AAJy z$Rh%K=dGfDkYfl9B}GKZ{seZEK`(&razP6<)C;(H&(3THE! z=LSwEG0HzKNjO_dSH}9uB56 ziF!2&J6Ly+uGbDlhi5l=e+Xe;8TPS&8W0UBUE| zt|!=!)zuWYJ`NRKmBt$yE?s{Xi4vs&BpFD|m0M1s3Qutw~=s8^>-XdD=n@7xMd&_P0laQx1{TMB=)~SaFJ#v zlSFF1uYG~$yOI9sOMhnW#jil*uKiP-r}NWuK)NYbX_eG+={hnHgML{LL5)pfB1DOw z`#EwI#v`LU!e<*z)vWx#=cc0{^>-u>#~jD|i24af52fO0;EKOv;B5F?9fWv-0jezz zU^yHcE+h+i6XGKVClF^KN>}GzU!BYr>FT)?CsQ6oRi?av z9ny94A->qw9AbY4=+f)6jpS^V`;YTXetIrQH?f`*maZ+!u|U5NAzL_w0l3AzD$Pzt z4fvhE_&K6JixCAV_2Y>j?vPI0QqBUtz(mor2F}Vdc)m?Pm97TMN-0S00=bW9y2bRY0d}l$Sv3pj)Gk81 z!l|iQ1E;^`iXK}_$A3y%OA}uibvpwr*Sl{yv>pvD#@1pvD#@5M1-@_moG0_qyHUDHj3c9cXB8Q|!oJd}mTTIq@?tz>`2qgfM2V=UA!Y8l+|?}C;miUW z`Ub!mUnuTArCU`7Z?My?0ZXj>VTzQ`fxBCKYCxZXZYK3CS5J3M4aEW|;|uAw@{;BH zAXcOUpItQL={m0T`JLBC~yGTh!|H3~iP_55U_763$BHstV^Tmdjl% zOMJNpPvgzdwy_?f^j?-L<1*^$7{bqyrs|1sp*&igQyh*V6Y}cY;!gn28B$k7Lkk>^ z*khG++s@kteGnY_^`#+ldvUIIIO@A<>1$gBK;3K*d9*mcayb4>jul^bM47q(=tqz$ zlTJxNKRF!ZKTQxv37QIEUbd93-9t(fELV$9;eUWjZhZvcbB5>zDx3;}2XePK-QKJd z0DfkOvXk_uh)s+N6wcp+uc52t40sSFoKGE&X6boE2_TlMZGeczVZW`V9o0GD+rq** z({e4knxfv@Rz;xo{|gUkn!gBcWT)dfZZxYiut9`!)0L{^A@m0}KUEHle5BlPC8p8SWXSn@VIvo!_PS?-g%b=e4l2je444%~U zzmMnE5vSvoZ~Sze@OJR}!l8>yrBzj^%k7*s7~Q5`E=Sl0_4Q?`0kj$8)JsuyCp0<9 zd zW_r9>1Z|Cv?&NS-xn%NO(phs%nLYw z4sLN8;}ZhZK+1V4KMgOi7h*z^9t({xc~};39u|VYr|kjDN?v z<-|A3{pd%MD2UF#^6QyQKOS9m3K|XsKG{}I48KVuFdE)ut$ZbqNLzao)l_JAnJ^5a z>lNIdM-K-<}_;05^miVO&7JI|H4Rm%5j ziMDfHxw@fPSLMHni2~ut{lQ?WDdw+aHSSBTvUFmY`h^1JtyxZ6i|oXMldVPmL2sRX zkIGglEj~vQ*Ko*%NLfC_FWXv##=ve8Uu-Q>^QB-FJ)WM-99`0%|6wgsON_ z_a!9EF8}*?p;BdVC%VLF7ZvCZVVD1Td9Y4U0^v1`Ch3>VpeIso`lkW$1!)NFm{e9h zM~ut=^~vbyx*G#GF8|cqhxtE+cwGKzFeTkX5n^2a`wc0K;w(kT8XT%PvMJ?ZT>h6< zz~@us7Z49I(3${t`JWJrBrZX`&44PUarrN~1S6H_&_X(U=b%r;xcp!0g)@%@P@W;m z#T&ZfTJr&VC3OMxU`UNvUQ|cM{ zcE_3?sws$ni>IMGDB)~<)r?s4XeOsv3pOV(qq;FjQ>BZ=AoUMs(4IeJ8 z7+C^fS%THQVO;*zyH9)B!sUP0?s&z>mH^vx%GzDFiFEbqRY~i^p^ar-b-5jw_qjDD zU1zos+EUV0W?ILL+x;(tb^NW+c5!@m`L7*YOY3K#U9@%M^8c(h#uIq^$cNBmH&6Pq zH&w5C{lVqm!b4xWNmQ#h_5~kCxi_q2_ zO`WwW1-ty8%fwTnKZv1*(hF4N^1lH+3ONJeZy8O}9cR1zFAk|7=oug?F%zt=Q;a+vBVygNSqf(XkzM|; z6s|7DQsH+X^Z-TkO53>n*DYUPLq7rdg`vq*skEj{F8{%?g*D^X11|v_UI{4%F8?#; zlo8V?v*OT`h@pBnRpN2^-#omku95}-ni|M<`HvYBruDwi2HU!E`5zZoRO>ULEi$@# z#o8|aCu_uNXcK^44C(giarswM1?=+Q$BAzw0hCwq&pnEvDvDk+?D9|31)gUeT>k$V z5UxNO(Gy$$;LvBQV(4cWT>ks@sHCAv0BZR{`auPk|1+b~b>1BT^d`v8TSIX9kA*1@ zl}kg)2~=O9*aN3ynsP>n4~xRIx)+3@M%4Xw z9%QIxjrJHDw4?n1Xe7~&_5+}?g<3MCVTW*^K!+`Md}V*(NDQPgMB1f+*k=H%Q8jyG z31saO@71W)VJ;yrBD*V;u?&H-zf40NRn!Db^YN8b)Pr)I-ro0y7m)$?7 zblfDqxPN{(xUhbZHMg^`ZOt8s-)zlIf3P(-{lV7Ub^K$* z`^b7d4(lf>nX%^f-;_`6LFf^q$m>e>SaZk36c?8u-Ddd?$!yIXaHoKH!rFb(g!^R# z7;ElJ%e<_)ucJ(5)R~nbhON23+ma+|LaT?vdZhCASaZMIz$rQ)w3kt&1;;c`#F^H& zrE-aJkfyO5V9Rr?KSOhgrI6OJ9Ae80tP?Lqh@FtW@sXuULR)i7M^SOksGbIDthsY8 zhc!2y5nFS2YZW5GkwMOWXu&wFD9Vkkxqphx!j{m{q70;Dl8che*4(PSQaD?4|8hOQ z0y`q`6^_8x+`*mVvHuUkCy}P|Vr%Zo&-03f2!78;*4$|w;>A{Ids+9j=5C)i zQk;Qy!$;SNur>Fq6RYuH-8y3hR(xEkAo#J~J z71rEsOC*TS00uEcr{c-USaaX*SxvkSV6K7m#o~PfB5|E;Yi_HECFuUx2714VkPKya#8X0E*4%!>bEy^dsVIV;-{LQ?wx3i=Nh?WGvs(6- zS1o*V%4gpIb$V>gefoGQ@fo7sr>OM9>z@>=e_v~Ex)0pMvNd;w;w4l9NdtZ2s-#o` zsyzJVRfn^f!rT7`Zd-F-FN6s!IRxA&999)-acs@~vPXCwR_qsh7iur)W`WztP5 zTXT<(!MkifLQZlV`sIoEVr%X`yRmrp2vpD@PZ^|&(}=1@=V+|CsToK&acs?PH9{#X zAS8ohH>XmvHTTwsHF1uupmt%REfH+Z-FyjB9SLz71C)f$o2|J^XXX{l0BmN6Qc;j+ zDztSMe4+&lH4C;Th_69id;zJhl+L+5R_5NUimOl9U~}r6sBO*NaWdZRc|pW8RUJ)U zmyfNvor_QrRY7DKO5e9;-+(3^!$c=4aTn-xNzBGh= z1KLz~Y3L&WpEE>dvaPwtKaIzW8{&@)@Kv>~xtC0bf#jbco-?JAVQcQPcr)16+_Rd6 zDpD33g5`c3y2-}Tet*#`rK|r6reV#!G%-~{vN_1MM0cm_uKK&QZ$RYsd^)wE&_;7= zTF1TtbDw(RFM#$A$7gHqI&=}(n)}166%{1+fjq>iu{C!Cn}#*_!>j@dlD~s|$Z4}R z_r*VBbZWVVViRl}`b^oHd-2Df_=(UelCI88jS8X%`MK)+VCi#j46UWD8*A=A$2qh< z5ZZ8CH`d(Cufv-T#gcQNEwpuG&HajBrq(}!w$;{+HTSmZ<n@_ENB>vU~s5&R}Tf-n!Aq$Z_WgWReS(RPYn@bthrCU(FouF0@21$ zwl(*#)mgd-L!gZ!T~{G3i_@1^x4f*M_;Nn9#jG1^?u&ac7qSt-+exBFr!FB|bN_ai{`%>cLoZ7pRaEBP56!zJRgj zt`tx}=i3H)H)8Cf8f)%O53B2p#zCLvn+{uZuUVd3C$tj4T9c4%&3z!eyiQ~vv_mEl z+nT%FvVXV0$j)n;uw8$7}@uaKQmD5;ryPrgg z_tD>#)uGkp_-xHxc`{ma2ZZ-BaXoFtSaWZ?i3O4g5N9yZS2ZSM&7HEp7X05KeqaE$ z=6-W+J^d)z1L?StB|V~aF52=M`vy#Gm{*gxLH+GZs_Zn0eFJ{|xripiMq&(rLti9Z$PK#6*V*$z9 zek`dwRz*QoAd;R~y7&{K!-VMeTIo~~ea-3G`Em_SwgjaC)stdt?x}}^G%%Qe!RmS$ zYwq{XVdi)m!sk;K>Ty9`O}6H)w5z%<@JG-;XNIbht+_9zq-o|m=$CyMsX(^o&YF=+ zJOJRp`%!16x!Lh z%ZjYIr|JlI6C{vz;~)C)5B+M_JXpA2pRb4W7g*IAU&M?~ zIkdLPTT@om38%whn%m)vpu3P_)2Bs>4tq;&fboGEzIWJHid;=%1F_?FhyA7c;&A6pvG)&yh5dTmY6Dhn&U&Ux-qA`w%cIS*A3rGjy}W5C4Yu*%^tduxJM zwpYV<@aXvmtuE`lH9_6&WtH9;QV*6@w=O7c70|<$b`q8WLQQgmN5_ z9&}&w}5@p2S0m-<8 zm&yo;lNs>+4#08HhR!#i4cn4#=*dv05($t(>i!?|}3N_Bt0PP}>&LZ1)Kv`sp?FeO-Jq=$nZc-(EVhQGz69_MLnMNbDaTT$7?8^x6X*5Y^Wq&eg zV4Nk!L3@++9&|x{KN+N2GJi7YDkboOPX=Yby!X*O(pO#d)1P_SH3^Rj3iA5Q%lYf@ zG>}gzYpB+3N|-VC=y=nh=$!^b5IKUtUBseuUE7Uue8>5u| zZ$dLhDI4d}vf_2%#wg|D9LVlR6yh;TnLV3A+^^b3DZh~KrYwh$bR4QU@-5247^Sr6 zfLEAo3h`wImJq;3DWO-8#1M$%8BnD(Mk)1*W9nr-v{kHA4*FD#QA%-%l5GQUh#|_w z8@l2eQ5ny%9|1gNsQIUueS1+I8KacPjqz*>odTZ-9Qyne?p4HluEa`t;U5E_zAvQe zU6E{*lIt_*r(sC~%9OX#r zdT0vYp?JS_MGtio#2-wli^4`J7eDjYRM2Y}|Krf7PDK@LlrlK3Hayc%@#O%d5VVG_ zEE}ab%6i_SO`x^1bz_tg-Y8PFR5=LR2wOKsDT%epXniiUMXVd6lv8t&wO-nVQA)mJ zr4=K01N@3$b#EA>6l)&CFiPq7T5-k5I{@!<%50P}^I)ph17FAcABR3Oy-W+Elq;f= z&a4!)a-^%wv<{<`iGBrj{QA(EaC|mOxrwh4QyFC+XoGCs7^V0WlKa(+Hl4T)@&@^{ICI_)13`YT6M=b}o% zMk&h*;9UHsVF3V#zOL#GPDUwnH(=Ge5W@ctVP_s#)A9cQnVWE3kr0sxvLHkv5<-YX z>=9y%eXFHvUyD+$wf3c`UG4k6)Lv>|TRWwwrKqi?*4p~9ec#VBcjnxi>hJf*a5L}M z%$b=pXU@#rd7fKXYm%Ng>rl#;L1BW%P?-#%Izj4@(1%hAhUXVFqRO_=I$OFvlyanH zepSrj(7qv^Z#>t|97<`_AWHR~l>j#DOr(PJp_GTs!c{w+1n_G*h|ins_pq14&uneS zt496};2l9Omt*L#hy8w?tYSAofn)H3Lf|^7%jrWY!)MUjgG6&EG;g)xkBlc(^{C3i>-R5+AUFrbLK_A>zQ39_!OARJ2h(=(&WDc9GS{~@?uj&`#% zl=rmP?d2>}IRL5YAblt$As|&9I%xx-lZLEADc7P?ls*jF7^9m*DXSi(2st0|t4QOb zx*tl}j^87Zdm$b(0Jn-cl#&<>+rb+U|8fWT+zI`*8&A>b1D`wT+Xf%#+(~=10dHiVDJ@k?nKrV~OMBUhX{Ifg$X)-G~Ab+YuwnmGT5_$COpr9SA``qat% zgP6c}L-ZF~lB-GAr%ql!sVO!fLdwYi=DI`L2D)};co5nMZI{;7>WW_zuz^mPh9^@v zBBA;}h^~x@DXfj@-J&L@ur{VYMCa|J!bJ3-jG=689DZ3He|=~1f624yznDUE@t?LfZZgls+PEk^?#9~q@N|^e zSR2zM)mR(1qg2}3cq|ml+L*ZivNmo2&9XLb+NBgP;qD^#DT1ve4HaAlV{Kg7TYd51 zH39E`1ZIr2G1bIa8!sM}OO!zlrGO<7T9ImFtc_z5LImaA5Yoq5mQ-5H+PJQF4bcbM z2%}R9R|wX|1Izh~SpXJj$g(!3du&-72gdq}s!0DMz@G```tcQwwQ*=nA~W})Khcb3 zZG7p|c##p4{Qv~6kE}%nvNij^n zMANE5T_pMxmI9Wwad-NYt&QvB>c-lboeKdgm95L7vvQgMv6aoi4GSHoRaUYvbpYGKzMTds!r^W)gBsDG1ia zH}@A;&?f-88Hig;L9jO7|9y1@jRG*)Ks=u`*2cC~FcVz?agzp|drIVYgRG6$jE9}T z5s0)(P?ezKS=Pp9X_Vo2Fj*TXZ}4DJz9AfI1x{jF8(+6*SR3E#ieD_J1w~=}Q;cYO z#4Ky$D!bxTZq=dHA)UtrexZ`JaqZS#Dt%{YJxqFjb(6Jm%?Y7O9}8`wrL*xUtc|}| z=c)8n(AHbJwl;oyteoJxbrjlZOV`%M@5=d@*~y0$j{Hngxx-x*pDOV`%M7w5z&eJr$z zmaeUhJx63v`YLE!w64ZWV{LqXVWfgi0iYFu8V9@3ozm9E^lp<55{0vT)L|5^FM+Bl#~Ehc4G;C(5HHAt&` zafl`QV=o0w0x-uNQhf#1#vhFfRnS%d`_e(&VPI|Sv&K_FmjT>K2We~LZZC?d$(NXh z+lb)2>)P6Q|Fi7M+Bh$?aHAV*W6=&9HA#rCN*eWe>27Uo@0Y3`i`D?T8i?8#yc&CJgX-u`QgcWugEMz+ZQOEe z29;(g_!%Y*lO~PRh8WhyzkKX$@eM%t8Jw++)m>)0WLO*DI-Q{I&1F#c4M|fK=|aNV zc=@X+RfG)R;^-=ZORG@@x!7u3fvk;J21lqa9S5|$!Rh7TgkWu4YO}wBngVERAbK-7 zAy^wX`Yc3s=|KR-7)X^)TN?+=_E9-42C&IM{4U~a8f)X1Gx1%+5fG;|OL52<0G{*iwv06`y;s2UrqnCwQ-YGc2O8WqC2Fff7;sk;ont68h}>rP+#gQ zZq~*G@R9}A#+%$( zH-^Top5_z3Af_esPVNjF8ZUzb1$-#3etz6>6{YYGS&NpS@4@FHf%zU>85#%U`wrG< z1IK7=Xxv#et|U5-tr}o#BdEsF-#$;m_fPG!JVJtF((ajt)9o!fW~utO4{9w$B+Z>h zWRo7R9~Ty0v+;;lyNL-n{UchCR%E2dpeAKSI7z8lAstyqIj}Uy<6w!Jq9REY%eqtg z++%C`nxeUpI92rFo2XiFnJj(gp!t7z1t&Y@d~5H-12zcYa0J^KUo>AhdDaT9VNX0> zR(y-t#iU$;BJZ+1qHiG)T-9;CIo?f>jv=sR@SRn;R1Pw}JsK)J@@h`vzRQwb`~l`~ z&0Sxqxy45o|bUM$mG>(=0S*Wt^hxzIVOq5N-c$0OQoLxwE#kwD$tjT!t&Vv;_H zgyN-rMW)~{yq?U%&0X5f6YbgGaS;i4=^i{VGc^&Ho^tkVRiy{t1<|QXkFp|oviJE*31|eCBz+Pg^iu@E^2%--O) zQeq=g%S0d*bONE|S=qBRisfu%Q%J2y-bR-S%yF`q2%eqYqe@1Rl~R8JX^@qg4wjMG}ojKpl%mD-yT9f*f%12mrf}UQ0}F3d{b2fFV7Rc9HkEU`hJ)Qo^D@J zI9$j-0KP)-I@blJ|Gm6jswWHgz|9DF+d4DRU3=jRm;f|xI0DlJI=ayXn7-R3mFP{? zYGB(JWCKkvYvD$RFg^35SfaPT^TC-PkV7?HtG*i@$n^fg=>5~}ZJOuDu{V(GHQl?K z8=al$v4w+(P7BBLd=lhunx_6@-9*{2m1~#tY?bWn7!K-f(L1Nc;%mXwEWa&)c?899V@VJ*BII-M)#nb zr1bi}@CSzTAJB#x-9fs#a8rBoSGC1-NXv{oCKM@7;2%*aQ#4*3J?;5d;FqS6XfGJ) zPpOsC+x|~tF782>Na5>-Z~9+u5z@$S=%^6!lJZ1Vy;dE}t-1r%O!b8D`LFX4=be@) zLctW)9CrlfxDSNiIbDsoNe^wJDwukj<1WA)-~I4yzXcI@ZF&{)DVTnmdy)wTFQKNw4 z1$u{n^gQfbhNQ?WZx&$}kD`P}4sOxf(q84F4>zgbLNr_+YEmV^YX{pSDO0opNckYlXKznJQxu>Q$27i&H6l{>F$AvKnsg#=*PJZHSKz zz?EyV%-c(ZG?#vL%jOh5XpdY7R=1$4sBY1uNoq+3Z_X$w;sI6CAomLfxnDGS79Pmp zcV0Qg$AG$MkUIv0+%cN0=p4+T@5|g`44|nRg*?0q6~iiSvZ2#oxQQma8=u|4PX{DO*ejTF9a z>4>FSMaWWbHIewDVj&SS-0z={s|W{bn*)JwJ>|yLl=MlMCrR*=#AzK8c}8F88icmm zo+SfbI5E_EB7)+DhbqAV`?6DhT(;K8TDG>zcGzAf8?2oX(VvtKq=bEyR8EBKDN6Y` zD7ujER(R?@Vjq{OhUf-jnUx7E$L(Ll1&Yt0>@}H$^J8_=cf*7LaSpMU5Nuyk_T5?e z**9QLZSfGX@3g{q%sF;5bH#VrN^G_HEyWm%z{yn+)xli*>7 zV!i1l?kpPJB=wMX?>Yw#(l7eOR7()|K>hDuX(`s5o>eLw$G2!1~)G!!k59i z1Ht{e6zfe7v1?j>l)E?oB!$xYb}lGB0oChYX(`s5o@>{%{B(DpGBG2iJ)Jj7%mcOV zUuh}Un;v!7wEVz#4?TeHe9fXQ7QmlHs)otXqavuYP8Uv_X>wzE|eZxQd zXE<0(-yG3&%6xM)mTqEF)?GEx@dtW%)mVv9fpDBEdsM!%2>HdsxAZ2QpON*Dss(~= zI^7;lricq3zE$dG1<@PoU_(r#jCn^mWQ(UgIuwUdDB(;1OAHi>}hR55G)%@%>@;74UJ0fT?pVKe6~w zTulwQ{8k_H>G3IsZKS)DGg`>%NYm_J>G+jkrQ`nPQ}b_3sE{tc6P8g!k!JG0((xO@ zO2>W6XU))PPB*^^j=KJUG)MoHj$as7I__6Kea@EWbn6}!67nh1*w8NS_2c)4m5%$A z&l4YYz09L3;ju!R#DAsZ*NK&m`_sRB5%nAYW{1!js26obE~Xdp&GujL3Zs9seHpS$ zGTQzO0Y`9mthG_XqBSPGzk;^a1~h>Kx2aLP*KzpYKu3mrzHQZUGT?PjW;U9j2r}TxPi-c zXxCc=ca-+TKx zvkdW;h>JWQ44yCj!tOY4#1;^H-6`Phn3Mc;9fyNz18=I6(53+= z;b)nX{Lq$?czajoB&)7R2ss4!IGu#{4>$=w#++n4Z8-Rpzh0e_?8BE8ax?HfItgth za1wrkIY~L%doa(<1Zl1C=;=ymRUBXX;kZJlwHUD9FL6Dtup|Tni$uH~!B)u+jjfL> z41JAvKoFt}X-U>1T_0C?IW$q!M}(9W0n~Jdw3CeA-{Yuc3$3X~PdF~3F8&WI!$ygm zWSm`x(Si0E=qC?7kfU*8H)&5HF#w5IBFLt6tsVGeImw`W93+0&Nk-D!SWf=HExDvW z|HDbfQ;b-7P#kqJy0p5YagworU>-qbC`C{!Y7XNhqubeRT+q5kGEOo&V?V~PPZfWW zX`ReOg^mpTx+uH1`*mGWF%b#4%MqroOsyayhgPZ~GW{-Xam_1=Bd$=+qM>EQcEos7 z5|=d6hZ6G6DP2ZHk4Oq;7aa39V#?ag;fp_-_>B3^a=|eHpN$I+`op;3pg)WYj)a+& zMFb*Z5Nti^BH9JV`dWUX5@Ks4GD2UA$-wsjHEPl5AXX{2G>D3^dwk(uCtW zs-|6V%wF!|f}`Vl6oG3NLyU33vEWrM@gs6MK{-v}>pNX=h;ymp24erxikxZ5<7}SI z;)Fn^UI?5}&bQ==w%HZqL@1=fMqX;kQ*CE2`-^grYP-oQ3*~}iqDQP~u2pCE*Dg4Y zqh=1Ok#WJXvTdqZjSEah}Kfb>1dJ4rS!IQVIxc;kZOEoNgB z@jJpFCV_FmvA8A7FW({FbA!H?al!F!nGm@UT|i6nN4l7C!4cXbUL-(EHoChDj$F5* zMN?><+;o+RaltXP14f_0AjTT%gmo3T;P@&zhnNp-mC?EVjSG$+`MxZs$D{$sh|STMG@py#6u=oFnI zoePfc?~=tw(7G6%QlxXiae7M>>@Pr!GZeQY59ai6ThFqC#>u2PjRTYmj>tVuuNwK0 zQn}aAAE*u-Lw$%pe5pSi!G{RS$i)T6)eBkqctGY&C?0?5V_a~|@%Cn^AW8gHkL7}+ z(rM*`(EOkxSI@ZM_-AxJ5szfODJgw$`R6RPKVQjj-qAP_G(#4p=fFX%aluhLr2uEJ z#XT)&#TjsUj0=u$>%=mC?LTnK1;?2OD2#jo?hS(NSL$)b1;>#**i6m78Q)JJuqJO2 zU|ev}Qzjjx8W$XSH)6Xp1u>0G8v5i+=;ngML4tBdVaLIbmN?8e7%@}b6W*itQJ2+z zIQZS*AkO&V$XO6Y-G!K+OzOu}RO5#uu~0!&^cvKAhIm1Q@xxI!02dNlFwi5Y+ENzk z+QtvZkjDi@K>&#cqFfkC=Z9l;OqfUm)%F7<-!9{a<9%D$+w=oA_TQ-GhhzUK_!L_J zV!fea(C=0Gj316ZKcFEFgE*rp_3&yx94U<>#TzQ|J?PIhBe}77icfwxau2GZpsZUl zf+DEq;Pz4w{BUH=E)*08puB;oOqL&x4*M$z*$`qo18{esyK4F2h|P&F82f`5X(-Od z_~G~quLsKy$CayzOv-h@H|r$UuPR2ZGBo^fWNsYFkh}u&ccN`MaB)6rLVh@6GiFq| zy&>&;XJwr_{BRs>ovP9YLCa4%r#F5$!uR4P7(X0`h81K;RtH(f$U!D=wu>j0J8x6tO9+>9TNR-eQ552EEZ{Ii>ME*qbaAwL}T{R*qHUxfCX zrE5PNUS|p_{S~x#mahG9tQ!`q@(LlX22hM3 zerG7Xlc)X6kz|!gb!c^TCL$yI+34+Q-!j&r(szc|!%DCHa9rzLPNg3UZK9Q4`{DTX z^O}NI1acL$^;UZAhvR?jZh&PV4L|+7HJPiaZ4Bk~?V)y2cO3=2?{#{tW1QH(VXfAwL|w zM6`l(?ZiqALA40ii2kwsa5NZ&_jh>^)eS}O8OsmH$zfQXeGIX^1}r}u@#|xRZ2;oO zX^s20^VVoT9Pv5I3)>yheM49+7HM10Rgb_1@X>M^xkp3 zqW)FI9oON3Ima%Xe?)M3WHSEPFi|9aXUmY~lY%@t*b)Gg*O29hV_CMm+d z^$2yJwI7bpgR>}d7^o>9BDEimm%H<++0|N5yNEOcLb_j7?EiO)B7Xy=>c=xFvq zx@z#V{f_`Y?&sPM$MMYY!;x(_4uT-4y3>5q_~BR}RA;xvK`(C@YD(jW zJGwEFLF0#GKz+Ly2w=Q{cwWU5V+Fwv$Kk1-ViACi?vR>Ib@@A4B(9S7ISKgD%nKmy zBJl12{h@XTR5kc+^X@=EuDuZ=v75c~R8-#!GnW7aPD6jVreWNyMA#gL*WQjLSYRZi zP4*VqQi^)z!`q#0G*e3D=258hl4+?TB{6#$nia9}aZ1TR?vthW*2pJP%J?;ek}3Rb zulypVobPdbe=-aaQxI(36ETd@CrON8bIz1_m2UtisN|cv6Q?7#K;EHKh(LPzrc||8 zYEw;6LU|7VTq3zQG8BOu@T+SnHS7y-SL4gbzajl&<|HEBdr;3)F@P}WCk+z#*0Fh%KCcy}{VpLk=L!{18 z+AfTRkfcRfh6uJdCLba%e@qn>sXnqXh)>)p<3r@pFL5BgggRUkYRYbWhz#unD~nkW zmuf&(r3)D!A~{w?Vsj4SVFS4P5cztFhbqZ+0QU`~a?w6SevZSdQSQh7CxWW1x{*nq z-an&VEFU5_&Q=z+sA5GymL$49mB;uHDYVu@&6(1mH8i^MAyRx_Wu<=#t&gQ^A0j(e zWK*6fzJWH)=*EXg%`|+MwHonTNu%2$shuq!BD1^TeR3S)6$7*+%H4-Zz7~P(L*zLi zall!NbUs9=n(2Iq`0NU(D2grcv@JXgqxv#Pn1bNhs%a15(=xZSJYy$!UFbo|mc+d9`K;lUeNezHU3HgPJ_h zSCjiwh`jKHgUJK_*wDWT>Y*lk_S0moszmx;3uJQjBMc|r2l2!saE;~<(BwrX*TY{D zl{2s;nLUH5q{(>$HQA;b<@qQ$Ht?bwsIHpqHCU4mnVhzuGLz2&ad7u*P}4P8 zV~8fVq*9)BWEhiPc1$+5fZD6c9}QWdI*~_;YgSm57l`dV{)GiWpeFU zj8%?9*mXhRx*s&;9s9@ z$-i$!VA%uejwWkN*5vWJMD~25`qGi*`06mjVNClGxb7#XX|iKIBCo=vmg+wD7RI8Y zpb|A1KTngN)F*O7L_sDi1$bjO9#mURmYA=}2TWF!d6?XD9Pf@{peAT?QGQMKEkSv1 zx?PybRTba@WG$#|np{;-lWB28)}NyC+=3r7mKQ+%uF3FlOg@$&s}+*%2Ki~JMhP=(>M*e$VC*f~pzFuCajULearZP4Tt ze@zaJA+r7bN=*7!#-ZTTpe}22kFO>#MT0bRAJ_R3&5)8_ZB0@xuudLxae*~$23j=m zC?>WDwmtsH!nnZdltM>bN7{W2m2K!a2LqF0gWU#OBNw5Qk|% z9;J(@X%t*wb!eSaoJNe4^FXXL6kmw9Bgq9;lLN7eIso9LfjBGU0?Rvtr=Y6JKL9)= z=saD@xWKv+7po>7nWzO2oVs>_buD8>HF1f67Gred0xR$@jCsk3uTC1bm2rU;@@0Ok ze-YnBYm#bZxxmWzbxH9Rw24MzAl)(pFV5Of9KSN(gn1oK6t#}8o z!)GzkrB=WR@A}?hqWECn@26t?e?!+jj_@hn!A>jn&e3wldqZl6_Yq3>#W0LvI5j5P4Xm;$&oxY^TQ;L z$l{S00e%K`Rg*l4W73x=TpKZoBl3seu)X>Ql>MZ$?mUTOat@PEFo`3w;0jE2!$HMr zk|%LY7UHSd3{2vPtQ87}hjl=Gq)DE{G5L_mdYHr!Iq4`4fb|14LX$j+WAaO$njOO= zj>yVy@rqso>N`#HB#uc>o|+xNB#y|?IT$BTg8EgHJc(oS7?Yzhi6ip6Ss2t`f$})z ztUFKQm~6z8xXzfw5qYx^yu;-O6|G60#4(wPhyPoc#1XkJ8pnKVf@-Wup2RV^gh@|K z;)tw&22-BTK@HI)PvV&TfhTd^n8Xp86~E*v7l2x=NuI2 zizayz$K*mLuVE5Lq+#B9=J4Kn>I+PvV#y9Y=XK#3YW$4%sDs z_6^i>P4Xm;NxzarPQ)aR$ayxLr#cMkj3#*!$K)0!Q!$An^1w;t`54q|P4Xm;$)`Mt z3&A9g$U-Avr;_^&%-#^V?mUTO@&-@hZetQh0b> zKf8naQjDKt#EzC3phFMgT|xVXOD)72C2-IvJWE#-2TCKSf?<$}3IAm%`_Py`kw zy-%G4Twiwh8LcWyL9J+rr<8|!xx@A4G7p=0g%~NDgXmx=&PP#jefc(_qLUg5V!WX^ zFXQ?$;~ymjRY@)fu$G{AbgkOuU)2y&R&FSETdrJrRk?Xt4n}W3)&T zlk`C(eAzWO8!hfVk8-?kU6PF!$0LC@TBNM`^c&^O=NdVoHd@@h09{QkMuL@8nl@B- zZM69Q6*|{$L?6+Tq)*0Zu{-`SMvHM*qeXYRCyx;OTqiI_i|v0Y!`Vn56f_5dtv6jt z8!g(`77@p&C88mfAel>|n2QcRw`#aW>a{;bJU^7H&5WXH)Pfmz(y%}uI$4%y-`w;zpH%3O7^WDCH**eE7 zd!3rTgI`t?e<9DORETwSN0pN1yZZZrA|rTT1Xd(f#+dWHh^vN)5VTmM@3HjyzK@&Z z=}LiC*XW0=?3>z$p3NoNLF!}V^p-q5?b}L};RgN&j66fBwK-qbo!;EGD@oz&hF8GB zNSMJ58(c^nM(h~`HrL>WqME5TGR8Z~3{ExDA=KXg!VJv)$qcU4uNWRE%Eb(>%tkLR zO74re-~}p*F@yUg9zX4Xn35!2CCQkSj|aQN#C66yg#?6aNI-dNB2IEpDo8exnx<6d9(@_&5_q9 z2)1Wb7Hv5;6Mo-5N9+)-$oC{`%dr7l@LW%aw$SJq=y6f{5Z_!d^_5$p?K8R;=_>nC z_M=71iC-YyHgfv@w!kj$RujHR^d5}NNvXBv*v9xMLB+|1sz|=BwjAS8F#M;pFa{j% zH*>m|s0^ew0=E|B#?_ScDZS+wr*)LU)uQKSOlHe5KRop@)OsR<;w{UuGC3ueZ8nu4 z*ecs$d+NC?T(&Kwgpr~x$0#2MMQh8kTyQNJ3F5kyiPLf{VoC{76w3chCfaiBt2ZU_ z{WQiu1Y1eUUR#b`Y*to8Ahxtt_>MV;38&@Qsc_8mY5{0sATFY@9P?{dMNoBmLLcDH zXv?v__dLWzNVAPxfr?=)$MSRx#_>#uI}N}M)cwgz8(BkbUl1$)fbrax!BLRb&Fz1#&3V@rI~L8S@qZSwr1k;v-s6!sP%q8Ymo}ENiH# z2MQ?YIHE5YXgJDbSwsDijE$NH5MLR9G8R5%Ho*=S)=Bm1W)w0n(iB6$J2{qnAwCor zJhF26sdKA#^V#>pQ-n61&!540||r+<|> z>vALbH@Jc)ovehuHyW&jMxY4RWF#Rj9KZbzE1|IKmX#2_3N0(4sD{{vx{EBHBG|sB z8cgS!Yb&7&WijGp`VH4dU_ecz_%4u@(51!I#2iW&4WP7!R4xjFl~B&=H5612Knnx8 zTM13BQcx^FF=Sr=0|{D0MbTD5G=Z_KgsNXE$c$V9aFv0LmC&v&Di3)O+EJrxD)q?q~cqGUWgy4HA#(NSqY5~D~b~U5a$_y8s5c9=-1rX#NGmR zuOYS*VOa_NSSgRX5tl*Sai;=){w{|I{$G~+r|{%?qRV7O_=tQ%a3AN+!FSB`FlA4} zLx+Rc9@$Ae(RW{z}4D6PeXRuz3ccL`wV0X79MJny__5 z{AXH|SqV@ZeZy?_1rqOG#Ms6I_|_frsYX{AX|vxvS^?8Dh}$*b?4@w)D~gY`*+=fk zsn!Z-0sN{Vk&*QYHhcR)6+{ZkEdPe~59xWR6s2~Rju6FH+d^ouAcAfU#4h1{o2|xQ zxkSMm*!n~88kVPq`lgps$Yt0hRGYLxlTKkp4tnHQF`_=mHk#&*LZ)M}QRt)x`C&H( zbnm7%F$Cl!P4i|U(=YLjuaiFbraIHD2E>T(KSes0n4LK2w%Ye!y@05Vn6ym`p< zSJ*ss=3QnVR@KOR#6nNe0c2lI^9CZ*Uf33N(tYz`RSo*f>HcCG$fcU*O+==Dw&=)R z!A!6IIzk)(c~;ZBk;rrlY$Q7K{``4nRqh{6m6VX4_f zD9ECk<_$%rzr%*2Gw%UyJeclL-X>~+Y@uo1RAf34n~F|4;tlp7k#|(SV9_7s*P7;y zMW+A4x5G|){U559LvU(hCCKfX=FLT>M`Lr*N#B2_+WSn)!r}tRyPD<=MyBUsyTeH* ze(uX9DYPI-WVnSl5CY$M-ehF@PkdACq`iK~#q^>o*+mJEl{L*9jZBZgMx&DsYO6|; zeRNjQ8e~sR^JXK{gR$A@q*t8D&Ut^8J4$>5a=xZ{!;$IC4GjI$4|vMaf;oo?;hd|$QzpG%}Ay8l8H}zBfyU3zE;yuW$ zcbv844M~+Z9YEu+pMs`N^x~3q9al_5gDk6Q-jrmzH#Q}mbPK7&F zlIbvPOgiZe3F^80r*Bp<2INdl^X4Sekua#it+#YnrzUnf@8uh0eUYwo-S;`*&);+c3?A$?f2KnCfbi+j!(?Om1&osvsVL z4f+!=i~e-27<{r!ZYduJi61t(eP1V#lYi`%T+*NaVRGB|Zh0}m&~#~aMPqW?&nK6l zGOQ$M2sMW>xeYp$p9^}>NXF!L-JiKFlUpiu|>1lm@&d9p;@uL>g@sX~HoMd1+f+$66O# z-5h^mdxL8hLyWQ2jp<_(pCOkoDW~;(eW$H%Xp!tTR>_<@^2&?TU~yPDBjrW{v2RuZ~`KyngqsH zH>iFH>^~5{o;1FevDGbvRp22+pVpH6lP+d#b-ULK5x1c|HagpwD!m9CGo+Ld86IGK zK;W#|NL6KGY;}{L;^=P_h*E}nWL*Wey4jD^5;dSTGCH@vvDLkN&Qo-O*5Bxq-nFQ( z)ooX_u$Ty7fq|$hQe8>g>iP~SDYgLEt08r>jIC}8Zn0&nJ8(@FLC?os&`))Wbhf%r zPFEFnJa#z|R0~jwbhf&Ov7c#+22t8j+>ZPLprONhmK`*}CUs1RRJOXy{^zvAJ&EhH zK`s@9KTsX`1N9;PP*8v9gAWmuk&CTv){ttv{N9Qpq_fo>*hu~S(ixK2JixNm?Oe!* zlNnoGuAZ^g?SH$7xQS$WA0a7yaQWvswZFTqEu)aix@B*NJ0-pr0KcGf`LX#mPXUEA2|j$Y<3&H%V$Aj*ZIbhf%v zw^k9aL1lZQ8^R)$bxD&?GK!$)Hv5@oC|fwNQvXIRTV4Mt5%~TKL~}!}M!#3(Gq$?7 zZ=h{@ff%SM_3&z2-PdRGiV!OCbm$8;Be}77w}fnU-!`wQpdA1X8;ILWL9o@mc?bTn zD5vWH?i+~8WZCLYIgM|f{Lav@)%~GZK8ECIkQ0b5Ko>W*x|{v6IfH1q0{^Tdou2}w!&bN9 z$`UI5PtZ=1&gqS&Z#cm~ zPd zUQ?E>Znpf`^rIpyhqjh~LIWqlr?T^q-7+u@y{uCYz|5u29K$2BL zW2@`WU02)cMg>JPDgB@0b%5ZSMBD1-vy#A8H>OEdCS^6?^(hGz(Af^!R<~0Q_^<5* zv9~**ZFTdDYU;YQ<&9N?pa+>Rwt{NRh#yqTNZpoUzsI73Cc4 zs)B0lPVyjWY;{L{k*f0S32LM}$vvDO3S+BVbqtPSQppwoUG0Wjwz?iOvEN3}AppO) zL6WLv+3GHxi*sQ2LA)>&y&WuD-FmxH&=uTiqXGBY1SM-3D-9Lzb=XipkjNKs(to{Db*Ff_m<_N2vR(ZFQ?< z#YsjYi-IcuAyV7wuKNbffKu3+g6c%184%Lh>Q=etugI~WREzQi&e-bq@O5UjjDV7S zzqGBcTrP!eC*pskOYm?(Q&(fFySODbK&imjq2D(QMHyS&zN^Y9#uKl-90)2OVkCpc zR<~ru%pwv%f`NFpw1dz~%D1 zJMf{UMIBT!wdvlBBJ~NL^wd3x#YJjbKnans?;cK9Vb7_~YpfeCZgAamDp5Nc2LW-J zhX}TX)Q)=3Y2uq$f!_eYTor*OxrTJT=Vbq}y4Z*aDI)-saEJ7sQ}>}5l#;ck_nh|N zV*br)V>^oWoJzfQ*>mcR9L=87S0nR_Hb^`XL2jdKEy5>j&x!JJkoe&}rwa`tIQcoZ zj>o zM?QR8jdz7k&5jdaBLVL@5ytnuKVrORQ|_wP)F_hwj_lPbj1!LUnJ5naE8%8JNZGh1 z{Hz~sR<&AAzkS%nF2*nv&>KM*=$?UA7V*+Sb_6Kx=M zH}X46o@i^HshSuHX}pm=1F7zu`BdAcw_ai)qz!Je06s zBeQLl5S~qB#s!`u@-?N(MprZ2R>PLX3o3n%_eg_a%R{o+w&Ev|;?1_ztvVGLtb)iI zCV|CD}cTlO1EwGmbzQ+4I}L$sDZRc`DYRIe9QyAQm06_ zZPl?5e$5KnNuyJWblX-3vcZGMEf9|kWo=v0Ai~>L+~XZIMkgI!n?Y?`{o?IVTT&yC zlJ};lApC*qz#pg&@rNt=LlHD|1ZCv1ZT0StqC5@hhA|@DwpGrr{8$=IQo3!cc}4TH zX11-kdS=@yFzs}*1f ziu4SjMZbl$u7M~QhSF_Y zt?ub3HiG)`1El-5Rf|WM4_yWJ;NPgTeBEvKq?aC;|9CotAA+jlCG>k$KC^9=Rvh2V zMS_Ual>4^Tv$vk&BPww%=uI@E8p~{3&B_p_pk4q58i?D=Y+KE$Xjjk-0E-MnWwN%d zX7wqJmp849T@1H<{ebw$%lT#~ z+p6XCc$Gd6w0xv<-HeO5BAaunvX_BY(bDy{)x^%eO8*F2YfIPLR#hLCR{8gbHpJ5P zw$;ib1(iM<+Cod$+g689*5H9a?tr$>()G4gi3fJ2Uxs$W=w{n$Z`lGuzNC1&v+k0* zkF{+zXh?ooPC$%s1ElI-dfVzr2$l}zKqPC*+P0eeLpfE1meAUgu9{H0n9C1SD&Dpl z1Z}v{^|sZ+sgdyikLV>N(fFxKXtu39x$Ej}t05sJn3TtXUr3ikZ(C)vl3?2^bbtqw z@;&g(nVbc5wu9cbD)#~l)VvU*+yT99HEnW|x^6P`G-9lx>TRpJ56h?vb%g$zdp>5{ z%J)bamC<+rQ*=hww$;R&IaMZWp>5KcST5#*R~J?3Pe40srPnUzrr*w`(m#Oqx0POR zTlF~ej;Zg=7lmES_#tmcA9OgCwSUrD&iaHw9dYwx2;BW z!HCrvVjlx^<-tmCTh-o&m)_S9CuzWJTYW#Kni@)0LHa?&iZL4#a)Zk}pK|n9QUwYf>n@4$t?K8v=qD$~_LGw*(+p0l1Rp1%W7a4|{(rjBj z49cRI9ng=tF_J;EZPopGg18Rgv4MDA<-ToItz#LH0rLxg1XUu2)NE?jcjb87D(q4T zVM_#29f7wm=nu7hp%U@k=IslgkMUD&PNh1~*10qCZ@AAY4*fI}iao8ss4l zb(wsz7rqe*_UqL2$G=ir337bFxR~>H*IcN||y>&FZNL{=xn@AevS5PhL zTIX@D+>-h^KYXk|#OMzj;eUY_byM_*oA?m%;B!>drksPXk@ixT9Q+EelogfW9%cN7 zcPfcB2m0I>bwg1w z2eB%HBRI{{v>0dbz&$Nz#TmFS>ZxrU1M~v`wRzvosqMAir%m|z`kU62|H^e|9%%X1IK^RYz zhFHx2l!dysS=4>`0zCjgF9T6745eGt%`8z+i~=?11El++uF!L=NH+r8|8LY<)P3Fz zEqf8fABNhE&aKL47IkBX!#MvP2uF5jL%1*M?j1=L!>Gjhp+{>*HI`Y_{hc95K~(|N zGZ43zS=3c-(v{Ck%~6ozK4V49OgR4v`zdHij;47InS9j#0VA zK}#f^p8~U}8-1^^O5XrlGn3vd>bCaBO)!hP%ApF9LqU!0CCmsC(SSTcv*t?FH#vH?yewvq64U_G~#EA^^du>qT9+ z<94MNgI3DY^`h>JTQMsCI?x(fx?a>3S{$hK9?(9wbiJsXP`fe@1acy@>6We+bt6BD zQu=ymTa9iObxqg73hy}LFOWucm(+c%MctG}*y4Wx@vR#mRsYh9x~C)H%f`>&A#x!& zn{Y&q6cKdO!@kfnP8Fd9v;@*s6Y53X+waveGg%*6Q={ueozFTrPUw#4FG;dWXcl!- zxa;agUGqEDn3M~FuSu6gFY4x7NwBDUIyQhwc@FqBN@sCrQ+J%ZGQ%0jQ^o{w47b^a|@W%LPvjyfZ2QP*6Tb||#bIumPA zH}+11O1}u&3M;){)Mbv0Q-#f`AnTrV-g7`?%xDixe(TlpCS!!Yi1hKjSrt)B=7j?14@umC65Zi0OEb30q z2~|VMU`XF+S<(=tdZby@jT{%D$d#bByOVr5v#8sZEsrA4g1YHW@*rszbwwv6DDo|+ z?18#^e06KkHH*5q_rVG;0yN$Yw-$9X?AVQ`;@1Vx+704rS&O=VJTS5C3u36D=sjaC z>gsLF3r9E*=V`!N)OlBgujkE(Kd3eC-_FNaFY2Du$6PTgU>sEdh-6SjOHN*Ri)Xtj|!i%<}To3@(38XJ_yAG4^7D_@=`g|?mm`fA8p z)NO}zSAMBYg0@iWsz<2%tQU1HzF&G#cMvYoZPgH8A3;4SG=VgWx&z%Zr~-F|-q$cx zTeGO^_pP^LCPH7}#z+RuqAs;)d9ek+Ap`Ne%6(Bc__u805`cT|keW@!2G-{FnP(2{ zZ)L@bAQXWYb@Ye&a#&S^?>2ug98mB+4ls1H|2D;r-Em;GbsGA^H4Wpu7NglG`wger zuW($!IwdLpk(_EJGPIC$mC^k(=ZDGq!^ir=t)5QlkpAG`3oDT_J&>KX67hYamK4WP zkaR1N&!WS*)7>R0-Ad$K?s(SBO5_EywN@fiX2pqrkgQY?V(Ej+KkeXQDcwqBGfLJ;EjQ(Ox7Z#_91*-mmL%H{0J^Rf$plc2-z?cZVXF2|;@ zE3gRp?RZTlWf$N*brNe4vgVMXu?U&|XDCB*Cdm0je?u2Hi;!1AaVod%(DsjTltZaK?74ccr=*Nc$6>u{79Rgznv?XYyc z2X=)4z`vMKPkl!OZCYzMsv>GA`%f%`!m;ST6UNSTbmowuO!AQ zs$PU7+^MN9v>*Bj_k7GEq&^N}P*vq!01tFV)*|Gmsuffw_Aqp91ZQ*V1JP?|`K$DW zphc6edbD1Ic>hserLPLDmX%&FLb~Qo5%+14(E(aFE4^NXbc=};zr&-M90hH>NpBV* zWqomgVKL&@>9o$iq8A}`PoQ!KA)Ye8G#;$E?motly!MCd_vI(e;?j#SAW)U*#g;3;BP&3>~e*LN< zOzKrt1Wn0c&ygIj@C`tBy5T~tRPf_PMK^mW=sbWsZjhvES&NY1csyfoK-dfFdeeKx zT7*0d!<#7>VweW3Maa{_WrQsO@u^x<69)Ad>qW?m>&e2_3Pfi&%36es42RE@uOLn| z053vRHhK{Gw&K-KvWsTUzWai!J7sxzp5M4AC1-6G_8P##6j z2Bqr9GbytOxzRpK0pAlao9~xigv@G)A>{<(FVZD=xS(09S%h?d5vB_K1o|7pP;Jd3 zX=oA?v81VY^#TC4-ldIGwj}iz3N-Y>mC*21brD-J(#2e8D{tae~;+Q z;8{|0FO~FQhTUJ_)Kw_7!djPSDZTP&7J4ZA)D%)L9917iZfB?nJlJw}3MeLh6INrra2MiJ1Kg(TR3^PaJDCyE{`b8EW{N^jG|X9if}KpK3L*g*(ks|tAUv!aX30bxv=|TDYQDM=nOmtg3wPCal zZk80?9G|0hU(F?&mGiA%xR7WHa3I1k1Y3WqPE=H>yrNmkRM1{BMErBCTDK9QY%OzU@jIWY0RgG-E#l zy{K^!Mh6(v+DGO#Eq;)F!-FcgcB757k3HDFFD#=7!P@PRiVdIJH9KWEECMVmsbA@FV?>>9?>ez2=d%XNMw5?HXW6Tb%*bhvLOE@vrA1}hJbUAwMY*(pKzV4X zw4_X1WcNE7%xcD>4v_^xwL}yxDAShN-?u8lITwIZ*ivaZnYPM)@K!jhm7r8JY8dSo zrFAGtGr1^QSf+hvKff$D6J0^|GGr7jE7LaG%ZH_~Iu^=AOQjWE+7|n;Z3#SOTLERA zQNw5kkk%$HitZ|{*3!1y!#Bq>@iVBaL~{3{l8G={tEIJ#qrRr_UG_R_D=;RDq3Q@+ zb;7xuXebkO1|o_USZUYnb$=7n(rjRE2v{&tkm{3Yg{L+f??p@-JjUshx+8fXcPo1yiNA&1&qUJ7TTF{l;;C@~qB8}8d~3mlw{&ZP$vwjht{W`__BH% z%6&_94z2NHBY9}`D2YyhplYXw)|YfwokOdQ$k&HD5 z+CgV58P`K=tpTx|<6y{R-E-7K>)MP#oXaA}Yu$6vLu+*FaL(lrGXV5T^}=Qjk=>^Ll9IXTv$PR_~RaiBwQ&CLcYt z^0VO_TE}&+#l&ZzzBGCMduWA08Vx1>J6qtiQ?vH)e+*a+tsOGte5?l_`XKyk!{;QA zzgG-|@i!e56jn+NV=`Xngt2a9Rp=<#>Zuo67}W;1eN+j<8Q}xRhcLcstS=*Vez5yw`4#mCOU!o)a0qh-_qOia6Y4;j5qn{@weqz z!D*L5S#70tj=zVKW4N>jpd7JO=lJ`)r9Z1Tpxm`o=lHvDNF3+<9!iEpUDGJ%_*-Xk zC02u>gc_2sjunrH}C&yGIj;PUBamoNXGT}d(U2+b9@gua~ZcB_4s?> z+aS&*KjdgatX%Z?8|7V+bEyKkhR(%|zm=B6aRzOmbkZ4^@i#oGAg3JyWt5fHIsW#$ zSdr7thqBb9)#EQe8_x0f=LbcYI0EXF$w!aB{A@VK-(RkmWa4j7|Cl`gJ^o@{O+$%) zfoJ$Ol*Zq9Jo0AzouCIF`XKz*SH@``p8pyM<8L}>{^_i07#m_BjKAxuImh4ip?G%u z6JEoO0*!UcTp{~U&H-^ALtoyksGwu6P}{vu=oUSN{Aw1uIL%VHJe zUsl%+Wy-$W34K+G#@}n0zH^xpU7D3L5ST z?ZJAVZ@?kHJQ&G4p3lSkR9~0FKGoO?-v6~vMJo|&pDO4iw%dL}wr3G++47^W$KxMi z_Nk^%!cn$AAwDy}lU6uwZ}zEv$XHW^BZ)Vz8;GDXNvenfn0=~>;gy+*g`Q-XCd8N> z(6mm8qAs-NMyFiWK2@{t;zc*af1$Mx?Nd?3goBzWA-E2X*TJDjzVc{`AGfFLIQdeyZYjxp8E@=s-qjoMr5P2{27Ui^X2#_a|DVccHO zAI9xPsU20sP(+MIuyv)9X}1@Z7FQB;5W8F}vMl+S7q=JtD>_6B*DQt@#6 zQ?w$dTJkvCk^(hFGf3@>Jj;?N+FEC;DZYd>+{g+F9c2MoDHD)K=UQmsU+lzXuN{eN9oc}Ay=G^Y-4t;=5I5u2bLaMM*L#_dJwBDlHy6~rw={bXGQZZG!E2o(Q7^F;UM z0!nUwTbsE#qp#VLC?oD&KO~E+wU`oNl12*lF|p4 ze@;^SyIV-nbKoG>oJ(xq$%8YhWl#OhLx+1=>*5ow;%(+B*%A|u-b1w0cz`SAsV%D29^vRjL=~s4MDoiPR0GgE(_8aR>b@O!gFq2u9!<$oo`Eb1u<$2W;nM9MtlLctV6Zmze8n zRUsQfY-a$#1f*{hHqI3`i zrAbGcf*>G8Kv3zZbWoZ!Q4m2uX}@#t?A*7zU;OHu#7_At8q!N3ye1@WFWHwk8QWyp z(t9Ax5E7Ni*_YVvF4pW;0Q^b_+!N@jI{Om4wZyYO0>&9(u^HKy*aU9|=Wx`|`SLSU zhooVQAeb&IH(pHIE^O>e>@cS!lXPh?D-!!LWiI;?m+wt7)|vxrBi3YJ;vdZ_oAj>( zdsEWOzQpX!@et&2)Q5)~FiFn`bBS0hL`z5Ui*)&ThmC!S*PAqCl0FFLPhwm4C00A{ zGS>b8c1NtqzQlRaNhW=CI#d~gshb>*n)7aMQ}!faWgXn^OT7PmxWO9(Yw6&2Ut-j^ zA;x|mVErAOer?B33~?9o#u|JQu=gF@?n@jOT9fAhy&BjT4sQ1)e$}kF!4CraNpRVh zICla>@pZ)CC5$E}P2I=Ymsr1KF=*`uc*P?)1a114-Iv(qDSlR^fKkD+oPCMY{-|M! z&)L&ZGn(dOrrr+v z-k>CQUt*k-1p5-Vugc3zeH-*gl!OZCYX`e8@zwj4aS*v7=05~Kg58&RV^Xxqw*v5* z#Bqvh_a%Bgl}twMfp-hEBl{A^j*2vfMu70PHRS9|ybzY%7?}@ji8bOJj+&oZ$E5!n z*d8Z6c@N@oQR_WVm`Q&Q*kva@c>>}mg}P4`sHFIZ;!|K@jeHeRa^pqame}bhN+b731d6JZUiH**N z8D=i9N(C~VNmmX>RlFB&=;@%g3ZOgt65R>K4e3=7h6j+iTF$=2u~43R1{m{&MQ;aZ zU*gwQ(y{*;;1)}8_9aH1PQ(5`#9y!&_ix{$vHKD~&sbEs?h~VlZ3Z(N*c&f?U5&de zasVtU1m2fuW&^t~abGHK@~j3%OJUJ_$3I8AFR@ku?1X$7v^RvxIm+RvS!Ik=bWH|f zh9x=s5+|#|+}lVDwY7SrxaY!en`gq5o)4#6Qf<2`s^ zV&|-eSr#l)KUPw*FEM3oiXk*5LJaqFyDxD@CmdMohWNge1y2_wHDzC7?xDT{zXyDV zaHuI|U*Z>;E1T-C2EHwTqnRlC5_69TQ9pukQAn(-0{126x)`VKgW!h7FePG=voDd9 zAs<}gpTwRg(EX|ZWU9fJNMy#F*IpvQGlacFq9iGn1X88-%IDzzQfWuS{Hk7+EcJjy zB>vQ|fT~w5He2}$>J3DUrOa!isWknii6fUtuw8qJMASiSfnExDl}#a+NYvY0OHo3- z2mkCRIGx%)2W^O{*U-JDVQJ1r{|e|gf$b#{6MrdZ;1E>FgJ4qAB@)yC%0u%)dbypI zRC@h|aQz${nUqYHl9qQet2+0egDU?xOx?V~J@P#XuV>JVmEb><#TP_6L) z&cZ*=eBY|hS^USY>XiB`j%i~+eb3TmRp+8GN4tOB*_aut!TN%j*T|-<>g2l+&V~*E z`cVwYs!r~ECiM+KwM~jdNKFUzt?C@f zmBLD|9gxm~+EtyK^i+MTIuGaMXT}(?CKHpp7nMv_b*Qh|Rh{LHaYhxibqL#tPR*v7 zZdY|~wM}M=#{i!vieu5P>MSXohpjvS9D+{oXT`4S{M9cfTgeSLKT*hvtm-VkRgn#( z0;z5d$f{0(PeM6u8z7yWw7ylHD(^*e+5td@NLst9!(+p@s?&Q?eP+xDYl+ygt2#V3 ze5*QFmTG3~1?#Zbrk8fxvW?b@o-&|M5pLAxH zkF=6WKc2_uX=tu%<4;;O(?@!rNJ*y#W10K;fl<@JiJW$>ifWL-H_cJ$BDeh1Sn(gB zHEO7WnD`>EeGsY447lh6Y^_4lgEnc*e<4m4MzQDJn z`{!2wz?}Z7Q1Ilr12J~;+#Z&o27sCu$(d><&oU@+=FYy|p4gHU`DGk$Ny^v3nLNq5 zOrFmbs;9cbMsEby5z6v7{*lR(?0E>Yljn1@;cUYH=Lr8S^L>-&Mf}H3p2d<}OxpqK zx0WuG=QUxDcF&umnDHA}e-iU1*_6ri{;5;P*&F}ZtDzA|}IU$c{Eh2u#~y9L6(M5o87nQkY~_f{8Wi<#Tw_b-B< zMLT&u|5p{ZQXX&(qBvIUZ2>=Tt;poLbdD*;Kp;b{0hv5|RKcCsNIL__ z$4*+`>*7t3Le|CiJ@G9H(;q6SVK_DlVFua&j+Jk%$7JRi zXb9r;g2KTw&@#l>8E9XLWS)T*Sn4xoAZ*I{X=fa7%30XinSsc<%s>zB=TnyuaTCF{ zk+R%|e*$M9!t4wb@@}-s(h(9L!SKIfzHbKFi~rae=$Aa@m{tqaG)tEm=&&$HyRSur zGNU(G{fK#tY|0EY9JirTeiIO;5lrQ@@Obx}wGBD-3P7I|%&DI>1C`r?-(g@JML11N zHuS6+=<2(9nehay&`#E-oq;yBYs_}yffOK=?bsRU$w)V+tqi1wlh!u_ZNV8~YK^u) zo_A2+4Ak~qDMkkZ8S0?E8K}e+Q|=i+K6X%_d~3Lpd3IF1hq5J8G5R|8OXib#f(~DH6|u^FDjYLK-Aal z43yqJ8`E9~b*MG0nQmvGFGRVwz3iMu0Shx26E3xU@NBqpSMX{viJcx@{=XADop6Fbc}6;Y8OmO?0x;QBZb zW7MA5slT=aqm6;I6qNJo$ePJ_VkcC2s6=rezVF+6GYw?UM1vkuWGK z@&(L{g(AwRgu$`Rn&Ny9ic%KAwUn$IWLRwEkQ}N$Vp~{5FH1uO8gRmqv912{;`}$T zeuA%}}ct=u3g$!`38?OGcF{5|I?45smP%C#Ep_bq zJ=s-XL=Qx8?IWlyLv3_y(GRMs35fpCLZ%JtFy2|mo=HU`t^&AG2yAy{DAPya=xdbP z_LY=!UPiK^Dw^}PeT54tt>EIm!f~ETu#z;WVyeoNusLc~H~dJTR#8PMJ|e_e{>FWF zXgXW&OO|gsd4B62xU(uF;|aM-NWqa_yav85MRq(CT_pC)`YK_s`}rfK)xThDaEvf= zz#WZK!m3ETIxI#Kx#Jv)eKjmiT|(@21XnmH>QzR5j2(~TwT}@SjWRRBPr^5JnR7Pw zgD=ySE&xK3khqBEplZVTtV@c8DymLB;LQU$rUaMV|DLa zx!wUdO$gjTPHUS!N6l?Jo)_g#I#wB*0MYS2@9uQGYDJr%`zJTDWMdoeCt1&HJ4Mgk&i|~_pM(+#1fb2d2Yo#!`oYMPR zJxo$moZY|=2Xbuh8*;Lox&-K^z-d$#>3ugpz)g9fXuXUGTuxI#r^!w4qc)HB?}8i9 zph>&r+3G)HkVDDI+FY zz5M@)iH<|+keGNI7sEY(v3QURAh^yHg$*lq%FJoVXk{QZ1pQwzfjt3KqNJ&*ChNoC z{`bk~)8#Ai$>>kxRDTd>BFrV@7yPr{{AXE4Hvrip=>KNzR-Sf%lm4t@vTpFo|Ce=z zzM8VmQZ}2qiBy?E&Vk^%>Yw$}(eaF?0I4A8|7MM^3>~Gcn>R99?*aF}Yh4PbKaur{ zJ=xUDAWlMfpN#+EpY^>}g&184WUZk8n>9vlQ@}~957Y6-)TkQ0{=cjrV3#7YUL2KQ z{fbl(AjcrM?)Ycjrbh&$i9kvU`oCFcP@a7-689}W;LLsT6_7>PF0g&EJoUx6^4NpS zq$&QSSC;rXY}PCwc^}`x4y2T;A(08aaS!WF_>@w93yJh|r+Z4Mls`h;l^Yhq zl>zuozXyi)Q}I5=|5W&AFXvM!cSGETTi}XQ!Y>26A$YD82ePP?MGmbvQE4EX+9kQO6IIeE@4Vd6DpN|O(8wvT zLN`)6EQvOcE~rwT4~?z+Lvd_QM(VdHyQNgxl8j9WbtkMsN%d3!vn{As5u*l4yp*Y- zahE;}Q|qWc`b#i&2C_;vqdcaCx@X_R5dnP?=q1ZAo3~6UW`(+Q{E5Bz{{Rf_<12;U zL>U>v@zB_RCKXgV7r??o_#6A&74O#-`OP?)>-N%O@BeF)f52lrOYrBaH9xm(Sw$zFDS6WCqBW9X8Bl#*fY zcI)wbECZfb90ETUP3`P!kOpDy>@)HGgAxF%2w@+U_jTM%qjH6BMmBxJ-1q;=Z0(^dd*Z@V;R&SE0CXj%oCcsfXyi1&`Pb4_ zEhK&oK|iIuM&U2#GyvK05curV05fJ~=Hy=lB-ixc|2PfM7oV28AZ*Io$!P$-ug9JSph7zr7bP`{!O=jR2Ds55v#&i3a2q!4 zX#l!fR89lXf8;a({YOp%Y&eivO-00q2(Ii@GJ6`}mCQ+M1!6Z?MCW1HcN*Z0NyXJ( zKt}~GNU)p+sMtN5x-3{R!mwcv5%x4dJnpbdqOJWo;)li1U8c!=B~vnSbkMLff@&By1+FFp00gg z!K`wxs9F?&X-==~j{n9ey@yIIJN{EX4O8`DTO_C1?N2t2FH&7Ki`j3OMq<%z>N{v@&E3AT(!L)jN`)U?Bs$S|KIJ*scrzf zCpfpU?D%h%rMikl_sxyK70{I4zo^*p|MhR#R0;?+ghW+gri!;a{-+T7XkKPo0v(HIy9HeCR0@-{7Ccl_`FHc~AEwo!0O5wzp~)Ny>r=pYy;g~e8Q zat*p@jApCUwf{8c4Lz6Ma5nVtRfKj!&tn*R_J*G9ND^T;^vtWCtm-17HG*pawFmvB zi6a|&*sk5s^X@Gi>wXjPFq=X)^vv(36(!Uk;-5JLFQcbuH}q_sSBA6E8vtz)*ly^# zF|wS2j{-U^Fm33e22dWlfH3`=t<_Zeh6d>>wQW-z6Tv%W%mBPoZhYtePI>(yepBQ| ze0~JiUMhrrr<6{I2&#bSnikSW3AgW*$GVz2P0gS?fY3dVWZx+(UGU?)X)*gwp?jj^ zmRx~+p?3;>a7NxKbd$ZjQwl6eQy(JnHwgMX<#h~yIqwv*<00_b@09MpC3A8&GH{Y> z`tN_dQ;yG!Q00Y9S(_Z?oihD)1x00OMx-m$9P&=t)IAp$w3k5gPN9q3ve=W6@`M%Qy zim8r>?O_oeLYJl|`rcSWGiFmm0KFryhhTYQeK1<7nSx~`OnLrA)$AK<96tRZZ!G!_ zP%_snnK<&s>i=pcbpckclBGD#-}lD4Tq03DL~J-xGNSW2@I=?5ti@C=K!pWPbl|D3 zeY%)R1yo1ik`6q>_0PNz)fQ0C09dmLbAIis&x)vl7WH+1`!?Hznt7;3@-};=MtM~R z8LUDieVxfwiE@*-*`5|v6_tKJpdSdXO|ZPpcr;MFyv=5JNn+w7BEMx5$lI)1ah!t5 zf%pW%I4^md71&o=l|^(l3+X15nY_(PVbe=XV9yI4_%_=)KSK2fHYNZ!M&xbw+FfkX z`3Q`~!fNH@g16b)7qY0$z;+AH?JsY$@6T0Gr+{4-oYMOj6>qcm6Y{AiAjDt{a4i|v zCMWwgOC1uY3W7ji$6`{@+iX4hkn=X1`foW!?{GCW0NC4tQ;MLs z+1^(9lxsE^ONHgU&1l+i#=Wp4ZQo}5Vd&X!v&%r_ZPsB% zVU-0o3L>~Zp+=*>G;!o@#&+%7Y|_ChO4kLap-mxgv$kW(D@v$e0`v;OOX(@vx7i0M zG8-QQ=skh$+w9X9qYS(d&~kz4ZD#J%q;GnpZ!ZYtd9F#AO8RXG3frlB)Ok?%YjRt1T+c+GifFKTsuLB$oPZ#D>tEf>b zfa67Qt)m)=!y((`LTVe~&jITwxHud}?yJrC06;?oW{E->`Tzc-V#d7~O`Z}s95&;B z;&33Uq*2}(2O1iO!{HHlk5f87hr^!@QTtyJ_Zx!i8!D|h9HzIy77eY3Ve1D1gZdCL z#NqJ$CYL%&_0dJZC=|d4CHWkLEkZI@tiz$#BW#%a0pJNsa2yT`QGdtbkabQ5b)G8r6wI*U zsQncxk2oBjZuB`6@&YR?xHue+u8T0od8z}e>)_VmaC>7_<8bHqS)MMR4QjI_>UhNG2SaF)iE@Rg-5O#g?WBUuRZeH?FC zi$GW-BsLWu@o90D@@}~M!`e9%RaJiv!XYBnrL4laBTfu=ck@&*9q}5ln-1Q)L>yd- z;qIFshnkKYK`n$}(#Pf(>=J0F{$$)W>eHET-V`=Li3!Wa2 z;?bq{yirSKD$^r!^cYNh)Cx*xYY8uCnGvzK_Td-#S>$_>@^(=_ewm?B?$1A}ftG)O z_>j>)Oh*&Jq>1@+5rpx^1YOxnf2k~gmk_gU_ zX7*qj(r*sQj|Uuv0ZLi*E0|R^1G58yWaS&5l7njUq*HO#eJy_V_5p7Y0zVKlYMd9D z=L!YHBMVBgy67@CA(Nzkog#MjNtW9$Yz>drF;rsuN+}wxe)ir@OvW#He^BB74 z1%Q1lL64@4ME9iS!~LRT08SP{pzitZlZ>V$%R%@;NXClQJ?`0FrS}3nYzdC;`54u7 zbkE%UY3g07*nKb`6MHh1M|4lcVdnA)oqZfO0U;P%bkF-?na%5>BCx6sZgtPn^I_(7 z(*{^4!A19s3Bk17AMwKpqsOADogLj%;Ex35ngMXJ5I!PSpzfJG33q-$_iO{{dm>UT zf^-j6Gg$Xje}uOtgvHhT_%d~7bejSQJaHYZ?bAIJ8(p>((wgEO!jTot{}tq3`P>MXvS3vR#mn9xR=Jjg z{sn>*&o{4fHdHESEclNf8Zq7zQ__s?SvuZN_iQ?VG4Ur%-a~Lbo=B6Y)jd)1n4Wd? zdszBLU{F)TJGv*PI2uBi0$SNJf^^U27MPuz0qkH2Iz(pI6zHD29dalS&#nW(7$q#u zNOaGT2zc*^H5-JbLSj>*d$w&#R#a8J1B7phl##M>bkC?f-p`TiuH|Y zY7y;>X{&p}uy<5vMROKHaE3IqS@-lGO1kGV1}J59@s(7S24+hH$;vlAB?qH>O18o+ zBHh62i@=TQr+et>D$hzh4)2QPRgLcHH3<)u9;RHE2MjV1Nyfx@{72#KssKh! zVX-aIJzu?8M^Tk@2N1dvDT(rO-rSxMC`D3SN6Z$`JuAU@mhK@hlhr-+2w3+}q)N&^ z1wY{6%Gr;k&(b|xFGfN>U~ko*AQ<_8y;Y)n*nO1L<8CPL-6}EJ<{$W>8SBPVwAf|0SBMPoOQfBkgVkS66yt*(I9O^HXvE7ovZ)knT;mU7~p_ zb4J|5Ps8*H2p-tx^oC@YXc{!e5=}0PnNF)bRM@!Hhw*f2iDv3l)X!c5*d5i9C7PHk zwNxq+cSq2JD6c{I%UPl!J01d`y+o7kL<}ci5s+NdfB$2N=2i}`Iw@?*+T~Bxzs!UB%Y53A|{$2u#-l=P6>H3p=?xAV!tekHWhT7;kQInZ<@12 z(-42l5)J)FmT2fdvP9G4W+8PM8Qwr}O{IFIW&oWLJ)2?Q5>30Z6;uMC zVgfH9Se9rOKgg&m3$~0f<(Uh6c8TV%AN-bR3Qva?y^>`G{O131o?8-kx-&w;+BeVVsvN(afKkNgYS@1qZa9JZ5K!<^=Amq4Dtm><|nsZaiNL*bN6Q8i%PIz={e^DT0=0#-xTRS2Zxwh2<>K z(BO8)vWMnLZ}{oL7|^42Kg!5Ah!r>>uyokD_GA|BAso z1^DOKcSs*}#o&@V#zTFPAimn<13d^EUE?=vca7ZUiUPS}kgF$G3?67$McqZR(jOox z{o(h|U)27AR}9iP@DNw77+kil5*z6EUk125S-Xq~Q}KtiyUCm7m_Fw}=*|^`q4gls z^cUc5MsPi#9w%1}etN=-`|l8c+G08+1q8WbkVcvI5Gq#;zHq33(jgz>HH5(SH2vjE z7;wd)hX8ZUpWEe5-HEri7ctEP6MB%NFSWgAklzg+;>kUOl`o;FgAwz#q|Qi1m3s#7 zxo5EG&`iwKy+D86 zCUL%hJo>1xanIo9MRk~@KLT?;vGY^ra?fBwm1JXWGqA74n%py3czjlq{uHotl3wl^ zytDuhLGBrx(>0Ds+WiqE34$3@a?jv3hmCs%Yv#laqasN>5$L#&@goSqS3i5tV8vQh4PG8t6~X16!RPWo z1vW=~N5W`!;@QNxXRrv}AlMJ!=m3H?{mb4nSTGg$ip&CIo@F`r48Hvx*5@&p^(J83 z2{(^`PA|v*9^%fB+q@CfjLsvog!y0n#8QlI?vN7VGgZ_`;YfipL z@QqM+$b-rzeGyEDC6r!-_%$}Iudmp92G9M5#Jd3=7D7p$tn595r|Z?g_6>kH zEJ5xWtY6h-rjpRPczGk3R;DS+(ltN~WDXECo{NKx^p z!re!Hk231&6|e>qQznF5qG+%vcy=N_m+u8l+}!_Uj!Gx*kyg4q9y z_>+_cPZuOz<(|PYALTX$z7715aHuKep25+$H-+EnaZm+?5sV#fZYIh-gHc^esLCKT z5E84UzS_3ACc)t_D4WGTj||~#^yh%S6xePS8x~o~z=r`H6PPxOv7g05b5Hu@zL`{d z`+db!>b0n%>{lDU5I?GR9{2aFjoyngWy!g^~zi&?+gA1HI&O;y<%kis`9S?!e_N!fQX_C8;fsS;{A-#1P)P@HWhi;RDlhZ%eAQU&c)sf5Z}EK5 zf5h`e{}IpEtkJd9Ib?kq!8MIaW<6gK!*i*B5E}|QBYGyoKF`;E9B0V^sGz`e2^P=S ztk76hPOwFUDbGyUv!1W}C;dEMAxq%-;+iEBM?7D@<|?a(!^&8)^eN}>^L)LGkNV9* z>>`Wk&m4H7YvJg!Y9pXs0&j5OsjlYT%c|pmE(pAZU^z}RG$xPwD*)4+UXIfgJq7O; zm0G-8%gv~2w=xaJahhCU6c*MgCl`3PTzzw@%E0Oh&TTB-tu>(;ReNBs2u|t!iwf`7 zsjlVJ2oRFK%2-DT2IPn@$&3u6Qtt2#ed1-zhYBo$>0S*~(j>T215K>hOinbF~e5unUU{ z!hcX5_z&ts{6|*%$4mI5C>iqeZYANn@y$kLQcF;TAn#WD^%)r2K~Ru)>y$SWW8&T7 z>WO!2OW}g*MUZE6~mqjhXiH_#s9nbe6c_gtKZ)h!Y8yrjNQ zr4+~3+YPbXNe=)zQW%ekVN7^k?yTcWO4pGtjX$Zzv>xrWM%wiD0TOMjgkNs)r@H3SiU|Ru^=6Q$BHQwMfIOydxMr zEGy8l^;y9}Dx@?@JQDZ>%hB9e;@J9eNnS%*0>bA);`S29R=X($3~3(-$Am;>avWQu z>tM0-2Ed0xkf-W6w$evoi%0BANFM~#0BlAaTWNSRIF7BSTN^M_*8#nOP2xDV#+?y1 z99!9QR%DXy2j(DRdr*JJvDN#HBx7wluvubF99!4I+$Q}xV4EbpIJUacLlDPSmE5UJ z(&xdvBG$yQ^@_uWV=G)^>i|ePW)%iHf+?Fgw)TJRvz8339N|1T#j#cHml`I06JV{J z^wzP}u6VL3`>Vj-aB%C``f6=1gHHxF!@;d%Ye|U&V}A{>^$uN;G)PYSbUW)L$H1KYv^m7r!9BJYiIKo)jF%*4SUYN;d@9 zA%LK1)=)9lu{C5Mb}GCM#+#PqIJS0w9Aka zCP2_1K=cWMoDzy-D}=kQb!^SNS%R7RDd-v21SYYLt&C0*IJRD{(tw$|4CqxU2^G-S z4%V@?@_wRnwFLNlAi+AeeqWEzvLQ?TCh*b3af)gkTQ6oWZZet;d}*K^aco84bPOBX z0m8S|kmK0uw>R1tIS=fLHKL5;3XZK|-AbDD>N6+-1YeJ~j;-2t>znlXfF%-c9*1>o zrS!|E=!c1}0j!>r-a58&M^{lD2=4-{yQCM#)~YwqVTK}poK5TNE7q~KI}MY^T!71k zkdfvK`G7`|v-l>^c7VGrK^$AJjm&SRlAi(nVPQ>El*vULTfNH_H%#~E`1Oroo(D5I zJ8^9FEgxl=rNOEh$mB^<99vbk)iulxVD$}TI+LzAwyv(tY3SoY{UCs@#Icp{#d?PH z83e;_zr)Awks zW2;Gz#yHsqM(qHW&I(YVW79bEnrVAZhC}y&epNDu2BZV%()I* z9}xbhM_9*J$(XE0Sk(rrB{5|}2y$%wJtm)Fz5$l0A1f(wY)!ac(h#N+A&#Gyb!?65 zT?_s{#IL0+c)B3fDvqs>PDYpl?*o2JI8p7S@3Qw zh+r~f2^8qqYF4SZssut>Ac@t~vwk2{M|sn?t-?vY*86by{DVc&Dm^W+tV&w;GFH-X zf7WM|qxaqH>iNJuSTw`fJy@;rH}AosNR|5L(qi~J&-Uad>!CdoHfK|LcZFrEiF<8+ zfUOf$u;$c^4e}85O<2X3lCW<8(f2K6^q5JyJFMNGz@uO*I|3)wZ799DYb@`c%riz8 zQJoNz0=!Bf2RndDtKG$Di(Bh5K7SL{C@xby`03h5MMS-Z1ZE#tbVT+S$!^lbb6%Ax z?~$nX8}P%2*eaybOSr}_@9{`7r7G>Y)RR|6A!$IaQfgc=To=+ zmX%Xp#MBp@tkNz&+ZL^CTTt!e)~v$2pW0#w5=dKc+2RVUfTX5#3ZDpu8_YJ?V~)dU zXH@FYIZ^DieE&-HzLAd$2Ejnkp9xd5vL{F zwNA?u58x0i0#1@mAx_JVP4X#9sOtf0MDQqT1bf0YZd@hKMt1|$OJM7?4Bb)9z+(Z8 z7nq!uvAJnuC^d(7+92%xr1Kd?Hu{~<*b9Pl5mws;>1rHR*CNaozG}a3A$9l{b{%ei>Q)D zdC!h4Va{jt_zGQ|()pdwcoG5+%TS~qP1((((#pBil_whGH{wSC7F$q%N(?!lF(6Mt z^%>Pie+|aIK$e`(xD-pfW-kE!-7?IemvgDDm%)Z>6MW|gTnfE`GLrKd)h5Q_1T?^8 zAq1Y!SQlE#lq3y=mO?UC?D>rSdr{e!0lsDl&iRbWD4KIV+SYj>W$cJIR63gp(W@hl!+NuIG=GISHiYJjMmwO6|7s=I*Dok;B|D>AUTuih$Df$un4risUbbzpz|3# zFy(wk7c%gS^BKX@V{bHV>g0D}-L*)%C9)u!7;=JU1M%1bL$_uQK;Qx~0fb$tN-Rbzs7}RuVB5`l5dI#(e z5oFNtiYLYT#x%8vcE+?lpV4*?_QS0QZ8w55q?ygyFn&Ek1O0>*olYUd;dHl^uDN-fnoPgnjQt2JfF4e!|!}wH3 zet+CK)!{8L5<{1S*ykX~1tEp-eP&09P5e7V4a3Vxw*jXUf@=)Dszivjz2;Vv34asV zFu_HLjX#}_@ecvb5ttthW#s=Np<>28KL#E#5@M(EKM`U?l{CuR<{qXax{-9pH|V;Q z&QFLfz;16HwiDw7!8LR&=T}Y%1DG*-V&JaR{`88gg_xSbVs5o$qyi$6_T-Hg;=BR zO|ZTL@PQ>bLhKu~izCG1#^hFOsbcwd!T*Ke+DPRQA?8jpTc~vnVD$tSA+~K~W%K^( z0<61(TOsz&1YE3(pLu#Duy+I(A=dgOlx7a%mk~yfMN>OFLhS0uVwk`I9u&eZVg(AZ zaXC_1h+P5cHW8^7K|+kG86?EwH;hPTZB!e(jLV_dR7c=xm%0{fBl^RX>w7NO!)vLm zjV_FZ@23yw!w`7xotqwGHnvb~w7V>FGYMxl!Wm2czeu-kze0b`0dEz8GtYDVS-Npy zrEmL>2O0t1FN)VR!fO5=KVg*{Yr^_EEZ#zJeK(P2XDh6RSH?ro;k!L57J)%c4e1E0 z#h2hP(?x-nwTvKPm47jMZ#uwMmY@$(CL*jpYEcM>^Jw~h4UECU;*3OC)yiAkux5ZT zUr209gjI*RxfNAaZwBFOBAueFL|Fa2vb^b7r+}Sva4W3dJy6DU%zuDA69 z;?R1z5lmZ&uxeNaA2lwGc(27Y)yff8-}Q`8&4G0loH|F4u;PIx!s;9uc!sbFp6>SG z(WRCxl!Jv;o$oM_Q7bsYD%FjQ4kF*5DDR)>30h%Q#d~jjF1ffzd%&oQ(*p#FlvtYXAr%g~WM@ zu-f0cu=)Vl62bo@!vVsIrY{};7(I4i71X2xdz zXf$UY1ZTi8n}yY?;Sg4QgU6C0cpH=gtug|)lAltd$E7@r(Z1gGGm?!`D*rv65Qb~N#(y$Z?8HvC}(QHeU(pPWSR#YWj7=#i;x=(q@^rJlKuGW<@7}!WW;mS0|718U3>XgX_u^DKm1+byZ&UtF zDKi-k+r#nwNW$j>TOzm&hxLc?p$g#J0qqu;OGg>`mo6P}+%FsQnc?6%;5QtIDrw-A z>4WQ}NGD%AgyBHx{Dwo8Ef@|-NL?1e)rCqc!(rZ)dg#Ut0JOHC?nMk44n?ZQtJkSM zx-S?*0$DN~c7-~emJu`@@F97nzW})167-vtkqn2E+cBb!0z4;#z~ONEhs>rV z|9}vB&{r3q6+0aEzYS-7E`a$g!O_V#&@Rq!n0&jC8b%dM1G5RS$544>I8?#P7BAcP z1oo=nG8|SMC}{9^flYRBI~-!)_sN;%z}5&Z!=dk0c)fNb{xD(mSTwb>GaP>L-~%t0 z0sbw7$;=A4x&OIBX*?Wa@Z@tMa4mv{164C8D}PV?zxij6);Y?H9or_jm#f6>XSGSx-3!8?lJ zytcXitdqI0(zpG`1C0Q0%td^#iY~_c?uegGUXcN>tk}a?1VeC@SVZq;tCJ74sE@vx z3?S8lni|s4$we>WRtDV+Xa~y((#cI~VsYwqfI}@om!?e2Fo6Sd;^KH!5iwfN1Y@DF zI3v-?U+>OgSX)8(PDpG@baKbP>nN(Kz5v2iB2}fVL??T`iZVJ`qXpauKHTc$zaJDg z9Wy_$B7%!fUQ!!lwJPFk6UJ>NI=M+0{2%QQ|B}Tt)ymPyyE`{jZvlHxaOxaEI++KW z=;YdD;2AnOc)I%o9U`^1&doZx_m7y!s1+QYoabM7IsQhz4=L~Y^aQO=&e;z?V|45f z*#CvVm|j5ytCI(J#VlGHm{;)62^XDwyC+V?wgA{k2!T3z?(|CPORB0K1jZO)aW7 zfnF6xQ?g+cFmAeY|CLv@MvT@jWFL*dnHW2Ug`4gsypzGlDhftfVX-aI$u-Z^R8%G1 z7=)HY>OgtP^rJklqZD5K6At^5PQDF_2!kbs{dMw?zPK}(9>FYd=8>rzR51|-xf-OZ z+k|fdwq0-$1_N@HVf+N3p9SWUP)7a};R}G{t~P^`C7qo6sGl$(s-#ig3Y{_=VerCn zJSj@&Ck)P1!$eRIsT(1XRj2VF} z5eCKIB%Qnx=$DoeBn;9t-q;5Kp0EV%fmufwoJ+;N-ai075JI3Zi1@6cDM{vI_%(>Y zb@5rT!XVQ!%xq-=rdom{3_@Yp5eDg)A2U+Lx`Nq**l|=I5e8HCH!$<-C}86R7h%x% zd!L_T0kCBbZiT_^1_jOhy%X4Xf{QTt=_Z=&XT)D4j2?@oc6Nlprz5K=*Asv-c&LVu zo5~xglRq0+jD%@l9k7Ff%W%k7ydmRP0R1j7myR;>FWu{a z<3dB>38CRo9A%Q9jYO3+$~(Jlj2RA@PGdMwI=|tNxV{+HW0AT8f~yjhR&?_8UG!zD zegKA9P}e4g42J=gva58ekDdj_qCl1mhb1R!fbkX3J(dwP94ej1z&iu*iY4f#l#%G< zhIimNe+1Bd(pO9!I2_uw!dWpYNq!JY3dvZp!(q`^`0-f_V45X3!yyZdJHw&jXLVF- zGS?T(fyC}W<&ohKx6Yi*)Kh_dD7XxVK`ZK-;qp1KFCE+thv*j_P>C!PgtCh=;S96^$hD*5N-*HO^Hr!k;kQ|s@j7ZW<+qEqO3$GcYPZ#rf=u%?2GPJZ+|Uhmxy--|GAE78dzuj4B-qYyvQVw!5@=;RHl zRn-DuYXzsy5u}rOpovaCM+TmulY^(bIJ8IVhRelSC(rm96B)IFqmv`*Lk#CRi?vS# z*PHYNtxjH2B^)iEiugJf)1!%Cb@K2%m7)KEy(IX%go{otJ_HlkV1RE6Ay6lO)-#8i zOjXsNfU#0ooQ>#YZ->l=wFiVFLgKtcCuhhTrY-~fTksiVI6x=U^rho>eN~Kg^3Etc zX|zanguoF;vAz*UEux(fXLWMnlh~)y3A8>4&VXY!>*PKMtWKUhx&-_`pp8f17WUK0 z^thDgE}pyh#PQNbC%^p*o)SGbKb_nl9P5I54=n5_E0^i%i%yQh&5&AO1n`FiHC5ix z$x#WA^*R)WGa_(ueS>Tm1q_|sWt>~xMvT_Q!6+{*wqsb($!n??^|6|P(Lq>jOLTJB zdkqv-Ne=*F2$AknUNZeC&qpYycUDvEUm{_U^PHbBNKEwC$tCfkUPX^!2h4m+rq)r# zL>TlqAFj3$ejeBr!9^IXZjy`fr+~tcjY-ZWp^W?|!ttQQ{TQEWgh6ePMHmoO(kO4# zmP|$%6uXEgMd|#6!O}P3Vto~<-=OTi;m2Tg^7Nq@(ZCY`d}u*^h!`RaHoR0{9i{r{ zwP0)wWQi~+Rjv>iKLP#4GJ=G`ykh7tcL6@N1bvz^5@9g24VH-G&cpSEz{Ly_2EX-e zY)VoYgt|g9R;(~cTY)*g1Hi79;0S|~$k`DFMfWvQ=c!`vfjNcPSExK949<9cJNiBa z_L<-!3_|YU`<=+fnpn@M~6#@aAsk+rby{tDi7v*9bq%Tfkyl1lNl^8ChX)vLs&W`ZWN9 zEvTs>9bwRG0LGY}26T>P1POzWren%n2XLDu=$B=-q8pY&9u-lq@oaSxj7!4ej6@hL zuUf;f9)l2Z$yZckN`%408&wrmRTluED3JzGRw4{uor?QKX$M0sU}+9+g+Yn<{H9~R z0Ia9rA`E)$#aJDV__2g>TZu4OSRP*8j}gDbVw!5@2!j^0%c!q_?Gv0jN02b!fhNLW z2pM>WFbJORlF=rqt=~5ewAGhUcWMPk7?dkpSm_FQ>Rtqwi+Y(A2DJ~N47vs4J6TLe z6Tu3Dy(6$H*ALh*!Ltx9!eGI#)p0K)zt6d#4x?JUq1A#ln3gaJJ+<%veQz4@?Tf*ok{ zuHY%rbMq4h^)Fzzm;MD7ekCiT=;@0vh@VP+nTG(vk(5EMydw;vqcCOYJY-lHy4vT6M6)vVCEc|YDEPSVK8QFE!COue}Fv|T!cY= zoVp-92eL>&Fv+^R7nG2P}%p1s#?FHK`EV|FwnKV zxc>^NCsB6YsI=lS=-moTy%@k}7Sw%-A;MsEzd~vt)kp6G<5(a|ghBhOw6oxMp!Y2! zNEnQ{fOly0HQaxXz@^YbDI*aEbr+%oCj(3sLZHWB-b9xvNlOr(7m~4Jg+bNtlCl36 z;9yH|gh5>t%@GDaW~r%0Q^gj7xt!SVQh7ue6rbh0>1`LV?*$iOFn>TrGyh%$cFn=9 zFetXDikZJ%*YOfSF!n_lOz92roPhXZgwbQs)Xt7Da8JUg8ma?qDunlm6(|hmw64t_ zgPtJuCnD7%NElExgN4EQYK2)Cw5z1lmq@o0fu{}XS}Y9c4^ys>xLgy~nU%G5N8ri0 z1^QD2o~F_#`-B0-M*l=?lkgo92K;{!28FkwcnRpH$q3Fo#r0>80T)*Kw*Ppb5#UW5 zgS+`@Wi9{je!?Ij294PZ7GFnjJzYSPkrf8_M#HD7-v#i#1vNFKBMiPQN`A8yKsQ)M zkT5u~ryjol3-Gul=rEbB_|psTY8A<|)om~y35zolVX(Dd2E)pB1N;9GxTwaI2!k%y ziz=$Bt^z_$BE?cxA`I%+En_-X2Vh+t+zNv;>q?l8IRw}!!9^H!uG&QDnTVfD7`K%O zgQ#(sZN5PKc8h7Ml_LzoRuxe{0lOqPb&eomzynQ$L3T3m3}FyF-PK2PRuBf+rl69IMLa%#>?gu$?gSd|@E z5y5wo;Q(Pk)0d8)gds=5plRkjsw>#9BLt2(iuH{+Y7y;>I4cY`ynu;!3TX2YoB_vd z76yMDvBF@%qLMiO5874)F1epDpvR>=iKv(NpWh1@VX*Ngo)SGbKVdLEQ!zBseOP!* zR`Sr(7h$k^JFQ`7MY6mI3~H*pBMdIwp-WmS0IeyEf@DL4!Ap}S5~L!5&Q)+56ILxs+b6a9DA#&%Y7BVso_x*i#4@sl0*0;QHMnEDT10G?9o@iy&b@)eI5_ z(s}$l9CfkuA_uc8&rVp1%RkkPu7~r)f8u|p>)||cg}OOR*V~s1R=q>lVICkcJ>oCF zG1Di-f6PQ?sH-MWu@b>4iQt+>C6Y0-+MTZE5MB>hBf(|N+^SNB@os>63CvZXjQp!W zMU5+Y5Wh=k%zO&6j2WUz8s*(uk3MRNZE92g#+adWeq&}-ariZkBK2v?ZXuOc#B_;W zN$Bgh0X(vxUQG-cGslKDP+w4eboM*A1OkCg7?zBgDw`QIO@Xx* zT*k}?J>Yiih4_Jl(PPon&d!)waS3;7OawSb2uFw&IA+GJN#HTF0i+#7q*?@x8LDQ` zn2A4MHH5`<+4peG0D4Uf0>5ghYq6N7KTNrfbGbg-U4p%A8;@c6%?o-X1b($2o9GkM z6dRqXB62ecyO5aX|BIMza}F0W^a5`Pg7a?Y`m>nk!b;!v9}hGFyfyG0Zx++9-uDyJ zOGD7dH^Jg|1lQ~UG2O9nGWzB*0Ou{JsUaOPy|ps+-ur~!^CkE7UKZr|e(Rpv=)E}r z7PJJtfHE<|1YWjYHM6T_h|#(l80o^|j6_WLyrvE7MG#&U5}Oh+o!+pZqN?h5L6}UW z)s&Tp={B>9nvS&`*cu17VtUN@Xwxyj2X;ts5!1zXVTk;S_}>WQwh}Qtq81*QLd!-X zFs7+ij+m}k5pHl`DS}hy2olpg&_qmsK?a^7rh})u8EF61^c-bbOlSEA6B)IFBc>7@?A)IU!W&w#q`WUSbf!-5dV$E^zTHlV*2M?Sf4uy?2_Pr5iVkS#1Y8xe*lI) z@EJEgek0EC{nn}b8mI?URh<`%;=PoljkVPpu(u-wjyQ_-jW}u%?Tk1prnlcKtMpmWen)Ty9J5(W?>lY9 zbZQT{8lQmXLF;o1n_(cMQF;2KiM$CXiyAQ<`Y)amJvTowT`C#cN7sRc24rP3J$(_= z%^s4r>I$H@1vORP5!2N_hM?DDfKC?1*JQ&y2`s-oJY8DtL5$X)g0Ws$Y{#%5ro;ZK z=3^ZOAN8OOQijjmtTc*Akw>fr8dG~7Ah>;)CN}c7X}kXmREx@ z_37&1)J1TOq=`_3L7R@LY69V%fprsHgh97WiHr{iG*)0P31#Fz5qbc}R^X5eCa2<4I9EKVi_W8}0u>+#LkhG%BqKg9|SD)sx{NGD2Wb&mo4H z@$f;^FUwX|3#mRj1&k_zED;7VJ29?w3!t4WBS;t&*a#c_0S>nWy__-^0U|@axt{;o|DTL8u z(bUe4FgX4S-mgmlZVhfQ$(a%1PKGGX0R|A@&bHj_%{3GH4Rlf zdQA}oo;Ij!u`r-NOu6=OxgO(aJ$<8e;;V&~t2OA|5O|vUYrIbwP;B&5Vw;4ONf_|| zMHpP_gID-i@Ma)5^Ay*gg#i~<`nLagpb_AG2X}9=Fc|mLPZ;d;qL2Roizg6V6?rnU zUbath1*6tC0sLb@O%3S?gNM_o_hx#6qmKxjJV+Q!+<{S88lcw_bXA$H$jkQgl*+0$ z&sJ@~ctKd4kqCp0`>Gq(P!PrmiA{+xC^I3mqN?fzAS@$NI%OroVA1B1reo~{_ML-U zVend_xgl3y0CrVy5e6;lVdj2}_z<+9X)6&1lXpO!=R|xVi)pHrBMi=-a;a2c>4H<| z2oeT7&_ozCB?He82Eo(aYCN6Pgau~#t)$MNic%{$!l2)39Om4Be78{EN9YM!VQ^=D zc|5=$5P#NU`V0}QF!*~dwjkUF_DJvxgo`j3(geSdGOG+K4+0wx5(f7dhN`PnRb2^; z+QQ;&L>Qdt-oUUrfzVS(oRBMkDm@RaDe`3Zya_vw|=2^L-?E3N72i!fM$_SJeYfVVBEsq&667_=VV z5IqOzGGTNe8zKw_xNEAeh|zi%7zc&Lc0?Hbd@jF_bq$O=!eUz@3_4HEpr}eZ8a2s+ z;Oaqn$@GKIn<1sQ?p6#7@))cIMTEhp)%=BlH#SkdTnUe$EjZ7Usn_s7M;J_cSWFEj zd=RkVf{QTdkux{rGXc#Nm`g$#`A>ue0LN9tF&7pFXYfA}21J!K$~$9KIU@{qhT%z3 zIzM4>xH+0SJS2mPLU4_w(uy#6QH5eCc8`Zny=1Xf>g5eC?_%>@wY ziMT`bRnUhc@HCb2J)bb3*yx+YHVLzmFyQ}-Fc>lz9DXu>Y11_xeZU6B= zBfwkmBss^h{Pu}EgNosoe}(Ya!t=0p`x`9&iQxL_6Pk>yFgW)T^>KH229+6sK}`+m z2!kD4sP`5FTFx?pgh3*%LD3BWwzdR)iZU_71j3-`R}IuT#NhlV7(;}`8Hq6XseNw4 z`T&Gagv6#q7_6V|Ra8~|6$m?sbeXaeVX!q^n~rrF*m(!H!eGH*pJM$Nm_jR=^dbyC z3Ju5G7V+_faa)Nn$Xpwa)iQ{$YB5c1gaB3u}7!r=B& z%;6a#FrOlDT786Y5e6TB51C#9U|ZR*FVs=7HC9fif&h%k65*2fwM z!YCneULp*xosCo<0$VEhSuz|T3~2h&@xNopkudld@AdOw|A`Pd;waWP;;2QmGvcf; z81(^`-(t|52?)-BV>SzeR!^)j=sp3XCM>%pJ_E)TVX+;z za=rW>dgUhoR|=tCWf0^p*Rds(s*WVPfgcvmzf=Ijd0V>+Jt@hYYryXaCtF@{;_tm& zL*-qr-TBT(s|b`m3xX*KSyl04QgI(`YaR9V%h^?BU=0xXvfHmwDZcEMYGhJujKyiD zm@G)v`ZTT)%<)D()dBSG)}(Zm(u&#yGk6(AsgQ^qJX;Y55m00VnH zoLL#hyKjZSAF~X0R+FvfR5JTPraqamhj1674_ZifB;0<5{4{p|Narq5%bXSrs{#E}m&{U2T59cSb5e?N0sYq9#O ztM}f^u8Y;JUJ@lj^hEH{A|iTEltc)k6D853MGqocB7%r6L83+wf*^_CIWu?O&%O8O z`^R{#IcNI3=Q+>J+~<9CMS@BLEC#_h-;^J4$dtTs9aRLeF=1tf@wh6xI>b zn@DLqGO;4&54OSl!8v69*z5jChmI`FjFc~9hPPr=>}nnPPj&ly-4^LkJ@dJE3vJes z8{V_ua_vpXd?em;Mc(QCLgqC|oHd!|iL<6PTXU!pNVbzn`A5({7uo-z&YEZkQRIrV zrhx|=m<;|6O>43;8JPNrEAoU36%D-*Ka2H(@;qlv5wjtMbv5v6Blv#hj1y-~S3kh+ z$18~M<}iJS1aa0xE7MU##aYwmeeve=0b&+N8vYeX7~-rcihy`*>f(nsJ&2eKp$Y$D z|JY)k$0p7N9YvmaY(jBLM`XY%jbNIgpR%f+$EHn<(9l9a%L+qLhKR?eeMRt;tPik_ z5SWE6TRb)mekY6S4Z>(4F&9G$^Vsz5(~4>qSgT*aG{a>}hAZ-_DjlBxpq=>tsh-EC zr@67&d;^TXg;f{x-que%Hq|d&6rX#HJI#=YGR##NXSmUz5t2&PDEF?w9$@AFM{3fb1AK)4xnCZY#^*lDc^mcBg z_knR#SSB0s*p%UC^iZ<((FIg&;Utkemz6h6{Rs4bToTV?Q;SQ&hR3Gfl_CsD=Z{ID ziX!+@aooifd8=GSY;LuIH6U)RKs+{0JygJ^?+WZqNiQCoszIoF9-Hp}SJ9C4G%!Dr z+{9y3EsqV4O`ktYXGnSvmz*4}K@WExoAS-fWAOxFbv@j9Y`WMYr_H}3ul@y$INH4|g7$%G|RHh28_~fZ*b>Y3?UD3b}yz>%=%Nn$ySg*mUwwY+3vZFgkNE zLEHJ|JT~1w9FOncgHgn>JdaK5Zq=|=tOYEQxS1Z}vFYUXzy7B=>bBFRP^OAWV z*xz1y=dr1A#+s@Q=SljkDO45&JDlRNNzKI-rjm${BW60m&K2je=^WlEbyI+^2q81q z3-Q==1@4t~Ux05rf_QA&IySXkNWX))739P2WsxqI};V{-ndq{0^A0^Vk&E zF0Jz20^?x_%k$Wja#$VZOPM`|%8p=LZ#*{H5q2J%UcU=p)n&k_D=a>Df=hI+$ZI_- zuY9k9_PS8b`jhP7%ke6HoiAd{xw=O9`9^^-!I4z5?7g$8_?L7D?|M?aJTkmi? zBkVXkk4^iq@njx<-+^`hMa=7$tt;}D^GaKF^$@Ix*uWIG)gKX?0T*?~6|byfe@hmb zzHOpq76i)@jFJ+MP05PZwuG7_qU5m%z zY3*E#n`F*QPfMztoRqV5D`YWqtn!|Qc~sph8F~Xr?!5FQj`gXzh**W-%g1G#e`#{W zOOMIddFjbDp$xwN3;0KuLcH`;ujp4ysBhz+`-F>dyquSw>$7T_VsuK>IURz9otK^r z2}Lbj6i{h_>7|DqprW{zB$nA-PbH?u0d(9u6R<-G$5hurb&jc`elMc*n~3Pk^yOK) zb4;}i=a_mNqNh5@%1-B)YQ#DGD0msLFC4DpnO+=Ib^bX@B_QT7@ROk&vsY(#+u)eW z?zVAzGUmPGdGLM$XH3blX4_LkIHoeYk1+|`npiAObG|7(C=QhulbF4^QtPtsI`Vs2hBEI_Hl@>RmKN&j4pGf^Q_7Dz~ay^@>!J zh;IhALvZi#v}Rg^p9I9Gkxg#qDhmoeEMf~{dTqr6iMOh<{K%yf87Fh4`+kTl(<75HB*q)tTeeZ<mo)Fe$F?ED~@wghrt}LNb;HSkV!@!&8lny~ERjS!=4bY*+^{yOO<$ z^^wEVRh0s_s)hp_E4Un<&bb^oqgnuLsfWA6(=h`g>>1Zrz`hY&4o}m(R}J^S5Py{z zheflWy~ER^+3=CvX8==TP%U9MS)qrgW9wTtM)^T1MIu`fc6iEWh992J*lmBi7(XmV z^+&p42s4cOaNuo8l7!ho_8_-4&H@NOY^XE~<*_Er^>%iIpO2U>u?)tmUdIuB1T<_|hNjemqQ^FsNs z^^4%^V-9TG;c3L{_^gz!4WOw5H9OKfJUvsAznAI`w4Y-HX0OQu*R3la!4K*u0G#Fs zdZ3)>@K)8M7ERO;bE2~bjIF{l1jVfFg=Dgl!_(~j3Mn>KKL+7nl18y8 zIXu0!sG^-%nb89|5CXV6Je^gxzCG2i1gxsya(Fr~GtMenBEBs#(^qnM`o>2%XXu0Y zcO9nLD(~>LL#8t7LtqO9=i~@GJT-F9!3$k+?s z;c1(qMe+UX0(kx*_>OS|-QnrGwXj6%LWnQxFnyi`cX+!0I1XXz0c#=nW#V#pI&NTD z9E|`RAcWAv(^4-dQPdsJFW|GaInW~|5JD}Eu>p%?k7&>0+~H}OPx0_>2wFP?&)^K(9G;d< zpF41?YHfW`dxQ24g6Uyv24amW>I%l)p8?-JG>4}H3SpEuZb65qmCH6pCv8Ly+nCD~ zj=mh8UW5~9t&agX??BDQdxxjn_v8lSKG1)KF`aqH;py6|)zxgoXq^Sc=R+`s*nH&h zv`C3^0ai6I5`|^*lEc&Plj|xrNp}IE2T2QAR#1Z{pcekuO7f|KTUC*TgM`6&e4ohs z?c&smK2@X&hF~qS*~F|$;eVbmsF%H%iYI;)*h#@f7-U_X)Zl*rdLXc=2@48d2;Tu5 zbEhJH%1dET26d8KRisK9e9rW(czdfVb1{q*(*+5GrhnpBCa)rOX9QmYYb(Ow=l-~1 zp$7vP>p1HCBWrI`82m}6Mi+#(~Yu?}nMi?~BlM3(opr1rA%T(gHfG}We^jflQ z!b21W=KmrL_TNuHv%yP(sp_p$ru{}3n99oB4z>qcfPYD~+Ey4mDjp;ZDpf~83COW7 zf-j|6jGQo-@GZY()E>Ya4%F;OPZ(q$h{wJj0d%5cgb9Q04I-RUm6b9w* z=2e-^YIOjNTOz5}|Gb`$zD3%s50CaQV`)3t(~-2CsY;p(>(B z>LP?%9Ag6(#~#t1#W`WHAvqS_uAsez;2E4@8)2}vVD5l0DEKTdK8pa_M+l~egMMn7naFKguxeCGX+?!!RRb3la~mCr0Zf7 zo1}+=Fq)()EK8OjI6g)_{XdPYXN5uT(m}#td#hk!aJz2}b%sN55ZQdstbS(0WXGWG zpgQU%@$10;5L|@8%A55Jo(x5#La@nAO;}LyLiiC<$23~r$O?l7Ad4^{RnjPb&HVwl zI;G2Dq?j&97{sTKI|d;O8Njbk6`LD-)=NPKInfVGuh8sw@^@At8kB7_3zlY)xu_ z&`3x&7bgrBpT+I;&H&$V1Wy>0L+PF{c=mN;m5L3U3g%3*)3ZJz47RQ-XJ5R&0QRNe zA`Gq^3J8Z|z)pF%69#2VCffD;99b=3zXa!Oy64{C{VZdgF3xkyJ)iJ{0*SHw96zSF>m}P@g%LoJhVe6I8 z)GOnh`bHS^_!6HnzX6vOw(8dZ=K?_ zw4XsVm6f?2Y!9>m|IFreEr}h2G!=q`LARmU`D%+CUqkRMUdYAB34^hVQ3?GvfKd+A z>_|@-6Y?UVrel1a4?FDvRa88aeVPFhRguymu z@B(2FzT8znr^W3!Qr8HBO%<__u@^jHut)ph{{*963&B^CBj|)dmjPHLbz8)Db(oGL zL4?6DWv#ISPY0jZ&$mV84bCYH^GWSR8vqdlu(} zLF^&usbQc^LGTRDu#GTST*V0k|DvSue+Ak~1XHshVZh;1Q5n%Df8Og!tT0$s38Tbu z3l;{ipeOZJiw?^X4PUi<&N*Mc@hX{i)_?*Ib zh=KnrFkTiGJ2Ak*b?XH`)C;isf-yu`CNB{NvpdyOY?7VsTcEDVwi&#t;~2reO;Ys{(#E941-!MRi);tF{rL$J6AgKW)|!E*sB zAh4+k3kqHcS0eYAlUuOI#T|n#Ad4^{RnjQ`)>^sjj={=!j1frttQh&_i z2C}xYV{oDlKGLgK0r=8^dL$Vl3?A05q{g#7`ZyRDLRlgVd=K#0*AIaH=NMtapwf73 z=4Fgap>iRZTIk6vNOla?`EXiY0bqgb!30S`iL+{-Ke@&U{VqD7U@C7@q%>)q#G=9CG&TS z^sq7smHr%EA#{8sB@Sxs9oni*{X=qoG6(GwhLaL-u{?Bt)wyVg!Y8Y^) z3tC0=52*i1q&~;ud~iGHts;$*VujIv0{F*)Iyo7#irh+FQl&-=zW-DeH(?M=7M3Ne z$iRlV!H5G|+cCmck+DZ1YF+`@%@K4)79^|4xi)C%5P;)_5W0$N$&}01WDy9fgk*DZ ztH^~VSgiH}Jm?7CDl!dC^j4AaBQvUOY}j2eACjG$^^sNNw?3KdDv<_#8iQbQSw+(B z47mF)1+0RHyH(_e1$nFrYzpjU!DSVBuL8QPC*t2G#$nOyXKxib<;Q)Q$pGgGp&(hI ztH|CC>CGy#8KgZVvK3*g2%Gt$RU~}Eaf&6*U!77#Jw#6aq)o8pO>_95dXPC}2{DJx zHOr~uEU^MORS`6zsRG!B zp9pY<5JJu2i^oZ9P1b?1T}U<;XAUdJ!M=|HJmmngaI!& z+7v-BN`6=Uz`HYJqt`R7tzJL+syF|acjtZM@$Ot7yfz4)lsBz0o9d?M*r~abnpgX3 zk87wAU`}*w&SUdx&p&J_gIJ)!Ul)3) zmbdNBIj}$~>aE@2l+qN+@dD|Ys3|{cI9lqQx!RW;OuV^wO4YEw$>CB_qtSo<2W6w}+jfR}7$uHd(A#!q=sm4lB8RrjWfVtW z-nQ>diw4*izz_#&Hr{*NewGBPNKXSgUlJah&0eeeu5e6MkmNfVjK+^>_HDN))3t<+( zF~5Bh(8l46gbk;s1|C)kipLhQV#aZ@L31q_YCZ z??8Q)3=sx-@5iZM*dARSj0T}B5eDn};{l;N0qx}&VZxxyB)nsd1~|nL^i>ul!r;dJ zqDn6TxK;?E!eGgqKur#T@S~7yE>0M1n1#c~n*i@Rf+q~NqYpe`@c66x>UTCQb0Y3P zA^84eeMA_nS|0HBRtZ>D!9^JS4~~xMVK&PGsA_!58dF62-om_ zDU<}-7SpCCf?0Gf)(Z#&#zy=-3N+ z^BQ50y=h+c37AV9oAcNR1O8!4@v%Ue+rjog3-FK0=(ECLW}_fsuw^*}?j_`U4Z-(( z5*H)44WBqP0(0|U0LdE$K+TTygn=4`NuzTCE$SFy!XV>5?DHl7Z0HC&LRKrD&-W`= zRi!kmRd+D@3Ck2D!eB+yw3amqgqcDzS&15{kG|a2)#JmUWk>DZ>s;lx!r$XXQ;Z0MKibx?tX?1t5|SxPghBSZNz`Ou^8_!-j6;M0moFWA97alEP}849odWwB zLa2>0HejRd5$)Ni69(y;pxct7J2N79CTiG57<3=)gh8^I#gr}xS{Ve>OF_bb!=<9; zqJRD0^@fidp3fI=ic#XY1qp)|Rk1y+dm)GZ%q0g$UxdNc%n+seeE=UjP_yx#FgW`Z z1EE&|{ZbhDn1=|1-`7Q`B8b8MFBlhuW%3bW(D*98V_;a1$v{O-RkX=Vgu#sI=@gr! zV?oGEQYn@t%a4jmKx%)lwdg|%gFT>#FqrykurO$HyOgTQA$T2}-pr~#8z#cw$gOg! zCGiQsrU)*=;M(3i244Ec^33KcMNXfel{Ctqe+GTGVaMQD z3yc)g1qp*!isF$R(+t0aLh!XU!{CHLvxeL8lvauy=)ql?JmS*+W?$5eAiffgOQnz*-3| z!l1*`BzFDn3+!zVcfz3e`M@^(RA4g&7hzB_2HWtf5x<2PheflWJz>y)Kzil-9^geG zOdu;%81yS>x8Wawq%r!Y6=A}F%?uL;vBhquHP7m;CmXBUNY@m>EE}9!=2^`@Y`vzK zdi_%@(nyJQ-&a+>VW3YzFw4}1fdMJO*yw{~+k}IcLrR$cidQjR5gLN+j34?u~Vpre^a(s^9YqWrikrM`Mu5gZLYMDaiK`@|Z zM|#5G>PlWatqio5V}uEV-bwJb+7@6}N6;-;h&2-kgDkfysWyng{XZ}!3dF4|l?#R=I#Tj41R_8U&kO zgu%onHF5rf_+rFNUx_ft^gX_vPy_Le9H!YSPZ)IhvApU8te@bV9AU!17@7!!uFT*C z!XSLP`w2rAm$(G(so{<2rB+zT*bAO8xO)*}a1rJH%CbM<2s&X<`$zm-NIyY*(w766 zUP6Ks23wP1XtDy!FZc@LA`E)+UU59YIzk8)22(z&q1Lgfx+54pg=LBnVQ?^aEz24Q z!c-xdvP2lP&6S{*0oyG2W@a2B47hyh*#BU16b9!q=1_ORevS}oaf}UE9D7827UzUP z?zvFCInbRY5Ilo3Y$FVQ#CI9(jzNujculMhS_1^r!$HD;!=<7&p^yESlR=zQ7}RNv zQR27-34c3yvflQVX!We_rX2|u*QL!jrW8>bT!-%(cb|5K^T3RhX{iP zS8J)a5rgmFf$>mSCLa+7zwT@lV5LPRvm=oOpNTI5wp?%#21_E- z8hkgP{Q{esu%O_DumRwhYXu5hVUVOvkT4)s(kOqW*ueSx%~vo|Ocx{!t~AF(yChPV zNAUe-hQSGg535kYH3ZPgf%+~PA`Hs!&!8T%J-QzlLql023|hR6p3~ETE^v%6VQ{D` z7@Gj@b_D&51&J`&SRTsZCxDlP5GoArPK~iOc?v>A+dx|axj13){aYC2>;UsPg12LE z8~x%5g9)P>zzWc?L@=9@9f|*W!l3TZy7s}}16UuyMHuY58Mx8^KCllx+zEp!V{6;> zdj+s{f{QSic@0*$AMwYCaac6_*%Jn<-o(k;HGq$Vke>Ao6$U%+*&F?7F!I?DOe?~K z0h<{v48AB>!w7@wZ&y+6k?suyvuto`8DYRbY`wBDed=em>KS1$AHN{?%>;cpf?1|c zzZDP$jE!zVwoTZQ!od7rgu#yfCDCl~enRlpDbs!<3`}KZZU@@~Ex>w>i*P9V1K_3@HJN>JM<3 zBj`IUM1;Y>B$d@)h{63oFqR0*6ePl6*QzR(wF`tpLNZy2FqnZidNx&G0pSKoPg#@* zgO_UrZiM>K1IZBrxDy6l(loRaH!rZlf{QRXd4gtu_}auwUx_eiu%7#qi0|ew%~pBB zpxmg6>Rn(H1?S`l69&f6L>MS8wJ#6`;mh3}Oo+H-EVKDM^RrGxN_SpJEFqke$f)fTEAL4zXEU>DAmnSa5;86o?fVTwLK?tG3 zp!vWIDxOW%gTWXpEK`gKgX{V8Th;;)RtU+ICBh)dkc{dpU`GW{V8$WBfXkPT{TC)j zVG#AdWGXp&Bnv{Q#W6NuaqJQ8S)3CFLz3b_T@kdp2%fnfDWPY`rYQrIz4bDPl z)qxEYVc@S-Np&T@4cKnMMHrkPmCxX30R1emsR;`TUI@DYj=30vyTsfvi0v383`mtU z%3pS1db?xLr87o~>4JnogJ0tD`&XoHjNp6I41*H}>-+KIURMBp9H{$}A;RFvlI-ej zwnt9{V@4=Tgu%d3Nx)bQbckM)rqgPZvR z4k|_idrxo?1`i?vZel+L_PK{UVNman>Q)$h4eXHMA`E7=$FcR#h`&jU!=l;Go-lZ& zK}GEJ0!)KJwS*7I3Ka$^_mnk9357u_MI=W@U23orAFbAd!GKXnzVdZIs)UP7= zR1|E&f>1(8HWz0OofC1kUJGENBY5VJ z5t)1D@Zgtp>JS^&56nSiA7y>S91gTEY30U8z-9?9=J4S6GbJ%z>a0Bf) zu+xHzIb?2&=hq*I|C<B2Gd8dgtMl;sTICinrvG(Rl09P_|^1`)A?4g&qua ztYd`f}{U;`~(=CW!kV(|M1FuoI(DM)nkgwFLX>naHMgk-W3 zo$T+NRI#Z#1zMOE!8emdiB8UfUtax&DWHo2EA8PGagCiF+# z(R68zPM*;N3mJRC)5!~SG{ODnH?T4z_-=6oolY+OF>i9jBfgHq^g|MyPJXxx%fc(b zx(WV_xaj2i1+dFH1mJifgzDr4PoUB*C9frO_h^2%#3o*nq{cN3>^gPA4}lghRy+p#2ZQ zGdROGI(Zk4`0WPTrz3HIG!nE)2&RXFbTWqvS5q^J$B}0=VNq&Bl8=dFd&`+aL@jX%owm0IF2Z0;t-#HOZ-E{2a3>5#;TQa7{k{e4 zp5P)3u5W@~ONQ}EhhT<9v!6X-Fl$~hg0M73jy$B^O_%kRX(#2i}8 zE2nx9zX$A*;9?HrcV{tp8Wa(OV3V8Lv7q4E%|Yszf@ko{8tUYhAd5MWDruDe%xySA zp-ztL2XkP$Aakf*57#q>BlTDo*Ppc&ojmdeukFqSu-t)q7#U&?t(uirW7rnlu36 zWg*#IoH-O-k`%vx0NBqFJah1)YMwb%d^u5l$cD`Zb0OI?SRXNmrGLd(bJzxKx8Pz9 z|NI$k@w32w@o;AjrJmHW=I{jAbHT+Nihm7h7t=R|%8g(PdxR|NR^luo&oS#|X?`lLtN>_f8f({J#eHy(8e3jC*+V23oF#i7FB?THgTUfv`+LqLXLj zha)VqqWfcW5W!T{W+ghg;N`T6P1R*Vs7z8cixQoj`P&M1VzmI)#>1UXe&eNhJ2CqM zdq;55$#o85;(dVlkBOPS5}kZIbpxeWB7U>OH0$f>4&}nhIGu1FUc^A%w3bPmRh-?|1Pn8{!Zr^$INnMP3BLv@ij-b=YmHy+k zrOt@&?J&KQ1gDeJyo^1m(ZHq%{xxyY$$!7r1ovM7t`$P4PVRiMm^#F!>ce216qYGQ zbn=|ljV$X92v3A$$`YNNyDePFqJ23LOgr>ZW*nlExqRu^*I{zh$wP8xQEk9}10mGn z7#pxS_K5Z@>aNxP+v~gEkYvGdROGIywDP=XdL;2pr{q4%$WpQ!{G@VvQ;)2ioLc zHU<}-dC1E?GJHqLa5D;K@x&B+GPLgUkycL2>{r>)=aUk6Yy(UbLZ~n}csGNsNf!|QCnTGT z69#?#*c})PaH1o4!r*08&l3il59d(X*su*?ZY4W6>m$OTOS^zDI05X8;35o8{ZPlQ zzxRRt?cq)seAX}U)1vh6;30ru^A};zuQJ~0OCmmw7>7l(pFLslXYcAzy#QYmLNV4i zR2UR#m)Zz}cR(6LB3ltA4A{(YVQ`^}_0*l~Km&CE>5e0qWrI`82m}6M>s7|oD?{rn zMi{926|m0@dKyeNvrOe%8W0AIjqXXdO*n?a!2Dl?LBU6OgqK9hY6#vsW!i5x&`f1z zZU@@~Ex@1q3Z5MN&03iuLBgQgN}QPtL5`yke6vD?!S43SF*iQ}u*89y9q9=JmB7=G z?LhZCMwl>o91p#D9^f@c(DPY{)z1(H{%2{`62xfS!^Eeb5KJv>K_U!_^(<;x1wkky zB$Jg0gM=8&LQd6&AT%dw1&b14@O7(#Rv7dI*3ZM8Fu3qf89Om20sBaB5e98GVw2}H z#IGS{`bvbsk@VQb-iP?34%2LvCk#fks;7Ph_CRn>jxb?h3{8Z=I%e(OxGj934=}XxS}=} z*mA)e6Bl7H>M-u>?*w>I2%*AY$>?0FC7Y_Rfboa0Ofe!1hOCLPtVq;9J%XvHHf4!0 z=smZ%Dg>;O;BA?4h%n&trDJ1Yaufzjr&Un>z#fGVYH^GWSR8vqdlu(}K}>C|s|!I} zhu|5UVH;r(x!ehZ-1}h%dqF#bU}_#D3^-gWsv;Wh-&>)u6$S@~VU##-LBilh3B1Yb zNFdP&zRetc5e8)*rv#WEKq&`mHr^8kAKW3MF3{$}*ugwR7<9!^%wELc{yP|hgk|!v zEC_?QeEt4_f6S;INJ-S zMwpcWZ<5OvHo*Q5Qa)qqpK5c`8E8u6@qPH6M_MGwj$j~tiMaD7+45cibqx_(R{$X) zl;pfgJ|3JF-xPJ29S;?C1ck*seh=Gv$=}nddW?WHaNZjj(cd8WlGZ`ZoHxmxzo$}X5c{h`bQ%K(yh$ER zUP3)4gxpL*9Ya{WNtU>jS7i_^J24gYHWD~*l2fh(d6WDP%#x;MCCL$Ql20EcRqcUw zVNQ8W`2lZ|Y5P@CgAhB~A-cE+PxeK})mJkCEf%<}2T%14C{SN*2DC@u${zf&@4v@M z)k#3Vg}~ZmVZBN2!pGh2J2Wssoj1v^(6T7DNW4k*J=;)KLji?Hp_vH2+AK}HNmh@@ zs95`YfEp8SN?5!}n)%9j@g`aSiIY!w1gC$I0A5^2UT+hr?gD!rg4;~Qo8+?}p&VjHWB&)iR6uw1 zN`W`YQ&$sJ1z-t+oBkJXl4;^As+WOv6P)RTs|s(D{cGh=?}9KNH_4t6Db%09G)&L7jVZ#sNw(9t7m*o^ zyuvd5Xck!-wr9&xT*>^e$5*i4BwKxm=S1viopLM6Fn_Qj<`1^P{J}Y7{z&8gcnyCP zW=28YBva4HZ8Xs$R3XfpWT9jB!sre{#(|CJO>#`8N+y|jlQhi}Z<2EkB&hF^EX7zP zY2&;r-`pc5wNOI1AOqR4CgkwTtojga9^9JFl`5M4S zLXc7Q97(3BgZYvQ6M&Pzc7Vx797+C+CxhomvTvbMhN-KAUdJWz97+D_vEfK^`)B10 zN%sbG0NFD*?&3&t@Gxa_n*waQDK_;DCxzKWXq`-gk*yn;xnd=l0>d6 zT)~j^c`z?aZsJI?jmL%~$!4o67?O^94{yi_wr=7`vQdSiHn$?cN)b24DUKvxIc5vd ziNKnA>766V;r~Y1y7vIq$HSc?$xH>|W($?n?*seL!{26W#w3Y6h@W<|5qbr%bsp{< zNfv37%i@QD{ovuwk>qd7tC}@H-voBo!<{3^^>^(Wp`+eM`w?vW#gXL2ym%lNKzwOp zY`a+$JV%m0RX|i-fUQCZnyZFN?Hoxi{TDY{dV$g3u{=kTS#aIJOv%Z>rV%&OLmWx| z5*zq9*J@xJ1b2=ke>n!LJc#HMguEJxBgxrjG7hepSH-4H8kzN%e1@r?f*vt3G>LO0 zxy(xfN0Koe>l&sm0eWR7VFd&I;2cTbI|t7JO#rqHB{)Zt`?e;v<@N_Yj2y43&XMFV z$8y?&W&&RvnvXb=?7AeY&1eS*dtFA#;ysc^_D)v7W^x|bWtWNPNHX|f#o4?hr>CN+}Sn1;!BIJ0<4CY-Z_$7Sv!+@PP`qkPLf_6Nxs?z z3(_FOk9KJTbHzE5Ja8UQl1~6G5yB!fnHn66Q?Unecx>WcRPYOlHB#J-!3I* z0Nrq~HWp@Pq&SkCo2G_k`X=MmAHj}?VVdH^k!0l=*(|d-SpHC^StP}gWZwx%EVC6@ zJwut^qAQLhe?L{s(#L^1HH5Cjk>tKi5tg(PgzX_D(=5-CWYM&^UVRLV^TOhD#&aZj zbr_z&e*%2$2%aO!TJ>UaqBR9q{}H^Y85kPpNb*!SfaSob7Q*rzNnW~!uVKCfu&WR# z9s}7pN0PJZV5{<7Fg_HP5g%S3rFM=a&lbw9e9J)lLa1aZaU|LMjCC{X`xb;_j^sI# zeC=5s(|^8Oz@9qX&Imis&XMHe&x=}S#t(3ufbjp#i0hZFBgy~m%4CIAHL#kHDGP!Q zbtIVyU$^(vaU?n9&!m( zuod19{FrdqmEuV9hticS=O*wcAslUp;z)AV(ZVVwTbPkRp^hZm{MSg81EE$Z z$*3uLK`>4s%`0(KJn9=)K4X646tW4Pyv`}4Ss|QL$TyJNi=9G_1|m)&-1jxfdO z+<*!S?3_X_-Co?n)dAHJm`))nja3xag~Tay6I9}gGG&y1N^@*i@}%lExbn+~83Th) zs0gPc^2_`4`Rgav#fux*0NN!6xKjSYSR&>4Zvps=9eo%!~F zV>X_tYEP;%eH3(3MXIDx{<}{z*psS%rsFM*>4HwGI#kCQW(}mShv3`C+R91Q$b7ZX zw;cfVaG*X)hTVe4N!7G98PrL(M~?;L!%&u-R89E=YrS3$biHF(Ic*Q3;QcV>7_QSE z1bD&`^f?wJCsoJmViw#0_&^AuCsmD}=e9M8o`$0z1k;v4F7BjiLJS^TB>|Rq1n;Ek zIr_yrscM5Af=g`Jt6+8}`x@&bCsnEPm$DoFLxGJJTu!R`J_~$gX&$goJ=~pCZR!~K z<-<;3UkffLRfRvpt%x&-|CJbrMYEs1ld9|gVXc1xFbace34f9mdQx>HK7%=_$_r8n z64{EdlPWed{G=*&PbzcOyT^dWsxQ(FK`{H+oLc6jihtN*ADenrd^No}sTzJ3^;!Y? zHUzWFkaJ1kq>8c8W$L0do3K4ks?7i8r0TChI5Ri{-VFqAu? zcy9js7bjKcX9S&89oU1d@;{Z-|1RW>a=e)x5%Y+Q7y4gNo1;&@cG6l&=)vP8}E$bKv=Y?dl zl9MWb{YHvS)ek{<9teom36D6mYUp0az8m<)ms|W#ru) z@vVuOzLJxw&wAmSVQ<90<1o!uc_&rRKS-*k0Gls3Cr8*xl`%9qscOg!UT{(szT9av zHSW!i%9xX?k+ZOnu@}6Psytiq?Z2p*c>h80`8dnmNmc4xyuwld@ueMxuLOeNPO9E$ zfiQq}8PJI<_)Qjwe+QHYHKn(IdqWLM@K50gGdgXwTx@N!7Rw zIN(bJtu2CQaE5J8s{Z-Rom6FQj|0D+puLS?df1wQSfh%%gmL$GF91hlyu#A&6O0na zE$F1Gb|;=1u15}Am`ihxzMNEbjm9MzeH6f12WmFnJE_Wjpbi-KfIbsOYvv&*RS&Dx zRqYXj@1LOfJP4)`n~$7StsT@T!19ApPgo`|IjQ=19`>HlB;5&w*GcNavVs~s4z-B+ zxG3(nzS2D+)m%{Qb&=O2GWYz{XPNv1E)s9zOo{5D=ua>N8Os<>(_3#qNa?gcS zNu&Jt`{lOkA$}f)lj(xg!>H-FbTk5~$FaCotgWbr_v1MV^8kG2K>Z~dq8@&#QA6!y zd-T^}d>_gZ_3$UU9N)hMdeA_)2CEmq{*K_OhpMQWryl-0iXV!xVROJ-MD|hEN7Ta~zooP4 zVLPxrf{S`se$MWJ>vO;^dAL&#BY(_n7x|~a)LfUpsE3TXvC)?a@p*`GSTy_DQxEqi z`joE{z$TJRle(>KSnTXS=LVj>VdJ*S@DJ&NLY#L!Tev;!?Q6s$IgUl zS`fio%uM_355-uR%j^2vQGwXW-YxU65lB1m6!UxMDf=(C{W? znjQ&Yk^?n6(o+v7D{{yDQ=n@cBTPL!n1!e6K7dCZLI1=;teHSPlzC85oktA(zk%_W zuuMUs9+o99Z&|75r%+iDOjT`Gq8`#*NUGRWT@Hjek}k6-Q4hl_*0&R@C9t+0?$kr% zn!v;y0Bo?}q8@gnz!u?$h@U~s^p&WG^;K{(unO^CI!v=wo_aX&yI*|=?7ZNd9AWCg z7@DYu>&)N<>LGl&tBy{K`vo83;s+JJT8M>=y@1IQxBy%GFfLHk!>Bh#@ZI7FI`y!1 z6soPeApU<2(+^2->Y-I9D339~J`nsFaZwK|a^Z@{Qh@7(5UL&;O#%TkTkCJZ_{n22 z8&MCpPvHqd)?E;u3dxiu>Y?X=A}S5qmkYtPLq{@6hfzrGlBt4<305dwG-EJ zbV;Cb!g#Eo5_#SruS~L>Q!as#R&? zDUe561dEF>81ZuzgBJl*N?=nH78JY?ZbI%cIo1S(L2r;n7?3Jy@KNc?aAwJ0#da1Cg$M)#+U|b7ji7*(w z5@w_S1)BWRKp|nmV0s+pXHI}c96>i`K_U$L=Ea1n4zPg`LWMz!cj9eLx`NP0NH!NI z48~TC$NL|^NsizNgI1`XCk)2atfE@7VH?5RMs{1)M}$G1l=jSA{|M}?;35p(Y8CKZ z{{Yw%4|l?#@%x$V`ki439s&q9e-Q@LPu5qu6yoEFaac6_*%JoY(iT^~W&k@1p(|OT z!eGFb!bTVj25BsbY(-DCoSIN#% zMi~6Pq^9yIOtiEJW|_LXC?E_N8@-fln=l=Pf%(4(gPnIU%%#Aqj^M3Rru{}3n99oB z4z>qcfWJbsf>s!mT^=M1=3T+VZYXjbjo|BP79%GNnnmD1OV0+d)Pb5E=?Q}^#qrjl zcK|)$7-7PoP2@9p~2HApQr3X|~D}27~(MRlfoIOK?t(FkxT}O@zT1X7B=G5Wd{CLD$Be zeJ9EYgVUd3A!9Fi!XV{ytT7!i>fI51137|D7>sN}VK5Z&6C9>TlHi2FywdnJ&^%zD z2|k{<2!r2F6jyo|z(Ya^6$YahHBgh;RQ)R$cZ6k%5n(WOQtEK8Oj6?Fk!;m_0}!3u*{R|E-z zg+qgdLFDb4Y9)st2{=&*zAxA?5eDs>XHYwc=L1$ma1jP2(p$VHp!x!vny{eYh46b+ zH0GySI0NA?W5$Cl!hlpsgD)+W!v`5C40^4`NHJZIFxc}iG}wBi-ooPcnqhFiARm(n z3RfQmaMpqP2pJ*_UTT*?onU+PUtl~BWr;8-VZ1_RT8Z~x1XD`RODhBg4Z2q9D$G|6At)?^q6?+MA~;)KD8PQnpvdb@rXj)?qpRcX6^7Y9~W za1jQ7{EeRo*GGIyVjLFDe)fbxiq<&8?E!GG5blx{Dhz6Gh%~}r8c6d=WGlji0h<{v z4Ay1Ar)ltMnFD)^t4m1tJAzp@IJJy0;2*YL4^6$U4YdBBW{-@*kKHiQiXfO}>ej-5 zFkozSsro3*Cj5OBgn{|L2!oz$u`yI1yfz5lI%V2#gn_B7%HRsskOg=Df4Veo0Pnu<-;-9YF`QdSlv z!r+Gy#q7iy2W*muJ7LfOj$Vy7k!8SE2`<9m?J+p<*^Bst#7tj_FnBLRK79We@i!c% z*{UF6P_vBs7g#iU-=x)?9AU!17@7!!T+HAF!XSLP>xb@*%kWDUBMk1X!$QVh@Pxr@ z%djgn80C&&*^0Bw34>Zq(Ih!I_uwBZbYt_6VMixCnzf_iRn(ID=*70in2%Oj#lfI{7Q88o-(ho`V^O2m>x( zI`#%kj>4ej?JQ~%*mDs=Esn7Pi(`*y&*GdgXtESn(YAqh7{N0*!#2Vo{aPmsh9y9> zp9k$4f@xlmFyL^hsAg!q|K8{2?T*3aFEC0Rw;*9q`(Nlzod-xE1YZG;z6gUXBVqlz z8h}IxYBt^z27jNxNa>D1dkUi%^AKTh7(WpxgBY#HgYl8DOgy zA`EWktf<%|eH?_-BvoQrvi#s%I_L_2iTaQL+%XuqK1diW7#1uH2A?UcDsTv5!O4r@ ztHOqfFc{IXxT;Is53Hu(A`Bj!h&OmUK%E3OHDN))3*j|XG-m&Ed~=(^U?IpN3`mtU z%6~f|wG{^AHesZgE=U-BzO$jy-y-!f7T3rOgA)ezw@_+d2XNnkx-}Uh4Dx3yrrNVT zI(h@n{}D_UmL!>?7-7PozzN(CXa=yIBj_$HNQA-PHSq@B58zNCgbIT$ zxNK=^G82TwLbACyVUVI9P8_!Y+~o+KFnAA5^n}6n($!QCHtZ^xzmwgE^$}qZ_bl)U z#)yruMg)tCFev>|5xf2t09MSy-Ht(~evx+lt^=&0;35njJ;7E`C&c$6#$nNHohJ;6 zt;`HxYXGMS;a##qg~4Bw;*Bs^1=5!!vK3*%fXxgS2AA=50SbfXCvvDKNS72d$1EG1 zT1FV~4_mL1re3>dS2sHbFPDX2DGPc{1hY(CT@VlkjE&w-woSO7!od7rghA0?ag@*r zynYDYI%V2#gn_B7%>I&F7*yPW<^LSwFA+0+CBh)C8D10~Azq_*Z9Ca2PZ*S$+E`@*R#0$Gjxb?h z3{8Z=4rcHIVGzFDO+??vZJ21kITy7R3mJRC69&DW;R!bl<$l7lCvgOwF!-wlm!GwW z-{vqqlLRLWI!(k|$5CKs1)oP;guyq5u*ly9_*4j?!rIlh{CBmTo#d_)$U~dZkIWrCs23)>$>~5GGg~91bN-YI@6GEuPF*aav z>=Er*oD&A!zQQZ`3DB+}cm`+KMi@kX;e^5P|0OE@7iiBBO!I<-0f$RP^+5Od=cmUf zPI;8@l%vFPQ`$cJbNkYDoK9g74x0jp;c$xDR6u6<<{o1|}naF3)PS(Yq6 zDk?Q1{qJokYK6fE+k%8a{}I8$ph?vdY7d8?0ytF=du&D_P3SJ1GqfRj~KjKM8VXzfs5eB468s*=)h#wfhj=_SR7%8R;5(dS8 zg+cv_)Hhk&X)_E?7`$DO7Nxdhz9AS;Um-(;LA#Cx)D5;r7X+hBC`*LF;U<_;x<1gB zju9pd79A@I|IYy5as+*c1&J^y`))kG{|0b|5JH7PecX01HCYG3b|KkZoG=)f0aZN) z@RTEX!eBAl?+Jq^`%0<5*s%Y=Ou7TDf6DrZFeqC#(yp&Lf#nxmghBq4fp0Tc16JF^ zoiJ#aCA(d}I{@n{xCn#XPw+9CcM(617>7l(b)GQTgwLz_<^o(Lgos42LWRMJtaiuX z8<2h=k*x?525e@yFv#90yAcK-m#w4HW7gzAFv|v~mJtU0!`3S$idCtzwW?)=!KO4R zm9HV_?GVf|b!mP;7%(=v4%s&0LJ9-(e-Q>(^Wj$!?}9fG!CR+H`;9O#m6f?2Y!9>m zf5GfHgf{-4zX}otkJl8!{a57pJ%X>gS&W=8=&R-w9fQK}WKj*xYSkEwHo`Imi7@EAqK0Ml2Vt0yOjaTc zIwm$$Y^t6G!hDjNvnUY;FI5gW7T*GFmxnuHP$*qM(Vhl&QE(9k*SEmtA0Yk-G1FHf z4004nrgR$gY&HagX|~D}2IaTqQ>B1a7o3wLOc)qL6JgMr8N5IkgfDkXFgkI6=E!Dt z4D#;5LdIV3gu&iII4)X+a@Vu$OdLTc3{K$3aq$0)_!AD(c}Q@=V9XH+gX_TV3tou0 z2!kE9@G6^pHd5KvyXTwAo)J{lHt%?5+*js{&Flcu-hry=+nl7-Z2@48d2(ti= zxw;eMN?~vu{}W+As-#i=VqXRB7-abdBgJ$!pO!KJ}CdASMjt|NHDU_1K269(^%ET{&uVVU>g zv54Rs!up6XXuTxxQ_M=hstPW`z`r_hIk7dc_8#tp!AHO2%M|#1j2;ARnBXD|-c5&} zuuVt&0%9B%&DMFspx57V_-Hb~Z-g+4tWaUF>YpTL$KWDJH%MeF!h`{v872&5@&r#f zPO-#$J=5SLdzeT5eJSvipgDYr|A{#?3^9l9`)a8PEU`T}T@ZXzSeTf@tBdQYPlyi# zHb!tUhwOh>G59<{p9*Yh$AW@uR|s&-L|#&-Ib6d3#2iSKG|IoBFu%|VoqXUR%z^2G z%%SV#4EX)~euw}B-vZWFuG~~!&6Bml0LnQ~uOLIrVZrM9Y8~67n}E?alqKfy+jm95 z=nHg+V}zN*@*~(d`3T@VN6?#DkeEY!XPozL0JuvCq2{o2Z#7$!(;!?ClFh}LLxD8d ze*YWbe~#dp!(sG`XAb|hDX(_0VFeFh6CA;}m-P{I=(0QDJhCpZMuLkul&@}oo2a`3 z>*e9j9BO?~#+t);V3P$GbLe{=8-|}Eeho1Wi)KH2=I}H#bn-rcKMLUpS)t}|R;4uN za2uq@B(fD@=D=o#nL}*WuVRc&epWP#ipQ*}gJ7PboLWXF^AB6E6Q*9p)|WLp`BcwH zf)I-}%q@Y<4>N&Aq%1mg*4h{~;Jqvm-s7Ja1rbFiHZA zbBr*ZTzMj_uqnV-96?`ZA=XTwlV9&pMO{aX)&s#9DJ)Zv=;V7Fb6VCM5S9taWFEvZCimKMYUKgB`BTOe7Lld3+fEm0%Cx9)Y$rcsn-4FO@ikW5*klgrgE zq~-!!CHPZj9HNuCeCgN{FgfbviTJS6Rj?l-gjyV90~W^~(VoRQot!=s-T*S7JM$rU z24~ntC;z$8>Es4i@cgR)S^|RUr68Tm;Zjj6(ZBwV72tW8I=RXbj1tE!NGF%>#v5)! zki#hE@`R%=IywIVn#d;rmN-ze@t#gD=*K~e-VSuXFrG6HtAOFO;rq?WRD{vVm%#X4 zSSBCK!j+q|i;4wU$xz932&O99f2n%o27nUHKQP8hUrjMXFEQ9J|?Y-8;}!X5e* z^ou79dd+R3>N30PVAdhK5$hwu;Ok`p=aH`g>n6AegTrkCP8~)78|UFp7*x!vtuR;! zY?T_JS<>@L7i^(R7<4mfMAvlPAwx0_=l}mdsDAGSrd&gDDXj1<(mxpJOr~$othgE z28@lKMYc`YlET3JUxdNB@_1j`0p52A-a2L4Z-jxVtjz6Td!PmQM?R}>g~3lz zUYI<|k0A!ZH^?kTP8cL?W_?e{lS(FHa zJnI7!>u+HHdAJh>>StUd;N;4J9?FSe(~B@j-2)ry>Fv$KG zXD+u(*sCw!k~If2*<*}$_YN0xCn#3#c%?U2=FB# zgbIWCeQT-_Y^v@9#=F8Y#fUJt)4r5teGI}vA(^s77!=x2No@l5jo{;$afmSB@}*<* z!Q?0mR&7tK{sTKTdL`827#pxS_K5Z@&IyBXK;pXgu%7VP8bZ&fJ3n+ zptVIXy%;16I9&M63i{r^_;hhA3|=~kQR27-34^R);hQ5dBEDM%QQ zDruB|VW);x82o)2BgJ$!a*LQ#o7hHrv?8U(P`w6fG9`1y}+BQ|JFxU$0E5St=q(6}gU;9A(FT^-3n*Hnv zgCu|CQoe@(BQU6zaD}W;VeovS-w1=8AQd5ztq2naY-YGH*q6Mb5eCV#7g4>CZV-Z5 zHaNA6FyJ4y*c-N9>uMTdkoGKoy}Jzb%?M_hIzA^L3>X`ow;@Wi2}@8InE#6~Se75k z?Id`=A$aSQX}=K$rm`}(gYAJ9;O{XB_W*gacKU3PFo-CgTINh-jiL>OF}9&q!R4n2?=A%Ht!aQ|Uo zVwM3`NpKMc35Q|xjS=6HnCUAK1}EdNyVwKqZ#hh}Rh}?NUbvxpAJ}ZcIXS|FfiW}@ z2E~}c3xq-Va`!JLMBI0!Ga6wq^gI?a_JSu2qO;&sCgL3C9D?rwN6-m_>UH?M$%FXf z4%7dU;Do`JGI{X*BVdgLPs(X3!azMnNu2@q7DA{nXgIyLN`WL=zX!&2VVPn?7|d!G zWm#)L*eWDbmI#CP8}q8~fn5|lEi(=g23)>$Y)hCNg~8QEX;fzPNFjt!i(_oS;@BhF zvp6RV*5Yb}t^?Z32%fV*NW^@CkUk9x}g6ZKPVZh{ZQHgno zFt{15R8_=iogT&KKrn^ad_)-JPm>T}RR*J$uuNVe45}AQuGl2q9)vC=)nZw){HUl= zsD*#yaJ)HC80d>Z!eBRg#M?1QULe{3W9-cXbgKS8@N@2+=ehHkXNF}QHW&CQYci&mMD7?A@O^?-{-mKKF@sn{C>Yb z4$qv|>-|3axvz8Xx%Y|+)Q<1N%}3-al?tYXLD8ysAr;|!fgRMi76!$CY|i*^fUaqn zb3zFPPK3_@j((vxF6$%=l8`68V?b1GBQlrW#*J#)F?i+-T8iufgu(acaGJLrtUDnX z8QcuEFsRm%_?!J?+nCzVAOToSLh0QW}I5O;TQLVNm*)GWh;8z~(l= z5e89k?g)eT>y#3YQ^AITIf~frDL+~mG|Kcpz56P#*EFt$LH&9mN;rH1>~jaVg~40L z3o2po9k3rYu7$yy&t>96$cVp77&VKedUk}tV^y(bAAtsqM^J<=oZq1DKfgN>`w>!z z+8{L`B9$VaFrZ@Q69&5T1a>&;V$HYU9UY!g@RXTU2BVLLL!q+)!(ms@aOm_$hUi5p zeh!PRlzx9oOb>@w&SZ&a3I7q;DUIvlaISj}<97f(&@ktY5(>=S1i;Z9ieev}hC>GO zq=y4hwT;N^u@x^ZqTx{TJca|=1q_EPFJt!W4eJ32#u&=2-avb+B^H))B7j*oD5nrZ z4~ONR`eG)PN4^inr@<^e99o~oGd*%Y&|@|u-*7lP5{DA60K8)p23>oB4*rTOZU_IjIUlrY0#b_Hekp68GgF2iV>wIKv?kDLcbqfG0z|P6Zne<`iNt zq5SCKP~YcO-TfV4D>bf%!$&<$g?|NXkAvI8VbAPjH5|?YyQpzJ9L8_LshJ0ew{Ry! zm6z(-84k7E16&} z>*R_Pilf-zdC^s!%QkgO^_fS`tnS-^yWfbc#5FTj~yV5$j`dgI+6;+l%2@?>E?Ua<($=7fAb#gke zYzMb>a;KCO)iFB&>#A|Blcz;O$UKYqF@$klX`S3J5nA+B#J_1{l1k<1P#+pTy=x5>1n%rB$&*}!o0L@}w zS|`U0j}TNOIT?f*L^?xh>FEbgr6Zr2pTE>X34@Bi2MB|DF9qu4PrP-+4QhhJaPtGX zx=k6F`Qe+n(43babDC$^PPom`9$eSY!+Lgn74!Ib>N*k1s#L&WEUwypR znaU%#gYiu;OACYj_h@9E2YTIRHqeS+3lIjShhjgz2RuHFVC3>-WDA2XxA4{g zIS#;d8Oi5{B&N!S0jFdjY%q%BMdI=&J_EBozOUSj(oy^ z2bvZJJ;}j;2!s67T^1@Wed_noEDW|@$3#Z8;0S}fcd>|Rf>v*VU^J!{w1vT`ziHE^ z2jU0Wn0$-~wlG-H1+(ZxV6!ydj&LmuW^Bcwisb-5(u81PaJ!5tI#E&OJ}{1I7AK>H z!I|PN#kvl{Jx$`Yv@lpzK0!pFeDMfe4%veo2MGh3zVKKyh8ziljDcyQJ=jkp1dlk1 z^^Z8J5$TLLTNw23iIPnLZ6Sg);F!(Apzjy9F!;L%wol&&?NbEKS%5I0=EBiblqqxd zk4b6+?V}rLC2F?-VX)`8z(;@J{u}*~o93GN304!wU~-boW>Dhg7)o~)wvxmR#NM|N zxsc#R2IF;I(Qg+K1>q_Nfi2~mgwK;7;BlwiteBF9ODsc7b>LaSoFt~ycfIMZ$dq<~ zyK4$3llEt~LL#?W3}-3nMz|qzb6q`hV+2J^0e7M1Q;cNZT-muC9AlaB255hwkjdssOuIV$IhyO&a*q_;K@=U;I zHfuJ`6+eO-0)~7)!-)jz&un%mQCU2%v5JHV&nXnr{>*0k-vU0f+59hjf09d< zN*w)}%_Y4nie2!spFCCO^!=aNj4ziiP9ye`jmVk~Jkxmo&vfxGpinqwLs`dx=NP3< z<6A?3%4@hj!E+^hpKpvzXerWyFl8_7Piam=?md)i{VB~#uXsdLI2n$}(PY|^JnBzr zMs`UPl*74zUMIK>!TM90+$Skse@gT5jj>GJiO7B0Kz~XzVr?BE&mjH+VVsu!l;$$L zM(jSKUAO&^>_Ca>Pid}NQ%A%AE35JCL7&p>zC2W90Lux&l@t9b&5~QO8`uqu0h-m- zNd=$Mth>Ip7!PcQ#<`C5r!=RHC@0P zR}hy$xNVcvjG;fJ`5HQc^C`_OUmJqPfDc2bJOWoLc1i{9_11TK0zjj1n| zQek(){Q*kiJ*r>5=)L=5W3i6#3&5^uTrYY*9ax6(5F}9mL78*rD4{@+KO5H3Ltldo zrQPsmAnQdhQMHZ8-1e-gcEeNeij$|FAmV|y@5FM7Yc6A8voK+oF@)!J%Ji!a7Z`ViO6djLc3`g0)< zQ9^pryYRb;LdF0rs|mr2-oiH%R8F!%$k8O_#a{GYOTtpJ8^As`!CCZHMe5F?xAs&i zj#0s00dqdFPf~vLqPNGKCTdap5wP_d*NfgZ-y|yhAh4qjZZCQZ{8>${O|Ap`OXGUc zyX0(LT*&XCi4nM2B-OLC=>6+zU11~vtfL7RDZjys-uCw!@}l<%kh&6)N|A5TOU2B; z=sl5JkQcqZr$vipuv>$`l9;*{FM8>R%GXuS*G7RSQ?gyg)$i~w`jenvLEypUwHLh< z8(zB=Qd5Rs(4v?B){EX3K6azn=%z6Ujs~Z0sTRGQS>3k-#{-Rk%ni-yWrbMuhCc{c z^j#C?yw2+8YQBJ2^PKU?$;Ez5F_O&Fn-r8PDn3$s~xMWSjIo7CkBp) zR8(0WgcKs(qonkr_njk+RL5!xtfhn7i{9R!MyQV27uX<;>qYORmC1PjIpU`g#&x9^ zz0=oW27DXwt8GkDshmac6N|9}1niK;sdMC8^zuN{i{3}%;6E0<`KP<;sI>I{OIz@w zciq33$fy<|+WhZ)U(gLhyAE2tK7#Q#wV=J|y;zDCT2CUrhmA>tx}m-3t$5#r{slHs zMenJl=mM_;T&@Yhi{6gKl0*bdr2G<$eVWC|=tXarvXvC;JP6k{iPO@H-eEV& z2ovQig23gFQRFyi(M!{p^liqFqebtctEr+T*qsrAM;yiaM;z6NbVi)L=$$zWU3Uy< zGZ35s$828oKHOq2deb_!?0dM+jR1m#u>gMYTt zHqt5p>upeWC59FT8^;==Ho0^>w5OACXFt#DPBH-SE|8To|4(ypl6!gyLT4uNwa z2T(#<7<7DuK2sfFmL>!XgHA7{sGPJ1p{FJ(FSan4GznLM(E!KU1Vn>@@FcFTR z{A*!Q$v{pM5uZvJHH)Nrc7(yRM=A)T5x^%jVIr}Dg+aUL3>F51KzfddREm7UfQp%4 z7_`k&&zTfREHA!*-5vyue@D~g~9h(crf-aqNB6%G-bW!-wmhO z@Eycfh6_j-@ZVY(^l6%dV#BgJf-_HX`T000XIA&^!0|vMAamOKVoDg43keeueBrL; zj(l>2HATdTIE`xn{7ruVNiV=gvVik zCvAe9Mu}))(D2!YVm4x=yaR?b{kc#HX<-msD@w6qK`5t5>`DuRm|IW9AU7g z1#Xu;fmZK`V9cWyw1q+CiF6WTDB{Q1m|Q{xTNwO^PjtySz!qzK1>ss4RQ?cKd7l8> zq6ztg!Lo8nR76uj8))R6*Xe$t$0mp0>2DiS{!eBcR z{S3722%Pf(VL;6#JV_{5=8$Wo-f*0BiyN&(?G_*mcCEs*zVZQi0BS6y*4Mk?uM8>% zumpfIHYlm^jxad(9qpIb0op_}-XkA8)T-e&`QzQ|h&718`#-=Ks9Eesv2dIGopxS7 z>t!(JY8Ly_!XUYQQ9(tLAA+!sNS{(#dioKb!zf5*$C;pzFnG*D&adAVqRM46{IThQ z!l2`YLSip9!QXK6FS$BM1=GTy`Ti(zobY1s5r?2~Eer;>OJKYfpiB*OPAH+kiEul- zN2g;imxaOeAZuYjRBaddvt!TTLH8b*#!uLDJ5_s;aymNK*^oqX0VTwj;e@* z0dfa`Z*5RsCWaOU&&Jjie^7bk?_k^tW@%xthb9zRAQTx!;6(BXgC6_K2$=-1mQ9d% zC?PEjnsmhqx&^=|H6d6Sym+{-%E>SgUeF}v#TEvYwnK2v1^Bv6aD+imWZn@5-9Ks} z9#Fx)0&@?sjXdN>3xfwo{5u9`fnC(N76wsx@i9N=Cp=-ojiBt+{Ec_MkLmCKW=Bb2 zpdTt>294juQD?aq=n$i5LXVg-S>ZgW7HCD^?8ff5FzfsJ!;TNp%dZlyZrB4BT8TnmFet#MRr1LD6RjO$7ZgQbsS z>HQtzf3`76rE-M9&cboxCNPQGS0$v*kxv-#K-0qDQ*!Vh!XW>2*Be!vK0ld0K-M?x z*KkZ^R11zU80JmIK`^8{oYIb`7PN)I)F0>zBGV8*-^OG`BG|$pnNDS`0=8b`l?m6v zpe{b`B=-Y6rU}8qpm=CWQIm=){{Z73&EjOVF!-oaE5(XN{!1coJ|(B6g~4>$RMZ64 zSmSlbagZ>e=}Y=5VaSm%NLpV*j01Z%Lhy*ASpSHl8j;S3vxUJA>&pqb2DGgR&VXY! z3xjtjl@g-W!_Jk&%T=(qdZ76wGsHh9-o6P!9BVKBNdT8iufgn{erTp_=N^)5=TKR1Id z3%vnD7Y9>v@LkO5uAC7%g@4qGpqY{;CP@BklAPhUYA3{pnX(; zFxYk;UYEh+`v^t?uTpGbF#j)9h};HXzYR*NNJki)lytM<9MG#aBcCvEi6maeQ~;y8W^qDV7{t%XRIHXDwAUndS%--t7))laCHNKf}EeyVT5nDp% z0bbXHU}5miZc}_kMU@_8Eee6NppwzT;9kQ7#YzPsU6VL1EexV=mlLgkb;&rg!M+hEq6zd;xR3p+EakelRmw>|sdqDdE!5MJOW?}IASN4uU zcwub3{0`bJ1TJ2HFrelVp3bPA%(a{8kRyb_^`dAcYPSGkkUE~$nyEl)AQ*M2^|dhQ zyn(iMasjlnK}m&ogu$2%=z($&(C0LxA^Fh4;JaQKqB&xuoCn4d&0;@V7`%AhW^Dpv zmu9gqEey6a_6aJIJPpFHM9QPI^z#rLv_?C%0_O~NU9GgP-8@dA>XfyJuhnIZG zIR6T8mQ9e~P(oTgSZ#4*WjVl)G$B|$l>4Wa%E?|3j%bqdVylORpTo-~fH!P{qaHS+ z4jlC`^g@XEo(fj11Xjoh#!r+VtsdIP_&>Z+8(3Y9YxQuYhD%M-PXg=g;P(B7vG*z| z^)L$9iyGJJVcjR#mz#(9C4^D4NUCQ?J;eN+iMK!i+@%TUi509K%1fsbfzY~#4 zkxxBPG4rd3s`o=#Jv=tPfk;BHsfECE3w1445A;Lj>k{Yd<*(I2gnRYkh0zQ25ePh& zm0sXi4-^|-<8h>>44adB;J>wc=z9b@a~^oh5uC}4%g^e8GpqY{;CP@Bkh!+MSE+}E zz5w-*H?*FRXW{W8g7L))nz3y4a5o$KRq|f|p|O5YQbjuIA>>vl7zsct*^GSZA#WSj zO7#Hd+61|s5>dkh>Y>++a^h>m;Q2o=hG`Zjq}9W!#Kwv>1B8W|#ICe@cz>EHsHpN| z5I!T)x0IAt4^KSnfBg9{upb=URu8qNq^gej2e3OD*XrR?I19KLHcN%I!8YBzynRIha=?RKh#71>23>JC_Nl6AtUurCms_S z)q1s@21g zk(ol40+_4`!Rq09RFGy!>cKkm%5*%WM#A7LP_!^OxFAp% zd{f9LGN>K;hVxGQNh&! z)U`p`i5OZK4E-Zj^q}&{_F(i3W@%wCITb0&=YdYP8To|4^smt+76V*q6XesBkQN5v z$;iqUfM07uurO#FUrpuY90*r6NqMn_LBDyeu>LEBivWTutZGQS@bU=i#SsSo&MGVh zQ^Bf$S&P^sC_h>lyx!BJ=GVu7wbQs32EV;sRpCQ{jdXBZ7}PqQs)WHDU<);_g~6@W zc!m9ki2s~0Y8FZL>yD=iGR|JqnkQRO}m z4iTvzC8dQyyj5OxtV_UdIJhkg-s$XrEYE`)Du|%$wJ`X22qb!0#3vEPb)|*Dz2_U@ z{ZEK*VPld?CWA0XlQ=Cc4DRRP9kswVXnYMh z4iW}5eM#TP7;+>GLb^5(H^6>`5Io{2)<5E?Mx-<1Y+KXWdPRLprpb(!r<~N*nXC~fF9C}G2}xFgFZN^J^?XOUIyc~X0abF4B~FY z_*sRJ$yfx=iez6}7$oH83M!JU2|^try-aBZ6u1NYXSR>U+r&v290x@UgW5|1g~5sb zB}I)kXoB(JOd(ep_{$LnT^p7WjR}7T*h-CSVenlz3~F6CD57-XNIhROxd&IV;W zVrXGds&Y}$iOM5~f$>5xOACY5tFc`t=K)<}GyGF4OAtJ|keYxSOzQz|vk9^XC8UMH zldCZI9s_t*6M}_7?uP|cPVR%?s^~9^--|5_ezdT(iv?K9COA6=S5YsHF!-iXGx0PP ztT~u@#2!rf(ZZm9asTuB{eTV8xE2N#3RhRcU>dO54sHvB#jp7vU04Ndt;V%5I8hWM zU^n6q6GqJ93<8%TpD>_e<`)J_52v#* zSlKB~bcWs22s~|2*J5EnKUBWPalSgH6ybvi3qNfkjD?`DMBr(v?mWLRpxE$D#8!r{ zkTBrCwJ?x4;VXB8cMQRqr?~tq3^=p8ZwHPC8UdL@2b595U~h7OFlcoEW3+Hmm?(~5 z+*(ePk-cMZ0%xW1{bK+bHYlkg9bvHQPE{~k1MOrp@(F_tMX>Ak48Rv{g1kqGXkqYa zbXoBTF?jz!7|S(_6Vk%qY;2lhZ3SVkCb26m3|{%Ekf5TR4sOyF!l2afn&L}f2Q~g4ISvvAG<`{5Ck#0f217q6 zB|=dn#Swx>9L4%a9My<)Mw~4SiVergvKDAf5S#(WY!(Jr_t?TR8~1DQZ-D( zBXC*#UTk4-xqSvUuK{M*1VW`~Y^uh! zF!TX%Ytfw-yGiyEa3y!Ml#& z%u`%`76zPI-M0hB1C4;p;AER0}1JK3> zB~_#&46Z)y0i!?AXKhA4Vc@#~=`aJ}LYp8L1POy-X?4V676zYyu|=~uAuSB{5AZ0~ z4776ys3qUu<&z)Cr|EezHUuckU~9bok}u7$z* zBm>WXBEBKyrm0S`1S4BjIL{~-+W zPj?oIoBnA9ddYI%uqm}Lkx?x;!XVXvpea}rb0UHfO)Y2(gIR^~wr81y_*yn5OA^5r z2Az(hf8+vdr}1)xYhlpnV;n&p2ym1p1Pg;%e=}Y=DFyu%WoGVjNltPVEMF<{o6zd;xR3p+Eakelx zw7Ra4d7yPga0VQ+Ss0w#YYPLTODkjpv=N=FDd~p|<(j+NRuPr({99-mAo^iLP0cJo z)ew{^q^o zKDIgM*i0N>Rd^?u*9OG1%|DI}BQKhFfsp&?be&H7?Tv+ZzBzEA#Wpnvp}hGs$nYHV z(37cb*v2-rtFq^i;ljJcJd;y`%?I1&s1$}ans?9DWYcN3skkDC_Zuv{Yor_&8l-9>30j?G}ZQvr62{UXhqdmFIbJ#ANhR8(!@J!Br;F4(F_ zhOU1`4wz?vP+BR3aQ#Ph%PAzhUrFYo*QL66+^eB`Q_OXY_&*ReP#51Zm977V%Ca{WUKYs$`743>j9#)4r!~Xlp8ICw~90q z4pw2S8@83Kw+tgh;mOh@^+l!y>-Y=j#|y8jA-oNx`P#e$wy#V$H~luSMHJ)N(rmUQ zf$j6OJ>ys8z#)Bwx1BUYo3~&G18fIqVa6VpW?^8I_{+Ajh`KY#6yAQ)tg~FDyG)xY zsJAqm0wSxej@5`4S_$t9(u|l~mlHc>TggTpFf&9k{+u*(demdne{ECw96o`D91R^Z zN_eMAv&&@{JBZJ;i_1gNPnaal<88{ZhioEHF8n=Wnzx+r9+76(E2Y^%N85pjV(5@G z`vW42XKhnazQ8%O(kkv`y5In@xo`RGR-xX~IT#wUL5Cq-nj}gsqCy^%qP;HJaT>cw?oxYeEuRRVT<9 z{t7Q9&FRyU*tCsp>aUXto)W^FXqXR+rwZy7gKc{e#ZbIqUiCxM2;x~*K_v{cPzhzV z!uBenx~{|&mSLFp`>Ll$w>b_Os%4m0{Ll%8(lbxD9fwQu;IOS>cKAD- zp+Z^y5{f8>9y83Aekh3`l|zP_8D?)k)Px{vSrK)ube8ZAHO#8h3$erQjzfk97-m^N z^n&A%p`M0W*AKnsI4l;@KzOGZ=8=dfcDT-Q$k0T?+~tS9bsRD@)-ZqZL%%x?qd!GQ zeakRkdCA8P&1_v{h87v-cs~?RkSa1m3kRfYmkG|E7HY8Q*7^F{g_# zemtLl#&;X$3qOT2elMT@ezE8SzZvE^JkmzpqeMM_h4}N&_<6&uj92~=o=!N`ia-C1 zpE1l|*C@OL;Q=$lpZN5oSDNOA8X0UqI-h^W|252`h0im@1%`OFxlP4?NwxZK1+vsH1K zHwWX>FX*qoYBL2TyUYiGsA!KIt9kV+32y_JnVccnDyE^X8HTc4W^F)ZRg)l9GdBuK z;eEno9yk=uR*yMW3_a#De*{G20glx;Gg)~1xXgbJtD2c%Tls5-@gAs|WhL441KZTD znLdNM3-9wT^Y7(#IN9%l94IeWB(p}M{_d!Y{@+1v{3Q^@ke63nX7yGUX4zu+kRZ%P<+ERTCrWrf zbD33f5{m3!%V*E{I@HCSOt$|#pZ(=>nZmo@WmfpCC|_g0x9vp~Lwj82vg`^BU9(L^ z)L%vM35ZiJ^OJ(DIMvX`y0sZP?lQjwL>W#bh<=Eu^fHBo_lC>tlbod1KaLecS6t?B zKh(jo+EEf?PP)zCM^<605sno@|GLa(!%T)|J60uETf$q!Z5n^6c3JIMF;viP#wI8| zwcD|3S36dC%el=V?-${6opr1jD(yB?I<@3-{o`0&p*l=+n}-&uUR<<^t__B2xXlxQ zs1m9Yq-rCBY9q&O9^DheOTHG46+=zj=2<_~*Rk43wb99K7T!~gttL8F47GQgU6-j_ zo^Lr;Z}fuIV7K|jlY zwX_|c&z|vRZnOM;wZNUH?Ntq@UBs1nv)e5CPeE1cAbbBpg-M^e&2N9M$eB15it5VJVY%8_ik7E3~+dLjpgiWW}rha_kM=8P^Vwo$)s^#x0ZLd~d zj0?-$xKu5H_i1~DciM>=sifh{nfS?Q4AHf%=UmN*N<#d5!K-;##;r;{Jwhw zwu))4%gRt0%RCK;tZEWOKSWBYZ*aCH;js6~9Bfo|!au@H>s6e@-ElHP;s{$GL}BC z3Er3!_gvV%7TE1u&>9Daf9OMD;pU9OfU5hl<*e&7$9cp>;*v zT&u~lRP0B72!nM9Mph*_Ey~zDt3yRR#q>3zzqcVtzeGH}`Z;c9_$d4#i;uh8NW?8F z+^0ls(Hz$DE@B_r213Da6z;yNFuRe(kkL2|75_v6ZVTusZC#XlTin}L%6i;88;a;Lwx#Sy-qWIHHx+U3#eSWKTo(TPa=MchH?}}k> zsxahN)P0m(O?0AJE35!c?y;@sAgUr|%|v|sHGY$KCF(!5c$$HmeH0&RsvPXJHqB0B zZ>g=2gS$@N_E=5##d6-75%Q3b#POL~B7R$C^5dcC@{yl474iG5w`(>LA!zR5juS?{ zvEB^J5e0zE(oV{4shlq24@TY^>lRB9`yPT(i2BN4M!t_MUN&BQf!G5!!YwAL_lX?l z{2V#ry_Q0r1>uS&aTe7y5q~zSPQz$H#R);d3nFk5su(H<=dFIPriw(W7Y)Z!HX?4J zOUGwQ>to>-Mk9cEn!pt_cx!f|i2tcbhl%C!bURKrBIs5rmf1UlW5aont7NH>RHt@Y zYS5;+F2T*Ln8I@(ts9;8hN-3_8ZmmRDn%K~thuce8uLXn?C4{-BoQ53^f$5_iuhw@ zWX4`Pp>zvY4-t$MSV%Tgj33R&!F`K^5e?gt2#Qgj9P=%x_zfY}zsW9^fcX?x4+t?0Q z(b2vl°x*KWd-EpZ=)O>YzCpR3lDqDe+XjM-LP#7SxGNyb|`a^RvZg3*m?X8@Pv zc_PG1tMo$55CZ^?vhky=S>a3`rq8$kFfg|QCEWv;dE(`e3~S#o3wWLEaph-afko}Mb7HmdxVI4 zT3SCe%u#rGU@01pZ}La7i2Iy7rSwp|+p{S~#0`?x_3~~pkqXiQv0cf6s>6>zBdxzS zM2e||j|Mi5aLyhdi-;R33pF`hj(3UVB4BT8+&!riiuaDna$SxYwkWTpbUC-T^4loT{y;;&I{S)?6a3)SWoZS{!JiX1qd- zpZG%;`1`KrZ)5)H;^-=b$eUt?EN#S&dIK7u z9r@wu*5nt9vGpWC)3tTtypk~@uV2k^uZD}HQMvf;+$+|fRzqfd1lD@(C~+1s``4bN znRBhbCdM=4C$LUypGkfCWQe?lR+|aA?B@=U2ilK9->^Eit;DubZE*+|L6s({C(i8U zEwRRQs>`{p1|-cvBTqLIdCROzq1lW+2Be*Xj->prw7%L}fjtifGQvT7Q?6E9>8Cu5 z&IB@7qlw&Or4Nsm8)+|?+5`XsMsji0GiM>i-f@-|u9qOzFbYlq7>g6dwBGYpBB(?njc z8r0VmeXG@^TRo=bg4&Mg)CnZhr^xOjMBaAm%h4^^<6yv}gFPzB0qgA&q3q>Vz;6b7 zNlI$fP~`n+_5Q_RFY5tsw!H|2p0su?FUStQ2lA8cKqNgqHcRB4vEF&FF5CVA+6WdtA7Gz(3Afaw|$PNVDlUH)0p8Svy+ly!M64wf4<(DEh$(<1 zqY;#osU463#lc##_N%eXNd=y+IV*^x*ODQxv=Xg=bM;C+md5#qtU*@47fRvn zmqKhoW@v&q_LgM}ZPxQVGGhh>XjJR94-ydQ5<* zsRuAu6WSz$pnKMtsYS(~FzE$+h~`|Q4CtOUzC;7&Oa(qqb3FY6d)B%>;o^N@pK6@E zs-8s!;GRX97appL^hw7X3;Hfor>9WfYnKAQ%XEDt^zC2pbDw;Lkd6IaraQlQaQ+J> zF$fIF0)*S&WvcZcMHE7Wlr=!e4kp>(Wm<9v53W9DW2*BA&jBQsUg%15&Q+OC$W_NH zfw_A3Ts?gM6;^Y|vIte8ovX{6aXR>2M1O2UGM;cdSL;rgq6{MN{c8}81(S-F=~P1$ zSY%bD1BjP_-L`R+vv?8<1xCjh%h2uD$frWly9>PFk%=|Gsv`=#=$qEdCBj~c%@tX+ zt`eT^Fps{FfgDknY>a~F?n-}%{$IUZO{DCZfzotZ?j40F-rwl7B8pn4aM!N=)QrcY zl$uJUo2?T>ntM%HRrHDqCy)=kYGC|({NJ@z{NDsT6oklDAJBiPE!W_`#Lca%3-{sh zu^&lM%aaXx*;E$$hN`6uT{PXbWq8;M6scC*gC&H!nJn@(vJ#ru{8S6!J`_HKtPVC0 z7w+aV>O7D_wGQnK7496xD_+YZiU@bEEIRk75G89Z{V-9u+sLA8z^oDzQl^@4x06=e zzFDkO>vcx|M=-vpM&LXsPDg3o-K_2?_5$AD=7>c14^@PFUpSQ{=A~+uaCec`vk?`= zDR`I-+5)1Ur_#hs`lzXJcb8VVtr>!Re*|cg4NEGO@b&Kr4wU!olR_K>c3R_P5l9*- ztx}y#aSMcdHc7lHHylMC_JmU>2#vYd+$-FxO;@=jDXzo5WEVX8iJ;2IDaL#RndlxT zttF$vnbQb(j?EFi`VA`x_v<09apf9{u81B&5S2&%MgRPXkFWVj8$(=k2G$dkK$ss) z!iFJ8%U^9J+%Jc?dMvIZR)MhACJCSKnP$ShKw5vk2hNSe6RWiGR?*H=YNusRlC zO--PtSEY69WaDFt))Vd@V*BFlB+^4%e}SbnwmCd>M$8M^LoE#PfJ$@R@B-I;*>v5! zjAJ_FeIbmNYD2CmMLLdeOjj(fNMC@kN0Yd=6zQV09*e{6aq@H)ge#gvSr8Gi)OQ-W zT%N+%Ad=nj{zC+AZ#6CQ4u!k5%T>A~3Q-zhicOH@RFt7blRHC(7pGaqP0J?tU1=S< z-iVp9E$C0#CZY^2oZNL}cv*-2Kw5D_YBEW_1m-ki(-b1IXzheeCF|0AWtF$(z*cE* z3J)=?kRch$ekZVf+TKme6?dp%W&hbw;pc%}a&TI(xV?tudpSbkW{)u8K~U+tX{q5Z zWLOh_Zzj0W6MqRSIe2@T(<>O(`r0*={Z3%}9DMY}3c{Ui zSWm^93O^6*l7s&c8!OyZ46FT&q6#;AqTeC}OmXfHXwt`f%tvmiqDU1$sIEzcsTag| zj&ce2krI!Orru9rnZLK;>Qh&SS94lA&9JUrX(Sp0mV@xmbL50)L!L}kb7mXX!*|On z(p(VU3MQ4gNq)0rc(&uWp;?UU5a7wp9j2XO8B-z@sRjty`ADyQk6zc&u>QTQZZFFYAau1!7%Ww@guAz4 zZR;5>zS@S_cr>ta8joPSpJ6Rofk}7);@=^R>xt;Xefm=*Fv;cm{i{qhEH;C%Ta&m& z6luQ8HGN?N)wj=paG6M4Gm6x{R)TP^aJkNo^{TcF>5ctggg}z@O{Q?~a=G5aXIOIv zpzSJxP=iREC*_GrXI!p2GFjvisTBwv@_G98Iy{Xttn-oaoSZQngt3}L7g^zE(o2T* z$$T~FjMqVUUy~>~Nq>sODlyn?`y+hha*0uXJ4U!4xLg|(vP5xEjDw(@*H%1w9aHy^ zZ%YdIOv4(nE}O60#=oEz?qe4`A3fsn*isnCZdXxn6=75WEsdzU>q^dfM4Lqh1ee=2 zvpXIqY6V8m|I1P$#C@bWv{RnjwJf`V%I=Gx&I+beF@*cvdJMm1Zr968BGlab9$4Fh znN%H|2fz2&cQJ46ce{qRtDl=GQ-RXmMd$LzX5B2)krf*ssu7yTDVwet1NT@Yq(}|1yce2S$LJQkZ!q}O;B_> z3)I&&ovt9l&7?<$75!qk>i6qF*s4i%W%83u(^_`MKj-}j!Y`Vn%E6vCS+0>6{gRws8M#xmMs{w9G<$a~o6b;Hy&~ zJ+n-xYvNB*bOAOv2=~iBCbbN8)q1>;cnO4g!6YUB4pUO$p{~lMqQ(0lYz!vFQ|l!D z(E^u?bKz}|VSPdbUemab#?taahJ~b56Er`Ey|A~upeTO4g~VUj(oIdpn(uHs?AXF6 zK?)|MYqP529bQj1!wo}-|FFF$qQpe{@%MSRpyfc={f$+{$lqu&M9Z7z1G7qq^y_tV zc$47-)P=pt(DDydPiI8*L@?^pAh$OemU^(>e-6HduL)Pg{HDJq{0EkEQR}BzBJ>N~arma3C=1KS5#(eVX1!?n&R=c~ zogu?%)#xGczrQBjU~CwhFAOr5^ymNhn(&B!BgHPwrleJh`fI`i>v#m^;S51DsP6RF zglE1|jx%~+Llq~_$F|1}X}oj0sz^X7QHMjhuc*S%&WGwYR1lGv990cp-445q-TUbE zm(j`{3%R^>iBMZGqW_pZk)r?ol+q%7*TgC!&DAHY?6X)2llXQ$-JlYFr@b|^sc`ix z+#Ob;-{c>JaLuxo?7)EtIR{Y-5sWg_yI+{Zohd z@wEJPO{mN>P5gpiYYW#iNu%C|Nc9{6o%OnFcq%D%>A65Lv>bAcs7?zZ>G>NmDmxpj z(+Su!kO>2^|AWBUk@YDJMc8O{H?u*g2QXI?ni9dE>220?h2f$*z=4{;xh%?s+G~CL z2_6)m0PIzbQwpjmiuAqJWfxX7D?#{7lPHxy(l1sy+Cn=3!eve3)y99-k#c<5x{W(l z@)7t22ifIgdO#MhTYo*k?r9ki(t}A{@63$g!r&%gp7j9jIp zwX$%gV%-P9HP|jXuZ-NJ-CY%>HRR23L7hP+fKZl5ZK!44r2So0q&2BZ6{Y>Nfi-q; zT06M#W#cE`!fRQ{zALca4o=$Nl_9MKtBWgq9I(k6kEeB&>%~;MsHtU@tC6(IFKEde z`5}nwh|FUno|aXvvDNvy#cMEEOH{(f!pxD!K|G^9v(k07mR7{7D23kx_DJI_!(G_3 zIr2thl~{=(c>g2eLNybqT#=^@a~X3Y(2L$7Aa>|?}!q0PE-t1zj*VSRqF zw#xf45H4sERhddaxf^O&`v;a+%tv4q9BSu+nd~!yea<$ltzR|A`@g`bqFJh|keYB6 zH?2C`4CN{ZgtnSQu9Plu#hF(BuC)|t2nf$>61BXalxkXQUMsCS#e5Ln)}-ECHMG`p z4N0L@mh=$F!pK$6v=Vz1RK5=&_M|rBvT!a~&bV?+tL7feM>oOv+h(cT!mjqF)giXN zkm1Adats93CN$!NJEm)tHV@ z3ofn72<_~+%7$3t{U}A+3c}uClA5ThgjiJU*SIcY0Woe=AZt_77RYXnw; z2!YA*l$IT0onPNfk&;2EosUH8epk~FYv5DrK~~usgePs1aL*&WWr&sXuX;d44g)qy z;}MLv4Y4-tr~OaFzfKrum5& zfG2_pttLTVm3QMv=~V!K+o00idV#>VBuFo_=BHIa*s?bY^147T8HDEvG=JjQI*?3N z!btz8%xY5#tDyegkx+H1WAKl@NtyXcHs7S|j>v%s#*5T#=k&23)aGBFT0l~sj)dUH zozTVjBlu1zlcHUvptR^l-!2dWZuFgnM|Gp`D1PBa-`2^Qq9l$Z8>6sfyE6~D(Z`V& zQo{Bv@0L7^f`htpR)N0`O`Q0ep`$e+EoNtLlZKqfS_;mHIX&N zyD(`9yuIeMUW^Rr8-4d{R%gyI;4f&-AmZp-$UTp?6t4n%Q{&`S_->48!N-w@ZZ05p z0y~1hE1j1b`ERFEiB#4pN|vDPwLS4&mau*;DSie0hV4?dJDR(ypgO^iYJ#kVhiW(d z&P)tP+&g@os_@vgz`JzCZsE3K6~xy~wmWTgt3i|BSXB)1Z7-=#aZ+7vqx zA>~jI#srh>dxuE}>*5>0Hm33^JQgC-`@NLQx$5>Da@F8QV6Ix#!Cw9zkiH?y-%^ycgxEGt7_vaXZF3r*6VP6 z`2TV5@GayaW!UH{I4L{Q`wD!8)t_8MPzMj|U%amRL8k!d5knDO?l216ysPvc12eTZ z{t+>B_rEzL*z=1kFzXA%9zZbmQO(j2;D4w|hq6E6p>(2mJnla(c?W;#8UULF;Sfl#Qst9bw5_%=za{)z-PD?>-&eQ(N& zNw0wvRpXrm;oBjt@y&%e1+$njh@~F^|JKT+MrxrFbav|9IU#PrGkgD#TW_izi}uUJJqfvHBm3EnaU;{h@wav8iX zt)c1IBHxJkoi-*dn(xZcTHg1bjCo@|isB(u+597|35Brbeibnfv}gxV8i=gfzHtkp1hTdOvCzc4;COR0e+?l!1KQvF37Y*;aqAv>CY z(N44KzhT#GMBg^(z{?G5(iI%77zoBFo26#a;Xj;ZhL)q^DwpZJi zq0^(j6|zeDV)%B``B7gB(|Yn~GbYJDz`Ucqm7x=)zU8uJmcwpqS}#pJgc@#ec7BXI!kG5xgSgY=hR1<}a2UbSo5i|cPCwwbS>)}M)IL|h_$?Q6=uo?ptmFw%AmgZ$yaol&^xaFEkO~zeM1S`0;TeR@Zv< zm3YPWevcMnoxy1Qx4uh?AsG!9l4IJ_=5nC+5}OYHnB1$ zx-N{auBMV}fc~{%$s>$=B=3v)3b?FVEix6e*h@J2jG)?snVg(r#=5L8jx}X6j3++RpOLb6E|3DWDv`0qVO!bU*1MmsRMIKlN=O910?F zvDChnZ@0@@nS)O&T>#?`&7w<);7u{#k1i`|1TF>cNw^b;pjtiRzIduz(jGl(;^z|HJYzA$&rczA`_t%(NGS(Wd%!raaI~Zp` zxL}inJ9bcv$aq4!BEK@Yl8r~e3Qqoyc22wu33Y4K!IC{DMjGd^(Lea?TqJ^}oK=1^7oIU8M@-^R-Z z=_2?Lcm!0s^2625L_Wjki?XbyEvtw`5UOhut0m_QpD)3(ifl^}%|K`$Oj2s9z>pFm zV>QI@+?|`u+!ggm)!r8<0Hz%_}&PlTUs4d#oTiD?GH^NMCTF0G}4UioJ~r z^*+S{O;))0IseyYGynIS`fvRg^jGRfTk)TLTJ+490jEVbP^IeAqQll>5vxy&-bTjs zY0;}>WuF#}d^zy6DDnQ+Y0;CwoYSHMdZ>2;4Tk+_1mg=NFF6nTwCFz%lDW^$1^&9t z(WgbJB>J>y?=INRgomx5?Ih|qR2qF+v_$(fLB4+m^qUP!%B^!+Gbj3xInxfojV+E&9;Hto}2= ztC~O_B$rB`7Ht%QC*6$cSjQs-R64WtY0)dgk(&x&RM)Hk0#1t_IfA>d%>nk*1ZsL! z+8a7q`?M%+%194!^=Z-E7_2g9AZE4pa1%p3pwb*S>Wn8&i{87MC2o`VT`)SN4Y{Ti z38zH|{6CDnd4SE;|37}tow@Ux%NXl082dJs8T&FA`!)*_krtICq(mwtM3N;-i&98Y zQI>=(5p5_-$&#{VX(d}iNq&##^Lfwnb+32*KHopi%)RIFIP2@Y&OPTmALpEoorLrY z2p5#ZwIw8|MeFW{Re)kjoSfGbhm*RXEcj~Cz6D`|uL-cRBghm^)uQLOVUB$Rz*`-` zQl!3GbQJY6suo@GyJ4ms2l_-8#8-<>_SsO2K3lsQlk9RZR}-6FF<&jZ@qHYt3~##& zf4(7{`vjgY#X&9Fd1ny`e-79s!a2OEMKjx_3SJNkpvBK>yryO~X4>CD|A&J3owUS@)sa6J#R@5TDpva7B1ozh z-GVNg(zKZ%bhwVh4F+n_BWvS?)Ek8RuOm6NXsz$%d9;IGYG#ciJyq7MXPM8Cv7|5H2k2$Nuh=0)S`!f z@uUIN0U`4`5(`107HxB-tT{IuHPZux-bCU&NlZ+FTD0xk&CNw3jRaxBbul@$=*UvV zIk~`VAS_lA{mA-i(Rr`t2x&72`;UxvE9mwMjOCkzmaDUsm{~a%$0a z!|}pUcHaT@gJ?SW8&!)=no>~u-o0R*iDpuDa2~?3JGJP({P`tJXa>|!aFS)pEmxuB zTh*d{_EnP_Zv<)wrL$U8p0ZPmroJ05%-&!PQ6>#lYy=Z((H-k5NfxGn^_nvIc~Akv zS#WC6U(%u5M+a^{0Cls{=@&#Mh;dMht}j~c=+ekq6>!AGs{8vD4Hb0Kc^Ob{a`(_4}>41Ngg$oD77dNoLZDD7q-HD90V43 zF5i3TH{;$ok6lrB#(70{v^Qh}2{glKper2DUFb_bGBxI7+o4n(w*ue(7`Kq3I28!& z*`H)$5x_oyKQmku^yPn|@I$j@n$2S*U8Vf79hyd>(b82KPO)^AY{~fW{~%p`1EaI7 z7Ffi~==}sX;bMR63z%a$>FT!rc;S-aQ5r6gL=!Y8T^%+87yHzOcQXgc7|5N>G&@)c z=PCf}<#1b)!mD((=dhNh9DE)HJ~o=e<|#4B!aT**8X_BZ#^KHwSHUD({RK?=%AAWd ze%{|7#?Ma>&*v1^J(QF)$u5B|#~z0FFAlN~67Gzj$wg|IN8o|`Ul4C19ETT@Gs&Ls zoQmpkn3UCoXhNFwn~!GVg&#&?;+NWZ%5>F_-!E(8kJ-SbTQf{w1i-U+y)nTo{A*M-GK%<#my1;{^}6>D#PcOA>e>O#&5~y zA3!0b!vXCZ1V3v#C08>kLjp(l=Qkx3dxtO+dK>}xSb&u7p=YrkoOUzJ7|AJ*Cz5e6 zolBNZ98;Nqlre$C*+J6>u?(P?mUH@yj1PR#GRcgB??i{#^*%f$F#dEyGY8N@g+KP; z>4E+;+M4x%wkW*Sho28zfQ9-1pp#LsX?Ti;ZEthV}ZN>R>Jci-QT8qqe*l$&p9~a30SToTg|~z^J0EvwkX2 zxK`L*)iedxUU9B}$wedUU0IWBZUZ(@aS9)q)wb6A6YsziAk0t_<-$rzr>QQot@l%# zD&|cP$Yjp>m50?}=}jn97h9wu9-=9+9Af;GZr0noyN;nZXFu3vZTCxPQXa0Big(F+ z!7&ZZWnl4m^#o_TI9F5RrMbB#rH}PqX;#w&N`p~VSxnHiIsaV_Q5Tr8GH-$@(5`0Q zdN8H2khfif61?Q!Xu%6vrd{Jgoe+^I)Z5ih<~nb*DW{xi*Q9ViAn}!V4=HcjH7oeY z(eh>{Jl>>qo8Ex>pih=KhBd$K8cMuHIY6s|9VTNHV=LgDE`&*@X=%Im-sgME7z${A z!k^y=ZiNUNKE!vd~py~a{4Lx3bO+JDOrc5 zDE%qfKDMQqh>-8Y*#~HFx)q=NpAu5ngb4iipORxe+jH>0qJmrc`X4_f1ukToRGOr4 zVwAL`s6QpWR%aT@Lu(=pMUMQRk{jZ;gty&$VbCU-KLS} z01g7{3jg~ZX6chnreS<(`cskkZkq-V@!d9!&C9Sl6dD1#+onNFWHvN~e8Sr_Xa&-S z=EBEq)5zF?5sKXmpIvaA9b1jkaN9I)O~4Sz{si#65^51aw`rt)hA86aqV2#*44g~1 zP26Dzxp60C#C1ejdWYHjT!U>V;V&!I+>deki;oN!v75PAVv@ z*FadTBwmu#Z5nsf%Qe&(>}C+Q6R9aZS>2{l^@bcVlK%+ol#jb@8rw$K=9X&X=Ar+C z3x{{xG)n)`MDWVMYAEiuY49vm#Ln2JQ3hwl5yy7LpPoeKcf)Pd;0dFUu}!1rGil7R zV?cb0$dYrGIAfc}KVPK?J`dQNit{=ZZPS?iX*)@56R^#Mb7F!+BiOW6NeTZwupc!% z?`5WK8V{H}f}7Vc5`YV*&)XibShlu8I?wuSSzy%(r}%Aqs$REE;}YhNHbkK8tH3sm zuO?TJU_Id5TZ47vr{cD0^!T-+O;0sU!Zr>2ce*r+ zCLpv^Qdh2;@HP!v-n1bCb(=<>d$Xi0L*P4BgK?=j7rIR&qXo`vdI^lVj^%IDNca-3 z|aKGc2-!Vw}#!*bg+J`cnh>QU{Vz^?hY+omzNVI9e4@%dPx zg_A1a(&;vhJY56QdYXZdMI=hZ3dwELNIOzb>gjF}9=MLgJ20?KV`yBQEEqlw!gEC8 zb|x_~3EMO>_O+G;!zCcBiH^x_)3^XDs0|UQ+ca8Etz&vpzD~gR3`pSWTMUY2G)P!qz))2#%Q)k0RT&#iGhvZiL8ZseOnh zEn6(!qc_5&^&H)n1>=PlVlisb`-n`pSlmR%);Ym=kIIo}sg*bcd+mLQ`VqLmqx7Ck zqxf`-#mMr|s80d-q7uHY140QJThNK`FXDjpaR~A*@DG%8>uO{`I5?#*sdGE#>;Zm6 zIgb)Y1>?o{q?rrAt|?BjVmyxh5L5tssADMeCPY<{@z5k@mFN7jt0wOCQu-)NtPe}^TU2q>9QhADu?OYI#1N&E zRDdiUG}HV6-*guG%_mU8WE}bd8`5#w+N)@|sTV$F#ms}~-7&te_yzx4Bmc&4qi^XC z?7%;#pLum>gnlNcE-Fs-Gs&A#Xw}cue>*}yvzS6T{Y)1i{%LE(`!D@WNnpNyCM`kq zTub4*7B0{Q$y?5Y>Sy+ED8+-_uYvD(9M#WINmM`c=sPvdZHVD2Xff{~y53Y8)z1w2 zwu+&+Qvg+Qu%+Dk`k9+QFJzhm%T=60M3SJNY2Bx|xeJ7Q9m&_vP&4uMGhaQHZSF_- zsUXfGGMA52RQ=4+wT+qc9`H4erL`dnrTUrF+Y<1i zjfU5F7n!f0`E4p%-z)I{)L~q+|3_-7pShzA7F)hU&>uATXv&W2XJ&89ffWuQ-i3(T z^7S(}RfO0h1z-&&Pz;t!rTUp&_hVh56&N=viz}U3s-O9523Y;Tct}|h1n6g$oP-s7 zBEb1dphq7rttRXAGxV0*5OGyM^ZWTs%zxqYtH!VZuXtpoX*8;zS@C7EnM~fXOCf@U zlV-y;rTUp&cjp&URS@baiHeCKSKjI-4*Hp1%Tt8Z351?XqAd9OnFrp%Lm32cgd<4D zsrs3Pw8>~Xz*igrS7IX>J#DQL^)fsLqtn)6cPB8@eggXEE{Lz6spGSupSg8aNhaBI zU|u3Ny<)z8re=;Ou@zi~5eJ;~2|NTN{Y=|wh1!9cLCPJ$DMxW zqfM!T4+i$Ak30R$uGAuiANmYnFZ#IC&$O!%K5cC|u+={9^fMpMX~l1)-39C$A9wnh zYxUD5{&T=C`MA^1d^N8FH$PkOU6da#Tt26t`4YSAIe)c*HS}?(pINZ7so-6J_3&}0 zpV>98xZs0V7G)eU{Df0>n>2VOAzK-PdGi5iX3h8wamRv`2`k60-ZDqK=1%w@rx!F>>P{jS26FQCxMeHh}O@vpKax}NCly;lK6?JekM>NQ`&Y%5N;+C z*Nl*ye&+0+40*OgKzRH*lGD#zNv~lRuR-;_2*T?`;%6r@IsMF)d@arHwIHnrVas(f zIsHticXduKa1?}7N}?ZGUq3T^fHcZL+zMzy;rxtS`l|Yw>7PaDXX<>AWL`u(fu^8z z(NOaC!e4@_ex_Oe>ioMMxDV7Z|3z2*OtlrAOkgHxZxU6TuH~GoerCe2RZL(#7`y(L z6@ICle&*uL7|HHQP_IPOsTfA}Gv$uglfE};B{pTkMI=wv!FdSB?({Q54;u-S1%59L zL%HR8kbJ9t=DuK2c`U;~eOl?PoRFvN^fP&{l@{jfV7;eI*3U#Rp`V$&F;TLx9jv3u z9BMKAS369egX(9t zH1W)KU-01EN5`6TAXxk`M?BkT{mg^C8k?dZRE{Rip{5e0pCN+N&(K^^PFv%d z4M|qWi*1B7_=}dT&EyBKOVUIm>+z`KLbs zuy!L1tliKD6QbH0aQni1rs>Q7I`9d$@$o5iKg1UL5y_Yu z(_7(KllY+b_73b_u*vK2a>E7Y(*w<6uu#zZ`dS>;|KQ)rVU|AGrPL5t(w}w2=Y@gb z)!3Tc30LHU=uXF0AG!$x_|R<#P?1iK4xWikr)i804I(eLah{E>OVdOf8VfJUQ>~y^ z=?vDB>|%JWhT}ZhO_Y!@>IS`ui!m&;JAr}_B?rYxS2{aa;8 z)b|9vdjnVnnt=eXX&~lDWV3(J+j0@pq}5<-k7n^Bh@g%PdLx@*8hs43OG@=~5kVar z^v(qEFq+*0~_)<(bsnP@Aytfv|>1U(@3iTjx7yw$s8#(tQnV zzmK!6^WC5~?9W7QwDvr(%RbJw&Q(G0(Z#IF?Y<^GpsPEukv32?9UCaergI@jGotR`F0CT9uEtZ1qLGSij>4HxM_PpY< z{=YZq1)hh<4Bj}=9`_#+&Z!B0IOz2rAC7+~uswuJ-OyYjcr55WnH?|jp9Xfp$JruT zKF<5(>3kA@;)mG(0w)b%KUHu=QUL}}#nx}E4)cTI z$lJtuCkN#U={^u1Qxa90^N^x^P0kZ0jxKonH;B>I|syhcbCGO@&Oo|l_jl( z&7CjAdGiCMC93a1IH@E`G;ET5InMk3?y^D(Zp5-AoD`dCI!t;y&dbv{B<-Ra2=$e8 zg6oFOohM_{XgZ1m@PUfGb3>dL^G{GB?+xD}8jg$2*${i@7Njr_eZ|vYOm(cVy>oAz zx9z#=#x8*WGKX1upThReYp99jQieKI`#A_-5=kCC+dG4vm;Z+tL%pm01=txMXM1Nc z&-(Nc%|#A%jSih!qmsJ1cqKYn7WwMRf(C*L5T| zWj67=pU2ge3EnUe#u15HayTX?HTS%f6V+vcHwT1;*TqB&62bPKcOtE@B=<20TOBEE zOU(7W*&RCyehk>pKF+qpn>_E+9+`p%K7wTlPRc|bZrGOC%k#FB%Oj*JAkXodpBH~fcyRl}vklq7fLo`V`g0Y^rbVfl*{2LIC5y?+nNY8lQkSD{*{SAVLQ9)#K z+-1!0ynP+wCAo4S)VPks*2Gz!_xN*_Wci^T2)T|FwkE#nd3O|QFL+;I0~Hr*;#;1# z;dcn;#=`#@!Z@oD*2MQb@2P?qoX-ciR0&)wVoluOc^7WS`|&BjFQW-;52UFGTSsVQ z&h|i>Z?HX(W*}@2q{)d1?Z*hH!uN0Ch{=|DX{M1eKNAO4!NBw?#!0E^IByuU*_9XT znmm)dW@|CFZwUWPxImyD-bQC&>i7^WRyV@?b_dyb!kvMs%S}y8B0Q{p6oj$SBxm-D zYlka}o^zNpdnLjn#bz)5;G4aAt;FD-%w8XWnVy3`zS-*|T;xY5P`w7<<#2%_RB?1GmRxYg50>IH zAxbCdz-+SM;##7E_@<|P(HE0ezefts7q)zcdaIomw5CMK<$2^)<>{b2mz8Q@i4H9{lsZ z<$o#P@xNdAU;6fg_~y)fR}V#)`JSbQqGrCXNXs|#rEHEc^X0+KX=>(c1jIM<5%0gu zd;ws-nQv~LHvG(AfbTrGz)RFDIS*>)OFrI#IctG$bR0GFQAyOyw|!1EGaoUW0PRAs$d;v-&%adZF8Z29*DGy8>T2er zPRfQTl$!Zgj0Ds6h1WnAnQ!J>`)dpI|L}juVO+ESM`~*3`}uR|rawZ^?HYVDWk=0? zbvvQcItK7JB~WeoX1>0Q(y;#hIXVnDsZNSPN~LDLz7N5xDi|5c;!5YTtC_FPE!Z~I z8H`($6+wWRZ^ULOPagz0NeT4mrL@~MS!d>>Ud@Jxt7g9P4`P!09(;Cb43qJSM^u_0 zjhgvtG>kD{Q|u=Z=%NPXno=|0sSoQ4slXR`O-VSvZTV)t?-nHrsXhoTltfwZ&3tF( z;Ww%~z`GpbIy2va%BY4%0giP9OWE_ye0NeWqh`KorL&l6-v<3%7sNO7-S4wu=IioP z0Vdh~U>+tmy<)zZuSK&&iS075e>6588;tK6x9~Kj0+1n)$AcX)SmoV9k8o znfW@5uOax&z~4okf2n4^ez&z3(&HdJt)%_*q@0qySbch7UVGLBvf!b(T- z&3w&Q))F({PGEZ!S2JIcJ~%-5SNQ)y7}ry@neWMk<>j@=zZ09Z;Dp3aM9qAo@|Tvj zT@QrjMB<+>AvrVOgUL8zjv8Q35c*t4a%R3B%TmqTdq5fm!bBo*o+Kt`<_qkpZk7>g zJ_t*%i^-Y!mJGugR7ftc6@=YNq90JBmnz-!L<=GP2EtV(QF4~Ps+n&ahDy3k{?x%( za~PCBg)cGch4UMN^@FOJui|~J`FA^TBdGWO7hTPKM}Dq|IaP5Kz9zbT!5LBbY1nEAHZaJ%<*;Sd}+c@k7RVG_)I zeY@fCSZdNWKxn8W$(=LvP0Q0tlIsG(T}q<8iJXi~J9@4Ztk%~cTc;nzE&{hgqu5@qHi zf;01Rugr66p2?A!k3Y~cT{QS(Gv9At!Gc4VD$xjr&3uKo(Lvk)|2xy0UxvIkv>h{? zjJoeaV~d@Hqjn=@x)KCC3IC>LBDa(9B^q(?CaUXq621pScM_J%C~Urm$62_*9hfm$ z`ecctI|(^nx0CR|638J6A=P9!4xu{<+h1;DD4?wks3F1k&=kz=B)s#sR-BBz0Z?~^ z-A=;nF--&>1n6OfX(u7op9#?zDrwt9Y&s?D-wWvW)cUs{V{NtmZP^?&GZ6kmIGaPQ z{wsX)t$#HC4iWh8)<0Y_%E1HFuWE2hU;o4U_vgG!Q%l*Dw4|ukzmw1BH9nH!&@~Q8;%q9!s-pjn59ql531rT^v5;;pW9saV>P)u;Ej9P1FL`d zCC00vx}fvWJ^TK28noL`Tk;{JR`vsRX-;HAJ>VnRS>`o)6XRq%8a|WYBuS6b5W;)j z51WuBI}hMHO2|iqaHiLKPe9FSKLWU237pHKT&Rz|vG>5@brje+#VG|TijcN@hj2TB zjoXJ`0yxPAr4mUx=si^)meJ}UWGjg$9{*NH67dPI=e?b9{}1?sGz`-tviP(2ehdov zBnY#jNnGy{%q!k!lUu_60oEtV^s~UsL@qXNV!F8x4jTRzwBP@Q>c`Ahs_YQ1M9N>@ z6y`D~egpdlT*TuLRzB-p+uI<_ss~1MW$`oNiBxt;>^{#dBdnev^idMek=T}%UBPA|usE4!NYhCU_BId(3v1wKx*i0qoy`=&<`Zm9M{V4wOp%>uIP zS+7swCW0RZc0zIPOS2zMr=K^OU}ZPA-lmZrb8O!I5JA9k;iZGj9#)rs!FV>7-P(FT zUM|8M+X%#FMCP3Hz&IQG`Wk&=B(j@<_0q@$$G*OfE8`{nFkquJJdc^Pv9E7=u#?~~ z0h{aNG+fT^ZM~=5`D^G$kOYFr_(w_AxoXIel|3Mh=36#IAltFB>jb>b9otLX)xN{| zKX6htTx!k*4{Wk=Qu~26Fd^RnMt8>&J61L}Miw}UMe~91f7D@?-Z-&iWe*N`KaXu- zs0+0-KzNZz@=DOeHTzK>%O%;JI`>gUNm- z;4REoLR!y75bS|)B34K=gUNm+;N?#&DfLtmgv!^EXmXOhBH-Ory0pw-T7Zy4q=@90 zv^wB*f3AVdVD16o!Run88F%)^fHysVnj|*~gz1hXcC74A173|m_!q?0vXA^kh*XodMx5CCRfVJ687ax;&Yd7yW3!YdqXzqAhn2 zt9fuzK%&<7Z~rGaYVM$`LDk&R`+E$Q^%|f*Phwpz zKZ2Bi3uMx>yMc5h*nGt)gxu$odUtiR68;+9uLaFO zcv#y4gq&!SGvqbsfj;jxhdD#uH^_L#*3uZEAXoX3(bRE)$Xum7fzjklgqlqu>ottX zRoz51wB_(#?;zWpaF?s{W8zI)cv!m+grm_UXUMDlHG1ibz%0_{_@fPZrH&$f6S^Oa zq(XzCIwhN0EpXMQMqkW;{;QKs+BYZBaB(jTRk*tshCW|+FAO~i)3j?YstoOo9s+h#kae{d2 z>)#8rEWd0g(|ciB{|%Df3zKJQee(x`)j0}Z`VjG_Gg`c)r1!#(p zFV(*iW<)-8EIq;KudJTeT&3!jF#T3RN;w*gNsc8iqPr4i#w~@-Fv{_K;7c9Ha%J&! zgRX>`^F%8lZ3SVslDNKvgezgj%_$+I-$1ymB+8S2B}}am`EdU62~1hxB3c4HRsTwu zsy(oYs~#B5mBrEMl`!+-3Wdi6bS2Ef1?`w=?*;vS7etIVbS2D@cxB^CnBm{oWRiUW z%vXp#iW29ae7X{*@YU85+bUomXl#PFj`2=y!_I$r+r9X6fN&14SHkq7hoD!&?B3R# zNjByOyuWbrI_Z@#cl&Hy2~#MeIg@N{FdGt^iOt*jwsDFLxzO@td$n^E?j;-eK(;fw z66W%Yg}BGJdw_kXxVsW&V{NpxGw}Y4AU}tCCCshdblsIOIbYObrY-*yhE#BoLEM!v z_xeF_B}_3~OF>L~6aL&uK`4XadT>|5yjnNi1ReqSL^Oe~d%=}3f0n8z>Anp74dVD& zWe%={`FcbZN$5l1pGU`|SHkSy8!o_c5PoqH8NCu__-!R5l9->-{^7#a>8^yCa;~yu zqyn(2g!9jhTp5EaVdgFjXQDN*EI+)v5~lo3t<8Aq|L+FY-w*GugejU?)2t_a9I%NR zUay3y`5Xo!ufqRr7dG5h+?6mruT+Dj9^f`5JjtDvyAq~jFO0H|0{qDl^h%ihrRz#p zVt#>C1x{+2dk=1rJeQ{{VPY0E5oR^88b>oZIbq^Tm|3053bPwneWID%NeUBJ!ptm~ zAk0xI$l`tz4gC=k* z7=5EyMz4gac)qd;j08AE349`;MB}c6IaeXk1YQSYwX*1UCo)BMB}~TfwkEIxwEaq@ zn)I)PnOYEgw4f&ooCo2uBN@FCrfqxqsR|T2nb(wui)az@oZXc$SI$)9nhrDtEBD`+ z?n;#l^Ezr+*fi(pClcqFA)!u-~yq!3mRVG=(tcO}flX0f;g z6aITB3GOavIH^~{T+Avg8U79UW#v#+>Xk4H-YhSif*72agOhl;x|yg~!VLJhylDVJ zJ0k2xG$Q-qbXMI(a4bJurw$&A(}K~On$Z_%hodCm3klMA<(N8 z{E`2S`kDU~`i1`;;eY8LpTsvOTl@V~glsL8f%dGjwb{qelj!LbyOAZm-NpPCAzM2G z#FwoR@4sYgYk~Q)wYA;jxM}77hM#`8Kz<}|IS(pZTk(D;<~#`e5yw&48kIz4Yqn4! zQvxx(0@`atElZ_Q+1k>?dWPa&2k0{gTgt63TRXU-mH7_XuZmNMND^dgqY8I2*FXrK zc4^G8tM=f?>dpCR66nj;X1$SUsv~?25E~Mi%f~6IY_0j#0?fGqcz4I~WoxHqMC2KDY*?%D3z@(dlQ@VY^gJt*22jX5E&M3cZvTLyA3no-`ina zv;RkGVHHt2gs+vvwIw8+PBE@*6(OAh;gXUl3%+cvStjJe`Oe{$hvRCH*_fVA@l(0V z#-;_1!9?Gr-AtbvFR1_WoyT0 z7nRtS0b8Z9@yv&0YZrRAmhfK!+pFPKwsyK*n&4-E{o&(Iws!B!6$H5{!3tcecZ{`o;qDz@H4>v@Np+wt1+&s;Q4>Y`X^jO z7pJndv`Wo|R1bvaO6p8c%E{Jl98RsDK$We1;y-D~)>dszF8N7l9R1{^kM}ewE!XKI+ByE+dc`xMIv#{2+7ITde_U5XIlX4 z(;m93>*4GC!h2rHFD$yxfUvNaDaL1k+%FU@Ce0wwSrC}%VjzrB7?m93R~u?hcf z2SOLI^aJO&8|%|mwsxo`Zg;5%S{tHj)3uy)m935W0h5GVz!>zuEE$BvIoVp-S285K z6G5F7O{ZcQm95>8QC#}o6=3a%W>R%<9>TFZ*;>MeViM*j@c-5@lv}Ps$+ybZ-fPfW zYP{$rG<`TJAcIj^`;SSr`D;Xl3&ApaO=o z;ACs{OqMk1S)eXZI{kvkT#s&dnc68sNFRdmxsvFYDNKTF?S@wha2D-x5Kb#e%Hd>d z^R~B=nBp(vrw>j_OvzbJQ)O!#mP2fk4n{r4@?~oyHsjilZ1~^gFnJCtTl;lJ9Wwyf zh$uWf{ByUvWVI=5rh)KkG)abk?sk{2K5lGQg79fH=~ikgQL;56IN2K6ALMij>Yv%F zMqYe61r7e9Pp6;}49nK$!C3Zx$<`kEiDYZz&=j&7_H1o3id|~Or&Iih(S~N=3+_(uxgDn8I%;B~lh1b(5-uSAVc>q3J zf$xsy@W}yU6T`^?ViO~4V8*~jFH1aRZvx>(?TAx064BN!h8!4YrBDPM>NUV#1fC9XME6MQdSe9Ibc%uo7n3|>Nm3S zsQQgokW~HFlS()Rfx<`VgW&WOFr zaQI%7=bBbGMFEvlIEGdm^j?&;Pu4JX6blh%LN6dbcP~n3EQ;#AC}bQ<=aQupNAE?c zU7(E_iCD%`O!+AixiMvYAh%VnnF-(59AZoO@RUH~A?aonpiK&w_2KD(4fzY1J%ElV zT-k@84=kD5#{3Q_2HE0dEy0$fkb6;fJ(OSyIyBrs-MuJdP_ht}Nbg0tx}=<`ff(Au zlT4O@I+U8;i_!~=hm`xifCdtrNwD6F!tIs(^Z+FiE{CWm%bJkc%w*Mh7IV-FI}%OObNMh$ zq~m{{MEL8D2Kp(vgkbX`F*zEj$oIvhdwdUrOYI&D;?33W@%Za)%~FIs0B1j@8v6~O z{O*x#`XK`Uy?ea>rH&lD42@njxTUZE(LL7R>Y0wprlciB?H=dUZf+ieS+(p9)z18E_To@Xv z!oRh{EPb*AY4|pT{tP3&yVZk-cJ5Y>XR$BVhHe0z7f6B+rPI=g4fQ6U@U0%S@?=9p z;p1-g*u1$8vNIb#Z^3bP>?lgZ-Rg1I2N2}i4FI<);YlLstsXVjA&SEQe^CPG(%tIu zTusPx{{fa5h)h9>;%@c0X&W>_6+oz~BuXWcbj%wy3QH#)K)6>)yu9;ob>wdK$lZps zm_~v>MZ+*XB8#WJ(XZrU{s+SQ(Il>SX6mgTbJk!4x*fEmO7-&)LB*{e-R9OZ_7Z3X zV_d;~svotSVUyo_Wjn{4NmOWC0|6RpAbuXg!mS<;+*&Hkx)F@qmBkN*m!jxakJdjW z3hPl2#wv-IqV!geq4(D|)EMk+5MC$Jbb7LStH-1@4P>!pJ+O~`oEB?>=z=eQQJGt+ zJqYZWkJDmHu(9=u4X7jd--N}w_}#4@G>edRtRQanc(5ldIpEl;_*0X}{BF2gJ$RzP zi(Wyfe^-2-!yMZk#M_80Id`{u>}*+B@JE1+RGil*=vI%iE$d2RGl9(}oD&lqw|exb zRZzme4{W`L*IPYa?ombXeZUU-I3GoWMUtT}BpXsM*^9tzFdV;~N!819h2VoV_hQ~) zLjczf?JD|x>Wgx*S`Dw6~xI^60p<;Et$90}Hx zXeP(0w|cx<3w{3UU@TFVG!>e@1(RdE_nyFEk(9y~5WZFtB`Om^y4Az`w2_cbfpAGl z6fR7{tsa?I3rV9W7?)sD;H1-eovS8%s|PJ>*${zxt4H?MaP#N@-sjvZ`bG? zLp`ef64+iJcei>J9n?^&;0&<82&aZ+xpZ>M&FmQO)6ZH+>nY+Tm@;sjNSMT@+`Jv* z_34u%^^^fZo9jrl1{B1t9(N6^DGNotKXdDPrqhoTndX$0n z%Z3QlTRnQ;RoX11%&&v*XA~?VZg;E4N3)6u=?DnFDrqTID&N@gON@6K=0A=oCLTWm za1jwP=eHPdTKi^ZB_&n`cpc(Uy}K$2-|A5#e=(_&To7(_r0_{Rf5mt$eyA?^Kwv|B zoKND3iS>TPO>^A&PXYF#!}%vx6wJW{s11(B2x&P8>y;$!GkmKDud7ODgj+ps?h$07 zJr4G13c(dcDe0{qw6+@V>u{^b#w|Hav`L`^Qx;B2%OQAqG>BV01}&^7q{bk$jV4J) zg6xrE@Sfn88sZ}k{+wKf(6Fg8kpldSp*=HQ*NUYCgw?^FlaKnYwc zdaK8`Ut--L7hsQQ0`C>1MHe0((;7rxk}}iEN?y8`gVtJf8ODTmVNB-k<)Af{v?^(k z`ZXJnun*CuAHW||y?TSPrrG`89rKDA8hC#M&+U}xI%=64XagCi#h*gB(Zi&jeGuE9 zo7G9Uf-iaL^qJc*@5#I8a52NAr28Mu#6S|%inzj129kXFN2$i{@{fY}gmYy6>RFhD z)zuRcOzMw_$QM?BLJ7NbWNIZQLa&B}m}3aC1o#@|^sR{usIYqVOZc$|XBY5;${9->6;|gwUc;ON_P62`LWI>L z&sH<}P_AS+&ZJDIP?xn06%HTj7%G?vQ8i^0IFHkDu!!^#)y3zb$VH^@r=q|;;eQ9k zb%By_i%4gV;Np&<@E+|T`xoJE5$QnX%Eq9wt(^tJf@qRkM0)eR7RIh}m|H}u1&@rO zlj?J>`sG8e8sv@4)s$k8x}E{)4+@#oBqCQaTQPdfgCI$84B4`TyIf6x_^vWMtgQ(` z<7ko-R?oT{??`80w>VsK#siBy!%NILi&`v!^ zkn~SGSbFJ%q(fk)-;Y232}udl<%Fb9;JX_x5Jwe9O_l#dy#|1jf{&F$Gl!b{AAD0a zH+?Y!>dtRw(u%x>dWD*sp0ZPO)935d-1PWO)6;jA6E%0Kk{B40tb0#d_`_xQVYB;? zT_X%tcOM?Zht%=a5RvDp4Mc7w!SH8C2%dp~OrvRL-1d2k^R!R4>J(A3F5H>4`VqsF)mf$4a z<<}N62f#Y}FHC-1D(TM1grq(Wg1n+|Qgn^fmvrC10$p?sFq$f>d%CMxm2~Ia4e9!g zVBGFl(giq4_YbW?raI;L5#Ucaj^)bYIU-5A2ff)qNUwnKrjoe6gak?VC(j0jvcuooWRDDVJjBB~ZdWGE76_ zvcy&qST(}=4OL0^k%yZ~_%^_@{qRoGePm)MDSIDa1AN>`y7yu2m@3dd32c&&J4yHD z+iFYv3xK`t<4)3j8mp?^wR1yZjGrB?}XP zJwrG*4_@jZN%z<##bl}DEnrI&carYd!5FuG4DTHT`8iZc_fT%SPSV|JTskxDAE2A0 z=patgJ;o0LN%zS?wV7#CL9a_eD1+g8aFXsX?#wZPEP&mj2~N`e-GCHH_kQ5Rh~sC~ zNxDZ)E-eYo06sf99+h-2zO{u!^gamdT|~a5d)DAG63ISb2VEpaB-=wV<2act;V%OF z+Yir@Z%Dci)h#07ix$Nq09^QSI7xTcxMbsXM*rUcSX08~jc}6gHVs>ui}d@@9oTIe zUM1b_7sG};1pbe^u;I4iB;EV2Vg&LWz`06j%8kiMx~DG0;@wJs>l{HP-6cNHlCESo zpyLj<{DyIh6!|qI-DSRs5$08}@)dK>gPEM1Fd^wa{zX8TRlsT(&E!r}n2>a*;S50P zKy4SW`bIPTPM6nBNYZ^*mzF{w1M1``x-71cqE5}kg$X1X1u^> zAZ&9aU(!9it~`^#55WF#xU>j)&Q8*O!*;_p9mroCOFwW@4G}HENxF+1ZzZFv46w3@ zsU5;~lI}h`TL^OiSTdaAk(5fhTUW~!!c#|!xFM0aSz?2XXkfa5aEl}P^7OOcR5F9${hWg=EigsM z(|gfZmZzUV>MBpCg$)~`P%2Nqg8entu7}shF0!!P8uIi-HSr6)AO5zqD-Bh~|B;%? z)64Y-w-|z!*WiCpc2u5z-#m=)8v$&s1gb4xp8m&`_E`S`*k1`0gXL1GJpH@QnDUJV zW16zK(wU|5^wALO+Bd*>S6LAR$kSI1fgN)*!0(mdm-dP#>*VRQ;$lO@ReAbz8Hgdi zOoFKbCr|$_r6EFt(*V9)!wlc$#$UQY0(z*hRWlc#%=3K@RrJAv)-aVJlI z`kfLojW`YLf{#0S`tR$T@>^*W%VFIIPRbYOWZE|_60m8}qK$#0aZphOMhjN6J2||Zx zlFHM+^W%l(_437r%(M@J{ul-EJ85Yqkf(2IQeH?efG{tbr1JD#_m>dTIuJg)j^yO& zH$GBMNJl{U^*WN1r|*9$OROTX6)+}%3qSS7wVRo^djej^YM#i`D*~&gxXRPtos534 z1^jaeqhCQAEl+>{P*r&??g8OJCGitcdHRGC;df&a2rm$cYeq;;o__OLRTeGa+bcTJiQ=Vg38kmHL)fiD1qUiOw~}_`u(6PPp{Uo4F7Hi7J|Cr zzvwDY-?|3dN_T^HoTyxUj+t|=^7NJ)aH{H`U?f)hA6EFKa`N=@^YTe{Q$cMMO{ZcQ zm8ZX5Fj4y68^IbF&7|tEl!tKaPM$vQPzee16!@=b7-njiu#~*N7o0QMN{z1q^)scj zJYAl$lc!(%Ek>9}z&fK$8mia`CgkajZ)hi32vvq42`-}6sDPYpnUcn3c)^Fh&XgwI z2-Ma}r(cjT3G(y<>%yelLFlI>`eh1}AWwgDdPmNp9Sg!UN|M|;d3tum{F2;TAgopr zC1*KJm8V}^htb3iFm^kZFHcV&0ZsNV@W0?Nc@8R1e|cODTz8N84_rj~W%%dh=?nK( zHR&KUizdnN&&kt2ex;r13c}sdq{7rxqU7mBaPo9AsE9nBtv__CDtWP8lm>s%@^l)( zusoe?oBwa}^z!Jwv*sq$gw^h095IT^6USp(t2kl5 z0=&~5B*P9TPoH=-5!y{)og8knD7?zkL%+2%o#AsY@FCG0mZytl5%P4gERsz%W6r~{ z`NFbTvMT0T+Y=&fcopYkXXFaRvzFo-Oi4M*Vpd&Pw|2t&TL;;ZggYDFL-(~cW8q=# z?;!jWO>&mS2ht!*%a3$9ZYiq?(Ht;o!(&)tA$d9(tyF(Wt6{3Yyt98xla9c*!r7-O zt^4uG*I!aRAp-whe_6LoeGdL=RB%gQ|3iOSykZBlL)nzHq^SDK9W^T&%EQkDO{MBk z{pE^vxt!4$6h|PHr_{2^sxynEoA_fkko9^#(;Wd=nMpLBMz@2HOrxdLyhO-z z>8d+!2(hH|k)PAdW03)UrnGaOXQvh2U!TSAU0}yh@p}`P^PKpdOnWMRr!Ojgr!Ojg zf2n=Qe23(|hYQT3ay#++_}=-<1^8Zbh<%6Qu=u?u7A6WKrqXa6&@Lxf#qSpiHZV06 zTTPe=J&U-U_`M&Nk5&9mw#0NUSvql4{QlwZMa*!-GMZxA!0F$-wxCHFA4opf-aH52 zxel>geRxWs)y_oo9-s{h@ATp6f$8rjnq7bnD!j*spAXbEEzMa#2HE0dEy0$f5b^t% zTX2rEL&MGAiQi|TWFab%ir;(G%rV~~hBojd3vb|iN=?P@AFXa{DED^*>QC@Vf>r#^ zZv*+O`2F=m9hvw%JYUuTDt>>W0t^Z7!vB53I4u>wPg~W&Y=iec2ic31n2O(@Dqr24 z1a?vJXz_c$#AYThtkosqIBF5Ub0jK$|GN}2P#cV<%KF<+1K^za^m-jxed$*AdrQf^D7Ta#qYWFm`f~w z$}=h5tk=Hd$fA+if(olA=~Sz{I!H%XN*K&ef7|DoVsr}w|RKYnfqn;waY z-hXFoy8Pa}j?udQ-kir0k^bJi(YJz`i;&yk>OR8_lBeKr^j zO$Ri%rLX_-do%TJ&txi_l9m+p_h#?@Ohb9NnMh-3x}d)|2i9kCMjukBYP-b7{Nqr! zS~2qyy!ls!2Sc12Qc6jKqJo&DoIO}gA-z9Ma-81(xs+CBK<{?gdH4Kf90nfjoA=?y z=#BZF+kUM^;Iq^F??foo`?uJfgOR{hc*WF411L`8Gu8W#Zj7og4*$9iv-HVUr15Yy z`csSePVdh%45#-mbQ;Snp%$P!y+5rr+0c#T6W06FQl1U@$*o<|L3MnnAHl5j>_VP!fSh^_dmI;may&xVTh7=ZBO<7-74iV)EMj(5S}Gc z2YRxq_s^_XUskUc1AEWMo!);?7ZSm?@!Y*U+l!h_>kVe zSgl%;*i*o!5Y8{6;L!VDo0}=&-vqW;!>isu@9g%1Zw9vA$DQ7PK!<#W^lkP>VCM*@ z_-!YuUZ?jTiYbQ;5hwzE=>02XCrhwW^)Um1lVIKWsW`p=Uo{#_-dlliqmrn~Bms#I zdjHIq+X!6{s?HvLC@eZ@}#`${xr{8U7s1LQTg75~Bq+vO|zx^eTp&r$4 z1ooMaJH3Cc#m%J(jsW|WaDMPyI@SBPs1t6Tu?-VUemG9V3d!mHZ*SaO>ZvLS^{yj1 zy???>v9j{k351?Rib&4s{g1^Jkd?QGKo}JrlhgZGg*nlN2vog)|DvtTov0l9Hhfo6 zu!y*w-v8o;Y$1IK!gosQOO@*M{_Fq970&O#|4|MX*XjLh+0tepC048v>_2c(RT9?w zf4irdR7q11S~-%h_b=MAj^MWe>+R!C@4xWzQi6{JHo@Wi6YJ~!PnWbpdJTldN|N>& z*8B59wd(!H?C;1#yBqAo6oM;?dOp?r(_*!+_un?F7!z$U19L|>DJ_Rkz5mlsmlIMo z5b8&hq@#k~|JU?dl6GehZXuGNwveFr?~LAz@@0pDFgiLW)%!pDT!xTlfw15@lGFR2 zOboAFe+a^-j^yk8Q@e)u0~`i+LUGmmw;x>IKzE&h?O-me(R%-J@fbaq1Xx)KTq~;g z?=~Jsvor^Oe@V&dVRK-V^)8ZA|)oESJgg(Lm%<27Ukt$*7ght ze?^m=e*5+(+F|Vn>6U=w_+?JD#P%Ex*;1a7AzPvTrY0uO<0&n{$UClNj`;>3P>iP# zW5%FD$ON+fHT+U17LCmMw(+R%0q}niF7Que)_0D9_+dP}r#r}onnh&2`;ef?4-aeK z24Ph+$t_!*r!9?}9p;v;D3|l?-Jdry1>$oCC*0N)qvjQ7Aic0q%*Ng(54q`ne3;MV>qmF0Z0-6it zMncb1RNMldq)#ELrlhYvD{A6%Qf$KD=CrN#^pO(yTHYoG3S}Z=lv&dtJ}utlRI)`m zn<+%esbT}$m!+76h<7Q)okX?MhM^iZP`7Rq-2V*!FCAvf5<$YI+rTrJjhp~>AquyZ ziF7!?(bTblUALx~IAnuk7tmCEzjdm~DPx;Fg8T+Tl*_?OZ-PR*p)EQHv6(=Lg}4zq z2Un>4y(i7oK#U#1=}OVmq38w1r77P%6mK%&GZ4@af?E^ZX7iWTO-?)OCBKd4^)&pa zYS0wMI$*=-hWuteyq7viGADSh_12dyY;3wt7z%~%XGN3wZc2TtYUJS+KTj`<5o9_ zF*&zdZ}zn+rZa*E`hY#ag)oT|TDLSgBdnLyuaXIjf&XNOS)!ZdX&kZa0XC$M_CAoL|N$PdVm z<#75g@cIaK7bbKU+E9fWMe5)!$lD56d8a+RtS~-LNLYr&Me;La1h2t~vz zmO0N9qxZV{vi@aF&RnnP_Z68@6j?}t3o|o2)iF5}xe7z%UH#yVpug^&3ZWjmPVl zGw}V(A+`g-67YoAeOy_S*b+t{I1Xq#5nO>A&|WD#VN%22rYzAK{@L4gvXrR?tP@-} z3g3-`E_kx3$=RPmeWTRyDKDuC+Ni0~dr1-=hxM`wz<6Zqpt2jm1#Y4@B8R~*t#|43SWJB2-_v388=T05{nq>K z9A@`}0FG9|orpnLDa<-#y`LWMWS$3Mk&^lnX(0V!Q(Ix~mY3Vw)FDf-pE1$CHDIpynwWq$dlyQR0!9sG(I?3)EBA6X+S06V zm6!_zyo^ma2jK>=?uiaV4JEQJst3HL-yr(OKzmB5R2Si@&Iovmc0t2s=Kx)(49cdF z)S#&7xEp(>KLof_2`}T3g%i4q9$8MnJ3)swodDyEV@Z$AJ5q8d2cH<-)C{4zjcPKD0Ht(-t;vU#h<@mq|7i z%yeRN+&t;X*>Ak83QkbpX~$V-tLDz=!7WKI-JB%&&%jRmI0I6D!((bl#yfi$YnK-QUQ+`9 zKquaNs-wvn5aVrbilKB;dsu?tBrkCr@oe2njp$D<)WpF0KRizro4d&jq4CvSO2yFV-7zu_N}sfzeP|^s<|x zQ!Z3CxvPR5^5CM*TwpyEr|gIEt--rCVRAJP*vKf{(&Lp~3OSe|&q{?p0sOn^uB+BkeC%! zWd0qHa&vt5#Cqk*bwK+CEhC!BJzWHKSge=&Io6pwgVsNqsfu)a_c?+|EV2zFkZ)qW zZ9m|r`xN{yIgEex!h{2{UZFfFXufP{%HgCLaGfPqz@DC*KV!Yc(=&}t2UyP$Oya#g zE18^xpqDui&n+ANHz-#5z?^`|9UNR8C~o?|cc?=wH*0?G)hE;QI1p@pur)WsDd4=O zJZ*(i50z;nO75s&(_tmdDv&lQksDg&Y4k$P4}Mx8#_R!fLSc@FIlSK?r)tm(e%r|W z2|@@zPEuBWNWR=V2RnL>Z7gR>(;z6CBzm8YFPAepcLlu{zD+i*LAY5-Ts(R(Vf8pO z9}aqzyFrQp*4Texa-2F6;%~--1_5?{s^! zEk@Mt=z%w|=fJ(Lrg?O4Q_A5I9LHU~kE9vJjt zNwWk&Qi0c1&h%T60X=%4@q6ja$pzj+Ig5#-M-Mbes%-`WdsK0Xl@$Xxw>b1KHa z@b5+;pQK83xw^dpD!D(rhdRhkBi!ZcAfD|D{~uvj9;Z|F|Ic}5c{~_njA1Y{_I1qI zmtn?k7>2QykR|&XlBkfSWZ#L(Qns?MMNzgSS&E3VWerJ$lIr*Ve9m(|_qp@c>-Wdu z&UwE-XS?T~=iK-CoO2PO^)wI`xJg!?x;`0h;_HBIx40>pJzfCs*>t=>aiCH2@Htc7 z>3*wd{NOP@}}M z@1-w_BKuytlGPE+*h15E?l*uq`(B9mU;AEK0CV=eRKAvv=Ps8JdmF)16xnMogY0|x zd$c)A%bS5$cLe6hz89*A?0dO%)uT!yg^Hk6Cu%vWjqH2Lf4#n?`7KXifSGR?^u$v?}hHMv+rf=1&^u$`@J9@Au`vGb9DB-d|QAy z*MR?GInKV98Lx*cPYZ1QM&SDBI#eKMAf$aSyXQ7hB|)g>CYgco{h7^G=!D>q77xp- zbP!rulC$q6cX~tB3(=D;q!W(f?Z){m!$A6z_r0VdciHzsPn`CVmF#=DITcZQ3!-+| z#GHLEb8_MZ^BCg)v>3PS|C5{Sd$}79M~(27vrnaplTf6V2XD}v4KH|#0m$i$Kg=&RmUIcUZPva2}Xf!7ovs(rGUXC;^sM=BL zNib?4hTKwygnci0)@L!KP9XFV61SEiVc*M?j`4;x4uok!q9Qo^Uf$hx;C?oq`TF#AOrv3x;f2@h~S%!TM8~a}J9<0hFU7|HQ zA%Ynm`0WgiP~L{C{i&q_ODCMi1cPJW%dLB%#=a-8{$eltUMfWAGx!_8rZ~9W_fqZk zD1)y6w#LEjzL$S-|M;eV0qkoBxBFfSrI`^>Uj=r{!R@}6C9me?r&5Qu!6Gn%sh{k7 zdF4@zNxuZJvJP(dz0_$6M{jf~oeC`7!R@}6D<9Q1_C101cW}G!<=l=M27d$C6bHBa zUV?iSG588#8!c|eOWF4lIj*@O9RlH`kUCK>u=`&2mBz#3BT)9eeB<0{?0dO5D@I)h ztaG--s~>`S9(XV{w-ftb_OEViNL4_n<0i?zmvc_KP-&J7$;(XL8}t{*#2KVby4d%U zdNb6Jrh@R6nQ^5n^9NSi>|nTcfgy+n+QGNe-={FI4g_r3J_tC`vN@|Xxu$z8Ym zUY_p~ZT7vy04pN6?0d;wq5=GW5MQ4#>hYQOz5Kc@mwCvZ1);Z)_>RcFm+yK7o4!33 zgsDX0mN6u|?`2JlzXPrWVN)g&Z)w>V67oBSy`z|H>aRgKMIco~PmZhhWM?Ke9^pz1(Th*pO<1&{Rm2ou)r!-%AvFg6w6 zE7?eSegh-4-T$!s50%~bvhYfRDQ+Q9%e(1R4JG?t_U=eEV{a3%y1SXw9Gc4DPu=c& zX;Ba#YNEk+6!^2nhM8jH?0b3j*;v!!YeC&9beg4TbC>PDm+@a#Hq29CT@xlvRRWpV z_p&NSx+wwsgrO;aixQO~?5MCFOsfXS7^4*$blhR@k)(eTU)10U5d-;1b9>)D(9I`BD-^+*F zp>teB{4I-_`yl&X8gC-sRGfs15KwXCj{n*S-t##6&rLGZKfCXx+2csn6od|L zQeWySu6-{=u=`%<^}_6Xq4AkN%Ro`QMn;oA_r4dJ!SKG9m%7pGV&jS`?gXspIKi=W z@$gcpnnf3nKE|t~4{tL*dc8?|Xz3hAG+ODkHg_zF3`?@*2@5Q|GV&98- z<)s(;sOuBqF-6|Mi(rt+Q+QKh%!3Hq5r`R2;bmhnwIjb+8U(Fh?7*L+RNvJVqE*rd zQ|qu_?9?83*Yo?u&ce0zi@gdf@r#{;S9z_YI^stV!83^NpZLWVYndDV?}%?|F-=$c z6?$IA&_5H1Z~bC<1hRgyvv87u_O%1u`o;3Z!Pl2!{C=@C)z`jP5o7&gx4nRuYP}3G z>k+s(dMf2%{bCcBVKqnZ2Y5^fvxy*nu^+6*yYN+jzX^d$Y5iiq`W<6K=(89E5S$!L zRjglZmPt75Fdl@OLZVy(Nk4fP{e}IYEkJl)NIai@QX5&n*i~(D!tZ$S=ZFo{1B!Ut zn=gM!r9S{+mz%`x&P?%(z3~&K!zV!dS*T7K0;uqdowgr)ss05mUng5}hw3CPezC_B zlGOsLw5|+;+G51_L;PYZm-hNu?ZN0FEWRl`_b0#D^i9!*H5!CTLgKl<_{EN$jZZ}2 zrs-uMd`P6lbZ5mc_V$cKvo^60*jEm2{bJ|TN#&lZF9W;b;MOm;?(G@_fMr3|AtY0jT%*GWBezA{Z)0v|?f!LkMrsUQy_J?0e8+;V7 z@q+V&ko;mV?apH|TLf$w;hdSl$uG7>w6Wg>Y@gVRUu+J1cYw;M&*7iT4sQKoKTMBS zqz37Kfcd)k)7Q(Xd97b;e+)_5N1*t{X4%`om{mb+9Wh(Ycg6a}&RrH~%HAG?zCz;S za0cQR`|%&840AkKbKFc$Q~YA*2H~p{AA+$}Sf;C3zu1l!YnY^tf^bGil&Rl$6@Iae zuT(Uo-$D3SNOb4@B>2Tn=vC8nihN!1vVdTQgpJ%Ze!p0HkI+5>#V=O(3NmT8Kx`K= z<63hm#4k4a1b&o6_wSKXGr!+L2Y~Ao9fDn+7^^5IUp|ILZ znUw=xgE)ar;`fVfR2?fN)c9>cXm3f5U+hnFLJU3>*vk%X{bEnggezE6v;ge`X+6cm@nMl?zmYgrhkyW1r;YUky{9^alt#0rKz#a=Oez7TKaC&eq%#C6Z zOi|r_u_eDpJ5>T$O91jeulvxv zSNTq3PG(S!<6M8)V1xxL)@#)aq<|V zf3c7bstepMJe=)UP5BU^wHN8;LSTDCvdbag>2&M?x0qcHS%ehd*Sl6XGtQ;kfxDec zyS@{FcP>PCC(|=jEO9RV%jH};d`Vf=u}Y+x4B|{7$Fu9{NN-gjEfGEdvYyaeloU53 zwWsn`Mr2f23_cx0&ZW=waXXhj-wOq!!v2?Y>8cOPs1T5{A$Y>6b;P;!-_}KxE{yoH z7Ss8OAkL+=F2tz1z*@R+U5H4Ib7{?;VXB)j7<8OVd!f7@AC*#^OTR{i0-Q@D-ma`l zAjJ>C*+9vZrR2rAv`gvcsv2Sr0y;`?3c=!B8uSzv=C2_BrkILz>A=-!zuu@i0;g(9 zF3zPV##UB^fK?H^KGJ5A+qv}M`w=P)q-Xw@=s1`Dwgt|m&x7%@i{&_%&Rfx3O$WBn z;-;^Ob1B_uaV|a4JC903X*|2ZK42{z=h7cPg(7km@z*V;i7w8i&EnFH2dBmj$WAwb zn$`GF8VA&7Ax(QUGo0eUtGGCsoJ(H-D9)vG`qEa$HiHzu)vmc-KwbEhI_fTz-#UB_ zB!|zcr&HPCb2<#H!{?1+1=Sp+wve*>r$0(;9X?<9IaRGg>~@Q|9X`k78LzbsvEO9>sh{KD`E-}EYAeh? zp(L(y1Nr?s^J7YP6tQP4;`Z;HvJ{_N0Q9H8x5(Q1cT$Ev!mNMiaB0x%-g;TGs;nqP zB0^;Z&s{e4`*+gtWc@pD91LSdJFq%AOuv68bq(v^S?_)Y^#_s}3)Y*&>XMFsXNM_xzt_3? zV}TXHlR?jm_;>zY51Zc;5uao+-6SXWJBxp3*2_UkX8`Ohg!c4EF-!bAmsE&VLqV7* zq%K72N&m2tT=CW?P&F?>;WHB3H#f*H$I;@`QSe)CR(cGXSgVu*j|$8TT}?lBnI z2l#8M>BfOLd9WI~ZK$IJwgvn_-#K}WQvIoIuLHklIhqT@ zZ{y_O8Iix4A%zXZtw-Qnuem&i1pm(9Pm>I(A_yr$qB1%DovBq&t=0g$34xoHn#l3* zoRk4)q!D0@7ZxWY{+%>lI{uw=C*@_PUIY3jG8sZGCH|c>GCTgAr!U4bNuL4pg3Z9J z@R5IKJ%(MXI6oBzhkxgCe6o)18v|=D z_B^*G|IUK_!i;@yU<1Y8;PCI9vN_z?zX@!b*z+u#{5!XQA7SuSz}7jq_3!-hVIhNm z3G9f2TmQ~V*D4wOI)*MjL{@%8bSYqof{TCWaOzv) z-$~<)_9cTR{+-e1tFfK#gV^UO0d7F??;J%5h<|6?*IAjQ=Y#n!vH4bue`ozX5Ue*N zewR&2o33U3JFEYQ{Xi!Go)^L>?t9k1bKJKHO5X?gmnA6i?`-#ZVRN&h24`2r5zKSI z_m{H~|IPxBN*QKturl0C&M(^U--XTUfy+daUU62UJ(1VsqA-8^OnP~qQs zU;2L{sQn1A#j_Cf9I;M@E*-E6z5+en1@vS zJ5!Go!oe4aFE5z&?~J>bPc=Yn8;h9VQNHiizq96~MDB+D!5JeweyrplW&Jy+O$kwR zL0Te2?r7G(vsn8CwFS^VfjJ50h<~U4w~0Cp!p}nDo8-7Pk$>mt8jaOIAmqSA$lQ4* znRPAe-}#9rRuu)Ix{&0l&E((tLu!8246Ns#z~nS#a?9`e=9S~ELFb+2eF^V2G}Fy8Lx-(EigNzUyKS&`gF-h& z(5fEtdKOpC`7@M`kH9CNKhx-HPBx$JGS~Fyf1E#a@2PBRr?4q&lcStJ6IrQ}qB49> zP$){|oImrZ2{w}=`kp{?-q{*dO3pi@Lh}kFU8scc**W+`vP&-qoJe*UO=3?Z`xe)7 zA{qT5Cz8=0aw6GhtFo&8h!}$4iKdd-6Uoj@i&t+TcBVyi9K-$-$ttB3SIYry5V#n@ zaw1uwCVA9u!O9SZFLWWoo=7&gV8DrFYlh=zoNJay966CJ*PUW28?XojPkGMYe+uIa#2Jlh(?sRn>D1g`7AGdwxk)>hpB4HUSM1JCiiQ>2a>3uu-L)|`Yn zk?aqhtd>~R-$Ct(WM80WKB|$NNcQ;I2-OT3oJ8c0WYwB-lM~6xj>@H|^!EY%MQ|sA z>uM{sQl6^Tdh>0aeKm(ppgnf*N#Ns3TUzqEGkYUd#_%cDg{DyAyHMB2`#K>O%>Bbd*^?xRWlIUS(3R~ zaw6F|>TYr(S@MhoMbF1*uqRmybJ$qSi)Biw80;Tjx1*|B3~asNWRdAave_e|lxIH} z$AsmaNJbB~^DO&lBufrymSVnJ7CgcKO|qV_f z!0cl(rHg=n22uNSPV(I{dJcTVmG74Qez*uH@Tc3Bli~!pJo4SLrv`>Gz38j1tc31- zx2*7ah|#(Rcy$mw!>Gr}cgrUK0Pp#y5Z~2edJGZdyJhs0X&<5T-Ljk0P@PGLStvGi zLU%pDg?H0!sN9G2W5>8tAE1O8I12 zu6MEFNyh;#EsWX3kWZFf$zD(C+5j_zKuMUq<&$N%ves1HK^P$eZ+c883@ zFBJmcO7q;>PnJbuANm3+acSU{Eypw#YXkJjvIUij7*cZ(o)Hqamm%SkW&OU3G^7_m z7$qbslaU5uZg{+h3Su5R2jDUxa8IDS>SyH*^2W`A>ahci{lem8^bfq{97p^)!l)%Rbsy)GWv62BlJz^l;4y&& zZTgq}WZ8y$m=@;;qmX5((ARUPsu*mne64>qQ-m78>JV;j0Vyi<$+97H3mHYFJ+Ll< z+fSCYjZRnkMMS?wkW)hWWZ6>gjP{dduXJq8OuY*9&6!N>C(G74CirAoxgWDKQ~v<^ zRWhLh`rE;NvMg`Gddl+{pm(f`Knvb6b%VWydY3Ty76Kkm9H*$v!MphU_mfRV4S;93 z(_uv?rcT<^Z17PbNd;7_<%?S;S z{Xt+y9eex9viEZpRrJ!Le+KreV{boM7LujB`ksFu#$gj2g1MdY$+F86Fp?BUe0jpS z5lmmPpDdeE9bbk_1=vyu%XzS}pDe359Y-4V0yw}D3{If& zypE-B1XFuH(!tze`^mCLDKJum#Z`3LNIqG%d2s_wqe1H=R33jq545eJYOME! zMBXPB{X;1+WIcqS#EEUUb%ydfkJVKv_``^mCq6)HhT zL411z)0i}O^nZ?Qb;vkuVC3KH7l4lv4n@f)%Z_&|YB=+OuXJ%V6Xla-2Oi=(79e~r zB%W6}5)XZ{tXQLPbrFQ$+$2Yip_w7eG5kvGyN-E(TDh?~v6%fr0hN~CGfpM^cG}Au zZ-d^%#DC~|6jnCg){4QZhTV8uuTgBk#@qZ!*jv>RzumnMJbP#WzD%tp8*i@-PQmfT z0N)fsJEY)jyp3rTq>jU6DeyJIDX{?skd3#=9W-0uSM2HevKTGvNjH*w{tQIw94z$dFwUw&DY&0~>)W3K7X|4^gmP7TMYNR~}mn;b=r z`X<_|s0=-cF*QOxqM9MnlaESf)ui$gp3H8Zo=1ssyN(z{Lp` zHEPRpeD6W9c*2x#6LPX@RF`4_YE+HMP@}kJiNp~#YDi=bwE(Fsrj#mh{(d#;*Jr}j zX2kBYh)#CkDW2J@!_>Ec&I(-5foFJjwFpzc0eUQOV+Wq&sW?8r%84QsLf~vQ!J3mW zYSf9Q#Z{t3GpSKCQ8ORaNYtqI*($2$NTE9-`;t`~%1zX$imMtZD*ZS>6AA82u&7bo zUnySHsHG*nOx%LV9bzDA)bC@VMjb=^Ny0cUQKR0O)KuL>^aBg&9+a7=QOCz5scci= z|A@e*el6x#Qg_PYWVrFwRXnhoF5Dy{YLu@BcE`5_qoc6;Ik}J;)v>M`4D40Gx&1|r zDt9=eoaIz09D%7Y+tIMiwAbcSts){x_Sv9IjHjnxNgsYa6Nsa1{9^j}^TlZ*1 z&qo*@t7wG47O-m6hacxv3BXbWCyPvK)XevS@%s-(Ct*2i6g}9^v+SdhEO|_i{6>wM zwZN}5{f-i|s8JU~f|YL?HoPEcx(N7ZFtxv1jiTqkM_f^(9;V@R0z^!3+j3HzfLo1v z_p1n|Z~PCsqelJt1#JQO61*b_o|mY{i5m6xFm%bwi2v1MdMpt{jiRSa`v?^^>f%f+ z5l2r)%OY@kny&nYPgbMW{+NQ_Kd`#GZQi7&F*#Z_in;;xFYrW-Ix`BToP?OUVm*yY zDQZ-5HW2h0pdSfi4lzWHS{+{r-#-F)RtS`Y$y?N@m{;P}T@bu8Y|SVYCS_8ies7ag z#eh}e2~4*d^~>=RN;d-S>HnoVYSgc@;afBSjFG~ck1lV@Cu-E#DwqMx24j(Bxz(u8 zR+do9WJHbHh+hUrje2)nDQ46K6Y6pQGmDEIy`k-%aWB)U-U&UV3sCVclh#Hkq&X9E8SZFHq@y2Wm%Y{8-m%C*jzSIqq@wg!&&Mc!1_sQqDE~>N;mee1DovFTQ%ym z$6xm4z*alBRijeR7dLg=5A2YGTQ#caxm1(>6<{|V+^SJox0f?`$ZU+v2>$Y0HEKum z2p$7;abWQdZq=yaRm&N?A+V-`iy9R^7cN{~5#OIM8kl$>anz{9>6q1y1vtw^(58P` zH7aoqJla-)vCgs_H7aItUQ>iGfgK^-+ybjcb=XwLs8QE}-4@)cQ8Nobya<|up$H+M zgrY|E=dNqjsNm^YnW?LSUN@78RilPGCQzduw8+U!-3RocWI_e>w}VxqQr9(Ao=E`b zx(QZ|x{DKKsf2nR{@G3(r!lP>RpHq}CZlhGpCyiyk*HC(aEuuz^bmxG`*=47>LXt4M4&5Fjp8n6_?`R69os!=b^4Knsm1AEr7w`$bmi8U0z488hpkTtrj==kGP#Hv6{Erpvldv*-Av9-)Ton9YZzv4u!g&t zJV=TfH58})(g3Pwfwj`jbOv2fqt2^FhQ1rruU&LUjVh=8xnBk0fs3T6T8WpRt&kFqp^Ks!>l(N>iTa!Fbul za@42;)ypW)41i07z-p99#;Q?CBkC&8HZTqei<`}9BT=K~JeLRh7ihPHN-QO6RLw`t zj2adC7JdW~{N1?#zOh=PjSks4D?c}~rv$L77U!nZ=000BYWaMhVYUFP%afQ^jk?{f zftgsn4AvxK%5$1YjT+G~hheS;D_{yHYEMNBnilf`<#5 zYKj{5&${ZSz#2bk*%1tfno`uLs^`sbtEV{dDlU#@qNq{N6)U8gfY4q@Jg;)AQH5WL zRnLPk+D&rQD4H3v8pW@~zG0Z9r$r11Q)xFZCaAQd{e3Fx?QVF-qkZ6ekcjLNr-R*8 z)H9d;*ttNAk6BA9vmd*l;bZjtLQHGRl@3&OUU@nZ>ICA$5IoZxAStWrY%5ChA4Dfu zNG~MZsyZW^mr+X)p>-n=TDeJ9)tT~EL#4Y}%&Iz6#;DWhF?A$W=k(hFs?GqUD5_5F z-{aI+7_UdrD=4qsxN=k-O2CgXAb&BT6qY9%qCNr`%If|;I zH)%y>NG8%6Y7Y6vRb<7+T+pWl5>;nL&lpG5p+c9wGz43g@J+=t3)0kwFyOCW5si-@ zUx=76`paJsY80K5wsDlc^;S{-Y*y;U%IavKp(!#k&3h(gqY~cs6=lt6`QiZ0X!<*j zW;6%aq8ZU2q8ZU2q8Yun7M{F_sEpv*MVVO5=rWMTh;3~VeUM?lX7tpa!m1abAp##I zSTv*V*_0Y5*mr~}-&Iu1YDS0R12m(7??N-;nk5oPG^3Fn!qs7jCg zH4&V(4ueB8dOSI+>J0387j6;}&8SHp%tu}aV}`KuJGnqJdKw3vF9)_kaBhFmjNZc7 zxF6Vef|GqBr!CmLim(9w(a7$*w8Te-K~PV!8qmL^Gm$p?!pk zW|TG*j(F*a=_odICzj9OAt_i==yh;PF11zdoKTm)_UmlcjCwS>BJ5sYh=lK6X)nMdZh{q#3gaLhDxtoHf9?{;DXJBY zMx3ofXd8R33z?MB@H!@IJxbu1hPLL|OoQ$|mBtF-@Ob#n=x5 zHq5cN!qLWiX~upAuz8NX6^@oir6_&{*Z^#sV{e6{&*!G7iI^AaW57;|y$DCo;XpZk z3-J%Ft-r5W;b`byj96i-u>OT$TB$t`R#rGlq|;&(0amsI5sqFdm2QTTW`Lfxu%;o( z(qXETJlxGzfn_VnNI4W}>oAMk4cv=W79GTI;3P&f(=2M_4z?7E>s?WL^#U(t(g>hYJ&&3)*$mcRV#U`vML4?qQi9=h0zS~i z(M%NKsABViY8(i&g~anJw{SFMF}{BX!pClsnN3w2Qcu5(52U zBm`3pzS}GzM+jzM4>`XkVW>AZ7|W}hrhnKzLS@q&@IqIxcs z=3N)A(!N_?NJXVwC~U;3&PYYXsW}OF9HNrOz?+YRt5*WC<%A{7!iTwZr=8ii0VMI+X9mQD89NR{c(JAn>0nLq|5I@GQ*LP8(?aW zjO>;O=cr1ENJ8+0q=ILUjO_XmezTh+`WXwE8K^xnvQ?oXN)H4!!s0p`*~^iU>oQVR z1Y%|Zf7i|7BO}AMYSM@*^3SsJQH`R`U&WB$K6{qz-{3Df=sz!|8`ZNV7MBS7B@l~c z?ESAp&E4>=vv#?%L|?^sJ8`F~!ugr*RPjLUE@4SqeujHK+GHOf+a@FWVojnEBAOw1 zwo&@FO>P`aRr)zZ53rECO+MQaj`c5KQ!TDPC3|U;=mm|`9>lBwzR}I$HaWG|-zIc{ zj~{>%b#xNekg5Dv@R#($mvHWciqac!F&v2rId8>jY)6)Y@E-oLoLCjAj? zu%tjtlh<5AM%zN%d&xKONxhfWb8PP=A3*NC6seLPPQ|Y>nlo~P)11M1@jE*YX}yEs zDNkA0<{ViDJBl|V`cn(JoAaZJwUzz`*pC+1RmonObKpy*R5D`j16RnClQ$f0PU8=V zE@u1zsYX#@y~>)ZSB8OfF?ME*db1ZMdDLVr0x|YuOlY06F+X{e0hwlTKY6cLv(55x z{5D*3vrwc;Xz>*Gq+mSwLQLJ`0r_j&7QKXPHl;`v^;>udM*KQDd)bXxP1H0Y?^s_K z9NDI!qMt+_$vR?Sm=^qR*@~)TZC!Z3roWZ>Pn-Wd=D#5GU$FTv#QYbE`}Re3yfZ^h z--%30b-`<6K+BcR=vegZwONdeS>+$w7#I#X`Cn;Z{xGv>=~qjR>vh zg0RF*vTFLzqi~NmTFl&k<)hk0z1a99Y*Q+2z;`dKDS8(o7aVrW3K7X|4`Elcg?Gw7dB;WauhXvvQAS}hDk)~ zOwA!``os5QxS-1giX@}qz_PLqTlB)R#E*6 zEbBJEX(pe~gFl3-g22kTaFdAW_s1Ki;Qa@T7Q#C0a|7dBP zdL7tY!O1?bsL=24_6}Abg0MqKR26Mg zG#;z$2;6v?^!pyK)KG8`yewp3G|MqsoOmZRU(2;@A=J{p;m zUz!wY^!sA_{M!6=l$b?&DhOXt9r%L!5MR``7t?W3iV_OY@2k`F{((jvQgfb@jJy%ck```a8p#FwgjgMhU7Xkkqr}lU2_w*e2h%5Si&qnz;fmi-3 z0k?iXxmh~Xm;VRd(eI;w!Nfy<3f>+B&uQv$qTg4~hgsG4h`(epeVGWN-_uj3eT0gB z-(fs{wzF@?n-Bu0r|HUX=+^HEaP<2-m=QNbOnbNC4Qd~gt<~?TFK8d}M8AKpFp4@3 zG1J8Q4i#1O`{$Usa8z5sYq2$Y1$TlD+Lw$;>45Y!G^JxYa1 zne_Xw)>T$fV3m0S)2-hJ=fz~Y4rnd@m+I*EVFyV60ArZ2N~7PK@`-+5xo{b!r+_iv zvdqJ4w;@07TwFb-5^n(hiREZ+EYa^9cPwv6-+^#JNZekc-?wT~-;f@G5b}v_4l0wQ z-!DF17U%y1j28lT2fC|{ejk4YuQzqUNEa3-Bl`UT{5&|0pB)DlWu_hq`peeD(eFbm z3mf|V{0F(1q~8N`6|p^O$XxXMJU3EIYI}fvDXEEmKX6h>WB(Jd>tZkZeInfi(eJlb z3uBVb`6)zD1oKRZeqYI9L%%=V*2^Sa3(N+@=CX-?zb&$19sWLt$sfeui8{e{SW>L z-YMm``u)N6Id~w@alnc>xYh5+e(o`NEnp1<7yZ83W2^?YM|=;$XmsMy#L@2`CF2~k zR{&0S5wz)FR==N@h-8+6vC^^}{l4IuBvXXlzzz^@Zh_VBXFQj{L%Y5R?3&rY@#n>ush-A>b$7>4<(mKWCar=ne=EY(kFX z=Y|UDCXt+eHVKiRku=;)0 z`{;7F0sd|YqTfdzDrSa~u+QQ3ieOrqhA5MZ==b%%Ofbx9U^Q|xIXltsn~n4{yMi^? z&E!E+^!w7~>zXttgSEuXbOv3~@4HX(C%gmH{Vuwr-(UH`YtlRq!mloprfNC*eW9&b zBMaG$_ap>UZ~C2a^!wtonkroaU|CCW^!vOOpt;sVd@GBYS%Z0ut$zQSs;@k~!Fa*N za`gK*-^4P&M1Tu~!1}#O#_IP&F5*Pv^4h=KmP?GWRG-b>-f2UWgNGkry#I$7U!nZ=002f{`rPMhM5M|Gf!e#{r;QW z!De>#0$AgSDFZ?#{l5QXoCZO&nPp&^`teLk^!qpdXkZAR6CslOxz+Ehu56|}ClP;< zvf$x@rmmvj-yel<4%5%!Bj6!>Er+5+zc0R98%`YXL>EUhQS|%Eo^Vwkgw{ghd6iqg ze`-x()fB_ zct2Sq`ol;H0Y)q-TI!k3VWo6NyU)X(p(U56J zn%7rpvqwa$v>)abS5Xaz;E1~RTSKxxi&Tt0xGf~Aen_kl+giVfhbijCAWUP4G&Ybl z)axge9tcv>()E~1Lsg61j~AS5m<&0pS{H8)Y&VDPMg-49y5ZJGwQF8@y?%k{Z!Dze z5N=hqtBFO_LPThN6NCqDl2z3z*2ZhJ7x{A9R#l_IM(ugAv{BUxei5LmHARZzqxx%b zZdC%ttr7GR%Bu&i9950d@e%lBRjpTrQfxloWv=PZ|4`L7rF+$WVN=#7M^V)pU#zF7 z3>OGmLCqnm+UK*OG9vmffkaiSej-3sqeAnWD_y9BmOX=6g}Qk#K!y4aO=8c%JA-Rc zq391$q391$p%Qu&RYMW+GJ@x0Dw$QG?q;p3rXhBLMf6^V{VLSg|JG4!0evj+L4rkv z>To<<9Te;+VbDbpVO6NQRRdJ0EeG&mbIlToBP!IN$CFhguzU!f?>K+I3YG7*!YTo= z)h(hgIPert@`GrV4(MrtuR8DyPqr;u^#e3a;9Cwn$5ZC-B5D$#g)Ufg5=Mo()+S1= zvS=n1Y6xoPqZ)|{)wyUbbr%_&L*!+$`h#*46)Inu+=@#7H=v*|0S7e!EGiWDSBe)E zs_EWJOiV;%Wib#HD(e)uqclN$3&J=rQK5R>Xr_7~dXR;54$4easM?dl)a$@z2+rz> z!Qn;PFHe$M4s5FnH;IS}b>s79N*@B_q_A>3xj=#UTjPHfzs8IA^JI}I@MzZ9&qe6`eHR7aSVHyZa zmYS#_d_i^K3+h9BvEN=qp&?6ALIEn&TkTW%o#UymP(1z#6fe?xMd~v&n4nB5)Px1` zjEM@x)e{wJNqS>74rZT|DP08olb_n(twPar;3KZ6P%S>h8avq?a@leeC*W40&K|1D z^!WdvJ1W$l8_2n`9(aurJcX#oi3+7ZFN8%b#P_k7jwgbsQ1p~(AEBZ`y_2IL_CF!! zJ+YxHzoA=&B0yBA`NQCjbqFyR+=dmXeN47ig`&QoeZ&(LDuj;r^?i*dL*Q!YYE)EF zp>jt;nb7e-s|q897@|U*9Dv2sbbuX&KuMUqMTNThTcjEc!W%-ORG5@Wg?hV7G4(cB z8=k;)t5Bt5aq`F)pq>0*s-r?xh`_7#EifJnYZUsuDW9lND}TespR!EK#9q-3m6O4j}Xt61SJAP`%3(H>B4okp393RP@vRg+p{V9g~pQK8n42sZY;fejRUQK81sO%N3-V?hLy z^gJ-%k<^M!^&f`!ro)B`HKI&OCh2`(ekEy(3e~txL6h2LU^gT+QK7ot3NrS=N3jTk zVCu$a@5c-d@>Xe_YRX;=SQ!VmDpc0qWewf{SQ7`gD%3}7E1LAX1MBPHR)y-ewXng* z1DoXFR)y+%KQ9jidKs_}9o(u=%P&V8d>^o{1Q!+Ro3OPJNb@BvU zQuHyr2_QHGZTgp0p&HbM*Fi}z5-iJ6q1F`ArU;FIH6xt62Y;fG4#TUuryvjQx)-nk zf?E~pcrcP4i|DBYIVBVosu_1(t3qW>NMNSk3i{`nOsop^jAH^7Dh03B#MIaE&pk4s z0{Yv*s!%`9Psc`@Z!rEL1Q4tW)u~TTlWzj>>cnx1YE`J8ersehY6-lfI~`G>Lb}G9 zgoc6es!hmIq27gN%cY$M>>Znk<3;MdRNvTd1NNz7Z@owt*J)_%PXhbVv9~Hz=n;H@ z4)Z?!0N7*4-l|Zs<&xBJ!t;EKl@J7TJ4J>1rY^>Yiil4kj2pr96{|vhos8WKtpRot zLK_~etP0hrL|LVW0({vLM1|UXHpvVnvjM$tVNFAn>5-yBExFj< zQr)C^4XlT5CJ&OLLVZ)kW0(=g@vR{QQ$5bz8FWR3deAMaq1OSmsf+HYP`Mua)9eAl zi!KsZ%Tb|T`vD#=lfjrLEc%^sRH#jr@vFTC;3i9ORH$6dys>1e2<>`p{UX%qtDKy^{73$}_UQ^)LfKL_bIA0SEb;Y+ECMNiP*%{b(*#-Kp#C&n5s$K;MAvAKdlH!GimJdp71xcFo$SFeRQmH_2C_8;Z) zW)zTeXWDoRt#jGm=)33IIXhoz;MF#)Wp#Ehmi-_Sh&qN>j;< z54|foHD>fMkmG_Dr5DDGUR7umkEcmx#(M8<9G6RshX{WQGoGfD8JoN}j^$-E>I{?+ z1b^LWO)+Du_oeFQ^}Z62BtiMH$mmwNDBV?>sby^U78sU=8P9^%o0#0aC|^~S=4lz- zD^p)H^qt=2x-`?KfI8nMteO6X#(~5cyS*Pj7sM&92fV|bqM;n}cGy;cQ~4I~S$8V& z@zAU@PI-&ftIMg}2mF^!MH%!5?|biM;RGV_#NfC&QBHvYy^3nK^)a*q}BA}nQ{t?nl&YFSxdH))CNxq3R zexr1zr=!nW{e0>kNN1rii9ldTFC^UR=V$STw*(P*{|BL}n`HI#H61I$2i#&-Ki`AQ zqOL}jH2V2lX9M)}l=6{|eqM4&W7QMJvk~+P%4<2U9Q~Zq@e%lB{X7;9*kpdnWv=PZ z|Ip7<+GkTCD2~aDY)y`$pO?JdR8biU5orxIhv?^12IuF3))Ywe^UvUeE+=t6z}eJp|7#s*)An3vb0DJE*x37|{nA_6zT$YBW?O096q9 zD8VAU&s-L(>IwE8Vam50xme+STuOlOo^T#I0@o~&I3m3F4$^80Qkg|5{lNMAh4+4M zK|af3x3l~?w@xXdQ}{-L>F!n5#fEtiUg$_fRQ1r{7xIev@Eh&@mPTwAIg!jKL;`|?q z!DE#Tfg3NA@LuLzgenBAyx?S!Nq8UgC|!9{!DuNgM|h`4-FcRMGz2CW_|t#n=V!n8 zMfGbaF{|rT5Wb)~@CEfDzNlp{rs1L#B@`gM7Y|NmndK<#GYRi?`z0}SgP=^p`-9OT zZsDD)C&K$(9RB7T)PO@DW#p_Z=^Wa{@2_R|0O~{m1I1nZE2l z=#KE7+zx`Z{sg?;2%Zwu<3xD>fb^U7PI^W<3a; zo~A3mp<8$-K!o=h=P`?KfS708hSjKjOtw~dr@o+l#1r9tT2s=$5HnS*Q>dsSyk9Mk zxEID7$Uq+n2h5Q_5=J*2$Y1$TZH!mCrhgvAp9#NN`*<8gm+(wk}CIQ{B$9h zhH#i};r;Ozc+S)YE#rTwj_{u67r4*#0OLhrMW>^&xot#v?>z#wm<+}|%W@0vudfYN zX;k9%z;{@V=Ef4?eb~(=hI9gi^FreG65;*Yi&+fmPY{By*h*5F9O3=^k6AFf04yT} z?hbTU9pSyxEwJi<(NtKRj0o>1@blmZ?=$8^GgH3+`Uq>{xPM005jKSPFP79`l3oht zN@8bF<|4emkXpv1wj0<1Nlk?J5|=`Z{Y7Bc#9oB=eHeTk;eFyW6_}*MuhNvw-zut@ z72XdzYzXf^wJpgcodRZkVsqI|17Xe4sM0_`E^qb{x`6oYf^qIynouD5f219 z7FbaSx5E3Xzf%pK0<6B^BD|;n2@jiR5Z|3J8l8AFafJ5@^B`fo3~-W*piTd>!uzWJ z@ULD1#tO@Fg!fOQ%9$d34(toU%`LF*pKtXy-e@)NhJJd^nV0@ zb6WS$BUyaLz8tVBg!6BXxPOlP*l(W!tgT~jh4&BdG*vvie;(M2j=dG$Ct#b{M_m9Kc{{gZb^&XUo5_Qu2=AMr2haej zCxNxt&2$D`5#9&hs&DAqLEYz~JHq>L9J9)4o&({wi=?Sqj_`iq2)ts0Z{pKG2&Ufj zJL3rN*TzCoEDkW<5**>Z#QG%Vsf+lQ7BjO3^B7yuxLhvo= z&er|2#Iy$7ex5jBi5BOk)8;-~;r*Y^=10xb7_4?rVp`$-@6;-0b~Oa7vBZ=CA(Q)O z+RU7W`94?y^Dq(K2WBf`2%iz53*Rp*yf=EMARM<4e}S^#;ew{FBD|0LHQiM0PvF7$ zSu-4JN)g@%y;$6EVu2^PIGTwfykB{`tf~h>D0^ub$$;_t4r50dM z&ecCvP@Z?dScAao0R3Uq0jG#Y9bgHFrjx9q^UJia4=R~_3iBNe-;aBWDZhOu#6>`IT=r-Vk>Cx?U`}vRkCXULWU)oR1B%P_WGJBL+S& zJMHbYqn;Y{5chH#2y=)umWmQbdzfRlhHXK?b?Ev&g;Z=8?M>d2${hU(h`WVct5=Vq zDz>-wj&D-V;6DJnD0p1zow6$Sqp+S4D6x;?Lwyj~Vh7-}6d_I3RH{hWuTTOIxH;=N9 zxmAu8Gn=c}$=dtq0A_v@5kJRbni|23QnB;3_h4O2qCW(X_!cxk!7D-o_iz z*FaARV*xR~*L1j9>|!1HcXU>BBkqFnx0^LVf4e@niaqZ6ehvJHV$j?r5jvdcS5Q%N zncl%g;;uL6+ZZTW4F)l&d`HlUlByQVi4HeB+sNPW$b6Qss?xEq_=Y2uqFrigv_jS{ zSRZ5g%UCz`TleWVCKh3IFkrS43rnMnvto zA!;vTk61)g@l+EwY!%VvJsgK{3D|AH7dv?Ch@(X?4-7(nIS@Eyz1&Iu8E>xeP*ns_ zMS&+4#9QoP{G&={Nywt&27AXgXwFTR4vyYP)&?2s{k%#^Zmph#aK0&g>H>^)wyb}| z`TNgUlr8GJ;yzS9RcC+i9`p^0QbD;n2Mi*M?ib)J)>A?sQ5FVjsl6>R*QY9fMbJ)y zVm^fzLtJYek?oso>O7*aAb9o?)Q+L2bwvMk|{t>?QN2hs=5H{C-@O6MUH#uEpe;0_gHjwqACR5xZ&AX{Eou$U4A*@>es32OC)iGk~u+H7-VFG zcV|BJ6Jmd{h(1rS=^n2|wE6-Mod+p}A+V*s;^1Q<-pW&6=_0@q1i$GdKf(L_-^Ep3 zKrIE%+&z|fD^|>-euGgzaP&j6Hr?ZWZ?UJFC@RkDgmAvqODzvqam&MKtWeSliu(%3 z`BtJL2?K`JK^tt(UI33%BDIPtMe!=gq`cRAb!7_Yd7e`K+sX3_Z->6|oadi}1ktlu z^tI$#D(GtL>_%d|(v2FNrYnwfEkzI28`079pHS3GO&YB5st* zs#+no9fBu{k{`&(w-GPB8>{*w_En4Uo$x=c%=s}Q%Zpe=nF+!>LgLiT`7Lo5a)&R( zhHO-4EAY?V98-cz-n%%%@&uss0vDoUm_C<`=cO~w3cC+bq4Btgrh%N+Hhqqo+jhLw zXqBku%~4reTBkuw&VeaGi6=jkQW7_0WKg!c_z^EfbfwGwTD!iA`z|PA=^p%e>E5t< zfwHU!3(aPRaVjWc+C{v)=t)553ZpC~%(IBNt-;CGB&et< zx#Nsmz#1XJEr5PvnZ!T)31jx3z-vFGBwO zH-_K#z6ku$F2`|L+TB0j<60IW=?__iq(5X4@{KV?)euAsNAT39dfP?FEVs+6DTtkK z5#5Ag{~~0Wy4BTcK-&avL9i@BzVu15Iw06ngehMcMA${hE%gExAvgVn9}ccrB5`C9 za_!wbDgszu1W!lK-@gc1;&2HSkJzdf(Y+jaif8wo1l0skTY(2S@C?uQKP9L>fQAbE zf&BoR45iE<4{4`L!EJ8-r#VHktD2q^03}g}V=2SRVHAZ}M!Zs(C1=oFrtT)Yf4a9!RP=nj4)zRdk!caKafe!JIk2sQlSQUQ$jG9Q zkPd-yQdrI+Bt6{Dv+SdBGWo{0KC}Dhg{)y_fw4a5OB4o#J%~oXD{lnFhMaTxb@~SZ~+eM~y z5%AAaYJc}4Bs~W{;>sfAjX&`bF+}|9w&kQa0rw(g+?80Sm;Tq4mC&6<$h>2*SwzG^A%h=q@B~FGD)z@s`i+CyfGOqL8Re&LU);eRv053~;>= zxI55YbrvCeB%>GZ2jiHqI2l=l45=^6ia0~QKu~^W>OVmL$C@~ckPR9N8;g*Y-;H3B zF5wANWf45bD05kaT-(2yNi7vvI^p~j$RgygS)z=6PhkDUUKSxo(oK-vKOHmbF-gw> zbD^Xri;!a-HWnerT)|f!LDGBh&zHpJvdJQ3!82Zy+E2i)6V7FmMab>BY8!h`5PY-{ z{PuPcGAumRl)VVBQVwnxA%DV#Uur;I4_IRdw~LTx_k^4Dp99w0!R;dCteM#iJ{H)U z4sI7Aul5h-fk3|pY?XuCMaVzf=Q8*nU|$L@i;%zP$2-C~#Q#hf)m>BfaTXynPT{fs z7hrgBAVHh{Wfvi@HHg6a2N>~|G>Rf>^~LG4Uqfc;Hou;}q2{LM8+^H5p|9Z|_b=79nq>)iwzY z1>t3zkh2K+ZcsIo$Q)n`Z6eO@pX1fjjQtj1I~;qv`)BH^3da5furrRmU4;DocCw;3 z0{uI%M~=N+guK%uN&QPN*||f*RDJ|=J7p2_1YEgvdBoQwj2pr96}t%OoelrRmH;~n z;Rhb9>>}jvd*Gou1mJK>kVVMBS8*~aJx((LEw!-bCyaBEMaaaiB@OcvunxJIoSiH} z_U%{2Ft31h-_7JfQWhbHe3{EI!@Xgu7=o!D=k5%;vIsfh&3uMl3)D0h-C2ZO^=Cyx z>IT9KE)rMES%j>aAMfN7!I&c~`kiqWAxqXsg#Hh3gC#hNkmU+SE6*2*KWQ=U-~Pwg zE<(OrrmXVZ0^^~JLFNOA53IG$g~Jq_jFmqECrTnQJzW3BIKeHYeRno+OU|w>J`mCdCijFy?wXwf{|AC!+cn{JXcet0jVHN&liz&Wc!M!6FumAyK^#z|)M< z&z>zHecJiO>G%8%-q=w~YnBFSV@UQ2LfRA(^;zGNhC~-CsoHLQ(He($9|Ompg#E__?szk0Ro_Yz3pLNtCWs41TgBvp&|H)U8ZYRdV?${5nCfu!YFyyo_9 z&ZNZ8(T6CdrGcauq2Vy8(Qry>Yi_(A5$TIS5*_%@r1mcnsqt29m?F~6K+@r61q`Vs zovj>|_f~=- z#x)8Qd~baN`((t|L-1T_hYn>0-^BQ`ND14L;3DhLzYq&&qs zR#sUTc;Bvy7vFb)t+cp#TGs0N1B#+huljV+DxcR>Sx3g~|2|A*8&{yInsA1k3Vc+f zgh#v6d4%wVKY4_hljdxk`{OOYLQat?>a{P^Ow%O7Rz{!^5ByED`!nkMHP7OWlG?9z zASuTsKdI{{ASHZv951Ih$9|d=!hMc?yOFjKd*kWe@i0Y^IDoaizaMW7O#ej5U~{L- ztRT{=A28ZbA-)eJ4J+#>Eg(|D1F2_KaKH<9@QQLa>>o*$&Y2-)a{YCG+mg=Xj6f*Vu zIFNLGlAm;nNp*(#N1?NUq+$#Fq=V}yr6YgQ!;F^lLf_krIRU9{eQ(dbbT!$ujdWjU?<`Ahv<>jH);t zb!#7#C3{dVYZsg)(i`O;gXlsfk9!oQvc!9Xv(?KTrUv))mW@glZTWXzyHPO4~}k&VrX^}icbvoj=(Pok@RN#v)y6Sl23e;O`E)g;wyr^ zHjzpCBAC~N9r|M{;8laY%Oa97MWazD5;H*rQx$`!1ba7!Co5_zT^Lwdi|fxY!l=-f zdm)jA!QS`tVdp(1k_tk)CHe96U~gF76q86#V1q60Pvp?NJSx6Tu(w^le1D!y~Dw?&!! zCf!RQ+$NHfZs;;s{G?1Z?%{@jDHX4 zoWMLopp0^LP9LPoO@J$<@=*z+PJfbs@|N{xkBPG7P4MO=s@SOH^S>2SIm=fM&ejb} zpryzTM}xHtE2_$O3XZ6(@yp!^*3A$+`Kh#J8JiyL-8l}cxw<=m=Pjs<6Qd5fWtE>1 z9M%5+82b+JDvIyznVa0ZxulSGLr6jhgc1lhgqBJ|?*s&-3evj>2+~w~uY&YmrAYgM zAfWUvf^-x?nxLp4ARtZo-ZQf^dlS(A$Mei(Gw(a+oH;u?b9Q!i_8G1Wst+Fz#*FN& z5^s=>souuaaTwG2TA*JjhU{#l6|=pKd*|^o!f}A#DFQD=iX`E@chI<@xb*GSVEbr}ahhQ$2;U(s? z8)cuBWyw^rc3^fPb_Ura7R@Eb*aIE&3W-RS4+l0@qW%=P1A>^?@OUGp@}z^Eup^SVUxzs|&XoG;z4bY2YE zJD%WcFPTBTkJrWu9%ML?Y!~P}7Ku;7I@3DFJnWv754mxKT&9!$#Z`zA0{fRl@b4sL zsRQcP5$hh`$D6+HT+MKEiFJ>?P)4nLq*CIEI~DS7jdhPyj;VEzjYlEcqa=%jr^twmer}07zzYpjyf+vtF zweE5G(Kz;;DjA4+gds^w)VfEXX`$>k!6gBeBUqGH>mG0QMC4H$*^O06K9!`ab&qB9 z!)on0v2;{8!I=z0+=q7-5$hh`i}Ys+B44e0j9-z(mO$i65_?0O9kJxCe$a+-1=yF! zJ){s$=IM2hm1Y!R{0CsSH2#N$HwxPDE|%gv2j+lIA(=n2#5Xri{2ss}02SA8&V_Bg zjRRLI3zt;|hjXf~Tz=Nqcv`_Jy!AFALa$z3CD%R5$`68m*WfD{{b=3e+$gj)%JNwE z`00=sp<*QoGDwA9_t<4sNugsOLH-2kb&n61)nFHpeHqCXLXckfm{~4?JwW!~3X*Qq z>mHZgLyHT6CDBNttdwo7d)#@YIwMO`K&XMG$T=Ijq+-9;`0i*=6+^ZDy_kB2s485~3$A~E?%gu)-QIWr-hS)c6zc9097O;hfE5v9Y0L zZI+Be?ZM$0R9ZHVWk&qzU`FPAM2OI<);)?IMtec7dlY%jMKIRg=%-3z-D3zYrv&l} zi=%vN-DBNmPNDMxNv>k)JY?K_5+ZcoC!`i3YTYBraZ{bFsRJ0aiu6yjx z?W{47BtnD`weImmuO!?u`Lj2WY;TcxweImj`$DV(a{DSooC))4HP=0MD2OLf<3O0M zNkY0__jofpi&1sf0pFILqt-n}4KKmI0dzsbZOIJ1?(y#in3BB%@QEe}2U*^hZI0aS zZ#KMT@QN!gAsk~1?y9I0(*bQ~5?+4V2O)CZqca8if}4q5BskfLiXjB^cbfVml3$5>xuSjRXE(sUQkTmc^= z5TN<2zM6#jtkdQ4ND>_|_k*=jBBa>>&g(*&Wp~w(zdwEyj>V8qnJhJ}HGGY^sg4EU z`KPr!=cbx0z&-nS5cwmLEubZ8sph8I9KnVEC$j%h5YJ1vnw#?N7|ik`gY$66jYSgW zB}vUqed3kHcnYv83RiPed-uc1o^wtZzc1Y+lq=z5-}cla2A!g^$HR}gRRY5IXqrCDoX`WX;@%uZ6K?`oZh`UCq%;mY(y z^{|Z6)AW*nlzL$8IrvL$mj8n9 zt?wdPVAZHjMOprmJSej#n$>EE{T#`5l$>ylTq6mIJZRNGnEf@tbDEI%Z4e^pE;Iii zZmeEdh@C@`yTG4nPOY*qKypTLW5x))KMPJUY|Dov6?G&|6x~$iU&RexH<^_NmZ5Pf zK{YJJr~)`B&B%D>BsZk~^#*1T(G+fhANOTF&!$l5vcdRZ34&mx5?KGk&`x6l_Xv=8K#`Xm^ zMB&o1Xj-6^YhXm%Ak6bH=Wj66^-+GDRT_ELgRA>s`h#`5H9Up6KX#%cZn@0wDze1w zM$6@)i@#!bG$_LyC;f9p+E|ja0tTHLJWmg5T7_smnj@L}4+w$9NsUkBWsaIWB(Vq< z)5O-gT$ee%_0L14PSiq_U5kg-1rnI_?GYd5s4aQnNlq5R9Cdiuw6l!GBrRDR%N&h( zSQwaz5#IPZp`!^mHataGfhSt4k^4E4tq;_5VT0tf;6^~pXu&xS{EXtTSjYK9<~Zt4 zHHjEe+F*{ixp8@Kbv6(Zo`CibQHO{NW#sxA%+VI>O)EJW$<9;MpA|$BFsE9vsBZ1S z!Rvyyw@b5RU^O&OB|J&}xMBB7Wi3Hyr$}rn-*Of<9P+1g>=W^{7T!Qy<7MAzi)W)y zejJF?i7e_TG)An)c6yE>+?fAIqTs9pzFBdYcmV8}?QP#&(u*BI_ICtPefR|Wu^s?B zwt3rQKV<9<2!CfMi3h+@%P_s;7~^eUKg^c}$N95xBy62<}q z_=T5!`VW|^n2G$|3KO3FKeWl>-+0?gHpP}I=TP*jEosG1dSb*gXtmLiVHKIGh@`et)ozdR#+!}Z0KTsXvey%nb&dvh`?dO*L!AI{rXp}MkBL=`jx_E+ih3Cb zZ7=C~!i{O?f&`QA0{wt0B3Cgw(z*XMi~WKdk&*cXiT@7f17g!9CRZ`y`%Ol-TPaeS zUpy9pA<336RxvufZAL_8mn>fbSZTsVd9hEp!^dXSdYmNjy1*J*IIR?P_}PrAANWZ8 z9boTSIIZ+^1lkPmzXF&z=@WrXvv68Q=LoeKJr-rinb5VsHd%Osvnk9GZZno`uOhBW z{td9x7EUV#9Z@#J(YKD2|2wb;7G5~ehdBz^i~*0*MDxQ(vr&H}vwqE}Pmi-12ijs1 zni^#ZV5Kd5;H^03C}A@iudFWdy1*J*`1#04=18;|^Sj4M{2gHZ6)yWr$7j^(m$ez4 zR+f^asUXbPq}6n$qFaU9nd3~}W>*nnISDNE*KUkH8Qg!7b*58n#)#94%}{NAw8A z#r!u2M-+)U(n@7AM@O460%vT`ZuG|2f&HTK0D*V485REL#QXoqcP1zo5S=-G+zbUH z?e?ZU<>km@K}gUfaYiI*y4}7trMzt0H9=@ZB;grJYL=9rIhNb)`JR`Q=e7q319Fgz zZ^|>r0lU3?9tZ1#V3SV+VF8iExs#Fv>5APxHawFJAkr5g?8_nPC#-LD46zwL$M9eY zwYOaY;b%>vL6$iLX@t!QdtBAWLOYN6+Ii%rVhsWIjt166s2}_rJ)mJ^0bIxqdyyrOJQcJh zMAc1~3(Er<&4D4w4*Rv~MVRdiFur+}WnNTgY9V&2@30@gXGn8zf%-T*ovOhc*FHy6 zTk5b^o?S-vy@83C^+fVePwo&ln9?Ixpx^r1Vc&V8q%2bv{N}n0Su6Y??RHFl4`EYR z!yefj4|eFZ^Z|8*rc;!{1=0Xk;=;Uidb|SiB%g zePTrr>S&UzhfufOu&=pNRZ8jz!uy&;>bTGuaQ$T!=J>_1rya-ogb83wRV)^82KN(= zzYM$U*AkfjLH;I%$@37EUWqeDAs_q4`)jbnz%FFN&G1iu@(f8B5}x*hO+ zwhfVy6HnfpbbFI}Yp;z3Rxjy04L8lvwB>a5q=+9V08LpwqZX^47QB`sj%d0osa`Hz zEL~06Q>Or{o*r@th)b6!?=@@#i$^!2B;3fQy`FqQ` z)$1FT4(FEe2teaBj7LPmf6PfSRO;_PV;&8QQj*Z?sYNLtb;hmpc9hsnr3^YdYEepy z`Ivh-i|i{3l985LWf(RGPpBROd#P~lPv!NZl%Ze7un^>gLtiYC5HC4mQHs3F$D$N@ zmrrvTF5jOJmC_dIO;JGJs0RCl58g-9B46&syb@v5w`*)9Vcnu9db4g<>KA6QgVVj4 zJ<*rWw7cbfTnkvRbbu_N^$%}vjbDdOLcGcO_80M$98r34XG0u|IT*@ht_=-vK4}i- zwpNDqag(;0&jGjKfVr>l#tH?ws-jxID4z_~W)YCLMzwnloEx9-&b)Csow--Co>Qk!!q@)OmnwehxC%)iw^M>8aY^ z({t2Rzd_&};E(<&&RV3q=?Smc6uw1ycS{_{-!D;_wa5%=KDPji2C*_yO(fe}RH=A+ zeA=Q))bfE91lk5jCyk2v2oi!Rihg2>!k0P8B6psjr0Pfb(aV)>Cz!%Dw{q0mwCtzN#8;3JwA6E;lJl)Vf#lf2r>AB8^m!tQF2eK}MU`knmBSs@q^4Frhn<_G|Q2)p#+yEns*3vkq~t8>ao2 z>e3rRNOkU(!R#?gMSvWGWSiirx@w*R0!;yusnJ(eP?>M;sUhC-j zL8Q9moocK%h%=Dpk?^UWs&5R96zFCk+co;C>QLrv3fO&lxl~;<{l8R~`==18UWRpR zzoS$j$YDsfS)QtoU{x8R2|&tf^i|a^<~&WR`6j7)Ke+#1>z21ON%flW8mu#j6Og8o z@CBZ#yX=S)=xQJvH2SLQFy_nx>~8y&RGnG=zf`wsnu}CNju(@`}Ax(LV$ zjlQZnoH?ric2_wgRhP~9FV!&_UZlFxrfBvZO1&V#C}7**srr9=69pOpBu1mJs?LiS z%}I6h%To1*|5QD=Q9e=~J+vHa1Y&=rk4X3yPt`--!MAylI|FH+MqgDOfsM8SyYFnr zat&Nu#WVk_)?Laalj_#?Bz6L&9w0p-;a_^HK5!l1RYGncM1>(q^i|c7_|hzB?$n2} z)+_!~wbvvYss8qpBvuQ=9!Pyi_+C%duE;Ovrf^?UJANEwead82GIv~mqNusZ+&c~cqr$C1R8KcoxRYx;tVZiQrt)=R)ivLyXE5D1_j7^VXJ5cIJ zq@PIm8Bf)5c#)LQ7lc5ZL|;{%pE(zU;;Q>C=DO)wenpUF43B5|59XU|Wa)sp-rA2l z3nF#&B&7}trH6vO7#%Rz!+NxQeY@Rnye93iySC2=m!!G9DazhG9#4XC%Tuh9b<5M| zgfBkCfkK;*Y*(l@>Xv7Lffqs!Bm1<1vfuLb8kEWS&%hoioZqDK`j)5kj-u>0yi&0Q z(Cbz_lz;^J=Gy*vk|u6GD3iJUA7UN=i#Ye9=xY{nx{EwPXMp>f7g}`o`=&rGdRu%E zC7Lpsd)RMr5Z%cys0>K z+?Tyq1@~ovZ+hOB^;r^+&E-(`HzZq|HaH3DzU-}1cwL#Hh%b_Wcz43pec8|xd01~` za9$XM#Ox$>U-s2GJWH+y>hnCzqT{yEBPzkD{{^vhoIOKLwnou>MAtA_f8 z;tghNN3sT3m&-4s>JoW_HUMMCYUXXBg-rF_F!{U&^c|MYxouuk_ zeJ&bL2C$k6SM@vg0+0AT&-;FdS-%oKqJC5(*UX}H+EBmgD6i{R=PR>*4@=Q`@3)pq z(sobMHz!Tf)-psof6z;ke)J?Yd=epB(8?$x*;i5T40T@BJ=9HpZj!#wPwuR6wYJza zxMxjz%|mV){0EM$O@j-4Qjzg85Rr*wE7J~6%EXVgY4D--xZK_Z*jf>!Tf{>}+B7&Z zG@MnXasxpasYp_aBw@eLUpo0n(n1hcYm%7G(3=K_G*4nNZ(#OqKL|&ORGUnRHa89K zPz#XWH26vJG{NEbKzyvpdeh+AUpOQlSObFqk}1sGG`KIkz@3z@Hw}I;vj}TU6{(Ee znpA?^VQm^5v9%g|i|}^9x)3hR)0+k#FKClb^@al*t8u+)@Z@d*(xRonRuL}b>rI2N z?#7_J3;9PC#>olRrok_Upi{g8@K;SB$K$1PwP|qXQ7n<+sG1*=JezHaVQm`xpnV;A zB8q}hIy*~m8vMY8>#hO7mPlgLU~-p~CrlCnz~fITT)?Km1TrVJLwA=lcs_+C`3I4u zmgHaj151WxB4a)YNuvu>Ey?$Hd9#hk{gRN1bcNO-Sc_*)lI`m83#Lpd5}~#$FD-;C z1PP|lpK7axcipgGvAZ_x*wPtmJh5&+4sAxQ+b7Xt-M%y&>-I%KwQm1F9!#|IOcbb! zWb-Azsr`!Eu1v@GACOILP(YlH^tyc~{?P08&!p5~Vbo5>BllxfK(E_h{A-{P!&d{^ zKyVbPQtS4A+>yZgkSC4+qNO&{618ss?A2h_mf)WO-O;7hy8W-9<0{x^*DRvS$A3|tlq?a#l`F3##lH>X+YOCoO5AzZ)5JE zYQkkNz~KX_v|6{{v0A3^Rv0S8g&#tkCQBug75xbt!=5 zH9=I)@{(LbE!>2gHK=h1^-B@!_M2o`>-OJ{%U}Z_VlauBPb$>9{i^SlW7CkkKp}i3 z!D`)p&w*Iky9wA%jjyxtVL?r5AWk|B?6Sr;TjIwUO{SG%_W^MjBNX$T&111qqi8zY zjzW>(@ZD5et=pd%oXp6aa)b!IYTdr*hO`&tx_y!7Yzl4es{M0|b^FI~IVF%+SRCau zy%CT?~LjI)*ybKU+LdltI@ z=9ghIrbFt1i~-Tus3SdUQ&gu0p}Ow{Z4 zgWu(hs?!m8uk0MPZvWU{<=IF;Q#AYwnW5M1mpXzrw*ufsO%M*Uye-=tx!d1tcyFPe zu5Hb-Fbd!=CSp^M`sX(R|!G|3w=jkRw7 z_7Ys#D*$fP1QKlRSA1?g?nEyFyox0ED<)sa{ffyUq6X$Ve@`!wUtB&D@hFkJ6b!A} zOQJHanYk%u#5(_EjbWX?8`|}dz+5{AA=;(EwxuRvuU|x zk)BtGy6fE*ABtmf*Grkqy{r;WBW_Kv4bheF1aX^(Czz_lLu)=8{Iv$)Ld2 z-@l3oY-={M0^1`vwgTJc1MniCUjyuVgJc`t9;a>+eyqTDDJDX=8(@kehpDXr)L}iVkf5=n*VQ{XrP6N!bJ2`grj1v<1}72Vp6Z=8!2Wu+4ZAkPd8D<;xTt zeiX!SHCYF?la?n){5G&ZG_Jmekp(YsC*|wFw&H0owwx*w*$^uMkc2y|z_wD~x@-;M z>A)%xF3i(`?Xxz8WnlXju(lf4fvx|@a?+v?fDI#@+--dgqs~S|PqUD}RAHQ)UBE?PC8_wvRlq1kg&WwD?ZP{NU|LCD}%T z_W(MiORKl@;>++_t&j5X?;hd36Z=>vl;=*O;z~M)zvA-g0JbVTm+3zCRE)t{z z+xeAZ*a&1#R*-a?71-8^VSG8j4VoZiTY+uGwoXQt90lQQc9ITkM=vVIZUcL$@tb6Y z4s3V6sL5=(P+u34uuxXc@{$ZPgqxVt2V&fN*QSZUwq!FauG0X>r9)iOIsI&@f_jZqGWKIC8!i8QH*otmQdqD=aBF~u!Defg5$Y^(`s?a%{kO)FlU`ukGl&u2W9|jg-F<`8- zgqVTtsmU3v2#~{Chze}0PbtlQK<+gpTL}`c0^5mihOkG-wZmqCh%;ect!7}mbS~EU zM1W9OlZ1&nu-$XJCZpwqXOFvw}MzJKwUJPOlIi7wsgL7%r+R{I86`^vb-(Z z9J$-yY=X>0I?>VEZ-Vhd82p1h(xB3R~}^)C*Es3njQH zrUTo1#}V=I5R{EZl8h=OSO>NvF|fT!1yew%qDfyspcUAT#%IU`sTBxaH0d&qtiZN- zNz82yM%Fk@pfXlqyV;E{aWlZ}NHVY`U&z3g93t#916xlokzZVMMq{+4!1gzk)?N~o zaUI``ePt=I4SY*s2y8!uc0D98*RF#IylKX?oF*Z#oLAQjEDL2)VEMxhnnhacyQGyB zSQdDjs;NG+wuj?wl8zA33(2;j16;QQKUQEla4ja{M+2Oy2sr{vzn&%79xC@42wN3N zO3{I3;o8L|=_Ck0XwoB?Vtr<5OarVzt_0CbJu*`!JZmNRRIvp>ia&j2yhMOva~OcU#{%LL~Il%HTx zS_PIfS0HzzLO7YH1IuU6V9-HeCpCV^TO+tiTJ$%W)QdYhLbi3Y|ijIx~ z%aLu-p-`4bVCjzymFTQYB|*XL(5u^1@A6tOtACVMm-IupHaY3M|iz^HIVirCu$d|T-@P?+`NJ26%X^9_WOr90WVgZ%W@T-BP zvqTarheBE4@N_Dz0?VNAl8nshOo-5{0!z^iX)nmYQsgyZzqYs^6&PzP{=DR zj`FR*a^u((p>r=uu4L&vWMl=`5juY$qy`}>up~K7%2t8p!t-G)3k(Oem_p3Ja&ooe ztPzkHLWB?%SWe$(u(HTaL$Wm|@hY&a(6a)ohupUnBF=<)wVHutw>;Ic{{aYtHAzU< zfo1W&k&LP{3;5FP92Ho8+0V(g0@|zL4rGQ7EQfa~jIFEyUe^TSAj{jb&5^tP&4xD# zUUBW)MN7mHGEpa{L#A#dy!^2vg=JtFi})dq=pKP(;0ruHhD^5cG36mS9H zbxoi$R$!Sl3&Xr0$^;|Hz><6+150v-@`=onZJt_eJ;YW6@4yWh}9)GAmT2P?NUeh_h}}*gzFzk5OPGH zXD`ICJ5;V12xS#XO3~40kEV4bsU8T;H7R@a`NS)no$QJ0sSgMPiS#F#qN2~xu7LEk zX2O+7!QqQR{8W>5^tmfBO5%Hf9n!dpJ`2GM+)4R5`kc_lVE<4>9w7G#l^}Om(dVLp zsm!|*@ZipP{~bx1r=!nv(beU&W@%t48rRY1g4r%ZBKjnE$<`rE5{>n$iarTsPHKnlVwZ7ChUoKsWU1)$MMF%rRYyi$5^{sCG!=c$ z*cHXvAh$aqx9Q@3NU#=(dT6}6Zc44mWV#T z14l)lBw9qD(r`qdqM(XCOS>~s>ANWK5XtsAc~(WAq5oj^#RJr(jM5m8V@5} zDsFD{eO#0c2Q*E?Im3(IM$cY8!ewj0;RUI*iav9{!G*UD5hC>JF-}IGvhvsAEZ56o zcxXV;r|lgKV3g$%ecDT<2o<@zLr@&4(9!4Uv>HN3ae@*E($Qy=nxU)$va2E4$`Yib z&qa6LtOc?=DoDCbN1t1tyBYrg;Al+{vaRUz1omquOBRFhS$2|+K9|f)Vtarc(|854 zLPwvmGb5Sp8o=K*K~&E2l8gj|o0#(o+~}U@%qyZ#M-MCd#Q9=za9aYBts1FN(Pzyv zX{<7GvlPPLAXr78Mbpt<+X3sL@g^2NENDvrCSryIo22oUmiRHoy~EYm5rVnDy^c=^r$+F%(+R3(5s?P(G6)Y$mmn#ISp{#_L{V<+jhSXa5*KA zS6Ce7ThZrFry_*T3UCsC*V1{&sM$PO=xk0%UqV#$NphT&t)kE9adB)A7$YqqX7u^q zN5$DtAPcn+6@3=?UmV+n+-*p<(Ij3)pF^fO*f+?%q!4i?%&XOmK9Bv0ZD#L-@IsS> zi8}hcdCSSDIxbW^21%%p)lkvr%+tQCET9S+o=Rru=yR)YGP5-X*hUkCgDh{$Hb?IE zHyfTW%yE^wj~gkecLHrO$bdTtB&3;UY@FOU7hEy&?2`-B1 z=yS+HJR;}60)36+)xmKB*jby-HJZn`>nJjl?9=KCVh@0EBegc z3!`{rWVO)*5^P1E!*^nibv(e2k!18qzL3!;IYg*7qfbvSkzZV^&){~SytEIcwUu3TW;t0r zRisF7tb{=l?yzFUd{rY@DZ;A*t4p{rPsfZ!yxcNo>;&u`jq8|k$(xDNqS3%65>D>6 z9)&wc;7(;J^4BShlM}3%v3)NL>H7hm&;)Y4N6a|tF&<^y1bR<1l8Ir>SnkBCYjL)G z;eaqCdCmn($Bc*0<02~uup*L(8OdFeDNGVg>=wS|BccuhnUmU~dsP3TB4%8TEEO}Z zDu6{JgOD+lgalFyt767!FY>VI$X!H87{%iS2-f16lVrO}hG6+M-5*ig)iEPMf+_T; zu3A}5#Ee0GVaFfyv33kImVRh6DrO|nB4(6^BW4r@Rm@l;50;|xf+!G&WV=R&sF-me z=KpyHvZ)OUh|`gd8SlUp9W$2bQj+~nb$J(*zN&zZ8M_UsCB*PafIcDk0jW|k7vEs)QvId<@VjU0TJA?;8lgZXx@=D#`yQ=_+PCIRI8`_j;T?@~r!l zdbrjRGX}hmp-?zS#f<-4D8w@0*eXc25+qR~^@AomGFVgOwo?cv^K{JkATWdR{=kN7 zJj22p1zr0l7dmiYOEq4_65rfdiAUvI0UgqC&Y+;TF|BEkaM@LGczr6ZW5&h#gtwj% zBJ}E!Nydz#lFl%|t{c-T$(S*zADRGVdBlu^aAQnor3yU58g|c#%m*n^@+{BzA5aS-&)LX=iY5lF3 zamuGvSWAd#M`GS16)I*Nm8&G{humQb;U5vKV#Yc_m?oYHY>CE4SopA@fGQZww*uR% z@o|>;F~*t^`Pn%@H!V0n!iD4bF&n{Rq(UYmbMm4pTjJ2chIf$} zI%d3hzbLbf12|n1go7+^%Qi>u_BR_|1=PmXr~##8g~ZUldUQU*}lNtEECn2s4ozC;|yqk$IFjIT+sju~IhD9?VNf>l7Mr%C%D z)QTB@np{nix`5C}lYYUG6*HbsMMyXfS<^Lv1Y0rVgzK1K-39Ofl8hP27cyofhlm=O zF{7uK$SL|GC;%;<#(K@SOdYcm@CAkA1-)FjMU z=DTA)3U65#kHTF>--?pd(UY`2%_QwDPo&F5Fr1L2@t&mM7bYp79+7%{Q$Uh7c#`Ht z`AVC78xm>uFX#qH(pgW^=Y>qtNuw@22vGIQiEsaYluaA;B>icKhrG?f|4$pgh|^Bo9RqCK;&|~f|N8# zlb~*7hDln21LnR^0JXylmq);P-3ynGVYzxulxiUs%u*(E9e80bh`r+}|6{zXA!R7eK{h~t>T!Sk2po%|l*wF< zp(UjKe`(X;>3WY%``43v*N!XZiH-Wm!xMGm{iM5|Mbf8FQn4+#_0=9|Grf zXP-XOtImEde0WsUnKGH{R1Go$r{gy*AE%=>rs;$cbiiDNiYG}@;9yl}kP^C>meK)p zP3?zwDWd)CeialcI>7FtPvPk8g{91Y(y>vntNR_yEfD=(Pda@JG6y|jFuqRdT6@VZ z=?gsR^qs~W^vu&Fz3p8rJSFMhc+%;MjXCJShl&1iM~9?8_oUMo7jw`i^_=v6_EFj5b@G9VUtPI8T_K~*340F$UShG9&lERTTr zA8Zsw+&AGMK9uz7fH-q-0Ri<59n}d$5(e;TWQ7ihd$w>gTM>Yz zG(l9(@{$Zrgq!e018Q6akKus=HK^VwBS%#duY={n(0Z}oN1CC zT`de!`CoCJ5a~-#(x5D{)p{>hMv(TTQ2WJ2e^rxkJzx99B+&tL?fJ*-o*cum^@#W| z8*J)!437{+*RXsD8P)qU_>}a;gQYG1Xlx02tv_i&7w`WDVj*FRNrvA@kapWhmW{WV zV)*yO-`6?kEv6ZXTRC4+jkTC+1kQ#(Z}69i*pFkHQMML1oMjYzSeCU!w`t3mKQu%f zaqOJ17(p6oU+t;=>S$B@K+^7h?-sgH3?@H;^SZ%g4BDXj`0s2Si{6nknY(N<-cCl* z{3Bjpv<1FhsfzZ;u_#KJ%-ygBx$mZ$Ak}IJ%4(k$ zuVa2)D*~HmGgrj_%1hF6Pf|oBlNA3Ik-o2l)f&_do>3&~lj%EQIr=rPO^EE>Zzdz! zg3SRVUtiq-{l&0t@%hyk!zdG9)GC3em9(osJj%+}f*l8zJ!i~^xvR}XQyM$euM5eg z*R98n2%T~?k`-+loyNvaGg=>sVkyAhM&iS|;cD_-RJS4{3U)ysIX{byt>YU7(8sl) zm-!m}Jp8GBQ>BD?#uqlG_I73qE+?w?*NQ~drq6rjsCLww4m=hg@NtaTKcwp@%s54F z4`ccF+3P0H#-o6&H{MR${v-MR z{XLknBkWU8><7~6F4P$-3~)-X0eW5IFLc{#hAVL_S}T_%V)!+bI*bbob0qe5KMvemr!>M~*#%ppvaR)LYkZgJ1 z@#rTyJiuu#iR?55W$z~j9YR-DA6N^8^ZZm^_Y)=JLs?XavmmTCzSEkeNxIw9TtYRxI;lHcqn;#1PVlH3J$4h4=)p2`?L`MJBvtez(L9SF(X1YG-I|CI zhSYWRBt45YyVq`Skh+AbW+*V;le7=tKNFHF2~x+i6{VyNo}{q_Oj2P%YVi!C62*LH zH3>1_kr`$$KnKh<><(7?kWEiD30G9Ve5RzTUIjs#^p)8&MNjl_L62f4X^T*oo=EWk zBvtVw@b+elpjYPU&~P`kg% z#7aNW0dozwOkpS_UC<=d?mJbx-sPxvd;zA;i1fmfG&8|8+^+$hmW4^C;RPmnxZpkI zg2Te51s!Rs3^rBQB-pe#&a{aRnCq7pX8hb;lVH=SJm&rPA))TV4`#pnu_x)*{3fZL zn7fL(4HroE?tdP$Nui&QHki+s z-1xhH6?PoLo+0s-=q9)yxGaP|)Z#ON`&UZGj<^Jd&@S3Oi@5(@M3J)iQEuEyvWw!G zC}b7q^d}3s_+C}!Gl%=vj|ySM1{9BDKIgb`6`y8*N3%(_)EG)%rto>njee;m8QCy~ zNb9KM3Vr56S)0vBUyxrIx>Q5OjvvIr=dN{xXdl zj6*&p6Mu`Yj4H=m6QaE3Im|UhooJjxJ^3bzyJDAkVFM={_sak&^lpU&mdnnK#R+A^ zv)nSsISIgMxdeEU8v9}`e1n5Gj(&;dRrI+SJ>5wriONwf`jIee2QAO5B0bX*mm$J}qJWON-O;H8Ob> zD3E5{!zYHA>wXB0p1Am*YPq-?Z~f0C{Sr#I*u!UH$R|?tROK3w=)=OILL&t!bt=t; z!ZlSi2@U7KDbqD{z+4OJ!+$L+a=#%U&ET|4e9J04f5TOgjIJaht|jhyCwspvQJS@^ zzzZDY^f&(1>UbeHtF`HBP38{&HlLVi8avIJXu7f~fUSYhjYzg$G}){snkwe ztwMM|g4NTsJ0WPxKLPtg;~!f1Fnnc`EX187-LXSEfx!?jD}yGXgb&4 zhmA#{EO7W_Dy=4(f~O=gGN&^kLa&-=T6F~u(^Dw&X`0A$j)S4Dq7HMSY2zoDXre5S ziKfO|%L;uvN$Pacrze_@cJvYozawZaL3*O8+uvo`9c2HBWLr#-o@hFD7H=CsQ3#TN zq!0B()2-|1uZsaJs|iB3HPJL8Acm18^+0Hzount4F7^#!?*bdB@lVMLJ<;@Snhh`i z1DvY~qH>m(hpXy?< z33X`ifhWyGhc;(|X$l=Mx6d~qAsSu>j%PG1KXkA!zFbm>IUekfuZUwY9#AH8- zvAF&FAuNyS=+|iuc0f6SkH*Jzb*?<;%ld-v`+R2tJ0FNvr+-+huTKb0%(9Yfw&MJ~ z104OsT7}r^lAJT&7G+;44tmSyt-#2l@ab8PiR~08Oo&jr3E0aH_nvl6dN)>WKN@!;qm)++V)io}UM z%vdkQ>4|$zu76moKX&ILPQCAnu<44^HxD@ahqYR_*C0;xqG+~FaguSn#q|$sbp;RT z!I{w_FT12Tw5u!EKdjaI_v;bo+%!DQnr+sr+j00%|FBl+eWHo;ILyroDGtpAa{a?v zJ@j)EXVHpqR#R~r9t21KuvW_kiF4W-o0+|%IMp##$Mp|uwdz52l5_g~n(Sl6DTYU1 zT>r3E*G?89&Klnc_POGGJsTYT!&;5Qmwv%%*CC#Lr#PMGfuny|tEa^ih;!{)0DGZ0 zG%w5b4{Me0jZorL-4nrL&~`;zy@=<&T>r3E7k>#RPW;Y_tg7Nn!G6VD|FFwm1Fj3B z$6`leA*3aOFRp(OBIMnMXOiiCxZhpK2tB#D8cXlbZJ+f|WfX8MMW&7tc())(7e9ot z(50Q~3buc80cN-&*-6EweTjK2a^w$TETc?0ZxI-=48M;m$yJEHQ?I16+PT1XeU@BJ z-d*IIYj#i+pk4jimNP&rBOcea>xzkUx*4hKN&kHJEA$jW|N3t{z9#wyp7i-oU!i9R z`hj|w@N9*+X$|O~8!IRj4*_GJUyruhANvCQ&E;7T(f38At`K`YMF`szD+js2(f7ZK`GmypS@(v2a31_x| z7DI}$_P_)85SStosVR3 z$BV}hRovT{+r>vP`5z!Xq9UtFTp`~88;g(d4&MzkxY$`WKEc~4nw}v@Jl}kb5lAMR z?keM5O7@1XF)Y5cw^5`41`LpR9sD%VY{#XBz>~a;H#5q((|)G zSejju7>Gp^agw`SeWu_Nrr7T|csdlyTN7=_{gpo!4rkQ#`+hN^oY3U~ETgXTMhhke z;J}6G;kw?a!){y;@)(xsFD`y)fM!Ma)+$J)4vz9+G-_2>rScp%ONp_J{^H_?{$xcT zFD*r-I@Qm^XgvH-l^TB5EG0%s`iqMn`jZuP`jn$mixbmX9Sqx>RH@Y)&CZx|IsGMn z=ucMkz*-+F)nO2xgJWR$MU|R(%q%5F3i^wSANrFOy%tuSO2v(<$HLUISr6`xvU-F0kWJTxJwo|E?g+7cTi1Dfv zO?8S=BCepnxcH$zS<#Lo6RA|)Wf6?VmOZM}V*DXWi4lkX;^K$?WJPOZ&IYBf90_JL zvOQL%ntf!J65}2H#l;W($%?KXn?_MpGG_@O^p(MkD&sMP9A2csczwkqX=FZ_sliGh;-;^K$?WJSA8 zPp4AjcSnhOolvD_A2myfdeL88{D^vGoSbQwqaSm9xE5m<_1gcSsJu?lYkyeRe6sqN zc(U4dXK_i2UE)FNcF63t2Op*3!YGSzmWGRlibR9vgwO0bhl|Q$SnG3*WX{axHx%@Q z-QHrHsQxYsz;nfe-KbMga3`iv?ia!R<%<=hQoflhObJ6Tjg`>BpF9VY-NX$!~ysR~3-YY=-h01w&cld2TH8$4RmU z+62=d&I~~k7P6sy(QvGLy~vGED&qds7U(5Nr9nv1q?_EaJ%S}(=SIVb6h_OVK=c;G zw?OKohV**9DrKsiten*COJ+5 zUq4fQXIWKI{FiLSIsN%BGu2yK6=PvAM=B%LQlp-!9x<~DBO6K)DV?0AXR4PSEG`VK ztDz9DiP6kw68EQbB~Bt4%v{AsXW+Zs^SK{gn{jl8nOCy~)zVqwLLRv6r+jP@@@FI2 zDp8#`(S;=OMLh86^U`b`ux%PY)`HAnLe>oV6xL8P^RQ-gw?r0+{31xUEV4$NLT%0SO?g>5u<9CbE}WvQ zIa;&0=p@<#=N+?I!-X}Z&(oT9opBF9ed~IZ5WO_@t;@3>Sq+4 zLh(PaJV8jGzlHkiR`|q@AU*LUEsAwW(xWr*ue)VaJORX{M)=CtPij=&jNt=#%Ik4g z+E`j+jDEWl2p+xsyV?3}f7yAvy0%BhO4XbJwytK)KVPY4U8<)0;$N5|g7Pi@qWn-V zH-0uU>^$p!1?$;ZCzj{M=u@(lQ zRx!B$^Q0HV{om{8?tC^nV@>_80$ru<7MuGWS7dso(Wn>u+-csIkrB)_<1QxtS~cK) z|KQIA(j;5KPiXhdeSQczx~X5;)!Aegz@NfiQzoet4UBGVnd{r5^i&0{usL`-w0MTk z+zQg{skEpg|0DEwL>lZ#Dn7@&6MaUGcTEVx9W#-Zd6MF=`;L$lEF^tiI9-yCc#@hZ zNvs!1TJZ@Uy^y3k*+@###E;3QjRo;ofJnh>JPaSMBsqj6`%Byj(MhkQNOVblfA*D= z-b9@AS!pSdiMIGQ=(;VA>L}V`{>ume&=%(bQ*AN%xwT}hc)`4^JFuxp+>g|Ywn!AK zE!GF<<67Cgh_w9#{?tzQpJ6Uh`apaY=l(w~N&3^1bZUi3YVrbdUE9LZs>uuCpLs}H zf7@KxkR=v2EN@ItnepT&LzD32N0aD)xf+heGbD1=JBmcEqSX&MCK71VZ7i0kh#3r` z&-0|yB8MDw+R6~yv#rGCMD(vc>9pJ-2c0&kVXiIMk4ey`|`(NFwF>!S0^ zMF#~-%J$rdV#+TcQA19Q99-`)F(NQ)Vr1S`w6yIIvKz^kNUcRrjJ%%$*PcWERfTan zvL;4ImNhYQeQiECIj{{1&oMDF&f8lGISRtJieycU zJmcO{$ZcQ`6`o^a*{seI-Q;3gdT(+}7wNwf|J zW-rhpVRl1mj;;ZpP)~)vbq6rrw=TFR`qs)HBdkZ?`WGIb{NLN<) z2$o&MFPl8N2!UB$#M0`e7{3G|*O6>*Q>AqmalJOMC&>5upNVlgvbqS8WpxqXb}cBF zJRYP{RHQ43(_O?Aa;u&@{eHZkAo1p4wzk-+i;%Ka7tv)vpdj%vU{2C(-9^0lv=HkH zZG0uLb%cwm=q}Fmfbe5>N#gPrO~gs=a<{oCN20*ZRx_ij2p%yo=$^ps z5nXa_aP%dY@0z&e*6fv`bvG(+hSovJ)+JZDxRa#;YlFlGQF9TO98s)GE*hYZtJh&O zoLPWB^(A-gvbh+nUJ{D9tCho$j!W)qaMUF?Z;)5c5t>S2&8R`QJ+S;|u(jp3rNZ(P z%P~HpvncerW%)h4_oq6Gsl#L0LzIgE&WCEtiDFs)9SA3E0~52AhQC7mCRc@<)>O_tYQh`Jnb#_n!K<;XJYPg=0}s0b|)g%6f% z5>|2jFNgxx(;LnEjiVZa0R7_LM5?r_iMJsI+|(op5VtuYUpK(#+#5T6F9324TAm0Z z2AqJc$_p59O4I$2CNebJ#*H9>A$ zh49S;s})Jzun3#?12#WZY0>&4kY z6uJTqKSrh1ilm86YcVqCDIr3yT9LHmI&ORQS{sLNt|!q<#SzTAPZ zkG#U-DBoLu@5}dGIiYh1NzP;GJY@79$b`;0gybVc?FB({oRqEhg6P?=3X28ffF;D- z3*z%?m02+$KWZUrFNpE&O0x&ZeS%~wN#fOB5cmJe&HP|*6p}#1nJ_P1b1#VBLeVvq z0-?Mn2@~~R5G9(VGOA8v;BB&V)LsxVJCa#nKtnW~MrP=}AU?j0y&z@)T&xMgLAuS^ zWt$^+`^X8BU<-uTAy_rM zr-uqM9!Zik-oV0#1(h9vQ7s)tY!xMSVJ*bSkYO_%0hlZEA6Ah2N!s003M>o7L_j(DP zPe^iCOXnfu=`<9?5Z1wrY6YZbh>9!N?%06k;~KHK+2i4}dh)LR7oeq%Ay}CK#0(*nqc<6& l zs{&2#jh*lfNi@M@yU_&6D=d!k{d5zo{=4jNKOo5)ES-mpF6YV!oihm8LWpXDB*#hF zstNX)l8=1}#vV(E*#v7}D#G>xxu}JxCg{Jy#_l5bA(HJdiC0Z<>FIdp1A7Z12}HI& zx&&X-4pTN@`Z*cY%9<{W)$Q=i@?=I8dkc7{>>Sk&ldhL$g8+@v@M$tgx5JKEvCK9P z;HR1(T&3Hgmq#!7N@J{n8oK*E$J{X*c@f%cJHO7Ze5m46w)=LqA|AHhsH==VR4ins2gLoKc!Be zy^#FK(s{^uc(#JjS&)$Dgs8?ya-5W{8sp6ccou~TpDas=*%&9(%+CxUowX3v7@s=B z*ayfRhGg@n*&@{#o8<{(Gm!haLd2Q)&lXub3lA@jf`3+*5fB%%0~k6$b{3JkBxXgV(*9A1e^t7cj8lEKKFYJ>>As#(%J zAZr}!Q=0YmOJVJSzpFIJhvnL9a*GYYcXo!6?j^(0Sqz$N`av{V@)(Pw{BYf5zlcf@ z_IycNYgqOiGF~Qz346{Fl0}GWvLwez*{aDd`nnQpM2rJkh>9KGtjd}Li6BG>QB8JR zIVXB_yaa({YfXh!ldZ8Pkkv%)I|>nJ>HluG%j_Kc1EA1ET~KJ$?Y1fwYf}|h0N&%+tr~VnU3@JEjPETWX2bsJVoCNH zkUzB$)v&!=*I~wCygi6y`yYu{4SV;hGOPe{Yb!*YrT@>?9l8WJudPwCn=UPE*RA`* zJqDwCjsQM6J4dzdDbaSe6wo>ie?|7`)?Ejcvh4?WLKB4NMC)F>Dui7@{;vwld5UD~ zrgP-~e>L$o@S1DnB`jW`38M-)hjPLX6Gr>UqVnJ5j=<@~+eYsk#rz1-J>E9@?m1@1 zcz2ZQhh#f}5?mD1ZyP07z@!zQ1a!7$93jE_ZKJ^92K$~0ZUSMaCY8dG^|sMBfkhDTdY}GKmNoOikl+Fkqd@7kfW42sS#&8|(2m6N zn8OBG8EK1c;F(HNhWWD6|Ccw(f3Cl;;C_a@ISGQaH$|mgPmkjIr`vSLH)*f&wmP&N025V5 z3NQ8cfgg@qC5+D1Ws(+80#M$pOw09`)cM3wBT=WWT8nEB#UBu}9P7SU{ z^M0o|4OK&x0sEb#NDy(@6pF2_LjZf+r0^!Qsc2*T_j*$iMT&$$`(bb-UH62fF=i{% z#`to2D-%VEglVEjK#3#aTo0G_9ES_w`j*px!oU^~oS*v%!l9G~q^W1)?R%X3)Ia4F}X` zF+gdB*5aHU+8*%3UTcbr*Dt$MT=j~fk8DFO-ynEe4?+(%Q(XJnU_uG zxXu(Wr}BLffWZVJ>pCDk#dUCBLzUA^02XG+Nl$SlUy{o5p8@byU`{-B#a+Zl=JHP| zL-Pormn*;-dO1aF;1jN2!axOPSqtpL?MjOkyDX?U=Tc5H<9hHS)>fZ-Gl%GZfmWey z4Z)ah^@HIr175n@9;}}~;_KTH>q;>x(Hn8Rc$+T5{%C&0c zR|s+m1ZskfAfmW)!qn@K`Sbc z;@kH^C4cHQ%$yOZz71-Py`!8{J=%5Umm6{85vZ>XnvssNN!n-MoEF0SCvL(N2!UqU zpdQ#j!Jwv{NT*KUIXP{xK@TJa7~YOI+vM z@6?XCZcsy*KO&00MI@5AuA`b@IkHk>}Al_mE943F`| zsodw!wLkdY=hpw=_K4u>ti-R_?zErsz9A=&;%Si zz@TV+&rbMzt)Om-oJQH8r&c=it<%fnu1&Z-9Q_XStseoS=Udz05g|)MmtH~htq;D( z5-`f~pq(ShxY#k2W20JKJfcy}o3(|1`XV$r_37|~)-6CA%d_N8o6PZaFsqEpy}X7~ zHzGLaPKO)3@XWh-QpQu)I*9aBSG&;q;ttN79f~*^@;}g0 zz*AQwV)RqjgoA!@3eY7Qs>*G5hxBc(cA4X;>y?)H1bIf6@GVC$Ec4aON+P3w70O+d zYEd;+WDN3_8c{`@Lc|pWPYe|qhkuIlGrAR}{Cp%U_0J16MaB^C$(i{@)FDg)At-z~ z#vjjHS!DDqTM-vn1XEIEWO~c>N@ZA0K90N#FJsj;h3It3ZgX+O~zHlPkqB8zN(w_ixNDP+bN=GR&2} zO8UR=9h9#u=R6CNd9Ji4X|>wxpI?impFr}NkrtZC!&)0sONYrTcA5P;I(L3rRwYvm{;GB$V@j;O)jIrq=#1PG3`X`5(9_ml?Is`M<9 zvB^7aEWXMGL`{Sygyil;Ig8@0##a#;JyNKzDfm|JbNT%Y8vy7qf>Sd}hL4la4iXt( zd3RtGkaC=bu!taTjtcUF_v2g5IF}6&ZzqVGOMHCuIwIqUH>iA0&gCe?$8|13NsoC? zywQL&_y>~vIs*~^d}g}HIN=Q{jL(1}HW%)v{0L6Vv-FI2#@wcyHks}=lB%9a4<{MV z7Uz58?^QmLamnkeT$T}?fO^j4Q@m7EE|KwEWxh8UdDZ)RpPGyq2hY>bSK(kitCa&~y!_p*e0m>4c>#(()YDAduGEFVQns z*n4=Pz{wERAyO}Kfk$ocH!pEJ;*aT>jd)x83XP^ED^J|kE+4n8J#{~w!H0cOBY(jl#HOxT zE-2OJr%B!)`_>ckD)PRG;0eajRAygPk2)B9@F7Vq1Qy9ClIwj@m#(FYJcy7o5rEV{ zklq)y_jAyCbI3YqxoWV?-_PQkes6i^w1)f@$iiq_RQrWE?1hw&f3-(R!bgVlN4|sg zPqZZ`@mJ^{9f_OC-5}I&WkvB3(r=>tr_=4jNzIm=nwdGoKFEGC^0xtUHwfK~&n^Bz zVK*U{OG(sbA@&}9kKeIae5sm;Jk7Viy;Mfp40{CTI zpOzsZfkyzpkK)`TfSUq?t-JaN;8=46aEO=^H1%P;s1lMC?&0l?)qaZ*A9!dNV{AxL zq$eWg7)B4rq6VS~Z*7eBg!#r|-Ry2;c>uw)n+(I2QyUijp3w4Fs&U2T6#%YlkXR#c z9j-2t@_33L>Vg%e_~wi(gK`p!V<_V(6vJ65W4N0c|CgRke7p+aDnQgB)EO#JPC~nu*L;@ z4;l0k0P78Cy4rhK!{a^lS9P=Zkbpk}b4ugX*2Da~KeMGLd_y~r_%Ck(@h{{8=H(iQ zRg$+Rw)R9UK>848Z(;#YME8=BLT%ZiKFhn=$h`KJL@Lq$nx_dKYj6Nvwco)k#ENtP z9a+b0>noRp6OvHkYY4Isb^fXNE<~-k`*WqAKwFApS)|ovMOjgkN`TJ62&|(RLYsUiw$H)z7&L^s%)|~kNa>Iux5EBY)oJ7rPv0$&(zu#q zBXcU6U-}Hat;C*3@2*ze6TS(cTy z63fF34CMiZq8J~L{Ev6$3xDyv^nQis@Sl;TSu{HES3auxx1(51h;fAI6Ye-d!!O-^ zN4wu?F3upG4{}+b?7<6l3F0mq-3>rU!moYsI+UIqUJKB$@ukE0#X0QI@jO^-K<=~2 z3A_oV_X!^bd16ug!&674A|kz?cTkahd;r{wh#ZXIX^Mg*r{a3?FCXRW-(I8$DuXQp z>v3gm8Q6nE!quK}+OHnLo7P3@`)%>M3wPK5B-}-*XKi=4djx3gcVu#g3jC02rndp- zrhVym5FLbMERq{ZuHD^^DdFNv(ohznd;>w+-PLN^3}Zr&HPmv|BQw*F`^eo@u2&tO z{)CL(9ep7~mDHGqN`;TA=fAQWLsaDM3Z0^Iwm{&zyD7)3i$zGklJajy>Tpuy?*4ii z(*TftYvj*SCCq-7yK^-w2MYG-kGq>JV5rUY#+TvFsU zUg@XP+AD>yS6UDK6JJ4&u2=eO55}fDBmQ{=Pk+*%D`>pZ(cpTlpBxRq1P!vh(!FOa zuhbzLPcJJqO6eTvl@8En#rFa63!x@afh(xiu2)J<$Kw#0THeAxfjTfuW#)RN&_WoY zhF4-_VW3KCq~(?V+<}>OcqIna1)#A3xn8Nynli>K5%5J|25Vdvuhi~t1wjWs%z%8Z z##mmdJkncUX+lx$m8femHIaF3uS6yK6L(;Yk-XARXDqKo)-hhG&vQbYMTt>B$em=j z;`rryCF-g^5dH4dT^g@M3Tdz8 z8FWki33cK;BU%w;9%52v*F?y z@F&4Y5{1Ab`5MVJ8y;IwLySU%l<@#m3Iu63%=Ie9u9`sBR?C$oGOKnK#D+H-pd|-F z#%#Fn`Fu`E`8uH_;rkj|`D@>9rr0q459t4K5j2U*hD-WLu?Fcsqx^lO4ktBiI3poc z`~ca{MxKi*Vc778#A>223i}&!nV&?=h8=Tb{usqYAaJ%y?JR6ax&*M{!FqK>5yOTj z&TBTz$!z#FGD{qVf?PKI@Lmof`y&2TDyTf^&lNOmcG+3?-Fct-yXi1UPMMg{hvTDxr6F)0>H+8~ud&_5D^VJb714eK`q3~b2A z1fVKtq{W6gi{s51Y{;ON0JJwCmkmR^xNJzkVZgkhaaq`~+>Byk@wEbCG34)QjKzjE zkltd$;f4*VYcU&=d2KeN68-P?$GZfv;fV_t8TtPM3xaNg>UKm)8~)Nf1^u z_{F4-X2bLC3yKVgp3!2}moyu;!#Aei1O8=*M`%bEHl&KUY`FZ*W@0r`ya)7Zox-po zvzTE+Qrcldu5haNHqOcc8&0|eHd)onX2U*7c$m101P>|GJVQ_i&4zKqOW^wlNKzPq zMY0&lH5=Ccv7v}Xgp{cO)C~k_Hk>yWE!6?C9$Ky}k-0H4Ke1s(6HLBmLdI;^s6W_{ zl#m@18}5fz{$XF2Rcx4g8T$7riY9T{u*moEViVHur2IdjhU27$4fA%YEPjFPtdV~~ z4Q1Hy!vf7jR}}UTa`_dBm<|8F9WQdCxB>{Atx`J+8Cg&d0bu~g7e(w{46*znBo>a3q!48VIDWU=8+e2@+Vr$#L9hBm{4i$*D( z1KF^2Ttgu*0db2^H>toVYE_pF2i3t$7MWT`LH~jX3{#o8Y?xLcf!UCe)qtv{kro?9 zPrz&{*pNZp0O(~vE*p*?9BrNr2{-|m*&3IH4eubsPyQ+(K7f3!##n6F4CyU49B9~( zx)!q`nb&4RD$!rE4hAiW4gdVpVnecyVZ%2f^9liJ0)kAWK3W66TsEYx>LcOf*|6o9 zcuxLOKypc!Phi9F#!rbQ22Dk)5)B*9?Nm{a4x34oLXBwHFttYkR`jqD88)Q)DmEmI zMZ$Zrb(js0U)AiE;+?jqq=>|I41%XB)s=5_x-@J^3TZYR*13jAhp>^s*CTZ_8-ATX zRCI>uc`a6bNwZ8e@UKHWSwph0Ayvd>!wXY$h!#k(7U+#Sg<(TxF~f$Ww8Ms6 z;nbvcxLv@8^R9tS?mcg_;j>r3hF-)+Ab7r{I%qZ=SsWX|iy=Bei{y7C*KGJ}yifdy z2q_x?&@vFD+3@oYcva~Q*~?n4ERnh9L=3UvpHuM4KLs*o!_Zg2hNOg?uh{T7s^Sk` zjIEmVrrh8!=-;>)n#5(p&)%;lb|d`(%DW{HcSh|7lIAK(dV zJmP0jK_f|juApJVuhVm|e)0nV)@qQ&hPlz6h7EU)z+CeIjZ!)Xvf;V+Fb8r6h#)BE zSwIDDq*}Xd_``jS!;`6HVL-}gn99s$!_yVebMS1)$VNbAXr#r41Nz|67i`F&{s0Uz zAeRk48ByD?Apz$C^Nz-4VZ*o9*AOfJE+95TzFlK1HtdM>78~|3Y)D;;*^ta@vmuq} z|6e+qf!Ofgb&CzjI))9!)OZmIX*C3SnC#XJzg#w?uIeM<vaBEyDMU&V%`u}G*Mn~mA<$_>qK zDc;L{D~f`+E{otfMRnyHoh}UTIQnnKvl;IEK6nhh6^swMhD^r{xCzNFc( zZ@psTI`GpVp06QU*pMpXvSG8nMa3PY*bekAox-povzTE+QrclduCV{qix>%_@!P1I zXe|9L$h?P~MNt*^Vf;4rPIKG@pv)hDiEY-`rp(?WG1gxd@ih@V{~L;mYRb&LtB#Ou z5Zz6S}H6fQdNW_#GbS=ARfv8S8nNm9oWs)uyW%_F#3`J{TF6|(a zskyY3JeT$*t{xnbYOs|CZw2nk*yN0P>lJsmxq1o>m>J?ZL&2JPgz^jkLHpEx#JIWYE6= z2$+OR6)qPK{UVp)VgeQeCPCw}aB-{lX(D1&0Z||FrW#{$@p)<;cP_2H;bQ8u%*AA0 zn~SML|FMgBBp@yxf5+ltvX0^6K{4gT+bD52f}BEj`vt#TE~b0HN5aQ*am|-YbMi10 z;3k)J`2;Q=(7!S`L8GZ?RifeIIUD_gbZAYWS=5Mzi#yjU$%+mzBE!X0U&Y0wu}FyN z8^T;X-H70sM|EWep-aQXq>$#~+zouJW{#{Z>bMe$);>B$e zp$N`S^(D>4arG;URlw(mxTJ<;;bN+Y%f&su%P&4gil#uf)F})XGd&tECZ!!N<_h~~ z*2D7(ot3x(Debco7vb7ID>3awKkRd_X$inbh982|9DX<|{QJj&{Y%;r7zZl`{cGio zgB9;S4xaFpAETr=1Tht8lXr*kQn)mV;9x~kl;^ET2jS2A5R6X;E4BkJkxEih*4H2Z zEw(9l#@YVkATZU#+5W@rT(LA(RnPk$h6bR*7Dz1>R@q>vKpL_NRHjHx9)=k$xT;t0 zf$Ep`vt3obe=$3F4yelro(jXz545Wq)dAHNND_v?BAHHd?Wz{NT2a(Rgp?%#ND2gL zS2g$F>O$6o>}f4my(u%IeMxdvlVh9V?H)3ARokNSJ0T;z2g!11&ou88$S2c8CjQ9q!{YJi>I=gXIAFj(IUg}mzT!vikA`!c) zwLJ@p`-t+QNX}NNoyAp=F1D*GaSz9AldBqqWK;Od&+{bvsy>K#)K~36d9JUj3rvW( zFZIHYdD8M$yX2aZ`NT?T+|}#{U^hvtJWvF!{mB0x`qyjF+4WZ~ro+kaK(_l3JlDuN zY(C?!&U{;rYcDSVa8-jWf3^8V%U|s}f#pMyD2B6A776rM`|@M2Zwe4K2o*-XBZ2yi z>#ue`kL4(29@!C)UK*w{bN$s9F%8*YG4c(dCTpbSui|Ur{S)sF3|bAqCkEvDtB*_7 zGyaNz2Y~rqk_vk~1vcR&v<6xw z1A)J}M;8&z6R{LpnSyLD`8u+0&fJ`6e_5=W?+e_k2ooiJsn5l#dzfwt_Jx$(_Jvd% z|Cx!7FP#0*@`Yr1;|oXO?CAHA{XqoTnA+qVe!0GoZZaPUAMXoa!+}YZJP!(RlS{gM zf-fAixv{8c(3Dn{XnbMsSqXx4=tiKYse>3_7#32M6@AT!j4!16Dqlz%i-fPRgNxRc zj1`fNZ%gsMTdk>BkEA;hJZ-72Oi^@cd?6{MePPO#aB&vGYX<);siS@2hbME3AY@N} zoSo{Y+82(GsV;g0Ukc*#8j{5qQbk-}*f1nP^h1jFKzGq8j4x#VHNKFPc6=dM*uT3d zUODK!WEE1H_mUrHJEyh|prc(A?(Bg_!8-Aufzj*4->*k=zX6*WC>1_>y#&sQHb~qv zIbwr^`qD0(4*d|Rg&U!j;666UND+HufbBlM?Hqt0lL6=du_X8Ykj zND0Yb@)bU+mA~;Hn6M}J(Zd7%57j|WaNS4Id`R`#5(LOXiU-2ZEB1 zrAiw2u{CFQ@h{v*8OY@%60!TJ`#~8|4N>)VGNpDF_d&YY?!(X9Rp36hBbjm^?Unmj zg@{Mp#|)I`x{ooy*zTi!p!+z7kKI1+x{m|dA|3aEBFKGo2#OSmn_&jmecXErP3S>< z1cGNBS%=MM+((zG&A9fmGyn-2WVw$B^cFMpG*e(bOEZm977278T^ism9B-AxrkJZ`^eZ5#fok=BI7=&zRG=&#v-BQel^Iy zBiMBx;*F}}29gFL&Xb?&$`nPH#(j`N+I?IeURxB0Fwx*kk~-Rbw3(MKYD3gai&a0> z?&HGK_&^fyJs}>bAz9o9Rm624b4KPD<&okopcm;B#(gmV8uvj;JMM!koH{-Xr(}Z1 zM|i=5^R+ zqyOFZT@g{+;HD@$a)+M)vSc#~py+P5 z(Qs|Mg^gH4IU9ia8f0;8x0fug{UkRYjW=nO(mIf9KcVfXCxEy_sMn~#MO1H>Yd<~Y z!M0RLWv~y#hrlqEnaj0*euktNT4H1pP?a>&;@XI}(f7c$3~B>FM+0)X_TO!BhHD8p z0+LVAm9Q@e znYflsLAIBCZMZh6GR%=1O<5Gd;aY=LTuU97xt5aKTuZg_AI}G;L|mI5W^pZLVYqh0 zIFD$7>@yMMW@?ie_~mjf-3UGsKAvlrB{t*aUk4pXW{HR_oTytlN_lv!iL{}sofZ*9h zb!8f%OT)FKkmlOGV{(Y;5Wa2jKao0`Yom6hijN@Lq{XVAYObw?Ro+K{{|VyXG$aex zQbk;@jr}%<_#G*NptBEwbu?Vd9BH_gly`t65b^P#_e-^55i)^8t#Oz#69knHXQ zKZK<$??Uz+f;>a!;`Q4Eao2B;hcLw7b_tdTQaKNBY07DIfXnG~6a(KVr&@$t&e0UN zoB~lewGOiK2=byShahe_*SMS-r?K#cmT(US)>^_{3I`FZCESN_zHkKnL+xAnFmy%e z<$>t0FsiK3^E0vFoBjbUTheDOaOB7LKr5@`$d6q&ojqRkL--q>#)B4Rc_%Q-E>*Px zQTP=>|A78;?ulntO zO|~y-5bZOk15vWPn$htm_?lYdLqG(AIBc(A8T+-zeTtSYaUUAQ$sc2QAaRqr-B?{6ZH*}V_=q5L=pvw4kRWoQg}&ak)NY2i_p?BW07Y%dc*L<1I4nR2(O-gG6p}NtaKsNX7dp;w6KoqE(5;BK__*5~RZv0!0K^B>%~Ltmq0O zifd6J8(ZYBH;alrh-Qn>Iya>a)m2%9R}}g?kHv5T^^ps?EQ`?c5o3|~x=)F-$SDd4 z8BG=`j$f`tXtjusgpaq#4jhb5$=d`Zmvs3Ai*zelPmD2WN~=mV7CFBb4@dwkCQ!Zr zi%hNCloj1%L~&i`)MShFT#!#3M>Jc6*1D+5P+gTpcvX?V+gEt-rTMlz{z!L13g0G+ zz_pp?!r0&+N9m62*S0j^Bg5-LY9^#cg;#wX*k7P1hIeSb?G-z3%(uPpIC#R`tEe9) zyH?qGcL<+rgE84fiZbnz4%jud5F;8i-?kf^;nIMi0%^!9P? z8jDoV9wtbKVCXE%2Uw&cwrG+rMU5z~YazUTpO&5lINRhIM6*R`?uOql8zV+d1*y7G z)zwr`++STAvneN!$|pwJ30?PNF9sp2_NE@>^D? z5lUnAn3AQ%*U0Z0f~-zfm-%d~)1aY`gpap+?hpN(yaE!r$t7Jr!Rm{#g7;a2rlM7e z#_H#5;?W3z;RLD^VD(>P!dTI_j41BUNoCkBV{exb`w`7nry)<+~I8{Iro)k%8Mefp~>B`Ivb2KRehZY$NfKd#0w5teQ5#9 z>NHkitX?;NQ&9!^^+b@3$m)af%kAqleBdMDO}%O_ZU!m~NV5rd|p zRf)#xFLlc)NQdhr$_TJ}*W%4s(FjyWiQ>{Gpv$7K=lwIEsE=s23k`m-)hi)Jjiso% zQPtI0O5B=}v26A24~vVwc0v~^;(nQulXL&utuo^e1A=(|1s zT-)d^7N8Uxbl|uHdZrSA4&akZ4C-KmdPO`pC_Cs(2kul>d}3yWFFT9&k|>iz<`Xkv$vMSoL|;Jgj3trz#7u*TSn(gCLr@rtlnu=%W_Hh_ zS6m3=41n`?KQU8%RGJ`7>HyF@5M(|vvnbjlxHym{(p&cBdMF*j=?1`+ zbFHKfsbw<>-)Ja&QEs1s_DLG3K7f9yFpQ=vXF$$pzhB3f%e%W9 zSs2(7pw1eo26gvK)@+RQ?rnnctU-{DG>8RckfV}La!490^KRVRM7%>9zYWClKvX3@ z!(+V6OSQ>4Lk>ZG2W`ZjJ}&qNfPdBr)Zicwh7s5vt0nM7x${h3*NEjXdYsjWJ3qtV zUEv~_`bO}qAe9DJ4>017i?B2UM!beXWh0<6G}1C+=A=|MB7^z^Fvx&hBhJs&+zgB} z@GXFrX<$|(R_dD#WB-uuG>El~Uge3|f zaD^=+64*8(fs2>Tg(?0fVWn^clK%ZKW}Wo!eb(<~*6&xX-*{Ru_!#T=^(bsjl%vX7 zJ^fIzM;qw~mgLKw(Dz5u`Ae#D{Kud|x2NyCjj<;*(whLttN3?QBdta%bqq7tn7v?7 zG|I_`z}X46kwU+YHI2l;G=LgtVAe+JSqjg}T_AnIAifK1q_3w1i62Shu|P}@M43jS z+Bl7r2Vd`%T<`{fx9bE}BN5ncBm!TQ&&>~WdwRYSj~emfpVe{xQ}IZV3&C@kR9f*- zfDy&Vv24U&P^hc`RH{Z=M*Mjq&&dZds67B(49GR293_kq88`}{DH@p7h`9^n9()(l zj|}2upbsbsZ{^N03%NBfweg3i7^1kA^4Z6k(Lpw&cy3EdLn}s0r0K? zxkeoSlgAj5fx7_uNdvPQ@g01QOJ0EVhC$p8G~%>Rii-b8<49C2AA-`KQO1Z=8^?&n z`rrYA3$6ihW1YYE3#8ThN zj?JQw)-{MosAoB`&TCUeF4DLw5Pbqs#-LOi$DqgN;~9|)o&oRzoxn0Efo+2l_@c}R zt?e4|X6Z+bc>I|XLY_r_*C@~Yq|*Cq1B`h2FHCK~i0c86k;oz!0wXOWdjC`_X&6)i zfK&r=jTriZ>%lg94>@o^x^7?En@7;(%^ zybW`~hXFpR6Iezhux&&FUzGdvHF1rY8uzFXPo-mWEG{-uq#$_8lS<>41sHMF{9rcX zT@)%?0o6exEhBFJLOB@*4FMq2fLtTCznaJNL_|h*TTLh?B-(AcqUi4@;Cp;0jwtB(QBn0$-G^awND$%q;V$5il8%lXLB$Z+p1Q@XeK0iP`F&zLo5C0Zxq-Dgndh$IV#GtJJ>@XnLh;P17%@~n^ ze*pBi24*$l@?~jwpNGoP2uib7fkwPmDKQ$RH2O2|?iOtj8T+STSQn2F3$aNdvPQF>)!!0GdJC-XQt}8qxD%O)-!(eg%lk zK$I~e)y6Sm-nMv6=Yrn@_+y>GG9rO(BNF(cOgfa)HDXA@qefiU3_NiM`303ltzRXT zrmP9*iS0hYyB~Vuw*bjvK*egLWyITaG0hB5&7cMVG&3N#C$9M3J(!Sz0|9zX1G5_O zwJNyzW)3tE1z!aCrcPiPk-)YQ34Bp* z!+I_AirFmjQ6s)7;a01|x(yLLQ%R-4?*|yMXC}t!VZ;akWN-ZIr;(Nsql+|TBQj_T z0J99pHDdOS$)+bVa05WwH886Ymo~t2!x2bN8^pXoBi86&RV*fr{{tcfl~ejN${3Mq z;~23?({Rj80V@*$PSpu4BNEs)B7ybe&f9q0*(xK_N{H8yGLwEhT_gLlR#@0Pl(O1% zyew_n?1JBoa!g7Rjl9ijiV>@Eep-z~LHUP2ix>4qlLp{ugOz`kUDc&#Ht{d;J%R13@n(fh3^WyHg$=3f ztg!j;U7F0?SBX}JVBG98q%;G5Pfv0_0MLL>doQsb+)bckHt5Zc4yY21diW202-ia( zk!0nRSA*8jTfP6c){PaYqzx)j*vUy&AkcT2%@wGz4WfZj9)!%G^=1CmWwDlmiW^{q zXv~^H3AEtX|2)>EGw4klMAHcjYDiAkJGsv6cnL;A(B((YW$Y?);Ck9s{n7%)fbTi1?HmRtHBl#uB{xb6XtM^2Ur29eM z%PX6++=n7^AgJW51QlfWYEspoLJPE?DeDby2C!)t!Vs1*0PNLtKDLx}Et@yjtt6#MT>vr+NagbW%;q8~IJO{4&dZgX^4a8a9b}t~ zJUIVLQ6i~Cc5l>Au}XdrvR_eV(%JV(WwU!<_#4hA z2ihw?0ym2!J39txoZZ{xFcwcHL0HoOI+MOHp>;+6;7urIV0LfYla-mQI{?&!fK&^` z(TZD#XU|L36qP<|pHC!>$X>8!QI80T%+ngT=cH_&!>6&tYY7s5jF7Eshk+t&NNz`; z_lIM}2PZat&F-JnBh$79HAB~n3l3XW5zBhN=vzW0RSJ$0b-Wpq5q22;unE`QD!6B>4h@=U@-eUO@ z1yxmU0pM!_&8DJ)xuK^7d#emhP!0VXWT##Ei}8MuG$YvCXnZx*QSOrrR#NFBxyvpN z_U0;zG1|O{k0BY`N_CQD!QP>(&^jrIuc2jiDWRYzodjpE6LrVRLZ)&XJhpkiMbo9OzJNi?5#li@j zRt_S0Bxg`TM$IpWt^5fPRxto(18(5tiZZ(Y(+Of2l1SMKh%N@j#c%^BH!jM#9ZDAslXCSX9WsOrha;=;EACQ3^9jiQ4)u9{yxCNNQ0oOQ->ru(Zd!utaeP+e6S@3nl6AAQC*@8NFdkISkS_ z3}QLup^Ta0@qXGOQLI6XluLnFV^ExrLN)Ms&-KM84H>lyh@T9K^WxhjxuwT@vRz3* zRg#wh_?tkVQdw4ob09J;+XgIPCto|gisfs6O0<3LtmqR#AqTsHO^)zzhYjZBgP5Bz)$B~ShMUtIeUmHHXyznf9 z@B;(*fl$l=#@C+R7|y=-8-Vr`kZO^|*HSf~d4u&~U(1P2U$grs z^~kiFM@lJQ8(H1*wT&j=L83i!?1JF=oNhMbYX`hQeeQn{ysm|kD(3pyg5C=JRZu&OF6 z<7@AP6ctoe84Exnfxe@njIXWVt+Hz9hLAOP<=WSF>*#c(=OOEFj;59kTsKzVA_A%N^hN+8-%{$N1W;^pgUkrREp_)_K5GUW%4%yXkXhnik|-=8)M|B zNp5^?TvGt%L-?Kn1p3;P%T>h%B$09(5Z@aV7h`g z)Jpp`t+Dd8y=%bGWO~cjj_imxT)6@{d`P(*rUo>=w&BxQFD-XM@S_&8{;se6b4W2D z{($svgE&Tc7++g?K)g7G7%9V1d>#ZYMCD_A?LYseIH+VGsv8vNWqj?5M#+MzB-;Vd znLy{MEW;r9yalvKd-ExD5%RV705R{iwc>4GyKZ0uag&U?449Ra)xWrMeeL~jO+@y| zkbe!?w?=M!?RS$3viuZ8=Z%;(p@M8Kdl6!PaxLtICSO~!rsZo1YSM_bJk^twuU$|Z zEk)@pU%PuL?&YVEx-Ehy47$1RwI{KVO3Ho^4ADZFn-Ip=KD^}-1*txAHV}&gQO4Kq zyNHP*xdGDc8j;1<-r9*#^q(O-r2(=S6=Zzv%o><8xJ|-Zj+$x&M-e05Ya2hnhp0%C zf&i2@AeD>uwZFZJfxGGu*3kgh*G3|9*Vl%eFD+szyS_laMCe4)$N1V4pF}EOIvKK= zMs9rVpwz}nz6P>SUAgwP+h0yn6#N5ZKN-34wFmye2-CY z$g~snaZoe8*AA;|`Px>r&U+_v{GL?$lgUB9*Df82hYWcNf-71msba3L{bDS6#$j01cv(TVZ3ADs=uv3Ed!f#BXVWCG=DOB-K%zZ)J%_5yZP zXKwjgvbgY#gbC8-bjhWB?ZNsmG@0J=wcDSHM?Iq<$%o(xqXsm-c0@sZuUsZUP)!RZ z>F@g5R*mC;XbEX&gNUL$jIaGH4PR12jFdxw7-LYJkMXtTrc`rKi-A~WP@I?XwbL`| z396FZ0l;?zDnw-&1`)pVFh%Nwd{}>mLjk5WKz(P^(kt^xxXY)oj6)*ci{Kecb<= zxK9?ZxV@SP9Z|4FGv{c4+!<&Q;X8`@_`gj6xzRxSMx-}00K4@JKr@~JxE4+O00H{l z2GtFBme*D!Q0mqp_1Flr8^Q>^hS)~DgvuI$M!4S6F~aAq*$7l4f8fE)ypgd?y>wj;T2 zkr2M_8)g zBBVFEhJyZda}!)yO}&i%->W z1k#)O%`4`#pO@#ACVii%K+9~&OfQ+KML-=PEXsA$IZ4`H7Mo}LV{UGvw-xX)ywowLvTAI1T#0d(@!6w#A)IAfj!P z)3$#@l=oRnR!H54k$Spk!B27I>;yblYe#P?|$2Wea`M4T_*s#Xd%vi8+ynd(zFvJ{tc+ ze0Ky-XR;2P&zy9#{-p@@@Gt^^F&bo@baP^ab<)kUOPCUVSEH0g^kV&Wl0HbfD6u-$ ze*$rcP(!J}1!PnAq?=zi;5CNKBmcs`dm5%Pb5FYYW(K$#OI8?}|LI6k1c5UY?gvTx z4h02Z-4BCm08r0>+>>sGy_sZ|tPrp_FfVJIdTTVNik&K)B%<+{G8yuj8e^Sw6OZ)P zNjHsOKs6%&zSWpdx|uZ#Pf~}F>MR29&Z3J5el}93Alpm6=0`*7?S|0ow@gyge|!Xd z8trY9cos3|7LZBJ-nQ1u(2>y%Y9YvtWb9V>c^lk}G{zgD@NtaL9 z+xGQ_6tT{rsc2QA+1vK_3tmAw>?6^a)bi%sg5~oPS<#C|WR9?*`l=&rNMn(ZS};2= zuc4q8P4|5j;DouFO4jX_nWJLiz}soC^zymJoI}_#a6feS}Sy1$D$r z5DnL2)dBVLnihv6#9`pygm{UDWI4iyD&iht^X1uiaSSPT0li13Fbiy$Sj_?(QrcNy z!xc_#@Lx?P=nq_@W4&uG3KmP0;Pu<9W6cqHAf9>_c)*&{M~j z6ax_XUqjF>K+yLf3r1Qczd{XV2zv1-KR?7(8GYS z33_QDK~Kf9rcYdgUe;CRaCe!2|^`cqWZCFp$l;hFCu+iZ||3Q%LQ`3ylX`d6qI zn-~B}XplwF$*)@kJv<$6&2=?OStO94C#=O{w?07pk5HAVzz)=BT!K#Q4!2L{kuw2V zs9`EIm!R{12+cvzjQj$quQbvk=$3QwvH*f+&`ALPFd&zp%TIGR^AIouc8f+(+65l_ zIAnSSL7T|RLY}NK7D4Z!^zN~b=}gcAbJk;mJ{N{(o&O=#8wgC$bP>VtZK^P`z2s{q zXzK0eG0`SyQq!Mu3C1RZUe?YcXfmlG==Y)%#Rimd8bQ8F#=ePPE52>P5zVMQkyks)Yq7PbUw zED}OGM>9dwYosRVkR~<64kSH@;2BMIW$vL%L(rsS%&q-8e}+ zh3pFri!=(U9W2?F&`;<1Km%jFa*uSY6zN?b_kj) zochyz4A$TYx>*MhT;DM^L3g_V7T$skzotwJ%|IPALGN4!ZasqN(^@2BNv;Wc#XpTi zA|j-`4}ceWb9x2R1busQ5XOHXE3M_q5}7p80D>Nn4R33;AY+22rIacYeti)>Y9jxj z-)Th!2>L#7vHUvY67;ofL1H-K$5Z}GsNp!NA?OdvCW?iStuXS{)KG??8|JLSV-sIP zEpE!u9<2spAI}1URE;d1T4#KlO5i|`A5=hWX8{*Y4tRs$oLh#(80`pLxaS3`?vp77| zBM+GjNOcWUnYjf0#?NrdAZSK*0_r)9vhXpzCzD2%1c42zv3YIMEhmj6skobSuumFPEU{?(vcE z@dPcuiRa|I1CmR+d;&pF9F|AiGH5DVm1qdM$V)i|>5vmT%c|7!hM;@CR-P4&HzGsO zRA2Q3O&W`YU8}HUeqv%cje}`|&a=I?=z*kz5Ii-huFO4jX$YDW(ggkD|GeT&2$vXq z6H-SLbf;&ki**ovsl}=TYJwiIG>>Qj{Lc`d(vU0!O%-toy4}m!L_4GifzFW#tfL`l zCRRhxq_jiOT;Wt%5Rb_q=&MK=`{f%pLC1CgcXvj7cgpk_)j<<2q_l{~5YmdK=`QV?{%@pyth4H*+OEn{aT)Yzu* zQ4{%Vo@$^7dJ)DeV|gUWCFmQ8A)*k)BX~NKIy~rP2s-=RY@#}34UPPHs)QlvkdFCy zaJ4(+vOkHKp!4NT5rYslQYTYtXCY|P#U|)7^DziZ1WhBWilA#MfN z3otf8*9;`+n)Pamr(J?R`>ZBt6hQ<%s~ZUV1`OsBbom!CN>&Z=^$|R8k#*R7hMhsmZcL0`vV9Foi< z4+3&r!&GK2L61L%u~-l^BmV;`2-W2bErRa;IaUCJpczyQfLH@^3A*?4U_;OZYz#~r zjmtvNtw%`lEe5gsLq14jEP{@NA{IfHVuD`1zbX^-w4F8aZh%y~5tyLqB7#TGl_|*f zlCPPdskbvhQ@l;kq^5sOMQmUof9#CDGd+5>-G%2JB`aqM0q9TMf4c<2s z>S%&4i?ROJ5OvjJ)d4j@Kff+buePc1&ps8q8qG{{;!E2MD#`@ip%$38lSi{ ze{_BEJEHlhCK}{Zm7%(dxL#hInuzn9Hat^Dh;OH0#HbNNH;o!GWCmqlPUY*mupIT(e;t=@us}rsuCw(5|CWd2QHW2?3^znU&6p2BA7i6gT5=JezXmzoD2VcC19NhA=tPNsQwm^(}w|CR1cY8J^53~5oUB#*mGUaEzEmP71tTE-Kkiudp z@_HXZRv=Su#4p#B^yuj$;p0vDQ_Gs1{8B)2NtaJB<>s6DMP8_*3ZtS`iN=)uet$}k z4iyNL8eqy9Q!BBe&5bDTpCWK?I2`t+N#(>Vh~^GQ56&uosw+(SGp5|oE6;vkUVLDK z*pv$qqn^UuH0mkbJbVlP0P>4uN(q-?#;eKI7*lR55GQsbufGvw4KiiW^R_9;N%%4*#XXEK z%clHeKm+k5qS=(>;*>U2R}t6leRN%z@{Uq8rw~ zs%?E$uLlN%dcEBW>wf9rIQk*{TgSmsP)^1B=*mN<$qn)<(3ZlK)7x`!-;#wI8W3`l z!*eMR{lNFS8e;f`a_VP5$Z7Cy2Q-Uw>fP2UZk7#tdYc0}8bif(KUP-d^tlbn^Q8m& zia{eoi!0D?Hs~x?_H*re)-8;jvRA^eDd{BpTD9BU*0FfIG%9Xk!32P){S64UThrRH z6aB!4xNFcz2|!J3kQm_SmB~(sj4G22@H_|DNe&EG10*}y?Qwsxv#ha{kK;1Z}cT%0=UbR6b zRy&~g3s7;%qp<{ya#~=6F0XVz(;4*62V??(w%efH%N@{(>SXcXBOHsLvq3HEI~KoN zm5N(k6mPs_%+UVY1gPCLoYl(~rys)KZa9Xz2vp7n-GALFj(!M#?O{%{w6;Ow87C+D zA^eGR;B6?Up*HB=C?_ZSA^cU_Io)@$0l}E7(jCPwR3XI^my(^3(+&fI;!z!);^>F) zuk1*K0nh~-^zUe=IQk*{JL_P=m_QK&EE5##UME*q92$P^x<8RK$mUM?Khm9=!ftRtm7!2bD%Z>6z`YhC?3j+fBlzp zb5=AUDBh;4Qyl#e{!(km0in2OY|uYgQ_HoZAHsjP2%em&`;Igq6j!*6Q{3K)WT%^> zi8BFu&jyt$<$&TDbf7WOAwb{Rpw%TE(9#sjDaSupnNOgrHfWh%aP|u?ILlWYBMt8&ZJJqb$ZDUyoI_ljkLCBq$19D}~Tf;D#pO1D8T4Rb(?8&FQY_c*sycN?^0umg%_ zP?OulzQ}2U4Vv>m2NYkQa{BC8hys0RgC=!xKs{>_sKs4}><`$WNDcax9wAekm(0ea zvl9kk04pBjq3+7(a+pk$_mEaX=IOTx@7Q{DHd_rCE;pi!YLOg5ay>eGdQ1V4i3sd} z2B3Z*NRQ42eGV7W5wc!du7*KAl-XaDP|8~Q{Nabr7;Wy4@kSwhR3rbb7o9tAEz+9> zcG-0O^6?{b@6L`nkJzAFy`2-0B^#keKRjk(hcB-;oN z^5bbPJL1C-JQWvPJ)!VH%)%8xbi5YHnk3gf;jh1|i28_-vK|0W2ZFQ_ej5Ph>j~LF zE!Revw*)+G8)5cF#|ZZ~u@R_7{*CWB4@g^(-Wb8Fjj)?vamEdG*!+|YTHo3+!p97H z^II)qoyU

nml4MM zoGk}bGzRMW!_xF|h91BrQb~%uIZywUh`Ny0l;N`>Fdt{Y$IvJ0Ty1#5jAtpccH!Sa zYN_y<20JQ{hO9W1Dg5u&cevpyE{(L3Bz!MQIvIZ zcif+O82cjSx`fIzCwHgN91~O-Pom`{GADP>nUq~DMD#KQ&l(b$le;5lVj2X|Uu%)F zp*gubvpVj{qY$1k0M6SzxqDg31VNho2S5l^<^pU5dP(o5UMZpwRmaFbBQ4Cy-SwYJ z!J;4t8yNssu3!?r{2lYI^bPMW@arhLSE~sye*`Q?o{yu{o-519@LV+PIY86)@odU?1qu=rsTR) zeUV%pV`wNT8i8|`d)@5EdmH7;B}zk7*@z$QREOe%xFuTvBM(w)-KjpT(}*Q~Ny6nm zzEcUSQ%!}U2?ryhM9!E^;i)6QBxDs4OYtE=D(4;Eo^Pm5_zmU!i(BR{?=v!#%RE7n z6C^4AI(A;gY%53k_$az`)MpJu%y-_{lU{KSh|u9C6PE1pPCMR2T!y3sNjQ^ITgs)1 zn0-+imT*fr&B2)8qJjm8bua*<4Ty7}%}1ce zoXmBha$!N$SpfO_ff%L18SjczkJt>+H%5G))KHx>O?o>lu8a9z2>&nuHc-J%5hX-S zvhanDBy_X2R#fSTf$1KO6&yLESr0$oS+e+IN1 zItkte8iTFRqu#>&PNtW__AUevqb4pV2M$UgTN9L@3O?l|0-s>OgUF}%iQJtVhWO= z%im^18bQ+7WTIIje{u=Y1+hI4JQe6X+W%^VTpQuD+Xc+|MG3<9In3z4kr#_ks8cpXLenWbQXSKR=2TaA zd!f;|+X-&n?bqJKSeG1xR3oX#-bf)inYr7?eS_z5ITzA*41(_ubGN^{s;=lq3BLef zmjPA9FZXUQ|4XU@oka9y14@EU?%iJJI57*N!V#1*l(BOw+O?*7`oDOG-j-48bihrm zSvmjfAEBD`VbbnKh7Xexckq)blSpdC`&IZgB;OcV=DjKQYsk^F@X=F{b{45JiL}r! zq(i>L0LuzQuhk+shvfP-WOjVCSbzvA_W*Dt5Tsv2e(r|5_X=dUwOqf3?0-OOn5-Al zqg$QVkVV^IEa9UX`9~~t-pcDhcU5#-O-u&Y)6#8?<|_1NycEjY8M1>wx;&pk51{ zF9IE*hVzd-7o>`tVS`fhI`0*k_%5;V*SgwNfi~En8yeJxK?(b4kP@}~)dqc-&&jDQ zgWjp*yv97VL3y8YKtT*@x!f5FD)EM8f&&`#>u%Di`+TQ%jcicOTux4(5-9b>-I9Cx zLttVDPqTY@i6BhEyo~rE2%ejEFJG}4qN#|Ur$zFi-OKx6H(~ar)XUcbur&~*d-)fi zL%#!%{jTM@m*?MwzGCE;cQh!9 zL2a%(qoSv6Q1|RkP8YtR;$Dhzh7!Evt=bg|c0k)06xG4GJ|e z##2rM6We%(-NcFKF!I#~@f{I7?a^X}$i~emf#vy#eoc#Hcam!&yL!B#c!8UE4ggC7 zLAr^*%m%->0kZ8{uABJVr!hNgH*x;&oF+c;JvTAc$iMP+uQF%&1j`8ihE5Ydl3Nz?@Z|AmJ zYJ+ME$L9a+pq%c5AvvdSY|x#%PELCn)aRZvR(r_?t^L;ly~m*axgAQ(G0`$Xk^2tl zH3FsnkR6YMXyR{xi7hnCZsIq;$JAF9#HS;8w%AR4>k{UTS|PfN7RlXq6K}cLNbKb% z9sr)oG& ze3d8Eu1_hcK)EJapyghtiGOC$M|Hp&lvAn=+IhpVRJX6GVa6mlHt%ACGH*Jd$_(1y z+qpr<*q|!698e^K{tJyz#jUhKSFbvtb6-($%|<$|ajy+pdCdWBVbJB(&KTHr8cHj%nrwrPpK(C{ZKLAaHpT`v0)1kG=ACswhZyv5 zgVPTV+o0e-9MBp9rM4&uHpX2%c`BNC#~izfpFe=n{|88$4KmMIyNUO|lPF|9M3>Ma zIo)pJd^_`qIo!nQ05l2&=_c-wI|fHJK=y)`nq+x)va=o1GZ`(4Ym&GWY{6=?HG#~qvB zIKeij8u`2ZA6;i2Z)5St@tNatU0h2X`?W9Ea<6^4_I+QAk``NuY*A9uO`9bO5h+Wy zLm3;|G$x?_){YX)fB>X;~@0^)u&WS(fzV0*M&&+(k-{(2+Gv`_6`6+T%yxb%& z3DfPB>aXM?^!$qa@rR?cgD)y`E|T*N6s5(rzD5|4B;?r-&?y)LW9LIUmyE_ zh$La!X-NVj3K#m5_dX!LRQT==yxI0jd=*K;!h0fwo#^#HeUF@pB;n{;NdhAZf88qCB%vC7cS$I2lkm3OY4F0z2&cOFqKn@~ zlJE*eX`}O*Z<*|`A_+NW`$^~?hbhgoXgd}qlxKl=Bs3_1-z>9Gz1W&`QQGa1&~R!- zRT>pqZ-!uRw8$pm^Ih=o6l~|L-6rAgMQG|L;ltk|Nf>!SlE8?!zVGMw@y#EE-&DO&{}4*(2|vC{H50J^Q9ah0^z(jvb1pSf z<0Za%5WE*HvhjWY64uw+4BK97xA9&1yM1(gcV3LdSLiQ^j}e7?@4<(ZM|%FLUVP=> zip1AUZUPxo3WHEo-xvK7V^1)SOa7{VGX#8-kV5BAbLsTkt)x1GWR!Zj*2Xce6z86K0*<9(}nBmIvWSyg9LX#Fk(YobLq2|qr8TLJdK zcF5Xo5{B%~6rF^6HzG;U8sAnbff0p2Zj94UOhSp*yd-R&9yt=e;QcvigYud$wj5ryCSAZ+3bP50t!GBpxkb%}4)^SHYT`)@5Y z$*EWXFW7$zUc>QS0rgc-LYYAi_pA0B+y|yxpt`d)>D;v2{dXvTT~&a%(E}iOJX&Pq zJI#d$C%`t{+HHInI^iqPk1st-B)*@sN_>ncTy|4r9`rr79I2X4h{Sh9Zk;M%e}lx& z@I?3Ji5$e6C`!`@dAvN5^(IT6KKTqOFcn!9I%>%wD?sVkj#Q}=a4h|AXscig{p%;;LOHC}x()=*qeV6eU%kgWIbnOy z+HDf97s9u#pM=AuB1xE4Mv}mY!uPxwnSHKAdzXau_J+z5l7tM0A}41K`J!V-BT2|2 zN%(An`6N%Ag(MWi>lFj0S6onC=^JR95w=iI25B$34EYKNNkLSXwkAEmPr}ow#nn(r zLPH4JM2l<^KI{up^nmSQYqv>g-8v=-Pn3@&;Zy}l0wW4{EuGUG32V^aC82>$!VI|= z^o1{g3mz`t`l77|B1w3LqO`lyn#qwcWj2!V=LdcgM(@N8M{%gnf)d*1C!xfyGI;+L zs>@rG-tQ-&|C);GYe_;g2s%cKY!a&F!tCz>*oIiUO~To~af-=55By9O9GDSVBy78* zsnQ9k&w&z3e#bi!G7QC8j8dqsWKFsn?Us43k87gRP=RGsA?Ok|Ue5ULSL!ho8R1V$82tPU`7CvQS~mxRVP3C-n(-<3)DxKni67d`h$Bnc^! zgz#%Pd50vtITuNoxyDbznM25QcGTxV2~G5qP~-%@Q_7*bx;5z;eiEjiub^g25;{O| zceKbRp>j`TaxiR9TDwic;S(`QXjvzcgcWrq35+OwF1?M3Z#&w%Bn+~5J+GI0m5-!@ zUQEJSU(|n1Bnk5=N}IW~l_cSsUIdE6BPDOQ^HR-E< z5+;uX-k0!5xC4UwqeV6ehwxCGeiXKJYqv@0^boHNdnd&vHjE^pb7M&YBMOh$f+Lmj zeTMcf2{Y|o-IXK>y~=`e6rJ})o!*ZmA)h4S@yR$QaU?97ha@yy?2G*o|&~85tTWaP}4@wfcLC_~!WRsA4ULCA|0o(J|Zj*54JbyuX zNl0lPNx~m3BngZt+`J-o664#C_AUu;*t^yj%Dv`CbOekSeNnBqBT1MjqQ5o10@zc5 z`Cd?h$$7aX}QQQsaVRCuX( zxXd1l6JWhjJ;<7LS=#O4a$_xCYl;f3$3pOGw8$o*?KAj_S`6C<)^3w9;{x`vpM)WJ zR#(+s%yX zC^X4yKlPJv$NMm~L;dY2q0xR4mY2dW)Q3<#%$oEBKMAWIPF7PS2@@fh5iPPw7@9dd z*8hO*BWt%wNSV#U(>oGYca0<==`KkEBMK+4z-Nx}Wmx3Jr+Y-E%>(4AihGYnM9IFW z=2MX*+(A*=Vmyw_BjE%z$rCpFN!T+HGx*M^zY8Vwqo0I(AI0^rfv6r~P5Qi_gpGYO zs!NiDDG{D8-6? zug5!LW-Y=&LDAE`XonrHmJFA64WAK;=KCVvbtSh?$z56D%W)k{qIN;#k}mf`Nuqe8 zk_vBnp@dQW=Bwmh5>fF+4;9|p2n%9xiS9h_czwz32%_S>0xEp(LoJM|j<1p{+M?p+ zW>9uQ$X@)$SIOy9QSsa+j)~;A_50 z&K!!0r}b30#~bC1YJ;zmQ){B)c`uwZs1#A1@>O!~NmM*Bq{74NhD~4b?|Od885~jZ z^a`FE{4KI7eTuJ=^B|()NeUG%*t(_Z>keNfCkjNxU-7t|3KWt<{z+dYzobRQpSdc0 zHAx%QTwf(WHAThWfmoROe?^UIyRVYpVxr;?B^54NB!Y-ve3krI5fy()sPOhBk){4~ zz2{{@X7i%rR2vrtdScxTBBGYBlIgCfI2Tpnhv(Hds{4GEO!Y*?c^K|j7~jOGp7&KU z7ZMdGDk@y9Mq{H|;;Uq`AS!;7tMHy@3LDjaUnSqBqT*Mb3Wq|GyOS< zCLztX`IC?uTksR`5ZWF?3FUatn~q{lIi-I`^}p7n3({_ zOqR()u<%~y%o)^lw4cZR7vSsBJb&sI-W$@onOXzeW@mq&-=S{dy;okzArGh?gI%wt zNfzFF?zcATS5*CNn;Exb7T%*T-okr3^qc{8qSpUkX&1-qsbz7z<%l44Ao>gz$Gd@h z1a&vm_dyBmVFVJQTO9B1A}!4$v`<6uycKziI!1i{lMkj4Kda|DVc&D{vANMLJZ$`v2|l z0%k6bC!&TBGY_d*dO5J)&D*%l=I1C!2=ra7c+=2gsrKxy?C(wBF^_13jOO)m> zMm5*myv2G-6FMrj%BeZD8PTn$RJdnGMISz=sUr8hTQF_lh_>SB8E0~f<1t>dI37J# zDHXs~u3pM#k?rcGvtCG2C9qu;B~+EM3Lm)X7RRGQc5%F3L-70;jQ2TxeY#^8$1C*L zZE7e?qpaB+fOc`bo=r-srqEA=d9IbjERM%Og2nNAo^PmHqs0!W_u3Y2aXf*aTO5y0 zM;6DEfYX-uDJ~FwZxtXkgNrn|>(ym%$M2~^s4tEZ+Q%5|x$zby@N>K-svBFAK0>>N z=$HMAs1vBb{eKYjjTTvm{%=|b-2VsL7;86~Nay3~fasnA_`JRan?RJSK8X|NA#|+& zk5Iy)O&CPh|6dLN<+oG-(HrxdsBh8!C-&cw?#RzB2hr2*5B)^@P&f>#l`ji(afw~Iy$ocKPvvadTC#b z7pz`75*mN?($UeYmuAb=PH`pa$nw(j*H}Pf2wDICLjZauLJa`@du4okUO~5Uu!Y7m zJCZ&J=pOy6n8*8zKv2qxJV3A9?g9F_6ol8zs!T?r0o}eeW(N;LF`TMJ9B?Od8UXtH zi#UbP^yn#&%(61mSpeuUShSCT7Uc%0He00!=(=lhu7T_Sv%8ZJ{NzLdpo_Q7|G!oM{o0&z0_cfPU_LM$tzJYCKyxEe zRxdDvG51=w1<*X&1<D~T6T+Mf*z}sdV`t^pt(+h1$4y`nblS_{R$;?fw2mExaj~*hb*8=-Pc6@3FCFA zze;y3pj#KJuWm#4Nhs2tIRGu7YjwwU&1G2sALd3@5(8)k5&(L}`64PFEgph;kZs`r zEnszkrqg!)e+HbE=U1G$#4NY{IsoqME+5cq>!)JX9H1K<%Ye@_s=v1; z{Q~V4&;FR#kG`e75m`WAU5lf(Fl=S5-DD!2OKAYmUsl9#>c+4Mpt<6n z=|oOjVf}yf9Ns%K#Q>V?StQG8tN_r9+m=zIP(POaC(<2h>HvM`f;wszY>S;eKZ9_9 z{(EF$S$ty)?79d|0_YphH&V$1(+>QY>eenRjLqG9sE%jCa(VsV1h(^}ve{DS= zdOP|IAezz&C!ZUjzBNjy1|yIV9YhagE@YMm=>tK3EAk*3hdr7Zw|eUGCA&@ZX6(hs3;Xe&4eNy zdJr9kXWj@=5tW9Zf)fQ0egBJK31|x2L(|RbVrIElCM)$a79tr6`v|M?AesyAdJuh2 zAiC*tUV&)%ScO<&l^36 zGNlfpd4J8N{=pb!q0p0=>@@rrK$J�*yC^=uh{hN%H|w%{4b~0a1;Wre-=dhc+WR zh*odgQqhOi6iw%zcM$!edz?7B-icnC@7JzA>`8x(lY>)@%+y3(;dQHdM=?9|`k|RuThI z1`D+li_bD)9bW2--!9EJXWf!L+9rYy+*`WFnmp z0t2Gwn&Wih7}x}&T#i$mFpu}*9k>W3e60tbSK}c{ zxJ;IUW|qt6hRrNDe<*YJN%4teQ1Rbc?iUy@nC13_#-HWp57nPFI=8wPw{T`wJM_YO zRaJ7x&2qXRs5)J6oJMf*@35t8D%E6NB}L$c~C93N)OPx%Vrfoi)aS~dz~l%bj#K?-7J^F-=Mi-buoaxqiQ?6Q?;_+;sKf|b$}jxcLnt(#@L0TJ22Vb;J*OS{PZNycymB^yVyvYXTtzN zbIr|L0D61rQmT$qb7(W71N6H`b1C}Jm7*@(^A6DeTOBWsKH^Lc(2UmrnjWi^iS5b> zp!tBZ1$6#z^Qa|gx(+4Oov{jgxaj~*hb*8=jjpYZ!uX@pKS*~hpg(-Kh58$&3>ZSX zGY6mr^ztfo)BxxU(E}@q0W<>%0G+jBWi=En+CzPZZQ%edV0D0|(-A;Rz-dS0%0Q!6SFIN`> zK>yx6P4!0oAof23{+p%_(4)ew)pM}D?Cj4mNC)VICh>|-y}tvy9!HY^I>YgtY8|S! z*=EM=7(mk(AJC<)@HvA0${cB}B12r1;6Hr1cnH@==1N6wN^(1<|2!eO5$OH7!y&j;; zC*qcvJyvBh5)J6Pf50`q-=X-Is(l>rJaZZVdfGr7i%gHszXRWYC?YeR1%Ph16F5&?0&Wf^V_02a73}M* z#sl;nXzu}fngIIB_DTZi_2Y2b?q9UZhCL&I=0>6{X<$;p+-unuK=Wu9K(pQlG~Enu z3*q820bOpF2WY0$0XlqNZB-A0^+(a!_)&Zc{{?{Nrze5Nn*;jgyDLibWl_yFH*W#x zB6&NgZ=IS$n-LwLf2oyK(TBfix{b7OfG)YImN=RV;TV$xG~+dZrpGGf>s;nE1eX`E zfWGJL`l=0@c1H>2XRN{=ZaP5IAq(jIUBYT4j4wKUNxEYJ{ch7ZH4~7im~|yo0(Vh(cP`u%!x{ARpTU2*u_FNV zh|!f)2GnOq3DuxGau0|D^z+}9RwZFeb@oOK!U6i@-}1?t6|G^{Eoc%z&pew?bw|~M zwwZA|2GI1y2Xv)1ybA_^=7JRl(7OzvccJ3HfNqEJ0zl7)#s~DSXh2unmaL8hr#L>_ zYXOZR2}wR_tg2K|3q*0DbLq575sp z!8xUYR%J304d|+Qa2EJgC}vYNfdejMPCY>92B4W9y$+IXR%SX20R1smMkAm_`2$pE ztkMH?mRk5_2tbP{GxCugMMhNtpvNWVbAYC>GBkCqE(XwRcNbFIu{KB-*t=Ve2k493 zL&16CaRTUP>l#3Rbsfo`jaJK01kl__lm$slDwumM+X84F?E+}l`+%mK;r3H-)DzI3 z?ehT5lsZ6H+Ezs!#~4|l(4R5ch4EhiXnuMUXuLU~f4?nGnsSHGa? zRg};n#wzULrUNt`vVfiz&ZY{nJ&MtvqB|DQIp4{t>cG_8n#}=d0bP4yCG|7(_rcuH zN@4)bKmtI2|3rE9D_Tr~`Ze3a0b0Q708OVOfR=#M`nE*60O;ZS0l1o9`hcGJ6`p$d z5gndor#)661`Fsn|G+7Z8>r5LW@6G0({2I1>;WuVj0&wwK~O1LWC8u=HC#h%23tpK zHh~WCPX?L-9RTwK(A2j5~nn1uksOWb5nrlW-s+ zGo1wx9d-uC4j?MZWT;ZD(u3%g@%Z@zh>EB^1a~-50MRy?S~-YPI2@YMRu==&O)E>N z)mVLJChT*q#)D`Y+ItWkB@kVHE~CtHhm62aobS=<4-|nYHxgyZACn5^Udy&Xlt-jM zl=VJD>1O!rLpTvbi01gpgD6w#Alj_qZK?zggN`WraVEPL{tF<=BRqk|n?p2bF5F-M z>(r>`nwz(P=r=v8s?AQ#wq`^J(M5AwD*AAYrcvDU4x-PjDJ+g&b|wc=#%mx-k5$Ul zRdocSTzY8JTE=5_*oY3Lm)XAWDZUM3=Rzs(QjW!0BJ6I~Jl#OV?1(!1R(e zn*-27^bVY{n+pAWn3q{e3`7}70MU)fwbV?sI0*GOwuOVJz|TRHPDdar0jC`)fCXRx z(e(&8dHg{iqF0~Egy)}7UkN3YbtPi35Iv9A6zOKD?r2RqFYOkhwM!RMg;Ak(e+V9n z7Fmc+KiLG=eqfts?IsiHT%idNU3eV#e!mZ!K$I_75hu*LODci8DBQ0fXikXs{~G?w zEyn>wci^7(Gf-V*|8u$HDsXKX2ho8oim3#&%Y$On|4NLiH74;(Nf_iF&r zv3U2CX0mltNIF=V=`4U~a!WZ25EA7es2;US5282f;{X6eMf55JuRBoy(Z+MCI*3xZ z9-8e|7X#5jD>AC>SRv^I>_1wK2hnu2_aOR+Ky*_d1JMDMaAvtI_DU@jfhac;Z*avW zlM3cu%eFw2N2EZM^*%)DX85itc%Xt1?Q+C}C{yYn`c0F(>LH9V4MmsW$K`GO7eJIp zcmj<#hiISEHKqBHsOFlRw}9xfM-o*gxMK$6&}Kvj(e?9MEBa8JqGaxQ2T{B#Qyi`5 zOb()q*FcmWtCV5S<`jr>fh!Bq=3l0&;b{6IN+^}F3Lm)XAWDZUME^)FrxwAu!s*lK zj)mxg(T&wMnD$$F{E0HPPK=2Oklq5ym@iXuKbhzk51 zMCo({q7rc0oSImN1Q5M}hRMec`4GK*5^hu+jQYpe=|_ygLNwoLoGKWP>S@-bx6p1O zI_`%OY8NWBejkE$(IN}c?d@>4!RN3YwRV$e3y~$u6Bz+Sf1eF3p*{^I)RykZ!y68wvyK%}9bmiL+3#c!4x(L7VrmcU zJ_ftKmnMPeo{@Fbi>R7xn;ExbAWC1NAlkiVb^qu5%wra!k)QJ|zlFDP=re%m*318W%zlkq6PgzV;ycwLe1?m?6(br7wvF~52bV;n`% zpEKFN;J*N(Ji-%byg5X_+?FcMlQ2NgTyygl5FNBHpK9yW9NLWNAUe2G9Yr7dQgo1e z-a)iWk5qB=S!Z$(WxNKW^jM{A|6g^1C|9hr5bZjrs9KMvpP_`lVXVRjZaRq4Aq&y3 zQnRVEFkW){({#r|bnoxCsRVSN2SvIw2cU)MOaI=ceu2Is%(bi}2BHikfao9JB&grf z;(n+fuq_-!1%3{qbUFf22{^4u1N>qHM4NpF2vs`jLo`Ei9OJ9d;W~EO=L5uGA*wFp zx60?JK59*RFzptiYm1dvkD)^AKOwjtEwT_j^AIk{Nf%--$p`uKvgyixf#~t~ ziU~xsF2l5cC0cDp5r}dlQP!a}sbKE4YzstrL<&S%??aSshU;a;eb1bUp7`E_C{yYn z+Ty1q^*hEW0)-yQWLL(20YrI(C(w9vh_)+HSeo~aYOc9?3y8LBUO-KDY7T8ibPz4l zvZ106?@~0Hd)`5G_Oh(v=niLc5M{guqV!m$3><^EGK^2n#uYCuMBi+3o4SFfIZziG z!&rq6+;kA7Ll&aPe#omT!C1%Xr_db>(Q%`5s17jQZO!HYv=FU&t%8~X{ZN=kSxF2; z8At%pS{+-f*=Vr<>Ls>?gQ&pIL6lBMASwZ;6>5$-B_KNfM?mPgZ+(cC`!ZSSztP}2 zJ1xEnF<6N1E`(OO(WD59m~=(jEkuXqZKJB9LhITPG>sNnh`#<4F5`5At&g>vOr&#B zR6z9Ssjxo{n?RJW_%tKQQ@1K%KfDU}$3y%A(b}iszdVH-K=h&04b)n+-^%_MaK}kg z2hk?oZc|5K`@z}Y=Z>fh5r%*!2nH@==1N4jYiY6->dR9)eKg?P*afZmc0Kr=mhFeFb}ndvM5^ne+-D;2X`QN9k<9INyI z-TxJQVF1t~`WS+(P87^?(}v`BfTr*?G#9Kc2GEt)msFRrQfmTon*&86@BqC8?L9!Z z6F`UKD+-{?Rl?4{9j$t!2%x!9MAp!w-Zpz-E_ZriMdG|zwmg65i=w*d5lU+StVPR*gs zhz`&l>oryMp&dmHxaS?9Cp?)bjy~i}4$zF(0Gb}Fly@G>DuCu{ycW=JKh<8ng{CV| zLd_Vfu!oxt&~(TGdeeZ0>I)dZb^4BU#{zoHqz>vlO#fK3IRGu7Yp*S#x&JV-(@ zBVJ0x9hg&ehKWVg_$rAEBmngPF634BphYvN+t?Nk&;nKmXgVDMv;>?Mw+`1k0qF8) z0l2S^`+$CWS!2AJ5FO5Cr~4R#1@!*yxW{resz0?ReS~%k=ofOdQYTQM^-&0Zj22ly zudjoj`Bz|z!`RZl$wWFAYzCk^tipxrBCrXd`Fdp$nPr$&0(VjPou6?!fwNq$nw)&c zWbBdv(9;tOtGiME0Q>JncVw*u2k1XviC0g;_JXqyWDpL}<&rDN63=sC*TZNMKv&CA zU#&pZ2HVWI9Rq0k;sJVx9`i#*{|T{e=PX1?RzP&vFMwzz^cg@jdxf^380yQQgvK)h z3DH4xMjR$(T!gw21TCz{gXsBV9z^R51C#q&mGL?nqVt>L)~tz8%%Ex!2i(cH1Bjk} z9KUavY`qGSPpr&z7C^L1wgiEwD33$+gH?JEEq4h(21G>^LOwF1$f({dcj2>14x$v6 zho*+r#Xz+FH~G~-K(qtwU984~=x=E6L9~%T^v|a41fqREgUd6}YB7pHlpBe%#M%95W^({qi2i%D8Tgc>?t;0umBh?)8At%pssrOxJX(y0dXjD7AS&>45T(-* zh)Td|&%NJNAUfncAe4O4hv>Uial_vUboe7Xoq!P>MDKmD6!?egI5ZQJo=&@ksJgv| znuQ9ji$G8=T4W*m;Aq?x)&REF)^0M9J{k+Y1EP(_;+l0I*aV_{{k$1T)`L+A+(qG` zXUZFh7P$cbWqp_cqPM?ZM$JL{CG0cPQ(c-|!W-5#8=Y3AJQ) zBz+Fh_b+c}7RWyZ!OvFY0lLX|9-w#aE}?WhhLNsJMxp_oo*la?1&Z2K_2GaoGN%Ec z>x{$nmFdx4An9Rcrn3OhuRMW^nV98@G99WHtkMJYv}w3z1VD>u0R&5&D46B`Sfhyp zG=+PhIb?M)fSyS&awiNoel@y1D>*MTZ6g=&AW|sMp3`X@w$y=0>8d z|8G*k+-unuK=Wu9K(pQlG~Eo(xq?fz1oVS{dVpq19iabOovfa~7>iK!d?x!N{1*V4 zpPmF7Zw}~(pK2}5&qOuX+`I*#CtXQU`QVNjj6<6d9iSV`XshT$Rf^u>o_B!Wy)(Z! z+TNKQpc$_LG(A=+uXHOavs}K8zyi8#gA8g6nodUvz0X*MJ=}DFrb8Cc6VE28H85^= z`gL^20{YzX%Ia&FPFk}$04<<(=1S^Q=>LW}1H3RJM$d8?NC4<>3)WNH(V{%mm2C?L zXaTDOG@XtBS^`e%`zAhEnC0fY1i(G@y$|T`2IJv|XVBq`?6mcVh{4Wsi*_!p^c$#N zWKFsY?H15)tZ1#eqeAOXA^0p>WC6W)1-^>DgYB%fn@pthjTHd&7QF0S$H6ZFbTz#2 zK%6k|p->6jMd6$yupkO&xu4R1S(+;Vbk-SKHAHt@B0icV{$8iZji?S$GrLEEf^gY|Kf)M~MqGk}ZaiRdwS&o-> zfTnODG$X7o2GD~(EvmBMT_TfVpKdiCpev)j2k2x0bRMjm%|&p!4aG_MqiA&oMF7o> zM0tUUNdadLPhqGn{b^E>#oInXY(%W=b8PdvA?b1#lR&K+!#!>~8ok z05m^62{hgu(A^JYlIAZ(HP_s{1)%pF%d6HpHHS7MIzUfpUQW@6uW0JaJ?{X$=-Fc8 z=x@&C0L^#}py{znnO79=>K>n(jju|vfPU)HMyf2fYoUY&GFD*^HyxnqkOg%4Q;pSK zF!px(5p>4^`rjPtahRU9W^({qKxc28rk;j=Cd>=1BxVN9KmtH#o0dU6j~08NK44oo zKnqwMpy_l3&=PQ3p4LSL&@2A|;6DG+2lN}ea2>7~>dT;nE-?lR=nY+QLAoxgTUe7; zYhbs4&OWxV%7hB7dqOZET4VuztrLFpj)rZ5wVO<&^KCi+^v3$Qk#iAj0_a+E0BAa4 z-jSmcE+dris#oztoq*;mZIWeG!~oE5Jea0VqWv%If4pnz09`5-uggZetSH7ljX^p< zKbEVlynm+*?D`Fw1kk5LEmbX4HL=Z%+cAKqFCL(G=&_X>`w(q=Ee4`j&|T^pxEero z^)~#5n2-8p9BL&akPsb2AD$6;L(onL_F0h!(f|G6LG7r00|*qZgl41F#Xz*#))e*39fj4`uzzbc9z=(uy$8|40@35Tr9iac0bFt@ zg1wT8A`s<9qPzgfq=LEEvMmth5h)O5y$?~k8E)J&avGxTbq}IUse@>*LhDMsw8zq>iaz9r&-ylg zq8&s>?MW3!t2vW{DC0E{rN=6z#CwedqI}hoh3FGGOQ?ZpItnF}pRo!bxalBDhb%;M zZ%b0MU|j6Unq(zY7e4*06Sg97%W83W(T`Q zqk4ih=}&045Ka1{n%ag6trtPCB3fi2`fI0hN^ghl3u`x-Nax$Q0MXx@;55it*aV`V zzXpiX3G1dat%jkV2;+c6NOFCIk0cb~`ecAVw%N}u>hA=ylJ_qRi>5a_W%lbhu*or(r*Zs)@^!#ktS+7`?$w)Mye<{}i=hvY4 zl&T3Fa2azN0Q#NW$TQQUzr~-Qtju&40Q$?FsRC$GW<;o2QKUl;(2LIDc1Fx{MN}Sw z%1#sjy5yk}Zk9{o?a{A+IYB3?H3IP9aW#shR-wD$m=Qvf}@btwUK$0_xd z{uHhDqX?k6kti=`GpS(iwQLKZd9(|lS?>dyZid@0!&^l-%RQGNN3hf*Q|bV{`SVih z2F9oeh5n4mZh-#+K=ad+K;z8;UHfDqY5qu5bIr|L0Q$lG8Ppu7=FnzD2k3jdW>xfI z9YtSo&pSYuADKZMJ?Kmh(2UmrnjQntNfP3tW59qo{c={?6>a(GQhJJ(?ETCT~ zhaa8AQC-oR^eEabps%DgQqQ46>lP4njuu%!&t8p_XMJHCYV9Ty>3p3a0A25kN_h1h zYy#-EuL96?!n`Cch9IEp#sko^5M}`A zSxfLk{QflaXjZH=2vx z*%gYORJG=S1DMkQ&||Bjj_J{(AQ@|Arn3Ohm2Yn%fEMN3P`ziB9-vd!XB9w;Xb%Jj zoG1YFm=ReVpeg(ln(I~<1L$KdORF=T3#)v{Z4!z^-~sw0wD$lVCxFg7(p<>Px*z9! zdZN`MC<171B+A<*O)8jsE!zTU9_<2X*86~_o8kL+<131QemTJdG*jvT{lUVl>Me}1 z4@Hk-vcJQB0igNmNucrOfG#?ug*4BF0fOe5o3{Y;d(X$IhEC0)&4>=r%NnI9`p})C zSGeaLpyQ@x6Gw+TlLIv4HGrnaDy4j~d4!iQyR?9=JiC;7A5A|&2~A_H!X9oqK+_=$ z=q=gWs1q=rar(J*#{xS0sw(OlOqnr+bY~7g3+PreYp6xg7lXNimBav=fdqgqFe5>| zix#&--OaXefEKViK-1|6pe5k6gN4oA5Vi4A&*YuI_<(NwD$Z;!Mu*GTY0_H6U;*7} zcU7ggpn9J*>15h1pbK=)tWr^-_0JIe5iPQS-oF%kDjwtBh9do&Or-ODr~vf&cW^gG zMc4$;dtU;e>4bU9sY>833eV_L{V3j6gzk5WW51g618s9U$mpMIJ;?o%JC4 zO=ohmeu4WX-hg5eRhKwmKJM25qU|PNrp#pP&5-Q1GSgWA(Gi^*2}DKt z3sk>br3cZ?LuIjq3?j;jeB?!uQN6n%4os=*AWC6vXqsAG3`DEnD68h+ZL;^m-rH(C zi2jTA9z_41fH1P>$ek#&+@IIt%H|@p`Vd7R%8f*Mg{?^ibFXDvAj%_BAj*0lqI5I7 z{2)G9gy@&qJ%}=;4x$HN$fQnVjN72l)p=-_!hZood4wm>cyowmZkJb@-x<|hbMqDu z%`sZ37o3_yn-LvEH?FL!=)+u!>Tu6Hi1yr-R~%jMOb()q*FcmWtCSJF%xMU|EZ0J` z#9P_bpJig3+VS=Z zsx$Nt!u*Jp#6XmR1Q3nS7_aU^iy2VQwk;e)1%3{qbUFf22{^63#^ejL+^2H_LIr>G zA-Z7}?i)LY27j>A-HgFPbY{O&;2)aYh9V|?kai2vo|{^!Z&0CiMF?s|i!4M(uEcU1 z?O?mh+D#_X`95VpbZuu$4Ti!d5Ir~^5Tz65Ey*fjE4&Jy8q~o+lrJ?-mRGw35IwZN zrFswTKVtv)xuy=HBd%9cdtp25?EM&ogJ`}kdF2hvf55H>(-p@hycR^bCT9YpDng=nuw%cy=Z zKI-(B>5hfy^{F-0SeRb5W^({qh<2M%SzUvEDa;>QNeo08NC43%>(){k)+MQ9P=9Y* zIEV`T97O4K1fmjf+6yIc=m4Ty5ODIK3qC~GK7vU{71Y;42@S^x4x;rY;(Bm9RNrMy zdNl18qFsm9S7T72^-u^#MT;y%|JsOC1yf*~W$h*t>3knMAbMmuTw4R1K=i;EK$K3H zx2&rK?xJvsL%Ubiaabu>hK*8tW^ z!I4TRq4peb2;&YQy6!1l#A32_TS)G-GSgWA(PEu(rWg>S!Q5-v7KrkQ6o|6khbY|)Pwd;kK=j@M9z>Z^2hjnWlp2CD zW})bbO!oWuFMueI@B|uf4$WrwG<_K*^g3e|K5)}P zlnz;l_O0DUErW5b(=VVq7NW1`PE&hfI&9760JIS8IirYL0{wZI|FM!7h%%4>qM_P# z)N-^a4xh`Sh>s4U0zU^)Ivs(i1e_LEER#Sqz7Qa^=?@>G!$LTz`Zziq#ZHT_M+|l* zn(+hD4`Fzgrc>mZy;F0J^wzY4Hdu^JDe zd(hs4=s|&KmOO<8q9?E84DDlR^#Y1OlpBdJ@}-+*Fy>y%wm_6eq(GGQK1Atec$mV; zDE^$ESHy!TQ|chP_4!2E5BD;93AgW4x)_LK$ISj!(@tV_jq&pU(_vbCBa-sXeDAJud04+qj zC1g@vp|1{eLo11aC<6%~`cJvq>K?S{3-v(T!a-Ev=O9X_BM_B<)5^xx7l^hi1_>N=)9!N>I5pZz6wDm^eyd;$U<~W z101!5U@K$oCKKsT70OMBR(S=0Yy_J?blY=)D4o#qo>Y~+dGE|9 zh}N7|Ks|-}G3>vWYw95SGp_Z%3ESJwK9E5;i2hZjxLN{;ZiZbCqe&oo=KTihOH>`V z&5YYI5T!3s5KS0XS@Bte$c3e!N?3??GZ39t91!h{J_Cpzo`E|N>Z86DN@zSIkPsb2 zUmH|SJk-4*=x0S9L{lz$5ZyWo-&*6W%6J_O(PNiD-1nhaN7W(@xRY@Q5N+}-PW3U_ z`fEr|TAArAfar6r@hb)p73B@6G9p~*(1YluURdY?Gf@$hf}p$;1rWWS81EoTVLNE< zvbva==;<|O)$+E5)g!P!ZZ#f6bHWi1qPqm5NB>F|h)!&S^TF%TYB!2NlpBd>&|f4K z%)ORvfhdnifhg;Jh|`{(r^Yvq|Kou?suqk* zoc=1^u@IebHH*3vre4-;4nPahU)o}cl?_SiNtj=-k{F0GkN~24591wTXz?!8D{Tu0 zQGuU>D4mW#R02-pdd2)XKczGv)c3Lv(QE0Ylvb$Egc6#75gbJ4_P|r-g-~6_n)Gzq zEkv(RF0W>xLhD8lw2c;7h}M~apX)tg8({4w6X^$sHz!1QKGXqU{ICf`n-ZdQLd$EU zRl;F}5^nu)D+AF+W#GTORyz1OzocKP+JyFd*ngsH>L6OVT0?abwqKk*KZ9@(ZS-v& zmAE)bWkTP&2u%V}RW_%}hpG}N(yehj2BP%EgJ`(;2Q|%=|5<2euKc$$41ihcm0^7g!nERFzrY_bJ3hT!?u8f5|K=?F6bbipt2^>YH_o8`7m!>v%|tjc608qi;4g}5UW-Kpxs0bgWJ z13-VmHzF`S`f*5}wKCIL0O&#$C{l ziW`X~1kjQS=3dLT0Gda;0Gjnapy_6Kb}tn(vU0RxF#Q;Hb&COc?`be9)D$S`mv>DL>dh+N*MIX9Q^bYsD1N1XR zn~I}DoXG*2@ftwWV*t8mc>(kj2w6p5^&lFr7#@4i~f2S{o6Ao#k$yjhAh0MDlMEYkFIa@ENl~<{SbGQ z1N4@YWz|p&witGOj3xo}3OqKv7FAnqGvjs)py`VT=p8!o`Xl?YO6X5UAR#)4{&Z()@ldydpo0~85dG?J52B}D##O?>R%N`7hG?JZ`1LXk zin&zfV79ApzXlM^nTC_(Ot#(t$qp+sodpmr^Ao;~Fv}I?X{dg&N)MtHYHER~h_WCb zIZ;3CLoOL*Qym2 zh<^1!g3@!*YB`EPlpBdN&|f4K%)ORvfhdnifhg;Jh|& z&iHp2BO4UD1Cw0@{{;}`5uQNf%^~{xj;7MQQ&e-!&09e9s~J^Px>Iv#GopiN&s8Bs zA7)b2g?rvX^q(=6#nIKy96SAi zF<6L>d;n73(=x^aQ&kNY@My$WFr0Uv}T;; zt_$OI+yK}FqHPILI-&O)h;kQ&SLVR0R-Kh!_Uq2@9I3DH4x%#3W}p?(>HsaE7cw9HixqA732;~JP% z8Ly)uI%_u|dK8KusoKB+k2CH7qMIh+hGHgLtE$*DC?YeR1rY79K^EHziLyLYm95f) zXtx(}>H;%S5p{&1s}oIfSN=b1SKC39!V%CsXLT_Utv|1!`m%Ll^(O4|tj2?AU9|Tg zx=2Pl!({ite*r{!geTB=bBG?B+Dw|i7S&vH^A-?&;4;4Yoti_N z5gkNFjBlst!wH(y$NVt-6HZo{L!tO;6Kjd1SDeW~l<^vf(qonK-IPKy6Ajg{P^=#R z(p!a9Rb-+GN+=V0*1`vFI*8IC3(@Ik6IE{*2RVH%x?>?)pmi?wEKC!v*&Kitq7RpA zukt~^5a#7p5(7~N5Gdmx>+uEtOP@(kz2p*3XS%~I*6?4l48lRQ(~(jt4TDvIU00_`Ai5Y+lP0KYZ<`sn zV<1Xjq9A$^z*`agoWHZSh3H%6=X_i(Ky)))4IsMxeazLDVVJcXsx2ds5FJDtOe-iJ z>MtQUVnrTApSjMvc+z4rtjj4lF4%AtfF;(+5BcL33MhoO$i*3BX5 zWM!ta0HSwx$G0qIqM{rE)o`oyAUd{tsz6jk(;#@mi2{h;xv`>yD1{rL*=cn#5MA3Q zr^*V5o`n4;tMMQ@8tpxZzAg|g)vcM#M92PG0w+7LSJF@fqTEPqj{YL4VD7bS3q*ND z3Pf4&LzHfYhjfiR3VUao2T`WfL3B?`yh#mXOh?ggGuiLpzW|~+k zl~TKHX@O|VIu?r6<16OOtVW>eSd`EQj8*u+O$SjrWFeY)N^!Lq#t)qSW4dD@+H-wA zwF9OD)@%+y3(+E_8mO(%pMm+JmBc`lfdml!?4Am0H(C^e&&5&1M+Z@XpMxl!jzClb zPCIrDA1pvr<5yGi2P(HW6YbLr^Por2;p6PI?k9-BLbORa{1BUf>gm>`ThMMHy0>X* zyks1i(kmd?5G}G0J#YoLzkUJRx7Kblksf|6M2Ln**aV_;o(4qegq~p_%3TzG zWIQHUgy;(Te|k<7M895w+Za(_7bUcxJ5HK9h`w`Avg!oeJQ&p!xE%vg`Vs}vZZ*sM5Ph?L3`ASk14J{S&j6y=X5#nvAE>&< zp)NB53DH4xA8!4jhdLh`C!vVQgXrEH9z@5k!7Q@ART;0NAzI}2`k0(RF_@~{%ytd# z*8rk#?gky1Z2b}>uUVPtEP&{dp18>X5EbQWsMcGh2houU=E}c_jzaLA69o|6GPtaR zs0fj_Y$(RN=qvy88{k=qR)tk**eh6#2hr{D+k@ytfoShnY6wK1t%`dZ2BTFvia?Yb ziR;l{WH9Dl%eFw2N2EZM^*%)DX880Exc7lG(Kj1-5M@doL~DIgPrZ*ZzD3cuGugl5 zzW|~+J)^P)(1<^Z%1t$(1j8UlSP%xP8<15pMNK=k>=_@YOPo>2F- zEgVDzeh#8^Is#D%IPHyV_|*Z3wrB(hWz671w9OTqll%}JZeXWZ7=wjqnKEs_KU9Bf zO*-yV*eyio_QMiGsL=Wn1PXmidn2+C&Ds|?vW8(RX6+^u=`)wMB1G@ajqmb0un9!x zjsisKguZGZ%3Ty5QVXAHLi7@J$!F(ALA3h*YHB#@N3;K@TvG?p4{@*SYp}iL>|+?D zgXo6?;lPq4wE=cLktTuY$*$SdUQ`{n&5YYI5T!33M8iYY7dIEJIybf#t==>?HoS zzZb1u#dyI*t31&77p>ljzG$_hO)0e&6WYi{t9nf>pfLn5TAfAn)VA<90QCMZ;*_3> z`Z*|}51AcFp9A!k$@mUM^m;7>pIDIx=%OLaYw*RS%VXh!#Cvev{Df7Rj6?&vd@-CP zjBA2NUr|B_IN+bmX#nV9zv1@{)1xauQq#&zX91vZuMHUhEy}J?-D8y=pj%zVQy@51 zAfk~FjCP^`&^@jvI6zZ4ADU%W7cYuOe+^Jo`9v)%_Z-3{UDF+*b1kfqXEpV&H-&h(__o3+^lu%*DD(vB= z12i47fIhgfrg|C1*PXsB-LZh4b17N92h&H^Yz{yR=-m(GP$|&whxr>Ti2*bN2>|`@ z@B*qD-Ga~YDB`07w1CwCnodUmEdi&UPj4@P{uvFEyJhqNJ!J=Ojk*){_p;NE7=s0L zz6H2>Zz!rqS(DyEy9M;-y!q5Fx}>K=FfUqU0X=;b`d_{n{`1PUCt4f1sJU ztKe?~=xoWk6mx(+-pT?RLlDptS_07Lkgx#ItG_}j ztE0XlO6XB$N7Cm2U35l0iC*6gK@ThP06i?u19aT$t*{2BRhf)L1Nv-JoLyZ6#R{q> zaln<#X#nWY58>OE>Ct;3Ic#O7vjET&s|kBUqWlZ0f34C3bgou7a|}R>s1Wi|97RU; zE{L5-FXRACVN+;2SX~UDU!GP@Jq$oU1p6SX@cf7BXJ@6i==|N*Rm~u=Fu*IX1xz+x*7hoJML*CppUoq0L_#-K-a&cyt;rflAzF^ zGuf5!UjS%+dJ<^7IiMf^y@fRI71dmG^A>LH(IN}zZ?55;Aa}vm+uBVg(px6wB%pWa#dpUj*aXm9 zM*z@tLT@#I{v2L~hs}!2plh{*{{`kp0lLbPeCi{#-@^VcbH_=-Wq^qtW7fL759zEpj&t;!hyI=zDhG=?CcE4BxqUq%=Kpbxaf@0YL9?MakS z6(&~F=K%fFzjY;g{SO3qaNCGHKyT0F0eWglJhoNLs!T?r0e!JHxY`1W&Q!JMfJ2zm z0MM@-$DEz%(Ssm)!pcl%0iZM71{r41qMQNMY^(GD{ajvr`vcG-+6ciGCz|9gwpDA} z%mJFhAE7yKbuoZGj0+6kwJ5AIBe&U6Bmxi6@1VU0=zbFrM)vW0%#w*`pH0PY1GMUe zB7o*bVgUiPq=LEEvMqq-(Jp{yy$@)*86I~GyeFX3J9>a-KMv5}t}3ObV2n>t^h756 z3;Y)VnxCEo8gCBhuBBKo99;fm=QPOugD_E}AY!3BAr(g+1JKfTlwh&`tI_<$Z)765%69nN8=#Wy1c3+VaF@Y{A3 zsz0$NU6FPRXjP$rs)`D&k3jH4w8#Q_LiI|x_5)i6j4l0}Or+=DQjLIaun*^elVB4- zzcw6zrW3mCQjDY$xQoJ*24SMaQyfR=e~|@IfIe^}PIW_lANKztcbqhJfWCWwLG=V| z&pZ1j?kET7)rV`V6bv>8cD;ip0rc7(71eT7t+&mL+cAKqFCL(G=+oyi`VifEM+`)V z-VTVSqR#-L#j_wo1yNrbC3Kt-NQe%i4Yp(w4|PKbnp=?v(frt>*j45~ovJGwun_la0MTn7;N(7&tv`U|V=FVA1rXgb08If=Q67Wpd#m&y zy0D|%8e_j~#f$3=VHi|%$8;Ml_rTjdskwA2HR}01J@fZKht$sw) zizuOHj8*u+O$SjrWFb24V4}){_C-;Qz9ZeS5N*GxwWeT>0E^x*TjtLJZ2XFxMC=_9mTh?YpqrB0wi>m>FYEwT_Dx}zPg{lM1B z+D#_XL)+!yhKh%AxuQ300@0W52Sn+Fo?sx#T@*gv5Uc4CqDgmQoRSNpAiA$roSKdH z@38+~u4w>Kygc_~*mgPlKnCF;TJGg`Y6=GX0d_r%CV}YmPb;a5sJdaB8Mk8~N?)QN z+Hqt>|6Fe7yDdZ`=W^E|g49RRX8_UhS+V{~57a-55*p74Bt!?%U#eFU5A_QWyktcl zLBkzNKmr2i(cH1BhN6pAX+`u~CWo&H5lRi$xZ z^(ySITa5?Ns%Y;)w39$|^QB4x(MH2@q4`_1`UOQG%8kSU=r58A=3dLTK$J(MK$P`9 zMCoR@d1w5JAVf2F_aMrYI*6VtSw$7XVbBUi|Hx!_$A1Atd4wm>cyoxp)xU-`pBU9# zbMqDu&0kch^-j&9&4>=7Ef=R0FNjgXl+1&_5t5q6rX8aiRdC7heh%*rsq5 zG@n>q3`9F+EU4aYR9GE`{kYY55FLT`9z>f7L{C3#Ao}xXxPPJ;_DW?Gfhac;Polp_ zDwumM+X7J@kpfZH`w*p@;i3E68i=;<;X#xsbr8M1cmvfFW4w%_Co|b|@m~N@9^nZz z-W;O8#3xGg15wR2H*W#aZof5AH=LS7n-LvE@2ZeX(T9TYSx@JlcMvW6PjPXyhBG;c zGF}5wdaP2C%izS-_|$9-dRiz}kH72nwrVh%rlW*rF;?LNHyuRjkcDWu1{u{H7?(Kx zV!C4?`qj&=)yFXHvSxDtT8J)fl}Rmw{v^!5SV;^-8At%pNtikFP>NCoL}`i?kzN!*swgVI=bYWScW;F6=lA~O&THn(GiT;JJ9l^b zXdHaXv#z$i8U5-;bJkXl?q}V7KqcZbmypDucKT&oJnq7nOJXeWjqUxT9Rt8$=70$pwqhyATVN1(-@PhHWL+ zgAW@BmG`ysbifG`DnExLxFrKUX*3B7mCOJxDL$8EQCadO2bGgD9JTVBo+e#@u7bHJ z_z|Wu7CsM^XI=%B9QkA__ok5~Z{?L-VOz-+_!U;|p;k5vE>7W~!VMBCbx4Bu zG0>Anldw?93{ZI2otRMJKX{KW4k}C4a8UVXu|%o>U2X8weqt(LfzLx_?nqF{kx!;_ zf}f8hpXZfa8K@Lv*j92a zON$dKNB^PIGABr=oC`_tN(Oq;Xc87GnE?v#^(U4GaIK{74l4IlcTkzOG>LXW*YEgg z_nAtcZcZu_#(+wWd@_{@w}K^k6(sT`*EskTDqB8ErOs|NXKm%^rt(%|6Re`g>8A{O z)e=DxY3-}%F#__^jjIkW%!z-@}oqaFZfqQLZqM<-BGSP zxs`;JT5+cGB77byzn=&yIr7O=eta;$Bo9Xhp5z(_pF(9oRt#mj(VVrFqnpa-8rCK* zLwg35^rAAZX&EW#5OcG!J5-vR6e=DaNp^klI6*?? zK1hNK36<_NnuLW)W&pQ(er?aLCiHSpIlHQZ%1WnkLL_uWzJh|6VJZ{h^HAA%GN|Op zCsR4%a;PNl=#^aK;8Un9isfCC-Du9*%F#{bx4Q#~%dnI|@m^HM)T}85-R+KY4P75B zRF)`NgRVm?RF*6cD#aMK6(@QAgvKYRJUUXNI8fmR36-Ic1RrCdCygdyp^_P(@Mrd7 zZj1lmgL*rt{G*D4$_FRQQ+4R-hM$(qR1Sg9LuIomppqk>Oy!x+GbQ;lujCpBpF-tt zlTzuV8_ii;Il8Iz>6@Rp4EH!H)r-m!r|L*S3&I>L$`yOOv{2c$SujKg8%F#{bhIVPh zWvIoVOfM>j=2nz~wsA+fTCYwKDtEnBlSV)+RDM*Gdq)>e*gD%h=oeNZQd#k+e)tK_nHWm(O=<;b5P+136*z|%(}YXlg7HbUZ`XSD16@q zZ2n{_U+C|klJ94BQ~6c#Y>I&PM)+y1naVcsd1@u!tQ^3RPo}c?!azwr(JQ&e!KYAJ za78kG?nZOgR*r5en@ulHT!!yCs;w85C!h0?g8uG~a%FErM+Uu}*j^2=-ppo+?}fIE zVOz=78@q1#5B^Dwrqs*{5-Kwwu`ZeRq_HlU7Aobw@9@-a@fMW}2RNwYTUy;z&Kz5d zxh03U!O-V;_tjsuTD@Qk#gD< z&?yO9^iB=(pQ;7(&1Y^Z|4eR3e?yn+RTO*(Q<)B*hf2P)EPx}QOy!a~fs(wJS8|Pm zPoeVY+F~@z6V!i&ml*^Q;3N8C}arr%~ttvt7^I6Z(^sN@^h zEMwSKa$SrI6e`Oeh@d1;;RXqnB_Xk{Z}X(Fu5S}6nE?t9`^Mgw_|6~)m3;q^o5~y6 zfz%MX2I8lUVk$?&=b@5sE(+ktCsX-XqYz2H*(7L=l7H=47ya&%MKa6~F`8JaR^rWchv&csPUySbxW zQMF=)%CB&n-9(6mO1^!?Dh=C8t}{i83zgf#gJ_QvBvfvI#JbAGlg7HrMW~cpb;57s z+D4(W#83y7d>4Y7%DBQo^b>Ri4ne^`WGW-!^H5m;2XF^)B@6d+WdE*3y1AQme54hG8@ zwv}8T&&Es*zTk=b{HYMAaD#+Og2cMe!IQ?i&_SqV1}J>|iX@B5F2fvD@`>nfD&JaK zgUUfyYy7mOOl3FtJXG>o<^de}WGY|llqAV#dnMO6_!KH{{8*RvxzU`pm7|-=!det@ z87^?tDlaOl^~^5?{l^{U>UR)LMl^={^opQ3SSc*=N$i#}Y%94I<5+yAvPzd?)Yb_S zDw{xJ9qaB%V;$=*R5Ak;-tJQ@n&ChA>en4q^0~-vD!b;-paIbJ5q{bRrgANO9xC}n z;{c9)GL-}WH6{5aujCpBpF(BRzjO*kajeWZYb!@Ll|wp26PKX^gTC;h(hr-NxGas_ zQLfJ_6%{IdHFy}k&cW8m(EAwa*bK2y$tR~; zrD0piHSt)gP}wayi*7qXLgjf#tYcU`X*4;8Rj8B`al^+Hx0~n7j$IOn zt1hws7eDO~Q`rbU50!iZVE{)ynaWAGs!8&pUdc5MK84E5K5_J^8_ii;Il8HQc|aO* z8Fq2hF)u2^Dke!m&$^>rF?+2)_{mRF$QM=$m3)4pWenR&uCbq>#e_fj=u!o#h7%-I zCO~2x#^_079mXhB%DI@~_bXs`I@ii~-*iyPrYA0HR!Gz%~@MHx~a@5T8_93AtiYaDzE zmG4drrwlimv$k?{Q+aJsEO8lLWYDi(RK`rLF9jXwjzT-fYN9_#El;x{7ApBHC95=S zE4lV#_J*l6?^U6rPLNQ!6B6rSCQlmcU?!nb&d3UH|CK$8ZociHl20~qQ`y3=IQ;@$ zMMt9Gx0y;8d>$(KOp*YOd@_|UFRU-g+j=F}IQSGQ)9=-zac(qcZRO~pa%TbJGAv@y zA6`_pUzZ{U-QkXMbu4VnIQo4XNS7cMD)~ee%NVv5L%;j28OOWv%_s^~xIsc?FeKLT zES@yh@hn26oRkw@CduAJ+W#E~m3#()o67TF7pAJv)fqqSAEt63d>$(K^nd`4d@_~8 zMn_BXC0@xj4nBp-P%H&H?nZOgR*r5en>5ym%W#XMe0Mvi$p*EnCSH74EQY$yLHmJK3BviftiFJ^ICyjNGf>0@ET7+MVu>0WK zMmebD-NJ4vJ1mQ%H=%0Z-pOCXg* zajeWZYb!@Lm7@;$5SO7kg9>_48Q;^8g0^-?xz?V?3>ey#1KO6Lp%4p|yi3{A&$eQk z?5@?WT-!36);U2!x*&F2Y+__$2h3ueSK~!pY3MSLFjsj zpBBzk299=8$vg4_IP%F3`$R|@- z_FkMMALEr=e*gD$C+J94^CQjxxQdOvRyNT$W$mQ7&xmz5f38nlh6n~#_M&pf=zLPp6nB)%6>Cv>s$Oks3$akidkZZ6Y%95Ld{98% zmDNA4Kr@^mq4FI_tc?nuG}cB1p^_P(@E;;@Py|zXeY}H8Ug_(ma!q1MS`J+&@zc_n z%3tC0P|0g_132=@RQm0xC&{CbfhW1f!KYB^lUkITxY3-om7|-==l0bhE<;xa)$*e9 z*trlX=v(e6*Q^C)rB+_~F`kw{EL8FeW2-c5E4eaoO9eNEewba4es+R{%I_etmMeSG zSj&}#N@jp6j30zEpW1ILPmkGowBi=SzK|y4@EKA5-+@J|-sqv^H@fO7v z3c}kGWrHP|ecP5|EeF@d67S$2pdfN`0ZYidIK;(TekvYCyj%Qz1>qgyve%u=?`KqD zEwy~g5g*jDMM30%8kUgx`3G67S~4$HBt8Lmf`ar+0fZ%FmMb00TIxNlPkh?yN(JE) zR^_Z%GG}axWi4H?l#mZ2KcXOV2s=y2oYuQ2YdJo*Eb+nD_Y{N=yOv|S$=tuPE^G0- zT!Z*th@w+%qtC|!!V)suH!j0k3iS*lzJs8;g7Ccqa)Sby70;xyma_}15ns`hqab{> zjm8o(uTQGUS_)jsAifUfO$FgAVB|6zGN+zNVJ&464C0&E7AnZ27(iG;rt7aT*0LqL z0`aY1dliIl`jR`#$gJEvoVENqEQR<|%m$kB>06k!ybzjBd_QrLg7Ce*8cWE0Z+>moa_x>ud>3v@1>yT|mV)qAdQw@IXPHy9Y=cf&7D`Dxqc6I|k*hkD}t!ZC-}{4hT!g zEd8_8-M}IhX#xZ_6lB9qKt2%2^gHs5=9@U5dmS$QP>_?q0KyV7^-tx^&}>*V9fDw_ zg7m!($XS78XeO7X+^lf=b1mNB3Nqs|AS@v>F0CqC)bf}g#Ujnu3i91WKq`rro-O=Y z%Zlg9Q##W8svviN281PKPR^BoK&xSKlnX%!+Bepo*t38P6i8BgU)J&|J{(i9(K=Z{ zR`P8kEFp8<4L>f+>ata769msHNd1$5d@Ycr%Y@z8L?ndATNIp2usM!Oip4g z$8MSwgfw#$gpLCeDO#pAEzV_W?N@?Qk!G8Mls*gyOUP{WpgwE)@Lomg0Kqv0`2q{m zw7vrAx&pV&f1uBP977XUp-u3fZS>Cv0AUH4YbMM8=>7+hv<`x@3Q}S}AiD)}Ax>v4 z6Q@Mdp_Nh8SV1!O0KyV7^WP}RWqJ64FWraW6$Kfx3y^%GrB!%s)^hNRNGb@(6b1Qk z2OumVGi`qwYw56uC>4UW3i9JNKpF{Tb4_V8-0Y&!Gb^I#dj3m_~ZbL<~YSxe0D zFd72ELj{?>8IbV;d2M4cE=!O3c<(O9q2KS@ylh$z2usL3gcI!9BHuc62!ad+xw96K zvjVA9AwO%`vNVgXE{md03Q}YhAS^-SWR<*M{@IpF1(D_*1zCZ{w-zH>TCOQf<2vX+ z7H&YiXYdn1q~Y#p;=EcSksc>S9LRDrT#u5tI*~pxO%J_>!0#gIG)GV4b?$4WV2GaeOB9?(zUiy;PER z{}w`{6~yrk82@<~Yx&(3K_jk(&{73)yx@a(6=O(D_bfVlA%wnBknx_kckgpj-2D>_ zn*CD<-BJ+8t2rUEAZvN}s3|@BBR22Ov5j`TcYpXOjUih76e@TugsLitPl(m48s7h))XI7R`!IF?P9(8<^OmBrdn@nr=fY%D2bQu;|X6MlzYX`i8pk^k#*Uv=3nUaJzAolMok4=0-hiYYF;QMe|A!ZDk0G%?_D^ z%>Y)jZ)_kfu3C~pKCsnzAUS>Z7h*M=+ayyZ6}<@i+O&g6AP$-lyF^Xgm4=x5Es2u> zY&9N8&h=3RSk3gdu{1Md?R0z2}=T z^7JQ*!vC{Qe7hR!AYDF%{|zcc>o}9;*CNQfPLSa@untxF!bgAR?{JD55J>k_VZHX{ z%gESuMA;)_e0YR4C}#8@lpqSmJXeQOh6Pg9c{VFvr=y61@`$oWDftK=Yk<;tu}Do( z@WO;}I`>{6y`~g+olzwUda?pJ#7a&>vj&TezcUJpf>~>WD0*HXZBh!nP6`wS%UOXO zB`D`zS_Ah+_41YpaRF!@Vv)i_rNHYfSW)mBE0DutJX39dseRt8eq ze47KW)4oMPBBJbBEk5qvn$R*zx2q}&lIrE7mp2B|P^G}@J_k|IhZV?$4{~#eHMwXc z|7|_S&<+J?nY6o zpcbOZ0j zEnI>Uh}WoBqT%jRdsS-pL7usC$76?(m+ZRMZ8ARFJ9NuOH^>}N9@j6C+ z_&*c+RU`W5Z~*mF3V1t;DB!ImqM$1)_!%D>>;_~7;x)ea_Yno(O~NgkrvqrMQovh- zL;-K>5e1*H0&Dxv7Ym?3yhi-ZbWt#DB(8D46hL>C0^X)23V4f^D7eN7tgTqjvjXv& zm9EC{gW54EgYsd+Ly<)`1+Vj`l^s0Gc^J0H)yq|`LC*k^rXczj7Kk9mEntF2{ho30Qy2H;7yRCfHyFT zg4L|R+UPii6^Pdu8f!hv5<}Bz;^P4NTPffTsiJ^4lZt}dtialI8qW&EYrJ|aQ%s2U zX+n$hndQrMiI*?YYB)qj-vgB>qlI3E~(%QbA%UO!o9C|OBvplt= zuw;2brT5CxNAr(Dmh6x*b+ay6)@fUo9wh`)sU^0RyfI(08s6)MwaZ5(X@p1T2K8pmW--_^u9{M$0|scd?bQoIf1jZ zj!t-wvlMUGz}l(QrfWb@Q^3rk7-T_ZjO=94^EG2=uTs{F@$8WC;^-CqI0cUz;p~uctZp`?KlafpepQFsSF&mSvw<`W=>^TlaEOt7goYT|7g6?68@1=5 z&~nO#@xr!vQ8423WIEI?khUrXd?bn};Nwh0!5UUz9eZ--Lnsih(XC>jC|LJOQM%tP zkp58$_;?yoz(>o7f;+6hI)Y{}D-f@_d<4c7bM<^b*Ps?J#!$?swgsm!>8i0q#_T}c z6+2ziw=}3KPrtFk$Q12X2FhW=%8Qt9p%xm%3L-KVz`h76{*cM}v3fqeHu|q~<+*r&OPyV=)=-aL^ zoUvHE84|lSE=L-g=U&7BXqF!NsYxt2uTpS8OK8Uy?-|#OEzU>|qLvW&eP&x6kCihT zJ7f%dv5;l)?z&=e1%;8tA;Ao^EIv3tMJ!%e6N9a2_qTBZ^IDwyNiojY=hGy5F(-!J zP}t(Ej~%?4#^Emniys`RC;C<^3}-ABe+G%&eUMIs`A|cZ&C<8Ds6;F{rc!V~TY-Eq zky#FNow@q5UFGQ*1P>G>49l)GcF5?MiN79d^&h{bi^V0E*{sOQqWi4NvbcyVL@a*p zx!TfEY3c;#wYYJF{GHw&)__j6kD)#at8vEcpjNkPBoB*T9(U=v_(kDeW zB^I1gDLA0{cVbdneI{*~kwdFfA_V^^hzn~%HFn6jF}b=}?7Og}RD-fBY*u91V-N!^ zi{ET*Rs7>M3P^9Ul@plP;(iCEIi2}gWvbdahF(+HqO6Y{GQLU8FBZ>Fh!cHF6^1hw zix)y-cRJ-JDswbNu*JQtlzENyU&cYV~(#!$jXCh2f0F;xUle-FoTVn;-VV zxLKASvLS+4uv(?yfHnd7kl8Osz9VUvu3ZDcF$KwotqdAFWDJ>v(L4N)r^cj-#eXY| ztk%EAx-5&=?Lzwj&v+izfbjr=Yqd=c1M^z^{2dvu3ym*9o4dwPQ-!_9`q%;0x#%0g z;ver66@4QVhV_ZXLm;uoA!Ibd{NW~Me6#d7K8hw5EKn&ppcUxG7LWMQWU5EDE<=?d z*rOotGK3v64z|SK2^OynHpJqa3L{%R1~AaFxPPT4V(}Z@{AI)=VvTJv1M^z^={NFs zs``9QO6wj&845ds^|6E3`Ti)e_(pHswu=#zehS0-#NwWi*aIgsIf?mNC|V#{`XMyj zSujJT;D9y>`H(s|;m|XquW@YK~mt&}+!Ul52?2z$X6TCq1j6eM) zO!Va{4C@n%b0D!tdStxE+`j|wLo~}qq!SB9s}vm2LVB>pHSPs+#-ChhLRBDGsvs8` z!Va2cw+o2HA7f^NhuDuQjBGO7!$8a8pgrjCz~VP>f)0-K&&CU_l#|{~@^5OV1SiI-O4AIwGVK`&4xH%;D(3FfynE}(Qu*Kyo)F&1URw+23 zO+h}ec#3?FdztaH0fM;-GJ;1-*dZhKfWKJ$?Z8`kGn^#Daz@1qZZW zUSV@;oB8-3L|@0eHdt&Q|lG<>tW9B z165^6?n@^yuQ}xhmlSg@_y-C2t`jWg{050VY$qdjX70TvY5U=zRrt$uGd zr{6zVDv_)I-YSrKLXf5)TN%O*8A~tM7jxb@SWTWzXN8en&GQ&&nKP?oq?pqo+ZrI8 z;soX~$IQXH5M=Diszv?##L!xW{fzaoLxyWN^r1fPtzSt794{*jXDk-~42eC?C_{{9 zpMT^ZSY&7+VuAlg+hPuAdyx;B3om22ORk=>s2tsZAYMU^Vzg0Xhm69xRyQFkAir4r ztis5a`1K65EH3d2hPdzmW8d?W5zV)pz`PdkoGdk`|Bwv2+c$<5DeRZ5j~y}|Jt`s= z*X~tD^qo)`&R8ry0*O5^Dr2JNm3+my%6@k(mRRtYO2GlGVL!H*67eU^)y+4us22oL zn`}}NG1929Lq^!&<>VO;uceE{4HQPsQ+SM{6l{yP|7*2~=amVUA=m*=K&GxM#+ z-}pTic!?7%=6nQ+Jxptj(CR7j(Re8@4rQm?Jpg)_lrvlmzx%!n!2Bi*&p&JTv zfFbOV(fMc89DFpkzm3B?@CpszY?~}6ZOmn$^#Gq~j8Q5)z;&_tWt_II6PVYWAtBN7 zbQWQJi3fFGb%Mp5K9JaBxiXAvZm)-0o~0jLS({ieOQqm|_R2suC+kINZmgZ>L(?Gm zLP6GHL|0=6Bap3Xi8&|gM~gX^6h_Wqd5?jXIW-c{bbvXjE7N7vxbPPBkH@HS9&=2e z&RE}s3B|=(H2c*Us-mzTvp#mn=w1omRQv-c?8Djf7=`YpFr2Yi+zAqU@L9&4%{Nct z%bKOXyvvtZFkYqLfc6{mA#+ZBndR;OMjVB`7DFo(Bm?8l8arfcNh~H7zv5#}GykA4 za)3_^11*d9HpDZ5#an)gl3{M&&uxntnAhT0`$x0I<7!1wi9s<`R$&uZA3J1xbvQyS zj-6CV^u4GstWPX%1&KXcE(7J}h1reR;(`O8{EM7DpvuUun#fQ~oT>icjn8#vcczf#s zeu)*gJcwWJ3!5I+D(1vNVvpU+u)SFUtEOPi)Y2iuf+i{j2ekcz*_;;!VBKM^exY(S z-GZREf_zj35O&CjnTiQBm@~9jsF*WTVdPw}kqorVDU*uM1I)QbHDs1xw-cDh95d(! zI>WhoZ;Vwu7!pHQ6?QV~V~30qtua@KD!cJgrsylN)wY;37K;r??8ye1XE6W0Rg^96 zI;$wLpr%T}0WE7NTfF%pR?g+>_uIu$UkExXNIXt|)7T;7^Humh!{SGqi;BhL6h;n} z3ud5Y@y6{K8G*&w4I^Y)W1|z8*Wy3^#8T;8y+{UD7Y>V|lL}j$^|3=n?_BFMY5tZa z`U-5bE#{2H;(U;(gX%bm`9@WI^|SQZ=PMG=lq9JX9MBFTA2K(k%iE#JabHS*J%(B; z$RUQXgV8h${jkN&(u;}3!xTnNBwWQn%i?O4tqx~_&{Ua0`P2!_Yw>VR-iPnegE8Oy2TT0$TEh8iEQsB1bHoAd zedI&t*4bDRlB>r~txY|K$557nWWeL^AN6w#UQY0mxwIJ;HRtN>r!=Ig01i;VZUW3b z3m|*R+%i>ObpLLTq>TW6pny{axb!Cg*-K_!3}8TG?$#hW1K0@!3PA7_ncWMq_;*{1Lvwy6xvOZL=lvc%=$)d0FUB8I9d>_gVa z4jJd$Sk35hV|_(mPlaK9LUtEO>`5z`uQEq$kiKqeeIgc2Rw+23rDCB7nT@ls;w4xA zA+jDdfn}Q%WEnjE{`sze8upU;Q&p_bM?3NNCe#nWa|(D|fU_TG5WIdoBZ36He1sp8E&#wW3lKE9P^i^{8Ax91B58&4dxK)6k z-U5)l=ucLVnln6KWts)xZ3Vn6z?(M#WG|VI#>(jA>lH)jmC-R&VyA6$2t5A&)$Rkx zUNQ^urWQ=QT@9jX0A?y+ssM-I1CYIBc3)AOfxlL*N1FkBMFBesaQ9sR*-PegH|06U zlrKYP0i2?M*Kn}1#ts>0hoY&5m*?OhEXzR?b+f|ol&p4uf!1?2N?|+^pNmwiG2^-0 z%T8e4=lsJ=d25e&T!SuSwW0qm+vXowA3J1x_j!^$=dpzYL|?kXaK`eSt3qN=0b8@b zKIt+1ocru4PAq7zQgA>^8_Uo6yCpJOm$4Z;O2);|7zG&ykH3G+-=K!Q*jKVZ+7A1p z8`1y(S1aHm0j~WEK=xvC;|m!n$=9hO%?9wK0v;0J!#@FJFA5f)$xmg_z5p6EK8AdD z+crNCU|qs)NcNI>B1@WoEwCbD8G!K$SP~w8|FMsuk-cONOvc&}EDGpTk&XcPtO7O? z;J5z*$X+sjuTqYhYnmQksHQx`F$#H!o-!|P{{2QZ*6k1bdrpj@&s2&(aeTSP4yMk( z#B>{~Mzu{9MC&nykyFu6GtjCUv!1nH^>^FllR0+UW1G&vyj8<@qRiQx%pXPlC&f@H zg}uu9*uk}c6|hVIuYmq9R}y{CDGckAs?h=xd#c`=rQgsM6)sEP*SadPV2Db=0c|hx z!LEY@`7`(hCsFk&F|d+|4`VO5Pc3sjPaxIiJj zs)lp@f?3T^J|nNhrO>r$82?o%K1~Og?2ys_vkFo*&Y9NG(oKbty9dTI(5f1GEdcS> z$ZT6q)?O0-5?L_&?c0g87{E>n*jj*hivh@945eKt#aaLQoF5$o zaJ&Kz7hoC#*^6(n=S0=$zqOpKSxIpM^Hz;(&&yw}L0UNNm>)wg zDr|k$#||04bi&vX-Wqj&NDzG!6^1jGsxcZ8d+Cd;dNGUj(D|+L&xVS`f>kO72ehz{ zxN2m6SdZB>!?Wnj$1(Jyf_wvyzyJLhu*qKReO-!eH(2ELPbm2;z?=DiZT5WujwuNs zd&!KLCo@?Ga+**%0Lv?2G(7(PhoS*wFBubq>rr}~uinSEv@B`vAxL|5d(eN^%kV09 ztsRlQUdHT+mNZJ#^A&&UwlIb!sf@nLfOYJUQR5pd@595ooKi^M4jUClZnjy-K>HP+ z+&~`Iox_Q;_U57!nD=3Q{-V^9eUlne??o|`@1RY^D%Qsi87=?8zzQB#CCn%CTAnn8 z;f&>BRffb~!XvA9%-D_9`C(PZLIf7HQz&EH^TL-|x>B7wS@)SkO$R;DDC7ge|Uf zNG#qLT7_PQpr3+-@d_Mv$cXy0saQPxUO};Vj>5>DTK8F(W%26zRpc3$i7PCtoAx<@ zc`aVvQ{FN`k&)E*lNkC%VKvT}9ds2Bp}UP|JPj=~UM&@M$hMd>7K@8QVlR@i)>fEArg#}4{i3cIsHS@ERxF_1*OIY##q zoW%E9G%Zm{8Wpn3Fqf0O5sw0EUEyltHSZV3G*y-!J2-?`u-{77R|49a#8^;YBPN+o zF<4clF>zjx0g){VMGx7h7)S{?q=~~pGS8ZEoN8IkI>gyk{nk!3^0=K!3c(>w91dbr zU>Ecivve9zpMFe=r5BJ&rqn-u4~H{eu7u(kI~M!M@dzi|Mo4)+RGPiYBdM4!t4!{K zX{QqSksYa%E=oEqY%ImQppdb_$PIyX|Fa-#8QQlN-JcRgO%&vXHh{2%%;Q<2 z<->2W9~)_UE6DU007(=rXS&Kx>cw5MC>3emRgg{30>Tmu1OJ9coun5Jj-(C{tXGgb zEdl8(kfZNpa9M`*Z$v{TN6`rdDbgGemXP`Q5XM-N^zR2&r;j1{OF^E=24t;30^bZJ zY@Ev6i<8qjWB=P>d%KCl=)PlQ8{=R7P0Gx{{x=12SoGbBa;&B2Ba;?9kNs~7;wWyp z`m#gogQmDT8b*v(5KmchI#&y%m`?iG(8{zJ(egy))C@Um!!>>?PSU|vspi4e={uEt z$6u&6Ym4GPB966Hsz*pU{>WT@SeI^ej-@+FnHxv0Zb!?AvH|a;Q{)lbXsNlcKEmQq zaV%vGw@CB%xdRR8m)uyY1+bLNjl)$Kb#dS!O7_#&P3UE%?9M>k0m?Plax7(EWXL?% zru7kY;H6lasFb;JI9%;EZj6Ak2KnmIMy2e%$xy~M*>WspZI@vvDOW$gBc2L&jir-H znHxv0rG;>i6SflF*bqtoDrMfBk*oH5-~ymdRT@yKqfg`lS0nAiQKJB@O9-Pz3gm4g zx%_8T6wu{wi&HNJ@-92&Oq+qnJVM`FJB;3CjQemEIgQ0SS7l{sEF$?;)5ql&p%IuT z+76(wUMIdNJh*%@XmAnIYQG+V7sM~?uz(QY?i;F7Sj`anRe_Hr{ukW6FIpD>uWL}6 zKB*c)#g5sU11kO(ytW3GhGCumwIc?tdnjqnI`DTa;ZkC95jn=ax z*4B#G__%4XD9Gp(hJJMb%~1-bhuaE{XW9y;u!73`veNFc0`VGC5-^g55@htQMoqQ` z&}pS$e}t`IeLY*jVOG$(5-W)71O?(XJ}YDG((010G%eZ_K#|97drOzJ6@1v(ULIT+ zQNj5!m4UJ3iBKS3W5g$zCP4}Q?SUOi^#UnJDJX_>?PPuNHXP9|pU#%7U`0hbz z9MA&Z;SbH&duY0$kh2~qQN02|xR(L2@<p-KD!fBkJ~JAHI1o61g(r4+WE`D}WIw7?>Nuo>gxYrzb(T3{J_>q!f& z!s1x5v#}GHrv+y0oZC>ES7C*+rM|zez+B&jl-D4!tGU#0bHFLQth4m04@wgY=BpGO z(E3baQi3~557%dQ0DTR?P6c_p0wC<*Xrp#$mZF1sE}3oua#=wtuoiYObW{%0^RW2T z8f)k%x| z9*fP16B6XJ(>a)qA>$SbtMMc#J7g?gV>Rz<{9#Q|4^bF?>1b!UgjU9SC2NYhP_J}( zJ1ubn^JZM&BODi!t4}s39No`s;Uf-J+UPGbjSK<`+Kh*#bz zFMp?=3d3(nZ4v`5i&Hyc0Tih2keDE^(^*bn9*d18e_;&`%-LO`oPZBH!D7xHNbFax zylu_V*hmg@UidkhSa4IN;D9!62Ai|?6Lea0^-AesvwH81!Q%yELnoEebV z9{}s!|AH<9W#=Cx5ev4d6dcely~pO9bIC%<^xSGRt};Vm523ya*$kNqW^si8S|Q1Qk1D zbMl~mKJ;rL!5Y*Cs7y)vm^n=;x=$3PE69q4xTZP@5)7Jz$5Q+qU;Vp9DU=_#8FmLo zPEF)MP%ShFD~1%n#!HlyjJ222DQCH+UwEZHv1FnGbHK&~pGc+nA%1$k9&g>G^6CFchKy3bzF>aL_7E{7pZ--Wy3$ zni4R(Yn$vV9$x2Gi6 z@`Ur@#%cQJn005#3IOvIo5h}DivqVp#(P-p%clxV8gAuu2@1qoF0m&Whct0GE#=HP zk)NC019C&i|E%4@8I_7+xgBzy*jA5fj`P*)Y%-`GK+^gd`@20V&Yqxqy>SLIce)00 z>q{K&R*_LYB}krIrHa-vxn7(JAD?+(7PxPNHF)CSB)5w)N;7|t4}A;iRF%(hhOk4f z=hqdHa(pnoEPbQmcOSu0UT*rcN6HcR5H+iKW7r^0Y;&%V)#3vweyg88bz5EfBQ=!% z<#J@T4jH`4#13hjO-6YsRg` z2PbM!X9jZ@hG`VOrh1X8okWmauf!G=i-wMhr+G*s$d%V^X4xYaHMs_6Q;UFEabATF z>&vzLRP$3`{f*-|$s1f>&uNAT0s6Rpy%pr_-bJaT!2UGYN<>Zh2{FZWJTY>Yo z{duaV;X5N!*bbVIpSW5*!$4-61H!g=nC(3E&ru2Q)WNwhV0*QhF6mxXxe10Ha{X@N zsP$?-g_AoKr}-Su$9#<3jVe8dtF$=Pbx>qZh^Wltb461sQWHdduS&FZb+A<;mTDZb z91dbJ@7qz7-C2*qhvcud!Q^wnem)+Y_~alPGc%r2MfNMUEt1i1= zDd_EpCNy`bF9nT-r}McW{FPIXMNZ1g!TcOf#FwF-NP&l)Ox2D@TF;>)B*7daP0H7@ zFPG{->GGUW{HhFk_;)OQ{Qt;kbUdv3w>+mhIM|0%b;|aq>;Is!``=XWcg25!ir3bf-Gb z{{-0jC;qO1Qb@4afFU@>_3L~BDSGL@kZ$8kvzBS2qNy<6V$~HS>O8bW@q5ke!Vnyt z`g)i^a-IO`ycx~S@%pi!3(=2NaQ>%)G(HL~mmndtBSVt(^d&(8S@;Aod2mHk9R`LFi6NuWk`}9G^vU}u0H`Xb0+3Q#_O+VXVT}1IREpqZS*s1q2(|n z;DsSc`YdclVMx*wAhYk|3pQS_fF**zqmH&z5T6Cm@;4-8&SXfEo-nt7KwkYX#CWDw z9eL6C9K-1~c+r0jjMM`TNIeh_zi9G&798J(N%82P;zhH*Mm|w;mX+{pT`=sBS*CUr zj}%|T;@DxsP!+G(Mkn}V7?>qwt`3jq;vRY_jut{tT0wT@17y8GsA?u_8D83#z8i}3 zKNaLlA3#__=7u0RN&1|5`RES_dML;v0>q!S7&BilO*Oas>M^6r%a>;rF#qef%-oY) zhKrK34bv0I=cvl(t-tXGWC@v9X4dAS{QRIKRfgccf^7W@ka`SpUEky{pXB+?Ln-;H z%?VFK$jg8|@=2C4PMJ0lN9H8V24^wxour#sF!JgUoPVj37rw`hE=bV9!t^F8$w%Xg z(4&Az8l)h7N1>AB599~ai6KdPbU+1x%y|N&hCfDflJtK9E7JDCIR8>X&fY{u-$8;= z{!?h`t7){@#S zkygBh^Dh--{m+1GhXga3r3+EJ(|Y@HvGg)J&Fuim^Nmi zC?2n$4l7Gj_VdYhcHRpg;lnT&5;D&)BuQ_lYXVvD1jxdkMOaJOI~AzW_6XXeAbn3k z%d?P>`7uM1^r8(*3FO*;Ax8Mas>Ch+x!cin4v(M$*X@6c9P8j4$}RrS|1D1GvFQ5l zE+W~C;h%_;)I2XW3yje{eOG!Uqr!$nK0VQu0!FU>CuKx8K$$w~! ze;s<&Qr=gT%YR7Pef<7PzYF^%tvEgAArsfC%!ZQot+?{EMy2gCKHU2Mu}A)Y&(25M z8I3`XIPtSNt*b&|N!uemgnmoE}Y_aGZj$W~~4#3_I{l47F(;5LCNR z2$odNlWDAJ@xGK48An?bKBpgE;_P_vj|&3-Vr^d@*BNvxnT7&?UE%kj;ia+T!Ea}L zE0>HCH-EJm?E${{4V#Eb{efqPO3ix1kJ|MQ>H)Y>aJx>?<#gR#9ZA0eKM45H4jo>W zHex)-7)iOaXlOJh*r z!)UY*BWav=@*3*E5w<{;D4k1Ger+Qu(T!5!^ifnC1^nhHkuD|jh3J1M(TL%}Qlb|w z)ukXX+`vntl!)yq$_1*;sx5Ph^21O?E5JoFu9lOcoj->YqDscmBBvrhDcZdMt!OcO zv!!TVI@TwC_*c9%O3|2QDcU(!EyP>pd)ZmOpF&W+`j;Z5e2pvAp~*3ERPUyPI$tSY z21Ne@bv1vlB;|{!T1BXv;HA-2k;Z=G1>EaXZq+}hl$8HtQqr@mnxFKOi`&roS8l<@ZCkyl6!(N# zltz_~qq@I4Xt2&qssYjeP~48)EMlKW`;%WCv%EA)aoKh$?le~Yf4m@O%L}6Z!f0tN zuKLVJy3s!&<^S3jcYT}>GuJhI3NJZWf_!d6!GHVNcokKyEsIPpRIDO>?1l{FPMKF9 zTfWxZrOngXmd-kE#^4_Jc)epvtn|41ICXeZc>1N|aU6=V8tVkOUm8$NBTlgyKgnl@ zW`EAuqt&d4ub9XA2m^{_G;3W}TK|vlV1NLCGn^y0{@p;_^Kh;*cb$KKEs{34;wbda ze_Cz3(8ZRnvy9ZeP)N?>v8jo+Mt9s`u^1<)>OKm!< z;`iROd+zL!4}QWL^tbnvOZXX%^G+QdT(A$M(4 zIG{DCfV|wcFPj3}L(qic$Fn{?OTf>o1Uxz+w#zU@wIn3>IIkf{=C+n-c4H9!gEZm+ ziJ>Yf2eda6k(W`gP*thnU88YWB1RT=03-E|1JVf;4>zLD&W3I70SC(`!xOH6ia10V zHT>U8K6Hr{{CDVqjQqVT_ietEB=JP6l4J14JU2lr#OhK^-OL7tR6P1E2(i1GTM z(~V>h;u)t7PYU;^Fh52#0vKeZpT!I9-1YLl|#B@}pNmd9iRBd9fU24OZfhIkEa8|F{gr;DpA@8Fi%? zTd&uqRVYT>{r?mr5~ARHFQFKvyK^yScHv^QigZaay6K730+J4138fgEMvCzwYp@az zbw3AW!VxSb`_{Bn0 z3X&^c38fgEMv8HcHCTxoa$;4Cp43 z7&W0CJAq3#kz!n2SC}3kvX@svDF&yJVsv8-R^k$z*mc@plfUvu*qy%=ky7P)EVf6O z*&~1DAD3e&+&2vGT5(#zM>+3|s?H7b5+|$E87pZrr3VID4f7+MQcD@&K-$Uxv-M$CUgkFqq<<&AsZ4hZ!;uGpU;N`Kv4MMe<4-hm1FVm zj|!xWpb)A8h`dxh<3X z0XnLO&~&BVF;KK_OH(RYz(+5;Iguv5pP%+vStQe*ZW>!;?HtikU&pI-4R^!MC9P%V z`|{gmiKW+6nK*$qC=(Cb4Cq*qM=*;2q|=8GMF07pZ=rhxqaDvp^!bYwpWjoNS2vdZ zB!i|u4yHOtC1iNUl8km?2{M+~|8!kh=sKD)+}~_I`c76yhbt~PQKocXnC|>V#ZDT=W`S_C+$iEM_<3E(KQuKw=tia zaRbpt)OGw_F!#8nL<(J5ga6`RAi(>$`s3@k)qw^59ItW6ayUrFt#2?3j9Jd^I40l* zW}90=ffSJg+LktW80;hCTF*L?)3iA*0Z#vKP7IWsB(1lDlT~n(3lD6TKhRhdZ1T!U zGBK+@#s@u1pWzcqJl1py5FvyE+Od{+;;t6sN=hxQ5fDs89@+6cIwiG~JyJ`fnxcbw zY6y~v6FQWi`2bru?dWej2jBBryaEMMJ={Rb+0{9k(ue8KOe31j(M2Z&2sY^7g8A}j zrJS`G{26~&e`Rqh9cHk5B#dVhJ>y<(_1Z^m_^H=@CYZ)I(Dat+KIHq)|B+6T7?Vp| zqwyhiVq_4n5hvq+FppTtKUh3Z7|rqwX_=vypk#pF{C1+udJk7RCLXkBc4V@VLz*}o zB%{=^y7KbKKGKkQjBA%l#sSG%Je+l&!7$H=n9=Qva`JyUJR)0^TxBKv@(>I=Wc)h3 zzGPIQT0PT9z zlS7&~QZdc`b82%Na`i5v^{aj8Jyi(l=C~bfk7itF{lea+g4@YBe*SKVCO&Jz>o@;c z3j82{#;4YyHs=A>K8{3}i5<_l*S!2&gqR0 z8(1CXtsTn~Yn<Isat433xP$BLR*1cJ_F7BKP; z;DFQ)@o??fc^X|2cOTR8NsrXbcv;pqK8i!r10Bg;DjyDLcc-HmMn+n;l`J~Dmk?kHM)ORHPiWd(Z8SQF9ivryu$-XBT{2SkZ@SV^8fp{HYLB#gFZ z>FYJAE=nRWfhfERh8;4i`WNRWLwKh^T0Sd|o&hA!q(^JiFzG%&j|AG=$?jB__0DS( zB-NPzE7cTEWh`uBbq2S0Zzj+A8!ivOgZMd%QQ~2v=FGq>xBKy1KC9E&DN}Be`>sPe z{eTmCZ$XDV=@aRAW-Otc(YuTBr(w=ye~R;w=w#aRUNNf`T@;ewj{LsRru4&iMZBts zyy!5h{nTsL*ye<*RjKF7IO>KJ?rLuhc6DMUj=tRKu54fP3h4G=yti=cY-$Y2T(5*u z_qq3G*ZtOwEL}32;6{OL=-dcO|9o=;rU*ci``Kqw2z6iw^#Dz9mG? z?w~GQyn3O4@K^TL+B5)tyop{3g+J~9+Wd{+USLkEp(u8ie%`k%?Xb#~Oz+eTu>PFu zIh)it{+twotBKeA`koxpbh3*{EVy7L>nj0m%T|8rju}uwSbr3nA)DyTd^YRcD-+hm z!>s4&v3a^}Mt4kOa+|!>mCB-|2`gdN1;Y;Ie7vb7vtDpqHuXTX&??*VB-$KssWvy+ z8$Adn?p#e$5DsX2cJXH_-z%aaptm0_SJARNPw`M!OZ(TRBo)nT^kh{YA45j#w4r{rUc);3X*V!WJ6w^QtMcK=@a%< z5a~8@N0M@O&P`zr#ky9dDJuF(2WSX6$Qmy0l`RR&KPH;@bp)+d5DsW-kMsZelN-3H z1nr;2xnaaz)<1xee*p*N|12K3TPY0;wAA z4!dXS;6`D5HlFpvtPG|Ok_smfgQwvq1?P3IGX5Y7VBN{B%SGx-@4ZnZR=9T3T?dQP z8kOwqB0o$~tY$jGS4j)5M};K1WJM>-ss~xqaT1c?!RXT2%O1p=({f`){;qrxi+ykf zq#()k#*4nzcclO%!BvJqwdR8R|B?3P@iAWC|L0EfWD-dTMvxF9%ZMaG5L-edA`%f* z`!cnJ+Dq-VR&7NQRZA_UwQDV1s1`-53q|SPQnj^IwWTd>(f0Q~=b8J=OnlnU=llBo zF?XKlx$krC@|?Ro=bU@5l1oSM217cK4fW&c<+A>i36zwWZ`AH{=*{?w9Phr{T)L{+ zuVxBx8i1Y-M4DYRW&qMv2`%3#2}}l`EDaY)eBw-PVX(m`B@?sv0TMTxly_PYlYXc% zeB*Jm*?Sch8%MJpcFFo*oeSj){*DIA*ZM1KkUU>O&-7>=>HFeaG!$TZs8`lav5*gc z)_Dl6Z$JrOJ!GemRqRHYx{)*(#J)O;ebcmW4qsuj#BXK$q@`cw(x!_wXg&x;Bv=Vy zDlZ&2dZtm+^)@5tvHGG9Im}F1O~hIb>4QD$4(o~>6Xvv`kpt|;B73wT36?A0y!b+f_^J&w6Bk?5V{3aMiz)?oR`b>EhUok0<{KHnd)hqID_)bZ{x7fOSQKU>5ZoQZw zWHrjT$v9fNO1a>!vV@? zGG!y&&%X3GKF+I49QPEMSR_5dah9epJwIj+T@~r{ub}6s^6F(fFpK8#?-v*P-Mbu#$7qD<6Jam_+^E%_a(>v@my_h^C26 z4RXPC^3N)#XZyob9!$R91Fbr`n~9pq|3RIkzf~yabB+6T;|zRwiZ%0p8V=VC_@Xw-Sg~rOyjD1H7IV7|RPIr?FfWH}r3hYVq`RTBAc$}`vI)AHIKW82aZoh|-!mv41zwEhxqOhFv) z=>3;zckMHxkive%ro!dCs?;2sB>_a*L>1bj=bfu-AUmAC)Z zpEiI(S_ZrFiSwSvjDPS5ba+SGn~rizcR_6zeqRLZzCsQ(p`2f8&;?D*9$e_cG2^my ze%Xs4XLZIzo3^Dk{&Z`6Fh!S_)ap{`E^pgXTP&q?+m?#&*5=O^=cfko)0QT|2q93l zf8`myEKxXZ93NF*mMBaBq=cf#cZtG=WaP_v_T{=fS!~!YM!d~eYsTV`&i?1_6%(h8 zNz0qciiwZzWo#|Y5LbksWyORn=s0ckoEabsI)WVcWXGbC1s#R(wV;HeoBqU$7^;O} zH8x1Z&Xt9I&D5^7cv%w2A_m8#GHH}}%HNpoUrKyrdNZ2G4PIs?eVME-;X6w#`|m{M zkAJ#G(6e4t7oZ*y4p7K*;dcJK{`of1$ zB$DCoM*b=gZs(WNMh+&p?EF%u!acR%>-=(oonIVZd>SJiTg1lZq8IE4GLc`=DRS>9 z7u-AkqjJ%?)F#d!Kb}n>)(MUS50PlTQ60l#kh|YD!|OiZ7$1^AtN>i0A$+4=V2I5c7A{7^)n8 zPo0;?qfl0;hc-;)nPVfCqJ+HRSk!{eNXAT!p==;T9v&}`Ub6gX$b$stW+WHZqd-MQ z&krlpZ&5MyAiz>w3rDWRH29nDHp)y19IN(icQpT(AB#RqUtg9tHyajiW-UA{GupD zyyr;8itY@{jx&F-r?6w#%-S+5{cS}B8Q_`%Bb6`a5DwgqEvVX;KZ>HFtQah6@Tyv~ zgMA2(?p210V+(X^H~%b63!&Y@oo;jF~Eh|NnDt;_T^{HR%R52%puKEy* zM(+~S+@lEJsg8a+ni@c*x!Q+NQnlPxmN5s@ZKZ{1KgCN0sWvd9fo=1VuXO^#gPgyN5l~BaFy;K~c`uzTl5a|O>Rwq#^b_GEC z0MSlzA0QXp2mGU&(%F16mZql|?_Oy_?2$4NJfu;Gp0PQeV5r0}fT41QD_tNj>SS^d zD?_EYvhbaffN$LnDyWk+ujJCq`|$ojW_6;b)Lp`(do_~cpn^K_|F#^x27RM5!kbXu zgyo5l@_hwLSUrY zSp6sEkqfT2f7CNN=adcLezE!8)-(!Bbzaixh=%S)ckViSjQ|AujL*-bialcJ6AhV* zg{+DLa(#LV-T}Lfkj=^ThmP(CCkj>2GsXdqdGv*w23Lb^7Lry}!_&5+< zt*ZjRQvaS+Wq4b-ER5bqDt%}q_D@ejt(UCqqGVYBHP6=Wv= z+`OhT5znaVHIql~FeU%@nf%^}O0r;R(;|4TZ3&~{U?2;I_y(mLF7!%v!ema^U88@b zXfnPVk}Gkmn1Tm&juuVEom>=#DX{Ef1blXILhYQFYs(3>XBhmSPpI{%lX^nKq-t|w zT1FB#LpTTba)@vasvbqV2F6fyl$Usz!WoH2p_j`-ID4*S;e56nYTS`o4_{{JHop9& zIhDbaPCkSp9t?L2XC%|;ge%X6a5CO;wMAVK&ST_H?FYrs8gC}1a6Vlk^|_Mz*};V( zoB=ayP&S^N@gbDdImaTLpE3tC>jRkBRW>$`_dU3rnnhF%RmWX7J4=bjl-q6U@U-$7)sE(7NvQ$6f=&=})A7 z*_(-3kLNjcJ5+_C>)Uebjn1&#{d^4TDWHMvQ$3e%oQ|Z}>R#CNI%ErjveE8b zTBZ$hoD9yCuk)Hc>IkMaF#m&z;-k0f3EAA7ief#Os>%M<=#Wsb8^>{>KU?s97H#3& zu>W@1OVDl~ghjy&%>FE=v3%)%RL-CWhsV${;oortSvX5v=3{GS$D{jUIEI`~a%IvD z;LFzVdf}#=2IG;pq_h)B>Hio69CG@?zG}1|l9J;?D3Zd0<(AVUK#;Qx3kViY=Ol@M z3^(fnbF%<*#O4e~&=?Rl?iQvGYWnyxDV)gbUehP4<6DdO{sc9!D9M2Za&4p6u5=*5XEz z7KaqRQxfnkjt1>!PO{q-5F1C%fl6MDm`l3g2~MSHEoE9dFNN5=5gzBRm#&lDjK%1| zcS-`jCD+MK4jvVTEUHFLn1U;?`zBzlz~k-EuX7wVm=sCf0W`fG_>LT#UnnEp-GB$1 zi(z9W;!AXQY_RZHM)*!iU|KyM-&Y{LRilU65I;V^IpMj21%!v_Vd%w48Gl6977h2` zI=7Mwjld3h@$Ql`t<`#s9cstwz3<$f5@Y9PGt1o+(=EUM%#>XG&03*^zYS>a{eb0NA8w5 z8ac^s?A)Iz#*<0^6IsHHHUTtpw%z!peieGGzC66H`A^r0hQ4TKZfa7-nlwoSqJ}86M zuH+`kfvFt4bQkR{TN&q1%juN`6YQoHwR*^yy+-Od+^->Gp2=kMjI-|m4Al6FIirXz zE@L!AY%t^Tgq#DeAed_`m|2tNt%#)M8nSXZn&>Dz!R{&yNij~pUtb`5?}fO|;ACrb zQk^=7(AS8T%C%g6Wyw4xTv%w<*|X5pV{hi~R0;{*RDRAYGz zB>^mjw{X_{!UijwW2?Wp>~&%Dc-r@L3_T3IH4iptqC@cLUW3i_k4c8#9>rNS73P@j zK7=y-vXMt-qMMn6Y3-RO)}EX%BBW3KWepriF@*gDfTv&i?_X@r@6yq9Mj4Aoo38h9 ziEinP0W+yU8RGa(Nx-+{{Avl#UdIkJd$2Lz_(*CEDyf@t%kaJ)z!P$wW=M+h{>fy4 z47nGgXD{7Xp!&{S$d%d4+ti3YSsg<=fw!v9oStpPBX8r(S7MyG{|+4bFg?40Gqc*@ z$t52`sXjKxxMwe2gl5MV{?cExPHQ4YnS>;Fso4S}rDiLPlt?xt@U_mUgon6*Gk&bg zzS|?_1=CRFNA10Nn5G;i!w!Xv>uWrCu&v;xFpfK24^z1^eYZD!k3&o68u`af&--Tq zmTbttBRgW~T1(_wv(?OyT3hpG7R5drL*D{lqSnsgQE2A^eEz(0fK6IqZILNQ_SITa zYt6#5X$m4ElD!F~))?-twYfl8(th`8B>eE%U`cr3;Wl&-3HS1$m&s-~JPJLI)I!f4 z<%DNrT`WF)k5!O_Tiff=O++sCA(Vs}?oL<WjYsv6k%iSOS)Z^02W>zEP9f(lVI$4rx5!sG`4tcTMrj5DEDg89cmtyp&UP>0NKdd3VZ=sq9vlb=1IiRAf*tHEv zp{0nIq14yTh7~+fPA=3;Y_Rk?V|e~)?H^}4?dyioqZrBo203mvsw7I5OULTaTI_M| zU&rgsn_co>zEmHV{(Kj2fChbNMFNv*20HS-aMP+q)Tp!IJa%ikBMDZAJ68j zI)9F${)m&T`mPUf=L{dC)25fB?X6?!8J`41CYLkcdklY)Gsxtc??UJXa(3E>&^KqF zFbC6W_$QzqNvbs&lC*FGtV{XEqHbk~Eq%dt-6Fs@>hl7;U)Lu=FRHY6Yv#|v*RzR-nP&oqA`eHJT(->=6fvip^*X4#HwFjPe&`OPw zt@5mZ99efE~yR?QcxfZGlSM5MYag?F7$9FS{YS|Ar@XK7~F}Ff0(m;*zUAkd)QP0eWv@*z1<83%dLtkC^Z4K%(F_vD!L#r=0rS%0o@-|FH zqb*|RgV)kTT04K6L0@9V^^*^w$OIcRMOwdO4yIM}uLLHRR3V86`UX-}Y)DqK-aV<- zz$5Qe#T;s;3Q4>)r9SP4%A&syp>L|an1h+>@0_ahSG*?(--NAAh4yj8OV7xRROu;r zF9}Tsqg%is3;VCwX&d6DTktivlJi=23eR3ryVa1dbsszYz4dI0XDh8GUxVWzob!+?Y<1{L8857lzI^T_?ZW_Wj>aiuv zrRWm4V5ony8cn<%Ns}~$Z&cAzKwPtzN{9Oi*6glDv!Z1_e!zlz!1bzm#fsm8ZGopf2VvDQt?!As=InAMaqRbNY|;D+v`tnF{zuK$N{ ztUEA$?+>LEjj{Gpbo>8jMTsKt?wT;_0F1m@wvMT?Fo#Ki-RN$#<&w`?lSKSV@hGD+)stClG2^&}@|0G%%BPlZK#oswrW6|q*4~#a$WtJ*zVJ}r|J5eP z8~yP{3a=Xv0AyY8)Ge4CC6`No62!_`Qg}O99ixL zjOEj3&B8?hF0si?3ewXuzBzZdkMeb|1t~uv+lml0~tZzfTcg0dvQ+Kg$euN&119DwC3!^o* zYp7t-##d7GygD zX+^$`Laf^4II;-VFKqt4;`c}~Or4j6&>sHXXcq8l7&V=SWAAVlEh8(XBW?C6=bL|A z*OPGj-I~(Go3_R|ZcK8oVTxQ^!=t!*7QbDV`T!#@o6U&I?~FTen=B#ZU*(k{}cYwUSN7ne53w>O7UoOp< zO}E3SyTF6A*-%b?)t8>4B(d$F__z!C+Yk4N_!sARPa3SMVQ3wj#nr;|Q} zQgCav%u(iGu9n##mZ}%FU?=3=#;fbn=vTzaD;fL_ayI1%B$fkgR`#Qurc(cz(zFi` zDmHWDJP_BVvS`m!jm=xJBPyb$64+!Vi*DR2Z)TK|ajxB&w!~Y)KcLyjCh=vDaSu15 z61;)2^Hk@jT1Y3lj(PDQ;A2vfOiCx}Yg=*)etjHa%}ka;f_O%IMKkd+4ha+ShxWA+ ziKik?xt`UM=Xxzx2h%s#OHo*JcSby5z9Q!*jO%O6IaW__8H{7W@x>`qRJyi{;97bJ z3Y$I*WKm&5lq}UT*XuafMF#UQ7#-e990`}{ghdjWKs*XPl4$xCe!ihV${fEqT%Imo^#bf!n25BD(fOQRwNauL`nCVAO{Pe6~2OoeX+d)JtwSdcK)_%6RP&DIo?L7}Vw z+Qp}_-QOsqT)P}8?tY!slmud?b0oIH!Y*`)`(n*iO22|krU+~y`)9*i zWW7d|F$ISmywqImCs{4riLmj?cW7rtq^$7wn8IX?XH%GINJ$uY9{X^AYcsN6uTEpU zIZ2Ip=)6p5U7rXLBX2;Xs*-^hn2h^m$-qCYH*&OBI1Up=UaD&o^x4O8@bqStofxb_4grR10y$FIpTrA z=4QlVtC9vS3>@J<2g>+2YKxy_aVskKtr5NmSRMo3>4E=Pt8YG;+$TN>?Dj^j?o9-^uresP1AgQ%~+IBT;H5a zXDte#jT+=@o+6j4Ll!roQN{J?C}a3tv;9d{1i6<5RqQ97(Fd(t(~lbCYh5GPxWA++ zb`MCSaxL9_*hnR7gUkg&t^?Ds{AH@quWS&dX^hCO6=0S0Jf!u$v0-IA4b&JmJ;@8< z{hROEw%ItJD|OLvX$6|kBs`z+HAInf`tQ&&6d7?Dp|s?OSlS7ka2$!Zt2n^ory}v` z8u)ob`dFtdeS;N{D{^ajE0j?Vm z6%JtRyO6}$d2v)NT8RhEHO>QZ9bAn$F_k z18in#oEM}smIR^u+cP>zhRJ<^Nc-UL5qa4%BecFw+@CRy)>!48=Vy$2BS-e?{)Nfv z??@!EW;J$Sn1(FNq(zPgE-LXsB)>L-tlkP_{DmZBd~TEYAaZIrf#x^N|1kJ&>4&`| z@fp;g!Q6BO%yP(2SBWdnPKQt>YZ=GY>O1u(XB6Z58s{=|1b3WqTqV9Zb8L2|)fXMz#t)m(Vc4*) z11xWbZ`6$2kZJwd^*pZ)c_J@RKD&N-zZJcS*aofdsht!&y4Rc*(9bj#Eg4=-5*q14 zD7C{TIQi@v${b8v+(|qm=Ob`ffqJCj1e}@$d)E@gOD6e7?YZNX$w{2aDf=r+CTGux zpj(Lj(kCsk9em75++%thjy;n3+py}OY)PnMo_FaalT0I-EQ>g|-7Z&ba8-VX>n%1= zXn}2fxT-S&@Msaf$2YIL4Nlx1B0j|P(#86z zJhnEK&39vb1qn;n-^PB1OKirW7qDwDQg81K^qm|@gBN-kgQcx&0gK{yx2vYP_~@OZ z0o2YGM@zx0#3xWO9^Gq;{Ij&Vs`GHQcq)XY@gpBXd3S8Tly`TWIhfk^?rfCYujXMl zwe@YnsM7PtqiNn8`<00}`Rh4!%C`W$i=QpfKMr09wI9u)c?D@F!AS+2yz8KOhm?Kq z&{V;w7;vbKpMVvEXuG=q#~7Lyq3l_=p#G3?X3ab!CH6qWA4sAI~$Y?VQ_ zA&7if11S7%yCj;o+J0#xmI}|p<3&13aTY1q7#vQVSdZI#>QbfZcof9xSfj|<(>&WEhG)70u z4jMBMtvDd(FW80+DQ!EFPjBMEa*gvqT*tz&9~;JJ__*xT(c-0hh1qa`DOfLYi9NG0 zo#HFR(bpR1fwQs~V>HE^up0_RrP*0YP92_Y{xPJn>`{IcR05VA$w#6<*g7OQgx}58r`r&pxAlV$!Xu}4;MaHXHcFO4%N9FEwXO|brDGtchXM+eywK0vUC89-0 z*hDAAvUbt(-kSMG?0SUv`Br)+jnx!vs}l-qV=Z%MX4CR^9)m2h@!0YVvh|9ir*$@X zU7F&696|eQ$lK}oS52`$oMq1Zc6_X4l7y@e@->&q$|G_-SFt7y8HOELx8aKblkwp8 zV4{22VBNzijw0vUaOvC{Wjg5rn7V2MEA073)#(c%b4>i9njAKM`o(&3T5=za@i~)y zj!LPPVIIX!u9a2s>Grh~~BzTntT_KGG^5VS^M`XWbO3YD4?q~Joqwh=3pZ|Bge?F%F^U6mn$@)W_(;w-mTsv`oqFLsAfbY}j4fB{ z{)t?Fv4@**<+O$`5Sx}xXo#4X+TjTn1eAh{R*^A05GU15j-wkIa*S;@9FTL|0d%$b zMo^t_I)zQE9i7}SimeA~=2}ijj0bCs1;m8Vx6 z0W^)txc1&@1hYQd9KL)M^F_Wh1HL6^-~+jwij%|i@?C_q$E;&Nsf>e zjVSedW%yUDPPb?H(`Oo4P^%(pxYG!lHyU$F*}utEDpw|m3#!$lpw8}>&qfS+dDf0) zrb&ZW5MjH4MmYZ(?9*CTrgz8rQyRzwJYz2@|)@Yp1fcGy{#$s z;2A|J!S5xqKyoJn349*P-6!5U?S#&1a zpt%~n=%RZYau3*f9$QYJhTgwYN>t$6G*O20Sp z7Kbtgd}m7zE!dA!@4C8knrOQR?D}r9%z7`6b5LX8q_MLlL+ZiWeay6LOu%y-wj9`8 zI*W?%-Qw@eDNM%V?qk>xdfCav*TNy5=j4pu0SC0XMj^c6>Of`stj^-zZ&8QacOV^S zHHM@ZL+1ty?XxESq{R@~5b7?rIsJ z!Jc>GY7akC2-EBro6>#2Nwu)mLW=A=okZlzZFZ>6THF5EL8ah%JJLD}) zJ8K$F{JXo~f{)=uvMX}rYe6yXM2?(`293PwFe*1R%_u#YNu8CoWT8=TroP~n$GkF0 zl>IL5%dCY+URR7z9E(G6F36O%(ycuD{%9P{1Y60;Be)0*JU0tn8(-pZh9&qoO)&=M zG^XoV;IUZ)PZ~1hqHsW*QE7RBPm})_(mN##eGv^ zAbbV#jie9C5uZCfOhd#Vu<|27Tw^Bsvp5AnU$q9&B2KKvms+uQ_%fCHyyygK>J_b+ z;`#e4(naLSgL#W-9!7-43~SU987x ztlL+G8fXfx2calozml~>&Y7h!=Sndyd{ap_3+oCh0gLAUceZ^v*Gz@`OA(diM^9*rhF|X-KJhsq=MRt8WHyH=n;yo4H#N=!alJDx zliB=Tx&?i!qsO4wM^a0oex!f7{KwNVk6IYM0o2itnqhD?Y zC$pQJ8xF|X8NT&kvwVX;)jJY)Z?%iKEvwgjdWc4#-t}P(FCioQb($Goejw>F4Nv?Nc|8k z>9Tc2(zW(U^Eyq#F%-V}O=J9xDcKcd%DR%t{=1!`-LEjy;`{u*s^&&(=a`H~7$2*M zBrDTVU-O9RJm4DncQEJn%C$DoJi2qsANxXPZJ%cDcg`CYVSkb7P=n z`nq)_MfY}>m2cDwf3nPX*o8IA`9|Z9s?qM;NNTAee4~c`#gI3nWz*V`Scmy?t4JE4 zA$+607RbN>28-T`!!2ktp3c>fh{kAPui*()6AVc)u0_-l$hLbSuCYnde_s2(6n%|o z5xsxd+tNiFwRl2f`UT2iTOW4;SzoN4Rq;soYojF>u*MT*=ol8Y#c4=wynMCF z0K^geWp#?KRte{yMu-6S2F4nVEj5CC7ynfnJr8fYs;2Vs6V7Hq<_eZcnbgJ}N7FU; z1B0EUt_E-h65?gxcqcA_D%Yt>8#P3ZvoEkSB;!qyN!O|>G#!evBO1atYNSA>%)^#G zIB^bLU^>peyQU$0qbinSE{ky_cAI_nFr1FgNBai%#Sk7C9B&5X+`AJ~^n7F1-0GyD zIIE3lksQ8JdvMGpM#i;LSJ{V~i|c)R4Pifg^<5=~Ozkc`cDH6VsBT0&jn)uezM(iE z=e}Xm8_#^a68#z(Ps=pqHf|ILjMd{&Iz45QkJuOWgk&pJQ$=QQu!vAtsk~#E(6BZrt8pt zJ)t3@nyekoT%tafNk+%qRO$s~{tgY{8+8;uF668|Nv2c_4?C&6GoC)w5D^3p$Z5O` zC-5TUk@1bGUP3(G)DU@%9N^6NmopiXbvK-*08*)+yI!7Ubt(rqMg2!W@{I*WQIrGy zYEuo#umN%sL};*Gq(sv$U~vO9y$@&z->5BEXi3f?*zpeuE$JRdCo@ zCbwvgv9-@YOxZAp8t@5OP69^TT5UI3)bgYH7dHbZtffC|};>=qe8=W-}IH=a^8hKI|5_LPIKZ9`+iO$mQJZngW*7Z#J&j@guBq-M=ZB-w1~E`A zuO^Dz&Ke`i+;LF;p@$m^!N@mgc~yEC4<7M>EJfnvjPcLpn%>^60VQ{gr&Stqs5WxL z0S;fiW9FvgwqQ{NAJQ0B1nb<~8Y#`a!D9yHbcv@+8dI71Z~#^MyP;C5=3DBEHaBRX zyX36Rb@OrjvcDQtJ!Lbxq&Jb#p{WLiZf6y8%Mm21%-49G!%ZA}h$VH?AqYiA$G2P1 z(w_13fF|%-3Sobwkn7>{FJbi63z0NUL-g-HK3Y@cWV3$~6^@Cg4LYrkxKSKH2o^Xfe_McIKNTh1zhmx`-g8o$ z-4!VKy)PZKSvTs#UVdi4Eji8wqvI~IqBd%JPEZIV7YxB&+OEj zT`}p9OUIe3wJ7@SipJoZ=E(@9G4}r3yO_eWbyHq;uPiGse&shcq>?g0DkaR!blcJR zLf!uoa>@ZYe_t%!@7z*$dVE7X6={ODI{@K;oDGj-s4Fr$7lqKFjqx-~Lyl|*gab@! zlBLH9tlxwVZH}ic8uHmTKsdmpCRbMF4g0ha9orI5$24RhLpUI3aI!2tTsl3ASYP~; zhVYFlzm@BL|6&;fZ`l&)tL^br?jd*ip56!u2jpzhRi;Ld9l#F#JL0KU=+#J+hC6`?KK@QQQ>Xah>!r+#&97JR z=oCdG_>Ve8j>Vd(@4_iD$p*)~(n9z#^JSk!FI{2RF3=TLEN?XHfpay7Bfe`~j>pAW z6o2BOECJub>uE|2iVmx6yb#n@CK6dp_Al+1CmW{BIIlKEqksWtQ(Nk>H=g=~QbeW? zZWIS__S=ZB87!cBC7J%L6iJV3NY_`V9C&K7s?>w5{oFc`paZ@ot*=`Em88e#c7PVd`6 ziB#dec#0h1&UGYi6bIzoPBP$5*%(D^&S<6~e4~<%G9(o{zCjRrPs<>-JN3~JzEQ2- zXUJpkAfbGt>FGw4R5p@kXb9h^F~=CvYiTq?b|3L4-t%LVhVYI0`UFEB$G$3nyfq`9 zxTPP}5bgm~sZ$K8@H(c8`Nk_PTM{2!_k)J;jmr3xA+tJ5p-a65-?T5{DePf)q1nSz zaX^j@Ln}+qQ@d_D)&Sb5t;US5=IL!v+i#3FA)H4i`{wr>lzcD(aKu&f!0nES2A}bo)vUol+ z4&SFn{WZ-+NpwSVm-){#ov<&7b4BY~)T)~?=QDqS!02m=s~5z14ksL(HSUaQBr~ma zABar1I3VZPc`|)*G|QvAU&T`|jT3iM4zSaaDQ26d8Z-u*I6tN#;_S%*Byp_b8hLel zBkB$p?+qHlH){Dg)QBr1G=>HGrJ2XVp(6Hc~QiA~9;Qc%Ess3BeFK?Vl=050b6_gzI}W7|{_pU~0&EwwMqxMNse^{_jxmY@a_+hhOL)&38CdE+4UK%FhV*2m zJU@|u90##w$`He5%*)G{K6gE5)W;{s^tv>G4X_XH@IV0`zZWR^H;evNMdSYCu~PKF zG5=n4X*pbecdebOXgtv?NJz8)3#rta7wMLY#wT4JLOSKYkjf}&R(J@GJwa{eZfs4@ z{2EWsf>J7cJ#G{S*rTIEFhfE@8&R22HR-U1Bs_!(4+q$IXElc7B4hNFmec@{%No*w zPqE;DoNcd*-Zpp8gor78k>1#Nt8NsjI6ScJ_zGhr z1u<1#M3u%ARF|&~Pbp79P5CPGQ~?F$@YRY1c@)%+ub!UXoPxUX)#%j96jX#O=jg%$ z&gjA6^(o?HA{FDA6!&*l*>QlgG-9EVE;7y*=TjOb(aRdbiUnnA7fy7N@6PcHVyQSF zi9Xd3{`6Mn_xFiV5#*VR8d8nGB>F=`*kwqWenSbRWGp!Mnn9OvN>`O}?!5d4R~W?s zW@R&QxMGp9Xj&?jDwjmfHRMPB;^Y8|Kg=}&S$8af29JuTz8dm&c|bUTzxy50c3xf{ zO7kO=Xr_jwOap`ia(@4~bfoQuq|i}7HfxCVFfyri9x9Y}7r)g`*8zE7L%7pY9N-A! zm*w-W*+WDxV*LG4Lu7XTZ4JCa=O;hGaF=gXeY83?0i^7BccH~x6cWRb(VMau^6ZFk zs*kOG>uZQ?_M0k@rBzFDp-UfaMnA_VQ5OwafmP&+19JLrD8pQ)M@P{)Oc^I=i1Y!6 zot)4E*Ttvv+KNKrcC=bU#87cnAUj?WhsMI64GK+2qJtX3dk&i$4?AOC!JN0qc(Z&q zJ(!wA7d2$i2|zf&(oe`SAgxXq)L}pp1w7&|bbt7ZC=SSZX#$)Kij41r5~%Y)>?ook z`(P1L9FX()aeNsU8JnNC(E|&Us6ay+@#xC|IUkM1@LyzXnp=+?i;`%#hVZ6;%G~*H zN0EG`cl#ipngH^ohVX(hC9B4qmq#Tsmn%Y^K ztEEZwxrVT*T#3Eh`Pes@ycHRrJz0;w2IP*0(xhaW`Ysg?u zhy!xonU0As#Cu#Qu@2_ChVYI021L;9G{6QP`9}Kl(UjIMlFCnX7h0x2!J8SZT#0{#p7Wb6NHitSLnXoll~DcrZyrZm~#l zK+g74MC$G=52UNRlW2{G>>U9J2iO4ZbIclwjKoRNwC&9#dQ(GWYAhS%IhP(a7&6jX zkJcVdqHi=rc1SU;y1m{;Lixt(=9%;=#^1n6?n3iw1vTIVXZNX3WhTGyw^H;XAc-0x z)2#UdIsbkQYWj|Gqaahp^$r?RVq8yX-=r}mtyF&JDmSCi0C|oJRbqunAneJTG++6> zdABC5U_dFJNwwq5NWf`gM@1a5p7g0-?5aAn4Mw%s7}%SE9FSv9tu{1hjGy0xCtK4b zj4&T_JYSHrxMCs=+TyosRW%y?EYU93^mCClR}@7t-3a9gi7<)x}n%A5G6jq=+4h=_du{cH#O$a9NHy z_4}ujDevvz1j;)&5V*b1wE%;{V0zNFuW2AH!DQr#`hHK^dZvZY&r5@7ehO{^lFCpc z%sQTEjhn^TV&?W=u@v6D%9FO<8L-tZBYL9Hk79bvilPMzBRKOhmG`F5f>RY)T7#~l zNEiI8hb>s6!VX@{r6knX7S%GN9xYg_LisKb_g9+Y{#jL~KJLT9=r0v$+^Jgz|6%@h z`j)22Z-R2cR^@jG+`=yUg?<){IAt3;uL=dR43`Rg-JcLPFgVB+ z!l0mWfc=Fx*T)ie2CaNo47a%p3@h3YE8pP+)Nx6~cRaq}h9xqvEovoG5s^uAI_%@3CG*gpPl$@I+Jj?pRw5#2;b$x6^dvg*{!QsreBHJ>ELM`+gy@&>v%HQ?ENiz- zz3;{xNOAIW?`5KT4}^?+OjqwA_X`f0(Uyu2+kY<=OIN|?CMP!n+kKfY_#uftVsBk! zqae^a@H3+)@)^Zu^au7j)$*w(qUu@E(>OYDeP99=e`>q3+cem?5Iqb(QAqO#ujHDA z!sIDK72kS^NAk%c>|9v5IkB*te&ap$*`=lN0YW$;!G!ud|=I zkVl6R^??;_W?iDU*a!8FqKk<7(TX;+Ub{*G74Nj=Wz~^-u7uQU;%DX9DubD%NVbf! z;`1ny?QB*$kc!V6J$IKPe$Snlo)1gqSS9-%IE$$U72h(JV?R6=zVQrUbv?f}^Nl;k z+pAG=b!CJ?kH*bo6=T>*oF&PWLg%gpt*Lkk_M?s_cBnj{Q)2lx8_H_Gz?!ugGqqv{ zaoV1oNM-z^dZ?n$)$M|ZS44{7O9{n!ChxU>Q~Q42XLRg+XMATm79SDlla;C z@?6%8>R?YiRh?8a9yiiaYM8{YV$yE*8YlCi=0;ps9d9myj9Nbc2e*&DY$GxMwNO!+#3C~ef)D-!cnybsE~nN67yqLR*D8)$IEIw z?Kbp&sO-0Ac`Ez#ANanU3X+*jwvEYjWluSgiti^p-k_t@%M$Od>`o_|l6oC+Z|iv9 z%6<)Cr7Qcr_V8!{PJNHy784xiXuh&4`zSxrm3`7f-P=AVD~J4$Yn6bYK7J-OUs;uX zS_s^g-RvcYnOt{31~UN{P4U&gRQ7Z5cUDgTv&y27JTsY9+2gv`6^?2bp6+EJm&B}W z9zWsEi9#iht%}-N)G(w1raahtJv)XLHHo~2e)5cZ>GvA6sHywL=?}Nd=9RmG@(i4| zGiXs}rFQ|77An)6?A~{R4?D|KN^gG^d|oaL73^2DJbi8}dY#Db&(qW=>jrj0Qn_#U zXiJp4>Q)mtoBp*v9Mi2Jo^`G9oAG%Sda^-P2HBFwRK&6mcSgCZy}vAkb!&RncmelRGE62tw=>W=6Ab(~_yFpA~2omkB;_1I3>I|81>Y&fx=pMCf6 z@`6_anZkpZy_x6MqkS zbsBN!IbJen#%KB2!~O_1OMC}$0dw5(N~$zq!T>PI^|RYCWfdmX0f^TiGUE&V>_gxe zE7g{XxHdZ8&E%1+8kE?<&;Cc#s%E+R12BR?TrR~|6r28dMM{kFYxIf~?w3wTw~L>B z%G}mwy2}At$0VG2lO*7MoRFS~#Q|0KDKqi?0KCT_58r_GTur_F>>sazvpv&Z0N{rb zocNO}vB=LJiY>>b74Db(`>el;#2RFT$4#J0Qzf?M|@OaqRC_`;^yl365{iH`6RPk>jBuoATF2U zE06fx#O~-SK#NbU`|6wNeg@D5CgIGRB$oKRyE@EF{4M}?ydy8ZmiT1$Xkc=R1|W$+ zUYxY}RNGzF~ZwfX-atWt&5xwje# ztHJ+2toZwp(!c1{+-x7~7kaf19<#l=*A>SwV_1#E*5PLx$GUr~eMG;9A>=hYKdj@_ zWCmE>YhRqC?r9&L+r@2}ai_hv^{=~0O9R=W|2ASf+Lg6JoT(6B+!7YM6MebYd`4nj}4|BOv zPeOTxpRHUPL`i>&_#Y02?GKOZ=qOc5;@zJj{?9hWX{(64PsjUyibMfyeTrbWIl-y1 z08e9r1diq_>r*6!pXg7K1|I6(_Q>k~v;a``3W5U`HD6huBAG(q{uCLwJk?C@8$fO{ z0T)g2)xUg-Eci2-R7KQ|6F-xJ%O+&jr^uDR5{0A61gI4Qxg_SN2y5>s=n7s%di%k3 z&9=I5iC0_YI`L|&n_wzckAueo=5u?LcU!FoOOjfT=R0(qBG$cFZFRLzSAW4?T9E29 zz`tUW072`vy2z)ipYLBzK+1Ru3IqI1wAEJE4=hh1pjUB-OXPUToYhtrKUkhRArX~> zxchW`iMHDPA&1FiFaRSo$kSG54|AAI<|1y1jxW(x3tDBG<=O$j>kQ&@DZcWw)tNqB zee2T7O!opnKQIYr-XyWwYWu=AX5x0dpK|zl@wK}8HI-~8rz8LxGRTXQZmaWq6_}jb z1JJVsC*4+iPijg0`qWqeCTWn_R#_;dt)465uAb{Pyd{W-G6tTGT8#a z4h`~%&%zfClgSapeXQe4h|lyP^~`ep2*4i<;&Lgz@`%q=pPBTi@D^se(Wv?){3P=x zi6uU#qOtCkOVA2{_6+jywZvy(a+b+y2moVCaMI!vSGBUqsThEjB{*sEITP5F`1Pq> z0PNKuQ+!w`yu^od)i}~#Te|wDPkXxhukPB>m4WsS(?7+cq`Ug~1ZV`qmWNR!eiEft zNW8nNUlLsn{`H7!s^fjT`n3RCUHz%Iap)y*stAwNnm1lR}s< znbp-l)2f+pREYr9WgwTt?CRMzgKgPRmFfMvILx*>VWn4FTIt;Md>gPUP{cDjaf>dt;{vMO030k++V?JH|9vnNyAaxzV zEemb6)mJ9hr*!U_!d5_5@G~=KwbdUc)~8ISuYX(1$cvM1t5wfbfZH5$(FcG5 z8f3Or77A&r4IkyMo_VD&Yo|={c@1CZmd=eEg_rmgX0|5v5}55_hULKCBR)5uO@xO( zo`0d^R3!#j;`58$l|y3Q7R{Y*B*N;`5GASHJUcihxuKkPR%fB|c}9%26zr zwjJU+alB;C5})3QPU#B(E-{G9rTEGtKCk<9^~Y8gn(2n1>MPcSDpCvskH!X)F4xQSSY;2hjW$u)L|Ga zpj8ff#!Khj57K36mCc)f)he?^oRVMsI#yV9^k(%r^n4a8^X6Y!iR~kK*=Nm!)f#VB ztG#n&-ux@8C7i3qW#%`Qv25#F&sa9HFar+P;QcOswivh~S!3C{)fmgp;PGW0rD{vO zdn~(p9*bh`BF?_XP3k+A$%d)gf^v2uw&Mj(B_n9W1ogp7@s%}}iIa^U%i6<@%{>zK zwjZ64NjXeC1i@H~ny;*}Ogurgx$2o;u+0GztU~Yt6L8TKU;WEiHvH8Dw0~gES`@;B z$*i#~em72#V?Mtj_>+NL5(=MGFN%Ve*;317!{;A09FLzYZ(_BsnLO7Bg}2`xOF_l9 zE*0y;p%kDZZ=kHl;isWu%=XZkrq1v&7qAIcvCU@3_f9cjYwscEr=g-s_97PNT5%6G zmCP(n$?rVS3fpR@mOAn*4D$GocmuXg{GZ-${uj9KhvC2WK|Cbqh*zr7;y&dg%7@E- z<9Csk%ID&{CI*GlV(fF?ttnLmKl3zPKJS(~?p~Kkiw8tLQoWiOiaU-RQlAzNj{F%q zW%lgoIA@=O77vTu9%dn)Z_gwyLFpz62*dl7!5NG&#Kuz{wJv^hn}=29CL$i>m-!*C2BwVWE(bq~lv? zAVHj~^f7M-nBtSQ(Mx=|NAsG9d~_0OiKdXR7WmmFfW1e2-slH!@h*7YSI4QD46wv! zq_6n=P)CqzF5tyXGEdN2d>-@>pJ#uEKM+7_4}v!=v?V@)d6j4>m-chSo#lARoFzWD zN4JobqUv|V`E78=mk^(s8|@~OssO}jkVkyx&dD;FWFW4!jxQlTe{X4Gma8uS!x+Tn zQhenRpYBpPEk2Wu<0M!HEdgjPlW^uu5=(q0KZH}vk$Lqx0BKYCL%pvIh0yHT;_& zmiO7KM}mG3XooZXEeC`_e{=5Lo`&y#JYJ}ylwyDSaQ8U9r#UWz$JL zPJh0pjNsHK0H0%ma_pebSJpUPm7nNwT6Rs>ZY87rob&^rsywnDg`Y{ySJpT!Yo7Et zJ?x28Gr5+4bYuc9n&PW}8K<)|F_9Sy%q)w7%O+&jI9(r(nZjocKpPmyC4qO^qfL1P z&HFiv)Q5O{0Y4e1FQNU)@XJp_WmT4$8>t|%5mceDdIbFv96XgGBPf@O(i^^-X?B4*&wF)&vhAa11f7?Sh2PDn3#6Q^Zq(c@ks%mF7OMS^xzpmzj&@dkZL~Qi<#txpmi5`T+n(1{So^t zGDz)3aL7ViU0`q^R^fqOokrYwj+e|?UEsipjfmg5x`QBKt2@3#7x?$&#wL^M0K{vM zrwjaeVq=p@CgR%Y_!3><=CXCna`gvb1cSI-imyCfARl_+89}ck$D8Rc2WTCWaOO=C zs|y^rsfn5RegNKMkcY3;1zyA^F;Wc|0QjK$@qEb$pKqk-(ZrkWrwN5_{CpTP^$OeVbm=&wN@@wqua&15naadUNi z3Gtb*-C>q%Jpelx#N|?aGLGdZ;apnVBWT70TL8$$g0)DQq3)*vc9=3RIZ6jJ1(FJY(&OL$F?b0#2tn zrS5GYHhQdGcM7WC>v()eN2%Tt?;dLtY&p>XZHN9BKOyxUYaayI((jI2S|~Wx0^kBB z7|PLnWsS8X`H3ED$9t%I+qYfA0ti6WR0MM^YQD0@+9^Wd*6)V5X=^686Oe;Uz(rGh z^)F-XpkSCkE&%hhMZskgGHa~uA6G#*s$f)OS^T6V=E$(&MQ+tEqrMvNdI{ggnAkkN z(^J{W$Lf&E0!eEoyFSdTvbVm0{VQE~JWxleI}-1%?8#YBvJ?>4P6yIEfU0WMbre9)m9eR|PGUxK%aVcvRf;#!+Z zH~A@V#3+IFb^luh^V}@`b^o1Tw9nk6czvYL+=Py}@7;5gg`3T-lJwVoJ7g;;2YJ_X zlL^S8o|`m9@_KI40@ukcZ=}oIWHl4GzwUd2gUn5&QeK$!Q$` z+e>iLgUjcG6!Gg*?*VW^gUlg=g+hjmmfvwN#`T)Mt#ZCOwXMF}OMJMy@DiWfcj4<+ z`IRuLhM%oD*n7lhT`7F9HNf*s9j9^_V2RHRAMts-LYg4eK)@ejk~V_Y;xpbye5T%L zEFiTM$W<2F5}%214Q$V)-Ho_?950!(#Aor(ar7-JL4At2^E$qS`0TC{Wiq(~fZAm- z5!YY1_NM(7Z!(EQT)d7iAwCbBY;Kk-6M!}h;&Lgz@`%q+AMt6sxV)L}D1fFg31{9U zvBam(fLJr}kcY1|H+laVhskL_0PmIHq{Zjk$?m$i0Kg9=I9cM;x)t&3Q+9|$ zD1PSakr@XIg_rnny*7Sh6&xxt2+n`iGYBpjfKT*X(6(jzZ%08G^o*k>wysuv@pzby zQa2>tJqRxQ0-E5*5I0}P`woJBYQC}tLD?Wi4}u-qq?yUpKv`4qlcFiU`jM_~RFeUk{{IO3?l>Ed@Bf*rT)Qk*x7uRWbyqK!6)cO@qxWo(=sik;L`g&! zy+s!_h#((bghUV_gdj>JK?)H>Z@>3@=9zi!y}#FMzkkeg=gxVbGjq;!o^$4zrx8d^ zbiC(-!y>E*czmz5iBd`M(;Yeq4$qcKouK=j-}TP#G5p5m$HpeXyNj}m30YH#E08TC z_=vGdg6RiU6;Ig8R~+jv7nu|Eg$r=E6qmG}6ZA{>l7f=phqp=vCBf*_&I!De1gpK3 zPSH&gl-W~>um%{rNrI&?M3V%IpnH=9V^R0zUWBwTW@CsyEs?a8iWmagD1arOLNP%QWAiG$!jbguE3lr;kRyz{3v81cP*(mnf z+tmd3x%?QkHHID%#m-EQ*Fp{eaM*y{B>4A)crD~AXg3W#B#L#DQA-4@>fgfe*FN|AqxkuS!@)0e_f33yfXkj38K9mkV%4Z-AZan;{cczBFRLt z!8HnNNoxVv8Y0O=vB-#c!QCg118~}abP}Xfs3_KDC6gfS8eeaHKAi-M>0B+9Ldw2?4<^e| zg*NVUN1=`TTrTHTAh`(0pt_Ek_r+q8D{uD-#KZADN&_mqk z)q|C^5U~db0g5BUB*EidqqLA{&|(ce#Cpts}YU<7x1F)7LSFTNhpCp#i8hi!7@eoPIeFoqyk;)b~0Js|>$+%C< z54ewzqX7`w+C5h8~;*)d!2snc$*njTKJr0QevouJNrGtl1=}UbtbB z-~*CP5X21sQxP&LU~3P~{G5iTya1E#0t zpl*uICc$k}>MDtHEI<vsRp4L14=P zLiHb35^DMGmC|{-<(E27M>N)X`uj%uw}$gS8NYE`(P;DZgsib59~wSHPcj2ujA`?9 zq5e@~1zWT5b(!4Qlhy981z(E3)vZ!co}OF0T2P*zR4bSM@8s#)8R`b*>A4tI*j@Bx z^7H|W$mHo5^lb8UNz{EKC*nC^G+Evf%i%Z~_F`$(sT@B6#+NArKETI!w2ehV!9ujDy zvoz2`dIHejfZROYAyWe_WFly@4Lu~#9=Tpz>$M4hodi)Y$(kEze^bhtJbmV^O1j^x z0No@D2d`Odpl$MX5#4!O?4L-K0J%2M%6BSjNwEM_A}BzT3AF2m;DQxx(F}kNA(Bj> zotCkn;O>({02pCFI#1InRG@9Qm3f-7;-~tnX!kjOD8PLf(E_Gad z@vi$+$%S>0U(tTc(Bz*4Sof)vHPC&!V%djCnf)LZK~X%n723E@Nrg7VitQF2ii144{@I&*&AyiYXR6~K(70A%Gy{9 z`4+U(h92TR#qp@W((4ugj|ie(k~PW;5=Od8OWZjZoHaR;bbzvy~*$qWJ%U+o~}s-lc(E~Z1VKWwB^NcK;;r_)>&@W zY@SwcRyBFL%+z?@-3dUhkb$a6)>6pR1wVs1oxqPlLQZ;?yFa=p-O^Hp{z-Q)gVD7L1h$`1gVvX zUwA*Yj$*Tp1Vk}iT6@@lY>_C11Zbs6SA(-r z>|)D`g8N)1f!5N{L!#K&c9pb{egM2>KyDN(*|w4vG7Yo^h8_~dvi_Mz>$MeteFRZ2 z$(kF*N-5<`6#EIUdr*VF4$vL4aPXSNMzOU0h3-5(_D>F!0J%1bz426NNo4@2Mo@qx z6UAzbmReFP0Fpx_nJD(~?finfPrd`dXamwwj837VSX_4=ShQ<=T(ubOKBtZaxDUfc zfcr#`$s*)22>XEoyW#ywuKP?LUQNgwXuoS{vOfXVeTHWWbe|gErBz57brjcsQ9MHw z+PF_2g*NW9WgtE;3XrUiO%sc@?(@ehv0^k|=hOqVex$3xS@*dSA1mf#Tyg?vvkX1N zeWtfAp@nP&V5;I8CJA;(!sVW9 z-(l$=MUmtJMR$_m+??=V1XvxiOeD^iXzFd1*99FPr%Kt#X&1? z=)t>Tb`lgf3p8hY=T$h_4&ZKN$OKuEHM<+sr!UQJSn6^odk^DA^nrlNdDyJ5+^pH% zuwF4ZyP;mm47$5-0XatoswP=WVK>~n1C5>l6AlY$4(g`Z>~7ezESr)j3j!x@D=mx`5jhq$Z@|TvG{ta{=RJU?2Ljbf4qGG07QiR@ zbu7KJ6(x-O*VhgAFmQUg;|D#xtg}o{FVi@`L!IAw_>KE(X~6Wd;0r9Bpy6-yC5uo4 zGrhbovGfUfc(4y9;KO5GTmJLem--%|G%F!4w_!h|4s9o z>E&?@E37s8GSkaP7?GJ?R!7ffdRZHF-?yjm3Ms6yY$AhmE#v|usOhCLikeiO zUXnF;e(M-fR;v-*vf@z0cPc6f4O6EjFhuH)}Rv zPE-s|z`WKjm+tN{AYsr~sU}%VAz)S-ii;Srz*MswYRqJ_0rQoQt1F4JC0g4PNKN!n z%kM3VuxHpGzU2Kpy32)+JF7$4yuCVbxs$d(H! z+k~=Abujg3D?j9{Ha3*y8m>PTWiIO`x7TknHBT<_qk)0hqMdxlP+jtC1_ovW)$?y)v&#(nNP&5rCQmxpM8PPOhfKw50X`yb&VFgt7%UgqHLk z0Fy!_nNYUxaV^2!CszTm-hgx{qf@9*mi;RpSTsq}Us>XWKTRV&1)p=l4P_f<;8T)+ zLcnbb>G&}`!i2IBcOgIZc`SXPD3a{1=uRlx`DG~~V?Zly=)s|kpJ2D4tY*8Li@$CUHX<;GFLERLa4P^_@msb*H0f34SNKHhJ-5tU>_uD)VD^j)5+yq6< z{rClXEsPeybE3T^=YB;_mcds=0Gb>KRC7PJMvyIlPhKUniVcbqb{IR%H>FNv6)?B` ztOI7n6FOi%_+I~Zcm7YnZ`{vlv;nip$Vichh8yThj-v)9U=FKaUU(o+PNd1Wu>o`3 zr%^$3zgNo#&;4FT2G9M@R8Jc;_sicJAB4wwh?)EC!?41dqc0OM|HOz)z+8l$O~71^ zx^L=IKG!-d<9aeU0rMCnxJFR{lUj-RMSJiLe?IE@&!vEXsrG<7_nXSw|8GFtEfi1q zGN@{o7HhA?o553E#QU#M6isF)zy{2M;ei433hqrJQdYtyo-BD3+62t(3T*=B$tqC_ zB)b9G%c5<-n?$-AoDG;WalXlYE?0rJ+0a7*X8UEew2&hJoG>6aV4m2P zR}1+QwEqk}Bw&7$En4f95q8XtqV$rixdAgwDQ5y^i-uvk-&z1QA`1tvS!}>8`EhF9 z`I`U?B#8EMKxXcjtxdR=G#P-oA(BkMyuQSzC4C0K?hr{PVEVAA&D|%@0PwQ`>3~V6 zPysXYM+QvVHL2PkwS<4-!gw|JOTOX;%pVrxbRp_8*8fmEO&7x>Ou(Ez5Ni?@(OlDz zWP3$-0%nQ(I9OYOmTc(30kZ^ONor1(kL6FRaB@7rGsut(S&}syFl&;*1kAcFce3~E z!ga+GK;vgFZsFoVq3mgk@z^CX`J?&nA>jN8NYn36&j|aR?clP*w&K zR47wMQK5`liTFu13hGd{^Vfh-MsEq2`+b-dm*_r+xE&~-QgC%Qlzq|>N&b7ZpEopF znE)HgMh1nld!o2PN_iCz+@N@>E3^q^1A;=?-p)}9BufKX(V}fA%Y3G$sLvy>DQK-o zSA(;mY-ghyg8N+d2kjk04+&-Gf32m3%m!eg0lA^J1+}BJVCBpJNIkkMVIYkXpN=cSmBeh+XT zhKm6A+4&J32f71c4=J!CukG7Rm}Wgr9U^|itM5@1O?D^1x=%u6U?zO;eVq3JDQjX= zmn?l1+PF_8g*KUR;6QxPlIVWeyk*hWeF_(i7sGjFdm3o7Nmqlj?h_poFZN-~auaAf z4L!ttqRLg*LQVs4-hf>9NmshM7IF`?7lt0x&54F2>npmG1dlDoSwnl!-Z1pwB$y)-Lo-RR#(AH@$;klEB|{U)lC0Szs6L%x zl3+oXJK6gm&PrDTDvx0EqvdAJCPDRuRdcJryb0NKcTWHbzwW3eSxX@az9^0*pHjfY zTMp``*lZH4+q;yKC|d#4fk0{^a!fv=P6jf}$Ex!LG|xv-XAS&Fh&mZyYXtvkLrY~K zen3;?s4x`YaQMiRu%H^gqX*cT|Bsoi5!5k|S^dv3axIo*=6P1}7PEVRy$EDh`&4y> zAtn`KFMs=R9kmRcAQ2^ah-{7E2RI}6q0Ydr_=!%b8dYXfYJ@$9)A>?Ahd88rem;aR zr-}Vfr~8}t^lvBU|GW5&`vHwM-QU<;MN~t>a`Y;@(TXPBms=Gr_OP`Vy@VT^?x#-+ zUh27k%kFAv$4U1)*VNPhopk>Y=gx}m+!QfzpZ?gjIdj>DVTEPD*iE{>f)SZ?KN3Bg zbUzk#-;(YLD&2P@gOlzLK!R%&mF}q(R;|xR=yd5W;B35})4h)RERyUDT2Dg{iPuR{>9mk>08BI>H(oC} zQcMe33ECz@4~f?$Kd-3u`WArG1W_-^nj5cIDCJDNK8FCR2LBYGv@pLKyk@cS`eFSj z-FY+su>`qtZMttfw2qdP2tczCNhV(3!RLSJ|FRbVLqa5(cwM3~PH#By=>W_zAQ8j2 zCB^1TRbehQP&Z`LMO5_WXTC)kCCAUnTUdyNDFq|OT z%K@3>KXGO~EomkIi$f$CXFAX#pO*AL0KN*5WSnV4n#zK^PhJGzR|C??pH30rKKx#o zzv;KQe;Zdny8PoN|0lA6ka=(6C6FkdCOkDT$$!i%X>lt%niC93wo`N``FCrKpx*|x zE`}bQ{L5v+&`k2bemh3tNGYdReo{Uw!=5B>z<|ce3~6_bQ2HfXZ*M`N49t zW|Kd^v}m8~e)^fy-9tdU&{wG@SxX`LC%zdiWC>vWmV>$}Hk!)6vsInUeDEtpSxMB*A+V-#OQR8L1J z7Y+%BJJ~DoUc|eA%3K(FVHC~Hn)N94b%=UR&-CYtO?Z`mYBx@;nRPisM@d#ih zTMp``*sMn#yNWMsQqC%D))Pog0xTXzC-Nm0!pjyhxjKDEdsK%A?NNE1->J^;Cj7=p zY^+D^8JAupU>~Lhr`$~0#-rZpP)!tM>vk%`jrFK9>Gb0h=6c%mN5St3SU16WJj0n} z9lqfSnq*b{L*vaP>vxPOY#{nFldMc=Fq5pq=-EuNj-zhf@CX^4N!A@Os7aRVhSW;< zb0p)cCET#uJ=YB%M&a7u+^ktQ zoSBu~IZJ!FbEVVWodo0;GEgoV{OXQ=1%~pC zF|4ph=*zg_Tny2;VIb^an2^9^SL0RL5!y!Srp)!h!hSW;L&)w|2kaG5e zfaw{%C185?d(OH-PKCHxD4r(80w1^ZNPNs>9on}Unrumcou0Mpfs+CIwC7pLX|K>`de&H>&0N3O$OHwF;rH?V3lxpE)3eF=9&IxCWl7M=kgf)2r)Q^Dl@#3P zvH@tV3_WCeR$)V|7SbPpAqM15&!WR)wUFtcEj09y>Dh~>+5&PL0Q(7|UXnF;dR9v* zXXg5E)=jPZy#df&vT*R4#ZJ$1B^KA6XTbi+i4q{!&h_7ElUqwF3qU+U0g}x0tmOlr zmed-6*Fz+k=~?9SIKkZ~M*uL!fb{f?PNAk}`MjLn)2{JT7gpD^`_qpC+=t;JzmC3v2%=t+HP?MgDCLa%q^*}-_gfUE zFN>lEuUV}73`kv7cis?y)&#k7t^166Er*sg0Duu8l8pPT*jiRgnghVH5J|>;Zv0tZ zaQDew0DNgc+I{F00q#S)`sY2t2W*f8Yd&$4;I?8|+Is?C@fgEhgNSD5Vx3N4jW|1+ zqYX)JRCFf^X8r=t%v1!erlAKX!4I=CD>-wqpYO#foO~1Dfn?YUS&}uI1m}~%B*7&v zcd~chy~^TUK;6?`ncbg*#zW!7v!B_I@r1vtP{vF`_ zpMl@F-_dB3;8VO|z6ly;Lblvb*(M1V*;rLc3eMd#SN`q~n zu1SMuAVQ@V~d^##4-9yKpY#N1Ov{7xP>U5U3gB!U8R^mFGk4E z(7xNyf5YZavV5b^CXRij&?b)E-sDvv8TkyJfTGbhjx}x2 zKzs*&Sr)X)q^rT%I95HWf#6=3uY%Um&_m+bp=J%Wkf8vKG$1#QEyCw`l#scgEjRR# zI9BGh2(8y{0KOrJdP&yYIQE57&cv~~n=9*n{{iSRSvYvjV&mAw&ui+=vts|`M+uN? z<5=s%RkWlk0MsTZK$3}LX}+$dC3OOzM~Ea7$A%NXIccg^FXXsxuAJ zB=I|@J0FO+@G`)C7%&3dC$=UW@-&2?sK#ua;Qy% z&##wNNLlCwzKxCI8KTg}efBA|ai8#?Dk_j{24riCw(irab6PP1{Bi(jLrGVIv+gsv zV_LBclku6C%mDPvVPc!QCeh0C3oVwENI00^Emo^|zT)O{Kwlcn(6ZJ2dlV{}CP` zMWpbecowXIE14yUZ8ve>W?nQGGbFi6(Va9{>PIZ&#DiAP(1X+9D1dDmyqv0>!pXh> z4=2NVl3BB9@H4iUG&sTKPWF!KTU~qrsN8_f7nYkfn+9ho1}6>r#^li5T?FJN8K|0M zErm3gldrML2xpe_Jxpb|v1#z+ z?S?_;g8OqjuTpl>;3K@rOP%{U=Yn<51i!gy#4Gwbky(-$f?H7Wb7ia}8Js1Fv1n6CP#HxfL24!9$8{^Fli)IZwak9zhTako#im}wZ5|6C zZYhdqXOF-r_V!i0c6mG6zce(tp8y-ho(Db6xME6Lg_M5){tsEcR%jE&?gTx}*f~wK z0!e)CTI5C1XdA_LFRU$&fnQbvEuM5WI2*+}&#x`G&t+@SUN`iRDAs&q9W7)80Amcu zjbaDa*U>^2fwtPvL!#K=pNnX{z69VHLDWmK=0>q=K~FOlDN%&WZ~d7i;ZHX z`-kh!b7B7!MhTE>li=a$MYW{r0MsWaK$3}Ko%^NNlDYuUH$;+&VhL&T3hqAnAplbi zNJlX`g^FUm%ENQQXxI1wN6KsWxso=(eHbnR+~@6Av0Qx)!Y)(b*>xb=%ZBHBVoB^j zw7)bod659?K5JUQn{5*8m?2RiWwA7o!iVDdRiTahEK+Eb1lOicR3O<3$PN~5-RG3_ zi$B0Ghk!PMbTv5ZKD}x+5T9UNat>(A3_Zks*0yYMrT;P8w1Mp0UCq{x8cd9fDGV3J_uZaivG5Y43wNv1|`_Bj{t+BhL=f!4^- zgOlLsK^U4j7i{oLLxq!r03Jn#aE_ccn*?W)!6d={E_br`uUQquBtYe7*zC64tl1_NF}tj=PGJ-`iryHdfOCVe3 zplp)_>n_SAcCs}uZO4sGg3ItJ`oIOzQxlxm6*@_<#UJ`!M)SztlLPv>7W14-fl_)w z)g-}27*<#q#%_|}FbvTo!8r76l3)eYeGPKxNA_MPgOdc?L4rzx$|x!cQY$>D(4n4) z4^J$qL)rTD0ildLJs^}_n*tA837P9qJo~7x4P}@2!#nq*{ivbILj>4R)+i{HZA{Ll zkn%3zkH~UVp-m{O85GLquP>}XGCDj`6h+Z!8_K#4s3%T>U)BVzF6nA;Hk9q^S5FK; zCo&nd-i96$%EI2MuZ4U7z+?k*Ls_2T^|g>sK>N(lLqb{2StYbyM*%oP5cQI*xuL8= zP$>If{&KqC=K!UH`PJYxiw$KzMP$^S7XrXXkSo__LVu$QT2eg#nuka-p)5U}s^Gh~PqIZ+gAa42h!Xlg^5e^C*IlhpyPPllHq zIcqkQ4I_gIWl1h~vRACEB3c0|M_@DFap(WzehL!#I}P0DB?4FP!7fZQmyp>Y{4q&H|o z4Lu}^z1F^}*6Skx77;|fBx`OIyA!miH7<8P-S0twz9$PUt66LmOUhnAcm4+e_Xu+3 z+9+1{UM(#t0y{AmN}wbY#omaH){@ErP@SLvNhXSY>1`yq`(zscIvJ3TVsr`>#TvF_ z6r)|^^L>z8C&A5G0^En;BEWsx9m1q!Ere~N!0~vv*>#^so3Z412<;~fO->=ex=)qN zfor1OhgMfe`4I5uWSObZ#(hdDw3*NQCe>0PStK%2#G+`lb)QouGKl%$mx-V?BwY>8 zy3d;88N?8DBHsXQfT4%D&!mKSEo2e^9~qGAKI3B}w2<|nZ8!7~_j%kkOzU+5fC~gs zFUgwgKG8}!GoOEFcYWQj2d0leQG?ek)_sGGn=QCg4;;ba29 zjmdBdvLtIZ3BExFlLU*n+{xZ|W>yyM0hOb%nQXaPvq>;kF*r#uPqWgxyDfkmAOlsC ztfi0y*Y9eA^?zXQSq|!^*lZF^+c~F_C^Nu}Sx}TFI^ItTXA;~7d-*0F#>o?sV72Tz z31%Llli&~U=-&kAe^>m*eSt=s1n=ET5YcGp2ifv6EpC$F{FriLB3pl_GThiC*!Pfg zbFz~JyS#L6UUrh;{y*b`7PS`aEvx@KcW%9i(n-XVKaFE9M}^2R8K+?d04$f#@xU)>?6 zFN&w)X8{gbFB!A0v1p!ZNV2A)I}W+LBBuB&K-*~O!46p$VC#_g(pFJ8c^cr$WO$Wi z)~rLeVT*CdPA+${xBJsf;%`7@B!-?BMRT)e9kQ!pa2#^!t|Z-E0wB%FK-DB`DLCY< zomg^y3z+vU2X#|y)*NKLfuFRx<6ZWP zoZk#<^nV|I<3^*=I^>nGM))dXYVi`XRMxkS+&hb;L9ZUHL~+AD@0?2xMfwhmciMUukFZUFZu!)B6MvktkP zEyf{tyWGj%v-nKiNI>NZY&KeM)~rK*qZk~AjNX?^cXt|)-^f7KBx@-+Upp(72f zlO08KP&dV99WtVF1|?C(0aSrNYNBmFm9?<5Fs1LC8`>c|MQMkulRk}d$TZIHQ0I3Z ze&hZMPvbh|?PbM9OEi20+42-68i%Y}Ia+LB>p3dJjdjTLL!5hA9fw?V!+HF}amZf_ zIB)WBP7~(f)hB_sfc;suviffv@;HVS_7r^?hkS$)8HfA}JsXGo7j@sM74_A!e+wvU z9P$DrD2G%=Q4UG1M0}c7S@k?`UfzIt9=#=Cp0}ZS23-GxxVb2v)%^qKdHavxhM`Z< z{y#&LpAul_dH2T$&ht*bUr8b56~M2N<#UBL^SnP5+RXDx^-NG88IdPaWJS?vJI}lQ zW;U?{{L%+ndD7M3>^x6qEh4zjWn<9V8hXe)FB)GqR6+&;Fx-ILd0y?EHb?2F|f1*$V z)4O5fcLKuBP+-_D5bavo)(HrMf1~|@p~*}HSodik zgWG!Svj4$1>nfzopD$9xpm?$?v~iy%3T@nH=8y&oBpU(Q+@h`fRJvY6HU{@-b6D~m zHh++Ts!7&TNP_hbm%>FFSSJcab0{HXvq^CK;3`U@EDw;MKx(4neaVJAR)2;${o7CA zwTv+K`GT&oR~^N}6TQGPfNYib1sFT}4?LRr0h&KDB$=S-jm-GfX92!Sh9)GlW{usFEymdGUG8LW%*wjrA3$X`47~t~=4Q@dv#LBFXw9{| z1eQI}d;r^-E`yQllJ=n+Q&Qb=ffXeCEEVSILS=Y;_7#!DY6vX+9ek6pplnAd>m zZ#k3@vRPx_E)b<8$_Z$lMj$oO#_skv58pEAc%?k_UZg&JM-}})4&Pq)5o^yv(nU)9 zh|2fD~e|s~JC=RG>f=xTi&6@S7{))lzs5d9q(%roW$aFGLHOX2E9@Xtv7-I`C z2P_A5Q*73wik-`=B+3f_{X!r$(H^yU77yRC;PHKWvAkN%3oD_|Rd>zN=c*%|-Wv~IDjWG0JhdkI6Rx`JY%&zKJIJq3)PsuPJvLtKP zA(yblIOGwRJK0J#< zwPbtuNPw~D-N4!2UNnDeNOGd0JH~#Q3Gef{1lk{l9&GIC09#`xibTj<|w2^K$2nrSUJdoi}b-jCJm&K(#{g{L5}l;#pkL zBYL9sEkg-TJPXs5Rf%T`$TJPZ&KnONZJ-j*dQd(i^(jT0#Pc%ACphwNLHXY0brMhJ zFOmcgjO(D>w7lxD*RGcXjs9mJ;q(Wt*EP;{y|?jJ9+}UFc!A=1eiU7rdE<>4I6tg{ z=GulN(;Xw-alJmd6NPLGT3164cD;-MTi1&j)pZ6-HyL$vkI_RrZldPrSdVAYq*|#JxRV@c~Q*74t zjvg$gB+3>5wIh(4Xpg%78ZRR~g|U6%rE$#ztMp?_>%{ZT4?6LD==?r%esiBf;^~7w zHt}qmS6`exnMQ1fY*~x4P2zcHkq{@@T93MNV-wHlf1S5=I*DgND(5yaC-IaT6&93u z?u}~_lz1K=){ClUk+dm>6&8)Lo5b@DhG-H`JM?T4&uggrGHu6F0TRz}GB}B+BP1vf zR7O#WhgymFn@6McBI%y80jpf}mVi~RKU*LtZHKtsD4z1Y0vAaO^g;;z0qqwJP5KG2 zt6YzQ?xm=IDX&7x)MaoJD2k_sLYq~tn+k1Kxpu8hr9iSAkX0<&u5y*fLPsL_WpmKl zlCB14SGjuOLwMZhau8^v3_WC(>yvy9wUBuLEH)r_m1|y}hFZw~K>NzjLsq%o#_h?p zfV>F64T7kbWX)aW`b8;cR=Ea`sI2?V1k*>MsKIL%yUOKRgl9G=w*mk)2y*4xMbc(7 zt7u8>0O%GX$*gh}i^-rRy$is^5J_g0Yhj5dg1b+y1Yn&3=~XT|g<9q6c@~}%#!f$_K_OyI}`!HMtxKEnNxDV_<2zx?-t=b2=Pv5hkWiE&LFN&hc4g^^D`6g%J8vCd_ zSkD7e*2X4@EUzoHai3iZZQSSAb2$`94g_+TMO*jjw;)_}2fv&F+C0+L;H>*hm>(_< zV9at0XnPDj#Cm9qx$g6Io1$9C1JESQq_y0SbSKijolr=`wn*h|yeqHP zD-Qrg38G$-HP?MMDdmj&+{1H(YVb_}YDX3hUbDFFb2LnMJ{W*^33BCH_vu`;zLqo} zfR!PVjQc$LF1?nt7l3a=BpLU4^G0RC-6wwq;12`R?n9>ta39*$-?uEDp+^$zQPE9; zk{#Zf#nkHeKr68tzrMw|hoxxOLE#ft7E!MQz0)V8S81B&swpH7o~pQ*1T~`ff%kiLxy~oe88SI^MVF$0R7g<7@MJ1C<0n ztfZ5myrz@jSHJ7u^3MO3_>EhGMwJrXxIhWau;QrB$#JXj5x#AFR3dxHVNK& z`if3!P7=(L+j$_!NrHKoJD0MZD}}c+IcICmrR#++^$mw62~NVW!ir$*CJAoGh)fbZ zg`Q0kJcqh3t|sCNlHe#ZI7x6GB&Z~)jG~etwG#0O9;`+3mDUHU1Vk}lCLb3 zhqy0MJd5Dv?xmg|R$@1uMf+7llgkOPQLJ~az$kX{VvIt{u*#7l6N+bzLYpY|nnIf> z*0OL01(KD4tZvaZihUj1Ky2he*9Npsq^rT%D0Zn>1HpYRhk-WM&_klw#qtfckdFaa zVL)ya+g-My7P1Gl!-gJ`1Sj3hsr9-Fz)gavmt@V2Vl9+%CW`rIRM!1wgXs&PsKIL% z8^xL?WYwMf0jNWeE7wM`iJ#zB6n57cfSw_eOcZOLkWEV(2f(xtNhXT58~uvl?vrZ) z*knLDiqR=l6npPaCPCUYJ|QYvyN{?A;64l&0q%35Dek3s3}G)RaOe*p+AD?WD&Ya4 z?0%dCqbQmjLx6RkM%nQCFMFkM^!`i=DeD8?ge)H_v~i!B3T@oyj<37|$)P}wv}o%- zi3=jcOz_LOpnXib8k}{X1HK~S8;n_g0onmW4{@JdyEAGbKLc>pfL!-U*qKoac?w$E zsz%G;BskjZ*Lp<*5K9pClB~JzQ$Z6fI5?fgV!w9eLfs6b?3tY7)OvR z*SgQTdJVOt#Q>}gk!0MbVMi@ zs3h1o-c5p^Rlu8RiouvzV0TUu z^cV8!?mW;n3yM-rvX(*;{GbJvAS(b<$8u0N#b%RW`VwJEqU;1vR|2Vtj`w~~S{$os zz~evqInF9Ev>(H&#A!RN5#3D7PW zda$vdEA5>(a}3#9P2r?K=jl;AsZPQ;tXX4+vBelW6aVPNxn%Dr(@Kh5fXY{}NwVCm zSz|{j2FKX-&sEgj^#|lVGEg+H7;xrcvi}f^Wx3Uya2fW6)fb;%8K1!F90sbKlcJ)fo}>zRiN^tc=}M9S;!f5 zBwmz5Yjr~jUdZY2epL`jPcM@Qu|LUO0L&Q3~DFR4Y7n??8*{{$h`Y#NcJVegVraZz1?)O!II+BHh*DN;rcimV?cRmz= zF$B4CZS-Gwy`Gk|2!Pcgl1%hpjCEsWi!T8<79z<+|5cO93hq964S>H5NJoDr+V!G_u=~y1Kg*?HVCX#3vd)q$1@<>$-|(5xJ#`$+B+GV>_UKb zpZGF?lZVf?Cn}`;0Px9V>7mfZeaa}bnLO-T)mVY#Rv>p;v~{2B6NTspet8pn3`IWu{P$W=-A`w>8k$il&E7VAFud>M4-+X48JAXlz+pK_Z@ zX-O9VxE3PGxX;4!5n9qq0K&1i10@;vdFS8qg1b+~08r9^oVh2QBEWrUSO1Y*cr^j? z+4?$eKHK*tRy(_aw-@>UcM;BDCJ(Rg!5v8Np?Q)a$-Rp1#u%5mYwDpD_oX<7@ zY$p%7XCx||JPGj6WH?ANYc`)9VT;LUCtU7i@Aj$r#UFsmj2L=u6wS?=&1Yv6gOkrP zAF8Cgs{u$;GEgIxk=oim|U$OB^RQ8}{~oiS+fX7F;q z{_iZ_966LV*!8+q)Bl~to5Ls4tN&)LrXNP6T!$kO{@eL*AR%fcC%LXuq!bqKqUz%$ z7{&9(V5?f;6kMm~ zFf7l zrjqT)UjmH1qAyN)R-k#KA;}Ah?ijmmizM914cc)-4>tBM09#{^#p4*n$$QwmAj5T% zS+mBz$rfYmyDoRK_v-i@B0WYe%VHCcqPbbK#?G!79Aoc(SyFe`1&~2xplXt}6pY>F z0&Z-a4a_Rbp~g%$YwWvcGAN01FIo>0NKLe{zbTa)#(o1z`I}{~t~|SHBiGpb2UWuL z53oET+m5RN#_pJi{hX;GuK%DYlH8~0jB#yr_aWnT$vAW1r|*vv!^yn_n{}3(HEZmPior4Vl3ycqcP9Y3LI$cPSxdp# z>5Jo@I|=JVqG%2^X0lmhAHbypC8s1nWeB7u+Su*#^ZfD@Ou16tc@xirNSt35XpHmA zA~O+N?D=JR{FC-Zf!F`VU5LXW>0L_8@hha6^UL9b3W}L%U2G`9=a&uc;oUaqay!Vo z4aA;bZWvr$onM{?^k8xHdUq^EvLz0UX-SMc0%Q0ab1KJcr z5B8`T09%jR`cF}Xlb-{;mkcXOX3cukI<^>(+U#;CdkRQZq}?vZC4DA zM|FBHx9+YWbS;abRFkZw;8B@Ur>Ly+z zi2;wV&<31nA+0WLs?%(~RXWXHUa5bZI{ydYH|`)BZJNEYBSHAka2c}Y0m?Qryh39N zipOj{OkKILY4%umXGUjcct0fsrP-4ICI+S1r|F#=2b?s!C8oGev(BqTS7TUVNf^7C z;hjfsW`=hKb>9aicqhcLjB^NbW_Vk{rMxcr6DXw8G&Ov~Vt6^4G` zK8)hIMs3W=#nO*)CVCOgHw;Pst>}*H&D?>k{|q#5GlwiCWO2#wbDzON7?j$$9S1EA-o-K^TqmK0>pBy2 z;e_EChU00DvYn$buJhxJnz;WT&4mm}?pAcibq@WB`mt)I__^d&N-;qqq3q-i9TA7<+R| z*Vtndao*b#Ed9y$KRA~)_UeAPxbQxjryG*otLTog$Bn{Fd=+S$4L#V{2kEKC*r)2` zRXBMD;9tmagk;vNu}`qY82gONo$UPyCxy2Fm02~a-p?0Vqw515YoLt*T^ ztzBb(x&l`a&VuD4+2;HiVC>2Z5hCuO`I#ZfMT+hi`;|$^IFYUJ{5y(b4K{WRz}DD} zHkMa7nE-HOGAt*VHEZlOY%#{(=yE4}r#8zh+5;*_V>8)uvu2IGRWUfm?pmp`?rsYp z2gpFxBx@-c`@)y-?B9U7XE~^wVzb81^mc6}QD%S{v!EzVw6V+XQty9&y?mv*W1@p# zSgnl?iZ{R4L2#q+V&9?##IxgRML)=v`!Q@A6mvcGh>2|dn#yov zgW^$dg`l8#p@8$j3+J?VWUIKKd|U)edVxH;;t`z4}6&6LRBIEXh zxVKO|H*jN@yKbE3E@lFg&_2h|&82W_yza5j0LTdp@*y+mmgeJ3uy*G zYXfpGJA62xsunT;v=N3LvTmI4D67_M4gkvtqF$0UciniaQqHU!UyU!O`~4Q6(`2D# zHH%$0KGwIM?)(-2j|g(*+I3@p!{S;}7VN})D1nm9x^dT`^|hqR03;9;Ajzy77p$8o zxcg*B0J<2EUN@#wsCDBer{Otav}@89Ts83fBj9idWjETpp=|I^h*!%YU=4*#{wpAq zl|2mkd(eE?kmL+ScS4!(P<71zK)Ye+!J+K6pK-jIyz$eHED9$h(0MKt&pgPItl3c3 zn+ztDEmmZ8DK*(!{_6sw2%xedHmxl;Yc`aF!1XGKCCOO|q6kDC>|0 z=U$tD*=ISZn_{z}Y~F(?B~kta&?N$?i9TuvE@miu4?MmnKjY>#gtD|9btvoiiw<3- zo!@27?*aVAWy8jXvd5cqhdK9E$Wn`)SJOBr9NYH>-ha6K?zfjP;EA=0*{;HmF$e|ckSVfH8IOK8+(KzIH=-D{rDb#)9O|1XGA*YhT zamaCypmM7+igHM5CE{1TT~6oLACd!dE4?Kkx4xPlyW%Lsoka2Ucn+eyCR^@WH6gE~ z{f?o@egxRu`t>$%zztn5ek!SuGWTn^2@b{cmO`7{x>KP|ZhdWgoC3*uKsL5$n_Ih7 z3KQ>uUv>wrFX?JyTT9dl*FwGn?I%MI z$*s3Os-X3{1HdzasF!5T&8-`iawfO-Se~f+je+S)qo~1a7Mojh70azVPXeGNL9Sez zTl;)kS4-*#z}q2`Om6j#DWWCK0$@pqB$Hc1uG+eVVPx zCCAW?!7Vmq! zC0TRbXN*$LxKBj6w7TD}0QDma2d`PI`@H?XV!HDQ0L&uDm22JS`%YdhX(IqTLL?dY zDSx99g=1ztrzg2W633~8l4><_5QHCCz1iuH^Cc*1lJqjmR0=$U~ zKa$LvO@bHLVv^u5E_bpwV}f7o22@_f=BDLl%_hO?ior>OH*;pu-DQEUg;11glC>0) zU|ek&vkovVEeCZ|Y&Hr0T&T8^DEk65m_TZxsD)(*kMG+<_$Vln;Ky(1B-k>$ zSFOHfcYen^ziaRt_aqyR@nQqE6%eyt&w3+&7SRn2Cm>tirEHS~594!Qf3fuub>+q; z!40|f>#xig6iZfhuA@4$vNI*C>sgwU1S8TqUs!ZzWd#O>E4oR7moTibsu;UTg6Ys; zl3@71ZW7Fly6;hXTt`I`oJj^J3H}8Jl?0VhR1&0CB5BX)QX**ro>KHr8VfZrE6dp3 zb;#NG;E>-!&i5#u2X~;OamX~w!g2nO=6?)HK2vnZAsf!b+%VOfks>3CVhwgk&s}t9 zP80rz_uddED_~QD4CzQ_%{pX8wit)Z=5i-{lc(Vl3AqPhGs<$aW*st0F*pwS;gc%5 zyOn@!B?DEHtfk<%Sq?R3vRQ{L@n0h)QNBQHYM4@KqHRCuCLMA$?B(y- z8PHM0GJ0b65l%!3Pv&CuMzkmPkmcZ^*lXDvMc4cZ7p4>tBq z*IXUF?S@uTIJq3)PswnXWY(;)AF;(4`-RJ$?5)*2i`WUMyn@YNmYX$e?9}(bZSDph zQnHcmE;DqEMp3Fs)>1I`?V~tpO#r63<)Chg%^JJt!&j6<*&CpN1X2@i>@sn@fS46L zD`ksq@gf~>P)+WM3y95X!13(`#LoEV-k`d(HNxmlNZL{cX%xs-%yhc1BgQL)S+l|0r9u!`WSj1Q4W+Uq^gcJb*OiH z=RAV`-VBu1mRB9k+M`CJJAcAU+#LarI^N6ms2d|;u@AvAm28Fn3-G8$eejZo6=>dQ zNV2%1J06w!Fiso30`0h=2YXZ;!1nM>?v+8|CyhBXvUCIj4?3|UBK%^EuwTa2;u zyWGj%8Mt#{IG}O~HtQ@mYu4B?ior4Vr~`F%cP9Y3LI$cPSxdp#*-OD2C9D&PqB*FW zVzb6hTw7B~lqCTwLm)MA4&Q3(`aev$QocT_mOgw7^!+~$-y)B2XWMW{dY96gP@1{^ zKdM?uF%zwe4JG*SP1`JwzWxvLZUeE0@1QT5=;I2iG!XRnOc;E26wRv+U+qyfz~aC13~RXXs44wik7`{Qw{g}5OCz$Cdl2AJl@sgZ z`ahcc7?P}_=#EFt{TFARV?di?=)oTK3c%K*@)aqmaPo72_mZIw$*ftAYQPrbQC?EK$7E{K4+}|~J!Z=*d*b0`NWXt?8z}Q#! zVfOevn$H`O%%SLxu`6A!F62L;JvQ`UW9J3f8hdktLJB7fp!1R_oodKSe^6DO3!(jHkJmCL0e4F0FT?dtcs*d96Oljuu zJ?Zz0SJB$hP=XKN<6}~(!}l$a-!Tw-_%2>vR~^1HL77i#H;OigZ^EKFg2%&WpnT!- zI*0GcO!0#L{yiu^T3&VdYL9vW&HWqj^d=AAhJ##>8W)2Tm^1_N5D+ zI~vWgh9nad-SMdEP4N6uP0$(|day^;2iSU4gC9yLoE!-7NHR1bnKkQCE!ko`s=dpd z?9E=dpqL1#+>FgG%gvhgsAR?9cvPRR;kvtD0Qr{;R86v$f=9iv40p$6fprR@Xb$S8 z*sMn_c{81oD60Wfi$H3kJ?h%qTzUBod%(Y=o4#RX$6(jkFP>Dx3;-;{$@YwF=1gTK z-z+^9&GQXOdLEPR82f|y$j%!<+hORz#!d&YHFnv{SrtxR2KY}hWF(n2YwT=nF~*K^ zxs$!m9%T{_0F`+$^kOKQn>B0fXvN?d`6U}K zDK=~D#u;lWiE=GKn+T*P+Sm_T^YCp3o|STM)_VH=aEsr<;hWOy9gp-YE93)6dPZqo zD9s$c9sAT4nZch2MNxte-@j(WtHZZE$bJK{hwqW>`tW@fls2UHplEaW-uy3C@ObDC z$`F^=Ief!$OEUd^Dk!rouR46SNA1N}{r@e;DlZP-@S(0py_P8s_y2!iPAl!5-BaVCzxszScNd2HZmpj>eCTluTA5hs3o3|}DYu2MiCEqk|?_a)R#bNB62+VrY|j(j_b9+Gm&cPrusnqO8^%-aQP@~vKXZo~+A3JjM5#kedY$15U+{cZ1nYpCJa zqfscq{joKIJ{Z`Q`mYYd58hy`$pi&g%G(P|=~xr?_Wu!U1|{NMz+E7@2a0DsRX4F_ zLsEhmiPjGdB{{ ztVM2I;M|<*tVNFdMqiLqYliGy_>8n#3yV$MrF9M?lN|!^ZpjcT~6(Co<4Kj<<)To zf|jn!-qrUeo9nmf(&_iRn5FBD7*^PQ^ktT=&tXJn>G~3SHrH=|L*19u60<}sT{niZ zX6bqtB&cIZ8AZ7vwG#2i8t2g$;AXrVZ~>0qf~m~rX9*&)Mq&yG!6|!tmXmbHBze1Y}a7jbcE0FY# zLij_`XnO%};lhgIf8dwJK=YBV24^q8onBB;aG%S1pfxx2kPC31FUQ9yD5MVngAB;M z0Qc5Xyf&3Wrhqoj&_gc3$L@dJp@rN$(nlsE{jsmT!33THka=AcYtn_g@e~D z_5xgvO?7nVVc0*}Py*!I3veUq#Ar!z0QdzFu~m? zhXe4g0qF~HbP9C=Zu~BIP8dy+H0M$=eCu*F7VR(>Nj2IHW$V)v#o`A99Hfw1FX2xn zl&yV;OSb3G{F@=kdW!CZvQ+7+<5Cc4sorx~gF{(kfNdx{QKX#0$r1opAVUk1S+k+6 z9a~H&d(Gue_Wrsni>L*t?2gSK%gve%Wp5}3CzK`rlSOwoACR?VplXt}6hhhC{xPFN5N# zE%Lg?7NZgQ64BhukYtjgJI0nOgV1&ZSxd1OGLo1S5v&Qbg z7Gvz!U2fdKytSd&1gQKGn=6)^HEZndior2iFKCStMB^cyQMcHjtnp zh6D)&X(>gD1d5bWyhza^Ep9FDMSi#xXmKbMx8lXUxD_bg^1bKGow;}S|2+BfJacz4 z=RIfa?0e?UmHoYDnD~OWT^bEVd^vX!(_k)o+{S=Qv#D8muSxen$6ToXJHyV&03DX19dM)+5gGbQ!p%^3{l5)jL z+S~8(inN0Ky*eayj4sCkSS0dMktCh(0bpE^|Szvu;4p9G>#b@{k z6q=RD(7ZLo-)?`A1)B0Y+14wOkakg`6qVkg&d|ZEL?+*|-s5epM1JpMJ(^;zMD`X{ zZ;sQyx}7vKn_8Jz3)dd~%6nZz{}Rn|4@O;jB{Bs~)GLuUs9CQ>J|OQ&Xo0l>Rw93r zf^`wS4=k_;@}TfagnA`n(k>{VRwB7ZI94L$Esm8)*&H~l%kmx0kdR!x@f$_;CCNOdL8%@5#{QMZz`y4+(?_s1}EUUU`-gPOGaj;hCk_* zNX$j-Ah$trXDyOHvD~7v^L>lp_a7k}r{&&Mwgg~9Wjha-W1L(M@D5T~M`FquD%(UQ zn#y+A+R1(+#>a}ofXcsdcyF{RXQ*rsD_B(a&6{GXx`OCyBocQ`QZ5B5n>s2CkMaQ1 z)@V@QSj|w`i8Q6yq8td&FalCf)VKB>Q&|z{c+#fE#Uz|etsJGOtXxJ#T`x1J&kolA zBk>u25ru}zQr*iSs-xg4Y|Gqg zSnB~4>m0mU33bk`PpS^!Y9Bp-qED)JpjqzPXuGDeduWlSvb(5RQ`uwWJ#CB99l(8p z7m$KQWrtybsf-7Osf>ChVgeFzs|39y^Y$1Am65kNsH{asP}xu<|F=#_&o>DF` zANq0tWJ5{LjWbmC%O7}16K#{TAzPy5K2#Qf@nR#p0NAfVHkEx6kxCi44%tI3_o1?o z%H>tR(qbGlB5}VY(R2KCpm$I}Jfb~9>G?hJnSyIsS$)f_ z8I5aVxp^5x)YEUU_XsL$GR~&5=-;Ao{bwvTfskC~(!!rKmCgDJ7wtVLuBb&an&lRi zEo_JDGj$^It!GgQ`+6)Y-SxHm#o_Xv=GNrAd1DVGA3?VgV%YCeokDI}#qePcC4Wi3ZkV~esD zK=lYnJyGAq1 z^%u)wTYgElHIsuR}-m&iYx2SCLYwuedn=T6TT9u`} zP}_^jGK7K3vZC#p%A(OkO=Uw-v!=3<$a^|`!?p*58)V6?UGBOr`A2i6Ova8qYC?hK&+oX)RPO=VNL zb9ziM1Z)yyEdL(YVQZZDPRH3-4JQRSk1hVa#RoQ{YQOZ(10Gj$((p2_Oxlm=P zCjf(dENLp+S0CReOrM+%z+4SdR7Ot0RCX_%zF0J_iIwN35m76%V|xNrR_F(t%03y2 zJ6kWqz)doe+8_R;sjNX^c+6`Qrw6H>?uTat2y%IcGXrm{x1cCz2;9`T|zX^+6+2cu0nLuFsFf<hW>(|*^TxWy~{GYUQ{-vxZ8`$T0-A_4RvWM8-*rnDvLwS zn#vN8_gs{?tO+Wsh|X#%>jVo-WjrWMWz;JXQ+BiU81R584k{yWaZs869BfYZg1P=k zuJpt`=24}oYjBNa63S<5nas-M6cFqZVfj);2A6YE*~f2#87cPweuz|ZFa_maCmLdhRTl4sUr#yWn`L+_>ZI-XQ=Gy>^g$xTt-3`tK~jaHhXQ7GSV1; z78+zz+2++r%19r`hHAMFl?Cp}uKG0_fF%T?eo4yNRQA$~%G{l+se1na=mM$G$SM^> zW&O%qmA?Ss1A%P2hRQ|;S5lU8U=j->IW1`_%Qhgp>P8#@i3D<3(o{BNu@E$U@^b(> zYLKEbatfxhhE<5lXk24n*0okY^5M4Oj=k6=qGPnW8wva);AbeAd`>ZMl$~%to zi&`dw2w)=N+5Y~*es#I$bZth;R8unwHeod8ZWD<5B`IgS&kpXKj)ZTMf>pi282TJY+<2v8+$R%W0m(fm13(;sY`eyN zrnOA3EHwe(b016EeZn3@DN6$Z80lk4yU&KN^9q_ixd4Er8l>EZoWkKg)T*et0pJD% z!QwOQAebgb2zdj#cS-*y@v07jw1VQ0Oi4KBk=^uS>YU=v+M0-HxSRAGqZOWM-7|04%5F8Pj zN>#TVkfWqPU6YhcAqaNQh%@Xbz!Q#HQ1Tgg7`;6Hbu+5 z2m75pT+RJLmlb22+zRkMQjqD;d&-%?_M;L#*g^D<9_&lgT;dd<@(m8D(R`&%IWySV zS-~3Ymv|XJHA|KNq!KAm*Cgdq80^SE+Z>WgZz?51P+CAsysxG-52(XeQ=%Y_*AR0x<)hi* zd^M#pBrQlfm`v-dDRp8h3Tk;TNc!8l)}1NOXH*gN#rP4D8Ag|v8p`#4M|-0#bi>nL zaJ>w3Y}e~{48L!A2r9=(ZF~R?qIDPC+fdvS_7{quYLT45a?AC~zK+2Cue0&%vPi7z z?Rx)iaV&N-pGw6zSsvgjq%aS*B;}0jW!dUj>@Fs;cD-sR^NV_b%CB)4Y_ut7T+g4h z_0^Pb!!xPs76bANDNxrW`!3s7!B$hs~OkZels^)lmf$;7KwYJJnFa7 z^m8IhFDVdBYq49h z6BfHiQI}rqHbxWmV)qnk){EWq$a_xLOUsMhBIvAM>_))?dms-AFLtR{BBt8Z>S_zJ z<9x?zg}lYFTDkFi4EDca?h7Q>LRwy!)ynwOpwK}mAEjk-83D{{rNB_(kyR>YwGtj4rz*b=z(WGrcFk&~%+w@hDJ>>3Ba+jSUafREU|9+S zpge&bmh@`n-#{T~`eXwDKGh(#S|O+4)k=Y`6hLWQV^)k!rQB!f0*Ct$T{zsQ>H0ui z{D-k6WN-}3+wOB}A1*3yNBIFQlamNw+~@WXr~B0CS(=gZDc~LqQ^8*k@Alt5SpL@;fDNEG>Xy9W>yHDVqYRb}=0QB;)q}?atbS6R5Cno?f zMT3<4kW)C^hsHH3XXCmY1hX!(gJ65S8(#ha-7}=$KN!xSgW$kQ*u1`v;(xSA4q>?! z1V5^c7qO;Wh?_u=Sk*fSjsVy!b{n6MVw{WyxCSYVBQfPn5d4u!^!jUtt)1)_biBN1 z2&nu9hv7z>awZ7QV+CvdRen+}Ro!wxwvYmKO;Rp}AUL}!7NeJdd1y4KZ>(m5V2!|1 zY*D7cFa{uTPZZxrO`&DgDvTE%?fbr>y5aXNFeyG+&dsHfbjYl;bTbXfk9Kh2_p#q61XFMtkm1vL3 zZ)+#}-PlxHECy5_#^J2drkwGpoUCAZRKlttRo#0)0?}XYnxtF`9(Ap7gplFD#2F3h z8><+}KO3{jbU>q(x_DAveNW7x|a&Zo;dKwWCGPEl8!Y8_J zw8b)TJwU%ZfqY{;l?IWTLr(^iGEXT>+#fKdp2r^{0$}m>aEV&H9ayCnZ=n2@m@wQ{Ax(8feKmG{Ka*sw`dhs>@P1K9GcBo!2-Z~-gS+pAW zZ)2zO1}Rt(T!xn*BUwiJb->kG{Lsw= z5uuC}fUKmJ`$XUF2dk)l)d1il0#UytP}d~oQXtIP{qYrO1x#0?L49L2Lzs)! zhOtFC5}@$}q@IY7V>#;*4<4>t0B4UVK8(bKx%ysC^?(qS2GM_3Yj}WeeT^?+$VW2* ziES$n#Fz9QU@x4?s_lfvV?kfVQ!Oig$HOiKFpm*P-nCF^5DkX27XF`!h3j{hMz=h} zVp1{NHe03GHu|1o+xK_XXJ_mG@%RkCjzYt>zvu9XIw%;2x@1?XOSA1z{N(OZD(!_* zJ{Yzg*r9~isdBa2Y1RK0+sfkJuWK*zk9F#7vF*#F-mkFRgJ!v3qb|+1PtYRGwj)rz zX4`Sddy3V>_7>Q-0XnPM_5>{0gTib}y%I6$GFMV;J9&+RZFvT4wvDTZ9*%>#ACX+| zvtty@sdD=&`0<)$DBqxE(hv8w+H9M)1ip*rH%Tizk6@&{1o#b731YNn+qd|6MwwQ# z?Y6(lF-WFcjZGjVg*I#}pOzKbp)X57R+{A8IK#GiAC?s~=dw0rjkMf{Z4dliP8sP6 zKraol&vjLDMJu z129B`)VU5h1+#5~$HcZYBr%W1Ti1n7{p@fbVl>C8@)sZB+RqUfJ4psVi*&lrvd?h! z_b$qxYng0M028ba?{vCP%^zwoQs!EVO;9A)SB%!~vy;)gC6H~` zxX++7{>oA}0Q&n_((aSgzO=G51%UZJmbClqOY1La`s5Y>e$yc3KI9Y*_n~o(+J0Z{ z7nj&z$LUi=p#K)SVjY_Mc{VtMj?--2YS>Yo2 zkxU0{6&2aICGK-@9M7SyI044VRpKuy+^>13!hARD^SbrvdXI452>+OHe|cq1vGc9J z7=W7ON*Wj)?hpKgd)=sX9Zee@Ot?>aY~74&om_qUt9A3O74Cz#TXO5GbYB^wq z`}}{Y+eq}uRYf$*eFt^vaQ``)sKfmp)U3n(A>=*vo7CcPACJ!Ja9ae|6v?&1E8Ks34hzTWC|{^$@>c>FzFt|($=9L% z^D$B$2K)r6Y-6-mht&1mm64VJwACP+uhUnpu8a(VY?PM!@O3WSFGeFE7Xq-FK-4cuIh(I1 zbLTW)ujpJ_)q578>!d;>t5ghM7i#2DmA?ZZE#}B!*YNe(1Ch#7egMJ<o7=G1TxN`jr%OCQ(knW#Z42)T9TX_XWZv%VtFwV?UH>V8>Zzx?z66b1!ZIo z01Gw9cAvPU3d+cK$PQ?^kNdo?o>TSf1^|x;ME#PKv)v~ncTT%c%~yp~y;(8z`H{Hs zO2xQOl0Ql0!l6Ie$%dN2(4!{H-OWJ*Y?3N^G`s6YI z)@YD&A94zZ`_Q;X1I^3t(gtum;-h%r-kys?t z6(qS8?&p0UEo5QHN@=-wxX%Ev3HM3Ei!x3&0yvoza*>#FCfpaL5*_YK+STwwyR=&uCqQ*;WNA_4kifh z{i&Ro80Qv4U|ZHA+d2r&ZW1eIQ)xZwD;-P_yf)0bP1_2BWAU;O=e>|4mRDB)TR|`& z-nx8j1;O~H>MasF2)Zk~v7>{!bPz0yCh8!VjGA>2Y>&L>v#og54$gTK&{-V>Ltue} zAP))$LFyHD7{AJ_SZw4D2aA!nI9TlEv7&hY9n6hDa^*;N?l9(mQ(MS6C|{~&G9Lj9 zi`}l`++o~Pyb>ejalp@#N)bkD7Q4V`&012Xk?jW#Uyym!0^gTBlQSz(fM z;|z<%e;qGq&Se~AiCXT%Vt-Gqtc-jPKt~O-SuA8qWo2X-WaG8mhsA~lC#Zfc1z%2E*kN)yOo zNwe75(HWGb+5j~2v7}jSeBpG0rcZVSpqBQ%Y6o%&dErbekU#hBDwq- zt=(s^m-|dUj)%JelCd~cGic*Jkwr6$%+%W!kbO>aZk%zSC;Ll_acGwu2-!$2_i>*U z_ac;$1pqA7AlrR@y%V8~?1t>9mixF*(v!TZUv~g_Mj+~!q@3+OJ-yuL(EPlr-kccv zB1qhLrDELY)bgyV@^}Dh6Ueq}+~>-ge9BT=0J{2E((d!_bVX%p6abTbENS;SkRede z^vTr#Y|tR(KI9Y*_n~q1e6hL+E+t0<-UhVL$G7u~36Ys21^%Lzw3HG@Hlt_ zh}M4>UV@F}Ivxfm)8Vnidx#35I9!Y5S(aPj@!%Pp4OfG#o|b!u#|O0V({IIbZwp|Y z{2JiFq;LtgB;`zad`SvAJl-I&zI&?q&QxM7pmH4!+l@BmOnAJ?)mhMUa#R^^Mg`csy}2kuAzNfD#BuJyDThC~mkHZW?({w)J=} z1LugHe^=pgd{-47>sz0{S)bSO8D16#6CQW%E-2=9af|t|EuWKZ9Uj{aE-bcE=^N?{ z9ZY!K+S~iBIN{Z-M**yJ#7Z;0U!vss+PYw8g~#3l^K4k31+G9;sI%=5BVE;+hR0k@aaISUe1gM2q!Pqv&2Pz!*5PsN z$F&$F^Xx%ikQCbRThKa}$P9fM3t2UibK?xZ^;+W+H0QDfWM62x55KKFkV+XD3cyGW zviZ%uAIm{9vH-GGTJFPdBa75i{n`(}Ndi&7B;{;=`-D5E!{d|ZajM?80Hwk3bK{kY z;kOrQ3aQHT0Z@WKwq3(-6S7rSmJ$G{?_){x+xz)tl%qVuW~E*xxL@vK13G|_i4ETmw&Fp*k5F@NH~P%9PzJ$xU=dV z%G2(%WHO8Z#(m}tcb+3wdlAD(84h?^Qi))+cArU%*6#Du?@StWV7~_0?$dWcCS~M0WDm96$9?ur zude!)7UP%^iTfofXS>gL+&S$&XZ-7_dZPeJAQfu7QZerHU_h9vyd?k~2}I*%f%Nmg zxnqU0Gz@_8K9;omjQhT%va}R{^*)xg`vje$3Tv9RzdWK~@=b0M~z!T&uunCJ6Q|SXap6C@!N#auds~Ah@+sF`UCd)=vBtYdx9CjOR%9$Y8kE^qS zVCg;;RCRX&c})t`HA%S?g5dC7I7iHh(J6tXG^lT^W`f|Pq#A5dRs$%JfYcKaQnd{4 zsMe^7OSc_R+yjYsRB0Dh8KV&qA=MV}-tomN_%6-^XpIx7UIK{{Yx9 zTi5*n6YlA^D}+PsxkD<{-sq)5?LzA_OKQMU-HS06cXUKaN$6$?uPSE3V9R)WMZJ z1NPO|-(2`c1RMqlAi194yJT}^jTN{LrWnd2wM@PxfZ@ul z=E{D%(+iqD`38Uw8l6`3oR*&~v3a_?LH5Rl@K(2vLXO+8l>EZoWkKg zG_FxsimOM{3!ktuUx}$@g7iF@HpZNlG3J} z3AF=R!3woYZWmV7H3Z~yQlPF$%B2u$U;hvH*nS7h6r(|XV>J_Mn@q~Z7UgPyHV}|{ zB0`2Y!Fx-mxsJ4l!bucgM&eLgD??7JdKwWC@}NgKx>%a=B6hpeVJdPW86byRDh(nv zho0-Jco)e5zBPR6OwYK7V-l`&wZmh_B3J(Dzsm2+u{BBQGL31w4 zKo+OvK3qBYbTwt92>>lM$mU89Mt<1 zl?)1)x80|Bfi!sk9m+3hnG7I+ai4Nt?z8imLdw)<@N$17S4Ku__X+iKp9hE2GDt=O zS<#@4`%J~Now=#EA4ArJBdA}l%M?hWz;0}SPUy^dR`{eO*pRDhTsCv_5=yM`*xQ!8U(;7A4}SO4z8&qX!_(r0G4TxavyREhx^dD zM%@Z5&S$lu=j>3Mc?|A~zX{!Yq+hW#oI!`$;vZpI{~wCoe_A3L$8sywc7BC3s=Sbe zYPol)O#s-0+EcS*7$@rk+>8|JlbCWQ)Hb3L9cn+bwUhmJdUA^nfXZ<=OgGw;GoiLM zD_EiSXh5v0ZU-R8NP)U0DVIX1E&jeFe#(W6VuX|i^^Mg`s7;<0%ob%1fbtQLdZPF~ zvkOmmx_4pBJQwZ?4uZALs~~uJfeM0it5hp&WMS3XDnWc`+>6$JmA7RhqG zwc8%ea^FT>ItWfe6Lk>$4mIl_I0kvooom<|K@f~XXLS(l2@4zqc~CeAQm;fz-tE>4 z&xc-guo!uZgT;1U!`<-xVQvVLD=Gp)69m`y#}G_I`2sDI@dPj|He#deo4yNEH;EYr-R_~ zDgmnAGXPy96&hKkVpy!tf&f+dTL98vjvRIki=D3@s4V3Jpag*&mNbjK>`_2jN&uj~ zk0s4wuRnmsJqjM7scbv#&m2VYX)Th4SZ-0-!|8ao?H*(=wA`D@LIE~ZHuHRC#>t$hya4Cm{7ieQTrBrNfAig^tHv5pP1ldOq_NMP*$A z)A3ClrL52S*5@vKhG)USP}z_!(PB_QI#J^?s%}oUHI+TglTEay(l*o?Iv6UO+rE?+ zl?~fj){DwM!*-sN%97)(*N$3LcFd#nHI=E=^?xQJ|^pYt*c%tRM28Rikm= zA*jra&T1-49hA;7C`@J4E4-1YVwhTx)xGARGV&G&mGy6ncTd!Sxg;c4QY8otl?|_l zD_Ct%-dW3JV*(f|OU#G6NX>%_E64aTQceMU7OAvgw5GCHMr$e?`BxHy|N*V%1AB9 z8fv)@l`Xo6OIcJ?X8`&Vi25ZdXH!`OcTQ8;#OJA1z4HNDK`PXErDCXTZN>_!^1T2Y zCy;H|P}zp=p~}*I0RHi@q^Yc6=8DQvddzfAB&Q`!Wf|V*5Hx+VEC7`>NKqL%1yk9> z4D`jKaZSuVB(sQGI;I9w+0N@Wm3?~)@$?%Q7)(YsC%_{#mG$i%hc_9Zc%Bx?JuJ7V ztQK}5}UkfO4I*5_mEGyPytSsVOg zsH|{ek~lIjoyZ7Hd5vsqD*NYr7EzK)Z&PRJV5sc%6LrB{zh*T0dL1t+8+S{+++Dx? zz51`=N^VhEt*hQQWmiYD+yhXTrm}8mqNcLfs996lf5?0Kb;ku3P}ylxu&x4p3=6y} z<3V96qh5)`fNOb0%z^XPlQen$a_|{Vx`WS#w8kD$$W8nb7?P{{9_Ll}>GyEEMzq@2xXFGtf*Xg(YLr@yMV8$kU@g&MC^44>`XoI_PU1%UYkvh5l^ zTXoaw!4?4a`dHF@wts6*W$7va_kApBK8r1klLTr-YPdr>BsD$!*ef{&^I4@~#Ah@l zQE4B-;jk(jecPt8hEoy15};d~^e-kksBB|r>=U*^@t0a8ue02uvb2>YZvKI6w3d5Q zS&beJD!a8HoN;n3z*|Y-4s1!v87gZ`3Yy9)+uF%~k^h8?1AxlMIJ`01lrvP;K9Q=k zsO;wHTB^Ey=xZq??wX`r3RG74i&!B)0;aXmpuVx1p|bzsfiJex51?-eNIg;C+Ker; zVz3KB$Mf$YEP+8~%kC&DEA&iJ*{#3TXCv$X-uMhZh(beUcN6_Z85CTCZTXyRYbu*N zGC{ne(l<1AbTCvlxQF#TiFJaRqj?%HDogxZy;(_9*-^YKl~-vNmA!4E?mg10vXy9- zyC&MMsq9a*NK;urU7N}>BJUYBIR{hOJW{ZzY%?q{mGPi3l~J!m%=r&l)T->}JqL@C zw>VfVM~xD={}JYHAi3_-W5H%sR;4pmWv@}5>aHb|PYGaHtnNzZy+eO~k-$h926zOi zykxXyu_}xb+2Tt-=148U3+OPa;* zei|xh`s5)1j%$!wm6200i=8b)uQa1^O*}o&Euy|Xk7qw7l*)uB9RQB4|PWC(dC_?-QsN9Uh9-~b;LuLQ5f<(mhY{l5JY*AJRs15`je>87J=o z{G1eCl9+O4u-{XO9&A@hqn+%xr!-!(f>z6-I7A>RZOWO!4q^psu>1R$Qq?sEqys5X z*Cgdq7;L}Sh?Jv%nPD`jZ>(kp`^tAAY*DTQXcGacCnDs#$$97+U~&gs89j^Q>qvZQ zk7htsPwuB?#kGLmYSawI7s^lCmQ!yvI{xVY^H!r5gK&x)3REd1S5tJ@ywzxXi-MvW zO6zHf_gjsYKB&eoi);_^R~lm8YP85dj^AoD9Fox_ZA+&0Ta8MzsVb=D3m{o)>soI$ znv^P!pfAR6knA(M959sYJ;gXgJsXP`?!fh8pV+Rq`ZBg|UqR(xQXA<(mHK;rGqfPy z|NJoqS7iim_PR?EFzZ!*Bf_4*YIVVwLN;4Y*vkHnNSuD6&P`Z3gA}N1l5#1yUj9dTs3kQyyN!aLZ3zaWPZCSWu zum``u^@_nL9<4=k1Iw+!j+u+~;XKHeYq|GeZvog0_NYz887Geee32CPl9+O4u#Z!T z9_+KWcCufa`z~=8P#OFbmp+k{Hs#D1AK=TPmJrN;C$5f?D_-*>)arToazJtWybD9CZgiobG!xHj{ zpxXT~DXK~Al2j;jlqNIHZUqosY{a8cG7Tmw1ClXgtEkESXdTf-RkShcLVYZc@lD!; zi-zj!(!j6p5nQgW0s}RXT*XQ`CinZlgRv!u;tpCQBUx@u?y*n|#Q?}gXu0>~Rs`5g z?h|)m#>rIxZz6?iB&M91+;uq10#Z*z$nIbB)8vlmj&;Ox z6mLP|S@|9u&y!1~LDWpEH9SE4KXV@8ggYN9^i(~Bp~acQ&FAXOVf2&$-p{RLeQvWp zFXJ=3G!Ev>p)xKb&YT<|LZB(TQC<4XA+EW;8=LnZ3^(zF6n|&eq3ELs(y*Ib@y+2OWaC^qE5(nyAklMxtiD2{Ime&!eeo6XZN8 zSZ5CTVZk00K69X6iI~E}YN#`ZJTDz*4m6UEGlyc2u*VVd0{b6GuA9{%G-nRAP9Std zqC8g1&0DFBb=`)Ap&0MM*R{^;1V@aPm^iFaMnm(Bt?vM^iO%LyXkW=uP z15>G+#x=};D3}4&@y?10LFdh zd%2G&AIM0#2=L{ka+1;7eI|Rk&zgt~43Z~-JZI3xeJ+m)6_=^E&menEa&DY)pG~7f zML`tHOs}y1L1MYD`%EdJj8p(1R)cK!8T~^EWu!4=ZM59Sedf4BRKErQFoHnTFG)Gu zeMWk@Pehq)s@^pKZ6OtEyizgl(`-nhs{AwnR|#a>HSV(_F}t$#27uIp^U9wPUNSzvsZb81^2}b&fmJd zhyU!iv+tT)RmhnzIiJi{rNFJ<&R#ihda(heJGI1n!JV#fXiwRmE#W7ecmL%e|Kg zeE>GggjK3NWpHs#DRVFD{y z%Y<3u!c}$E0BJ}H)HO-D6qX6)La{B}3z+YW2K9~A%rfEEOGVhCoD0w*0#Z*@P%E~C z7Th(U6V+^F3=ej~dwZ}OyuwW`XP|PC)LKS527Am7oW?yx@jES&9a(M-_KNaY&SiL) zQRG5mRqw&>3a}aMX`!KvlT`ptB86Tgrkok<{#2p|dx))_?ANhYZP64^IT(l0Mw@bG zu*b53HQ0Z(^Qh|90`eOvP}d~oQW)$dv$4%{1DIzPlhF*Q>dF1otbEu_E9LbV*oE@ZZw1wfb?CqUpA+k?q7;7r z6m~~awdc?y^F2@3KUB;{=@KpR{+@TvtWKpd71jrX5J zs5}?tV3sBXC0TN9L)HO-D z6kKn98f+D`0H(9ipuVx1alI?0{Mn-X9-y%Vq@E~`8iM|0c#2m8LQo$OblVFghIP}u>89!8sTX0ZQZ1#7V9cEM&lDNM&< zDJf9bB;`^V>;ZRhJ>w8C7mNlQAvH7DIq&9Yi}E>2-w=>`B0{RqtwP`PeT5LEvSEY^ zBJuZ}W?x$uY@g#lEXJPOb%0ths4YZ6hQ_8Oezct4n?>~awZb+`sagtJli*EIi4?pc$J2j@A-H6bMyCnFC>RaT9!=f@A-@R)djWuCM5T4U26la zRZ_g*F9sU#hop4*d#+qB0xD4(Ct*1Y*Bj)rU2n~RD!Bg*DsiM1QJx0TddqF~4EYht zP~1|BWCfO6uD2r?mkWDBHb~38T`v}3<9Z?6>M~9)0(cE6)FLtEjO*2-6771Q*xJc{ z5nW4&ZGg%fI6N}ilryf^j1?@`t282BRhJokEri5flax!r^{VH_*>-hc8W|1h8><=D zYdkIoTa;Y@`kH{$6Xj8dFVput1UgZB^1{n8*jZBBgPpp#N65ucSwU)|f@83cx5n+> zzoGcB7RdmXTZ28Y8C0)9_CU+M2RkFcX0Qj`F3UKXF;ym!8_AW8#FR6Gorg;FU>CBr zll?L^ODjSFmGyCGX0$1120NS;tif(Spt!2;TRYcB4cX6H{4Fgq>Twk+&>npHhH&$7nqPVpd$-ykQX613d!q`E9Y>1Y7 z&&qc+4*CwB8>Ol+PA&m>9Vx6OG3Csx{7NNyR<_yN$$l&T&M$TXD*wXasnMpKnU&qF zV9m;+;4-SZ5cIVu5_e5fE`?e7?+5r{Enu1$4eA@KnOQlQIzL;K-2v)DK@*8kRJ@6U#Mzfr0! zn&pl|+x6zv1T;}^UOhm~dh_aUCVeO!Ba`W+Pvxx3rxa1D7<+^y%I4wiYKVe zs|)^_MECaY3?eKfRYvj_$L_=CL^$6mm^+W;YFQIPvw3xAJDvr2g7UXoCfgFg>^}5b z>)gDGdzi#XSUvB|O zgR$g(Ny^!~4=uQJdh=>%=aQ;k4?q=3g&MC^%x9|O>WKr~(!NMH8eG|Zzc z^#Ne0k0rhPFyUdevNRijB|et)?nA}YsRd1++y%gX4N|)gM&78=*YL*Et0 zlI6|D)M^W2j~h#tl(rzc=f(m*4^V}WTvIE;QS_2!&hb>D0!piDiT9G_?wFjsWN8L* zvWA!?OP9kQUb6IsWDrSzBGY=wQmSYwK_omClG(PdwPY#SC!3%czX6ghMwh=WDvI91 zI79_L!-Ff}>xclGujjVHCFloG`J2@CMxjc*WI6d6=sj(^Od=x^i{xRJTYNq4e+bG& zA&by*Z@xYOu;J^}X@eLi8w1>i6fTjNa)z&OP>JU2yS8?+UxNn$qAQ?s3J&v)HsuUo zKV}7suT#%XQq}DP{7LG*4+duElKTqCC6ZU;!$-^6c5rO`H(^M;q;Azv{D6i%Wn?7n&Y~ zq_in#20J|~Sc9EwRbf?K89=I$0(DJNE``B9nHJ}6?SScJG^lT^W(K?efl6#qjt6Km z0jVb}TgY8)f9gTI* zV;Fcr_8XyR<|{BVAMUtt2W1jjkXR&>S#HhB$t`&1Fch+~TJAk7espb6FIm#Asl_BdbKd>7aGV5q-g7NrS*9fpW*p&Fbku+_p^!7#RJ7d)FcN`5A?!lZ>74z5BB6x z8ZtVVg;A}7)~(9cRlPOg>AV(3qM`K^wzV+&@`3e26l-C$q_&&!`l?>468K4u#;8j# zjKa}Gy)c@Ln)SkHCi0#kU&B|iFj_+j*1{+QEHKscpzy+odL?4!WU$_q_&lRyVZ?)I zKXpB1QE?$3!rU_?SLq~l)hvuYorn7t{4-<{nUPo~qX}RZMqB=LE{r1Z*b^aTG!9ir zr3$0+0SU@nAt% zCM*JwaIclqeG1INy1FLHKh`qYi2%lZ*3l2)n2W-GEn*od`vLwfsdQ(wcAx(-TD#A* zud6UfE(LP6K^yle`@Du20DXA?vSTFY#u@i{u`Niv#vsVMkUiIOANSdIHCP!5%8V}o zl4V5y7F76^U}eMuSw$`Pai0UfB&vRW3_uG4QNJYRZ1hLLl3&xvDq3a~);r4*)LsSkmru;YCSh=>-5Ed@O1A>3zW?X!>Lh zxI;c9H9dS$n4H4lJ~XZi<%~p+T2g1r`u|x{$1cE+19pSS-ek63417c{soS-R7o$-+ zMN7Pw)GOTucuBnq;`JJ0mehyG=i(*xVMtDpv?-a^OX`w$iU?XH+=Jwat!pi*8_x?7 zvlSsh?LGaA%4 zRx|5~iY=4aqWl`5egveRsGv5Sa_-#FiRzRIO9Kpccy@cRC%3_EdMlvvGpX$*upaE& z6S3CYhvE}jB#*P)8thbOF?Y8i`&-Ms2m36*X0W?8D9SjQEgLq%kzChFOgS^y52-{C z_A^^M*>CE{_@!JxWg{GtjW*@XV83PsYp?^ymQ>Y!56Dzfpsq>Er7+kbN!Uc*1k3@W zL49L2GuU5lj%17SIzV>_NIk*&3%~z>#oPR*I3dh}5iX3xi#M79Uc6Ch5YLBd4SVB+ zI6YH;MpNh*g$ko4bdTVfnw7(zsZ`yt-fsjGpCP%XkeZ&Ueh50U8;bjDk(|YHYo^BZ z!%R+uY?hXL&(u!u{;g(W1WoSCUrRHA2Uy{(<>ckbH+aUM|l0f%&G zzS5?gnW@dJV9nH!=rXFh(tuPZ1?rllTnaOlZZyuF+W^zUXi(o+&CFEjwHUT2#{e{m zfYcN9wU~#;9fi9LbfW%igR2hsTKMO(2YXpntm<|{w6$b`L7igWcEGPWI~-7cOd$ z_E$LcG1`y4+}{{ZHy(cm_dni=d)?aHu4 z`5LAF5s-SK20IUS!LeI}F*VYRaR6P#9I-g_}>{QO)w$9+kc?zP?#JUU(q#g(*3 zo@Tk_QO{4|G@?Fa&9vOxqb>k!JgQ=tV4VCG;4!3djl`5Q9(9{av`0O(wUhl?Y$+_J z11fjmaL{N|&Un-_RPVuti2(C7F%l zC0Zosv)mf&t?tx#|1)I!wcLBK{|B%c?4R>hW}JKk@M}^yLSo99!9GbPda%#g+R1*C zyVn(I&}tcmLwO{nO*u2zH(9|N>}(T@sp^^o@)api*Cgdq80=pkR>S*`fthPGsBf%h z2D{R*x@=Ky1ZX<}sV8c%S5nU13Ob(LcOrSS;C4Q>S@30VwOJ5mea^N%x8pNB0}f`h z-~t{H>DMbzbC7ZQb zu(*9aCAT&U5}T^GrRvRs=V+08IO@`y1$j}RHwyx*+M5NLkoUa(4>#;$v)~dbSepf+ zkF{CAgTk8y)GLv=`*WP~b$MmIqNYee$0;98y5p2D`!_h{%bOo71SD7Qnh=^DgUoc_ zP9>CA*D^VP0Opi$t=FZL?w1l7DLVt+lT?N>TA%VQ^177L>CYMrlCyyP$)L?C--zb8 zUkLhgD`dM!&W$ste6G(jiU%}N7a_Z?x|3BWF$Xi zVOs8U$~XOJlImA=02&a8`Xwo6pYl!jx|H(hN?lcNUx0>@3N>D-m{YzBSt3;Ba{yRM zAlt4v<*WN3x3aVwfTKQ^^eNwmqLIqd9RQyBSkk9_71|~Unm!o_cgTdKriV}Y$SL@g z?`}V$G8&So=;xK0%6=$hQ(5zmun(RD-H%ECXZWRA_6>W4D=8gN+(V1xHkMmdR(EX@ z?qr4R2QBxevfTh1DhrE_XPn#w@b9E>l*E)XRCbC=G?ksVwXqT@B*k$+Er9fo?bV0Z&FddBs^^MgGm3eT_4%-_-kMJ8g*!TRou%fcpBNdfxwmwUYQvcV%XZTzc8Y=tYKvm(7f?KdH{~_C& z%6_d=Q+%M(chng=7%F={%6eskMP(0SvU^e4-|4;gJuCbd=0#;sBGq%zn#zWtS?>C1 zyQZ?0XpyF}plUXiWk%l9FDn8fsB8%-SX4G17I@{tgThosy%I6kN?UK6d{ETEV&p9j z7F$^XJE^x|?mm)hR2>KnivRA-QCO8tyme!AMyK@F-H5$!N`D zAzu5QEvBVrkZcWP2ZJ^&ws=!nu^9St2xKEj&W$rHHlj{8L31vDf^4Oh`>Q?~9F$9VGB`IgKSURtL&)O4%RlRWlC6Wp? zUa1%sn>r>|RsJ~uT?jtxL2_jWN!sqy z`z zb9PJxQGz)8H^}ypoEvA{r`34e@q$5+*C2bK-Nz+ly{+5j{n5cNw^&UT*~+&R7P**HR|dIthDl2oYiO2xQO))rM%BX4Y~l={@*NIo(R`&% zITHj=)1$NcWgx#Mq*c|G1SEzOsB4mPDFnfUg1DcwB`{r#2K9~AOb|?)#l;ro2!O^B zka{9QZuY>vY7<@GcPxV6IY#jzB;KW(*gm(~rJ~Xx`p;?&4^W2Ar2KF?epYvyYgq)Y zOWXzQxl>uSozN<(unBi5w9IqmEKbo7YIB!Vq4wBJ6{!`%Hm5ZoliqX2NP<`X0Xl*tWf&_w=Fw=%`g77dOMPS<;btO ztV@vASwq1B-Y<4;jApsFpe`M1hoXr()UHF#`V!+N-T z3|B5YR8r8K%Rmpl`baGI;mXq2JjzG}02MUI4z=B{c$ATkAZxDWK3w_dRkrV22!EMD;2|)$*sav<;MUxPaxZ_;mSsXb16&D0r=0y zlIF^tZLG1#j+rimR6tVNlrx06niZ_G zhRsEbsp?t)(wP*fYm#y)5N0CYTO=m{GtX#H-&oBMW|QT$*rMD7&<+ApPejO#?omXT zmFt1R9-#Og5)RO0%0JU}k`QibU2GL+RYvKPxey`L+>e zC2qVs8Dru}yNmWbBGc`UP*k0JoucN;KdaAr*8g4c8NM5ZhN{P{3l=3&@D_E+x-@p0 zsv~$CG8`5Oa!BBP7XR(T!Evmk|z`6}q|J+TW^|mN$gXd+a_bsd_spSZ77+V8I>~rfTYyh*`NUzgpJHvJR@| z>9eW&Wf*pT-@=@Tga`dhPGG2d@oOw>LQq~n%j7l!7^)sJz)96VB2p7lCgM<^RCY63 zQ*|#!YpVVtM+}2xA0P)Bw4v%ghsuki(3jI7n?rJLoT2L42g(bYbGZ?+-CFKL)jdvB zP)05Ra6^M^s?LA3f->?RGQTobB|cQ0xuL1y+BoOsWQqHF8j@&s-)jJ0kR`q@Y zP%BcQ#w!&=)eD|TRrxmn3@4Cn*HHDFaYdA+xd1Hlv81W`)_*Q#>30B*`B>6ao$XDy zpy`u$0eGxIimJ&en5w_rNK{SZns_EJ-VQYfTb)GJJ zs&{9^d2}rlH`F58n&lQ%cbx=p`vS6_TJBBNbpSS0EoOTdCuab>h!i@Im~w`ydr*m{ z>b|yivR~fCVd7^%<)1j*Fxr$eR6UFpEUI3VJxo>SkG^I{;;u=`r9jmUvfQ>0EWzxsg}> zZ&BH;6+!BMYt{MUZbANEQ`tMT$Xyz3*Hl&%1)9oMqGnBH>yY=<7=be+tU5=Mf<P|M^I z0vIX_eC<3#&-95uBW0Efxc-IYI>~5FWrER~%6=;x$RJr2$XW(%sOQNJYRY$|*4iiSc{+2|fwRK58y^d*qE@k+%|S?c8Ks`3N?>J!MeYpCqOkgUp5 zM*zO|v81Uin>$fi8V|q>A4{6bifqdwX!_)O05)roqB3#{rn098iOOhPW6BOHrXpcL zw8MRfE*$PtZWor3siSZc0FrAwn9_EiSGWgM=0|y$mdVKkFz&O>D-s5U=V7F52zXOc znaODFK5M-qp-fePL2?L?BMjQO&zpJ?Vm=Y~PmnDkIXBL@&zhtNaRP%NcR_YU%YEEu zS(8X*{~BWWw){(mH{#K(Qy*NUosg#l2WK-4cuIoo{}c}2oqLHSg@ zEdlCCD%5zTV%%qR*8o-dFaX9A$hK?Tr%|^2%Fd5$?4|DP|0Tu_f@RxdKOz$DR1wM5GYQV1gW%dv@xs8m zC~l%fauCa{AUL2G-c`^Uvc6jG9Ry=bW96>TSIXQk%Q!g~;ANyR9JVCoOc31ujpH8j zjr5PwPWBrzIIZ{y5hXK3-j z)HfQ`H&!!2u+<;s*`oXcpe_WYo`{e&IfLkAVEQTCATtZaE0FkFRU|9Ra@IN{EK3Oy^9;P!l+44jTt5p>-p6+wGipSP^fG?x)Uo8TW4LI0VSPVBxI zDEy!)XHvs-1nn?Iih@)+p9Y=|CW2=CCP76+>tx_~&;06t>tvusA1h*75%g|->uRtS zL0dL1<`qGs(JXfl)TJY6do)o;&~>O;N6<~kd#YB%o%@KOhe*MSptWGZ9u$tC)GHA) zX=0+Hz~IUb3Z#*AP~c}x;OKsF7;PliJTR(%34Ni3f#+R9YLq=mkg4_fgEknh5~~k>xr$0f-_H^-EID zrohc!5j5Y#XjN}(fW9IXYP?c06xi-)B~|(N08Ao~ZP!p>d;IVR55o!oe)X}WDKP$g zw6b&@fQvqsGzFf;%c^Pmu)wW`@&xPucK8h6ghE4@rA0nb5Cs*hqJz0Gc7`yE#Cb$RDlI@G zN(V!j=T};9z_SQ5{dx6l%qJFM*7?`k*tQ7s#fU13P%XlIgxA#5DT+Sjn1yD!7osjr znETKoO_*g+vnI?a_K6|q+W@b`9-YVllbZm!sHpScTd*; zf;(d>V(j9OTz9Ce=9J^gdEA8f3Cde(nS4e7n=sQkcTZ~MuFgm~0`PI9@|w|xFd3~0 zvwQw92FYK5+-lHuPsLzww~z#CssEdRklBv^7E!pylLlQNPM zfczR{6Xu4scx;%ARD!I!mirK9-a#>{U#$T6l0eiiNjV2$a_96Z$Ej?19XB=p2Y}{~ z3N>D-7{WZ>pp2?~BLKSzWZN}_Ijp)+mM#Hs$H$T;%yXT}DoYYG9f;(#qzUuP*gAry zPZkBBqy{O%B&T4)eEcsFCJjkUhl90Lyd6~2;XXu4j(FRp9PW4N17icp;3+IgZTBfP z5?t~l%I9jCyg~ruK5GX$-RJUGRT(Mw0e*y3ZZTTB&mu-^_ZfaakU{bVkZ%pzxKGC1 zA>s-2WtJK^|3y-bGw$;tdx%JYnk)-hyq5d8&zs`el#!+Yw9+8keMT0|ri^?8*>ElQ zai7U`^Q(T%1z;J0s9%zDw);%y&T03#elV4)_XI$fNQFjLsTlY9=X`2a`AY!s%#zct zai1)EQY%Zj04Pczhb8SkuZKq}OO*kr<6}v?PwI=s1x=r92S6tcQtm@e;cy=s*QoQi zu(m^}&5~$`+U+wvLN12x3erE&1kRvOIY#%ugWkWP_^=kqQ!KYa?Ul~pqHB;n&~ook z`)_u0($L)R6y7&TIGM2)-hYDRIuBctawgOU=Wv{ITqUs%wNK~g5ut#}`ZzQ*+LSY) z)}OR>s9kV6R8{vaAQMS}x+W=?La1H37`KwG17^3;;5L(*3AOw1l4`bd5vA7&NIelD zz5fZ}s~imig#3WwbQon0wM}Nbtm`iEb#tFMysrLV9>&QL0RKP=FJVhk&UoE>D$xr(S96Eg z#m5yE^8l55aX4I9D-M-Ah2yqQBfVNx2lf?)KOC9u)_sqS2tfv6}I^ z#A6xRqWlP;#ss9Eh>*3#3X<1t^T*Bey-@ru5_{dTFLGGb(}*aq`+tO8XSfqZ8=VOy z1h{Z%p-C^&dkgi_FTF^wL8Jmz7@o@cp?VI%tOC>m~iW??U?J5Z+_@Jemg@NDmg=%}VY9&2aq}*n0+# z&dRDg5m~A9M|I)pEg&Wa9G_OOEtfU@_tZ-&EXT-@0+f<+sIuG4#2?6$C1zLJpiCczmR2R{~754wR^zUXR zV>>(;^bpO-TDXw@QgAmbH~zvKFS7wFVDRXye8m-U=CzCewS+iL= z%NCQBA3Sx~e=tx*^#E1Bjl~SBX3b{hXA!tr*)^qrGu#eP4pYF;G;1-kQsfBsKEHwU zk5w=>k=d*~o3^C%s54<2v!h5P&Sdv05QoYB409IhgZJ=rvZuH5CVNLk{Pd?WBrPf1 z*)?LvwD@)0@#%%;!3NSvg1eL5dpxdtz6ET$!J{Yp0LXT-XBKNBJiQI%eH0ua%$l9- zlWZ}Q{jH}C`v;uNtd4`K?_u$uRkLO%`=SWk$!_~cQD?Y9h_xb$gr-@GG1*i0u z&6;)8n^c?A-mN=}IKvGG#(C#eQ(BOdd9XtnI`3?_qRzNpZQGBN&9Wdt!o`-LN=1vCE7b6{z zvb|RNC19fr9_@gSyc~A%FDqR{czQ9&Ybm%w2e4)xP_lG{1M>5OaX`DtacT#s`bR8o zST$?b0p&&DI$+L&V$N{s+vBYlC=!}xErtV9J;Z}AwZUm+6^u<})&XsX2Bk;cAEdz~ zG7=TM`A#wBrb0(NOScfs>ro^(Gqbo3U~3##(_I^xAo(2Z{%3hQ;K4dP;?di-Dng7oZ&Kb#J64)2~D#W z!vWh{;ete6aGtXY#wIfBfWL}1kRJ5_kcN=RNK~-;By7gheH-Uh$MmCl1B&G4tDLR_ z*c!*xbk{~ENInPG;>SyHKtudJsXhkU`AD&|?qsXrcT?gy%B-e1EBO~BKXPeDaUH-c zYN3k*IKB#GMj*>{mo9}Au6=Y$`vAI9Eo(*oX+t-B@=idWd8P)Y3aqv$zWccDWlumd zE^Dazp>>!+qEA4+y4+MwK&AkmVJP;8@S7%7krR+LKsFLpebKc!0Xgt6ui_gj4g)#v zY26c$9L*Xkz7ORpkUy+ef^^yH%egt2Ub-S-R)lM}tHMIXPQV3G!vY!wtm-`AY7>667=>vx)9u*Cxmo@1|8 zL}s_;vd1$?kGcv-HA!S7*r3`BR;!orOVUrsK*=)Gu#=otHf1C zBYkFPxEDYfMgc?9ti?!!A5U=UIpD0c3dSZfn}!in{nDdO0_gyWjKoPp%?z050B3%g zE?O&%D#E?jr`@q(m--$Fa&4yj)3O3vaXx2DQ7D!(d z+;vp)_i*Vq2rQq$qaAexWb3GBa>NTyHw3vg1%DD|%{uBHTa2Ur@zh~|T-G|OJE;0C zET&sEYt~WfRSxGm%9pOGGu$>%4pP9-G;1*&)%ushc>WojhgQMZL}nee^PjBJqt1Y7 z%!(qBIF5Roz6tEaoQ0OQ!(}B*_SEOS$zFPt&p$)boU)NvVLRE?Kg5NXo@gFqApNG` z?qt7vES1s|fK4@c^klyy(YxP@x^Bh)1c9eN#UhD<*@RiMll>uE%w#X|)M0;UTm^NM z>c6mfV%4nK$^KXb?qq+_HN7)j0mND!MMBf8#hB~?yx-LpoL*KT>r9!Q>`4!^NRN6f zTHhp*kvNl`>t{}OVXW3NJ^MO-cgh-fzUT7Nf4IDg^8kt?cCLHW^MC$LY1we|0eO~avf=33*>Ds6zkXa( zt-X*_)dEuA)5?b1)L)=tNSy%M6-ZC3l?^xZ8a#Xzn29x7riVP=(0R#F>ld(h%H1^?;0x3pR26m?4RsY^Sb@AKFT&fO`M5~n`o%~!mQa1rW;$#2J?ca4*Rq1OQD{CstaJ%OQASw*6aq;Uj*(3 zGp2ARXSg?3tdRr(|II4zV}msGq;8rLv+mzd@oM_4)a7;;5>C6AZ;V zD#ae>M79HvZbVP9YjeEI`FkdocU4zM*7) z39T^layWIkj7E#KEd(sDw^*YNdGCg>!>|7@eL|vA6)uG5oxre z?tyF_H3=8e$!?>l z3C?gULD@zDL({Cqa8&&_YT*Bmz`1S}vd)xQM@<~rNP5&7)96Q$NE}Cney5{WW4@N@ zX8!~oM?C;1=00lgn|QLX1z2s_+aSbY9aVotQPmHv!weGbs7%k*5l2k{Ji}0|qcY;K z*W#H>Y7LN$L`Sk~+S8f@lnt*n*=l z9M!YG=crYm;fr3bemH$c@x8+#jia`_gIg9=&|KF*dbZ%^pneG5bpY1W;L(oy5M=A9 zk<*e3PfrGU76mH_vt}K&jxA;%wb@gL{aLanQ%gYAhp;$h)vQ@Z?GSPR9ZaU9j~HXZdClf6t2++J6$ z;yHQZi*VGa`#|hD`6PVuj(XQuV@qFxo<3%8g+@j=>dwIi>NB(+G)T0g5<6uQNBs!+ z7eld*YJmGIbkx5@uoAv9?Akc0%h_w51;A&9QfdqqYw49M!oG4qY!n@(N|OIizt^p{uyr{|=gG8%Q@5+;vnq3HR650NZNt zXh*dM**fas^g6=RKY)Cdg06&FvySS?7UQUXo*F-PPEhJDs5(1Vy&#ICX3aY4B@wue z^4F>E4A&TxP82XS%~}jcE&Lu=9>#z(#VQz^$gHCpB$kpM^$L*IlE_FLN1di`0?nXF z($D>xRehjosSgLjQicAFv{ZR~@+|e$Gq|RG4_*Dk?tWx<=CHSIY(13@`s^rzL|ba^ zQRgC6S-_PI#ac>#;T-l_0BK9~2D>(v+Wu2*#VhgyfeiJu?oMi0zMP6TFy007zSWAQ zoU8G>u&$v}qnwAsQoZC^>fYhHc>gCPhbjAqLmEq^Xn}8q7tws(Ksv=J;;yCAm&DiX zC%{q-a;4FhN&~XBRG+~(;|5Qc!XiXLkT7f3Qu)|oEcL9X4*P4ZZlW4cJpha0R?V8V zR7nxImeLKHJHssmWjzH9O|urmQg?CstWSV*$tq->DYKRuGRiMK>W653N+Kh1EH#e4 z2`t8(B& zmfAEKuTjG9Sl$Huj-gmfP3~M-_EJlMtRlLeT^mac-jPZ1n*ClN2R*H8so8H=Q~7b{ z;71_8Sglyfxx62T#!${iak7`{{IX}Mf5za7c9y{zRZbM&9u8?NwY?R#oib>yW+45Q z;Kow7@LY6DV4V#fZK)F=TT7MQo?3W%9LQ5B_?|Fp)>6N+#aQZhPaXER7#LFXK-EcD zd~MaNSxem!forMJkE=Pu-6jR`N@$w37?x`H93Jh-4^BC&U~D3@mO3=Hq4cPmfYgda zM&ej%FMSiZi^)#XQ#+P+EmeES|FP8JmUzZuKDt`W?wXH|u+$rA8>>&znq-h@OZ}D* z5=(sx_@bd$OEtJ$LM(M3$YY`%*|o9M_0|OxZvte(>Saf91~v!2+vA!j-eM^Wq%u*_ zilv!_@`lISLq>W-W%L`tk4YUIFJVt6*#*vzF@GwyE@}7lE{#L`LFR zszfSUssS_?d=LI=rc%x55a?Dao$5Ud*XmhOx;fh?{f%>qAECWL{jrHywbNKA0l%t< z`+<)Eq=d==zq8=VLs-RZ0CF2p{im##>#$N|AlS|wZCz_Z_BC? z{H%6<^bBB+1s_;2(>fd>zX5R0fM&hhq=+jOwCnvMVEqt_r{w+XtoH!){&pP%E@(!q zl#FfGo2?go4c92FccQHKxrp^<`_mhk@8)RiNUi^Etalt1syV@d?b$PcS1p*yf=kvH zQF;=9sRlIb-84m}jdksM7XYj`VzG_9?9_j!8=&_Zd+=8Tf_`m!Fze0M3%<_};U?aU z1X=H?uX^*%_E)ZC#`j+|W3g*ZT*ogYFB?Ao40X>%AMj52)?KRcfN+O%Gi)JCc*4Xl6{PC^53BO2LtI5g1bAD$#?Pa=51h)4epGVd-f!UeV_S9kjoV#^Z2vpr2i~d&4n%$W+6M?%k>F+D= z3^yH=#S}0!&036|$v>4bfnR`g$|@L}$n4JKlct%aNBuiUcSvL;DwzEVqMqwJIi)6U zdtruipvaBkmha?nhv$r_U~fMjmYweNFUGFDAxP~biSpAkw#G4@2%VPCa(rNixaNs9 z3+)_HUyctIMtLcH{ds&}o`f!@qWDhA%DInfzTOPaj4VU*CkE0N1b0(`eF^Ni z!J|{^s~M5fikor>Pu~OiKMF2$j#;xQtwe!IX&Fx)_P5%fUZow0$t{OP4HQSsnoVhS z5x6OxyF05hTu)GjQoztOYcW!K9jDrQHaIJ+g0YFrrgYM^8q%ZQ1JZsH8Hox`KZ{o^ z&Gns2#CCHT%{NgbrL#ICr7}EcL6QR@cSyIXl zRiHQ4EcD!C=Pc4c#!Km6S?b~WZ*);ulrG{ zEirg>N{82oNa>oEnT4khfqaUB2Ixz(W>dPB0+Z4eo;vK$J}N<70#&CNol(W3IBM2x zN;iqXO=+#fL}$24pfsR>p=s7)q;yvc92xq8Gr}quo5*ZRKYFF0^r+tlX)cM3Li_)5jriOC8g|8 z1wvS}P$&Ee4A0`ef6Ytjf>{ap{)Ze_K=I9!m2bk~xz%P+^?59=ST$=l zrGBc-EyY{$iOz5V#F`yNLes3pNNI|9up6reP7|wOY$CHMJ=rO}^r&9|sXvK~Lx>K z4pks6)+|X6Iao`DU5&#ukakMX%vDjslV0Qtbvy5DFg z{+_661KkK@o7F}fUM^uOLSJ7;YGK@xZ+OOi=~i;3uR(H?vc4SBoG~=I))Z&5uVW5T z1kytVca1wRCuTlBu#yIkHttA}?bnZ~kA1?^&w<>Hf;S1XW{o?IEoNUd+f#@AZ9Yn) zUIbOYkHrG3X3ZM+LlL<9qQR?6IKzDb$|(vMnr1DAaTj(*MEAf+ju~_#$Cy$c)NJ4d=V7~`Ty)oXiR72co(^Vl!psX;5G?uD# z0uQdYMRV9dx~$-?rB;6gH@*Vw4TDEpsxrvdQgN9x3s0{Ec@qV72(xA_)r>91QqOtn zu)lM&CTb6;`WGy2Sv70cQXNI$T1vlP#ThOOVl9Lsp=s7)Sn7j`1#n&nPJ63hY$CIk z3a9i*k9sgjuad|}97}DdZvxq%NzyMBYow-&rFu_*rT*<0X{pqmA}w`k9%eELT^(R| zhuNLkOLe-FsJ=t%F9wOW)P=fAEcGv84C9E+4;~7)DI%820VFTc)9l(}+?FX^z<={G$~op^*J%6pK^r0f?CX)Kj< z8BV-cp?Qme^ew?%OYQBO4o3`Nrwtx$srw*XOBLyuM|k=X$jPw+J~bUMYt~X}*kUY| zfgj8r%v&p}sSKd%%2?E~YSyf!vWmd9RPiYloZ^#D)>s8& z6PdMCzA2TYNBt#8Uy;a297}b8Q&iwZXp;2Fh1FDoSZe26u++ltk(OGEPu^ZC@g+QF zkrAwHD87m8&RDARjgTsV*2)Hnwp2=8KrGb)a9cyMd#UA>@!CsVH5mwGDAD)XwXxKn z_!ny)@ZSaUzNd99HL`kw;wjE*ARDY!Eae>D&m;8Elbp^g8w`1_9-}$sL>JQQ1$Qk~yAZxX?W2r~||$3fWS<496y-tcT?)1Iti-&n@X&{qh`%cc6JfCll@n|L}$1npj4rNp=s7)O!m-v z*uixIr@vJ&Hj&xMesxfO=}}JrX)=k71aBj5kcy|rHRt0P`!Sj~qsZy;dtGsQEW>j~ zR4{RV4Y@Ay60Q+n0_j#HQBIH98pn7dbXq>k_QMWUpg7hnRBa$O2&DArJ6=l9A8w#@ z-pSYqP<*>&<=k&PFAT(2x@u@{U?6=!a5tri=Wt!3GqBzUk4|Z}$dtZ4JelzH6p-gq zaD;QrnoVg_3QS5HcoLS8o z!MELo4Vcm{C4x$C2I-4PqNJ3qaf~NIr{%MxlpU(TQ{+ALY;&X!Dg9`Qm(p{s@q?DT z=;9HIFZZH|t#rfpnBjEq;!7ZkK)R^lZc4lUj=Nu_fQ1YmoznMkCTX|Qy|_$Hp6-rC ze+o*YFU^`w={^cfNs#&usJtP7*r6(pfcZNF-%8wK!k{_6OZC3WSuFqDQ(?1hxDk6f>fGBMxuhr+GSu$w>^gmZHDF!D3a3YU64{4 zo-?9?srSh*-NvuS!RRfJW=9ewrEHC3JP|rApCzU2Pz6pRVWG1B;@4+L>5!>jN((>4 zzUCmhIEvyM`cXtmU-}vMz<);b9|qE61$R@r3V&{-^?P{q5k;iYDJ@VvBBc}hR}`Ku z2XYMx-b7!THJj4*6quB@^3-8}t`m7xV^H;AEJj;3Yc{1_Mc}5i&-Fsia7#ehL;*w7 zti?#_a|t+UE3w8JNvgr`$X!$yGO z`;&9bnoa2{3QS6udg`!$ZQrUY7pS@}7R{}iHJj2;MBt|M{>WO+aKk`(lLCgOS&Na< zBb#w~btO34tb(zL%%*f(ohH(wJ^|8M5*djKme`+?DgAtKW9;_O{1in}IKzs-ak-q?E03j3+{;<+G%e9jd@^tXU{+)9jMcY%{!+uGF|U zHvnA>LGjgI9I=%?_!18{y^ZD>2GUIhcem1m&(y|=0IGL38 zp`bPT(yZB(KBB;+^sc84`)dtKp>Bh!gYV<_4=9eBHJj4^MBt{h)9lpFa1B9ePXR;I zti?#_?ccE9841p0t6*#*vndT;%P&3Zr68>$k&$43y3o&*mRN~dIf&+yD3Ynx4Uke9 zo-?9?*FUPll=fUxPwA&1r9(O_QBum*IK~s9)ACtT$_`baG1e?}CMEv1^?<M!ucPXs~AXU7Tir~hYC2>HUrkd;L#~PP&p!{8_G5o zo*oPGWD0ViFU^`w=_3kEO7D8=u>TnT6k`skdN&q_teQ2O(*H!@rnKUlHJssYg7Sm{ zhNfAIk8q8IQW>5zf-f2B{R{H*2 zFQo^T;k@ftbn!ciZ=kH4yOm}+i@V2a4!#7S2&6{{?xyt7DV%NP1yplUu0v390g+&nN4Z@s0Pxb&IwXJ5*djKmTr-gDJ?S+JMUU( zZj2%+9ash_mEk!fDmd~-`DK5n9hko1AiW()l$5eHj`2k3w0xG7vO^X45(x`^y+KJz zUzq2mwD29Q!47n>2gP?%R?bc7n`v-E<20IoG?2a^xSP`L^>Hw`3+$=Eqf-sQhLZ!hyB}gHc-_;)z4!w(5hLpDLpL$lhPIWoZ)7I zvWx`LGzO_tBvDez);Pu!q0{nNQpyfhU=G$Sq#9t0 z#8z7QLocNVnpDQ0pQDTSQG6AbM(lVuEyS^OC7L%GNGAyHrc^I(juQc3Ck-B*(!VQ2 zq_pAQ6vESglRO`#0s7Lc*_0M%QcOwbr9S{?l~ph{k=c}PYEVUb)cZg>L?R5GNp?;;iAy5 zXugdiDJ@(WDV5`4W})2E z%1KI7FZ5D63qL~F-O$AgD86~Ja_&|-;3sUrBhmb(f%J00-IU&0j|V2_0b6eH=#&nK zOlj{QN()ba4f0tE)^LtlvngFnfl29nPaXE(x#?4vLDi`i;HS?hj+!-_(v>1`Q`+)C zqBC3wltvUVG|gI!lrGAo@%$S&qpgCmiOi;S?c=J_qn-uQd=eRn3bx%SZ!qfhc>;EO zX#NsKGPR}Q1uU3T(xS}Y)UVQz)k5- zQ<^)&rA4gIph#$%wHPUVZ!AtiYl73E!QG7jQ<=m9+euT?odC*+UKzcxAN{{`FlZjfu8XG)1 zrK#IQ?0AcPsD!5ngFKpoAxtT2Hl;NwFk5LAPaXD8Eb3E}K-HgMvD2zqvnj170ym|z zmuqLZOQ77MfT3yDVx%-_JPsupFgy8B90g+&nN8_typclssR~j8iHrnq(EJ=1h35Lo z-o%c#8=Cu}NJ`6pgp|tgoDmfq^hrIY^x_g+SxDknYaI80+E6o~{^k5b56 zrY@d>pe&lJqex0CFG5OXc+Q9l9=Rj`J~{Up>;(IPG%}JXDP?OMtBN|Hlp~7E~nqzls-CM9p`^&{>DJMg5YjS+s?_;pyNqYy>F2s_09zW>fk#1tz8YJ$2ZBzf&Pq3RK+&i|$sB~xQ(xVJ!=O#wsGti?#_!+!Y7s_oz$vI@o~GMmz|*BeNW`X`XClE_F@a9Ad} zdZbF=I$v_kaC#I;>G9b}sSM8u@VgQ686KtAo@uk|-%K9a!(xNN9l%6S8TItu&#RL@Jd$MwFN(UCib>$Dxyv#uO#FckbI`?(l*x3Q> zOM^$J^hT?Qly-Tcn(*`uknd5jh;z)EO=&46#iX=|WXL_^-TqNP=@ppVXR#=U;;31( zDXky^H>C^jS8|5y07_p97@B4+MoRlX!fxyxaOPVDV-uN|(kum~N4*84og^|6{4>b$ zy4>+*_yreezD4t86iI258Az!N&l$m=YUfYQlzx$bT|#=KA`glsN=n%p$9N)iT0Tok z*`W$_#+rpTOu*h7Tj{?adnv87FAaYGg)Z8m`0mKcxhXwU0(*uR(fq1`^uL0;DSa)z zDozA|%`td%N*CeDKYMen{_V!X)4M@FM1gMw=9o2`(o+mo z!0J0{)@(}8i@;6klnGUx;fjM&jRJLOA&7yQoG&eGk4h!z4 zG`=z}2zCS3&*0H1ozy%crH!W56P}(1@_Y((zRYnU!9CQXb;fEzbL+)vT|-pM<{$R&9DYv0#F3f`vrGX zdSN{tk17kSy1}DU`bV>fls+oaRCxM%kOxxmHG^c$rnD@RVp3XMGUV=fr|!(DMuDmq zW3kq%S+gmvECM&BZ?8$=40jTgpDAEynza}yy^;+le9167nNS=BV-uN8Y5sYsrAJ)? zq;e!O5*2K>IKY&q&4HbFOEh;zk(8dCh?L6koDmgVGA}(-+T;xkHVLE;B8iexw#G4@ z2%VPCl2Uev|Kvi#LiHcxn-Eg^%6c!Q-%i1^8DF7`6DYo1E4gxRO8=cx9Or*%zGWa? zL~xVRsrdCz@^v^7KoMzlN~<=FNa?m24TPsFf}B7>DfFdTvnhR@0+Z5_o;vJbov(^& z2C5#4#aOFm&8GA%5x6P+dQoL(xaFXHN&!RDti?!afl;`Nat@rUR>9arW>dOjn3f*( ze}Gb8DkTyXeEA5zZqD_++8Ze?faX#tlG20;NU03Z8BxI%9qz9>BUH* zq?E03j3+{;<+G%e9jd^`$VzBKH`f-Ri=`;Oagiw<`6hNl+tK`m zf%KHfl>T$Q9!>;+T{d`hN;fo#Na?4~XBD1K^$9it6yFTaF>5xZS12$kz38dK{$0BZ zsJx)+23WMRYSwH@Z-~H6>HPPqI>QYI`30=H`;`>!r&fQ9TpTUmz9W>82 zkiIRryOloK%o71%TMZta((8>QQaTtf<0nu5fW=h`9&nCXvnhRsNiiu6N`~AWZ`V9^ z)Lp8xZ^ZjQP#iUDHl_JR;HLD2K|W`=#-MbffT3yDVx%-s5EC>8oGDfz>r9zV>Gs#_ zNsoF3TGx`uNL28P5ppNyn>5%)zC!aE6iI1arc{RKjHuw0)0LUh9uqKq8mY*DVu_Mc zw#G4@2%VPCl2Uevw~SyuLI>Y9{Acdr4PWbo*ec8yHwi+fTFPk#dPP6|4qFU^`w>H8Fzl)me!!~VZ2q)`V! z)qh~|w^g%dQ#w}!Zc2yVZRiY_3$d0&k8(;C@E!Y z9OH@5Y56QEWrr$|7i$)38^WFvJKlYtdMW*2lMnB|*n;<8p!n9y%DG$VjBdE#R|d`1 z45W7m?xu9bOgSQN&hSuzE@1>2V-Wp|RPeV%ynKGHui*GX*zKWt6pEy@2~*0M)tnI(%oLXQU~hg4&(wYl()LKA zq?E03j3+{;<+G%e9pWt;$a`p0C0w*cN@s2JQu^ay++4ekF8)OEWx#{r_VlRf=n^>p z+lns%C<5tc1b0)q?sHrbEdZ>P!J|`}CNibBTND~^mbPoozgcFBev4ZJ#m=>JpC*dcCm zxlel3chLGbiHt-A!x`iW!PzS-VYi1F&Wj=`UGy4KD#LR|a0~n^c|tHZPga|O)HRYQ zDP?OMLbo~W<6Pb!GW}*1Lla+H*8XsRB=YMGa z)IjfIVMqf`X7)}?8I2VagJHDDJ{yRn3NWf47oer zz8^PH89~)mu&8U*tl5;77J-}6+-J%;!}SAYBn1pjvlb(zSC8OJ`iJ1GwF<^2GMmz} z)tw&oevrNV|B!icSABf z1ThB9ZyQJ_3ht(K>OA;y0kDq^9-Y$uktr?yZCc^!V<4ZWpgH=|tl5+A0t@wEavt%~AXA}i-^ zrK87`MEq#(WFWmxa5tqrrsFEbOTb1MJUXS98boZRdD>MKo?Z;{S_(FEj#;xQ&BUab zl*UPh+^sal3t80;Q1y>k+^}laY)Z3>z)fl4?=m>UrAMrJP$V?XT8xzX<`uyIzkt)q zDj1u{Y)Y@S&LchQ{vZt|k&&q2-X6GTHrF?73{Fm_qIn*QWa`V|NGWGlb4KtNF4yD< z!8p9CSsw)HTqIFa%GNl>6QR@cSyIXlRUjkQER_C>@{-amNnT2y{)@c%_n-?DUy3ym zTj|nCh$cUpOBzVW3+|>gu?BWAb$~T9cyvntt{;)oQ^m6jPY(roECoUIrCGBnElv-Z zlos*~5%zy~Kbv|NRJ{p{-B!(-O=(#XxG8-SpU)ZYGAMshz|b^nF;bfFKE8lt#_SYC zaTJVAWHzO`wUQon4Up=R$Vl+~NfOR_@M4Ef5cELv02E2->{pOd8J;tuf{Q20HQveJ z;ml?pNUI}>l2W$DF`fvWmd}z>cBlfsBJZK6`;h}|r4#pgDb4!_BK;m+{Dk7`CoAV} zrDyKqiq(BItGzCyhY9Ycv_TwR{gDM&UV}%cv~2B&ln$&|U3j_y$gL=@ zcq(Q9NMj<2l2W$DF`fvWmd}z>cBle7keSdjydZ=r?f9ja(vowE;0>7QVhf7zSY%3@ z^acGenok=@pN~xG`XxByy$#jv{=DPjRasDVdn}%}YSwH@e-eS4(!-N-I>Su@;$7f;Z5517 zWHzPkGvKYT?B@y=H%Me8D)_1ilj6hQ!2xAMpUq71>A+3?pys4 zE;H2vsYN7FQp(mi#uK5_@>x>K4prb?tXYyCxx0}H%dgOD?Z?;2Sg&7cpBk4XK0w!t zP<*@AVx`Tm(4U`>OKnE$=LU)X6?&ZCdHds8z&{v@{T2Gcoz6cw-UafA=mBZCNgW>7GrWsk9q+}OG#uT@F-uuvVp%suLn(%zE+`_dP6KV;~*^6 z`P0Z38}!2`@5KhY@0C{icXV}!-HkvX)>3bOoLi-UJ|0DoXiJr=pykB|ML2|^SW6`x zs3|WtXaJ-s(FyF@SgK56oY}%sF97N9Y26na3|p5)@t4fwflRVmiPHHM?Yde@rAK`Wt@lY}B#x!F(>H;+7&uA$U$5o9 ze&yH0$ak#Quk1;Ndlogps)yn$xIV&Cjc4Uj9njj-AkmiEb|jxzYBb<+hGH$XylqOc z)I1;`5iQNGjivq>l|}J)#oK{=?rB|1?amyh_)F@uKz^`Vv6S;G;^b%y#oxqrTv)30 z*Pf+nrNUWM+OKdlK=CDTNMor!k8rem7R}`hq?-utS}JV^{B)-wu+|2Twp1&St))t& z&o4YZ0^|u4bRo=|wNxLr7)uTG)M5Yip{3PyQ1vz}_E|M+)>1=7;9BZpu!1w(HBkPh zfT3yDVp!^p(+Rl$gV`yH;wTuK$gHKl4CIj>bsdlrNn|9BrT)e5|5V^NOm>p4ou#JA z=HtuRs~mx)Y9~eRrCQ*Vx0l-S80+>vx|+-GF0(sx;LF=9UadvzHiJZ4YRJ91VyWYR z&lrle)YGE{#8Q6%`HSdHc5N(`cY1ln-!-Sk>cyit1DgY1i;m?Kf5}~hD^FCkVkzfW z%PF$D}wvlhcr)jQ&X)&OwEScR-JW!6${(*&hQJsYhblE_FLOD&{t z0yVIvNqS5=zq&04zITryE0_02TIx1Fd6w!Jk4F!GL07-AyT$CzSgKK*8tN%p1L#AL zXiI(nLvgWGKETf!inY|uu(Ow{1*AUFwd~qhD#Orc6n|IU6-ZA{>+YrA+gnKSm*k^? zjI&y?l=CalNOn-Rn?o8))jfgZ9u;^pe ztXWIl5`o!EH7V;1_Z}zV16jER`2;!q%0+s)6F`wjsh&KXyo^TA{Uz zL82{Ha+mY3B*OrYG8Aj6MWZXpUTOxA4~X_-*TzzlI~7p;UH3*H+dQpnsa)GiDE^ZD zIFK_|E0%J8#r*`KhxQN4AeL%!(zDc97g zT}!FGxa^()tck&+E%h$Q)>5~guP8kIGRR{nm_wK~YpEq{F_!w+Q-}Qvmo!zALDd_u z*k#qMSxc=KforMqhif~-{RGNw3K*JZErz99?Zy?)jF_GLD2{@$iOgDRaFZ<3qpk*0 zZ4w!YW2sB@P2dVLkfg&Olywh$Wl#N|1K+M3Shsi3)pT~3dSis8My6`2mZNopL82{{ zdcJcmbqMe=L$Q`BH5A`k@!qjtf&5N13%fR!I@A>RZShm_WLUibiZie|@ZGwYMe&#P z`G7o2RJ39#=U3z{(P8NMhBy_+UTV>4&r(@N;Ie23NV-v0m_r&%4XRWd-+$0N)TxKJnl)>wjv{a^_3q3@&T!>HsY?Mv)2zj?)Jw;3eYY1luUG|R6PdNtOQXt3k9sOd zvq)qlj-?8ILrYbHCQ0WUSin8-y?N&U9QgjOiQmOuL|0eX-8Ob-4t%{@7E^z-6@3U2 zZK-Q>>&byH2jIMhVlDMb&x&HHDnM!y-N&wtrP6n+qxie_c0js#T6Zt?+tp_jf5|@# z$SA87OF6${Uxu{`oym^#X;|vcSw6q`7lYpDSlvpB;|1Z55d3{A5Z!&33_xN@}voWoYZ*hFS6)jv-;=~4d-(lrtp ziDRjLXKAU)&?M=}N2|IAzHiU{p9A09PjT<33|N&=e9t2gYpIWOH&cz#+SVY^mU^MB z^LE66fQK52wNzZ2Tw}W1%AU#=d*HVE8IF5#Z z)iZdsrKW*wE!D7AS>fsaAiqMve8Q|*ORZpwvD8{mjYr$F+%3Br+1mQlH?G zj0#-9bSCKvCC#~1vG4wmr9NAO>tPen)ns;;ZgYgCe#72=ZA)1|C8%vG*v7X}Z`u`<@Rdfb62fj5G6BOTMkON3wqM{W`Iltm> z1X-wNP5c%ZmYVy$XQ|Lb+@5U(NqfqQaY$pSxY~GZdjOh;8%S3a+_hB65;$f~0XEm* z(Uz(Pvb9tT{9uY%!K<@2SK7%BQlavs6FEA|+PeQL|<()l~$p zr7E>aaE2=ZN_7etnr1DArBc?yHrEZDK~^E_OqsP*Nne8WsNY8Gdn7Uv$5MAM(Nd+c zl1V!2iP~zgoSwh-11weLS3vep0~_FzcY2<21+Jc-Lsu8r-BEUD4t$0FsHkqE^|3*s zE%kU-Avy46!VuX}9I<&V@XE1`#ZqN~R3`c@yEc|;(YvVPdj(nmY3pg-z0?n%lu>+> z!ayKHtyV1MyhUIE)+m&esbe4FRRGH&Swq=%4rwgac>``$CZYLj1L?bh zyOw%=Z#mpT2KJl5qb>CaWNWE+`{os%j>q6RQGCg_0A|fvDm`0_rLypYd97&owz*Xi zP<3-GI$AYr)>1h|;99CnzDmw;uY)p;0*0npi(#n?{C}QJ;Ow&s#wIdrslrVvN{{+` zkbWYOkvNu`Mc)L*LzAS_maXJoJ^$t+EEVhOdEsHjl`aZaX%ycGc4sWrU|?od7p=_= z5^bs1cGVY4^#(k^Q0!i+EsirhJ%1C(J47e2Yh$U!c0-;g|^Y!-(!mdbj$8NUDgh~Gb>2&6v} z+_lvBKXBiz9IzS&kG9kbkgcT_j4CcX-4o08O2dBHj!CNO}*%s9(75O z%9F@Q97}!o8!dGfQ<C*^H z4SpO_bI`iPAkmf@In(*mz@31T48>Y%QTg0rsc(T?BwCwY8%tGql2!4&3ip9L_O!00 zKAc)s@l6t$u=3ea9IaT&d5c0FNJ3*j!KpYbHT`GLQr{%U!LTtTEh%fwA&sSKb-^X8 zUT7X{Al+4P*HW)d)JnevY`VdtE!7ibYpLWbl<@R6koQqAh%jr`QX|=7EcLpl#y!43 zBXt~9eGiNOteQ1zskcPnT54A!UPezrA;ek{MMBf8#jw=uL|hAQ4^AJekWMJGmdcP= zPkPjEpmidNjKs0jDEcN)1Oq4O-Tx-25n`!VFT+wv_aiNJ44*tpy*(p2zWk!Av+V93 zyEA*KX86VQ4Yb}fNVKKq4|Dz_J}rjGh~kLNf$z)Oj-`qLDMR!>c5N)xX|7UyuR|h` z=APEIRO5h8@l6)Jfef%(v6S-`huK)8(DNVSVSZTZ$`#L2f1bdGxe$_Nl*Mn2uvGSY z&*J+Jn!hxV&MCNSsSi?NFLfT+6@y1xDj&$!Ql;?=J@Rw_gJ(wZl_1QTwNz!c7)#ag z)M5Yfa=BGLP<2BrT3a=1)>3sv;96?(zTD1mBS4u%0YlTQ#jw=Uc6fGh4LG~3Le`lw zYpKIk3Q3RpG+MtSk&!r-%K3nn8jUz{;9H*EwbWO?{2xobQM(Dg{DM^k#kZK<8A}~l zmrPYhYeR!XTWY{*=P$jx1MX!gb}vN0H)-qyl4P}FDd#N`*AaRsz7ZZhg{3N8^(?icUJYEzgX95ayE&xUOZ~ST-{{i( z3NDI3`jFtRrM^Fx0`ElyR>|PemO2Kqwba={iNe!eK<-1qdBUt&OI=}$vD6Ju9rmYq zA(_>AxFX3L}o2@ z@RfSfqb>|m2@)BJW2x|CTIx8nps#josH(_4zGtq%QY%#Yh|}{e_~h-ShEBnKh*!|n zXm;0aTZE@7NqAfM~*~W4%wHfdZL$Q`>blkZv^$n17MEkL8W2w%U8Y;fm z;!hy=JgsY~YzbKu-{g@NE1wa?(Tb&z<_wC&xX$x{x%YY!rty zmRgq{588D@b3X&=w*+@B6+VKA7z=E&!J{qpF38qWl}2V2p56fRE(+!lX3bh^30sV% zKK9gMf3tSQ)ge&zO)MT-HEY&V>qX#NYWdDYXSm#mwG@hkrdf+&sjQ=L54|NgVXI(l zBD0pd@kT}IQI7=abrKngV=1lSn?OEjl5~};HB~CH)Jr$84h>R9TB;2`d6wET1JB$Y zMpwt#-80)GEHyJnDs>sHHw_YPsWv|}6icPR5NS{xu{rSl{bqSNmns0HDA8xxwXsx3 zok{V%8nuBm^t7&}+LfuO_$HC=KzdoNSju^e##F3P=+ntK3&dXP!f&3XzU+uEIddRc zKv@+IX)M*H44xO=gy!7_(sc!QE%o0^xaNBj*ad?}TdEPr)>47{O@ya?7(5Qe*M=}_ z)>7TsVl36mQ{$&CwNt4apy~uHnpib!)>1Euz_rxkfrXsmUIt|x1q@BI7Q<4>A7CR{ z2F@0%U~D3@ma0FkuJovnf^>>RM&ej%DSZIjeL*G2(eB )3a3Jp|}Te1Cm>m{lX!Q zrHXgJHk(Q7Jd2y))gh@v z*>nzRETslxcibM$&l^a8D7b5>C4b>uYB;d*29LJX5|FK>@=T~EJiQ9!EflOH%$l{- zcD5KxeeS8l{u#s5sJ)=-t61EzYSyf!z7&CLsZ$-BIm2Z|tk0rIXqvScmP%Y-Sm{RK zbhHY_CNgWOy_4!lk9r75BS>T}s})N*Z!virLx$eSk0TWZ$3 zLmEpJ?15wA8Z>V;kZvfrYpJ&Hq*D58U}p^;ZK-A;TT3nbAcOGqV~|r~1$-R{vt}*T zlP$(l{XBKp|JT=bR3=b$2#b1F&6>5;OCoSBwKZJL8LmGlqbOi#nza~~YLyGS>;>Sg zvkJy0GHa=q3nxgA`T$5rNMt0ArCtuwQUjq$(x-pV@Qd<*266;$Ka{X%wXj} z@f~G%W-k@^+^0&RHDr)zOBL_#T$gGExV@p+z0{FKMa5Dt0eOY!x9r+jY7<`4#`mhc z2V|zFH3z;I8!Em@W-X9SRx6fr-lFn7LJz6DxGM@v6}jhGs#{s?>n}rcowDm3(paix zDeTmqpgGlD7t(hHcP-VR5PpuD3s@0@M_cL<$ktN7JXctFx;e-lDM-ExFl*LQ>Dgi| zm4zS7fv?pEA=MjHJq3%oR?V8VR8A4NmKv12xij47pd6!sp=s7)SgKh&eB-+X&VN?H z*hFS6b-POw=}~9HH0DN;NE}OT%SlV^#hfMS<#j8o4su;8-F;Z<1z+TKsUi5}ou02- zi;IJO(A7Y8Hv)lJOBHP!RO8S(#URm^x-lN#fU%ca4tR~BSWBI6TU9Le1(3r;C$MW{ zsfm~CDZba`Cm>fnt!t^rCvqsh$>u4LnPYr zm^Evu&)8xtwa-(B{h#1x$~!^Tm$3NFs#&v^IwS(uQdhf{afXXWta(u+G|gHJOI>T7 z8o&Por?piuHj!CN{X4O+^r&A1>17fbiDRk9d1lL-MOcu~d)0VX1_|fb2cK z=J@1UYTlr7`0|Uc_OiQ5pG8}|RN*r~bmpcB? zv(%H7b#Mg`lBtxn=8(ox@g;GbT!!XP45YgX?pi8;5OS1`)l< zu8pNyHOs5`UYm(Prg&O+FO|@usp6Y_mIGO1wPGpfEjFhSdgzmacv=MqzI=~8OO+Xn zbEzL7`I)k(9MV`S|GK95{)1-yj|=I*ZsM+`o~*@1h9Izf29LH>dXTN9mQ+hEJlzoF z))eF*%$l`SA+{Jx75CJ*dcHZe>JF-Y3ybMi&6>4TSrNFFnsKPQGu$>%4pP9-G;1*| zb^Lv7U)RBTXcdf2WY$u}Q`D9obp}jhRuqZEvDD8vTULRcn6o6kYjR%KQvQGckEOot zh>bUlu6nb(sR+bcYTrY@8im%k3=(ar8*mhmZ}qmL;C?6X7FfBeFn0%)P;BR2~W=lc?AUr3A1J` zb&@T{Qr~*&uzz>HylOM3`aBj_teQ1zsf!|TEp@4HNoTkKV$F^sp=s7)SZZar1iZ=> zoF-Pm*hFS6RkcSQ=~2G`QhyQ|iDRk4WofCj&?M zbhVS+Rs9^@nY~nlbNSRUw4OIew58_ccAl5I4fuhfSW8_9XO#nAI;=ou6kmOIZ7j6~ z-{|>XpAtaI5tV_>fv@`W`4rz|)EG!hs})N*Z}Az4;X+Br@T3C0eRGG)emJ>jhj(tyh_1TdI92Y`iJK zN{iyV%I=J%5?a?)dC^+jAkmiEcf6eJrRoB1WGL2B-(627mg)hd579g9+E{AT-^mr< zEA%>$w>+)8mui_WkK&t@76MsjwPGpfEkZ{Tdg%NUe7l3Ca`-a2zG>-yF>f+l%7f&4 z%6xkwEVT!hC%9{=aCv-zOpmMjIZ#9zZK+HkTT4A%l2&*+0punWl(SZZT6T52=qD@mXIwxC)idnuI+mb#w+$ex}j zs~vfIK5PkI-qR6Xb!T_Wv8L8iZSyr&L(w|cAkmg;{ApUT)EvMI48>ZiQ29(^sZW9I zBD#TH8%v#BU0w0LMyG(B_q4910vk#yzRBq}kOx*PmU7-=ln0HWIZvGHQUjBFmiqD( zZ1yZ-E{WpX%OTBPYG5-wa#07(%?zZE2<}?ytASaSejeCBgGXEHB*@lMEpmi}r)Pq^ zh=L1*S+kb9#uj6#o1QxCZ(kAb$pKZL#NvWgvt})IR|Kx5dgmzb4Ch0v8BruO%~}jg zJ$Ds{@(SS8vkJy0GHa=e8S6`rx+_RMNn|9BrCPx$Dv$yi+~b?wP(2h&J(mKOn$sZC zQp@ql`^MLBILadsE zo6@t?wfEqQQINbr*;o!~EcK{XetiEy^I`+($%4C`y^D2|#nYpG2la4ps5X=Z1*wxIN) zfT3yDVpu9+B7X1jHaK&wg0YFrTB=;GBGRMY2+}qZ8Hr=5zv!Dl2WXP?o!;e?lH0~B z{jgN!rjeGak58VZ7B8%;^nd6o1(?3Pdm}7W^3S3wJ6a1GB-&Dg`zo-EZP-KUJzsDiexGqb=18 zWNWGL-GK0P2;_PcbRf){wNy{G7)$l@)M01Etz_nD|mMYG0 zt3cU70YlTQ#jsSq4CV3uBXEAR3dSZfYpJFS3QLbV8KyCSB9Su`faTdK{j>|&{DfM*+uwN&>G8DuZD z9>^A=-?D3CsSTkAXsorTk zOKrt1PMtS(CRG^4cb!8TOKs19ZMGVk8yHC672LJd`tR{Qr8BVJ29LJXBap47ss(Zi zPfr1PE(OW=0cOowDm`0_rLypYxq4o_Oj)%YRQ)v;XRVqwYpI+fa4q#)iX6^xk3mU; zcqKH=S`16I_z&OsN`h0vDj1u{tfd;^Uaa)f4x}z5G7`s9=UdTIPchjz@Zr*!b1v01 z?f*HK>e(EJiWTT;ExQ}R?#$_Vwa&#=5?a4DNVKI|EUF@wx(xWbp;${5+m%i%rLY1i zQG65FwXxI_JblCWTIB{(kf;o74t!mI$*lM$uj)YRSglyfd5hJH7%uekZCo6NrFO=7 zmf9a*2$%988A;hJ4rwfPZ&Fo!|3ULS1L=DA-7tHEXHQ*kUZT&r^r}SBAu^zd_ZxuXTdX z!qeqIu0cT;!mL?K^X$piYT?GY1Ab=8zgz8sNu>dNfh$t2mMMe2P z=iHsyO~m*2em(z@mIq;5LuU5uaqYLeU+n=_H#wJrECdzAL`X9KUkL`^~G4Cj)t*0YBzcg^?rPiWw zf)!@F;^;lxfyad{f%MmFdE)r#N9qgawn(vNW(CBmo8oVRd!EZCf#(@EDvC-eu=sD5rc*Y<&{^S?X;2pW@ z&1oXRi656L8VgO0|0=3$(5X1_YDOhRmEL#K5mRGvf!9XD2H4p0Eo|_PT!*SO6OHV{ zHEHx5C;et>aA2g$j?_Sy#{CEl-jOT3cB*Kc`z(frw!*L7(cY5BnN{|~->9;`RHELI z>+H^0(fD(Vo2I5?wO3PPL@RG4zp@${~b$` z@Qz#|P0_reF$C9VO-@BowQ}B)k8*i6E<`}1Mn7oqj$G$%>Q?gn+Kv==(MjD*jT1|~ z^%3=f9hbrjpuszGEnepm8=djVwOUzt(ZkeuC)pdTp}P^QL9Ot33-8F)?s#L-_%5hE zCex^o~Ot{CLZpbBQY|e)U@5puF$28Hndb}o8j>50KrbaNnS1uZv5zuJW9vZwO z*T!XNiQL+A^6B>m_}Y`pTk`+xrf(k7C4b1)k=ophqK>A9r?$5oZJ)#sku3(n2JcXg z&FI*$;h1HkE-y#X6jQ@q&KuqJJ5cAYXQ9D6a(%NH=ZcB;z`t+ANGsW4YRoUy2Uo3u z2Jgs~wKh;}9Lb2G(XTk^D^nx5ytm|k<=e2D+9qi5j$BPPv=fb1D-9|-=%kqP-jaVh z-m7un8_<|{6dJrkD+ve{jaM!;qxZjcQX5m_KYQ(lkvdjY+a%y?kWPBi)VLpuI!S$W zJB&VPd=DGEBiD@+)pW`4Au0J5Q{zrYZ*-*(&Mb$y6Yt3N(<2Vis6E&ajn7PtZl&s@ zM?+}vj$D(s*Ai8 zda!6r#dypmA8Bg*XOEg!p^j)Q-PTSt)|eXq+0SQWw-k+EerqKfXHAX&?C1SLvqa;a zl18Et;P#gMPz!IndVC&6hL0;I>m9jL8@3jWn}gemMz*Q(pZ$D(W|(N;{Tb01Y-;>x zORTk4+t?hg>%7p^So4UtB|f+geK0W>Uh$4xM~=G1#-@^5V&jylQ6tAYmVEETMC6Hl zXz-3)_w2&BiyEw25-b}3m>Tz#I^%Tj2o2tmYu#RqCD8bGZMRNM~s9 z4(GuyFqS}L;j%2zc+k{XcqMPsq4@t-4HcE|RjvF3bp z(Ku>qj6Lm*)#PsjN&WE{Z19d;x$bz;`2A8#(YR%5JYOohX{(^YJ972z*-A9p&eqYb z8UKIvk2i77;BtI1G*fKcZ))VV^?K#T*;uW612%X^t~IlDISMx>i^g-NMvYP$L*Il3 z@5puSQbVy3o7h4$icO9GJlFp9BE~Ct_0_d7(YRr148kk)GVYc?1g|<3!3OWh72iUa zV@Ms{Ka%ii3z^{l*VvxYJtBaT9twPOqK9}Qbw4Z2o0TVt(7>TFdKTLBCQKX*#2`Nz zS3@N>nrDx9HKJPOf#<6f5R>h$qhjAh^WFKM-=eNO{3hogch8X>E z?BL76`8M8P4hS)Rn(idWgGd-g&zTsBWp5Zfbq_H<#@@jJn{VU2AsHdYKg|20WURrZ zi6I#wMjz-K#yhi^Z@dAzRz>}^|9c%^bQrV@6%J#{~VK=gIMm|l70eu4TS?|^;#iUfLrIZcl7P11ijsM1tp!O)*s1P!g%0ncudc!bXzF z^z4kRLxqhc;qmO-n@ELCB+=XxIkXBDHkU*j&#u|*A*4!xy~Um*z;le0xE1HUXS+N7~O#i3nVed^F;&_Z9~VR2;)4lFIJ_( z4$=9D>l9DNiPfpFW8_#Qrg;V)cT!;|;{w>GdlvkH@^p#(9*L(tXNT6O!u!Iff?&R< zAgUo1_Ky7>lBJ%o=TXx8<7{>DBaLV8BnK7t5qCCw${&xV!oCi!|81TzeG{m#U&ysd zPTJ#XS{p&^FTNFf?5*I=gN6ZuBa|43YFHjDRT5&9z^l#b%Ie#DL^A0Wh3eua^sMA>MaEemg6)v-s4@6btHmk$0)uR<(X?;q7kh_j4 zoU#MLKW$;$#wxLp`;IA0r8=~ti550wp_rHExLI9#q9(1pPl>7fEFr67@N+#@*-a~^ z*usY5KPyZqi`c9>a=AXzel0S^^n8GGWE)y>QRi<~um6=vD+-k2h#qDXj}wwooJ1VB ziEmcr@1)U+R29j%pBN``Cg49oiXR;3mDZDED@&s@V9X}PRHn^pYB&1mJr znrn`Q&Y2{L(YQ!>s&d`BTv|^t$xO+?M zZC2?G6KF*Xy9g&)ioURM9x9@05>5pHV@8l?Rj?#H(Z6H8L5FuluJU>gtqPGufM@08 zdbBE35{9Q)N+VhoCW$c5uU7#JmxR;v^7uGf6(I?ir%p%%TBT1QZcoo+F|^7ll8T=D zYT0R3lq3>7ZCOQ)4n1<=6)$e`zS?n@qd?$OQ-Vg%dgv%#TR=ngY%VMX}Q zy?chAFlFQ$jcCP7mG~dy*O@Zu+@!^;V_D20KUB&xrQAt$(292|@juqSWNCBCD?S{{ zV%DY<|B_Qmjr0gw@oFXh$M`BN<^!*75x*SE|5&@JrOl+FJZ(aS*NeqZ0q)j5U_FFQh4g`8}5ZF~t#!!iT{G;SvV0Daut3PF=EC zysa|mO9+cpOf3lQsb=w{18_~C>0GaaCRMDei~^$iT@Z!O_bw;5rzPxboBL zv?jRRJY@_belwq$6>?`}YaU@+uA1Wo=(s$;Z3?1wiIRx%tiqCt>ylg-VADPBnSPZ>;TSe-rlS2N-J~3@l_%p8YU^);{EtiwaNufemQw07(SkN}!-l zFsFpgX;|P_7SnKlrV$p?&+8k~+96)1byx>>?a1H* z%=CqAnEuCZ8J?Mgu+ml)k<{y<%BzU$L*P^wLNg~NJ;d(V% zJ4RrsG=>Y(59;2=kod@`*0jDt1s?Ab>pj$l*4K$)$bx{>^{rIEH!j)=@N}hK0<3k# ztBBfkwQ_Gdt-bE*(k-HBa@@hSF((ISEOov;t_?sd-ms+?a3zv&o8Q|v>$dz~{GzeX5F z0`4{&MmqK(-o4HliMyVL5s7;uhJl-bkm|k8>61c~KBaXF>c26|7?6&(dJn#*y-{Bg zlIQS%G}@@I2n~<%hl4gYmCH7#r(I-Q+Sp1GZcnGT;%H-QNhEkWF+rXrQauNlAYT%h zo&Y9jBMFaZ^R@=Gu|N{dJ+DNh)5f-vXyf^(cV*hxPVO9Z_So)6qP--#cuqB_O&dE( zqN``nJb2Jq68Cu8ugRp1T_w@YbADV!+SpAJ-949LE78W@lIY=?ce@5{yk8RcdX6z$ zKS}iTJj84dNa8+E8)kb*61_YhG21{%^!8k2w&9Yv-!q5VMo6L$W+jvRw5N?vM({{E z1(DPjXFR{6V*7;%6k}q{JmiNMm22TNU^g~3vY&3+p5Ef92W<(VO}LCkbD>|&@!ekV>@a9(^r2m4&>6%541mCcyRL)fn2;(-zJDxp?a>NRfIfye_>Q zr+DJFkFzCoo7GfgZyKU!8}Iw&f5iDCmEzKWXXO{Espk^u6_@njnA-l{>8nYe!>C-~ zX<@*CA&=9hZuSVSf-J_eOXe)DdbcRr^rTJdvmFb){`~b=CT;3rkK`H{%=zqq>~NC( z?mVJiXqQvmOjF66B|jV0o?d-KQX5&ur+VLYF0|jnt$|G|wf3>QAa2=e-e*T!u!n!g zQXZtr7V2uiG7j-5TlgJ@NP~JlpfDIOTfC!BueWT=I8~y;tKw%zB`2jK^3Z5k`H zX8x5zyTwPSJ8sbqJ}y$*c2%WK+vKD@q!dhBDpJ2!$)Zg=W$OFPT>zVvdqWa!a_Ta! zH4$K5i91oR&X?jG=4_e&^MuKa-mHNY=+%3~&gT|DCTN@0pIj%s?3KBF-)12RlNnQeGh0dFC2F{0Tu_@fot6nxkGldcaL^{mb;P5k z04UFkZNWZCCTq`?mZkrU-~63hLY%vh?-}t@8f^)ads*c9U}per36JDkdgzQ-ZnmH; z5q4JpHlRyU5*4OJ3~iO)9#wJ@c_9Yh>5~%}-_??n8&}sMB!=eUrWOEv=)f&4-0Q`m zVBNSgC0KOvU*XBtZqe+BgY$2w*KjV_Iwv9l`=Jl;_gW69Lfl3rZQ}JJ)awx zL|cE8(9B{0j~cqgZB{Q|%cZ!A`XYBL3(NFiFGcqX9@=_aDBdkap`)`|Ex5NMZQY{% z{D~2)sK+5`x|>zwRr$2_3ylb`XCn9rB}oDrmzduoX4`Gjm(xrL@{ecWe)XzM}=_5{D|Gk7&<>te}XVrFZVTegG0 z#?sbZw(wUiJGX2dFQRQ_s_-Lbp2QfA;kg&uT2DLnT`8eBrN@b4dj*2MtrEQ=Rc?D@ z+PY81@JRe5Y*NkWYoo1`b$Dt?K9AuN{u}D0_RVNpy6D_c8-ufHTZSY;jI;P%$dLzp zh_SIX4*6H+5QBd$a_|EHA;t-~=oo}ULk4W^NrbOuICN9AlT0cRMJ5n{~c?E1(yE{5XU24e$QKNIrfS`w-*(C2XqEGWK8rn_43taE2inA;vP!-Yp*RVTkc{ zA?wdx&U-!%Fy!)(Y`M5gvAx3?aB*aJ+t-0XD?@B9h zxzGu<*sy^NVj;%bS_mJf#@JCGUCiu`1>Wn5LCBHeG!p8Up>D{&YTimd#6`NL_deJyRXE{n{vrvb!X<$~0&(^3~UNhGqAN zvEc=_lgP0($ioBwU!4RT#1-NyQTC|NKRvki4UK9?ZL-H0zao*$iR=mCi`O-xK%_&l z*_uGpE14OdloUu+klVqO=!ry2X79wihj}-RcTe+9yoo*yv)2MT4t#>L$2tm-pE!3; zEM-6L=+4KdCx=k>6OQ40ye-p3+2eG65g(6t@WqtpLTCYHFOeTk$kTlxyj*IqQN#1@ zuq?`67RAP$9_rkPvI`yTgxfQtQ60)&?qX*WJkR05jqDYn>{RM6YRG_ol)W;Vz0353 z*F~sSx!6k&Ile+3RX>#k-3BUt74XqY*#Zpw1sG1HRDexUx+FQ{nTX@B|Wa zl$r{}+|nq`(_7i+?8Q&GY8|AnS%?G76C^x+l&u3Avp5o3x74~;G z2VI5i7fYUML)&ONWB_FkQ_;~Q>gtvh)(~%kVY9s>HkRhsr|d^m^k)XvWfNaod#0PutfqVIG|$x{w)Wh&_82lXg> zmNGuAmO<}veAWQE(Sj#QTgq3CUV?t6j)$`6s;H2PHRv3p-v#mmA6i;l8|8Rd(0|9e zDSM%E44<4z7Z`n;k&VpqCP{lNP>x3hU9c%14Nti)maj@*GP)*^nLf0%*tW_sNziRW z&?%Rz&=LK@=qjUo0@>F>2c5$g0J2}gm&I=u5H~(G4zfv>e88;_Lh)G@Rju}05fqPV zQiYItLwLlaDEqQ1pIs$`IH+nTWP4dxTHO_Uzv>kz`#V+ftEE-xEbB1?y<%Gx6HM9v zsDuem*ysiI_cQU7jTfEbQp1S*d3NGu*sNp=lg1GL^UDs(u3}5R@^K>l0C5}|dlhWD zF??>n`a~1TKB&rN--y>926Q97JC31`%^kFlK{LuUcDSSF&n-{67`{^Z0BcFGU`Hi; zT!OZDdg?3~|-&oy{xErfm*!{zp5}|mYY60zN z;<{B-OIJ^96FdNb%`2X;umbJq9%+k9ktf_$Hbjlfh2BMq zA?;Z082N637LS2whAGw)JgKpm|Ltx}J6&$h$|q~#at1kdFU@HOcp z20yWuy38ikbku(8)7n%XwU>ZR{8y(rLf;2Sl#kli_H8Fou3uVC4z4~rWPhkmCT(93 zsusC%@@Y6SkhU)heFfO+)obL@#KAW8N>OIVh;aI*8BN z+4Fm%4g&I1(wb1>8jVE`452{XGyOD8F^=zL@j(pFujuevr)zMfAMHaZi z4yq2g2n^5PI<6T-KoxnB2UU;O&T!MtOi4xTRt@0q&N`9`*{v=!OI=BY?pANS3zjTNh3!_c2VlLPq#}O^7={Au ztS>1i{M=Q6c4kW|YPaG7?aYZRL9C-&zBwtZ!WJU_-$?#X4(sjxKSlme3ESZPe?tCG z4g1mi-?f$XriJBr|9A8LALaf3jQ9Tr@Bd=&|MTAezj^;hy~gvHd3f%AS4wF0Tui!T zQnPDX=nu^ibKBf3yymfMnKE1-Omfq%$?sZtoX9$e`-Y&Ot-*r z%~qavM}!7~ZnmB0NV_+L(o$PMbWPfYaoXLfKp4Dwn#5~3#~C_(EM`-)Z{W?|oLFN# zIMQ|u3ZNXf{pcAF1z|FJFY@|ev;7>2-g(}3Z(M82sc7G~pfT~K(4&?J&tTy=y=h;R zA*Z{s9f=e!Moy2IkC$V81b{anPhSm<$gI+Y!sS{DZ=*-HhDtx0`j7`h zvozhQ1KqTzvE%B194-C;qECI|Y74VyPcz59Z)<4rFA)8~Vj*|`Z!he&sBur->Uhrp zXOC#a%8kt{9(A@g?P(qJQIdXnLDhjM*Ay!*ne52@(6be3Pd9t{0oCOBL)8nQ{tT4j zxgACEwC7&u&6<%UO5-7#Y>Ahs<%r|(JGdqF9A~m?`s{}Xw@{(!b^gM@pVQK~0;z5w ztzy=F*Gd}cd&pBBJY=o1@n(n|k*}$QbEes7x;%x|K0p)K%16`x9;eWRBTtRPRiIZ{ z4k`Vi!>uq0-Dot2xZ6?m3$>|I?@8Hokl#XSchW=q=HWD-Uh)AXfDQ74xjfA&xV;zl z#d3xfC!l%(d(ID*TP}|Vcksf_zQnKrgRojI{5%PmS={x@s)(O^k*ABxW%e(rq$xcD zpPy#s=+-{vpo(Fle7ltLzhcFG%Z$~#m?tvLLw+;`_`)lXENzgdec&khj{Lo0iExE` zBY7t=S> z<^qJ$xO*1Ghn4%nO+yjaI7q7aq!+qETDknb>>R2Bcmqh9u$1X3O{9sgd(+zyd)*z9 z{-#uqooUyNs#H|NsIa?%bdPZmPiF`lSJj1O5xRa+x?wA>Aduw{Z!8TNck{=ZR8%)m z%3}*CkaCcNZKhxBiKe}`yzwPe68kq&v^U1lA8iyvc+m@P1T64(Jng=>ieb+;0SS)7 zc<}>(k$qzT*g6OUIs#u40JDCIajUrJ1aIJ3K!4b7wrfVsjeTx|C%CpY2*1p9yK{c$ zZFQ*|VEwS!SW&&t(MnOGAc8TFLCxIRwyb8B-3lVHb~c ztrZW{bI>a$Y{Pkmz5AB#qZxj%Ru%H- z0~3~dhG9R!0JC-XgH4>!hU`3XwB;j1H765mpXJzUEoUO}>T4ww5b`mCDG~GtyA)djI=a<06 zb<#;#_Fm^U^aAqNShAO;ldx=m?Xdj4cG?Ni5lhTRsw#GrI5;qq_SbOKTwH-XM!z<+ zzit#u-JWyXiT2lXF{+JplDC>9{DRl7U%VVcHWbdlW_w+C64#3nw8h<)R0Pl!Sil~d zorE*fdtjX9`nm^Fvz*sHzI!)KnaDb)3mr#-C2Qwy12 zOtp?U&);EA>ysU0^pRlDSH4B-hLKnOj9h9JAV%g}My#oeDobt6U&6hF!ENCpY;`+=ip}E(O=Ah9;Xc{BW z*SLH~-q4nw?wEw>(RkS#B^w82GdfXyaR;Cq@jptx@^=lfc-+t8ljkal#n;%P`0Sfy zmG$0d)_Yy#{23=WRA!4#zDT_Ch3<9bkXM$_M_u~Jzs2~B{MQjm(!L*x3d3{!4BtEw zBPDsvGHi7re}1dafaY)sPNJZ>hsda~?6-hYS7&yq}=ebl}quaV!E;B8kJ+2d!V_WBM|rlof=B6a@X z{9?{a{o!HVck6WV`Cs*iT0(6@p7!60x?l8sd1WnG)!`c~!+m*P!@H|vnvOEf^)pSeNE?Zxor5QE9R{WdpLZT4f9kkMItvIo2`8k z&I~g4n3oNc?=O(Ss4m1FhTIE?b=eTsxC8NOmAH_ zoC^!3t0==h%c5^1=g-x6sdn_UHD8Mok)p(&&7b;f7WlfXX1lwFvTL4LfP zG|A(QWO9Sd(TzQfG?epD1E`QC<(}`UZ1BxpM4dwU)<~~ z7sv9o`<{;UEHaieWieAS$d8wk;a>8dFE*2a#CO*LISQFc&i~n1MvfTaLRd!f?yW#S zAtMlaw##gUL1is*Ubw%x+q3aAEm}rDF&AQ-c7eV*c2*0i@KhQVK-!Io!V9S=5JCL3W;SxcN3?r-kRBk2;5g<>wmI7Oguft}T5rW~nAPeD4H ziSJ`#5lcYih|(Q1Vk96x3Q43<9{TL7kUf1T0lso^EMM1;sX+fj#^+4=6jL(DkC&6- zUh*&SLL(2Veecr&=?$4l&i~n1#=HNp5iW#fWIA3vs0W<_Y_`>Ggh6F3abCE;xrAr& z#0px-Ix&aGcWx(8Mgn$LKswKDO7kFH%EZT*Si}+#IimDGUTiA?p1*4 z>?;??^7XR^W9SQH{KS;sF(rficsUvFC2xPqCIK1cC7%fy2ZG7@KO4)q^#>c_LRdz2 z_Gv|}q0~vWhQRG#3Gh} z$PuNFzDIYNsNOmtq0nbvg>2HV>?;??^7ZG|wdiMLDBx|qn36$$yqpa8l858PYHlG@ zyyRUVGs*cs8_PKI3mf4=SVo35YDYbwGmMQ)WFrhJYl-v1{mnf;(=GvdN6ew6{0#aS z?5u!fUCg4_A>GTw8<|+d5)e6}^da@@OAD!XzYa+KpR~_Hw)sc)m5XEf8nZnItAF9C z4{WyMOvxZWUQUL4$!lDwD*;*HC7%Ep2ZG7@KO4*V;RiOtg|Lixx>cn?(0PiDe9uN0 zRMrybh5MVkT|Ynqa!$;J7<;aPJ_9=|Aa&lXO~sI&Wa9EGKrCVjh#XP+q{JKvNLC*m z5CwhqRmjGC%f51PEML<$H>PsHCStRBn36$$yqpa8l4sA$m4K}ElIKHalJkEyma+FL z8{tA&MqCd!pmETd$3}Xy5eAjD#ChTV=3ed}B?0+H%!L?>u7ExWJ1ZbRR%=dYAicoE zGniP!5)e6}^b<|-;uL&q)mI0k`q$cLAzS(t`^v?!eBD%~8YKZ+51VZhQ!>bpmy_XM z@~`kQaPBg1c*%!B#(`jR{?EoTKD*3DxDb|+-qMdEx%%8XirQ zfZP&uA;zZ9L0<_wD`vfc*5C_EpFpLV_~5 zIF_#!W^|zXz_!L_i&+Uu2Kn)FGTcl4)IC@>0tG+fCC~OF=YX)WjF&I65iW#fWJq=+ ztQ~*_AK6F_8(~mcOPm+(XD(&NuXQM=Y0Q^3ZFCd2+>QQNAflW6MiR@v2mb;#eP`}$ z{aC`~1~973RdC9(8jYFb=8#z7=%G3Iid#4uBOQJ}(Z7l0n!POBG(5@=wyZ6!h3z%X z#*f1a+w;sOSMztJ^j$%h<$ZWAej;xg(-M!SaVc(>;r)_%*~-6)E0jK7x!_J2-LE0D zZF{zwuyuPt;%0eSA8AeGqvRuR1bu9!k0_-mv4DA-u5T;6OEoVSRZ2hI;$1lfcLB@R zhuoY@sWsg}IhipZEwT&8+nR$bP~&w8{>vQ7ZGD z0MTqduyp@VL;7EByR7w_--NRD`5BFbV@ny1=@y5tW=bD<-SYAN*Ab=U)rw+Un;&wE zt)I%+I&azH2>y$g!$>Jj9_hg2N?K8>aKt<;W#;y$#}(b)5~o#QO9t+m@HpGchmtnX z8fhIFzlO`2`o5>;*q$mJ$z${=xJq;I^K!y5=WZNjeho^wv(CXu@%%iz)>sEf2O z1pgo1z}H-LcMQAJh=4C&`)n&}or`m6T@l@s>XF zceP9$r1YipD5d*8i9>$dT(R16Sm`*_K-#Yi9O=7TU~&x0lki;2){x@(bb0D+9JJ+W z0Dkf<^ivGhZ_jj7D4w4tewI%0lW~-?r&%l){M65N#D$~U1um9U`xbV=SM7c-lrD=5 zxRi2aL>RN}pO_u(l>1A(vRCC zjd{w4+Cq;JrQ_BBX}^YGHM+aE2CHT;`x_C*ak#xh^A?ujomqVS#?Sozevu!6R5q!En=jU%4%Exu;$%b6s0V>P@QdVzKm~bVK8c=HLY{D z2WfvztHLN{UYGW)`E}J$(R@*BvK491PqC&IKRGJvx9>MoN~zOE+R>se`sDO!8Mb#! zKm47MMs4jVW%S3G7iPxnSXn_lqES{9^f{HQi&9R#C+)v0mNMoNx6&xf@7hTFTCBgz zeq|qUN6J3oIfrmeD#J07IsD@7^U-zYo3?UB-?rW%XY_-B>i*9<@sF7&b3UzmPv$j| zsn6qlMkiOfS~Z1Y2bK`$X9WUAdntZoijlFkaKDjv&td1KScOBilXoXL({=P*zEFJG zH&-a~O^UiW&=iZ@ccEB^Op~H%1EDzc3x+#*vEQV4jwyI9qE9TJ)dY{UgFG&r^jXQ% zCrpl#;`xa_PJYtWV?Tf5wg6gErjfXNp^_I!sdcZ#?Oj;@ z3qF$3%~h#JnvmV-pj%gsN4@ch#6kFQt16zF2Qsb{Z+xDz!af%gi4?i|VzF+XeAeBd zAK`k&Bzu_2B-rdvc>R%Py~>fFj}mtFOtl6^*QOk4Z}tv9T%Zvll28-Z=uv&V0_gkp+YKPAIBZ{Mt_9|AMY8S=Ph$uJJ zZvs?)1XLsfQz%yk9{i zy?m0|Z{TID-I3e2G!n@eOX6Dfeq(Xt@v3R!#yW^()LV)J|AwfC=*xKcJL*uxrhL1+ zdH5C+$;*o=9lx_tR6)(Z-l#@7c(wH-mdMMADYHJULs2;?4K8Pk&d)4i2)$(gIXR9z zS+?+oKi5EN)9iHe)Ubu0n&Y7>_9@peqO4HWoX2oCbS!ivg0f?6o#Z!d$o1Rz_<3}s zs@VMfI`X>CeWl=D`A}7H;Q>SBV@x?cXJ7bjbGrFX0ha$#6)64fw>;#|ndsaDgs1gu zuslf?U&NwjKj7pvSrUflcHeNynG(i+I4NiMi9B)^#|1xwK36%cdoC3xIb`u^$2*|; zF`!dwT`Eqge3q@oltk8};u_&OQ5pFBH*)B4++SAso(o>|5m{Qy;7>CFM1Ha<`OY@L+*3+}w6!|XMF!q#84THDmR#y{h*+!_KIRri0yPK1 z#TFopeUcYDi?b7q4vM0Yp|(Q4r!?kG`>?A{DsB??*6nmzhg+Qm;9~~4nWOU8%^^ii%rTZ1Zls7FU8O(GpJ>IC20L7gS7}(M*dQ zL)5|&yQ4m9L&Y5`9_Z3gi+e-#fF-UxFCm_at2v(MJ1p)e+J#VYm!!xwZ3Uuc12@mY zx*un8_oUNDv$c35L|c5~haYZ8#rGvuz7eIx??80YCmzV+K1u&8XsyLxL3GU$SH5Sq zhl;B>=JP4mJ-$UXDjrhl=})2rqMYOK&@(oPp04x>4~(JWk(CBdO45)lh?|y%xYO{s zO7X->lT+Ji%)OBJF)`#GJgF%aPpvd(e1H}|4$)ISF~56UJhQ?tgY%`;sTUwx?h~)Z zI&j5vV&D3;I&mvhJ0aTZ6EE%3k&5TV+KW19@%s>c;uH6K1l6)2cHKQqwfF`^w|wH< z$VycFLPF}W+Vb?6iW-kG2%By`F{|fhQ}MFOC(7e-vs%b*z#{H3IyNzZ#|$c7SNYrT z!b!DJ9x>EOfX`d15-I7l?&6wg_3uRu`ze?X@VWh0fbLj8 z)+LSkK5E2xy)374KZ$4Ku=(17_QMbTbJXBZ3N&LLkOTh%?dQL?@@LdOyyY&QOoGlF zOOMAHuP6UR9UhdYF>4{;ZDFip$*4p@QZ~Da`sbH04L<|ucmDy`Q4Bhlgdw4}ithG+ zeh)$gPQb(1*mPl~h#FK+MR#1Rzigpu0#M%q@v7^jK^_&|sB0bFdO88n%K}M(f&wu^ zC~mBxUG?#L5FXZ6V*r?JfwVaRwNTN2oYtRoQOf{WZ-KNq0p+Raf%UwOzYV}~3#83` z_67#AwkozdK8q^Rx(vWC7KqJx;~>aEDz@vI_8J-f6yAErrsKdi{gJ~|?9SjIjcfo^ zdkbluApGH@Rcz~9UKbt)^znbgDd-|QJ4VHRve)Bn|3Iy=kbcezaxq?G4v*Hzy+EC? zkbceza;b{F{%4{_eg)JY7P73f0$-_O`}~um;pHY`)W@cyV8JP9cEuDbeo4j7J(aDI zO@QiPA^n^cHWLKz04g%^43+d;qAYWCneH*wmasf~qEu>}L8(D$xP_fDO2n{a= z^u2$>DQFwM`%}D2#l9Yxppn;r`qM)CIV;G0_`cfIIE{2o!dD-${i_)WvRK7#+wa!M zra;|eAF2B<gw1z8$- z5U4X2(l4@tJf~v6f7hvz-vLFF&0>{xR^S&^?B}0)U5E#?=D*<-)UroKD*jT%-n>{> z6So4Yn}zgqR*+X!?7-J6Y2-+tCRs>7X9am(#YVneUn7?SwbeqFbyncNs@VK_-YAp+ z`q97P7ny;^}v{ZRP3--`V?Q*DBYrfd z#l_*aDwR59%2_58^5&&5x@E>sDu>f-D_itEc;bdfx%&VbU;#NOlpQ#~GNu8xC@v0V z@0*6yG+Q_uGD3sApHCqNV+;B|f;p-E%p+{(^3reR*niQL5K4K;8b z2Ltc25&mS0;_EE62M{kNJ-YCA1Nsv3>lTJz&L?>pHMfq@r-uF#PvBIH|JZC*nMSTc z)A*_K+y=&s=R%}7sSszCh9sY#m`1scjg9ZtrF@9GvRE7uAJX_upWNJt;gc1yXG0+# zZ8A_==$UZJ&5JZP)F*~N4asvp>7wOs%57)NYSTunZ-iuvPr7h_66JO_Ud*c^(j$!gyQ$}Axa@8mO-=P@F?Pa`DGeL9-wu7)q(J3wJYBkF3XAC%@bt^-X zmG?D|O?2{rsN}$}wjL$|iBKsCr;UMD;N4}+y7{9zO z0fmkh>aMzp_1W8a`ar1%P3t zL9&FOy!0<{SqR}^f@8&F%P1UtAm)bC zrM1UF^fZe(%>K(LG{zF+2Okc3Vq_{lW`gQ}37}0DPfQi8B6Dzr@%n>!>25!=ODqnF z?Gp7I>T+2yLU+!Tn6GYrh}MzQ9j`mN4|yVw%`8XbcE8zy4mH#(V%WwV-4TvdGzZqO zjeG6Abfn@q<+R1H+My=)4UbhJTNYGXVzWh}x^y#;Rqtau{@RQVwJ^TSYEQPF0F3xI zNUuQX9yPr!9m)&Y(KJwQiKtmX&Si8ww*YtkWg8u87c#3_iWYBx=v7O6)5g~~(7IjAXLe4i}VrYF0TqXA8n2it6w^A-||`bhxcv@u@9i4tKV% zo{LpKf&mN1hL>UCK(}Xc*W_@w%>K=ZE`U2(pgk*nHqJ?PfXX$IDj)q&F~p1M72kMH z9(9MjzlD(-G3RZqo3y0Ez2kq}6EAmS#~nlSDwlhu`688h%>Y2c~vvK0XCJg$fQQ9xv6X?i}xHs{e97B>Iu3ZMQeh47D<5E zI{?_jps8Hs`1>Mi(~-)m%C(Le64x^jePoF%inywZ?o}s=i|vqN3k1@SQ(oXS2wsj;9Nyf*ey3>2<1Go2uk=q6X<1Uk%Dj zux1T%!Z2%aU?0Bw3&z>ZI-84Q*5IX(bXtzwS51jp$UgJVs=?iBTG3Ku90ulH3n%8x z8oaq3-@gLn3dBE`fp}|h##>1Oam>M&aIlFVvcfc_d28^{N3{e}3*rW4Al@46G{Gg1 zP7wDh4e{0BQ~Bkk2FC#KB!gCQajY6V*TW@m3@?IcnI*Pr@OD=$h>PqZ7ICn&lU5Bb ztr;)sXCeRCr*78ZBCKL{9oe@^tD^=bV)8=r2P&?x1;&l2k*eHVsKTf1GH^saqC!q# z`X#818~9}tOFJpt&sfM&Z&TfF(>dmp9dMXb6gXAGXrkKBV&_or^`hFi_0?{Z!p+y zV`)FVNLhpCVV@&2|6m6UoB0#*q>9R4ozxU>8mWSjW5}L=u0aGADdTVj<|kBYOhNfQ z@xxefe!yFe$jA>XNW%(cE&s{I3S}w<&lcpn?I9GFU!h?|tWWmWf$9{I&jWR2{>7s!K~9yu^{J$$ z3V)V)Vpyr9s{J6pf;txH32bz^h@x4PY_Cz^!TOKLUT%uCKU%ex{o#RF)@d7L`%Jmo z$old#K}mgA<}(Gf1sUoC$UiTQ(H!mU+mq35e}(LhCEwxWxYyNShL^+rJT4`$iHk~O z_&#*W$dDb!^iN!>G2|^6vqvai9#)=89*sUQqdewVaHBUw50nu<8J4yria0*%Nr>IAPzHK59 z^)+zcS-kGZ7t*L?U;N|$c&P#wU|DNy-g*@A0oPY0UhPDPYOq);$~_$)buT#_nRm$> zk`@rP@o6tv)Rsz)g}%}Crf5Ds%yHO{Jj&oC(3Hf}+umvxs);lQg z*-+=-VVdV<$aj|E5zM);<=tX6&uPd%F2ge#--|9eAG+m6Yt8dBLM7#tQS*&(s)X!R zY<_X57`L$+mBcIK=zt7r0$H0f@>A0~P)VXPHsVf3AIJt_Z+ixE@H-?T?Qh|q`#5b&-s6bb3M;ut6@mO=`3t$Off?1J`PZL+KHKDJuTG3m` zJ;wasA)$Wb^^3Z*M$?dZ`ULr3-9^?FA27Iy{J%^Y9kaa+o%pHHJ|oN19BUPG^H99p zv{E)FvF((_m?L1b?SNZ=1>hdsv4ZH8%~JGLy3MH=))A2VOqh}@@-yo4n5uNFSM;Xw z$tKpu*e8LTVq)pIZTbxx9qZ?`&5t2iwKHTT#G6fsraNxSYhR6y4Rkhr5KD^KoB@n` z2blLt)%gXmmthCk=cCLzm4N$k?BOyImH0p;bN|8x_4|X0(Y7Vr-=sQ2*GZ z)Mjy7od}>~k4M+P)zHKSGqwP@`$}Pztju#pHL6>ej!lWqt5@DsFj(P9fM;3?>>`4& z4!1mx&2%@d_S4 zm%F0s$qraW3~KnH&e(^58)9MsTU9=pj_r2lMCc_w0E=MQGl0Bc z!o+L%+6yDfu>;PBp2w-*=8RU>RuS9D@tjU4_wfe7FD}I^)B#tzgwXMpVUH$aq^J*k zBi89K4ndl&M3xtp)E_JNbV2_8rcAe)D;hA!{@(Zwq#lKEq6O&540`f+d~n>Y{@K(Q zIcf<2YfFQ?EW_>JycvKBh48Qi7~sb;BBtTd49#*8fUBiJF<)~zM%k}+#zPx_LuR0K zQV#EAi*ITl-)X#IUV zVC#4+gPbSAq@OKZ!F>|-gBcW$h0_Z=(qn}P*$Hrd$d;bvR5|QO0Mmk!HQ+l4f3pDL z4Cu(r1Fkf}$A6cVqjK2%v_+VC4a*}{hal4uie?OJo`|s5i9w5N4OwSPF88ptJj%X( zgNxKa$VOXo8A9*KNPD7u#F|xE)=0}`Rto`K%HXc-;T;)jPq;(AYv3@4PbIj$z`SGP zbVIu%W9^9wAst@MFu^tfe*@rk6HFInusu=PcyJH800vvZgcf2Dz$S(@;YAs3PgFJj zT93t!Fxm>R9ze}X1MkRidm<^svBfY)IjLp!0D!~Gf_c0>krMLD%@7kDB+SnNxU4Lg z2iy}iL*Bfnoe2&W@IC-fmICV$_l}IXCo)1F{Ij*GVTi_cpl?|kGUVQoQR_sV$do1p z)~F@m_~n=cVbk?+QAXPn*~Z!3c>E5dtzeo$-qXaWyA8L{!*F||iE&_c1J(-_y~m+5 z%hL0Ww|8W`J<;5#b~eHqZ$)D>&~KE|KrQmw;sh4fTN%RC4gsD4=u#=5?znek+&$68 z`0ay6CUt~p*wLdRu<3f2G4i4eyeHZjN#`)6Vc->DEr1%71@g#yqLWd5Bpw^X$Sc6^ z01dQ&zMHI<{-?uIS+poObN$74BoR6w+YpZ zOLjW`Sy+!*sd#4|_+r*M!vl|3b~!BT&1BsBNB$*Krao0$8hwc@RT+pHrBOjL0OG?C52#FHX4Mja0tQ`WmLM4c*Cl_s>q>65oPeWpNXTT3X_fQqcKf zi?^oHy~ur#B^*PwVh+lgb{~G3+P|>eoa^Cq9=T6KJjGJ>EK=#T|Fz`vn{Z|eXg6@Y ztVtZY=a5|=e7QK1gvPZ(16o z?YQkLKWa@~A?jm^i{p!Y~%%DWxIF{Gi;)s>bE#vuH6ytILSvxQI3F zeLQjW21IY0V%-MQ_8dv2leHrj|82<0^c=(&7;-?&@A%tAC+kLbZW}2>{I8Jw>66Zi z38Rzs!-H=$Bo0XQOZZ6`n^$)-zJhboV-%0pqh=-~*_JeUer`QF+1PmWQW~{IZg-Y& zfDnP=!HbB(2#qYr^-Oj$8b4JSJp11nUfry*LzV)j^Tr&VZHlTP+E5;tOYfhQ_n zg?PIOp>cmyOQw?p^^ceKabIKFcJg649Ah^!~OmG zXm!o8*;;}gx&-IW|anc zSuWaBU%|ItRzSGP0&MkT`K&^Ee7a^i2*9b*pqPv7>lgNO{c)G{5@g?-a@|^F7JBNN z$X9-X<2Fz427z?yyV!&cc5{+=K%9Oy zDf)CHs(~EU60(jgug$5#M|_|KNk@lQ)}X-vJZ6D-U3QacNz&4hjWuW<081@UeKwce zzj`9=>t+vX-GR11wA&JQD|QmbK9S#Rk2+VEPC@pOC2z#)dhpuUGwxlgC^mkA{IAlO zwnEX{UTu<{loJuCfK3-hT=>S$*BSfnkE?($n{tWkLEfx1#?uFb`@R8T;uM++hc%SP zY{_tN;B&72fHUvDp~2@c#HSAF5QH=Djeuhl@@e1ju-j+RYt%fDE@I+9_I*eR%SPDm zgy6)gUPk^-Q>NST5K$gw@9-$5L+S*CADRFaDhvV`ZNGQE0^mCcf3pBl3>YXSdn76~ zKagUPp_CJYO$BV?f@aa6N24YNxV)hH05rEi!X&5HePg0}ZA#Xl`ylUUV)WT}-+1Sd zwaN4#d{E;cnr@1{XWu6WCD6XfQG52LXo?k(uQf5`7V!+{6MIrL#Xg9Rnqn^n|8l?Y znW%&UudAO!e$B!R-;WOO<{y6dJs)+eYk6(Pj{1neri-jGCtr-AeT$v0MUAwr6v*o` zrV>Xq#(!_eaA zAX;XMwV9DphcAbH(G4G0g=i0pxqT>UTDsxA`dC|=y%?Rztm-A(BC-=_4O=4@u}yK~=5GBcS>GUn6T;1niO z=x_u~5`!2?6P*_G@LAb%%-g8W-$O_~(6y)zHZ@@aWTQ@+uh*^wQ~7=m1R-)Jg?}aI0FLknL?j3?RHoYY)iwaU+H<;ZGiTLw0~EAt(>} zmT2N<6RQK&vx614amSPFWQXXKxx(QBY1s8AoZCp@L3{zw$nfq@=emi_=)i(Q!f#(K%Cf(<_A!L8GAYS9tBO>OTIT51>uXn~7 zvcJ0wUu}bLTj{?;qOlwy%C2IxehN0L@n~KZ|7Dl&qMRGJ!M+d$B43x!f%S;AQ1lZ+5QcUAiIuY4Zcc5 zIu%?1W`$_7ui-;{Sj4*W|C8JJaDoo#)qPYPtWR$N(vvn$$Ta(|LVGXX|_g zN56-;uW`nk6oha$aPx~`X?Z&L(3~7YcD8=nAv$HP!?xnuU=tK0b8wW)ViP{@!2iDu z2`gCdz@-ARrz>jUlprn|QB}oA`mo>n4X(!qv^#)r zM)taa_D-1Z^%thy+8D?_Xc}9eW9kwxKUswVy25JqQM2ut_R`h$WGciEwX{~iUqQo*a^>mX?rSxEZyPrtxG6mzqVgIyF6UQ zt9&)mZ%`NE^Q^i#WWTY0w4o}^J%Ksr)VT{}@311v{#m~A1{G zQ+$?Bv_eT7XSHLwF8o3(#THcvXzgX6-z@{OWt`pNAdbeMEanHQG4wp5v9fne!;@2o z3%apO50>`Sg;WGJ{C|KBmo{>ct>6@2%Njvf1Pmo0%~?s=98TYfhHQLd*?YbX%tFAb z{}1%%l>)MjIK?}>J==$XlOhn=M|d>WA9cf0J3d3 zMcwLbAtny%LL>ZV7%JwGY*0s(Y{%KJn;?g5JOHj@0O%KDJ!lyh8w_Uyx$@s;`vMu# z%Rm$zSb)Vk1GxaG9K#K@jDhF1aggjpGAbay$hwHxOUxnc2uCJE;a($Ldz0V8RT4im z@=)!`_Hd*2DO7tpd$=R7;#7O`J=`hIShZJb52sw0q1waM6-PEHuiC@a6?G0psrGPn zMMIoN#g~VxD<|&N-;GtTymn?=C?zK|#&UL<%-V*%r}epbBdF}V zH5B^{lkh%9rrfnT6m;OLyic=UIH59UkN$1MtsBm`pI3D<1$k z`|as^BHyS%&Vf>{E98)mCbD=-tYX2MNXT*2X&HdeOy6VrSCXeBc`VCSr$5fn;*7WB z?H~TrEY5~Z&{E}lk$gYQ3s>ipDfjC9>V!HS^XjYfUkmdm5&O=V*H4{KC(Wt&my-O+ zm^V+IPrOs}w;}nPFmJCqpUw+Z+1DcZ7cuX;I-lNMtMU(3!1M1F=6zM?lNoJl{?tb} z&T6~^{l|aiKbgYD=ey~QtdIReKgipv-*gB?TUVU{wUE0NVq=6pf#OqAQ zT9X8n>yG(-2_~I{-*&urEXl6CGXh9FxoMa+N0Z&KX)4LibISk{PHqck?b2jVzhX(U z_3xBoFF#(wtSg%ACpAM!cDGIC*_@46nDs%E?cG&IvMp2HfXwvcjCSG_Km6yV-CXZZ zvd<-Z(D{~}4`u~wvfo9Ulk78(jpz(gE)}z?YqGfs1|-|u-WN!qxb~RURg>+%!;fV9 z4)UN&;BezGYleZys8)+SAMuVocT_NOI@Kqeh>r!nh-COZ}9JmsV) zhF#=eHh6?tFErVZW&%0Y6q792+!0Q`3+MmfKkqlquBkZpHjVXp+%U^umCYV8**cn| zcXbs*>+_Fkl$f8c!hr3j5uT9KK=I&945Mg+S)DZ5mVE-V&Z)C)`)&_|oDqtE75c2~Ld<%l z%I4*hSCeukm;~gPWh+YSV$EcP|11I@xL#OCms65ecyVqkS#4K`uC^;@tL@?nykxaqUADBY z9$j=-K^NWCXN&F{uyuBE30_>L)`+gN%P+vITtACUaIJK%98Pfk(8~-bxGt)IUk6Z* zei;m!BTmsK{Qw?TkkXnYvsiE(O{#jSTh#is+eC(26oB^nQ36J^lDQM$|${# z*@J^J2DgsN8n|^*n!ez0`zsTdIaEGSID?C_0Jmk8(}uZ$t1{=DAGj$?B^ZOdGO2+A zJe2*EHsGmjKBf|sQ?5Iz3+0u!(Q7Z|L9a^St&CY54?fB=^(uj{vf;wA;HNy4ZVCR% zvRHb6vMiQZK^cPEKxGPUgOvMbTSG;kfIH7KIiw zYUTxP@A2MXLh5<%^&OtuFKAd0xqrZS-7!^8d#-8o*m_Rg{LtNfKATL*xa7aN%WJ14 zP4C1W9`c(T9)GOt^340@(`)1gZtMO|e|*Y0uXXL8+0AcoBVu3wZ>2_8KIO5x^&_j< zb*_f&>ix-ha{5KTO`TskE^cx==Fs3@IwKQLxUFpYp!CdImn-h*`N42v+J8P9I=(2g zu+hz^0|S5Pja8m0m(})(&D?s|!uIt2Vmh^2cEFadZ=9DlyIbkV&_4%tx`oF;?g$;r ztb!2uUIyAJ$XmGQfvN!U6m^4#VUrPLxA9=*g>L>z3!C;6`IGm1|=HaQBps z?=D|emZySSMzVgT;h+`#l-JDctI8^;;0|KeF|KBOJ=OW&Q9laYM+4HtbF>XGKHaL_KF1Q9>>Xv#O}F`SCZh%4gnIgV^O&S>@9V^E(jC1VT6sH(*QNtjybCP*z!i zVMb#9l)o@!+$?Xmc|vELJdP(KmxcM8R2T@qgs<}D9k3$azz42 zlNd=wV~<&BxV{7eYN|oxka2Ws`&f>u2)4HBic?71A)==mMZ4KNAj0~`4+MRfqDe_?b6u?^8p=U`0#fFK&O!H{WpC-l(%xu*P@|wET#8?u6+&awPsm3thZ_V+A zyjCu&@N}n`i^uy>tPqk-jsVy`tX&E62ABqQ;!o2I%&M)<_M|7kaATVwFKCCU zJxLBZ3)u-0ZqwHo@adz3dgFBbB?u+&A*4(lKoJXYZ_Ql2r#QvcvzC6YSeTouSO7o<94 z>Jr&K;}HHv=(3|x`6Hx0ORQc8I}xPcz_ES#zX`w^vS2gZ{2Bdl1*#G3U&Y-s<( zmcB@1hq4#p8yxI?o-6uB4N|LVkooZ%ZJ4N(#?^{FfP8liKpwt`Nvia#(P^B5*#K9QapOvhmh0 zIH$1+Cf879asH$@W}4=il#0c*!Mr}|d?JU|KMy zXoMQ`VFW2$<}+?rm<`|hO=Ms zpJh9duSMni@>Jw=Q28?*E_@895Y88IL24}h-c$Y@htPp|cZI238g6Q+vz=z*_nz`A z=r5~_FNM?ZJ>@TOTKK|_Lb$$&8?46Kk`FrNFLv5+$DU=+!K_7^Y*L5iudw*Ojc-+V zVb%dnHu-!~{u=9yOdYxeA9n?_3N+c|wUzu0mL_hF>}}!?nDtGQO+MX}zcrvs8C>v) z92kq^*cHHk77_aCru;ntMw3%nb|7YjX|hLrw}AXZ7O(aBRqv}~RwGq5cb1&(wts4( zzT-i!XE@?qUJO((%pax3(DRYLK+S+Tc9HjS8ax3mM$lRjkUcWXzh*sYm5k{*hWQu7 z81liY{96HDvpg8&A!fh%3(}``OUVCb@v2t_{7~W%QsCucg8$5GYioQKoZru3QWrNo z?=d+P|MBfN@|j8}qj)Rvd6E5VHyY??_<)(tGSDYcdcxIQ;{(_9 z4J|H^oicWgaLv&0;eVy!w>;~+JzO)g9AJk=v`s+$(0_bLg|pqR4|s3UvH7(7TR%-) z*tB~6>FuUY-5o!#=cDkWpDXoRdtmmx{exF|e|;by7b5p_KI_w~dbQraE!*vUS$>02 z-?H@wEqa?f_T^B8yYalHzgwq(az3VNzG=Fb$k&w%dH!-9hGkf17_F zSU08Yktyefy}fAD{fYB()A32J@c~`?_nzE7X1u(NSLDZgihUP4zKa_=u@r3wYeap7%f5IZ7hRNC{DTXcX`;4xrV^{0L_*Fmmhk7j>S>w*vxnB3Te&MDM zcXG(>H}=L0gV71YU97jY>S@$>+tW7fokngwKEEnwH}-qPmvy7MAN$pP#ky~|?yc*g z@L6=ua$0wVk?u&Y(!;XK-cxr6ewptnB}FZu^M}{g=>vMT5kCxpO8y|EXW***?4X zEzd<EUMpI7*KerQiAh* z3;R6XY71Ihj87WT^v?U|9ijX7Ih(zw*!=9>E_c}Z@}{?n}7g7Y2f2aT~b7%hwUZE>#i z$gM%+8}Hxo*!Fdc+>3e9OQ+wgX?Mt=N@qOb&%1Z@upC&fk*xNPW=rq&H*9$7X3pdG zRikF?(pg`aRO`dOW3?VFyx8VM_iYoGn$-=T>E-gkto>c%mG`Shb+9$k)fsX&aZ2Xs ziWPV7UHI|2Q|Feiy*><_Ik$c-&jvo+-2KXh#(o1cURUg~;Z&2A!L8T&{#bnBzXJ|s z?f$ECu=KH)i#I-ace7>v*_QJ!M{GFY>NNGt$jr#go7(EW-?1*$@LEfsR|7}HHkxO@ z;(A(Y#>m3y5o5|&+DC=S1zRfG%x}4OV&$;35BHu=J}_B1b9~3{mtS1X|DkK~ z_3=%|HEx5fuYM~Xn%A>p*Txszey`Y&^6<=zo@*+eIK4O{E3r#rnS;B|rET<@xpDS| zxFL(~Kj`&FdDG>>z8((EV{2Od>@GWLA34+@xM81JbrL^UlF!hs8gRCAX6^HH-Ir~9 zSi6<%$zFo93i1eEWxOQ#zqF0qtW&RFZmTxjsqyOt2Y-hu zd^&kPZ#C-B%-+#M1{xWjw2w2|ZFr@7@S0EV7M)f-O%FI~S>P~e#_63)l15CnbUYHX zp;zj!3vCaW?L3${eo^`3^M_Ab_bsU=PFlV7uAT4g3KO5djIjNf6<_~;)iuK=C5Go- z8Pw%%?T8(z6MRE6qj#S_am3-b`NIpBUN5_Hwe`-Cd)m%^=xEjHTlCNWc2*uG?^C(= z?@^OK-ZI)_9d@W`=R;jDEtO54hgFZHW>*5QQl_&1+IF9hB?wPNAw{=4=* z`_$pZV4YaEJ*$HTSWQx{A2Iv8!Nt!74?Xf{rR1ls>}>Y9afC^Q@}Dzy%h$F!Qe)`A zF&C~MK2@h?tBfZbUr+G8Sz}1Rt+62ObGo=}?B-c3?sVA(9+j;ZR&Z*RceqTI6U+Nt*%0R6 z$1r#E=0@1iY}tJ$%vH9zOWV`B>FegdjjdF#%kdnuQzl=^?cRGOrr`9gT93bM+WF?Q zrCaP|&wy=fs&207t+yC*cwh!x0JW!?m>esn# z68~FLdijY5z5a^{4%=;J@bb^{57C9_%@@b@}#A zUwqCupZE3(%IaCSZ{=Fc^yVJ$+)}CDWZP!bYX0fwFZb(O-?f~<>ZLs^!H#O9Z&_C{ zJ$qvApG~LIx72I&rp%S9{gy8E3CnDD%`^MV{${T|o0k4#dVcII*PcdU?uW14Z1g=s z@8u=a+NjT)_>j8nr#OK9I4%)_LFMYld~_n+|CQ~s=6Z)R5DroO=;Q=hGL*nY%0{_dKB*<+_(3)H(+!{o-@smr?7DnIt+ z8)XiDN$Zx5JI?)3T#r!rlfD9e%z*4)kF;(xq!N_E6xjfK)y*k>Z?^ZSN;pg3UT-qy zjO8pH%;AQMH7v!k!w^h~Bls=^Pkw>Z?AEDp!$F_)kT_eb`37*q-LXnzABJm#kWR#) z7aEWen^w)>MqFuXY53IB8Ax}hcMZy8IuCstorSAuowRAR1`OrH=V}Kff7@H&DL0ymyvi0P_|0u)N8EDCK8xHy2KwzV{g~|d(^62-Q*P*Xt29_~ zh9St&Tdv5uiVt6xr&WQ1!E)Er^Mhg1G-tSr2J*Ua)6N0#xdsWN4o7ylN~yN^mEwnS z3~n(Z04%0EgUO}8;W7p{rtp`+41Th{Da=E7Oc6N8;5SnU`}ad$R7?XLSa}!i44oK! zD)wo4F*1(qaE$%&P6n?x!XG-uIK#psaE`$lyL~Nu(*j*>aIF+&KhNMk*t1r%lO3M@ zO=Si@z1fAbe=7p#I0|R%4Rq5ed)e<*D0}&A8ku9*zgM%99iIJuNIHcZb&a9$dJNuI z<0{9I9geYYn6C!~{p7BJW1=a12!l5x2)enINCNotM=E7~{>Fl8-c&>c$59b{L@p## z)o&W!n*IquUjp00oNQFfoyX2 z&<%sln1N?Dy0R5N)_50>S_*^F<4QLS*8Za_W5Y*0uJO*$jlpQd>4w4jcgL_JTT{m> z&^pN(<}w&{l5QC6`I0yqnC6{~z%|$vPBWNYb-~dgJN(f}eWpXC9cs86VXejDrJ4xP8ZB{T){JsMnE8f4av z)WA88?C=Nq*+)Dv@gj0c@T2fs3|@;M=!ONbh7}XB;Yc9WYw(1fIX8S36MnGzQ|PjO44CD`2nsjG zrW9LES}xr%)&U2x-9^}&D}l863uZ9t4c+jCm&YCe!dB+F!>|NrXvARjKc-kUz=MJn1u|e0*S0iO&Hy=DG(y(SS)!RgaB-%SUVCiH7mfdzoQKZ`_KOp^su^J>PeGNHZ$p$RH zH7!q(`7>>4%eGB}B&7?GP3mV*r04c<8r7Q`Bq?1%Dg?3t%Oj^biu7F)NJ}quP2_t5 zxwy#&EQ9Ye z)F6rP*NIND0ZRwG=*cmfP{x;a0yIccx&S%-sv||#dF((N_G=m>DP3^f`|ZKX5ILiL zuv>|3J9k~}J3;QX$I{6LEPV#E$anD2rOmsK7%7Onb3G@@xu`)E+QQcnBLz{?Tg#UY zp|p`7Dn<%CGBU+Z8|rjjm`+hnu}x~xHxC<&0$%*#^F zrj1f*>;6lFB%JAllWf4U!M6~K?0-C&w%FxvsG~$U1;~TPu*pK5tB-qAoi#K_!WmCE z$p$Pt;Mgod2FyvKItOZygj0ZgxE*UoL{_x1r_JDM4U!xqfk)MP2U6trSMju6%@!jC zI?uJmekXGJ?=+>H-^55k>HTlnFeLf13)VVr1!~`kM-R#QNj6}awlbU|hlI;$yO6Fy z5|0GPeV*72!_rNw;w)g~?5jZ%&J1$Jkqua`X=p}~>$jRy&gB{;;S?ZuEX11zQeS+%t{y^5W&QswJU5_y3~ zFXyu>?re&Z);1|(q(*RGDW$>j;71^Q3+t8~i&atGdq|3O2|M&(5|~wM5(GitIkPxBFcJI`cwq zun2EeEu&v!Gy7BSdp8lkU5fX%V=>Vpf!*B8t3bp6{6ZqDro!s>1nv^vo3M5YLG(QX zc};2;mQuSXfYp*M_)VaK%W{LCg#6}AlzUa~+c-l`K~FT~E(Z*w4#lh*ZuU0)Cv;t#g{uGzrK&4R^)uAR!i`{ja~F!)4h_Z)$&e0 zkRi7EF`QbFTAW&qOhhZ%lzFvUgEv;FaM5=w7$vs)IgDDVV|X8L{`eSr+u@N#tJLil zuwQKTYbdo+$M7hA`<##Asb_e(qt%>WiSSlz^?L}lQpeCX8Le0hKM%H|Rv%2{;CWXR zjX#5_l{$uBQqYRU(5^F6xLhv@8jG!fS*c@KEe);69(aAds7uRA4o-p@Vk>SCElV9k zy9~5qG2A`{8|0>?e8ahLxJsBtPqjn4KDw)L_RL`iVTsAY zSp2Qx^tk{iXkW_LU)Kb>5DHC|Mve$pY2;Xt1Ji&ov139P-+-0amdVJ+M!3TJGag+n2uNQ@Id_M^9%}qBXb8gJ&rai7CcpOo^|(=}*gBnJ&~W zvX2wB8(hS$2eBjLmAoDOY5DcHKefAwOS_TRaCQ~3TTkrBjSFu_e_C$WRZ_dD_uQ!6 zt0H!V#Ey($@^@X14P`abS{Xa};eZ}kwVZ{MmaNgt}wo2dS4p zF`Bu=hFJo_|+?- zGx({%_ZG`Nz!4Kl=zApNY+W56wh_*8Le8(c<)M4YIh!)hN3HQI(1i1rkh4=wTpzFG zoFf?LSY!N(8R4w(OyGN`5x!uD2_^J>4dXmSKEFaZM+rIWy>)`lCFeZPI47ROW{PlL z7jpL2v4&$M=X}jLyW?bS%IW)D;QNNAwqS(`CG_2*0zH?i7~l)tMCULeXYV=o(7xoH zp^VdF5F=9o}I--j^H^?CS>G{QMZ z$l1v`1X`Dza|PoZH3et363!eEXZ(i!r4#Axi~ipGE2^B5=hUb;W+J{a$XQ}y7lB>SIIep7-t(BzD{;|CGfpcFAvbi zgcAB*mvQ?4(Puh)2{{kpqAX2H&N+Z_#$V>+848x*Wq%L{-mwoh?hk%@M#D+QG|0FE zxvkiW$;RZ)892evgMNZ1HLP>M2|kJq$S8h_-e)R0Ts4UgdFc4-JhSZQ-EUkPJx>!)?TnMO82=arC_(HS84Y0-wZ&WK@C} zJ|~7Os)B)uq479$f+e$%MZqBh6U5LBlliC$#wILR;$!1Nxvy)NROll%RE6i`ekhGPF35k^AN=lkFM`BLT&VVkn*-#nDJ~q|Mkg_@tpQ z6i_@WhN505HjYS<-nA@*e^$rBKtNF*leNX7%Sr6DAcc9u5E!VT&=*j2EQZ29%aBrR zI;eod8VZGgVskMRF{|)F9;&!ukE2>&)Q;#0DBcxA@xT`!2q482#|TK$Q0NLMf-zZJ zEH16VhptHRXqytoYA8wxDEbydvD%eUcoo{ic@2eJK(W6Vij)VDRKcn5U6Kns*h? zxW9P*rYH=@$%#nu6u+W$Q$sm`-c=Osi=o)@4j<;CBL&`aIH;i@&AWD+;b?C`j|J zqBvO$Mbun;Hi#6pb|it#Z*?q4^RA-M!(?raB>chIR!DJAmH`bl6r_1qQKS|_(X9~+ z=bOq7uvkMuns*h&8~W4BelJc*LT#&udVwX!Ya_#BRGU7g z7?YW){P-CC>D6IAcGA)2CoU(KT*M}Z*pSy(cpLiDYfK%S`i3@r?+Ex=A7B@Ns$KsZ2mOyQ8Z#001MQm;m8x}We!{X*z z4Z9#%PWx}>U@PNepf1N4Q;f;{WL&zE%r&Avy(W~#MiOlnquSz$4T~Ec zxgKx5(I%?0F$^nW)0^0^xKSGxx1^ysI|*&NS60BzA~x%Y4SAacUpFH&n&?lj z{WDoPb4J?YT6w%bWE_!dyG?9Z+~`0ei`$}~c>9Ot{5V|(*FX5%#vD`lxKSGxw=tP` z6OT4CHdsUTA~uzY4VnIpGk(duDf-hZG9Pcb(dNu{4;ZDgAw!i^TYqB1;zq|RS=?r< z!uv|J$v5(Xy+v#`5gQgaYQy4oTwoUKX*rPi)BRHTaofGCPa@^!ny+BC>$FY|*3(!{hsIuTP-Q~~T&cEg#D>L6_!v_awv+J89l-K2Bptf>WM@u+a#AY9{VR552EN)%(ulnS=_2T!d4v18N13EnisLDLu}ajMr~N!dg`&dw}Xcv%qn6tp4hPK zl-jWCbf`xdEvKOzzvrg1A>-my+i_w;UK}$fFUH}jEM(3c+2`oKmuY1j)kc*^_lJO) zg7-{q4VYp~-uN;mZ?y5&WS^t^UN!Mk>%_W+K@4ORv5qCyNyM7G-lZy=SkrwkcYhq} zL+h&_WiUx)O}Zyk`5(vYINxQy&!A zEupoWy6msCY>jeOwnn)>u)6--c0GtzS(9E7RUU{bH0FG_h=>L+{nhwX3%z}5R0X;f zv1v_gh(6wi{`C5w%i8ZlZzEw<5u5q69E}YLHGa05%v;Om$fPx}8Wlq(v*3gzyhz|R z6YLCv=)>}rU-?N<<&9L{Bj>M>N>C`~)F$CcRYNdG`@bSkoCJR+L0*TStFS>Rp~Fpg zXf5W{CZWX-J(wkiONl~FCgH0we#Mj=2!CTi6pUt7hNogqZ4z4bl!Fc0moN%_n1oRq z44H&&LJ5vLJfWkQQ=5e5dK@ee!=*%F0h7>T9k~)P3c)6VC^+0sgV$nCZ4v?oRRu@1 zFJTl8GYMj(gz<-rVVM{%B?^U1f}KY^lMrbth(ehjGWaOw z)FvS>$sOF#zJyWGtw=A*+@Lrn;gC>5`anJCE9TTDp~B!4SR;l@iGnwi;5LPy=oDur zh(hKz1$-BCYLl??ab@sA`w~VWgGqSxlz&I(giu1WWG@&j=F}#kQI`bRD27XkLJub4 z?jtL9AgbUi@~lO~+VR#E697yQxTQ_P=-aUnAcjkkFo#LV3FMfBb3zF(n+Cv0F{d^O z_xs>8K`~s4g#AoHwGixRkc+aag&+!MoE7L|g2X5Ue@=rCFg(mH{J7kQjwC_@V45Fstz<(DF8}a)}8NqripuK)e_( zMM4)Q;d3rOpZBIvf^umT%ocNMi^7L%<=~_kE=9r&CSl(lzJhLQC5S=?Ye%rf1c_0Y zqjU$Q7%oM^ZYE)Pw^Hm}xG$9OaiI(riaE7KVbcX%G*JwfBB6junCOFdXQW(#3Eo5`HrY5k2_#1fB>b6yz#kxtLR16h7_6l_SJ(DH2>mXuY7@ zrxJ@o2OB{Y!n=8aJ0?hsLfi7LP)iJ#BB2VC(0N4_CgGJ(g2{Um$P#mEi-LcyC*+Ia zQY3U>5_a4UWD>gB3Zk$pF%i5mL1Gl%H*$jpVz?9uQ|oneSzfQIJo;M?*L`Q0AIG1Yv?83S?kiI3@Q- zA&xvFH02-%O~g<#w#RAanIg$w;|S&H#W>^xc`pNtej_aU#87Y8B9d3!?QY6%55_(1Ba(|>n?Bghi z!k|^2PyrJpM#15$5i}FSrAX+*Buw|5QU8~`fy6jsVxd-%6RxFhD(vqgGneDi>()tFhwY#&wf`(#srB`*kHy% zUol*YggH#Y-L$Gqg1M_83cb1qz<**+ZBa;^X$s%Pa48b@GYRc=__5|$LJ7Ui^dJKh zBu2q<-4giZh|OOobCd-Voq&QIMziM0PO{erAUzZ(2H`RJKsTD zAe7+vCjx4VIkic+xFG{Zis4cucrpodl6Ba*;OH)h!j!Gva81mqO~Qg^CZLP$>nj z%&AR+w+Vjy5baACg#sp_$7FtFZKF_vXIB?!Bj(g5;p^c_Fk1|l5{2JPLcO{CuvS1h zK@_eOM8GpKr#1-{tCj~_v@c;4TzzT1P=2H*I~TSKC4?W4Lnkq(HVGbULt&v9E+q<8 zn1r0xrc6Rec|jDyGMwOzm{XgCOOchq3GGW5g$_)D$z%R8*j}N8;9X(RL(Hj7!omk~ zST2T3iNX{n!EzjbHx=b2h{CnmGWaCs)FvTia}2nneF>wmok?i7g}<9RER?W%lP>fV zb83?iIlCfciQ!VBkk2Hno5J5s#d`~)&_`wiKg68cBn+}i0B^J}VHCbI2{rGNyD7Y* zK1d`CI)l?)aBx6&C>aKbImzTfPVi1}i@HFe^2V01S_~y~kvPqZt|&}Aj^GqZFGgQe zQXgQ^VLpPQyOnl?Ph!6RRrH9cXmCMWf#<~)oxqCTYeYI~SoC&b(RDxTK{qkq|0>$b z499@Q(Bg`2!-}?AnZ{rLzACTr#%%%41AZgsBSq7j_`idW@EeIK;Dq*nc~0R)0e4vMzQ=5c3dn?0oF`Pu8I0>7XgfFM~ZfcaDAPN_2R)tSu zPHhs}MJ9kd+Ltg2c}&8l^T8|%hlLWdPPjooF{d^OU;czcmKZK23ZI#TjmnBlLcG5q z3cKp#^)KesCgFZ?6nLY338P>iMlZs^Lwpr_UVn}ql($xuNImm;ABlQ5uAmq|D;lrZ&}7mN~fYLig& zTpa8W!=*?V!z3);$zKua6$DZET+0;nFhODz3LXYSs2DCq!g?m*$$S1OOO8;&oV);- zAm-E-g??=UVV@W-MZzT}VPHdkN<^(dK@`pmv;t#HkQjw5BY%h%!=*@g&m^>L=FQH9 ze4zy8?-ZCO=F}F21oK!pB8E$mU>!~`%Jj|rMcE)o5QWYg%Rp&NkQjxHnG4w{npG4;;azT7u)_q2Q5dnx z1CqsXDH2972~h?7ElZ(LLdZ}(SS04u7KPuL72rQHT#AG>OoE$?zj13FEQmtWX;I*e z2@<1Vxibzj#BeDR&NB(!()bUCJrzpWYor4!#GKlq;Cj^>a>Z~d5?(V2m&fwAEFD7x zQ5e124LmSGVib-y!*zbea48ZjBIrdKHk2RSdo7gECRrcWiaE7KVaq2|xF&{6kr2ux zIL_uL4|Wd~M4^4181TUaiBYKD#RwXT;Zh_tWD*W7;iuDn6iP50U<#YXoZ6x=`9BA^ zErv^xFoa2X@5oOc>>DPC!l@I95Qqs9qwsrNMQARDOOddGNvIdYKLGkJln}Mt3U-P) zwMF6iT`PDXhD(ufnn_rjXvCVS!Qp}^%yvkHFiembg_m6{pp6(VMZ$9?;cZ*~(Fa5b zB!n15!hSKQwkQn$U;@v?a48Z@BI!kG*P9>c9VwIm%{Yj`1c^~_!?));iQ!Tt1ThK0 zxB03`H&PIV#F091RLrR@3Vn5w;EfnAMM7OB;np+q8DQ)|j~7Zf`8*90F+pM!B1gqR z4>4SdgaJ%~MV1Fjz&mPr6h`4J;492Hj?n5xG#nRmk{1_+A0_**fI-(ASi&nYl)R(G zX}+|De*86#;7-qh+f&Gm92PxPSaiY*6NteCf%pHb==Vp>p`937T+s=v=uxiZ3c#Yj z3yZE2p9p)!eE+NHDqB+Fff!m`(QR1KNjvy4<-yUafaAcx<`?B43={tHn%|u1@aX!#0Vsu^D=?`Vou`y|B}$PLMeDAhD(vKnMs)D z&3D#E3MK3t6b3PvATbK=yKtRDF?|w;iI^ZU3SV0%K@TxpiUfNPdJ*obtYlS?Q6)hXdTcHWr^TGw zqR_)u4xhwuDH39t1lL#me1xe&2`B8LAO#a7M&aEo3+N|?OOeomN%-@tEQ^9goFEGM zCye2Om{VI6$}Wn9A7Z!^31gUq?2AN_`Z4$l=(uHYaxRfZ2U=o&@xv?mCCkdj^u5C0t6mx2mFxJ2dN~3)Vqp*fa zSeT37SR?1cCZUAVo%Nutm{XgCj;D-aju1ooeGr&QK&fD5T1)UwMiJj zDZmcxOBjXMOoG=?{=J4BLJ7|L9?)6LsZBzqcdoEV43`oGi*od$d}YHA^oCXuMB#Sd zSa>Vu)F$CGSb#IymoN&UOhQY!6FV372_-CQuMa)NoZ2L~_jQ34Vz`tjG-ML49Og&d zqm_av_{EsQXECQ%36K221MN!~g&|BrhfIEk&Jm%6I}LID12LyI3CDtCV67M~B?>E; zgxqYh+z4Kj3CV&ed{8F9Pcf%92@a*If)Co4Fbb!cgc_Ils_B$af~}4*3>9-~laM;g z5H^e9QljvjNifCll+!3ArwF2OYrZGQFhSs!HVM=1?IBPMmmJM0(3rAQdSBm_1thX(@hsH>)76v`YeN5>I9DD}Vq69iEp zAAu5n75!dy{3;3#Ib64jf)Fv3d{c(geCiuX#N!A{*bv6gEPg7)8DY`sBT`_rnD2iT z{on?`Rxz}=qK~tphc4pZ=18b2@H%`R2Y?9@yx#h!Jot#A#TEUS72T>P|3=R#bEVcui4Xm%|duI=m;Mh7z~@ma8f*l7YV#(9Kj%pwuq;~%=qg+UFAJ;>Rz=0 z87BPY{r{4XUM>&<#c(MS{F#ID3!Q2SqTp!|4W5`F zF$w|ST%nE_E=9r}CgD;q>_!j?Z-f%^6(+Dw%&9F3H|9ITbunCu1X&EdDBBO_$C`W8 z7DU0PO#=90g2X8N*cJed#BeDRJeh=>>-i}WpM(lCj z9cM@%5f4PaI)W%%&53~^Opq9b3z;&=6vL%R=)xrQG3O7&521t=Z%kpAm{VI6j4q|a zLor;6gc(f2i8%g145=%K!ie}J2*(78QFz%t721m7QY7qV5(ZiE_l{gWfrPg%fp9>~ zsVxds4&(Tb7%oLZ0h6$x1wSQXlu&~Ej8url1c_1D(_8_a#c(MSelrOPUHB;xdi4cS zxN$!Kj)^(7MZtAy8F(v(OOfCjOY4QUefaL{1fhiR0uGWeL1Gkqx^vJ|43{FI3X|Y+ znt#p4xPc%Flgfm`88N4}C|rC~6+VmMQY3U>64qPrQzE7bB^-5AKq@9kj6$Ob6X-97 zOOY^zNqAz-KYc0PP!NT0N0Z>9m{VI6yk#=@DTYgtu$@UrYtA2tIYJ4mcIiNMOpq9b zMMDB$s2DCqLOzr5)`Wj8*RGKu3O0pda9PZ$Eea9Nz92(;fnq5VzB38ayLw_9f#V3X zh=fk*_=S2L9Jn&3EYuWpk`G-Amt||6&!FvIMZjP&lzeNK)2uay!u}dZaHbdIxF6(` zEm*W^V?oiaeG=fJnD2iTeOlKYzKWs66@8o)-LX0E^#ozjZS%`PGA0Nj`M-+JA8rCY z#n9r4e$0xV)W?7|%EnDpUgJ%?NvQ-lBjzJT)0=p~T6i?*2v=@(hR<|6La48b}nS>d=`ObRjrh+JxuI~vK#hlur za5Xa;ev08zB-CUQ&b!E26y^veRJ4tQ>X;xg3hP7DV5k@_MM57Yp)be3^=a2k5Cxm{ z25?!-sVxd`b_IY8?FEXZNLaumbiBs51&f3d+{?Q|T`{LN3B%&;V6+%6MZ#ewq08tb zb|9Ra3!<=fcnsVSb83@t>0~)jpnVCWP{<_wxXE|WRtP0btY-#|#hltCxRomf6UA^T zQP3?*FUr_nd~@WHDTspWV=uTX=F}#k{rb{ig7zhhf;W>8SDT+IwN@yhRr5q>A?DO3 zA?=|tOc%qYL?MGosCzGtoeMrK1W{=0Y66eMoZ2K54yyo`XkWr8^k5RM4+kb;vrt0F z@p8~k%&ASn!*8)LR}7aDg*i-uLoL!g;f-5hOFpQh5byz zfeQRoshvUzKh7pV7cr+c36H~!VX+u4B?@<#1j8%*Yx!ZV1W{;u#~j{?IkicMJW&Q* z(7uFGkh#)}vP<1Wb|Cf(C3Kl<3cbXf+9b@pVgW0~a4AvnWD<-k@-uW|S_`6Z>}wKy z5p!yjFk^ibJ~*0b)*V5+)`Xz&bHpN));<3Afz%oA|^w zf+(aFl!jkoPHhr4Z?y$qv@c;4W-tj|5An|(P75XI_wa*ZVoq%mN(YsLEn>KoDC}ku zhScFdW|Y!a5QTR}O3=Xsfm_-n)L0k-L1MTR2?b2T_(Xoh{en<}@qTj{Bj(g5;Zw&{ z*d>Nbk?@;I&@0Dx+^V$`M4@&iAJE4HiBY(AO9#Tma48a8-DtgVViVs+`F7rZPJUqOfYJIh4T!iBaf% z+!0d6a48bLGYPIe?D0V0&HO&3u(aBSOAF&TLP$qLIOzqjM9dp800#&zx#JVe_jb5w zI|`%&1oD~`%obztsYI`|%DAp6{E~5TI?p_znDR=jhIcJTD{}7XMj5};0$G9g8xmQ}S^SQ^rai=kE6Hk61{F#v3&nJfJ zzVvV0T^aYrQT(y2+*J@0o8Re_+Zz-AAto~!cjQq1SS}ZGKWh|7xySq)_a4UW*pEM! zzl7YMKj896cq|M5jr%6!j_<_ZPBrKzh{?86j5`(+{~;!S825p3{IT36bNiww_wj$@{=vBO%kcMS5j_PlIqYplxt%caA7bL1K+nO2a{ij1 zE#$sb=s~#${2O;7<6btFKL_6lxt(J}D0lw9akpjMoEiVTGrgA}Cclr_Qtlv3{D+uK zWZV~5k_nL5>0Bq|uG@uiPx&|QEsVRybv`DgdJA-yb7JlM^MB*M%(#cN6>H~L{2OU5Rn8 zUlhZRWp5#O<~uJMlZKf14>4&;x%bGnZ{#mb)*;3Yn@wY5O1$IWh{sZ5SgDiQsTazw zrmOJBuC%3PFAFHBq3e}gjdf3~`ts|-k&Y`_@74elj5pe2Uz>Hi=})h0d>xXQ{chz0 z$tn|amqJWph)E|0VnQE4(Vt$M4&d5=Xws`)2n3dHuiC<9l?j;`O-xo26NM8gg}!!6e|lx+;YyTflDW(q?yF45YvIHspP2k~BqsEW zU-YL}>UAe-GP;v4SoY)HC!d2OCi<9Qyu_K9(2tGNpI$4=*;A8kx12$#G9h20CnnLv zWCk^%ON-KG5QRSMQmIjTtgjw=`l4GHmOX=16pFT zf|#&~(doY|V)HKH;(2KDWTqe7Q<;#t;l$)BF<}v-ZyvLVd2V#2CSx-zLFxXy`{cEB zVxqtVK4SEvXDnihhq$aCma<@+Csa|HkdN{alPF@sB1XTr&LZacO_!Q98RG-}RVL(P z{KTXSF<}v-OTDp(8IMC@optqo`zONzl?hpR$1MW9ucV)`51# z>%+JMwQL@S4ITXndt`}stK#J~QCE@0J*MZqdv10BDCmJs@TNdy$E)-}D+%n`tT zmm>ybt*-lW|1xm6CBRVRn}GjxHAS*xd=~F=L1r}H1vvWCe${VWiXKf|gLSFNc0?Ug znUEdhNz>FO^r!v%bX+VHO(LpTQIkiAdaE)aJH{5%)h6_({S$w@0Y{TVrUBH%7PY(K zKNUuHjOS@g=uf(i)w7^7; zM1lL>6+@vm?|j2vAB<@llFalVGa?DeK_Q8b%$Sm7sY#$)Q4*SlBpv&b2ZV%#oNH<& zH_{_0$xAhf?*BtV(~u--K|FsPcO!}VI1ZhM9_cMil-`PZyRWVxtO9cG$Y8 z3{KZYrIB1vt27=6;%*jK{W6)xN8Be@r`mmw<$=`w7YO2r8{7?2U8&94fuL#RKpY6- z-QR^IYWH7L_rIM4_?q+eWO~_}miTk8J zQ@ek-Y$SF6IVN(%4ep}aePYhsr)fy?@`M@h{!Xk%)$TW$FFg8XJ&Yh8v#a7ib)W3; zNB@Oik)wYv#2&gKVxS66cKD-D;q2&}pToss(f$1HcCZu?>s4^FV@#J2mQgtUN!>qK zhqzDN&EgbS(Ld@wX#v#k%YU+y{~0E7#0@Tyw4!Qr<~~hB66*^oy!$(_`H!GV4T@f)r1t&ZF(Wh{B^cSthz6rWNbcQ=DLBu*0 zob0fLo@EqH7kZ}pFV-dQ6L+(?lcUIBKRNoO%~ZQD^YNia|0yPN#0@S??LIMQ?$b0R z8Pd(2cYiy!)oS;n76^~N-$_HLgxRV1Pu(Xw{L$}fSLEm?kN1Nvi0H3^lO6u(Q#d>N zJvXuHKFG)p79(P<3Ql&|8r(7pr)zLi_wUpr?h|*jxLhTfCr8{Ty$rScOON4QRo?w=*t=1?zn;3UI{F76l!I8zPQicbKH1@qet+X4NB`=D z5a^7EekwTG;g3Frv!nl`n+`qtx~3tp2oYH-IN4##qsu6qE{{(2zo}2$C+=o(e;SdI z8{$6cm#N+N8KR)>KgL9kxWO${yHCuS`!o$nww<%(-QSA+LACpK3x!Aj%~^ej!R%!G zr|y#-{^(CHRpjU&?i>T15Ybl!Cp-Mnr*Lu2-#~5y3lXtK1t&Xv%}?Q)njabx_ldh% zT-h(=g-_x>>B*|y9}pf+-G79M9C1TX^AmICzN+T`($krDe+%}Y)$Wg@?yHWz@gH-D z#%v}2Q}@XZfAn|$z`ne=<{zx@0v!?2M+GN4{L!ayan1i=|g==bl z^G3vd;%*k#v<(@9CGL~HzuNtWdiK=)hnUC_Hv~05F=y_pYJPL)4Bq|Cc+a7B|BLA8 zKk|x%D9o;c|I~f5!yo+%AJBbqyZ@_GGIT&hZxx*E@JFA*#WnvQV_le!h*c^$+2Lz` z3fI*9u8oQN#N8||-~)NBhqzDfmelU&72*O?c)CBpM2@&2sQHOGb6-{Sb8l2N|0cXk zQ@h`05jpu)M}LTqGelx`W&Ee^lO6u(zj;>V=m+`xL3>2>Qo+d%fAlF_T=OTl@`ZVb zSgC@O9lqwLa81o$u?caXxSPejpFtjQ68FjdrP}?jt*oj0g_y_@Hv~05F=y_pYW~xC zF8tBoi1)Q>_peg-RlUCbrz{`>vyG!g%!WVA|2 z`|469|Hs;U$5&B&;ls0g1IY~}v?M?Xz1`4z2^SK2=twgZK@mX^L;)30q>2bq6hV5E zCZIGCL5d(H^de1~pcE01B27eipL5Qg*-Z$(zxVfk|CoHv&Ut3eInT__PT9NZL6z|I zyX;b33C-T663#;4S)SJKuK&XrU}rc|;gjY>EVl^lwvx5J3oiSA0SooT?8 zyib+zq}1Vr=Cmm}x%{YBppp~rN*cW{J^V*giNlFgiRR>JcI`>-(JoGMolay=LXau* zLx&S*eyBO|mZbM;D4fpjN~(;&mJRN;CZ-aH6Q>f*Npg3+N?qixr`I35>pX zCYR8asE=LYytu^2o?JDRIGi|@Xim;yS3<(6>#n5XI$1=QL8izWhmEKux8#8?+6 z_pud1J@$#I#NouLL~}B5QDUHyw(d&4JtvjaL8d%m9ZsAltmY)NIUV*9C$rp@T)rQ} zo_u5~aX4`*(VX0VQP1>#b64`uk_4_~KUKmHJBJhJVW&CiG)mvazHu&22DYxnm1IMv zRN`>rRH8YlFp7RZOX2i#S8{AaJXexLmGE=N;lz3FXil1*X&9(vmAjJJFV*BqW|~SI zPMk`FlZbPb@h?VpruY6_PmUk2ppbb&(dOw{WP}~#J$>GWV=>~4jRn_t>AgK)$j1i_ z`NpR@3W!8Tpcua$!o^~LFM=IXShGJWrCRDDy^|< zT|}QAyz3O<(*rKjWnva=2Cqt=@5bHgijbbAx*VBeURCUK-Al16jhB{GXSJRhid^}L z>K)ClaDQMiHgqobuMl+P%>pXRgh0UhR3bex0!8>#f{S$7f{7uyl5TcWMMzI2U5-p~ zKT)xBuV)cfp5#*9wVrw}xzds9jpm~Zo`$j*>tB|WT^ZP+l$xQ7=+ll*oFaVM!9}_Z zyp1Mm&`$#)(%?_7JfV6wuq*tGVKKJ$A$Sj04!6a(FLV)o>QfFG zfg*hB!$rC*!v+Gma&25yMMzI)U5-rg15Cx<#L`O2={W zD&NEaD>O|YS0a%SxE**lB6rZ{K_%IhroD=(mb!>OrAl;)@F|ttK?AU95O?DW2-6*xtu+k_T6EwauxPYCX-J$dx%%uiQbrg(r7VgW}R{ZG6CPh?-3C{7=PxSgjJfvM^%~RY~hQ{>LRp3PDQRbMfkK%?x47I`Pr3< zZ{c4=k}EkU1y~_dJjqwFjk`v&E1ia?hu@5V86QR4r+XY6jZZkKUF7Bq_U^mIMJ(vKPu+4 zVi+d?Y?ZaF%8imfBnFSTL!xnW|sI{A{h^bf|t;j89yogWXz-hsuwqm3rw#oXhCc}ZuMF`pc_MwtU6`zN_eL>w@>2SAgjLlGRm88=08&A% z6z6!!ugHIv(AR!Z#P_+Y2-WNqu~39_BKKFP*H3_M<3^JU4%!Bqb0QPgOLFCU`ru`NSKrm6zIv6oGSHyPh+HLG+25jtVYXoF*pvvBWM#n3f{Xdj zXH*AK@D!3g6z>gNFHq?^TK~c61nkp6`N#l1q_>L=;A9I-GP9z}+(=TxSO%zog+;NL z%9e++h6dQ1iO@OL(hSAfhvMmqsflE|Ef8Ha%3?bX>cF~&f*K6OXoF&7W3ATBOLJWZ z@zFBRSgU9v-auKUX9gidIu~1GifCDLrJD;wQvc86LodFz8L0qex zcq8KglqWSHHF1Z%_0%aceHRD?@5B+d;}CbHZ9_rj0-}gPv9V_oM+2#gIAS6#;?%Q8 zZf_!-^3T5zi|yFl!NId<;+XZ(JxuBI2IuZO3pd&lxUz=aId9lY|MCl%9~KGDc{U`o z{6OLW07nUQnYc-xfu^#>g7cj#oJsMkGdG~SYxGO)+3PW@YZP4caF%oounWbG&#O+B z4rX_3ZwTCYv&g4S)X8SSp>(0_Hw*rT=~wgM4D)drWDlf~Z&359ihihJHA7xIBysWq zIl+Jp(=rI)4Jh9=fX4)I#<2fjXC$1?ws&`kIvE`t+zdmme@~bEplW!qwPIF;DosLF z)XM^S;G+`CDv^IL-baJ-U}L6SGYrI(TI3glcI**1u`>f?1)>qlt! z8$<#l8VBl_(-EVFbrIT|8sTm+So!E!ysAf4TN&Va6q3u5v=~r%?1lkUSs?rd#nxP0 z#KVez)E8yI#p7VB%UlW=g0u<*37AyO1q2}lg3ts(h=V}RaB=e8M^%HJ8P_q1gtBf4 z96DH=0ohI7G^P+*aO@iK!a?kIS?8g>W)Mk?m>#I3;LZv_c;IC?64&8o2UZToW>FoP zome0$Xp~qIc7W>Kqiz7z42X6H#nxQxG=d-gN^woVPIyRgp?MS@1PYG@2^cv-0R$lh zg3ts(h=U+hLJ)Exkb7JYMdKr5fu4T1ZV}amqWBYvzsbEGboB<2!9D%h;UEEIf=|&% z0(hGMF8{lusDJ2$IQaioh}HEAG6&%VvRR<2Ss({k5SM@vCn1R25F`)?5*h>v7J`Hi zK?s4sJ%G%k2ZU_Lq#k+UBd~h0w!quHcvf3zyJ&=) zQ&>4%K3FYeP6q=qTBF30a0=9&F7*Pa1wgDYD7NO}^fkocf82|n1S>*)(a0B2=kUgx0^D<9$d)Mr6t4rg=P`IsU~T(PYx<9Pm_M&t?XO;cq!MmIfB6QJ}6HZKue&Uwe8st@$-v6q5Rtb+A=_|ZF{%GC(ARzW;7C;kY4M| zj9W%GRh`ImMIaJ1%3?bXDl9a+pxOh`)u7neGw07lk!$|^B3-7zku27vslCKwvb8uH zefI%$U0Up-t7t-~ItAO8$o?x&)77X8&gmXO8H`c_cr688YhJ@w401#1Gk`Z!(A8uT zW)?M|Y@h+D>58mUPrXB?y8zK!qb#=L&~A>n_fS ziL^MA@htVY;=PKN@h7&-r83j%tjbc?>g-rn_?rkOTO!$$3~^?4mgA-RvO4PpKwkn) zCtA$vENiX&vN{_J-9)2zR%dxu*H5)NBX?}?LSW76j4o2G&O&5$HaSa4wUFGp4#)$l zpPgJrPVn^S(K~eo5RSNVA_-s>0i3l}UoDTMm`9jBxRF<7(sm(jx&!f|A?K=ynD z!c$hS#YT<8@M*mZLnF!9cjVh9GG-8~qcEIVGof8<5Zf4`*J7up^ZXau-5TM3%EE<7 z70RgHRJC;xh?^QEmgLz8)V#G>0;mk|Jqk(uVQXeBHp(im>=ICx(*T(lwne8>-s-5^ zpXgWM01K`ifbfEgLT5q9fQp>p9P-&m#R4j zU?r3r4B%G+@LWJVsR@lg{xch%VAXQ?D>4URAhKDYs#zchSP++h5ho#Vi%~^rF})VM zw1;LrB~f1$B>Q)A?oYC95U;PQ4n#9(+Zn_iM(DLzROetI-h}o&jc~UZEH}CrqaIS# z*5^RX(}{N1N2&KQ)k+%XaX;iknA!PQYG@(AYS+hpFgpB zLp#tQ{EX0RF@J-aKzsu242^L209fAKF|(>gRa`Y2B-cmyfKuE3yBBWG7PKO_?!`P@mv9#`+Q3{RYvB5qd55PBEN( zSQnwasS$2Y;liembE|gDX$CM5h2*j%oB~y+e1ib0ED(N!Vrwo=?;{TX)O?k|Db`}A zfb&t7Yb_S~c|}D(;>7b0+L2UKSGwfif||8h>~9ImngQh^1K?kNnzh*E=e<0q_yPJI zgy}O0%;~X!! zlx%f=r6S5;R=VaJ*R1r%KQpNX4_z3WFk4i4}xhog9I=i%H;;olmMw_rB4yF9zpxRm|qRVV2*eATN-8{^g=BxkUbxP zFy81{Y1A^TB&{sK#vs|P$hUT6%pjJ`#XV$|hqi`6yvPVWE4?`yOCYN?v@dFed%S@Q zHKtTjU8!p8T_8qklvtAS2B@O>O9fDKfmmixY|YF{=WQ&g?5$8HYk)i<&*6qw-k%Vo zzq-!?2bg+xSWBTppwL+mGLS<~a1QzGqhf()b<#*2+L5nCP`E6T-G{FFQ^?#zN#m*s zpgxo>4B&kN@Te%B97nzWf^#av6U^UU{f^8*$czFkP}MAu11yM3z=)F&xW%XEEi_mk}rCVLW3Z|LXhww2q6%-2atL606i=HxqDT#-fyt^CwaV=LOM(y8$_Nm zbb^)+er84z#0f^|S!v4!)JKX!TTUa~Jph(#4=bR~Qq@)yAd)mnEJ+Ul>cOJ$0IDw# z?;8|bb1~Qmp8D-fRlp$TU8k|-%o9ZE0mzYnAqWK!gcJxu69gd+f=~%T$b}$uLy!i5 zAWZ^68VLf0z(y%zA2|e0^_;U^*OG{|GW@EAWM81b|6s;6Vu~Mca9C}j?P3r&8KGyT z7mgF_??O9LBix*l3j?CmJ?3;S5X&@5ED5JT4Sl^>0JRf{!v@9HT%3+Y9R8s59&n0T z>6;r76=k_*rClDDR`k&iJpZ7LL9($f!?cKA88)-hsIK^-eK{zr8vuVP#LP;=FU9ek zq7C#N2op}_c~<(=Blp$MjW)YxrHO@OR5esm56NCmOr_k8U$77VjuVaxjv;(-3E7rKQ^!l3D3g z0A>?tAJJlFrEkYY%dB)AbeoOdnU%I&lrz<=l-#kshk-S-Qo2YrE6pLZ(vKfBQ76c) zoN&7kk~n#moZtsRu3|L=5C>&l1GqqdRI}2jh-nV9{-wWE1TmQ7y}E^l83-QK!UEaz z5eVaro|Rtv7|R!H5j3mF*dOHEO)_Q>;df{>*#+$pgSf{CJu4lRSR06|(B9Js_jm&r zzPXx9J))|uEbuA@$z@5#8=$s*-7tWv3`8x1VrynrYT=bgI|<5;8X!-|x6!GTw>v!W z7k{xPIKb5NHXuCpq|jLqGN2+SIEQ@pQL(_YI{39XW#grNYXcBF$wmlWh0}Zd+(dH{ z@c(Wo>kO2?8$bjBcvKWmdO_nK?8^gBFn@czmCQj{h-?<9Y8J==7Q`iB#7PL;VpI`Y zOwURyMADl^ZBbtrBs-cs$w#&g;=+5?fOr?$kp@wO5qef?zeekbnb0oQ2zQIY@^ASo zsS;GR^&=4bHA*Z=ive|iU|0Zk1&BKa#nxP0+$S!+d$T^cz-%pM8wCtOS_OgxOe!V? zf{+41Xo4WbK_F+iIQj0Qs= zV1%BPMz_xn#P`td&aG}BGP*sQ7IS<4&jS@@34p3A7trt(K<;rp6pfFJ1$ue`yzJbN zqF4x%E6Kg)bk&B);GQ0_r@jDoLV4H#IuO9+e|~8EMVjLOT+(DT&kiyNVFI#QpsHCQ z2Url7fDtDlh}#e(5C{?)1PK;`gbzUofxtb0%%cbBS!vW>ddFhG_6Ris$?i-cy-FS% z#NmrHgPIQQLW6jn5qefycMRo*|&)F`neJpib7p~V9z zFIb5{5`Wm5i^226V6{m!R$$)MXD87FL3#iL2^fM<06|EBAT&V`;vfi>5QJO^LN^3y z00`0~5TubHPzY?4BKDC(;8f2!N35YSXCM4OK`{)Zz=tqn8ZqYvcDbw@&^|PXk&Mu@ z()}K4mn?`a28ru%a|#!}{i31zm^rNsL@kXHOTsBo1J2hApk4stRfA${E=~)=5C1m_ z`M@b=rNPOFin3g@(ypJBQ1p>FJpWM3DynG;UGlf)%&hdYzwly3GL%0V0DoD|%u1IR z%ENPtYtY{&%zTQQXQfX)qThi&vR6}ZMYGa@aO!_OD;+%zZ>xMue}2Uuyt4Aa*O=+i ztaLG~nY~hirJR*=PUN9joU;-7m>W&?IcR!Yg>xeJobd}fXqT&QHf&xzp%+y+C)w&6 zL7T*w?3UW?n(VGiuBm(|UmnT6M-7m2vODG5jLK>NWh(=CL=2nB?!Z;piRc04TLxg~ zBY>Xlj>wIL#TY0jYk>4TXBvAgd$bBc8GNe;=bsv7u^k81U?AT9A=Ggoel;jI_RPsH z5$T%j{`b2l(q>+wGuayRChj_-tMeWgU7085P|t(E&PaARd6SZ^t+&x%`$IY003w(T z{X3NlEAg7dR4C^gKr{n1U2!Y$8>jV9Zr6a+bj_|^UF9Rw=YY7XQ5M^A==%I(6G5ei z|5=d4T{iYix~P}B=prW4dXGZhJ#qDOZ+WqiqIZ@V~uHWQNQo%K|WiFs44 zr^p@KETsydOsb_+;70+^-wRR=$)hfS^rouWzvg70XD@pi*AT!5P);y_HUvnulzNJj zIS9p{zj+uq!NO$qetI@SXo^}`AbUOn;UTG)Qb!8b#G~XkH2;#Z4&+;BGG-8a%hv%S z0=4Bu62z;F&`YWLS?CC$60|in!u^nh3zG*|R!+;oT zP;AXCrQZIefwI4V@=FbnXON9pBc`(twMQ?hLkI}e5eu>;ye z1~G*ZdMQ=q9!(X0hxRXxaJLvNZ^@ER&7!KUaCn{*$z@4e45+UTmI>u|FJ7gp_QtiEM-N&rz= zqr{T1161GRo&c%^5HA=MTXV6~6n^-(%`6Lcu<|K-n8JfV;jthABS$EJAf!MLnji>q z5QItyLM{YykL#gmd}J)p(_>!ts0|dwEhrw5dt2!$naJRtUZ-;&0c3?wxse30p8ziZ zd!nd6+g=O)WBJhiCo%`&6tY>Ms#zchSP++h5ho#t+YlrW2of3u2^NBc4?zflz&(J> zqX+1vRQz)40o&m10rL0=g>;fUHi%-!@Jm_i612Ar;v6INQmRjau{9TiCE$m@RbBe%92N^3j}T1|qz6Ed zfFTG45QG#6LK6fb4uVh#LCA$5bVHB^fFMl*K^h4Hg}_EBVjnpKPW3{mNfd1^kTh|~SFhVb->X*ZBAgvD2cGCzqr*NUf$h_(?b2S+1qjcO7!7p#0Q^w^7T#R8uHj zX2u_9gTLiL92jLo@jOTZphC(lr5-GA%+ra=&{ro+6sdVB_0%JJD~u#tv;BdebPuK? zPyeLb=GWdM_cS6HmR!FB_P<_oEu4bY)?xg?YHRu>+%mL&YIw}GA|ixJ>Vr}H|*t|j0 z3XOA;t?9)v0@H+e-f`E2c=!hVLKWo~A=&+lqNP$UB1-Q@OKpTQ*#O=L%2{*OZi-dq zStzdA>--`U758l1Q;W!SD)l*M)&y4tO* zBdCEud|*&)?3r}YP2r-8nD8VezSWpN1n^f>PIU|wT|(mCN*96N%2b@j|AYwi)~D}3 zxp`f8h<6bv@_UaK#$bZwMw^qa<;K_*QTYA=h_8WUe`RZ$}$^)UB%S=ZqBAZ$c73*@JdKrpJ8DbovK{II6N+K^=#@gRc(3TRXCE%l0m)tcYFX93q%EjVryoZ z5;3%nvKvC#Tm#f_yCD8|PnJIT9Kt|kjz&tsCS?ZLC zib*<6q(LCkSP&YKL+*19MeCztfoWgoPaCW26y@(g{DoxyMOTlBEbbI9JSr{#FKUlK z5js*%1R45wp&aqYw_W#}LRV z&LP)XkYJ#a@E~w6p^DH;^wQ|3_6caa&2V8ik{t?%tw{3PAnwee$>DiuuNg!xM(Cwc zok4UK0~!}mp+_<)>lieb zkWxfWnSsc;)-)sk2OjzRKBlJ}7B4-|lcNxNpezgI6bs}!3la=45*`Ez7=lm$K}dlh zG(iyJAPAKZxYv<+^g6weYt#&rF>CBOeE$T=eu>EFNwFG4%v$WBTg#zcZxDSLp%-$) z7dHmt5VWT?!rkj&x!_i>>Q7Z$_X&ZRU6!QR0X6DD(Eutp5I%!qYs2k+;Qotu@ft@9 zXgVXM;GQxA+;b3Va(@@x`lE{leg*#US0Wgz9qDxtgenL^9t5Eif)ETrD2E`80704s zf;1EaX*vkfm=L6SAyDval(_PdLud`X!0Vb5Kf$$f!LK4n_WMNHN6fWGRBJ=O)~OC{ zqCt#jgkAy$U82*pPSEzy2zP71g?w>k)g*3>4}h4UQDRA21E{!$-T-PT5Z@XUTN`fI z*q2^){s*1s95gqPQnUtT23muINV7HG1Qq^C6Dr|Oz~XTmCSX3waxETvURP>5ds-UW zI3#;6T`r@D%{kn-Qg{!gIh5@UfC@Q_$4s&EJIH>}4Cd0^C;D^zsYieM z(4SfKXE*&BD9g^W)ofW5j)hm|kdLZ!j_qg($vKhP+F+hdOUUotXtL{|X$i?WkfBnz(V$>44^($X%IfDF;vHy3s_UgH8`L=R_`9OJ|cnA8@0IZ3i6;H0MMfJ&hOe z2>q`cP0iw5x&j?-ol8X~+KL_mvnsXu(j?sTI1bVC+%?+LO@q%=`g-{?W zqk0YMcaeDcKo?;Zc?+S(biIYZ*Je{7G^my0T9U2mAK(ok`dGjk6iZ{*Btzh10k^7Q z8-O+svi!pk;Gs;+_}DRR$xiTh+LT_?={ ziQ?f`Tog}#lBoRPygrg0MG>T=_`;YFWI_3=0pw=}^&zhpzfQK^hw>u>D9Qj$aiQNq z@oXrUXh3R;Ux}}vN|NdAKpfO4i|sfRzrVAVpsoUO&!E`YGbyIK#ziqP;Yr#(A{SFU zHKD0$1~NJ!1%_&isiw>T)lA8$2H1Fb7S&#++BHUcbHKS|>+_yiln~z!{&4Y~uX?On zhU(WM+2x<+yXY1CcxNA!Ck&uQ3ciPz!5vl}|-8>D_{g`pq zg0hJLv`;~LreYZ3UxxB^19&L~>7NGG!m<#`@fwht^!Bd@scvL?2@q>E%3?bX>1t>m zK^+9*ltHnvXOd2j9T(}ugeR#%W*?LO(GtI^05a+z1xWWW>695Dohga`FVcgV^fadg z-ha`0$=0W@6%*!fU3W46dT<6c9flSn*}a}-etKJMS8jlErvda&!Td>V+*oIz{M`WF zPr>}eSMiJf$547tE7Y6O`s1Vm|#ve=Hp{E)p(1l16T)&|AKp2<8-4_wR> z6P~0uZ&YUH%YB(oeT|AXAqAKZV&*9`z&um3s>^?L_w>yCvu9-9cMyBBRbv3gY3lfI z-Eh(W__HV#{bz(KjAV~~n*P;?v9DDP%6bMcDFywd{207DK-tXzrl+94^lglCgP;mj+6BZBgJNUPq@U(;F8YZHPtv}!IQ#$+E2835 zNsv(;DL{WZrk^qc^fM(x{)_%}O#lDOtk`=GEt+gyq`h}C{rIM<4LXd?uV%sImq_-a zr`zD>66}0zhH|$7tV+=a<&J{9b5LG2fbUYY!O;D|SfasZMkF?ox(!~Rfq#xlri%hm zPNOWg}^js#Qa(4>xodd zG=Kvs8e;K0oTzq(vY!DQP0zg|jEU4ZCqP;Bg(tw9T3 zS8EUxo}_-*g(f0mPv%z1sOSVzpfy6cH7GOC8cfNX|J54u5wZVo>v!)bAS~IMzYn(? zd9wDdt4XSTo?a!QjankvkDqRmcQ;i*|A(@l0eFkKms?%l#nN^(l#>h~69e>G!1DyB zQcI!yRs&KuN&DR4DiUR^LqMF?D2welO)~yoNkQEw1Y!_(+1N9igmx@kO+rj~l6tSh zhn7La*{ju6dsOrqQlLq^+$5A4XcDF*{=b?;&UybY_Eo%v5J<8${;f!9msa;(?Glgw zGI0)WTt%{T5jiQ>Xru07k!{_>_>UxjA}QL%?u=>+Kv~iNN~CC)%ja>k)rPXE2BdD6 zu5GfYa%8$Y5dAdDVmnT|WcoClpvD6+&7j!WGuwssw_NQ)On8!Bc@ZBEM!U2cQd#|l ziaa1B&@Q3eE|eK)7pCOncmLxFB8S8Ox0_zx4ItjXYe^~WEaD`2-vif4@}o)EVjTqM zMj_c%Oyii7cedoUu6eO|8s- z-UP}SBz zAU@D2v1Cw_Yv8{}GHMPGOAU&xnPcY$&(~J=7ASXW05+4~j~%@t63tI@rV6@&RTK0+Ggt%t9k_$bHVCXnj;HaJ==^FrVsA z{&zy*9!U24boCMbNDoR3PZYoqD90GUcmfQ!&m>k<=jUO8um+m#NGW2V%s>oWYnm}+ z2FPEabsfZj1KP+({S%}MrKVTW;OH! zu>#uf4dO>e=ws)}CGgWA>oBxuG{Vg*T$os@p4!E{J^;dlcwCl*SD=RCb4P;814OJr zv9;m$?IF>s<34NKxI?Bm0aNQ3A*T zWkCbDM1bM;jr>Jbp_BM6brWb_KuQrgWd5V+TodGtDc>|A$T3G|pCc##RozQ)v0 ztOhZrBONLifwrtc{L2V^?ELjPEK{sRXj^K8yVt>T?`xLw6bH}N>p%?9DC<7eBfSo& zl`+KwsLz0yZBT4&xcvvXpSu!|s9n(fjFf_V$_#MNL8QrjPjKtc(1Bidz`^R9RwM|9 zAiWNPPz6EAgCKN55P~5HU#CBR(7yZDb??zLl z4w??2IVW;?`@n%{4>y{uJ7_w9=A6hq_+Aj#J;jY?H_cHf9YAwVvX!}hh^+0l21Ppm z>+8Vz>S_bZ??kfStchNca&7ng8BCwfK>51?yaSYT#CfYI?SDh*MYU4HAO`3)_itq| z*yM$>mu?RPAv#hS7G}e+5hNix<;Nsn`A_Z>_`F_pMtJTlkqvhl2BGL zfJrIny7D$Abxomcs{yI$s(m1rnog$s0WnykEVkp&bzymBK}`c0lRU2Xcn1yy?;YNcP;PIcwJfkH!Qj8yUc&6r8ml zieJ{e1Z7VH_$mcw6^ml0Z5WheH6S%-$!o&ZYBIeTh}9Zpu^orA#rWAaqxJ*wvq7=3 zXL3g4fQvI?A}!8jYbNz@;{69LlWfJ$qgxrBtyhSPxJ@O4RT7wd8OdJvG;t3v;iZKE zP>wKw%_)eRG7rZq)1X{n0NYa#w`n9^()=FE9U73DxC>bdsy$@-JP_A3%3?bXal=bx z5>zn!&x$1Ovax3pM`N~&IAS6#;$-h6&B54P0S@`^{ZtV%5p0Z%^r8>x|MgY-rMme4 zZ*XNUxpT;{XEsK1w~LpJk?#T6LZIWsjoBEPezJmWjGTh*ywN)wBR6suNcD<6x#Mh% z+(DVNH%9*56s*pWN0s1lO(gN}H?q&u=bh!N3*dPuI~%|i0yuB6@B=h$V*0?gKfG}S zh{0yT=un!XLCA-ySRkW50%6*uH!hm>rfs}k&>SI)H^{-eWYHk%Y@uoHRcP-S#6OJC z8yA~yzaDjWVtJMHftD}X0ZhQfjXauUGh$#`h}2`?3cCs=Oh%Sh%RBqN&z zs+t9IfCX`h!0asqZZWC|Ev7dS=J{xn`Y!B_B-?q(xx!@IAbySG?;kP+R*sFSlAf; zV=0=DiOfM*g=`k6Y8J==7Q`iB#7PL^HUtR-f`kS^f`uUALl8nBa1S8!=mC0J{>x$d z=H~S9NVO2j?nWW?B99H?*fBgpt&Pwo8$@45=wJPsO1RNG?OmBAivF<+1b0{ zBME;%dDj4_kh4VI?Jvx8ig1{SK(fCi^?y94r=p_B6VK6`kyypfK#~9TD%QNHhP7*a z0r*d=(cLg)UY<_*GB2%Ni>cr~HLwR&^*74d6a;W?rFv>319rc7(FK0sPAVy}sN%4(r;%P>$9B z=}XS+bbI-H%2OMr7XYzBqb#=LpiWoJE2up{95X03_RKjf5$T%4{`c>1nJDEYI+LyC zf6|N9nDV!YaM9H}I8HSJe{GO#FZrC3uEi~|I_wGMI|h&mD2J|lZ{Ushu~1GifJg>t zx<1c_Ul)80DuF1m<` zXHWTEJ=a?lM*Mm0`nZEJ(SIeO{uz`>d!iryUR70+Jjx4?iy?`B<;gxz^cQ@X zLjVaBk^xjDK&pwpn~wId=+B6c8u0YKdlZe=m|J^M0SjcxM<5K`dU{{39qlhKhGsR{ zi6^(}k{yGnISO6U+70bdgJ{GEJ-t6Stqu^^puMjV?tvRFe6lrzdXB2Lvcjv}NG?k< za0Ar|8~1{$0z_?tVryo4?@fqTc3UVrX@E>M%A?yTZx49jPdAsoe}OstUx4r&o#sPSI?8L8-YkB8!yn+%M>!V&Tn~x1#lM1D+bV=0Q`6nPhN+{ z?~93tCzuF7$xh}VEJ8L5R5c6a01M(0FybTxa@$80p~dv{e&aE`?_#w>eXk(dy~(+^ z$hJZJd7~r{??L;aLA=WdJ-x4a7%!Dsv!MM_Bit9;^U%FdIL^>_DLKcv>z2 zBS$EJAf!MLnji>q5QItyLM{YykL#gmd}J)p(;xm^PtB$%7Qy5ya&G}$EhRFzr#Bs7 z2_PBDpA29n0bKqUfX2V>n-KVqDQ(_dWDde7$Yz16W`P`FL0kexoP;25Ly$lqNN5lw zSO^k61R(?h_W&}F9-yc9_6a(t9T**{K0vb9Qb-%gV}p3{6t(XRXcrm8kBrdM`=#H~ z8Tuw@cWH#X2f%XG5=GT6s@l2$#0`xSOVR^?@=eMWK!t*pNF?!xt+^Qdl^BeA2I^szsH>Ig)4gJNqgP7A>g|L0jUgHufJJ-HDTWx1yJtMMpBL+Z)oay~Lc{B1%;yUzy5$11F^Yq^RbnXvT zWaG7XeS;?TZ^NIIlX`1~b$+g;PPPw<9={JO)D_l-iQ($xOXku)IK4Xgs<~X}FQ88L z4qUP|{L*~$%nUwBP3H~YFlsvZj7U~^H{0bXg zgyoj6@eIX^jK9Xi*XI0)Rs&p%m%@l_k`70eWF|h3%M3Z9k}}!F*~~w&!74Y@C6U++ zU4#{6Gm9!=pPS9_wJ}4>2(HC@hRpnUq9#6O3p4lJW;oQOk=tspnanL~s}WFtrp?fW z&|&Yonm=Hs<&IoRwO$q)rPkRs)B-?OATdC?WKzNX_%Rl~Ra8E2d36N33rOrXU4%8E z@~-;$+PF<;aqN}14#3IQTD03xF}uPBw@?2oa+)R8(Xz z-I7XbDRkdzy|6IB!~fzM0IPkdx=FSV0dZQREVkpIK6$5-pzad_L5RO>D=_&-T?Z{i zv;Crp(QKPI(vyE^|Dj7W$8|`O!zJn0ULrEt`s=(cd=1a<;_K-4n(7_c9!B=>rNLLG zPKaYFwDS$(QA)nw{6*)G6pLY5h}ty_BFrP{gv}T1F)CgzY$d zwW^XwP(^_#XHabGnS9Z$;Npv@NQ*BySaJ;p-j;AS*(zP8hLESUjOjoq95(Ee-?If)XdliemYH`8 zx|W%pN?FRf1TSupC&dkOW|^6FVp&;cW_V~Q^Qx1jgV`P1TL)P4nOM3=wbsmMf5NwK$KIBzNq&6{%;#i)y{u1O2%^6?S`c4D zyUrk*5TdE~or>zWj2JlgKygefrRkj8G|k74cPlvY*Bn+3ynTfC2@BH%1HxOVg$44) zM<9Hgz{3ktXvaRH-S?`6B2^tEyCu2akz8>gjds)Or32L645SM|!g)oCN>(<)+XaK5 z9Hjx`@k5(OG$5MY&7gWv-PU|SmTQ=pa$q@27YM-i0CLR0#N51LcpV&-9X}F_*Sny+ ztpRueh63vHIJ2_SA@Iyd5)0p0JWUBfr&QjDAi+NlzvT4Mf;FfpaRY(4VL{kJ4*AGA z6qApNDN7n~Z}&p#4T_~VtPCVeZ`0LKqC^mfv*!}TC(zC?hz|+Tr9=(S$LxOnuY{8Bu+NZQ3SAzvPyQkFQz-0kjo8x>Z_AH`a*6@WB0ve= z5I_~<;_*8LYdN&*H6q0w1IXo!nbdr$+Bycvc?}awfjb7U+UszuGVEVKLJ*}m#P){U zjnj%J<`mXSztUq%^R&ZtGAC1KlArR!QfFOiG5Q-oOX%K`)2tq6b@|BBI zY(AXJ(I`#1wicGi!V-C z*Py+x5h=(6WX=cp&Ns7?4YA}wa+wN{2Uxp;xdN~_KQmyI?x zc@^P>f6*xFXjuMRB`A6b6g>;V2XcfV2*MczVG)Ax3PG5LAlySB@3|g|)T1CKn`|n$@+vc1FJWV zkjJm^{$ug08AMpRU{hs(1;|EMxxiwMeI$4wLNod`4clH!4afS;I)G9qkp#hw*0kmF zAoXr=6Mm><2Fp zKR~<3Anr1P_lSTvc#PgB{uSEm8j<3z2c+FJrT(G1EiVF&KysN%&r=bgzNs7|s8Rq_ zH6XU;dVHjZ6@R(4elR#U#F|`+Xo4VjH3SJ5f=~cKNP!?UK@j2~2$c|oTnIup1Ze;W z(j*Y1kswe6Y?LDQkwbcq2+Bua(I>fX1M)A$V4>4m+;e3Y35ZwU%m+jSJkN_Hh^*AN zOYj~M5Y3KazGPK`wx&i1r$QDGOZEk;98`6C0J>_B7?Ld_fTqo=Cm?Gm5FZ;98*@>b z2uJ;SGADpiY!SUu8bMK(Ym4Z?7v)qQGH2~UEyt*)LUdW0+%{W8wS({m)HNvY8vqq@ zwumwo&daNS%rFs!WLG40)0M52I(8KAM){Cep7Nc_o4$57gQ_wGw(3IBTq}i+wCCH? zVLh_(8+iNYWY*A-y(khh;g#C>!D49GwgK7ISY%A0dTKAl>nr#-a&WMCrF>Y0hvqD^ zteR?y?3K`OAWSR5tf2oh6Bx==P#eY9-X73LUI@mIz{pJbGA=VE-z}%^K>dWO==8Li zgSF$?Ob!$(h-CL5Of>yFJ(n52_8uZLD@S4nip)%Nn@JavNwr5tSE}Nzr_JQMSc}aJ zhJG|*h7jg``XBBtGkoo>0)51>nx)jq=+Hc)`;}4C0o#SlR7fcvvEXtrBUk0fuPVdu zzfvN}b%`h9%}dylLY)WZW>NbLf0O;E1BVL)FQ$!TED!%%N`Qj8Rp9v`A19 zx4y(D3u*)xocT&wRd8B9)zVS*pNPdr^Uy+f>x8R{)AOl8DeJFP7c;uk(Wa7`!LbwC z#i}zi@~J6Q$z!fxn?CzuMzt_kj9N`95**V#J_%KCh2~fL9hEM%y=^G}5AiFcqW?DT zlo~oLL>0@PUj>yjDqX5(u2B9Tk9kQ&{|(Y9)g>;y`lVog73Zk9RE6N8XXiz!CKyXw zlZu2uDa1AVg9hs8bA{C#j!N5VTd^dsZO4#`mcfp#VFPiX-Kwx!XjDam?K&!`fB4d! zWmL^6xztW*@Nhi+JGQo~;aW*c!I`}o6R;&1qkg5b{Ld++rQpn#a*b6~wi4=JN2Nn3 zRR-M>EafPVQv8qBO{sBpo2pOn^QrQVN|%Zs?NO`HB27s}f)n`o9)6WEs)XupR8N9l zdVPR8Ss+997DF1z#>2xkbf-r?s))BDg9ce4y(?go_E3|ID(HPHr1{{?N>y5jUl;@p zwxat%WyKt6VyU2^R@f(7F&Rb8JCW=s^$>p&t43H++3v=v)5yN8RaP~;BpuLyXhjV< zo?ktJE*KSZQ7g1Q^mTrss;60=Wi2u3&JCqc17uu_R`~ux&VNG*;4>Y*^_`Y1br2pr6NuPt=7#;tDvv5 z#DR?A*6P>mtDxoS2wfz8UQ-pcGAwa_WwiqJeNT1XLAcfCvo()&93^)kg zDXkZmi;$I|Z^9mb5u_eN<%Kbxr_x0jyCbzR1KLA>uMfk}9N~8G4^<*nDe_lI@TKjn zi)ux2@HJ3X9V9!7f{(rwnh&>=?aN{pRWZnU0jiftUBFRSw4-wDjZ!6%^Bz>gOzEN4 znpfawRXggdM@1C1)(q(88U19@`|T_{Kg_Mj*A38ZB|TRkC?9WUK2#9@s|Nbh(EV!k zF~2^kt%7RXA*CABQTLJkuTh2f+k*d&@PZvy{M*8KO|}wd$VgJH=sMbA5jat`N+Ul` z>#S&M$`V}oMcWfMCIgX>Z$T|naRfA%81U1PHyIi+P z0F}Nnw*HVfWQ$$msYB2=cF3M#cnlPUvYY{sebw{jF)=D=yB#%oNkzGj8w1dSK-?SB zZ+U`SDcQ~%G$KUwJ)!I4=$Cg#Z`o`6_K(8{7AT^Rq5CwI{-9l8`Rff-BhoK}ZWZad zy*qsBEv|x&+P+_RmErDU?S<~JDKGk;?E-@<6yok+U4!nnDL=~M3eMUEFV4u${mBYL zPtS%FFmU%qF%|Ti?Q7pq`lD46y7Hvw06Vb$ie2E1$<@U`6X;r+@}n#U+^`EaEn7zn zybfhQ0|;OE32t256LM-jHUmc^e}bc<+b`%1J8a~9Y}BV^*yFKpQBEdVhG(i7|XjOjR zkDu3|A$095kE#h(VH%IUeLvgOD=^s)eK;?x-<@=I@YMZ3o+%q1tXrJ45I%p*a=#@(XkqjoukT zr?#^dwbm2p(pA&dJ40xX&B1CN^m(BxZ1iRb?UJj8ibHmSQJEq1G5(Q))du+;Qt2>+ z{%{5dfqjraNb6D!pG4KX-ZLaHKlD(AL)eNCm-*2osBfFPYS>HN(4526YdekuJ#v1(w(wiZ4 z+DG`&(HBsDtpTZq(5gG-&5oVW?<34sDsP6+n`Qm#S7cw;#mx}v+e|}f2=doh?W6H_ z1cuPh9ZE&NY*H3-?mjmiw6(QPy0{2%$VQ|T~-e)I-5W7Z)52dzsr zgpNz1Ip%R_e>I4!aKjlwZ~1T{Z2b*wP|bkb)-6WpA+*sFypCbzg0_f8q#8nhES^o> zM;R*tfJPdWY6x92rl5eVu0XtLQ0$HwLO<$MG=TaTh|dj*T{1&x=p4K;`W2Ml835UL zhS2RFRFEO^FaReA#J#}`q1*Zei~crr4;;N7Lf?Kdk-LMHwHAH^h!iN#dJLiW&IYM} zDdKX_RUtjMw;4h+^e)WZ#cBavk|}S7&_Zt(;O=1cfv&$P?+l@5ca`VqyT5njtj9mfB+AC+L1Q<((n))AQkC;0}}z4ZsYccb<#J|AWJA7KJ3y zg@+#~%tPpgyEqUlkNg@)7((gBV2BK8`gZlw5E5BuS_b8zA+$=JXAYs&p>l@Mi*Hm> zqfzq&vK*d>b~QujiJ4()0kT(Ul~n@&g4J2bkLzoz?a&=G`kYkW4543b#s1tSC~s*% zsv&glgGBBaq3|d(lAV{zn<4Z})=H`{vdfT)Tr)%H%po*{HbS*)tZB;=Fid?Fd=jX% z7xIMGL+FMMF=_yiBdAUf)tO=zTJ{n)#->6yU+V)yX#M6q3$3`eyxIxXVI&?x=^~72 zk=j^9o3MxeA$LllIWUA?NW>6I{yIbG`9F(u@WFLa1Cm{gf;U6xv`P`GB65mARf5#j z95sf}k1ym?`OzA6p=xAGJ45KFIinT%@(OgX8ND-vUddfhQEPnw-5Any_0ACbd(j-K z9`p;KTWa)X2yNLpqxu2aJB-Q1mnnhqk^!v|xlDLjTT#qj{?%wB0o#)eu@5FViKVj5Qp9aT=6r z2z{?`MFCk~0`ZMOu{&l6O*b`L0CfO}lLo~unISY=emt3PLivvYkbP$eEs7^Q?NV8h z_3;uElJo{Mgf3W9PV^O_tLEtS5IXx#LGBJ#Yv|gi(qjny_We>SivEA{ThP5rdTwts zgl;>6PmPm)5_Ho|c{7Cma5)Qi2Wt&<-<$Hz5Za-n^e5{Wl&1~A453e+s~`sMLHCIC z+}+I(dc1zL7>L515`!cTIz#B@{1v#vT9u)!ZUANo{rM6;B+(N2ZH+EwNldT`YG{Y_ z!pCg}A$t@O-?elR#Y$8Eg$2;6DGQl+js^+@2Q(J3Ih%JCYIYIJV*Tv6_G zi=qFTF#X8B8J({`%&C4v_I_R5jLy8>>b(L&)>t2m!VQhlxqriFjn0JDqw}8f4ODg@ z3n1AY=?+aXI{z^XZCn|;T3R0%oj2yGN)y~4s%KHJK=lR^w=-RYF)dOXi)a(}&>z(n zZAGJV(WYRS{B=g>CX3=2wE#6NMY0D|@Md%l#>b0BBWD{_yGcF4QDbz@9bHS!M{E2B z)fH3P8J%m+t*givPb2IJBT4Pf=-g#0K0=MwDhORH>A8Ambk2XjplX2fwV`WZ^k#I< zhcBOYM0OXWGNbcPSFz1H5cxw==`cEP+KYj6GVRfLK$>+>pN(FG>GXGxEY-< zWWYqmItlGXgP6+*Jvx7VmB#N!&;~aS1eR)a{&;enT7)uIegH~nP^!^+z#mlwWYq_v zg+Z}9W^~>-Eqwsh3yA&(#V(oAd1(*)UUVFkQw@OZJEQXn^b8)IR|Bw)K-?S5=-l?@ zzy$Xv=zezedUU?L0-w7=b6I~u_jf8iM(3`x8mMxl&(;L{-$>HlW^~RnuOfFBt2}g7 zNzeAp=zOtyb?y#UOX%8~^3LddyJ$A&$y@mYyj;>TCtV*#*&}V)+Wkui?%!`EKmTn9t zKn65@yLxE|DcSixeg~8AVdjgiutK7Xgi)UJ74xemDBBXr{)$>oG!wJF`FWJ;hTJ|{ zWAV~f)GM+!n_pdxgz6Kc=FLMr1PIEuKrW{i+01myvE0+PMI+-7;qK#ZZ-y$I7|m z7S6UoV**t595r)V(8^HzVny{3KO^V_)hmvg`Se*4hpuPhE#mi~8sVs?97BjDt&si8 z<%@MQq58s656E6ZC6u=cUd7l%1D>{)SqiZ^`D^n*-^I`7OfJx!LYa zZPC50qI*%dUt(pc8ZNibQRvSZjCA|gwxXW3R;E0^*HI96*Li?db!=~c<{ZES(osYN8-fMJ z-chkD*rT!cZtT4g)YwZjw%A)Vme>=!(MU8JyD?&H*b{r~UGu$bHhX5t|92O4r8g?!vhQ>RRnk{lEst_y$GMS4h_n^PIx@3)Exj9qk*I5pf<4SE>Ca zg!2Z}huWcbM!)2(B}7~{hnLE}o&DBNuPx%j9GR%&bvXOYUWqsh$LDYnL~hQ0k8qnz zTy}>)b*%2ryN-ZY)kR!{L*1DmWd;wR>ts?iaiOwyHc<*v zEXf)0M_NLPjPqRgVBD0HO*DbjQkOOv=ktNS;wmZ}1Z{-YO~(1}{i5PF^f}NLlCG*Z z8RxHW5=02<-v(`u)^)}iG%LS24V-CME_)TY{zM>*__bI=-krn@)PVYVY(#+ma$22lWI zgbGfZ^_DOuB&j|(=N42I^*2c6c_eIQ{2pppS9hz ze3NnZ@ywzcgN%eBnoO!S=!`QScbKV+v^ul|quUwhfIfbzImph?zDTDdp z=Cc;yfgte4&)PtJKHy55Xlnsc&R1!XWx84vyqe#kf}Fs!3M|Dy0!nVcu}SeXYUxU5P0l;;26N&5^)X41YkPYH1a zHQ%Hq|4B9KNqgJ`Pw^Vqi2_QJYKfMPNqeW9E)fDPLhBTiX3{=>Dqhc67GgEqkZ#hx zJY%eC7_FgqB+q>+uP5#0yef;Kz+c(L^`za}V(OB%$?65L?=WdE)Z+g)X(zWmX+QI4 z32_|tou@i~$KpPnv~Qh*yC@z&du8islXj2zP@1$4sar^dVnLDksYyHi;iGnoRof@W zWQpUIpTw(^vp9sfYc4H^m8iMjD_%Hj&UN6{qMBXnloam5 zleT4tjQX{zxRk}8is7$uQ{2U+5cQYbA-~vIQe4Wa{&MV)Zn%PSDOCOC+#wrXt0^vJ zQ-8VbkW}TRaKFR2Tg<&|k^Oaqe@gxg-ffVf2L&XEloH;{`eCPM_ij>Iq!jYJ_a!>O zG6#OlE>cSS{DmdEqKaItBT~w`9_0=e9xc)S@pJq@Wr7cED=AW{g#`@35y%CcH=R_= z=i4}Z=ph!krY*^K)OT1#6QbJp!{wS*(0Xc}R!7omq?sBJyo?KTYMr#%H-hguQ3CJPoBsBU7-Ry$UCBQ?@l6R zTCm@nn$4fp!h9r22lm*!bwydNqvKM>=!OlRs^0xLaut*A0TZpGTnHa@_lf|?h!)mgQJj68+qyOLL%jdOe@BR3U%!7 zLi*Rp(e=vfM`vJLJDcSij_AYbyB(0F_$7;Y1^huvI!$urz-u>y+G6&HpH;1;Ax(Na2bYZt zgO{_lbZOF%b@UioL#VB^qYHHkI!ziFjSJgy0MswFqap2@f_9ozxjD`Z$~jP%*^YE+ z((sa@qAAMA-7p-oP3h94zdF^mO!68W_q9u{j!u)F)+mVPDh$l z2q%=MQ|GxJ+E&ul@z!Zl1ChaM@yDV4q08$u=|bk5syWC9(4Ok@CQT~7H;Zad(g!U) z2*0!i*!Mr)?ODuPKp|+wNmm=F)1>eVk=6q0LiAG8>)n^?@e2gVtN?Iu>u0C?UQAo}?umiwgpx@cn1-Rp~Uu;*>h{{%i2xY%N_Zc8kM}jPeT9 zJK9m1HeAQzj!g7$*nd#HI;O3*tf?G!ES_A7&RFJxTF`c+i^Z(L#YA0{ku_jwXq(c- z;;(glEtBjC#}MsOtD|EvC@MV7H4~1-+ND-W$D-G6+{yJL#C_U8%Qvy;csZY8JWe$SITYHLy1a?S3Fj-R_9W*+T&fLv%2LnNkT}MxA0T-ZM)#Xhr2JJ0sE#NIgfpD_6K*wU{M{Ys}gNJDiv6vr!2+zfcpR%V* z)l#YEl5Qr|yx0o|;eV!@RZ&o_0?$k*)$;M*iOyj) z@K#%rfu?9eRK7hJmkvWauXS3TPgBi1H47sDhvEIFCbi;rs#&-vo~R7yf=$Oyt)F~B zOUj;>YN~4OR5R#uN#&DOV6H_4s2kC#ru(q`mfRjvS1sGArd3r(E7iO_wyv_sY4FX` zUOUyi@I?+wUJq%jmhDtizCt2nrixq!va_OEovv$SLT+$pMgt896Q0 zyy#oTl4nAiXXLb0v&Adx`9!$|(q3DZs$JTt<`=!ZRYR5Mpx@S>|C?&YOsj4gKf>gR zcC5D3YB6@Ixi7Q3wLZC_7y8tbmTEe`DrQ-#LvE1Hl9p=r!6d8NT^ynA(EHdP5$qb3 zNu=}%Ts{aFBBub)#m@~hwVRH#mUnFyh{-?kc+O)VcaaS8o}}*hois9r#E02q#SitfiENN^ zk*rFaAu(r@FtG)Cd1#fjZid7o(}To3)YuYQd(u_)W=Q<(r4WBY9|3K&*7cAWi|fX7 zffs2>4~eCR;r$j{!FQ(9FeJvk$dC8`fM2k+bVK5wd1Ek~J%;*DJ1)|O>mhNqJ8qDX zfjw~Q6F;@u@|tqkLt=kP@3ANWwUX^fHzYO~TSfehGO{@goorLOA#ohu^`T5M369Cy zrB+7|i8&9urMXtYv01y+D(NBd`IoqZ;x~wAw1KKNL!zfg+J`jn!|;Sm>UioQ@#&4U z$M7mhE{20bJ`%30i6z;S4QkghgR4~eTjWVaUZ z0NPVs-VBL(it!RC&qdUs>q4Is`=;-vc@5ZuXILJguUz0q_$cV+6EAokm%-O_hNE>u%6N^*N zMu-^bhoPO&x{1X{wjWjXn|a3)Ew_BV8;GT;wKZqKv!^!(-c&E*6Kc$!3{k#@@K~6+g8Pq*@&viy!BD zrn!p35vyHlm2@nIbi=JdjUcwx23o#}#U1X2t@LyV42fh?twG1)s!SI|D_@;VlG|5sZz2bl#8 zIUGN$yotr4IQ2m7NtS_FQ5$qD-e_3TT0k>sZAe!esADm8WuUcyq0qk6%KYly{LI0-5|W6{9ZJCc$@MG4s8pI%IBS=@(n`m74DpmwziF2SW)Vhww%9|3zX5j5w($ToMGM>6R4*o|v z4bhmx6QkxG@MpG`E*kqcLk2Cq`r$eQerl!U#7|D#1SQaEnSxMDYeyLE8hzsCB_86G zvL4jtwj*6Mt}0kb z>2yS6k-u|_nrJSvGqf*ASI1jNxc)}>7>j?ID>6hXfZ?H{e1SjpB+uRgoeVxQdo^EPhuE@3d_O-X@)fSWIYwTl)usC)rxM zSX}VCJK|_A)aBZ-l{Q?Dj7@y;W&*hr>aW_dQ#tIB@mw)_RoX?Uw{1tdSgb9>#6H^i z|6uSMl(zNhVlm#QfMt@Aa1_xlwK_T$x5Vd7b0xs>xpt{l(y@4RcmaI>0pbvCpyiub zZ1qi?6^q}%@GY5CYtXUiOfF^VTcGVQx*dznp2Vu=AkRQMpH4?Cex4sEr*Syt6KJnU zSItSsV#X4os=3GjG}J8kS_z~hRZP2mk+>Uc&r@+r?4Y9a$5XIu> z2!OH+_T>?-J=ZeSABht;FaL@adu|uaAh%Y@AWmeGJN`rC3mk&ypYU_^*n~VcIVXW zV$Xd~IwvhXiI3>od7wCPDEqN6H|3KJ09)DxB>l@eFSh4_XU1)Gq9&k50deA7_9<-~ zR)Nv5O#QS##DT2QV$U5<@3leJm9G;=i(bDGqHJ;tynEC6fB!2%?D@wtr}m$UUKv$D z+2nP2ACR98r?o;ioC9dB_ju$liDBI_3~&9xPi>+y$T76BX|7*9Zgwq+_dmf==2O>1 zwM*#S*q(=;kyy4==f+N?hBYXXUzueal;~uavd)d2xEU7yRg5JkLi$R}A#@h%#2t@( zjjS8>a?yN1(b^yCw!#GzTCKm-WiU*+NG-URfa(6QTjV1U*jQuqYYLK3oH1G6IrB3-6%Yr zyBgwFZJ?$pJih5)Q=G^v!MNf zS#hGM^mtzlr%JQH5pKJL@6_i};zTLwF?=Jm;@}l+&Dyz^p}h1cITN3RY7DW{CkAWx zRqQ@MoM<3D>fKDcV3r8eL~T^DUXjk5ooFKcqhbn+6nbOx3K-VdCgBl_tK}zJN>{5V z#jLZ9$?#Za}b_%h+^ly$Yq_QyOc!d1@O`w zbEjSXdqNFpqfx;Gsw9qbQ+-|_`Yj%P16c;?dy?xI8QG5~&TuyZ*$?TEE^V@(_n*}i zGlTH{2WYpoZnB>P@2iLsDE}Wc_e8sTll|mf5Gh(i&jBq`>pJ@hZe2r^1CG;@&VJ-O zJjd`kc&l_8vY)*j@k~>H@I+fnm;Fr7igW#PHq@os(Gja*vL6|XL_%(ddO$n6D~Fx^ zZ21g#y2!tv-n1R*vY)$GVnrX6ksn}ieVMlP>9U`cnmH|#jDRCXyVUCF?5E0NndYhu zM`P_$tE97^=XLSYiXIRLX#*|aWIu`H0?a{t z2GtzoDQM@?>BxSj>aH`a^C>IlTJe{KCXwZ z{3E#M=(I(qi^VIr*G9^mP@}cO6IZiMEDnxC$|tKpt*ae=%3;T17hl}GB|AawV>{Bt z;>anvMP`(dlVF%_o6^N%j9VqkBsamaTf5Zi=vbWJwMLrj930oQORbWQ#RGY;u$K_! z7<;>E`6d?IAo)`FTV;bG7nxLR(6KnTfwc7U&?+0T&cWmI#JEupnfry~{v#(9a1 z!B~kA&_1MPq=Z({MmlMvOOrA?pkr}Tcig$q9K5a8 z5Q}uxN_bvFcoy2c%uBc~XYl`XJZ{!@`0v3dTmngCde zw{z0`2(blpudT>`zQJH?$^Op0sfEO8NcXfn4F6J>^`v_|Hi!;unmfC5be)poBlPgG zX^WA@eGX4Qu_Ad_RL-&ZtoJ_;UWT3%%^hn&V%{H<<7pr(;~!=6@x?#$aQRsi2mti zadEkOVD(c41?~4vPBWX%eQvc;_CUUF;FHNHSg2_ak-y=t`E7y0UY~fP`)SC z+OFz&=Hv1p|M4qpDs>-}gQjeQ4({UeuzZ7q0z~i}JPi%yhEeH#b(a&Jo2y$-vA=~P z&;Zp|;~_QVm6u0*HECU29g7es*^L_74{J5fZ`fSx{Cz1Xu|_S`yM(wr$#uSvbw7M# zC~a+3MyUPSY-F&wJk7Pt9e032nn*HjlB7dxHPPR3C`6Pp&Uejz?JE|+v|bzOxLB=f z=p#(tFMl7Q8oy(qyqG^iOvIxsc9U@&-D~oJZ&;;J)h<0I z;QSwcj_s6|=z(ute0H%HBob0I$;n2>!1u9WelZoK8l-q#+6;W(?=385Lhl5vht|!& z*K}cNF%SA!Xp>1-)tiBDWrlFE68cJL>$I*1zKPdMi2cBaw4?{V1{tdhc?tY_It>G# zcSGEo{EGO*v_+;H_-3YHRxGnZ&7&Phuo`CIyHGz4j`C1zXvazAum`@!Eiq}6t)X_a z9q9(XVij|VKT$@Gfnl0$N;mK|zhBZa$u)5Ns9kDx^uXsA>Xzm@3C9KPQmdo~z54;K{kfgBAt$b z&*yYjF^aZ$AhhA6tK+Q)zVf-z{gFN!+5%l(4}2FVx>R$JKSA5A%bS6(|NLUAJ;`$r zFKL4w_^KSOYc1dfw0ESd=B@|6WrJd@1!P8~WW&!|K{N2R2+d_JpcKSdZO{Xs9E7|x~uTO?N(A8?Ww~!f8JODqnbt2?w!+3Ee(c87RoQTJB9GUa84IR&B+N(kwaQ2U_E<*4bM?WgJeH!k0ls$n$ zM~7o&Cd&zx@J@>cIbvozgv&KqUOI-04h^k8{UbJV0U|ND4MZ~vFe!W%=W|^Z}x=NU+e8m9badS+=dwY3a5iBl@UONcRrP5_YdFuk z7Zra&yQy_m^?m0@&&ndTrn5yIRIrCO{39Hm)6@2a)heRsrL?KFUC99fA{c#QwM{ph=W{1TU17y+RAHs>pUupY6GPcskLd_RI;4f-t|?kKrs~bE4v&^ zmQy>s3LUe$l7*0$>%uCfP3`V~*|UgBcy~eBZ&x5JwO8=VR#{ZSdmhRaQdTLpYM<0Z@h+m6GMuiGjWS@EjLXyl28q(-k z5z!d?a1)emx@^?eDKR2-wyVy0m-rp@PrIC&yCA1UznePGD`t5S)$cxl{L~cQg1JfR zB3IbgHF264F@c{I6V^0|64BJvt~-N@iAYH0wEVv^*3_-8nJ06IdN8%oMw78d73%FS zQh#>kKJF0xp(kpOT2^#Xk`K6kx^GRr=RjU)Tg1?lo{pM*p$Z-vbREbWEi%u<_>Z6CI30tqjNxeS9@4e=fV(IHTuMvLR3DyO zde^BEClY`^w3otERrG||-!_N@rB#sLgZH2to(MkA)>L8Hm3l6#hB)aw zQ4Kx%MmTn9*Lhky%TP}`r=7uzA5THNs0~+?;rnyAbk|BcSGB@}ERP|+vkg|nx1wH2 zwv&O`W|bB%6EPrByhOS=>u z219N#Su0O7M7ckN$mf_^}chXC|$zbWNpmi`h%^~GF=^8#g?fTY8 zXk&~%wiR~HHtD^3K>=&y7eQM=y4rwF_db@w*I{MI=8lB~P5tCPXjVJxbm=hG?V=3H znH*$k7mfD*en#B?gImPw6}7aNT0?e9QPHwPf)H7&(fJvwxW|j$4&mvI2#L+C!ex{r z!>Vu5?p3_Za##TqrUC|UARFxE5H(@0i=QJu&33JbS>bT`4vG*RfV*j_*+t5FesBce zx}94j0#CFhYfBQ8zS-ejBnv)Swglo@ZLpfNGHi1=w`a$dfc+4UYC{QHxUVuCbU4R* z2(ip-yl<87)EGQT7Y82mG2#{Z{Ee@J7o~RwGGw> ze57!?3Jt>5^9|rTZB0^r!lNNRNFuX2y)R#Nw_42~F#JWP8cJ8KHiy%zCLosKb8+gr~YL$(k zH8Z+u)peZSKevdr)~`Rbp`@#(Vzuh#&hWd7tErPVayB%pE&SisNoREmP3~Ju$g8fv zOE5OR-2~rb@$^d#x88$%#aS!JDXx0?on4B{s%mf%g~qD_et{+C;dM>)$5nsl^YL}X zdsL7C%+ZjJj5X?wBRR-<`YJ{MvfHxNGRiE#E7z_DTOTD6h2DILr^>Hp3mgGg46Tlc zB6Ky(dA&_t(Gs*hm2X7{VxfEu>DhLi3b&b6r&#J-+|pZI3(iQ7ayyqg*RBf_*RrU; z97~0uO zs;ySwJ#9@DPB*44FORNx3>?$6E1cF&k7>15qt$;0aicawD1$ww`Lw{a=ODyWwt7!+SNznTkZMtSOe_CQ3A`;1H5G@UESU<^ zs^~E-=%cqNNBU>bnj767)BbtxZ(xZ^F$^!sWUahCrcIn1WktU~+E6flX}Ue8z3iIB z(u+VVWpsN?Yq87ATKNXhJ~z5OrVY7*_a)HI=?iTz>1qS?nD)3=Rj~)HQ_h5DwKF}Y z-3qB|X&a4Zk7+fQ`B~agt)bVrhCkGp7N^EEs(7KJ7rqGe#}bUV?tu}WX)kYZ#(4TV z0vF-z3Y~BfH7jkv`GuMbWF6rUV;!MAON;Y&aB+g_%MVux{2a6yRwMImFeF?!kGWj6 zH{;vJ@emu>2H|^qLT%yP<#JUUg^w#}Sweh37|EWVbmFe)$MC6<2-XZ_b)iVc;@DRwJSPAhrcfwTfo%g6QfLXR}J+R&YN!jhqmGp zq{H8b+!NuP_dndXrSno=;XLk^WzMA9A}I{V$KmfyB^-NTKR`v)PMa9zWOPBm%D1u3tV zr_0J0{Dd>WnX8ePht=X8{+_gy+VC{CeNw&1Bs5Z17EXNMC8evo^7ev0!1jpg@)@;* zcaA)`&>m*F{H`s+qM>HMl`7Gyr}Q8dF+Yq_amZOiE-msP(l&QpcmtP2Y%?K}T|irPb2|jzOQgJl03m z5;qFTEZfHB6pM~yY13htL#7+FHXbo8yv2>8($!(S)qv$DXxoh5r*ovZQBr1Y^<`G^ z!)}y61?`;C|G8N}+$f8$fekG$JW8SM{0r@c(H{ns7dI+Mzvrdu;E`gSSM$SO%8Xyy z^0REME^btoeuqE!T6$qlt(hr>CMpNnf7Po21uHgM` zP4>Qkjq_Co+{qNS1M{+p5NBz_9ol-|4MBOujkeM?7u|te3vrh=wBCh{QPnh4$F!65 z{w+(eI0?fAGN~i_#|964I<432BH~6r8TQRB)LH6`BNTTw$|9tpjhsglxFzQg?n~w;BN+ z!r4-0-1Z)uMwNDjufOfJde7-HWmL3qwvr`l^(-JdAHi*TIKfBe*)T33b1_=pOlOl5 z$bj3)pv+t9*kR%%-%jljp7eL-wzw#OYSS8Ym1yN=UVCIq_cp7S0GDM zrvtZ2l2)CU;y+a31@7op7u%PUmS>3(uki^p2zj*_MPe8gbgq-`4;K^@vCwL2y)1Qw zYh}N@=v04@*%nWB2)B@5p`k>IaXhU(e^u!NaqY!4sika7D?-A z%hbwQ`cdi3S06@t-?j^f9zhQ%#YJd0+#=(ACcP(Q3lOxdrjT0ek_%-UtXcN6cnZQUV();1*a2mW#nn^^?BVZC z0!s%(C$PI}NwE$^H&gKer04|JJ%3?w2>6(mbOKud&nsU8zHduXRjL!%sh#Ll{7Q6! z_*vEJlQIvl;aq$;cp+O;h0`Ul9lX$wR)wRUc6~`}rxVx=c|3*e1hJ1cj8z6Zfz34% zXKltooM9WRJ!9pjNMIklDI_MNh+GH5HrpiCVsrv~_$|JlK&DeLT+}ADD4oFeNGc*! zGk*@lTQbd{RnZA-o_~DALaHRwDqR1=FHN@-*poPZs(cA(<&178u*s+LSR-O%Xf3p^ z6WFFBizEL7A55A$iaLRPn~eQ575rOUle9f1fgLxmj8*v-77W~rcw-ea7#In}L7ll@ubhQCGfo+CY zgHn*n&!AcDOee5!@YQgo^)Z^Az$WjFwzMfm+k{s%$m24kJRZka<+ejpZzbD|fB3W* zf*m8(|0CC>9z^B3T~)42{)LXt4cx?ico$0g*XnL%#4Y^gI6=Eqd?$ye7RN=Sn~|>U zt8wLB7RiqDeo@)7;hQclQQ^S9J_0XN7Oo28PZY8Kg(6=!txQGMxW=JK4AqT$U5ly* z@7x*}6Q^-b+^uo(wEvb!`)|p#|CUPoZ|SuEmPz|>*|h%_Eu#J*4m!^-i9wg0+$_WW zA8_mKKVxx9N-lD1B9hyA(z@b8SG$H%B)4~8NPmR{m(NpDBzJapn}*MC90N+h&k;xu zD60E!g6lo6B$B(jmh8=hm(W3b_la(qLPle>O77*#f7E*WtRvGJw14~@A*e_M*L&=Z z$K_q~kQh1&1D7O;mLYgM?5$==DfwM(r#-p2;!GkIlbX5d46ENAeSXaK+13A z4y9|0ZKLQt215M zm!K?i8ho>KRoSQEU24grJ(qp1z3W2%S>vWGau`?)kMDm2R7NNX{0y=x@kzWt`&jJ^($Jk~RuS`WRRBKfG#u+@Sh+;fhqWEtpH$wRBk zJ?DdRbS!Mu(+YamboJ27K$0)}^eUdus%I4R@pe5TM9J5JE-cM!Rk0M(D!U31LhlMm zzU{N2VQH)MK1c^mX(itcO5B^%Dt!ggO5DjYSSs#v!I-ZwlEO}4LeExPU| z9IYUB{j|7x>?3)QbnQNY^9qR&Cu##73!xsvNFFI&olc_WB@kEI26V7$11hZXPrDHZ zTXFe|;f z@@naNIJzRPwdj#GxUH@@&W03 z-`XA5zv1}x6PNI45LZ|v|0dmAbg!d!l)M7NP1|IhOHX!pdR*Ov3j=S!-8S0WTtsa% zJ#v$LO}hN?L01_Hv4A$vYNxG-a@~`zeAAHC$H9@HT}$wAk?9K!d zIcqw+2X&6MoR#5f_^DGJF3Z`;;a%`N-f2!1d;!;}Pn~Rc(4$q!=kb}G!e~x&Vf+45 ztJ)_Xbg44AkTdhHM%e`&Gx-Z#$Fx&5UCY$gnfVMpv8PP8VR)=fv1&u;-lpV#9IpI$ zzbKvVl3s|P%=kH~P`y@7baPYk1BWX*H0>Rng<&X{&a|A?>9NBVS$;L3W{nX@3 z=e&|XI$Zhab!mejjv|AKNEH#fbE3{@oO}L`<;1OWnK1IDH|)x2gp`Y!@w-qV6)#t7?y`$9OZ6a2=vF-XgrBt+LTT;J$hlv`SFdoLe`;B$opBb{Fx z^}VL0W>cpSJ3;B8)iV7q;czwbZMMjP@$R480peEUz{$CT1Rbd9u+7odsM>hR)2(KH zc_)O3wV*$dqK@o+I!_#3m*iHNCSJ%S2pos}hb}yn?i9S$Ufn4e=NS6@tcSSOA&?eQ z%Q18at})%}=tHF%I2OB2f0SL^>Kr;^PhFv^clf(#vmBu~K-M-1PXcA#o?DliYT!DS zeftEJ&@=TDnLamz3^Z_2EB|THm*(yLx2A*R}QdO#F>bo!F{Mru? zx7Y@AscEzmBk5uA^R}i6r< z{o64}LkmMJXB*PZ)CXMm5wlQ4Hin_CZIWs+dZzwvTV2aE42H4Vq!y)T>Muv*LKCOzLsV^T?Tm_-r2koHI?U}m!psZH;RA{%1ZqL-SRZKf~{SlhmHoJa3Q_oxl zS6D;AbCRZxqMoUb>zWf|0eEFwleE!hrrvXH+NGeDFtjI=wet2%J!^xaR%Sc`+GwNO zGxaPps#*F%Xv>Xm&(z(!q@8Bi1MPs(?U{P6Fb`|vFG0Iby4nCeQ@{U9b=C36574Z3 zrf2GzPnWi|tZ3h9nmtpm@GOs|mDCzKZHfvt{q9(uZl9nYEpzk{yx~1D$Pu`17cPa< zUx}F@OQcAyP>mpMpTpvly;|N&rx`>k8 zcj#Kn;XbUyNJl`P1@)Aa(@0&K#EQfyM}Xh~KYmv#uq)g(P-rdFI|cBC&|W z74z77=}T5fIgI=rZDUb~YyCgg2X4whs%Yenw0DX*T!-H0RM|vRNUe>0mv%#OhpTda z>q5|ANF$B>5wE68Ea7kszZj)T&w;ei$luWJDCuyG$YNa~{t41=pEGZHb9R3emmsfIqNKK49{D%aQSl{92s#i)~Kj{r=pq3L< zXf$$oZ^RAI&dJ~lNTb-0)L5*h+>c&EocNi;Tg3$K<0C0(Ic>DJBLkgmKHyk$s-j5j z=8#FdF=nggh(9R$+-SBFsam{EFbauC)zEE%Dnz7e-!^d)Q6Xd#swh%5KAYg>7a~%% zZ=2w1i-;7|tZkBkh-%F?aS^dsSFNEZ2sXi9Yu{h738U8H<|w;}7ZthvXH`S!L4e4T zNLjSUo5=oMY|%bMa^|XGJibIC{zF@`YsM|7A~MFegduo7PDB=&S=5jKBANH&+m&{y zKq8*?a~gsU2~UX+WTi3@xm?)Ico1wNGRt7}T6R5nDndj)`$HOn*M*44qbCvJwg+!x z7m-=n9LkbfKpHsn*B19gRXHBdqk#ft?_1fD-&tylwTFwqmaU!R1|C#obAkxwoorhhj zB$3#bEC@;w+3;JE3dnyoV;-1ZribUFswK^iTtV%>G^F#XVT2wtf%Q;d5c!Z9 zX@%W1dJ?G<&w9FEM9Lju2e~(q0sjP>QhkUlYt0U6Um~uywT!19NK&uvy$$KFNc*9N z3?S0J1IBZEeFhR4-o39OgFzy1cPwtm5F&qOW?jNiBK-%&8_zHz`OC2N8BQd74Ms)~ z*(fpK+e;rwq(OOB&qN}n&$3dJh-`n(_UdFJ<+*1HkvgkcgPKYtt^_-#(}*np4Wquj z^skBRc*+|1bRyqAV#j?3k%3jpn^NBpIq{4&@R>y3e(hyEvxvBsvST`%$noy%xX&Sy z_k1~1YA%tMQ;>JtYxpgZTz!#2*<>D(voUUl%qLPD_cm$I0wUgn+4?LbGKn&NyPicv zM!V-RWHAwy#M+)EL^>W|9d;oJvUx7%Beth7ky$QQPZ1ELBX-L4%oQvmCBEFMi`0Rh zRJ->;L+a9kif3SfT#rcG7VOZ}C$h0E>*E^`IbVbwh=xRRhO^z(h)9kAwktjZNg6P) zzgc=?^88tg1#%N2yNa-s<#Qr0PP4JJDUs`E*htlk$b|E(t7%T;s}H%&vRV+?ewlR$ zEs6AaJ9sqa||HJV5PKQ>B_AyU?hjgwyydD4kB!m&g`d8u(kQkt?xIG#vP-oPdh=^DfK z^HL(2@V<53ewGm#`H%(jaw6e)*`@ZZAmTsP*O2dsw3^PyN+LP>vQBjskzC`L=X)Yu z=d<>+nn;O$tkfDJ0|XmT))LvClXXYyh*WRK*7*k_cRRC|u%5`I_AHs+K;$VewUNk? z%548^B2ugaE47)(p&YC|Zy_>f6l)1vi7b7}j^K|(PNHko`|u|s7oW54dK;1Q&seGL zM1EQnVM-wr5s|(xSgD;vD#o+O-9_Z#OGb7RSu~Ap#U3IX?#G*&_Yx`i#AV3OAW5A% z4>V*SdE!>GE^9xLzOy2Y=NBT~IXOV2!d2E4Clh%vjYZtAM5g+))Z!qKXOG!2Jw#;4 z9aic$B2!+n)^eE0SWb=*p=&~V&;L%O%~duK93?UwUuo5zV??&&w5uk^iTMAV&yW*D z;-0Zy;v|vLf3Obx6cKuBp|0mN5z&ARd1pY9I&~jxJb#cU@?sf7&JvmWhOP6TL=N%6 z;vA6)+>=7&7w$O^k~FZbsrWClEIL}#kPAd|@r7R`(!C<dDsT9}BQ*?ogqpIoWZzOC<6S*7w{a(&`>-JNH#RZ2cZsfK~jE3f<_)X0FJJ zMCABYtn2xg$iv)h!t$8v={2aQ**Qu~p;iKCtB3bHGG1a^#lI>lLA#bRjuASNF|CT)87G|T;J0dGauzmWT$f0AbfA~P8 zYaH7>ABn^tV~Oj3M3R20W0sDurHaV9MOZzONEJ>TL^j`GW3-dV{9$ZFb|dosOV)VY zi8SubdQ=Z0uW`*r$8`oGl`^p{brJbw9_C4;oCizwy+F_idz*SPlIN4jh&Pc>CL=yX zDlA~@>`Ua{A8dW_1x#dOgF2X+{mHXx4BPVoM2l{eLAD0O9b_EfM%F4#OOdv^p z`gby(%tShNW5+$1NVm>IjVFsL)sclo2uLK|rl6OVl}K}ZXjhX^BBiIXJ~kWG)1^DB zIgC7=+*nM86X}5~wYr|{ME)$sx~CjePw&oMjVC90zWOqUA-RaGFCJ}3ZXz%7=~G>E z1d-Kss~eIBB&h?RDn=6N+lNg9qo@>JQ`hz6B~mSbrTY1(R7cz_q&@k`vmf{HYEpnm zp0GlOL=zd_k?q5RMAo)rU0)25OIz5CpeT_hWm%qFj7U9BiWA9xnsrEJh^)xPI<2xq zcFklpmm~6MDC>U86M1llNU}t?Q z6B(YBdGKOzOs+YpN@UwywtuP-nbnkKS=EVb;GP;FNuB$5F?+ryc|3k(qd#toM$Q?| z+D2_6qdAEuGB^VpToZ`Ie_hGcT!%;%PU;d_l*;Bu^@z;oq&|_l71>zYfXLfCY)f&s zrilC(eT8mOjfnii$!A1nRW4vWjfqU*qzRFn>sYJ%oJa^KO^J+L#^w?DumPt3oHQpg zZ6a$4Er^Wcq$NmFcQ!b*BFlh2EVx?}X?%h;pEg9kxb7u&G9SB6~RL zK%^4y!aEY_aj%5gJ)MYj;G{E=In~${p$n1iyg_v(QsffLsk#w)mWds}?nG9sVaY}h zBKI2RH%tG5$ahcJvU(DE$^)er5%*9wc=jeTa53vz`lx!gu+Fb9k*t|nseVLy^67Pd zB6B}uL)-u&AC|Ka97yEL!>rR9M5GVy;nK|z4_2T-ePEr$5F+bJv(9EH5w{R_E^Qc* z+wH2EdWI8O_$!-6j383ymzu^il1Q=ctW+YA_sJ~%|B^`J&#Ze+B2xJ=+he1MjI0=9 zY939bLKQ~F5E+=6P36BLa<&wkRE;H4Vk#r!h#c(_Z)zS-WKUhT6%&XI`H>BS6Nxlx zB}}PFL~5q8^_fg0XaP&>rVx3Xo25}xi43`z*VHqO$oxEPDEOL)=XKULrW1L{kMay6 z`Jz}`{f5ZIfoy$d5?R`Y9rsy8CI_E<3b{%xn~iPb9l*%u4gfk_|vRqE+NwUIqO@N61mx) zZRs*1?NV5&;cJP!<`egIM4rcon>GA_$XD%Hbgw6}Wj1RY8;GnP<7P^2B;vyd&rL)o z@}Y7wk#VP3DzSygOg<28CGv<5Uq2E#e2@*xKM`qw4~*!&x{XNhCoGz_6GG%W0ImYz)H8j=OQKEHw_we8i*kb@RsCnOG0sV>#ncKt>q z>?al}hl#l3qXW9;BSa46@G#_eB2VtKSUyU`rvhsm$B6v2leLZGL~c}LE&K$L3-~Ua zUe-w>XZcl>Q$$90VSDv7kr6!II78&?$*h<7gGjeYEFC*bWZ*$Bv#dXf)ZvNMIU*_d zSZhfkQXD6`^ctQg(l(wQi@%7()Mdxw0+BA?veD`ykwZOMpKyuD_XSvbc$rA^zgWaw zA+mG^YjvqaGM8ZSew9ejLN>l%BXSU}TCekUB5!|Xqty)}6|b^B^CpoM%`%u$w}`Yq zoXe2gMEZuamiaf4iQZAh^AC}aIoMX*A#yc@9l^UqzT($I?hzTnYraq9L;KvO<_AQ^ zkBJmqRM6D?gveFiGM^He@{S#^XGDH%UDcF& zP9*FLMqUtk#nITxaq1 zo=7u3?fyU{E1qW6E#V`PuQ$aR@*j~ODzNnt4v-(?*k~?^+=yd6t%FF7MJ#)F64~6z z)70!n{M4NTeG(P2x#pu4J);mm^_6A7XKl zk;re}tduvAKZdg2&xgp=25i*B^F1Q6^(@u{_z|hLl=W=>M5^XyJ!Alp4EXAx-p_$V zf|j#-f{28dW<6~tB2x~qsLTwK)V@j})$d^154kJ>9XTsq`7IkM!&rW1QSC;$bAhM$c+lM(plG^v|X|^I4d5Y|) zW=L)#U*a1xdXGh@QoV;6l7~ElcqS7`bL_lCZtP|`e?B5_GO`(Bej-bM zVf}LfA{B2cZ(ncnViRXVu^(3V0~`|B6;4jQWc3jK2*z;szl`Fhq{Ku5sA6PdbY|$ z+z+u6IaP>^2xQq{RU#)ZvL3P;kz;f6nVPGEBz539QVsGrn|m8iO(FtMoa=3`MWh8k zWmTKV`f#>}@kG4%yeWZ5?j3AP>kyebfUQqmA}tTG9;qIYeFIsRQlCh}3v3@YAo7UM z$QlyK&u3)#xDzrNJOiS);xi)dIarU~n8@h$Y+3YNRpd`a*oaS$#zb!8q$v^4M{LAz zM&$K8Hm__>WCJhNg2-PXY!q%uq<0IJL$)GPz98$_S`%p?S=+!I3KzP+VtchMk!rke zX-6d7m*s5jiMXdY&EDuhq}NV1=5{0!jE5@q9>c5(B_XL6SQ58DN&xgFHj{9QF$$$9l1TtS6C)scd%Bi%7w0Y;5XHWY-JU z+w=j!vId)F^(D`Vy)19(M`Yat)}H$lnaF3y1BlG!W5GZoqxo1ch=@ldt7kBgvIALW zK7>dseDhQ9=b=RE7h-#U7?CUGSyT=uvK#NQ(WOQZ$q~i!jgdr(e!)m0ktrdp)R#mq zEn(SC5|P6R?3j)sQtW4z`;8{Dsx8Z-#t>;clhynck%+l$UNDwOa8cI6#}O$R%X-N1 zM9$-JYQ0?(h_o!lMwy93TJK`*a1xR5tSncbOoTp1q3fAKB-dS*hfgJvuLv8drx7{B zd&sYe^liY-;7uoTGJ!?Y3?kw0%a~<-Lu6trwhw0#dGnS4BQl}|t9d1n zKgzJUSVe@s-=~lA_e6rGvWQzvgq|GIo;5@|ZD#r5T9Bm9{fuQDS%z$7)4?BP)R{mbIP8PMqY>J7WitulKOY%}yeH_(`K(M5;|^LAjeqrVzG< zdx(fE>;Ud1GQK%0^)r!9SO>lIeX5=bF^23X^7CePNPi)cw>aCY2Z*$p&pLo)B1uPB zlmC@S|G!yNJxHYIRM!0*0!ixHv6ET)Z{)e=U&D~YL^d~IrH&A3R*)Uj--#T0%=Y0? zkfg3WbdHhd`gS&SA19K!(a$XF1d$fg@)>fHNPgalo+1+T6H9+i6B)Fbt>GCW>wadQ zKZw-uWBd6mk<3+n%(DIjN$T34h5I@39OF~-6e4Gau<$xhq~|miDSr`}S-F_0`2rC) zXPhAyi44oghNeqIde3HE=4B%8^;idRg~*4Ttea0IGGh&E30H~SSy$35>zeW`VQIp3 zB9}X|A@>H6vQ60_bCXEb6t;%9i1>HQYHGesWN{DH1N=?o`8NJL^7UX$KoE5h0k2ZbDv1#3M>j95D7WT>Ul_{l4N`S5s^l2DPqV@58IeavSqku+$fV`*W}ROU+5VUv)0af(U52^`ctzxs8;D;M zsp890q&Gy~@A5LG-V#|mKhltQL=K&2Q}g#kW|U)5`GLqE%~;3(k;q~%cD(*0GUhsq z9pMDY@5Z815*ZrHT9kuGBYfsTZ{`cR00=3&|{xw=~xCL zlcNfmnq5RnJYw7KNhGcw+b%C6&yTQUk&#Fdev;3dNWXh*OMQr(>%-R2m&pEKSZd=( zq()H|DgGcyJv;U?YZySDITzVvB9O>)o;cur$(RK2!~yphA%(ooPB~;I(xf?CpI{>Q z%CHGy7LcT#eOMwBLY|lWB12XppYux!p+rKmvm_-Ok=p#^cNmeg?P{Ct3Ri@m4$n@c zJHI5JgNVmlcHDCk89ySAsV5ha>1SA5%}t~Qzr+ziWaSFBrFn=%@spsDM21#nttE=c z55KZ~n3u@>eJuIPM{MYbxv8xGDNCeJJd5sfL<%ioqh5I;WuCGgGM319 z8Q4*-KxF0sr&)SMBGWsudMXh)!zZJ0M1CC1_C{qQzWwu>da4k~$VpWq%j4KsPz~hM zM}4Z3=j(DTPHPbPd6KuOxh9c9ylvp47MLy%Wb0F#$m_E#(~BokKaa!Ig9n^&6AH2y z-9y$PvY;E=V|9r%=J!X|Q=Xfwr>#$9%O=(zHXxEx+{e`1kjPMe5~2~24~JO>@)?mD z`&fS1n8^L5EZ=BCB={&Bg+C|ayPnm9ClioMY+>!N8Ig*CEW2n<g&op%M275O0oR_$j$Ii{JspU= zY{)jNBavS)AJ)t2M8sz-E7h4ulV4ddbs_RaO*S}mC31Bl3$JcOisB7+y5{a6Nxj$! znI2>b=NDnWAhIEq^=mze)WNsebT#Pl(L;r^T(CEhm;AJCA0n*=vJLG^yfuGOCLyNT@oXMi2U@N#mQhInSkwiB7vivNONUzgup714+C3#r>nna|;3>E{Uh*a#t zW;>&aw7@6D_4lCHrB5Pqj%Rg~iR1}mM|ld7NS@J7CDP|_c66o@3FA4z*F;XNW^>W$MCkq5dfR6Z zxf{gB;BSaj`jO?hGl}dwSId-|MdYiyEI*q~#FwAynM342Y7SFsE|DjR?09{v>gkr# zc;*p#x0toV`9zwuWXoDWEDEnF-wVr z;Y|s8|12XiJr|p|EGKf4UtC@RlGJw)JG=26k$(M6H7lvmq48``UPUCClkbV#_bFjY zttPS{0}G2aL?Q;VZgwq^sRdZ$T}PyP1GY1MATqiGJAmtn+%n|~tWz}vDl*+xXJjx=OD zkwtS^f4zgqwj*q^ypzcGt*q28B1JN?kzhBG#g|x6?jbUWpZ(oS+L zrhLVwt0#&4%ct0~I zb3}G8Vo6I1k(GDYb%66k4i9FX`d>t*^E;0&5UI}-$csd>=d5kk=Ms@8OW7cNnMgrC z0l7lt$PN~#sUS%M_=9{`$@8OcAye};BAI_=`{z25h0EA}zM<;r+S`=6sR)16;TDx@ z$M15wO@vkVHFytR1|21NR%pD?M^QqrmkfZ@T&$~yS9(-bcpU6o*F@Hd0 zC!d% zG|i;Sv}X~RnLO|Konyg7wk=}Om4(RE_l3=}Lx|LlWn*wwBD2S{ULIeqMvMPDyD628 z$c?&eJPRX|^=CFM$Gr_`VNY2L%TAO8I~V~YDClc_n1$fcZKhU6tOh~6M(zjq-Y5%*#KhU6zQA)KYJ zc%3IsjfJsN(L}=Qu(&Nqq+~wkDMZ90p7pAD+XPOjuV7Ofyq5_lZF{q5FGA$wd)AK> zCE|X!hFQa6L}ol@^%N(v%;|1CB|wq}_A#I&S;p|24oVT(;l(zzG?8L_mQaR>3}&HS zmPljX`II9vJ%}~l@6^V3gTgsHGMC8IQwpnpR%2Z@s zXk{Yt{N|S`MCR}gswxq8-a%DUrE0OEygHHgAGRTqZyM`(+Y&i& zpLMkDhJFb#}15+KEVWUv^A8t9m-KJ=TRtL+M^deHTQ4_NR zdsC@?f3n`O50NPq^B7NGBJ257wI7k1MOik2Q#B%~E`L^i0FefKT0D?Qqk3%T45E7S z_hC(CFcJFFh3=b&5NXU`_BE8q#?7prVMH>`jWnf(gCu=6sDmLR$Wt{B+tQIl>hs(q zk;o^nefyH?;cu8rB1vLoF9uu6XDV6BKei0Krjez1KTCIHI$1IoTFP(+ zSx&yNba*q!^67m`JTu9%+FDs=k>&48mKJ?BS^l+J=8)x!B9YU8J#i$iAqgVz!#L~z3f&R8st>qM9Js-0@`3h}0qmuXo;mT94uS%9@Tn%k;jMUfgD~%&;E1w}mYKKC_Glx02;$MVpzPOtP>_7RxrW?2NQD^xMfYxl0K% z)DE(2m}coD{YaKO!Ip=XMV1(ACut{H@&{UeK6jC2W9Qsv%)805{gmbId&n|8 z9E%Qfnk>s(TRJH}ljT)M%RI^%va|@dG-YSW;&yz;IkLFzUp!A1W0$4zzW|H--o$Lu zj4fm7_Fg2*B&+2TS*BYpm&x+9ZACK|zmR3O^}VkvWbt`uDV?iiDRn=_40VkxS$|oY zsq18M+Xs1rEQdd~`2J0@{Azup^;fdE9U=J}S?+GM_}eX5hIHs|`N{a5G^0lZnCZDK zG(9alnD3Bgb95e4bC)ayJuLOoNtUQ!OXuqzS!VxavD_!iBCF*ASyoyt56QCkV@oUe z2U)%ewfOd*Wa%-?a`#7Md1AMC#b0DOj1N`nI`)_><)b}J%M)0JbTa?`Tm0SFH2gzB zR&=&Bd{2d>lck-1CM}jG^|`dPGvj$d@wlC6_mV8JYb_1LE3(Y=2{A*xCd+5miOhe= zVz>Ury&+3aYf-!<%Sda3^^PoatW6KYtV^+Wi!8t9oGcz!Ee*OsmSxsecMMap;%M!7 zdXVLFuXr=fo@9A*(bCX(k)?feOC|Is%l#D*W+*#Z>TL`%Ee^6&?qF#OeaLdrx+^Ld zEJHrGG&H_supTk#Ck&QfAZ*2g!ZHOG07J1^tt_q|NS40U@|kG~B8%IewqUZj?K}%1 zOM`NjpTAHMkM+rfktMd9B{|__DQ~spCQG8#5X`8qgk?w<^Y22WSYFh!JW`k} z|1`BU??uQmXQHL4EJ~KYgDrkkj4VlqEiN2QmbmklhOszVW;L~RXG)O8(arJ`g#+R- znzSy9l_JY?YxgCFETdmq_G`q#g1uM$&3wg?W@vLuvx8F^q3LT%N|T|R`F9x#GHJGD z*0U^G%I&taf#pOP>ybg_N%MC~Ni&`bWO-q=RFt7Cb-fa4N?Cu2aqb43I??h=SA{I} zwKZKW63B8rJ-2B|B#Ya&ovLKHG~P0(Ng_+7mX=@rYGkQ9)S{_Q7PrklHOS(&wWlUo z+_v`Ag2jC!Pcmt)U$-=jwaL=`m8G?>LzYqRe9T!aTNlQ7` zhh<22GZvghfu}XYGVRcSEMGsjG+t?BaXWCTAz4;hm-^Gma?rZo)QBv0d=x}KiN<8v z6Kt7IX+oAphb%pwreyiZ+tNFJk1VaMO>i@^Z0Krv>hF`~*a%CXqd8ex?y@*i3$mR1 z!s3uE$#VZ&OQYP1EX(IvT)#D0+%^b&K$e)|mWKO7vZQ)jO8FzQRQ$=}75MxZ)|;$9 zNNvaxGSw356S7pPYRSc?WSLPe(JaHZWI3?Q*R-^QWk^r+@Al&Fwx5}n&nV2eJ(dQe z16hXSLmB$nbR^5)hb&DcK0l6z(OT}Glf}c+GSu!&7Pn0fUy#LZlS3D>xV@j%l`L-W zXLTdX>28+0cPGotZkBxZAj=BtIIkyJwzsq7;!Co`aZ9LPWN|xUtv4+08zK6T=5?55 zHm5IH-j1}4T>6p4u&%=PC(Gj9mb&{DS;pZoMqR1{$kI8lWo$K&EbmrW=8^`H#bc$# z@-3SR&kVA-9jKf^mS3w|Iwv#9GQzsZF^ep1(BA3iKAS90 zT3UYm=aA(=4T}%XB}BGL?gY-vLmlVvtOYpT<{ge)&ESju@RS*~ug`~oi{i`%N_4`gv$^;}Mt-zHl6 z5i7{@taDBCUMtDsaV*ZXtRlWFuL+hFhNUCbHb-mb$T-EN&}eTVQcp|Jo`IUCnG|QXtR!MNP{#vh3Ys zY3jCEWck-_iDw^K?gv{!?T2MZKQopCq!3^8QOlqQ$>O#Kb4Y~g zV5#1RNfVmvVa9WWEP1S!qhv`gXsMRR$kJh)r7=8C7Ke3=c!DfhXDko&Bw0LW#F+Ow zMV5g*EEV-MS@Q0&G*Ca2Z3tWI@ zh^5<{E&jIj=Pn9^WlVO7;wWYPWX)x=j9X&q8vR0+9NuQE!svz_2=|=2x-|2(nDa{6?0OOS|S}`(Zw+k^ZLS(3vUhI?VSJ;v!Q0P03zM3NZc-&@);^ zn|1>`A$c|K8{l6^+;|;=#qkSp+T@U*8$ud;E_3{(LB(3||hcipkYCB~Om4D!-%q z9#|{Q^*1HA)b(Y&JM4XkiKnQ)Df#6Ld|GoX>@yVOlxg#uk}ZBNE+bzL!B!H<#H(*g z=3AK}zZrQN?1IVFHzkMHjhFlp*i)0MZ%WpPsU+hMKphOj<%(Z@Q}XtL3NrJh!736L z8PMO9d~qcY8;GjJn}W%DroSnf7e6P0bu*dzreyP=x{{624BwP2pC^y}rsP=hO-Txz z@T^l&;Uu5Nnoi=4UB)bW`sHNC|F!R^%GgQ7at@cVluA!K$(eWMm=oqdHPTMYghqB1;BHc3TOFn8G9)y+DT?qO=OA0D}z-txpI;ffz@UBmS7*7Tsg_4$@OJD z>ksyg=GsZRU-K^AyB$$NBsIJ;xmZ`X3nL`=hZ;K~y0c5;yX zBQ~%OyInhQ2?#6VY*d=vuI6}LI-C`WtXGNMuI|WRCY(8lR`KNd*zKBj`g}RNY4n20 z-o5hVFm^k|cMv|6a21BTxQs6J(eW?w50f~}H}w00>@|>AoWj?rE{))PIInKa`|b7tk7=63_Cj~hIK$5_#p`RMN0o`V^P;a?Do>IZc$|= z`;C~$TX-cZnRT`W`)V@&oZ`eo5Je-KlZf_ji?iE1>}x}VWHrwVUWhbfC~2dNsSf0C zm+#Df8!);VuL@R6^9X7KZXfooxH6dCBK!f+ry3uV!7q3|#gMfM8hAbsY$$PhI#glE z${duVq_KF>D$rvs_r8wjzRX5L@Ia6C+`;SN?y;ZN|M%wsbTX&e>8 z@JH`~%wr4pzdtyT*;yuEP$CImc;?=#q0Ho)yQeUZOde3OLl}#qSUtA$fL_H)GH=A| zhs(%KiHWLJ*N1uR_O7FqwIr+;8un?5@mwVEVm=j^#}yttr@j23 zFJBA6W)hVo?!OaT2R&|c`?I`t89xg9c^!dpiCpjF7mBjV921b9FpqD`F&>}aX5R@9 z3L(#mFKBa6u?znS4I15$#TT}PXRR;C$l=J*cr&_H2)SN-0hgHUC3{UR%Hqq|0yZXx zvW6sV4I%%8)Fs;&FT&z0+X8m}XlLyeW1-bzd|7-|TU4HJy%|M2n2-n&x&X1n*RVyU zjkmK7K(kFW06Q(>Yuf@2Hk75XMI%*b51}hO_s8KGt{+6^L$DuD8t1 zcyGV~N}!vS?{^nrd?L)V72>BUYSpY%Z0Nx*&yN1;Gd=<4Z2ww(7U2!3B*L+RI@T^NIc_>U~aAA zb#GR-tZ>L?r5E^Dr1_PSrkj;MgG)2I8J`L^OLNt%WZ=`#3sIJQBhWUDb+dAPUtYFj z27VOJfL$a`PluWnvN8wdD4|f7iYzd@q<6VWc*-F`X)L>fZyOw_%<}s-uf?+C9Mo#? zDtkBQXW5ng-Xa!zbeS6sS$2ZGYDN(jTRxnX!)5HGFyVW*R%F?AqWTPiZ>PYVo&(kQ zJKqcr*AC`x3h{kr(6s2{F@B5>1DmQjC5KZoT=6%JHV#&1d^rT$+(jbcG7&?gHdPE@ z`~(CSwdjb**!76YEIY;_qK(+-hj#$RhEK(~jdc7C{;^d+hE9=o5bZ3CN%_4+J3BdWsF6p1?lb<;Tiw1SZ= zyNsh;O;5%n`ltD@>^b@Oyw^ZT_%!HdYi&fwj)^S$`~1D0my>uC&~_8kpuvtI`9u2hm$rYBSsyp+-58lV7->uR{@aJu0Z~rrd{er^(N{wOU9iM{CqnuIV?N7PbU08+# zjUoJpi7Oi6L)+A0+hAD?v_gmemT&FQka;8+;ojx_*)H&XU`I61kUYr5C?+RbT2t2#0Bs&dm;N)LBF3F?ldr2O(4SON%45CAUi)iHEXKxa}`YFiNEMuPUkwe**SGhnM`aQMlZ6Z&4k1FzRpw zE*X9_7ZF@Hip1s%7WIjj2Tag{k7z>QqNI+voHSqf9o~>N_W$}X1bN`HBabWmsLYVwQ(6o=~3%mb9%RitQE{{G%|b5 z8wQtUJ(Zo4|5GM)4Ln?ql$5A4a{D%5$!1} z)GY|^k?0C>-D{qQy{P;h>^>+h5ji(|&9YRfIg9({C6#KH{dy8Q4QJAQF))xL&MKm_ zzbo(-fIT{fy83(kUPKpX?<2wwDAK52xM6oe_J)^-q46M$xE zEV^DY^s(@msRh^?pluqLFhidWAC0f79S6$RSai-}gsa~QUV^W?#6$ll@JlTb_m`4K z!ThHXAqhfmqHu}Cb8(aCh?G?3@y7NdZ*^7%sH(!U(UI-F!Gr(#DoK?1``{mtrUE4~ zN(@uHxc$cuBUula2ddz*yuHNlK`vfmU`-6D5W+Scf;mdWBoE->-)zq(x`4JVU_TNU z@%xJwJA~WE*XPXkGwj(qtnU8hL8iCz@vmwzyT=0WrqyV4(7Z`$=CLp+5jAkQtmBJ3 zsk0cJy0{VZSmL+%acx!r%2K$Da3qZXI~FZ@^x;p-Gmn)v`=_C-I#8;@vNY5!!~R-N zW7u!Fy?XU`~|Og$j>vx!(r~A#5^Y%b*zBQ`xl==sO<% z?^ecE!o3%htS7WAHlXDC#-f_OfRw#vcMcR|4y^o5zEDILe4; z7l;&v<1#v%&u#&?5BRPci=liFqNK@Zhs^$t;$sdf(XIS~!yDzh)_W6FuJT=HE#LK? zq>ipZ<-5_(Z+R>m0_9kW^J87U4XP8~W(%c>(*w z9rJt*8~Xz>{Y|kgrOYYEe7Q3p^R5xjf^iALSDE-o_~+PO8Vyud!2c%APD+Dz%!9v_WOkHQ2rgLwq9N3d8B(ei+lLUgI0(us zk?EL21{Y(iQ6{!@ux7+X{Ms=a4~b;9&an5;VNr1&g;FPys>)Wr6sM|?W8R10za5j* z$}zitk&kVMaz8~k9|?1J%+Sw5*hQe<6xQ`d-10o{d#V^KBV6-Jt(>m8k}Kc52^JWr zopU8VsuKrOd0fU5DnIR~v9kR`io$Uji48nDgd9_5B+6^k=*b%vBYV4?zY^ zi=I}aD2fH_Q_U$k?z0)AhhZ_RF9ajpMIzy9HY0r!mZs-GuuO|aij28tGeop%Hlt&0 zCUpEb#AhhLc+#rbjBZDANqisZFOAh~hK$NJoAJ7nkC5>2y66Pql2NJIjQOwXN?Z}B zs>W(IBf=Yt=V~@%1-_m~65ba7bkbTin=xuoK8c3{jWn@qHlu~jHCS2zw9Le=*^E2s zRhUTlZlD7ucFktwYh6~(W?TchWn$NC#()nK7mry;OOV@11N+VF>yg-Fa?3&Fu zGDvpvcp^|;h506`O=>pd?I{N%18)P~U2FdDY=%9cv=oklXsVk~&t`0!n@6T+HF&1G z#x)PvvMfcYYrEA;Q|D%GZcx9tF8y zkxnVhks8hda2dmJN7hQ9BItO3T!R&bxwJxjw8>+5c>F)LSsk#(nol%&1s)#oUJdpc zSP#v819kkBd3f{Te(XE2Nt%C6;p>#6q}AuP4TmtXy9{uH61aZ2#K^7zkJ@=DlubvF zqu}S2MkX>+Xj<^7;@?ydnm@o_XpKmy9K3zTgWK1U!{s0pT5ep%97>1$0qn%>?>eTl zSePplq9U>m-n#M<>~|lw5N2Cb2wEzUtB=@|hd;tsJ0g&C+m~SdiHrDUAF&U&*Ue01 zwh6FL(_!J7;(0f++6bw!vFuc<|t2EJdDm=oIRr(jXW6* z_c50Vcf)A9>k;g7)b9F42`p;iUKMZ|Ka0Xq%Y?^I__I`)8*5~`Ym@V#tgW*1D-_Oj z*LAp^#a(-Uh&c^D4eHqx;#U%DcRgUo%;I|3e^iVgAc1z*zgiWR?s^`AUr6+jxOUf% ztKz5kFW6t}2uydStTM+!q$li)PE{F?TM=Tvg_0J_gmJ4v95ga^(0Z9zCX8DX5?=Z* zZ$<`MIOD5nwM8rw#;tUT$u5=&<9-YYa5&NbCE-R0`8HCEWx}|5pQ5ezv1-0Sd@$U)}pAhV7^){<7OVQW@QaIjDr^ z8;`@2n5V}c7>*S5-HsSi8Yb6bET!X_Tv%C(HA*le<-r0UW=naKgsF#=5g&&#&y;pE z>*Luw20}mISkyR>dGp*jGX9Rg?=6aA0W2&fws9du(f#LmW-!Mt*thZd0Cz$INrb;* zAh;7I{^HwsHwf>9i@%JxRuhw1K}>vIZ2UjQ*H;1b`1(v(QI_5agCJbSps!FsYJ8pV z8-Lab)?SM6ktRM8{&gRJHWFxx#;r_zEWBOgP__(cgT~`cd^Y?|tx$FV=(NV45mw{t z4Xx_2sW99HexfDf{%U-k_i(h3_#ii7xJ2T)xQQBHSHzM-NuUY}TL%vqUswIW5T2I~ z-i$P}C=Ggi?N!&CeGYSP72Mi(V&ZG_v@o^=A#4*Mn5IPL_}W$|k$pjYE!bw_B7Qx- z&OJUKvmJ%~v<{0pA3T#hIvAPW%5N2`il4*<-VK`QAK)AlU!+#!>u>Rm=m;o_;4*sR z#_r?mczoleGElO@`uA7dGHLzN6x16m36*&tOG1rDa{=s%=H`;nxDHhqRgb4&uZfH3 z%_X6_ee1I};9*TM{=p@=UJ}}}2lJmWSJFr?2{pQiPDw-9-^;^Q>kzae<5L^+m(GG9hW5zq5pI4g=vlmQpX(Mv)lekkcu)qtXbR*59(C842S6|jE* zu!|PZ{mmtzZl_&8SR)`9MIiVj_2T565=gl zSUTnSHrNB=;(6;Op@_u%qPlR;X6OXtlK0n3LKBMzi0Z(LfyEFP;mswXO*rb3>JzUA z*iZ}flF(>;x|-@E?+DhFxX7Sh5?U~>yu86ku<<&)xg>O~VyL{qa=^7(pqGRi{D9u! ze%Oy{7IiE(jCnLL>^@&Q*dH*zz?Hj9fokFpzk2oY1mrUn|Mk!9EW$ii820|hqp{@! z9$NyJ(9<9PN3RxrTTXoyVl02P2yZ=BS8LN|@Q!Ps}GZX>HCYEe*h3{&PMls)c1fY4gB}tN{#7 za2ZjQGBuyH&l!x3sxWuSf$IC!NyhG%p)ikAh-cC5)O^y&rr3kI5Nw0ylpOc@q+uH{ z4S5iP3+^J3a5bNF{+AL62EkJ;DolyY@0w2%(W?2R)f^9tpKgl~jy zo7SrNr2hR1OZ+oXwuxQyNrCu42=(>(BcP`ycFiZnUW;NP;r^}A6UHU89O0TzDm$Q( zoKK1aDsN)fe9~8KYA||mydltgCU(syMQp7saTlO3P3)RansT9_OxjqW847a)PfE`x zEyBDF8TdNz-CFa1=aagXuPB8VA^OctsOOV*ZHt!ac@6G`x-K&7+WI1Q$8`DAmE`=u zC38qz^GWSIJXv{qLdoFul}7B2q3y^HN^e3rZ>#}g!`YILOqZG;LN)KdHk|c9EP9G| z3qDHtJq$}I>Y7%Us#hVD@$E40&VlOt-F_d9_&J!bE5sAcplQ*Lr&aL#4`%zo6)UIY zxVu!#0@&UX0YM3Okx005shWOXcvA?1`dU=a>QW+F zi6;O}(^$EbjLPLwcb*my621evJzA?=%2Cln;!8l+HC8Uwq?*O0YW&26gnN95O#rxL zs+3Ey^a2tW0V-)?mrFI6>w0A-8K}OAT`sk5cPSSg?JvhDqXqky!E;YlEPvYG`2Tbg8spRu@rAu7{aw^Q5ih92QBLnBCnnAck z=0!!2wLxt-tDm=m6qbUhqMJ~=RD;D8WqKNcw|3XKrf7dXS5iv81pYdQ#N|?FGms{F zLNmbUDh;_*?a?%rq4b{T)h_$9683#dZ1zi#tmN5sJlMXaw%Ac9FFT28bvU>Y!~44! zAFnZz=Ga+%`@StU@6K&dr!&`}H^Eo(4GVqPzD!%x3PfTr=-&8+a2g)S?m&h&`vBS^^b zf8i|A5c3!Gtc;Xam@~1Z)P7rwVf%e*_db{d;l)FkSIXXsDa<=%lN~$$Q~KcWB}dBY zWN+q^atRYixl*>SPhh?&zPY@aUrOzA<(PlUi}Z#pAm!$yP!^antY16}N@+W-9t%#X zR<0-uNmXh@YEo}QR^hQ|}g>qiAdFL%!n|%fKU|h!K!FVQ_ z{L{oV=DcnT@IoPE@^3v1=DcC^r=^KZ?t8a1bKbQ1(l2->_up5YIe)hm-hegcOrG$q zojIL0M^(hd#(MT;&U-fRSp1R62OaQX&il4Nn*55Q95^4?JU^Zn$o>I-Mfd)l?tR_( z`ty9u`OxM$@1Nq#7x@!*)itDt5lxTFxz)DnQ35LqK@y36p&GiMs+`0-Z4>_qV@-fQ z)Oh0{!c{6G^j_PG2SKa{&_IRdE$-T=OYS_FuYPzT_7~!t20mYDxQN1~Ii2f!bW};R z4Sb*0$W%BMA)52{>a~+uHqceV&nSH{)Gj*j`BWX^$GW3-{sZz(QJIC2ymN2nd}-ei zQe8Bix!YpvGcFNGo9OaSFFvpz`Tc$cmz}& zEJ{3~F5I@cN;!5O?1tuI?lbb;-ZISDo!b}VBb$_j7vOJ6BT8tT=Qg~V!t{^61^gOAILoA>AbwU5S&snm*KRf$sq*te2@#RIAhugJQhS~ds!Fee|t!%9?6 z7JK~rf(;@rZXOw7k6_M$JmgY4UD zP7`C7n?y6`JZ^7#!B-@HGI$1QMA2mMUTDvpi+ItFiSdj=YyittaUk)6FXcrG-bxbu zXRvG?KD6n?8qB$ZM^qh}DC+EEAbxfV-8p<+N#92EX!p#w_OEIc>>FZL=d5#y$Sd+vk#1XJlDh`y3 z;OBXZt?uIBaiozo z_Hj}WT5|5w_;oF@d|VHxp^57+0>+D3%N=r0a|fUv2 z-BWR(XaxV0M@RpXD!8p9UVOqO^CoZqnA>W6U7Y2Ixg;TR`ziD3_M@lQX3i(vo;gCc zB+20Qb?^*c6~Cg+r@T~^Y2J)Nv;+HG#ev%k{*0Gee#s&DFtE`&yuAHOZu{w45c?kH z6)GsF_rO`?sRgJQueiPHU_X)jz2Jv+@C?2NKi|$byx7L@+TzvnU%~FGIBgbHlcGK`M)axi}$_d-`Q|wXT#W+iGKz2r&65qJb>Nc2XydApMHtVS-`L#EM8KS%{8D~CLW6OW}U?hdz@!Ef&V4+xk_6GFJIP! zIirp6HRl{Ga}jnbfE6ZADL~N)Ufc+8;V*d-Sh5ZuSzj(3MH+pGBydh=6; zk2h?wdE;3&*sq!gp9GJ5e^DLgEN$2~KC3Bm{|cOSHgn&uOaOD1GvYgz#WrN5C=#q7 zaaZmIFK@&r_A4uRB3MlwKJv3?h`xdm{_3$8^QsiWS_6Gz;)Js0@hXmyGy%i@3(N z;;y?3Uf&3LfGj81qU6qTVk-DK>ak9mA0v2JH5RQ z8wYa+A&Qyr9Zap=v?}NozGv9;WTuGJtq0$#gJ9g z3nT2IXH~%;fj!mXBhT(B&YUfcJlO?otnPRC8SIMrKU~(vu!~_^i5^r*m@5zxsq6g{ zrS82G_z~`E*v}tK6{$-HZ>EE1@N5jgoIQ+uzih6`C`31~-YO2HPVk;azO`L)3qApC znhq}uXNY0z@u(nM0rO@RR4!UcKFqK^+gzL-0XwHTKS0Ubjh7;v!wvfngDQ&TJp_M3 z8du-y%Q}V7aWtYI|A0Aagz)W#P2jlHqZNFV5&fwLX4SxB!O9VrZTYmCWtnrdk>_IN zRQ3oXsWhObCf;!Xz&eSm_FR*cU)OP5jA=D$F^~uwQK!BN8_TXp+VBC zD0;!C8Sy8>eFV36$J58<%G-4ej-AsDzs}oY*@LvatT<4d#;wF?W2WI-q)Hr1hB=*( zNZrs&l)BRxH#=t;c3wbs$vc8~)xk4(NRK4uoNJUyE1tqA#7MC5Dh{Mh@Oeh5xA>+O z@gKm}=|78tfC3sc!%m=Eiqku7dHFo|H;tF8AJ<#H3~jt&`F#*U&`Vg&SW7su#l z(L)Ln485ySee8kr(%zHWG75~J9ioOPw_oH zdJav%Tj&TQEAEcM(AV&<`JES|ZP7h|`f4m=KV{h3JT1eh};3|_M@ zesyjb0q>qyV`TE{i48xvgb0s8@S8@!ZG0`8cpO-H;?e>B!1w8#zZ!N&^I(zphCuIW zEN}n2fn8mNStpo#5fZm=MQ=n#9;iZdB}VXPa{?Gy#(_`C5&VH+OD*1r{Q&bu9lQkX zRf?Rm5ySmI4F6RHiZikt1OGWk@F#{Xyg~@O4f9_*_)HP}Q@rx){Kp8Lx;KH5#qlL3 z!EnjE$;7`jZ2#_V#EQaPnvh6*7cuC38i2&VF#=Bi8pp^|AG}eH;D*h%;Yvx?9_H>k z_|pIyv+np6?(1O-@Y$8f$TA9iLXO}Ln{8vMdTcSwYjtq(3PWUIh>baeZ1#lq@~!7X z;3stO4F1nVjOIgaL8~GviuWw;fIU=k;K2$WW(&%k=P$TdFSLHRTn~27&zL!iumvQ~ zsw}*u7*LFf8_lBLhRwcUjO>Ef0cxzUYdRy}YhU~-+k$Fg{Q4A)c zVPyEpbo_$Z?7xjJFDlU#h-NAwi!3`oKXZOy3&@@p!svvL4M3SD7V~W#Y`L~wEWu`t z!tWo@1rsNydZLN8*~{Gw6_Ni5^ipA&(x``94K&0yyWgYS%)d8={|~qWb4^`V6~cO2%!fc6xw95p^(i)=3*#e2`|$~Q2c>c0sWzWp zYi$A#1{$L<_ogU9wjV-&XSmJrH}3b@Ks12P8Mb1%>)Wvd6kC8cLCxuZ?6Z)d+{`)K z7U0Y$f8K28h|v^<%$fhwIDJCW+nlWr7dd+m|-+TAZ(qIcR!rt}z-f`{eoxb#9ui zB)gXpb$(k#d{@h#ZNY7R#DzKKhNW^doNKaa#0E zVs#P)1Xp0oEBdpyBl_L7>8bVgP^w*^)IeP9*K1*P49lG!-kji zWK$uSB}FY$b(&5hj}TQEUy@xdA|i-@S7CP zPt-9@d6l0;&CgY46YSDK}rqrUC zDpsn=KAPge(jaImMfGA-Ob={Z$K+%8_T>uu25XMvP~N;NH2rajGLnYq)NIVO=C8VVc^7u|@D$ejOwIkWRBZW4u2xhG&B0?k3hqMQvJEpXd{qr1A z?hkxfL=N;IbCB!-B#$k!;G2Hz-tfScBjqnQWL_i-?vMAsak+}NX=D()H^ld^$JNRyEiEJuB8E@r9>p26iG`+KGw3Bwv}VpylZV|)wcU@Z&y07(=)rQzAlBI^RKncYP;ua184OmdTswx+zVF*g zOy|1}+*^4pidBb`*1=`8C!!DBi_EUZTEqN_M*6_L<~T~U56nXp;v!Og;NJN8Nm%~_ zoT~)pfqOmr;^eK3u^#Vw^H<9=P{jAy?#eAxI;UOuRa9Z{eeo za*?|OSQnG41NX{2sv!9YuyH0=2kz}@D|@|s8Q5x*s{{A)#Yf4^9|Sv2TvS7S;9kA! z-YgK$ncoGI^-LeQmp-SMWH!`q7gGoBEhwE|vI3f+kt~%GB2T5X;Xlkl5hwlPW5Y`c z*y_3yWAFGfMs9Qg)afU6FLB|k?GVZ5xQr#sP!v$?Y$Yt;0*xZP-o)7II{l*p zYy>O|fqu}T&8@BtUZt`D;6H&K(%jtY`f*Eiqrh*1-61ZbH@CX3`ohMB&>g?VCKz0j z>#eSL8dYa`VJ@VR-s(CkApz&#!Cox~!&cX$FL0(pGuYcGmSd}{-wUjS^1h%$wPGtJ zLvMBc@lkmwW`Zu(iY%c}TU{@1M>D{8fF4i^>7RP5>*tdTvAqbxuS0N8iE?apt%TQ% zgot~5gRVa=Q3hNjM{jj)a4^ZGDg{MFtrAJnTV40zm1CX;*g^{^YICdW9H1bzwDUxzohx)v;hEs7+V3OGv(^j6nfrLom@9qd~)gHui{;sIZz3Dtz8 zCpbnq>!x>2%wtt~b}5Nv-HszC#n#bH*bb2Ox9{9-wXydR!JktUqo^$l7wDCZ$6`~6 z?+{G=;Sz|ayoWxAE9`6EDn+p=Dh{xU7SMzlOA`?o+_x$k4c-W_wHDBTgvAPh2e*A$ z0kcfJC+Hxpm_TU|buG)2J3gz>fK8*uo@YR@$X(?x;_>3PTlhW#-v+o(3ucREnU>MB zyt$*`(AqMZs}S7HA$pTLf@S%0yK}sUm=NTAD7t~Tj0F@)e%tu`EbD>jhjRzn#ZOgM zDEDtX##;z^d9tbyoc<@bvt>PWKf|L6kwWrz;Y^mGqVWN#j%*6BF* zJ$7)=xm8&uc%cW_3rISifqxd|2ypaL8q2!txUK_~;(3>~%P(~NFvMpm06kTirqM?4 z4t-cwX>PAL6@7|(pfBB`E?EU`FIfmDGX;K&S$|xjI7CEzwI~?cDw_3&;}&ku zz39cDtVHhHbd({~JOx40RWMea-VDmBiH#fnk&<@;>!!IE?IFvm%k90&%l;f61vbIt zv`Z|j0r&f&k*6qXz8q|==Kl0Wm8{0xz5%PBd_U}`6q9)uuZCnb=k|reaqh?OfcKSv zdA--%pJjc>eed7pGI1WmFzbs;=9zLY2arzEcQnU4`n3U@xd~+PRmYZ+tgYOBGNi0*+q**4 zUkiorSEsFSS=)GUr^`XC^=5d_6bNQ25%Z$HRMu|plQua-_LVk*Z8Q15Q64PoC!TBk zOPd@&odLV3x&NhErCHW~ZvWbjy=#BK{(_kB645-h63xmX?&~|L4D&;a7J*B0(Vdn2 zF!!CjBAUff)2g}X$)XLWF*|e6LrO>)86)<)PD9J54pEO;}S==y)55KcfJ%NvpPy z?=2A~@i#!jHCFpwWmK;Ht{vvq6cWA?y7gMC_PgdkTU+8IK&LcT`&~EX#!yEcNl;{1 zAtB-aK=(>()qdBLb@ECaG7^(uxMZRuT>D)s&GeA9xExRw6T9}i9>w9pBH>MeTAJ9k z-_=>Jw5+~8f%=(P9ZAr_SM06flYnNJ*tOrab*)&5*8y!Yv1`BUjAAYaI|+12VJ?23 z)sY0oj9wcoY#!~p5@Bf!Tg4Nb^Mm!b?5+2usb8tVN`Ys`+3$Idq4SfblzWAvuvc<54)gFx39RkFR>|{cD`KnU=MQUPt7OBKe%LCx z_MIoTN*euvN11zzWzZ`DCmIwo0Zw31i_Yap%)m?v$Skrn88Y{f!+gPfF|u(JV5h zXw4LsH)X)0G!~WevQ09}m-6C#ZI(af3{q7f#SbYgm~s-K3ZAdDak;_^j^S0=lLgsFc$~a6SViLEdF$a-*6;D6 zy70ze&2@M^yjs<+n5YiC2UuSn-W*=#T@Wbh6Q2y2p#^$)wX0fgd4mmLnZ!i~_3*0e zrcil2fEfeBeV5dJ` zfjT+G>=L%30{PD_VK2m@yM+6X=VA+C_<^EsPhnJ-a7BwojAy~THwUWk_hcJhOv;A& zSA}>58o;S8;ny8;3dS=quSu?0IVH!vOSmN(Z%O8bAjVxJ60W+01GdC5UK4^eE$S{Z z=6b78M60@lemCk09sd&I{uH1OX;qi-P>sS8PX@}+Sak_yRIV;zr{mZl2MOPafA(pu z>JqN)ilu$RSMblT8mlg0Ni3!4F5z!?qlAQePsY2xxMZqSm$1P4Xo;hNVomJo5|*kV ze}Z{kpfnS^x`ciY3o?=L&w#!#v8zk?`^ZGuB^(Yk#>B2JVdeGlj2;|c479?;t}fx0 zs(B>d3v}4Tt}fwmE|;eHO`rz~bE@(B6{vBu%E)CT4^+)iTq5)T(Ix!l!=h3cM|Idu zsJn#4zKD?NX$t<4yT;We?7J#mO8Ni~$suud2`?5%XSCSGXMxXG8W#TTqY^B;NAQ*V zF^vBR^Fdr5wb<6)crlVZgQwD3HJpz6iOfOyOlW)zZjSGYJ(-SFyH;>kkxYFu>E?ud zj7^0n;4-!n(JMG@-fzV6!dyrry@JzkLIYL_=2{AI5vg9mIe)tz-hTr8SP9G(oS*yP zmEYd54_8bC&ar|MI0y#^%!XpAR%KD@^a@V#KdBf14{E_)Ay6wgC+edUbp`N_63A9b zui$)K*`FOo5YDFI)qh+f|6JTgui#AGo-RejASk0n;#PVEXZ!48j0%7^fS?J9PEu0z z3eMBZCD}RRUx4*Axmv-Qi!T$3@MFOyn_R8n9K{Ktw3N zPf@Sn3_gnK_}^gvLorU7Hdk<-{FN>v_eLG@!{ti6TEVe>;wgQ!I9Qy?)e4TiYlP$} zVCg1TD>&^hCd>Fcfps&vTEVG{j=#wKD6k2{MF#W=&PPYBBrUa&ww0(vH17^X#LNO6I zhm!;aC*u7-D2{4XQA(Y5l7j2;g5V9n`&v*+2$YljS_DqQGEo0o6lbe=&&0w%F2Oiu z+H{h~-Ce&Wu@ICakxaaDlAhQCC|*Tx2=<=Im6Mcj#wG6x*30C|N$L-iuek8>U{g)5 zoFrybs?7Wvuua582DFn5shcid{o$v;WIfYPl4K8-?7qpAlZ?lgIq8)m?uEK9!@)^v zCb^b5y9g&CK|<2d7~v#Ib4@2nrZQ4aGAJlNYk*jq;4;#QXeU|KK9zNVxr;{HNyg@j zWJ6&drw|vBYA4y$7wK9AxLgTLCt>~21Kb7sNyS9q98Qv!3qPf|p!id(no;VslYE~a zv&|l}vHpikq@1@F0_7y#o>xNs2P~rmIh>?HoEQ59L3jfQnkx|(x6w{={JWx3)B}Qn zS|o0zo#f;CzKjZhPlsR*i8@eHw3AfZnZUXb-vaid$(57L>lP-%p9lNJ(}eZ$eTjYl0-7`%1OpJ z3ri>X60E<;m6Pn))JXCvU^7jwoaA|;TutX2!M2%PImz1n4w?BgU>AvtYN(wgl*Nem z!1&)_vYu%t$$h$>WC5t(E~cC$W)@ol@Q!bwPwkhs_oPBMPJ z=_FrL87U{}aHb~ffmr(DGQK9FouvHR8f+}glQq&#axI}ATL$w6g}8`RJ4x$>_>DgV zctQzGC#hTmW1Zh%f1;QOoWn^fCt<{7pNFA8E|D-kic+VYq)ua{VCqnnW`3%1KUnCrBsA0GnrW5jclGHQR%wwpLKI*Q!iPo%W~SM&b~x0f66W z!A>Di{^UQf63)K{T&4s${HX_CIoXFGd=CW2l!%MlXnzU~E-6L7L-0t8#I3YHeY4(+ zQ2}trLi`BgGLBGEv_Iv^N@b^rmjo+qa^+9$W~a;WsbGywuKa00ANh7M?+n&MbL~$( zKEbYt5wMRVCZ3}9rzHn5%v%Ke8pSwe+Vm&?3$C8iAqY;8NG4wS)5l*i=`VM{9-3VF z)6N%lCHF!d^Tp)~ul%X|pK@n4j|Pi1x$>vx`2u9->wz^SE;69~sWuL0pjTk|=U}p) zX@BzQhPNGveP=S|Pfb_Xm29DAaFS(LTuyRRI0*$#_-6z@i;PV#6_>+F{&UmIIqVAM zcM#7*T*lVLC_lC7W%WsXhLbNw)^Q2M#gQF))61#>_%dZ-z&I@s=fvtwF9-LaX4D33 zqy_GqUXG8(92xHb+EXjCC=GhkOY}{H?S+N&aZqHqtMsOqNg?=^TLZXF3l38>V#^n8 zdI>5PFQYjF!POihwdrNYOJDI~D}M&Tza%7@zQs7c4i;h&5g0bHA28tVYq^wRw& zhtTnz5Fe%h^i*Y<%uO%3b``g;t~R}F#|A?xU_K9Q zvF3Wy%d75ap|``nUon|?z3JtT&yn2AfH#!D-1KtcrOYaS3FcU)@=UpxpVZNtUYtYY zWHyRGRo-1GKCUD8v|!WA?u!^mr$X_*R>>oQ)TWoDI3(x`z`j}_V^Eu3mNzSb=MR{n z1@r>8Yd;G%ySxJrQO zO)oV*L93e^_QHy}idSuV$vlKF*;fIq?deh6^+fm|8 zaXu1)@k(TFdg+3rH^j*L2e36JSDRkiAM=)*UJiks&|GhNDLMgz#NS~5gP8CV_f0Rq zUrA>Mnod7llDltuY1%xH6$XpbTy1)xwt_k6AtmI&JGwN_`oSu*qjH{pr&LF!*3mTf zIb!HevD_e{J1R{(7{5)a(;XFm4;)Qd7%)x?9teTzsI+R1Ii}ixjg%lq zN2TtGc=i}U_~#JxRw8*D-BCGvxTq9OfMAvuiCgK8%9@Au85IEE1i^L^J*TATj>^;0 zQS1%zvtXA@t~x4@{*030pMbqEx$3A?o9#MmHh3j=0^ySJ>yFCg=lRe=z+Qovc#68C z(j*B5*#P$DigC)c*-;r(FHSajJs{{yBAIyAQF%SwwYzH)*bI}aj>?`z@=zeY9&D@0 zRYzs{^IS6i(_j}&t~x5$8dR2Zt7B zP)wZDz!yNUj6}~V64gWufC0Ag+-kfUBc2PzOLqWt4X*q2nDO?nwdYsmeQ;9hFmo zcy^;eXSzkn-%yUtj>@I&cvB)i>ZHoXfEyK{%QbyW7Aun9H4jUW$HFtelbd{bG; zJ=bCq9G5I*-BEe)Izf6>F|ZhutBy*4eDazKnAZhM(_D8{)&*C?_kUpTs+i2X?x^fq zg`e7CfTNYb?5MN}@so*L47N#ePPvz#UD6$uHuylaFrR?xs=HEr1xt5SF5fDT@83Z2 zMyteUwDj)W?Ko(UhpfY=-*L(MAY)J+m37(pR9OYU8d^Z-jhG#k_4w+hNK9)8KGPyH znH?2JW=Sa;48cS;#K(h5Lr_tP%#O+|?k_tkjlr6mTy<2& z^z)S+l^$SyHP;=L-n5uA4)zRU!b{vcDnIn8$X0`8YVO`qnfHUij)P@ut~x5>JSzu1 zq=bOMSP;fwvE$bN)dnAEfz~Tz0~$eGhC}Sa6&r+K*|xU9RC+wj2@3I6N!B*_n=6zq?Ae z!KLP-_j4KWcP+?E(dag~O71`z&07c@8(mrDQl#489u>-n(`tAT2uhNu5JfW1!}W>; zok<$V9Yl01{BxH`A>}P$ZKuL;`rrR(h2JemkRirEGu=JJ6z+OSVhWnIK>e1)O|Mv? z>i|us~ui$ZHH%HFD!KY4~Sn-0D7`ARc1T<8YhA9fK8b7#3k}76uda)+H8j- z_TnpVl_06-E{h|X*bR+#c#T^-`~liQQuA*3r?(1bw!vs5SY!Vda_o4dCU0I$=N4LX$ZeY@DAQU6CN<^jG;jmR`sOJLy zpan7p)egtjz?A$>z{6TVCoh@p@U0VtWnzAX;Gq^#`pkCtRFyhXKDgv#7nu-B z-45qpha!xHqMTBh?eNu4@S9K{_GXH?idVJ6M?S!As?LA|+ywGyYSj+Uj*v&i@Tm}e zuZ6<@bvqomyQn-NcpC(}l*nv{C$vbG?eImgYbICiaMJ7YvK@X7_C|Bv4p-|_7409I z$NadYm$yC>->&jawfjH*uXecCXuOgz z7jZ12coL}csJ?K|2(+S^Fz;1}zuJoapq{i#dk){e2>6>87%lMBbvs<_5?+dU0cgu~ zMb6#Z;cIU&vgDDV#k8UZr9rpD$1t&47Z%QILXqaK((Q1qFVLR11?;K?X%vlaha0X> zkpeLY$T(fU6BYs{^1f?7*%}((y0x&({=y zo~pcq*#-}OiibN5bO|XapF+V~QjX0wc<+Av7XJvzF?ZR=qF`K;b{E~+V831gLd_o| z$a59UY=fJRFC%%-b}ZB2lBKNMU|%o!30htjtg^{f8{E7FP7*@_^QK@eHP>yhZ%Yi} zy21XHVlwZ#4eq$MEaMXarz?Tk7w(%;NG5I_*lxu+nwA z5Oo_|}iKB3#-o$00HT=}fgncnF;U)hcUFRJbMe(=sncKVMF1>eZAk>5a z3B80KdX0d9fPi!m>7ZbwNpAwuq$nU&dJ$9zRf^I@P>NEMgHSC9@;=|0omq14&p#$* zKF`#hwzE_6w!u4Eq_ACJS(@iP~m`dB!`3;lNoz<0FZ+c$CKmTrS{51^vI z4>(B+^0vXhxM&x~XP|4fqBq4sx4|jlmDoVoaDE7iAM>ho8=TP}(^j_tA8Nr+@ozzpHAx70SJ?Jb zE;#-5A8qiwGY;uuE;OIzb)nked%8)b`G|@xq2q@j{?WsM*#^JV4$ub2ln)m={tROF zjY_JBlhp>lH;3lGL5q=s(kWuiGbzPZ8{FqhY-FtgNwd7NNg`v+Hu&vTnE2Pbg>M8l z6>2^dPDU$dRvSDuJxKC-V2d@^ZSaIMv2s3Q2iRVVt2TJavxXw8`30~mn(H<=bQ^~5 zk6`~oG5I`n8{9e{-l#*q#r!9Xe8hSj{B`H@GIHg>YAenu^}Gw6N2S}~O*INjWk;y` zm#r$)?OMSiEWZ1ApzfE=m#SoyRZ^UJr3U25)U*5iYiKFwZWx7#SB#o*mqFO z%wE+7e|ZhB8G`^b^9W@2sWy0zpSjO+5k#xBP}F_h2G_(q4V@;+zk}eg5?O6G;~Ur?BPQyG7(fJk@|nT*jle|@(~#>|0q^A% z*c{Pqrqt&!#y>D3X&#?3Z5L~@f^fpi zKu|@AxClmn#&lJxDn)G|=&VH|C^0VL?{mKay?hyE03Qj#SP~^rRP<-ed;V#xH1UOC z%Pg)wV^-vKrTaZ#-&tIJ#+-c8Tw?e$*fq`dXUsbr@U`U`?EfJqo}&JYdDY#7@!%s^ z`UWEyr$k$yF~7oL6yklm0tD4aBqOgrW4>88NmkZ3z&cu7ea37V?<@H*u=g#lK4bQ| zZ(dwAA8d)m)o09exn*SHcY^IBE)t+WWB&0ex-C3oehEyLGyNH}|1wMp6MJeg^%?Wx zMt{kIQ1+!ee8wz~f5al+B=H%O9N*#Z*RIO$Ci&8l%5H-$5^CYh?qS5vnikG3eUDA* z^aslza1M8Kz6CFO!`Q?lPn&@KR)R&}D8)v>eu8FnfHdwJhr8^ahVcN$F)adHr+L%g z5Hnf}!NRT$4`JDTjiReB;87w{$06btkNfTy@R5cyqGgAT!Gaed>k1aFbyoG`8)>j(C}_JcSVk%{~91Dp90*z8w9{3uNlT(ys?=qZ;jVmx?7T{tM zSou7tU%D5|-faZaGmbI_RhW$DyT^#6<6(N42k1iS5L*-Vb(!I7J*sS_}7aJbLk^8f-K8b+Fr- z$B42}fam||NI^!Cc?IM^c1Z8V9ZuPWcyMpD_f*CU0&$Vz0LqQ*lHBMsyCgFzz`i~? z6WL`YZ9lJY2~A0F%y-ikf%ZzG9tK%XbI^@sQQVnbfs1w{{)~DYjfrIAT&Zzb$Zaf! z>-FS=Snp1~J0FKre%NfAkk#hKvU@ey0VqytRTIi3Q&5K++v|t1-vIAxK`Roh61U)H z*X6z?wiFS|<-O5uguuwKgfFJ39``*vH$aF=Lr_tRgimqR7sasbw%pk^oUl%?yDN8Ai3nyrVcp+CX`8T~We9irYz<-`!FrBDqCwyz{=#r7PKefXXY+v=0YR`) ziCA2*LxouOSKN7NCudo(KBW+U>5IAWxUauOrQOb*Q!6B}-@xwY^Yv_kH^-Vb!d!kry(-PRWIcLG*JAuFx{-Lf#f|7CZv`6e_fD@K?90@;7i z3EN(%Aj|GOEj-8@Usw9r^6$PDEtd3G18P867f=>Z_n7_@Bw8k{PI4|S0v%r>^F;Y;nn$Ht(JZIH-nMq)Sy@t+!l7KHe* z?CM4YJ@d`RAX*8YUCSss2Nyr@HU_m@kAqh08_{%w^f6=5gJ$S(ni#Rc=M8q+81zeo z!LnNzag>jDjiW}}^?1zfjKZPbI9rh$H+feKor1C(@UXWK(?h4xTG1b@9oc86^MJEm z8?ba(w}!EG6A$Sc^0KbrjkzPa5PH_$fCH6)&E$9c=0`c_!9NU1V|0NJw*}c^z-PlO zAYE^&YZb}`2D9vzJeZD6cJAk05af@OF>6H9-6W&%RP#!fp~Yr4WZ8qcD`H+abA1n+ z@wV?9VB;@7x^dcP@)%Yx$t9Iq>2kTfXHjJ5P+Y7uCLc2*(I8{|cj0Snt`|j{{=z~w za{ae$K?NL}=ght?y3#ghRogvc@FaNgTljHAG|EUIRrW2T?a~IU8@w9;V-w34#brhk z-!@wQQks1PG+kqHS5YxB#JiJNbdf_WuUiAmg+od1CT{td_IzS6yr@!YyQ#+P$JdyAWOH=5?7H#IkUerf@tv3>zvB5nY=&lSI<6R zgtx~+F}h_0Wx!|;Pv*faCeO;z>&Hxfxu+8~+Q`>pZ!+7|2$N{GAUo;4fjRh^j;`v< zK4^s0eqc@r;52U1-GX_hbexXM<~?4bi0fWJKPp6fBwo`0Pb4A`#1cFe(OP z|3*v?8}WTIu&#q9-fThMq^k){J*}JW_|>S$vX2_R!3{AlNrH|Fnt9?j!g2QrhGNHM zcI-N@vLlW__DRF2)(ZDO!Py3K+L7J`M@Qh32t9u}UcN?`Rc4(#8q4A010x?Hj2`{VrL*i@BajkU z+-&zA%62aq!`l*7aqp9>KBkZm`}O1t<%ud{Wcj!4#)TA?9CWvelS)_y-p$PSFtYzP zy*#@OXOGC~FVqX{pd82`_JvOzvqA0;BiUYJ3B>NEIf`BnVTFK-Da<~#59`_wmtDn( z)jtLI^XpJ11Y$yHAG~j)6+MxXNn_I?-5AcbR( znUx{En=xJ(ibSodL9x>_EA!T38A=_%W?E2J2-M8V*ZZ*xr90q2C6HMoCrxmadX3%H zSR**$lOgy-iEzv&g3&W8eFim@qD>I&(IOF)o>>W*;m;@o_&ErECQ%FWM9-|aHkD*= z5Pt&px5d@WO1@c1(tYSpSpNoNx>qwR)!#BVSC$8>tht_9>6C%F@6t z7Bf(NVLwbUPKmZ=R^DzcFE`_}Aec`g8F@9c68M{W5XWY)ofcO!E7R(lGb^XTE?Qj8 ztXy85E@y5Yg8gA}HM6pQPM}PD5Xxu-j7)%@St)Z|3k zW>yyc!6oaX8G2S~7{$y=hL~BQ$&|_+CP#}(viyowNvNXh>Bo1bHe;LM%XS!BPa?XK z9G+L3eGls&HPV%&w^uoK6V|y3ap9@1BxlZ`?mB+P`d=97ZR2a-VtGY%?pJb!7vqIs zUsf^U*rSpRy$W>$C|YS%e~O*1B*9au@d6yI1%rh^RgxmlYoYuDexd{(l_Uh0RF8xc zz6pXoO2kDlx{|y%DO8HiL2y-zL{Petd^Nlr6J5$*5d2G`u@n_uNoJO5z$Q>gQJ3*0 z494WDl2kogK)SC4R>R_|l8k(>s4Qpgz~0haSCRr{5*Qx>`%%QiQ`D7YOBE;MvtYki zF;0oLD#?<7cz5a z9Az*T#`IrRl3P(#Wa3l7suC9o(3ND>_=@7TGTs(UmNQ*Re*Ck7WP>fHDoOW?L6XhX z43&g>Hc?657nOvrK|5ePpI91WsT_YX_)Pj0_t}dHN8X!2PN1=W4L)A;3-IR(j7@B9 zzRd*;F~+PwL3{-J7mD%O)ZLkaP-F5Nm?I7O1w$Vg;g~NVfm_TM@Xd@it8)GpB)yeP7V200&a{FoCpt1Lp$hwAI2I0I&?nSWA)HO( zr4;5X{*p(soVdu=DfQIQ%r11>pxN`9u2+*{ET=%w{(uBoIxy%Hx+~D!culvsx{Kw+ z2els|FGr=Jp3oV{Mi-0>cd4C5FF8e{Zv@n4U(lorhI&F*Olz=l5Xq~2^TW!b$!dMw zYO=)DOsZc`+vCl;!c|XlIhD$?Zn8d0^k?tG`Xh~WleK+AMYaIeD;1LWJ?SQ^Fy?>x zZoqGq!0Olg|BH1%=V5%hL5Vw@6fHCeg1a#wsY=>@?663NJ`CTrU{yxCBb^fA~B zi>oHdQCPT)+-d^+b!<28#9t|`!Lu6&Be!H z=|0kKEG!zsE`r_A{A=NUIRzbKH+J<;V1Iz|8)ne5CRXqEN^h@Z}%kuQ#ZH` zum`?=gWEv6zQL^>vNrtmsG_Xk-J$jnx}z+P{<}NOZnTaIX79q{Lkg;DCeljZTevt1 zV@D8jHgiU8MbUW#~Leo?0<{0;J5Wjn!x3hB@ zToy`a!PtU0=_V)*&p!^`A9IO73U00~x|p4}tuAIJt@}`2Oup&G*daJQPRLohIgog_8+j25?Ec#wKh0rydvxyC?*_xbTRYC zU|~lmD0*p?Si7S~#gG5S3dk{lleJ(qML$4wF$+)oGQJFOof3F-F|i|?vJG&;4?*yw z63Hb(x{GnGtRO}AAb6@pA}HO(l=`zEqYU5yw{Q|DjBP7LMRzf!LmIN(#FM~MEUvnk zJO4J3?wf(NvAF7D>X)o2pY;H+A)4zhX7e_zftUjOS;WLs)LqPmlQ?i_HSD)2#wpQO z7xVl(^VRMI1ZPMjBd@xc0Nfoex|kfWM;2FIOwZn!4W};17iBCM#&oZ`nA#)FaaAd> zvKCie%*gtoGVzVUS`rrt&|S>DMv078b@BdSvYhEraV>mv6>PG_R2MVTQ9!a)nxTsc zXk;EwSGqCnEhfi_-(QPo$(g>+U+$v8rhhX1j5Yg8v0vcuHyB&;r^ql#0z;nVq_988 z%R46F;!E`a;V{AP(b?pl%vh zMc8ugw$!TBS}Zfox#y(3^kpN#B&1`da#u#N%$5j z)F#zML5)C87J_*_YIFZ6z%qL}Uu4RgtSST7%`2Jd=sK+=%j_48o6ne$6w!528q569 z`Oo==%;*K_0FsMmF=AS^QY>>uoXA>Z#0FbFmN{2U07e-jy4uUI%umFB1&t9jJD54o zoE*%chqSs_3APJfWhw7u-Yw?G0FUR^W~UV}I|GJ`K~pWdKz_^> zV~P9ad=4#Y%C5lX9q>VUH3=e|RO9&=JDUJBOXC}&B+$~p`xShKVb+R#_!4+qUP+_# z#hbAE)r_hRd=@(fc204b?o!phF2nBE^f}w0j<_8<2lx?5?uY<(A1co7*AD1-zJU19 zGW7x;KxD%<^0C@Th%Xwiuge(Zj_|AiO%8D&U4kj54r+aQ(QZ|r=6 zPl43Kbpz?GX~L{pDeQh5=SD1c{s6Wg6QP{v>u42m!tEK1-G9s3>R3Lu1Za&X{-!h8H|2&^OBA!rnWY1) zyLzB>|5tI*IR6dSnF`6MtMkbZcg1(A`CzLxr|9H8pZwnpyuj^+;6z@Lh`2hRe9wjI znE!h*<$cZBY3tySlfFX~rD;?Y2v8msfkf8L5Ap}MkW(v)UG!q-5zQES!teLi{1jSXdd9)Z8et1-_fpZ6V(a-$T5p(+%B5&0nt z-e~iD^4}IUW(8N_2~`5Gp)@RF!ygsteDX%)vCHU9SoeUz`Q&Yu;S(y=41SDC9{O)_ z^Sq0am6+NinML77iHZ+V^ZwV364e0ojS_#kOS6sebsGg*gvzG6QQ|zdjUR>WSw(mm z3m=R4^o1J(m#13zY($6FxCj}jjK-ma<#iFihwLAY zd%$7Q0JxQwi15{o5|1wPEl+G-$t5R z6bF5yM9*1e*fCh2Q_ekZl&JAnVb%mLjQbEgQ6i>ql=!lJ0oIDRKQcTNM*1%naN9$< zbMP))CR76UWr&Hq$CF(ADGk8hL$2=N2kVt(LH1?d>mc~wH%gFN-6-J*sLw`_lDvyX zzc|S88(00AF*r#-0;>Ayro1Qw*u8(gL_O##79CgUaLfO#MUM#8QHhnoP9^6mt3hQ<-;#^ny z$DantW&x^3qo{yuIl;{SrOi2SAoe4C0LjNnPyhTsy{hFdYDyAso(W{h59T-rJ;U1r z{qta+->HFR*iJY-0At%m*)Mv{g*L%IHFiBtVHd$}X#Opg3OOc!urRz=7aw*AHqXHw z@K^dFG^QU*j7^0bvKX+Unje47k7eOk*82%RYJfM)s~NGeVjO$0vUOx2Bp1aOtF;a>1A7!@?pS-t zkPnhrzRj_0FZcn{%o&Q?YiIIJyK6Cbgxxo!xw&=37UWBQ{SNa$={Vgk0&(9uG0Yum zkM4IaT~B|JI0$ixfRUcgF`oVjPlN2vKJPS=V<}sZi*zYk!^HoI`?`d%g6Pd1fBi>q z?gOBE^W|M#Y&5*cByXzH$~)DY2fkH+&4cZ7MRJ-1jP%{=&D;DP#;AD2gOwmIytjJuAq<;>!0UiD)Li%GA8jhi-hy>^jdX8bqg#2* zB*6Yd4~E`+!^dfO|AYMs#XNfRmDjLhf$s)As1;)=;kq|3G#%?L`DM^sS}{Q=RBztt zBDM?g7ogtH&4hXM=EsjWWK-dW7lNR)5_$CIiB)kP5{Y;NC|YTih>q^fe{1YCRsEqD zp;aPEx;OVPR~6?!04~%5^4jXn3wG?JY_K--F2Hl%~JEp2^q0WK*Y;o0_cfQe7 z6bJqU>~9Z_-hAU5PWEys5)+Do7!4zzx9-hTCt&LW9yc!!R++d6U-#z2Z`BdSfwu;0 zuiaa{`46k>i1NgT0FKfE-J4J9R#*n`3D_dyG5}g^XTN7FJatQ`3}8Fh*V?_+o4<#v z*r>qr3xHR&K=GhOqku7@xBtT--)`+z$iy}(<(cqJRd#g`pcVReUO zW*^IA{s6Maoo~vSJ(K%yOh@^LO+y$PrxW|y2{fKZTzSxxy#?#u3h{Du0Egt0c*MzF zA#4oTWX*RGuX~QXUc`-&{qZS#8Q?l4c)cB6!b1|@l-Ir<1pl5i-W0xUN0)J@vuZrM z3hR5yxok&~;_on@%UpKY@8I*NG+=i7MsHdVYPNNof~$a!vHi0GSuO6uYgB{qE+c0e=J8vBby8e6b|TDb33pPE!D#H(VBBP+8B1wz?rpo`kI)z~flJ)Tki zKLqW4E_`B5MrmmKy2CE}|$F z5%z?Ae-DPnE&?x1d>rg&DCW`F)jENRG`cfcCdo*@C>fwx367du$YHF2;j&AH?+QylxH=*dE zRU%5dv3s@zlYFBACu#wCZ8dh|e>6Y3EQeqgX)~m3;shO_&RM{8X=)UAqwnz-U4=gFkW-h}rk4MmkVA)YV7~3S|k1QzWx8*7J zz!#~7)Yn}`Vj~U8#-u*A1-bS5x8>pX=q0^F^x&G*{UIEz#4Lt(9=ZqED~%}Y!wMd> zvuQwcG@i-Dr{ZpQXZ2srTiWocn7r8z&38&CKNrWOj)PyF?Y^JQjTSq5E`eWF8ub~w zgFUcFtWzbxL4Gl0fESF2Kf7;;&)9A4fxFfPDxIBlMYM*w=>PBy(Mg!{L0^&2i2Sd< zqA-wC_Z2yX>ac#-@Yo~IBWX>s%w`y)782ew1=1;lkBlpI$x1j7zeojSjdHaf;4y@E012$Q4-B*Y(7u)D2>=Zkd&vg7X zG0ul^3VAEcH{`~-T9wZ;T)d`z~yUR(VH+m zNIYD}^OanqaXfHObQx8;7-@}3_%VmXpMHdu(xMufvHjY&*9G&Gcf^*&+hD%cTJb)+ z!Y<0~_P}1P_(w2S;!6YDEwJ{?3JQwk>d5A)uJ+xRV`AsHf9sy z`!ovS0`khL7Bd@iwgT206ylj^ldU1ti~a^X0CrOI>*QWni?~;qzWELCz7lv;i@oLK zT1YR%$PdPLo80SaQ926yA7EXY5JgQ_i;hSzE{}(|JNVH`0is%5gzA4+3sS3UaV|8D z4TbVU^7A~c+Khe^Q z7*@;hYl6pt0oMc{47e!1)_|*P6M5B>;T_3bFt$gOcs=0i)gzX@fK6eb;)HE$G{%5y z|6lo7A3Tj(Kn=8OYrr*)IT^*JD_Bp>tpQi&1z$!_D-&!IapAo+;41VXzDy5~Wh=ne zXs!ob2m1N5eX!2bNDsL755bC&OR&G{!7$(|{1`ilp2Gf>Vjcsom?}*%!ijK3vO+K- zQrw>st_NIWr{mZ=UJ100R)h+L8gSh_hjqsMP0$`nAs?ciEBF5-q8J{E#0O5`!% zvbCx%MSK+$o3%jBq*H_MCSz_);hMx3~JI5Gvz@M#$+b!gG0eW(hxQ%l8 z%d^9C5BOtoy+&j>i!tQmn1Gq1+WzPRf^;3x&L%d=#@_tf^dW! z3nL_~Qu2qTnfsq;iX5wS#|N{pFCs@Y3THu;hQE!T;H4l@%rynRhM%6#Su%A1seb?2 z0aO99I)I-?OY9Db1^sd))yH$Gs8! zcfbcq;L!p6+Con2dLu^uFt$46UUvY^zfE9;VVy{bqNY0lN-J}v!`mIa>>90QInpGBa)IR z23C_kaBPZrfEVHR7joZ_lB1jSzoYO;8aB^>{vrIPh0&z*`%M`YiZK6376T()TTOb* zOCLt%B?YXC=2nyL{UC%<{A(omd78>vu%rk0yO)fDaoEH@p=Dos`ImX3gK8$2s?VTU zt5qUOx=HW)J0krZ;BhUW@U132V{l{Hq~C-fn?#~C=qA0^5OeLkJus5_z?fV$>CHYX zDvARy0#?$4qe=f{VjQEUh}Q*cL|i;?-K6)Ofv<+dyMy)7?sb#?*VIr^9Qb&!DcZf& zq}whO5#@=m0$i^Jx=By$SwjYJ80-XbkwD#~FFYG01Goj2qupCg`uKw-WdL5NN`5dh z1-eOZ6;T2EKVe^7GgS>;RN2E5j-erQ$MTSOgCkiS{pZfdjqpaXtQ}muMXo-eXA*Pu zSVQJ6z?}=n`HN>V9OwhW9}6s+=)&B&dsGd9=L4-!n9miO6<84GF3O$l|B~0(ehq#& zuO{q^ubVJ;2_7-CRb!S4b{&R$*)aT!x$#{PbC>46r;cEGBy9c$dZnD>V;?42-DSD& z)<^U{jI}jLP<|L0!FhafolxdZP8_`#I5v9f$1&MffKcJ{HjeCG{TAQ;pxW@bQRSU!^l=NF)n|5&rT;7Csx%?(=k3 z45*yOT?ng>W=(QxvE{I61e~EIB7F7HEdQ$lLed|2gqDav)JL;cpZKt8KnoOpy`!@6 z5U(M{MakF_)as+zhgkhT8p?^}-53NcJHdpbUCJ~2TBAc!S+)phmBP9Y6hT%iJssZ~s1Hnu zL?1{Iu=>DbHkZh$%W!+0+;63v(tY5*i0UFI{swwQc(;Yo2Yx*@Rb)g|7|K74bZzy4 zCp*?+RCp?Z)zIAP18Wp4#He_*1AB|O@ZRbJXP>XZ-T@y0Hb!&Z2aY>fg3X2XLXC7E zc#7T(zk>Y^4~9OlQ?(+P|AGBQ#XS1JP6KEO%ze%{#U!l~(b0WimnW4? z)k-KfYL$qR?gMj@(Dxq%{6PySe5((Lu8=4!+!+zc0$@z8 z`oOTHNKqVkaj+y0jy{kV@?mu-#pz(ph>Pc~`@o*Js*B>ndw~tm?sXs7;zB`D9Qemz zGqiiF5Bws32~nQ-TEMTgK=*-LBmHFn--G=~TqIEUft@qMWdQfUaH)%EZ~1M9JZ?xs8>Hwt|q{pW7ZjfFX}tRq~!O|CA` zGm)qjH$uLs!-m89BZatlvyi+CH-0))n=Jrasrhx`-n9Z9V1Mp>{}aY`fqg^#w&nFe zZZyx%$Iim~SM5}^7t-rN+!)n4l061{srl!*(Nkq!0YAJ#2mPgpSdUV=%2?#F-p{1;N6vm#kBk+p1Ei7Qcv1d zV3cYIXi96$J#XTOt<9)qm0n5=^biQe&%qWj$w^B>*sw*b2PJ^r)`D-#{bQekYD z)*(36{l?v`$Qr`7jUxPa3m=R4_LqvR7tmmh*IM{^#F(NL*d(Ai8sD?<*@(gUE3nl- zTQvTPuD0#V&>Jf z@kq*mr;_Fw#X)z!Ynv2cjbNRjoO^V?Wt-Jt|G#;br5$rP*^XPtS zPQg-5-Vb!RR)kQ(b@%JL4>Lr32IxYqh!P6b{f@7JYk&D>(7j6G(fwAAjAi-ZhF^f- zx)OPGzZ0(KlOp~P6khqvgmMub-TgNGEy`3Cf}*rmi74srH+(FnS84+`)dC9N>V9h% zGMD-HgrGl(L}}37FSiBC?spQ{bc?I*H>YZGQ5^Uhu#FxZ-EY7soSjI;_bAvY;^KMh z?zipx4MlO`cfjsz_qzM7+NZiG4%{0B*&jx_x4Pfj?R-Re;>D??Yk}^5zh7KX29OTc zjJQak?tb6F2QeydycgI2?cVBsC$qXTfXRR}wLo{jA>FZ$bQSE^Yo=>H)%Eb<3vo=R zdk=TcD2VR2e7!X0-ph?<-z2araB-Vl6`?Xwh8~RM-*RI?8*laxm{$SC=fb(k4{+m1 zt5g;XmY})pqlN!Rxe--7id6-xuQ~n2LWk_fm!q$_i`oep*!SG1RkI4~1w2@b#rBOD z@z#5SJ73}CAZjwEfiECUDn+;|MOgOCKl0cwkJM%~OR^b!uhvL^Xv2#8B9C77fQerB zJosg$F?-#sJbCqbobU!up8)-BVbQJr#$!7E5X*KWOG1zl`Cwj;MLKodH+iAfzZYb; z(P>ryuckCCMs(bdxwGhEdGdJ%PzQ}=r~HbSpy$PmjBJx>`oFE3mjh{!iQhH)?Cv^*=C` z_6MwAc%XFu{Bfwx!G&lPXS$Npt!iu5rm~otEDKgkbBa#htyw>9!A#j35WJIDBqFZ1 zX6?d;Vm=yziCQ#`BB{1!32)WbtPk(|2p!)5@m6v$hqR-64Cu}-PLJ4xL(<6`egZ$v zXsotoNw3VUSzo@BDkS_*=w52A+L~2#R;a`gg^>v`G9Ken@z2 z{7BbYwKc2a*2)rh1?p*Gb8A-M7xHzUe+V?u!sgbjKHt=qQ*XQRddHk}Kq?3_KPnL18Z1hp1Pr zTpMT31;LdqFffk5F&TI{#%U8J}t33-W5rty!y@HI$OC zfWP*Tm|L@MOz~xNvG}(c7pBGTn}Gv#bO6Mtbfu-4_|ifN8P>; z`zIa@!(<&s3606}RIiPE5lF9}u6qjTzmHLo2hJX{r2Zw+}pvG)KmY4ItvEl7_ax>-mbP)inn2RL*3B_OuKP20 zn$0=tOL>Mg7Rz{%&WUJ9znF{uPi$8^MZ6#XT;f08kFNsg_v8MLE3xMAB7?m7oC=b9 zKaSs9oArk65Jhm2x5mdJE*7i9J_eerF|`q<>*EpK`^B?0KwCB5Zn-`iv9en{I|6h@ zV`@S4`|%I8O0%z_zX$wOOGNnU{kZR~R3Y(0Y9e7oi6S$QQoW%tcNP+9`DB)laocWYRrUSo)TH_$Km}-iHc-=1-65@ z@L#_l4;YJMu1~=JjCQMw8MW^_czm5w;{CWu$^Uvkrb4OSk0Te_SPYa!VQilwDS6+I zyRMFAm4WIitVeI6WG}XzKby*Gm7+p@d@RPTQJ8$CfcwLYAUE|lsu5qYr?Bn^ymcG42*3Q(Mk5os_-QV|ENTQkuBVUcW#p-G5I=r^vDKW4gAl~pu>gvI zJKiuB{*LQbZUW|#fP$5)wand8A%^dh&x`7{aO7Puq8$e#eG{w8%vF*8MquA6@v3Io zNnAw)mTBx=xEM{Yc2HdY8IPlW1S{1( ziA{t30?jy`f_Ilc+mO!O$wu_dkFi-p-2LG$XGBbo@nPeTTSp)lDU!)Dk0GjW1a3N0 zNyoH4M0a5xkl#t0kf&zyn;O2WmNT7v(gl^q(gzr!lS=_(0=}up+!=;%y@hc?n*v@% zX_#Dk+TQS87Ah|}Zw=I5|;}nBfL|K?0HY;z9+nbk8hX zLyW+FYr<5nG4j$VfG1cIMrMa7TZ1^-ZTDy+ipoYoV^H)@_%isuA+|}-bsRmZ1WdqD zmgGSPa!bL<8p1@vZ}J0cqFJ*0BUwt$V2(ZM!4RC7mO=InD*3SaaJGV+uEEpdS)!B> zI~m@DZH4_enjIr%KK7|bU{I(&rOMt9qH8d>Nc;oo%*Q^<@Ll{qw5OdysS2Pdx|bWqp)X1^<4q`fkV>TC^^9I= z1SV#Y_3)qaYM9&g(Z|5}W|vLceK1KpwgliB&Ht?xZA=ALLoUF=Yv?K*{( zS#9)kOmcIR|ME!^a(aWzX;(6GdhTG#X|nG*6E(9;B=d!_mHG-fJz3;5v30+OvpCon z(`@EzIlaROba%+7QeeM`Xf=eeIf>`JmeXGwzTa1~a+-95pm|^EWKIWAfbRW<@pN-> zX3POxN&*q1*YhpQ2t2vBh6$MD_gpvAez~P_n`@3w6(b?XO_b3N-|8PMS#+2i@}* zof^lw!+M}{?$Ps~=#s|fz=bgxf|*KW_56o3;uxI|Wvm7Jin#D!_x#~6lbLZC_9wJk z^!#+-5px9~(K~qEjN(`?wamMIx&P|7?cHIY@HFYygmQOw{h+xPz{A; zZm1xuT^d~<(#R1$^b^i;}O`O#D(|Pt7Y%MQrIT&;PMy* z!bq-PEx(GZ&Pv0&oJRW9a!fW>S*OFknFqtGW$ZXCmhBGvfr@#&S{}WOZ6|yp=xnXn zP6^kqmOo_2L-8f(Hm&$tDAcQE_vv_P;KxBPD2041dcAF(H@(?@xZ$}FJXa!*SIgT? zBc+Ikr{KF7jK~8nqN86e_YDg%RaK#=uT>&S`qgsJeVjei5wNEgQ25rX<*VcMW#^p< z!6Xui(x6{0Pv)3A8CHO;vABA*T!zKl;@q?YU`IVTUM=t4EzRhtF@6K=4sr3k^{Zvc z%jHCI;T#3k8%Bn&UoG!!t1gNIF9cSMxNvX1TDBi(7v+i9229ri{c726UlSR?+hDzj ziv;Rd%hgqz$^bqB`&hfTUM;_Cm|q6)IpA6?(65%y=VG+`9qbQlrmCTf>Uv^<4}w{G z?o8*xDm0Ice{yFT@0G5=o7T?( z!A*1RuJKMBRq&tgO0~6n;o>>xTg_9&W9|ue!^!RPuMy^Y`XPo4;&`{*lR;a%($w&X z1(iy$+*8pMi6CRd*yoK|?hoR>;>L)zU7O+p3o*<}Ge+2km14Q)#ecPo5#4uIWx1D& z{4^+%)ixqpKWM~qe{lxe6PR%wfj-j#(@C*>BE~B6lY7Kl>k!^mXU`Q_TqbO zMNbiZ8S#CC65@=$F(BhLjq$tPfaNk?u(R8nb;5M~QlQT*9CRa;Td_xZ75xJA{Gul2^38;?3@&=li+N?Co#bUsKhyLs=3Ry?+ep(~S3jM%^ef#Y! zwGOqc|Mj~gNJk*?kBv(N&@B!BMauyKM-`>hK{)HybgEvj==_h zuw>2Y#yNf&Plm-jXywgv(|KT<`f?=O5TZ6(NJlKPn5g#1v_{;y=0O2b3;Kf(RvH#F z=XfEO`zCibTooqpG@v;aURfeP%k9L&T{!lB(cAa}2DC-vxOF!QvfM7*S!`Enb_CXE z2#G39Vit33o|ENv=kZ@p^kHud4rkfm54H1{_i;*YZXfQP9)>YCs=5oUL>P?pLL6!t z1Mcivn@v|hc45p#CFv$lJl~jgaW{2{_iCSN7+lOr+{x<+aLi^5!P5I4T9ZFs?Y9G=DL)Ix<{N zWRQKs5#?BAJcGJ0w(*o1$0;*1h{aqUAM^a(cq&0P#J#B z;Y@mLorxR>z7*z5^5-JF$dlo}IRg8(_E9OfllVByX(g6-hPrP#B35p!E1KGSK)D*5 zvl4e5&LdCFoz|F@Aa8<^)JPZ+rka1rkq;&;fKupw;4pHpH)2LL$QqDTWPxbf#AQbA zhYs`fQrrg#2X(?LGERQ?R$`UaWO!o<&NsGkA9lp!mC84r6&*L1wgs7j4O#dngy`RBj(f z;M|g4Dq4(!Ge$u)L5C;aT=(+{@T$eC8aN#Vl24U4D!a(?Cg1o9pRe3UWp5*N> z?9L{2^?U_i3qNkcv{#fI%^kI*=1(ed3$8b%k|k($}Wn(`zO0GsW~qB{19GDBySpSx3cS!J@Ckm zqB^@MIAbkD+jV$mc5Q%Hwco&tK1q%$Z&Y@X?BTuu)Y$j>Jlx>N{E^GSVu&gO};DywLXAi ztq)gC$fawCNXUm^4bBBJ5rxhQsDw~hH~IRV zI3wR(Sw5GZwN*;Y=TgPt9CAXJIQ}mDSP!$4{3_}BT&g<)Z|*Lr%8Q-Emtd|cF^chb zGn?X@5BiG=*2Ej>7y|Z%k)DcTWtQMN@*&#Ihgi=M{rsPztRt0Rin=@9VYmy{WJYbc zPA4CF%e1cpY;@e4i8yyqK&73+yvZ9s4oBL*muat1$0O~>9nJ$Am3dUAqr_~}ovIm};UtL9uQE*08i0YG2B*JWwa`Tm9 zQ?Uz&lcc@!Mr9XS-sF#`*pXek_?*l(p!E7eK+@4Q= zuOraT(p9|dBt8psNr~0d|5iTza5M4y9kFF=HDQ(L=~Han2OUP+Kg%&A5U~k|u}!AL z@8^#IBRb=-Eb_Uy0xCIVx!BfZtv{O&WDd%{u?J3>pQfV3s3h3l!H+@oNB#FOn=P`+ z{HV0e9{BW#Rdq={QwcGJNB`gCBUj+mUH2-PVLj@5WY}tlGj)MgT^7U1MVK4p&7Ynb zw$>4NYK`t%>?Ahep$kTcRfcVl85U<|*d|Bx;ZpV3M9MIVjeCp3D4KvhF==pEpB&fO zB{FOT;N(;+2PL1|E1)urEN}9adz&ITndb9v;E|k<9L_FD`olEu13y;5Y$m@#J*&$E zN8qlTRx3{8voM#GSS9CUnVcXqIa3_5`_od{UsU)hHty*TqY!SwGy)Nua2VU?GC3;% zE8(tl5sl&ssN_&^H+hR@{zy&{nVjTwkK`0}IKQfF73a$EV+hPx@+;LdImI1;&2KbP zWyen96)>V=#8(QZB##C$I&D_nronLg9H*n;_@)^0BL}`pk ze}r|7ExjW)6eu8G5&{rP7>vXLBD^=ejkDkRFry^wlZjD)Y|zpbRhhe;cYZOgfc<&cqwz;!m-u6`tw>q}b_34L)}f^Wbvb@RtE|+EP+^6LCu(gd;Va*`> zN29!15UgThY>z1}AE1CT!D_klv(m7ysM$DTapcbZgJBGrT7xwK>!|r!oQlF}BGP;p zyKEFIQZAIyZ1PZuMw77MUdl?k;=z5@C`jUhcs=!npHtt*dM$dzU#dJ*1&=JlpJORZ zlsnty9MKr%g8p;oxI%`E!1_h#e<3%e;J5rLF1gzk@=1IV_6Y2S=5juYZhUh;aQQy0 zRat!C4u_|)Ft)N3KUs(#yPOAcJ~PdCqySa1uqZ^&ToHwe6%&Q1HBftv#cj9le_Z0? zZd|7%%FPRxctONfK`y$_xN^WzoS%KMg!hSN)=Jm3_Yn5~yo6u@Jl88-PbHLKhhcGo zLcDPqZq!QGIxBI)&~;ef_CV?W10r#(+CQ-NYGS(L@6c(zYNhL;{) zT@TTFS{ei`@`^;n)k@b3p@>8e2;SAAI}}N^(p7k?R=Q^6E+^9Q`S`Jf9NZ_ZTIm{k zElJ`XKzlV-D_x~m=1SLcKVl;VB>V<`+|gRK(sk8hEUYKYnj#ZmWJqeIYxkyJ)=Jl7 zLz@W+FAZHetyL>s5B?k?abuvC7B*M99=&ZIGTjI0T??BlT_1EvWg_B}fo59RTU`9`X$gN3!5ulAAKIr=)v)$K&LEhu5=BlRZC*}wBxq0xze?~k9j_}1H~=~Mq~p2 z0OdkoLNK^uJw_J11bAhw`Tr|jf9@GAg{>g!lt-vnx{j?~S;l7=cxGOWxzhDoHS7}gGw)zjI3N;w=-2DUK|mo+ddGuVq|^>_CCMh+}&M<6;$!gcV7MKpWp$FkmY zo^RqplK|^G$`2N?J)r^18s^;gX(48N0c-CTDtJ1bMpnzLF)ndbji{DcQ~Z2=Y6%@r zgt)wNz(l>v8tbA{khs)N4;lDTnvn4J&~+xQh*A_)%dA-;Ih`F6j{q8@v8dr$bN$}z z8z)0r0JM}a<&10qvgU`Z>E@KKcLMFxu0^HJS{SmYUtx(a09~>0+Z7wKtk3*DsF)&? z`xNLoVc~g{z$-$^&Gr$lZJ{kASu~7{xo8ryK63S)i2^4Uzh|upJ9XYmNO(=?>XBA> zm4AA{7?zdk8u$}BFR5J@)2dx5A>nIae%}3V8u*g|K(yJ*zv$R(dZ;qLHAziv& z3-pzCE&7G5(JoOB#f>Fd+x_Ei`U?ra0NoX>jp{roon`IvA9Tt~;-^5*Ej%HtKFiwY zv;2$dvZ)Deh4jOi>HlJUHI_BeED_XdWgYO{|D>Lf@S4!oBdtt-LU0(%nq-y;p*A^pq-f0oxzsrf0irZY&FLZuf zRq1*r&^+xLEt)(RiJGttwrNjrPNXxd9{29m25k*l;>gnAp>~$lfE%k0a%TGxZZDJj z9lAxCYII3NPB-F#ow`+Lw!c8RsNSN&2ju$zMfP3p^j_u1Yyt2$8s;_U0l`%ov8`vXF{|O_=R=W=-PW zRoWpHD-j>Y8$a;!VOevy-|*&Wv*y8r<>bkJN=0+{fYw$h7I5RzE$k)S1)8N5C#aDK zxKCazS%{tcYJmEtWNdyQna7Dc`}JzsBHwzAgl9^2p^meF0tGtc%*hojo|V`}BcgXdV^w zYLQlkWgX#;W6ev5jN&sPn5RTcET_nF7+y)m>fx(7>~|{0&xkZHt{%j)u5)L?Fig3e z1pGw{DB5N`gesdm4>!Wx>SNfy)T(Xyu@pUEa66RJN8I^$4{VMKeIt^^!&s>|y%8$I z<~+6p$*BxQU9A${6uBBzk!4NeF6vac*X;Mu)i`Z_`*Vv3CGQ9MaP5TtVnuGHApz65 zi@FulNu338g5=QrX?+=YxZ|g>^ot%H>tQma8HPKiQnWjYN4%hpb&*smK$t1Sc!e z2%j65#ORsv_n`VfD@D3%v;cyYTJ#e=YcE>&nsvwKTr*96vE}<9I7*^l z$zSOc-O`ukwmBcZZ+-*00fE~?L`UglJ+wLJJeF(MxEE?-FwAQ{1=NYbAmU$}^UxB^ z<`e@gM*>kJMUBvn6E#NX`G-Y{V7Qg_x#1iL-CBOJE}Eq;jMBd&Zcfz@gfwGRn{8+Y zT13@k#hqE(4exw!q8T7h@EL<|UXDt<(}=4HPn-n<`rk`rS$hm)QF~O`4Y1ovk$6V= z(O9s(hF|01#c}@)>@O)MMoDI&6{=qiBkdD3Uk{-8TdQQ*7lIpx;Vz7oSb^;^X#gXF z;Or3PJAKy9x@|=Kw6qx`8=e9|&AcKR(Hz6LvlvGvWdL^30x$AK3LJK0YZ9_^7+|Ir z$j2*!@v%n)u>cv&VhC1ik?=*vCD0x*=0Zs++7H1AEqcwT684B+v1Uy~_!a~YwMZnp zSwG}#r_pJ+e-_#A4Zj$RN17NlCjtuz_d4%%_>VQAZUDU|bls&u)(PHr zBd@m|K^CapAdwx*vEi`&ND=<6g^xvKVsJhmXobd0EPOoTL24q~0kmJ^Ll!<8QLscJ zI}dbCRp?AEBvC2So6_(ki z*M!!aCHGBr0`Eqe(-a52CbZ@57&ZddW-@<<{Wa|tPqfvHXuKO0M#6XS`$71)k2Rrhc0??AMhhIR>Ty^R zsnwd$(|gLW;!u`{v3-ev^R5Yv`JoPL0MtrhU5G@G)zWi{$OHRsbjFJqMZj7Uniwfp zmP~@%>E!+f<&<6%x_mkA&xFltpbdoG7RH*;uSy4rj5rK*Lc6xsgg(ZZmsEIef#qm! ztqFa5TqvXB;pl{qU@+2qYfb2uK`!VUN5FoZVjgQk2W+JkpNl|OX~j!QxLy;Q@e&<5-v@eBD~zM0P-{ZdXV59F zS3&P7g?ucc38DMC9;__JTyVpUw=fe7BU10NCUjetpA_*DP^4&;h>l(pnp8a1R5gd< zO|249(rZGOBw=&vyMUv$fWo)dgwFAi=f3cH5G*E~0Dng^+|a2nmJ|IwYZ!P(lkJ9aMTRf*>Fu z9YjFsh!jCUI*K$ws(>^_igcs}P)g`66ybZ$%-*@#{a*i={mstvoau9B?zwYk0PfHP zy(jecP%FOwMff?5>5N2M&dt8&j7a6tX7`#!T7@ z!JV`?M^7QacdOEaczchXxFzxvU}IhJ(7fpd=|R!B&Z$M|kHbiG0q`Zvk_zXAU9qWc#6h&dG#UXbp#~!F|CGYS+GM0lI6zs$%s*Wm0>RZ#2A6A--^y!<o-cwdY zA3y`Nw7F2px33U!=lK%YOpTihm5{h9#NFe2U>g~i(VGjExmC*01IYgY?4-u^LM7l( z65T@ZeGTb_N}IMAL!Mo*{()1`DhvyiJtZ6@3L?Cifw?VI8l~ev6p;=zQ!^fN!SzCA zm<=nS;4W;~Y+W1(_l7Ve7>p8=h2FjTYXh03hqME^oYtO4OCgXFeQS$EW?Na6$- z7d1;}M=w;So-|lFU|734i!5Z8^g_jY1;&&xfQ2=I(>E6?M{lR9?L1XLsKF%J8}vfO zzqWIZ`2bia6E_wr{{HE*JBZ=H#=7BHs7yGTO6Sp>;u~N~8JF#?7b?HpEhD>&*a>W( zme&iF(#<1fcMz9=UDxvFLS;goG})iTTY#SE?al<`I^&wV-<{ zJ{460Re{yi^5#P2-YPGZKs$h4G(j&^hL(hxZ5YBwYs?si4j$`CIZM504s>|a9{-`k zTLI8I{N=QIbQm!mTB4_u%W|a@r^+u0y+WyP*Y_3h7P~?Gw7gv26(!r z$V3bse)wb%Ed%tU0lVq&T|-OA&UhI3G3J!wJZK#rHla4%K=1=Y+)anC!r2j(AYt`J zh317*^`{>;hmMXXUVd4N0xQY5j9=^UaO?%Q)W=p4wHs}7HyR+lyc+GW66hsz`@eS=%y=CBUW>;oOn88CJD z)ET6_TrFHo5RS~i#f_lOT4rgg#!qDNbOGV0xC=DouVJU6u@R>Dh5qF+< zfwk1Qslz+(!3qU2egteV<1%_vhmVa5q-e-b2R29JT8C$1nX?wb8#JVK_=-|kPai_~ zQ8x@aJnkxX2j4*W0|Rr@;r5lRp>4ge^%YKLO0?vHYaO1k124lx0?@LW@quI*I{eQa zg~6x?w7J1>)8Rw=`cWsyh`t~UHArqcJf~4bMG~{Y_*S!IcC-%v`E96^wH=HDnkBQO zb$E|kIEL#Yz?+)D>F3hni;Vp;y`f0pRC{V29-Qi>bbKLT@0hru!|U`(mfb<50n2d1 zp~HX2hq?0kpaZaOjLY`cI(*GPzOuWB(ZD8Xd9A}cO-_{EK`aKgLd%;vd?%h(xj%`$ z01s(`e%SmnCRQbI4cINlWrx)|{Ps>f;Gn+=D@KX~PG!*4;W-fjDuKcPi)w<_;WG+i z_OFKUS{l)lG_XaM+gSf&SOUw#)R4`bl($^>hF zsiFkOL5cEvx}qUPbLz;=HR~X^nPorb)=cD`JNJ_UKWkr*_=|-ONO7V zSXO?rb_c*i4JJNbmPGf<#2g7vAYLRmKFm-4aL$Nl;HBvO@^+D^mVHHlB$?=ibGWOy zf<27t$UagNNIerR)+!y}8^n5t6(;Vw9f5Q=(SLVWrTf+Gb+7>8hw4#4J~L7INn3sY z)vKP-$}C&oH>+@1n- z|I-L=p%m-j#i97pY=nGjJ!Tzz=oCJB8zui8#|gXx(C#I3a1;bOfoyQ48OXtT&sVWv zb+YhnDINP^UH!jr9PI{3=v(Ag{1!l)5Z5-2Z!&+1r*BL7)T<&LL+}Ly64Ol_x5y3d z>_ZQLz0&v`#`SF}6}%I0{vT3{fRl0Oo@#O9Xzkj{KvfQSMdp0N@@lHZjicXU&srk{ zw=u-kREt3&zsPe8M7oga9aw0tm>Q&-EJIZwsBiZ;RS)T6wDtu&hbc2O^3#f zqdS+w5V#N6Q3KcWp-i($S))x}e8L`1%LVvXqOX=flra zYtkx&!~#lSc%2DjKAiSJE!vEbG(Z_z+MEw(+$~C6{~dsJ)3`Ywwu-_%r@%)8o4~k? z-kcAA+FX#f178emg~s)K7(38PI}p5ALwY`}yRbNka|plehG9M|h*x3aDZ;70GfDUP zFa}4ri2OhcX~rI8!<-M7&JG`n#CK#RX0?g3_PT!mlySG%E z8%5v%?Ei;Ty+O~1r$(r&IYen-DJE{rhb8)v><;2xU@hHn%!hOPSm`8}_#?|lY1=26^NmN2>AifCWezt7-Ml$Y$>{S3iBVJ zlUmvwi=8f5s5$U^z@BK_9E(@80_pkb0+eSU_CLd^=*_XXt!y}zL5jtJmDRW&i&GMc zQym01)Q}#F3+;G0-WlON-7t*BdMmNZbQHoT8<^WzTu>Z~SFr@>_nMK)1=nlSv~Ub{ z@hi|{no(6UjIo#$lMKc`Kpz_z8;I7 z`-;f!AgTbXq2cC+bi$n2a&Wo`> zsq+;hKEWgaXE}$VVa|&wVa4bOLMj5P&hTXu##sESSSbC8kk)|OX=!sTZY^DurU4%U zY^27`u^3mn1bqd3F0e(6%jnIq*!9;GS_1rMV7oQ0$KsgHD7us=lj zb2kiQvF8Ukz|?ODR6m?d(rqkm3d3%1Q37ZM&A5SVm}4;kUqlGe80dSNaYr(YvG{0B zJuvzM9d0n(#$q?znEVhjVh#vP43gVe%m>4}B#E719MUYA9X%Es1=Mk}u7h!3vt*X^ zSnSpxJK;Tt!t?>B%FR)mWAV;Pbw#Qu3PMRH$=;yHV)diy89>wp*2u(-u^5+DPId>; z16W@-9Ahzfx{Vfai6;V^%D8NAJr-+RvdHctRsmb5<@H!BC~C^?APxgNrsd7C*tu7X z>`&qjz=xWk$Kt_X3aA8pFbD$RQ~~w8ILp7bN+1zfDaK_6&9NASTkpBUiuwTG)dW2j zha}?uxo!yWqcM!d*A=}m7Vl$tCRM#0D~}nih8gjH-hx#VfVTBre?NjYBc|;f+Yj7U zhOIBLMJ63X=mi52n@sp%NUJnF4gz|qVIGFs)^~1W5asz4Q$L)HS8Oxk6CtUqf+z`4 ziiWw&+Sa$GbTa*dfJT7ZXo^hVu=REQK21{k10Jp^G7-bp7u~B2O$W5VfYtEO?`tj` zQlDyX2ELs+2RIMf*0;B2MLLe)zYK9VTi*<9a%9bAeFnm7gJiz1d0X38_El>LDm(&C z#jkCBpN+)_vZc9|7?X9cTas1iCNXTjx&@0DH~-t#$L(&|`nD7;OryY@%+akx#$0WE zGsc#rMSxZtu&NCs$)r~UVz8dtW*# zP|64X1i&*iMW%1`n~~vpBxN<=&6*+;G5SrJ%~j|ypi>6y)^8r>Q76pY1O9|L1vrbk z-wb~iPY%>oFr2CY`MO`(fN^^ozE_s3Pit`y${HlI-<<0}a=l^A1lE*s8Nco~?~klc z)*cA&t7Y|U%spk32%QxL6|m5D2ZaCKZ+PxD`c2f&#b_&-`#8FP|6jjZbf^KH1$51T zb-$4dfYhC}i0tA|o6lJEn~R5Gbh_=^h$|$1<~eM9cx~TuAb`R~!8`_MX(Ox3Co+=+ z4p-`)T#m{jxPpdSUfu}{Y9Zk}`+hVvL~v^Z5;D>}5>FP^_U-Xh*caeHgFy8pR$2JQ z^g`#KjPUseCWS@U$=ElcVyGsp2P3eST@S_%&FaFrQ-n-m9dR2A=05>m(u7`;uv+#~GxdjjySK~Qz`9=~=L?L=VA+$j18A|h-wmVR)u{6Z$9NPR?LK-F+XsschC zO_E7f;d2tiFyWg$$&{?j6~ z&=TQ&ba%4C+X3rh;^F-QNURd}hG%0`{3C&lH}SFWp>j3}-&X^1B?uRO5wK;9%K~(| z{RYpNdqik#XanN0BK81N{jAf41IXKX5!$qSvcfKz*y}q!ByNh(d$o%y?4`yq5_dNZ zqc$Dt5lwCk1EYf@4mz0v_ygJ?X0rx}otk?LXNdkv@5F9*2RAeh6Zeg>%?!ki&)}PD*WD6dP8y*?m9%k2OgE0KFzz)vzFqMwAoaEa$l@jWtRB zkPMoS(B%dqZkX`Fkl*YXv<=XH4a;6+ytlrNdos=gx}o7aru2!B)f>F&1t5s6laT&i}laz_G zll2Y4J$?kEy=UCXwQ zs6f2{4KiT8CXq>2x|>=D{RXx{uPQO!d&D1}<%#uJQD7w*m(iQ{ z2z#|!QmfSk)=1;p9`Wh?VCszEo*L5jh$=AW7 z4TCLV7tq6+5zGbGw!s|_u}~0Kfo5w)xMUdih#UvL(-hW8*#7}13#OVy?>$907_Br*W=Gp2%C1XsvIc-LLbGI+v^^qh5?(IM0=P&MIDOL| zkr%Hmd37#+24OdoWN*;+h{Ocvd3NW4T`_UP9xw(gbaf=#w{E zB`^=zw~WgIYJ0@*-72aCwgLM^%bWIyGi72`0)GL_(gbaf=5XdOPIwFlM0psf#QN#?dPba;orSZasRo(3Y)O!#0(c4`WZ05nm< zvrzB)xw1fqGV~3gl^V`4rB8&!XP2TKfDURno?%0WzZ_+!Oaxp3d`DAc`i2fa^Qfkz zSW%iha5D2kCSvGt|B09{0VNr*n+~7&TpdAG3wV9zH03;K9scle6tzciFGJi-hYvrE z+kqir9S_1}gJkOP^T{P>3hKnV4A>gRW&B!)-yVf)LH8j1pq52TZf?$2ULO>ClQ{B8 z6?yGR`jr3D;mkI4c+0KDC;-d|I7=~P%vFbPeCCoYv z0aJ$u?N#r0`a*Uf%eUj2(mMQyuOnqmOaU~D;Vvc&9e%K8sH}+Z0d3UMrVj7rRhzi; z`~mEw#!Vew`*kpJ_qYe_3F9()Q->dlFGT%-=a~v40Gz^29sX}BB_OzjhO`dN~vb%^kz&y}dRr*?oU&+dl-9bbFE6li* zH+A^&zG=k$NmK<`QxmifFZp{xl|Wlyof($})H=NIrxumK2w>y1ys5(uR$FE_15qzQ~E?m>;zmt0cf&@yYLKP z=8x@il z&a#=asCD?RQMmsC!6g~us&Uid3yajE?FhEk1)-5aBE5gK?@vL*wi9a)V0{^v@oODE zWm7)re}qrgvS>**L6GAp8h4Xeh8>~24!6zvFCEToLx*o($B(F9N{ zh7X!Bba-T?G+7ZJ0UE5OO&vaLUN~{*nGS4@#!VewA9FExkM+Q|FfOAvb$CdCuQbY? z0Cra6T8G~`=}C7H{8&R;hlfRY;Qqte*#8EnqE#4l_`ic;w2VV|Ndt4!;c1s)8WFXD zzN;C3aKW_>≀f7~O#m(2SFkVd!wGiDnX$fX+4;ZaVzCQbBYcGGZMF+YFMM4j+20 zxFU(uU|iKKnH{aeS6>iL)@v|qUpb2`WR|oJKeDS3iAaF&Xac8i>hO2-IS=(q10jP+ zvNvcQ-YQ-w9p3?1HxoB>_@e0+*&W1aU=!SM=u~Q1$PD9q zfF0EGT8Ezqs-kSPS-@^;c~ghish%PGldz(vJK&rN7~cTY2rr=$hy|9wxGbR7;qL}i zPzj_1tE1&j9qwJdph}<dil@ep4Oj6 z6~z88Fkdmd1K1RLXm4eD{Gt8uwDPLOg1^Qg0C1LW5{@rkg&y~{UvzE-6=eXWn(*{t z8T7cHy;y(eo{e__wbXD(a<-KokF%$Ci6zk+!J`dGi27)~_qXEmm_D4&e#^S5CF$|a zuoj+WWNF2IOOx?c>G7?wAK#BNLN@$Ziyq$&E8C>93JJX%P>~+L^|yUkUL7iU9Pytq z(jc#@cdOB3fvcI9JGYEH1oqm%#Q;vZrK}rI;XAiz9ST5Y#ly+E5#Og{@AnXp#CJF^ zi=wV^Ep~B?huuRp{5Mdduy8YN{fs!4gD_6_W{04Bj3UlgUMA4P8|oB3<2N?IYHv0kZ@iqe6W#DR}{98p3Xub5b zB8gJ~uNVZ;8L?2d;NJW!sKC`?anzIJeF=gm(pITRQi#k&KC4#<&Popl7N_xk97(8* zBfLP)9cLU>fY)*5q_mn@m7Wz0li4OmKctd$ATfbjVliHD(SIz);{oW!_>`Ri^c5mq zz!Cp;5$PI>@k+Cc(vJw;ZXn{g2_Fo(jedCy&;<=|HQ~b{)Ar&5AwVxRe8z-NgnYWa z0p(eU{U2~LS+R>@V=;bsTP9sZKq0>JMnR z0juWI8$254DC^xy`~*#?P{P6eodR)(ji6VJ^now&N|e&wz@>&TI=m$uJh2FYzP-u;e;B8iS*^wBJt9laQjKknsZeGbM<&5~Kti}88S zV5eUNaFZr*`sQMMSh#vJ6GuTf#U$Aq^kTekZ|BEe_kle%abq!_r%Vai9fbc9yaa}G z${R-UDG_z($G=eG62QtaF56o##<%*#$nGK<0&Aw_^I4EZztpMRHHq3Hw$2=5DLw^4f2ni2 zcM1(b@Hhh!8-q2DFLk~vS&HTY`%dH0EU&-R`C&c=!Zv{W4MJ``iZ69?_NXtkF9W~9 zoMe_)dK6#k%;=IpFA?mC49l!3J&F?Z;(;77z&43(SX1$rI-|gH>N>YCb(n2@sngjK zNts}_;y4qLoYHm1mpWgr#U_M4z&t?Xo)m0ItKRA1`YS3=ow4oqctJhHTWH2EH~ z8(F?G7e_0*^(Xx)9U(^m{mF106Na+OUz%t!>NgwEBQ0$zySB}06W71*GAx4NRP3g* zd$7%mxc-X)E6uo!-c)vfmMKXMf!7E2uEw>pyS&3nT@l<{Lt5G0n1;`n#vuH2Hw?<| z$xb|hFGTn%19MY$%d)Ua5x)TaT{D_-!L{XK)xUVc6xV>>(~LHfVJN%AqtLd(b2&_3 zaI#=-$}Xd6O==GrQ51x72FXp?)qWJCNTLxKZ8S?}M=QI-V+__nFh*&X%#v1iy}rZ; zNM8Y5tO=aHsqD&@C-t1V9fUnhlD$DIyONFbDf};BStf2MyY0)X$nGFsGqys@8_Moq zpVHKoOB?~L5aY7FwX(a@u(0edq7tw)Ew7bb{n_ENJBasybw6Rf32<(SO9DZ7mbqJ?Z=}73CYi6RZAX!FmROUak_TIOIFEr5h|0tjCkBy+*~$AcoYgz*)?zGqy43{LT?h+{t1?E0Bf$+O5E+JJ7yaT)=2{@QV^I?aF$PyuGV_Z!RSC;tO)}RKPtlZB*OyIg zBkgl7$*}rl)sufG*vWAfIS5HaYFUk~w)c~9{wpGD#}O{NX;`E8TZ#nwQ&#hk$FqEB z8G>Qm-QTbdWwm@~{l4NVB=luWCS`q6FwZ}E=!Y$sq81>I6-GRQFB_)Ngk@zZYeJxR zTtPbX4X$AR1^98z;nUcJJd!Q+_@Fe(npE(wiZLp(+aNvFL>1Y*{zWM3%UE0E7+QkU z;YFTxc>fP4GfvWWo%NmlyG8hJECIo(46WhJOWUpU~}HZIrb)e5@@=5x)cJkS6k>aDu04 z!DeIe6pbsoHbd^Ch=qI@aqVEF!S5lT1W)00|jx*_s@0Ovg8{xOscqrogAa&Ek6{ukfVWVT# zKlbLsuz39x*cisSPARlSIQqlC(GRZXqqW=`^8qc^a0K_~EF6%LQ8|WwLhufTxauhQ zKoN9E%*LCV1&|lc$vX>3+CeGdVGTf5;eWSp1ij$k^}7A!bB7{$2>+iy3!v!r|BoQH za0-g$Df~xvz<03!gSC@c;cM&Gqb!>U99bqn_RbH$><*{uRo(&|K5$nx%Cd{Vw&yEI z&S>CYXpZcc5wyM_WqAwxff0FVF|hR-mmN=$PG9t5l&nA% zfz{HuJewn;!;n&x6(sC;VA*N|tf$6vC6fFzauO`;YrFZ;NDyYZlH~2^GP?5j)`d`3 zhzS1_$7$hA?GYVOz4=AeQ9ohznYa9`7J-x%CQ81j0sDFRP5q%53W!#>cX`o@#U5iY zoU$TCx#qulQHJkgkA4@R@1jJWmh*Aw0}gdP4y9FTP(-wP+dT|17T*Y+W3E@4)z6TR z5rsD5B`BPQZgCU;?HxlNr7iD_tl%Iogj5Gqn_*uQ#(HIPxp$})Y{u;Xb znAo2q{9iW=>y^62fCX*B`VUSf>9$_!bBIrNCy-EYv0}9LTBGv)htmXB3_jxQmuWAi`ghLXR zLFPx6TlhE1_#J)>LN)7ooOUV>06d`ytC^tQ_&l$kXHHfn`jO?bL3pl7G8QK(EzkBX zc0~$6;t_BvDH&799?nV)>%RBU`AY$-WZW03klqhl!Lc}&$p_-j+6(uVdnYI!56oBQQnkaC&knA zZAxdMwl1fmTn^TLb=sfyrFt50z!6TJBGDnWGNb)Q#DBVZihj* zOGQ7bma|m&&VXIL?M#0guM(35t-FH6I5W3@AbuK)q_v zsr){>H|4AsRo3m{W;i;x7UgUdiJ5WFGD5tK) z9v4!LUIVjXWalb)`P6$DZS^fhj$Ox!lN_vOQV)M^r$#UG+WPOtc(i1^nBr&vsWGD1 z$!K~p)tjHk#)#jsC*Z|2AI=kw>>lz@i1Wzq7c=to2D^3=$W;3XSb{=+7S-uRG3zT2 zY{nW5^b5{K310D(W0j%s^43%Lp`R85TdVPk65q=5)vQlbAioFL5d&8pP5%04US4faT5`~8-n*dG;8Z?jI-#dMlIX<;_OoT2XJZuyUY0qvIUAwH zaWw0T4)$qY6_mlf0iY%t{>;K1-3!spINvzUwoXm>;5#74-H3v{mHn@o)aJDZC_uP=3>K*xC`vD zmOrfINBWc+UzAvR2|qM&5S&W?iscPp20%k{aMoMjzR7vwQ~)#Q8U1)-u?F zVTxbUq71$G)?RLRd1?))odJ_iH1_&%Pw_c`#f0o7jsg%#s@lDh+#6CEYUzI-{*eYi zo7#JQ%qx*MO#z)L`bnqC8#K%avGNX!!Wh?zEGDsgWmvfOHGmK8>aTT*dEA~zJP~nicG{X zwWqxZq$z;r8nEijdT+>;rRoy34ZycDrzdAoo7yjD)TKWVe9jPeGqqnH8bNVqicYKT}|yNFmNpZw8DT@Z5T;bx;&zY+8dI*ACIn_0rS!I zgXU^W`$foJWBEZ`Q~J?0t{txOL&$4}_QHK?!gzGG;bDR+q5zSN`<5f_9&LcNXIw^aKDyrfyol5|LxGLbxPEkP)hUW*BY3`s^rP#q_3`!p zdW3Ir!|>=@A9_suf$(z%=Jx1%p&yi~cmVX3W{ly2>qplIi!;Co+=uU9;AFPN=aOMO zx*nc`qkKeJpj8cqY8JgWr0T!vGzBuE6$qUSlG~%}+U4GgBu0QSQL|)r^rLHq_Fhib zcVMj3ESV+!=t{WjLhJ{4ToX8b^U?L`6}4+p+yUVslVp49N7u4X!j<;+`4#6M!8vi` z(X~*m1lb)#BCt|!I38WQMfi}My6ORI%(!fC{pi{p_h)i<5j}zR)AIV!^>=HUnt&z& zo2KQ>M_2z0lKn}n0k~ch^rNdM_5{iVeg}4faaln9=z7sVS|yMT?2(o?A6+Z_R7EA= zhoKY%r%Iq7UDtk6okRk{OK8j(h7KO<*Tn!Uj&I4H;}JM%KnQsqz<^{AZmZ`z>$1Q0 zuzL(Iq$(M=lUG74gWyJUY`t)cxVoX-dFA#!>L#QQ1)|6$=ld|&x)(H~$=32evV zm|q%z>c!;X!0X~hfqE#;3{P2{Q|`r?>1mIw>D*6*;`jis1=kqPVsS6dSDyZ!W0Q1o znA{)k6N79NXRfC#PJpvG^F0fc^NJzQKTsUbP4+@h>x;5g$+`}Ln_1XT73Vs@!xwC@ z-XQRh0T{*M!0TdfLkEiE@RG$@deE(PyuIvxc86}A^AO1gP7Iu7#Td8N@$>S3y0nHa z4wKVB%rMACaq@b};)FSi6X+G5e5pFM<>GK|vV*;>i5ZwrMnG^J3vWP5Xq`O(^Wn&T z!7d71YI9i8!vq-RHQrx);e!#!2Dq&p*APy`IUGK8 zC9n2M@ZAV@ic>T$O&W*aEP5VLdp67g3nlZoXTMiz>_PmCy--yAaw~CIpo0W zBIjf}ayUfg@F&jUDa<8C4u?AY8ZIxRa~KZ18C)BVtA4H=4p%vha^`TPBfQfBobSas zp^fVCWt^19Hr(4Zz4D2VNJ)Pn1IrC#oEd|IMvSOmf&8 zOjO$x1P1JOh&<jJcv zi^I9ep5?H1ZHX??9)ew2xRxr;F@Q;rp-?&IAqHR+hXbz*zjPk33siBM{q9zrZyfd( z1)LXfp*WKe$9}lu99L2I;w*OfpIu@WhslrOUKnJfINzz_#5jwy!V&&l)Tf`hIGmg8 zRSs+0;yA^tByv+8&QeYl#~0wjXdFDmG1oBwqc|LRT^xyNfa0uI#VHI8?J_%TaM%lH z2kHUa8ga~mTf}h%xff@%!#``Xr|xV_-UD~gAREQms)|$4S)6T-@He+CbbyP)xyjz) zu(l>#{qYupp2(4@w7eq2Y>!45=`0mt+sAktQ2&oS0PwK zthSe9fqanaLD@-G+s%pi^W=sfWv}qC&CbD}1wA||`&SR!pbPjDya?0ZB@bJT2l&&f zKrCgCwAu=PjX&Q`O+ZUoZGj8%$Cq>P+{2c627iwK6H3`VJ#7^no)olsvJYh!^0Wnb z;!nf#^(lLVr|l-;&*GBE&NNTkZZ5*wup*S5*VDG>vIoUJ9-KSgtnZckST)Lf)lVm>d(~5Z*$_pill^Q*`3bb-<@$6w)m|eyipn9N zBAjLO1IUE@f?f5eYeZDsjF%{l5Zv4i%J9282yg0pB6y$yiN85HC0Z~3bzgxI1Z|7eqDjASCZqUn=F^D|8Exo~esOeq zVn|3rLCF(d$8i4_oaF#>hj#AMiB5kJlJ7~Xg5v-sYB-ATY(8Dg8CA#706IM*VhBY_ zif9a83(XA;xRgPsXGhe!9HijBfCg$fYDP>roi6F{DuNYE=%9Mubb4;o+Y8>3BIbj) zSaUi&^*dU0UeAD!ovp9gfsg!wl5)5}5{u8$#E2QL7< zHQ|LnCt=@I@ItDr4xkD>j+cONszyTjmiN=^N;Lf3hq&QHGN8&P%-6V|{;9+tYlJ6+}HZ{QDIIYZkIpwoYqSod3HMR8y#1jES!tByBx zVCRl>`bFUI9%X5B7+NR^co~C3AyvjE(CH5Yt3OUA(GbC{;XJBSr(l~J8DbdsCOj(q z&0o$sxLB#{a8(SCyiKRfk#|CNP`+T@edLWojC$nN92rYLAYc=mQvBX9cGSYjPU z@G&;sZRfQ_J&Dta;&Ntb()SI`S$M#bHE6L8#zLk{LHf z-l=zqcujJ_n(Q9Cq{xio&}yV*BE)HzDrT?LO|bX*cf>#D(A?X zd!($Sh~L0FuDQm@Tb)@>!M6cD(6BM`tncI+c~ySOCn+N2Bwp#jsazQ&uj8&_3N8nz zq6s@kUd^S>laQMNYHh;Kk=H0+5t14259kvUc8*?1HyzL%6LyZgL8Cp08&3QP z=qD3)j=bxWQWbm*&}kEPj=a5%)yG2OKA@KdEV`mk=#kf`Oj+Up5rnQ83nvRNJA&#B z#>jhn%|{U{fmGXts7Kz;PwJ`sv;*GLmE#1}X@6wynWak!I4t%Y_ z;URUFyNiSKn|l9#gd9@#bLNnG$RqB5PG%<&iyl&MaciFCGy>o(mVa{%sqdd-lR_kd zW8I((zp8^PlSn~ux&aBhDOxXn8qTg1ErE5@IA_OoNcom2g#Aw-eBnxx88?PhkPl*7 z48mGX%EKuclbMXx7*Ycp*OWZ*C&=emAegzvklHh^sDd8>%F(bfq*PSSA(i?Y-fw{- zqRwFT52vCshSX1)#T8r`P`ZYVAvJnKt|3(w7ql=%bjQz!nrjTH8EZWiJOR*`ChQzi zO>>H<8Dlx1wI=KwQXav$5|yiVKcL@D*g2${{T8N%)D1v)OxQW3e(@_z+;GC9Qd2`20uQfZBG*fpPFlLn}P0gIo}z4VY;pOi!#Ai4k_s5$?CGHco* zP7$YoG}ncwht#t(_zIk}y8-x4SB`T?wZPG0l5!gGRX2)rNM+tEOrLQJy#!7eGOE-V zQd3&@q+7%Mk2_SrFkj$47A2o4MEtJZ-0@(MA%lP$6iDmig zV*fwF4{KSp zoluHO1FCGmsy2)yD-E@KsvBgBT*gx_XTW?Zv2b8sIbc45>|mDf%{8T8N<27TTh_!h zKwmLD*o5&?VrA1pvLe<2+N`C`ml8+9Q;9pzF<_@PZoZUwFJCfo_jmy88RIf~^QFYN z+Igu0@VtNFB^aE-^-GB_s^_Cb1eel~ekrkV1dJKA5#GQJ!%KhH4i5QX=24X*2~g z;vNVu43gVRiPg6&E0PGli2eU?vJQmIj(#aI>Q%gxRRN5enkBQOUrJO9!b^!Z0J~@c zr*FQLIQ7(NaUBW5cqYl-pkGS7z|EpkyDkE@%*2hC5_>Duk=;S;2DaY~$4iM3pOv6k zG^e-%>~F?pd+V1H({~2T?jq zZB5WGB_3T&R0*^P)|GKtK>bo8e^RTSdaW$%Y7gIto zMr|~lM>DBHa&e6j?qD^N^v=vks)q9n9$xv6>ERgwJw0rASO?$JVG@9|cv#U5jCa2%T&o!rO!ohK#IKiWUJ{t>H)9iHzwXCOedN0XnSVyrwih53A`-mjK<;a1Kiw z)5E-FUK9epKvsR>WX@Fj#`MsmW1^%K23$f@WFp4&@UUVar31<|V7KYv@HaNukGlfz z&726%qMjc1O|sBv1b<-EhM(J^Yzko!;dVzXj%r&Zyd3 zPY>e@ddTh~qJhORF4Nc3!#{JZvO9>Xz-nrFb9x9oj}3)9KH357q6vC>7>AD)xIc=K zz{WGK5-=_Rv;Uc)5?BOmnU*)FhqC{cQ3?D4@PH=h>A`n7wu@dw_%)5`ka`{C8-{5K-x{e?v-8Mf+* zH!wjom0oP0j3EkIS-%dQpC@cx@T?cydcr%B9yO>6J**T>Jrte2BGZLlo#~U7`AGWr zWlB@q0ul7JqO(`{Hq0G+1k=hx6a>ssx>9-pdlk|48QfUx!5dXIr!xy9shOg) zSM>inckE4ZaZvDq)@3EVGj{bIEF4W^6`j3ewhO%k(~Ea@NV-SGI&>i{f;K2Rdj-wS z9lJNvFI`KQ^j_Vp^az|RMQ5+5>O#LU2J|KaN=y1j?)m^=i^-aH=Ufd&M{x`oUrjyokZOo%550ZE-gT1*dF|k+Lz9i_>8U3{$e~6-o1R z7v$VS&|`fwCH*`u5j~3BuU2&SimzPg^EhXnz73T0y=7|A7DRbo(b+40cA*bsdRTS^ zNk7s(oxCbWljEi=_p|ZABnn*3otgB$pxZ5~zsTJ?ZLzhrtqVPh>06`n z$=H{@!X|$7_^yind>mpgdNX(ICqD#ze3Gh}%AaK7{H{ovt>`sAKzG zDtB){ZK_1muZo^C7O^Md17&k0PGtJ@76oNyihf>6_}&G)mE5jz;X|p68DJIMYADyH-{@H)A4b0%BjM=mBjI`zJ2+HB6sU+)L6c z-%X}5>Ctpq(O->1>^0uxj=clZV-6-tdatjdDGagaxh;!Itr7bK7kXu;pD!CK=}nGA z($yN#R9Vqqk3{U%ECqAbdnnVt8d6!(JB~@GRdXY$m!jWqiP*=w&>!~(eRX(%q=%M? zpyd$}G*8jlD}Ht4j{QJQKlJw`e@WjswF+H^9y_e)>=id%=qow)WjXjD02LY4rU*@e z9($qa>=o~N=Z<|W(<^L)<_CS!s?xL!oOkZXnqjY)=0b1F^rD44CH?w=y3`ICZlUPx z6&qdX6`20)6viiFe^kv%yK6<$ct!tr1S;~33q2pxEB&DQ%q~YHO+=KN6rH`IfKTqc zKj;N|^Lk20?X_e7A?p3QqF-u(davR_|BdMY-FQ&yx`hpf!DLo3G;wn0O#gLHP^o+_- zdTqL-N_y3=GpP`U$09{%uUOzhPv+R)mQr)_4xi$b7>Q-CqO(`n^5>2{km)naH;}Pk zhY2PPoHvTjUQx@1ezzy+pPmhq^r2U=M--f-yRv53E4sVT4>EnCk44fwZ+X)5I??o= zqQCtV6*<|3zJ%#R`sI=Ir0wM>9@Q~P(b+4GxzL9*y-u{6HK}3|dLQ-vv!ZWrhI)VO zLT|^FPYr-t>S>6W4sXJ%D_+%;2{=^HPp8UO4T9&~R`B!%9SMP1z#vEOr{ z`!oG)$Eq^+dvAm2TyzB0R&@4?60x~s&*rhZDnNCDiZx=XGZe=_MQ5*g*M&bZe#EXzoih_AIhS#SA1VNH$9E%!KjS>6rH`o3r{R&W?roXz2WErGVkB6C`Xg0M$=Ahk&R*fQDtBfkGyO!91WC`z6G<;WjHHo@&R((G zh2E9vU5BZ~RU^E#n}aBSP;~Z+YcBLOrbn+=GvDZ=0aVT#PjHI9=p7h_n>UKfRWl{N z0DVBBTBXK@#?k>)w%=n}RQ8HbTRZ?{Jiuo?|8ahujOOz`_Jr$k3BCc`n%iM8=n`-aJ_9`}Srs)N=Fuitn*O2a>=lbY z$sPMoOs|F;QaSd2(hJeA$icsg&R+4W3w=7%TP#*<;4|~WXb5sp?3t_?_KKS>^xjOr z(X_UV{asvi+Y>oxqv-4v!-nL}Oa{}B)~+e(weh_E2Oj+=DthV=w5qjTe6F#YSq=0G zU#lM3wrK#JYa2;h75&kCTsu?Sg`U9lA=Q*g;{N;$ibVNtDmr^bug8hGVi);+5PL(_ zswcLWp>p8le=dv4UhzSmV!7x|dSaD2^P-CV%W!-!0ZuJNXRjFLLO;L{{>3}0=ce*a zd}sz9-#=D#_6om%+_6_{4*Ihb>OrW@XQgO0I7=0sy<(UPy#Uki%&aae@}u_!?FHvg zMQ5-0#fAQ?8R%`EsdZ$Yo`vZMRwWiZ7{QFiUeP5mcV=RFL7teZ^n1*@dNeXEno21; zd&TE2^f#424-80_nendWNt>_$Yp3Y!6^~u$=a_zXu9u{@d8ag82WN_+vsaW3%AJ{w zO#dNWSs~UeEKT1a!#fq7y`q^5eG1bPXFC_QJ>qB`ICmADy<)Hny$92~or;v1S(PLx z899h}A!~-c;;IXMNiyjEZ`4@5cg;qXt432jMQ5+@2+p1N;Y<%ark4AY*A$^>JV6Xm zboPoqF7#GRuU}BbzH&quWrDL((b+53y3orq{U;n+%|ko0mOt5%;j@a)UUAHY?$7k7 zG!IGNu{Vao!LhxRMP;ve>_Wd?9`yG8RWpsBUySDAsk6MIvsb*AFLzO6OM|{>kXp}_ z!v5uZ&=#E)oxNg&3;j(g(0{M03@$Y<d{R6*B)C(DYr0DDw^<3yvn0|k|(le9p6sO_f6oly)!-Kt|j|;sA z=(NGY_F*5GQ!iPky^qCtge7ohbSV8|*?9>%;vI`mBR~E(>pQ4`1dGoEL|5W`qQ&?A zT)Xo4%yW9;4NrA=&|f~nVtl5h!206Ba+~{3&PXj{{U~2$Zoq#(wwdBYE-h^;J3^zzT zSYC{-D+xA*%5t1z3DxFoDRs8_rQLOxX(Y3RYIB)}YV$Kcxe6~>hP-jt3Umouk)_02mQdrIjm*wApZ{9b)hG2Msgjb=X_Uf$XPe73RGZ&z zRz;@qs#|61>IIS#OdYIC-JJKOx}EtN)@CGq4TWXajWEhS#Cgc|2;0C%?e^rd-a8r>=v zpz=yWr;&*N&Ni26s5WnHRkb^#b|vcOmd5)mq1s%gq1yc9{TecjBi;OIu3H*YSc0u% zI24=9A-xGXg|Tu&$p?>#rQekVo8DzPeqjmK=4_dFws}+orLR8N>q~Ex1Y6&w#AB9F zZO+DdXPZ}U7cJ8$iM{(JEwZk38gcmVY;&20YV+U})puWK22ckjq0?x=5~|H*8mi6T z&#!9tVqgK9>6XUlETMF}Ohf7RX2T5~|JFlD&~++_*X=4>8!w)unonD`-qYhG7Y8M@5U)Dx-~S0b}I=s%*%3Y zVhPpeY`1r|dFYHnGL7yj_34q4VB5Tu_=hD_o3rWO+2&uLQ*+dpOKeowLzY~p5r+TH zHkWCrHs9GE&u%EkcYCT(b0wkEsK*kj&1D*@&BH6GHm@?wmnOKS@hM9v-7eEmx;@Aj zCT67Z??6F6D+#uw%X0j{5~|JF(C%#W8|$&$heU%&8)YjAwvJ1Qt1O|~oQ>zsHt%^Q zUP|n^QIMiMWyy6KLHKXX=bVOWbISuf4k3-OakywgN$50cvV>}LnTBd}ODU{wAZoY&(|{r&vO@Ih)pjJ>}-&DfvuT=Hd%6=h7JFnZ7$PLZGLzP zjF6B>|2Uj#C<&cLDody~muaXr?{8BpnK?zv(LlE}da{Jl?J^DN4c-}`w9n%UCFE*< zjgn&XhAhjsEX5lta4s>dA3}N)#^i@(3^`l>AcZa|3B7jzlO^~G&AG0Y-h{yulwLd> zlT3N-vh;e*>xut<{JPb-9+lpN_SH+sG`v43OKD0%uN9M7f}f9JI*_Ym_9l$jq@F0| zhQ!eTC85_?-C2TP{$T~n^R4tIY!6gV6nU1Fpk+!zuTAE$1Y0NYN}464H(`s1TK{~o zE1u3O2|cGDVF`W}MMBQE(wmTTMAdMo)-mMmkR{ji)LWKNbtUI1Rae9Mt6cSa?njlB zgx2q+@ZVWiQopOZs$|0(5wu739Wm5LNoc*;ktI}JNxdk&2{l`)T=hF0PK%U;)+e)A zLe-VjC#tS`e2!;vq;a%JK{}x%^f)}g5~{A`I8=32)EDpJA@Qy)m^@%v#>!9k-xn;Q z>Pq%sRaYb8lBL9rooQ4{Nw5uF_QaC-@5k>+{Qi%y^8mBq3j6q(*>!Eb?ylXn`l{FB zuFkS+mu2-5okXwEiRdNKTl64^6hxE|y_e{sMG!rNh!#W%Q4+rYd)_nW-kooKpXZrt zXa2wU^m*sZJB=>%;HCwvXAO`cc6s$gZjH^9il7woygogn7-dW014@Ceg@n?@98STI z*GA-eY{N5#wC#2|QtYzXt|nUw6o*nAUD?XOv0RVM9~Uoe=NwKJB3RayBIvT@ccrbj z

V9Zn7=erQ znSrNxB>poiz%@(uLbL6oE?Bdp`+Z-4pdRsiariJyI4{8s$^Hol zeXmQnwgfjFb0_ke{i~}BR%NjQ?SC=gU81MOEQZfz|B9o(M6c!?;No}7!%CM;#@Smw zjDl5NDlWn9dY*AL#VTgk8dL{XEnz0fLyskTeX)}5k$42ML~kHgihX9853Gjak5RBR z`;8Y8U^Nme!~QUh%*Jv^Gwr~{g0Px6r-EP6(y*F~)!y!TwlS<0&cPpu>tTuBQmicd z=aVdF+vo|{qrrCla*41yx?hI1$PRol6IN$ouCz~-jf2%iO19B%F$pUR^{(REVVe)Q z^==Uy3GKGq;OkjS^d7=IV7J6iVlB~o3iF6P_0tSk6D-bPeq=vrSPa%gX*~I0!Gfh8 z@qBF`F4GX!6p8!6PHe=SQ^mSryH;0$HC?PbcFCPhVa*8Py}`qyetmnwnk|;b>3}DP z3gc+UL>@;azV7>C-(lJ@=gHJ!y=72DbfWefJPlr37=ou$OFQ6U1cIUfFV=1IW4KKE z=jz+`v%(`ngHbtnc^p#R6sAB^O%WRbnYk2G8|pK%j25>!G%Pn9(e)9ub)`u2;z8Kp=GlY`iD-we6#LwDZ z!=o+L6V>H%;aU?I)s%7bZZwQ^Wt72ad#PSS8Gj9{1EZZXKHpvjMtfyUT3rxE2W4#8 z8xNzS7#PKl%fwA`VANz><(zU!f@Ovl$Ne-u2n4C>rU5inH{3HUYQhPEuVCZu(WEQu z3T_+*d52U@PHHfmkxJH?WEjp!Wg-UogS;b^pWaD;;f$~To*fOt8D9iRUlBslIE)(OTLny|x-`EnXi7qmD9st`&sg48uya&4kfFd0OwU2qRM&51ZA6 z(NGzWw#2|_q>Mw4ios~2jN-`+VKi07?YJNq&6KgPRv{S8mC<`yJ{T>OQF&-J7%i1? zETlRNTN$Trv23lC@#1e8Fxo1k`T069+9_jb1AiFpm62Gu3XBfQm}l!SIx1sD28+{4 z86O=k1*5YvR{dNPMi*s_c(DeIuF4qjH~~gCWxN!f0;9V!ioe_xMh|5S`?NNUp30cC zFcL;DWh5mRhtXRZ+lzC5`Y7X2r%)JK%GkBO5{$mec=&r!82ywHG>9?%mGNMCWf%jL z@%yb3FtU{~^;7P`KxJ$loCITtGKSO*fH71VK?PX7!<13`N?91gmC?Kpw=+T+@8jBD zrH@ob)QKW6Mk%9slL#22mGNu0Fc@Q$vF@!{7-N+Yd7Z~}oHAOUD+6P^GCDNmi8MhO z{jn@| zS-Y^d+B@QZ`)CFX=K@?aH5!IA+vwY7R=fjv_(J4oY%v3o7X zPaQDaNSN*n($?Z;Q;h&muCESyjPzA5J9lFSSixqPPRij*MOBOzFBqN0Ia$ZoDO*njr$f<{9L^gs(Y3W`(m7t`r6N-Kgn8?gE+hS%(II z-j7{}t}pA_#`4;Fy_eiE488M1JzQVjEqw4^gC`BE5(!k@s4f>405;JmA&2&an0s5GqE zQYCn?@(xp7Y!U`+VL~+o>mh~N#~$~a_Mn2c zMGpu*g+v)LGk4CLH$%(%r=||COSmTtYn~qDntm565iz{CC;Y*?iD20y&EACvOY?L~ zEMNOReokzjb8E!gqL~Pqr-%4lMsB?9f}J1E9M$kAyTI2dn*Ywbj50pFN64Bsw+M{# zQV+lAC-A&q4+^e^A?$BHy?xWP0NstDTx2WV+Kdzpd@rVAWTuB0SU2>2N`PiGcG zt90p52*z!P>cEnXH8n<7Cq3KjUj-qCvqERWM{={Z|m#K1TAvA9@+rJxAl@w zifWNN^u*oBs8!-Ri^#3IYD<1Feg){N)Z>EyE%H@01)p@%*fgMs7Wuj!)B;P~bUh&D zq*3t{GM@Az{y^IFkJ2Jfc@aM#%Ll}3kso^zUm#`ilgg2wdJ&(L*b7^+@iTgmPeVU# z3yQiEe?~j*TGZoNCgb(U27 z!$`x|W+K1Q3t6P}z7|FrL?y z{L)(eAFi>quGMs{-tk4nWdGyC7TjPkn5jSyVxo zLZ{Fg3Hyp)SG<#@F9LOim#(QqKl(N86{GC}y5H+jkth6m%#GDb-U#`rv0LMwoq_*L za)x4KTB**c`jWo~U3^+Wytm-Fm*Y)ry0?f{@?KDp^TC=wb~Frss*qAgrPksom3$QB z(>zInOTb;06iJ;}VU)jC@@Y_dLNTol&}PpEUov8DqF~n&qw-hLnuQn8`oMqCx2+S6 zZwsOrrCr;DwUtqm1D9bZq%xykir~muOPhn)q^KzcYG5)YFs<>!G%aeX)VV9|t8K|J z+^jfH+Qr0p81XdX(*FChm=-lH-?2IcwWPET*DWn-y7di=l(c*piA2q??!rh*yU;m7 zi<+5!W2~iRq*b}lSc{s~VB~yDYoAuWUr8-$c50ing4*D;hBKqIs5y~V-GbWaw98@D zw5YkNXGu$5URaBom#;WHvo;&$3L(Qn?**`EvoWSwZ7piC1NZ~j(7urtwbTLp0L;hl zFhwnQ0N+Jx^pne*XsP}lZ5Hamm{F>rX0RL4G;2D8T7y5Uw1paUKjng9A_(VOz!njZA9}VBcXYt~1>r>!ibhLJ#{*~~p z&x!YKt#P3+-hD;EPx@uQk*w+NvH?{F_Dq9PIbgvOJga^cbWJZcvsq!yofg@nyPx(3 zwtW@oFHE$L1@Y|3S4?x)3|W9%kya0I1#0tSHGljm_qbjIdur}lAy=*yQtrZV70YpZ z#?&sVxf@2T!@AU^*rZ+qu7+N>{^5NTKyeyJ%(`kSPIvgTb2!rJdxmO>n!A;|L6Zcn z`7-300srD0j>;pPHtt`SHquTm2WJ=j`*S!d4-bp`1=j^k>rZ+aoHOu$>E&qSFdg6? z7SwHmuUx_UEqMO+diBFnAGM;BH1~+0hKs5yj%z`z7L7lrx<>7-thq-89S=%W?y_)I z^}3xDpC3up+@phb;uFBq543{6OCHXqU=(yr(9JES6=xLuQ}S@0Jc`rYV}qvW_bAR9 z__yWZyt5%%bB_y3>Rm%|4#WRR9!_v%5zRe5XunxPaejjTk35`9thk9*__9o`-9uDd zz`|H9AO4)({>u&bBx_gs#>$-tSMoFWrz=p3$(DbgX3AX;t`=UmQ!9VA!QQ-N9mXqT zvW33z56{DqQcSS|OH@~!`S7pI!`X@}$34~hU{vYyi zwlqPW>DEJxC8Rw6!0)@r*<+{7@8&Z!_Y6z>I-lYcgg+?{=gfHInQ7JgFidf3!QUhg zXA!G!mQ}~EisJNwe{dd7#EWjtJ=#7;otW>XWX#iLpArZh?>cU*76a~;?Lm!!pqUdapi~mWp_wcuxdeW!F4ak?Rl*P zhQq7PeNQ7a_BVPkTC_0yIoCkDZ_rY%vj%Jl*VrKGCE-fRahK{`shZ~A7!j6k5nxU2hh8Nr@34&G8Zj3s2; zHeO|G_Lm;iY-9yZuMKb09ERk0T`?Z$K~?=w&YtiNdY<9`=xIsK9c=i0KCF>CzOzAC zNYYAfqsCw!TFVGcORKEO^x{tV-pujm|L5mQn%iaMTYVu$dx#qR2(B|Z?$bI~Te_nS zi}etATJPU$jOKP5A*>NsrM;EQYi^Gb$%epl+el1BH9pkiu9Qzg$B+7r#~%()bpwY- zYO`MbVw%?M>DC}^-JLp`UJ?Eb{E1TyR)U}We(u=K9*o|8252jouulo^v7`Cy*j37X?D}oRKv_Qo!pWSNHVAxn>^gU; z490ix{*uGUCga($`^CSkrayt#M4{AnCC9U4H)wXMrpLis>Ul=av1`@GUmd%8AT%av z7`Ksg>?V&dqK;iJ`1{ugtj-Ax{&*=H_^9u0B|N0Zo zill(gku1IQS|4@AjDRl^*wlnTCSK&@LkRE=hXT)R}883M{sGV8WsM3);en>Pp$j=au}G;D7X;! z1t&_NUYjCMEBxW2p{-KN^%B;zp{DnOYcMy(iq3Vt)RpT+j#lm6RkT3&6C>aY@RFnv z)Eg6YYd#@!uqssF1;>6S*@Wp}HP$@fBR#+F6P~oNjP}!3w9==^;Dj+w2@VN=(LG0y zq;PP-)tEFWvoIawV74S!%+e$t>5CeHYzUn62}VTF_eqj|SWX5Ry*lZO8DYa))Yf>esq}Q_ zP9JYzVIvk$Mhz^Of%Hj6ps^|^eJLZN;kfqUbXbGoNuL zPu-B3Z&lUY+l-)gMbb563-F!ZoLWd|rwW>TyAd?#c6nS&z)wE&wK;`B(+(r(tw#Qu zaS8Z!uTK@jU+!mi8bRGB;a!0>DP zP9x125AHnj^Lwx8+?BQ82xe_Nm37Dn-SlyqqVGe@A;!f$tE_j8pq8sqA?JW!^yYLb z>pdgrTvT<$d}QplBz5lgcqJCIgRTvH5}0@aTd*e+{nkx1gffg+X!C3AkA!#683v@NhK}4 z6>w*H|;Y1b)ry`=68WQzKYo zy(aj~E2OzUF@h`N{ZGTZhX8e6er|+T`mwsoP#DZ&WOsU2)#r_%%6CFEqZ;s9^hps^ zZGC}rFECuyRwwv+dt>$9OyE?+MW-U1aoJafJf;Y&#PwG0uZ+U-q#S1$K0D+7T2+%r zRnuuBct2BO)Qa-u=xR2=hOTUyr{U zD3Piin$LQDwq-fv?i0LUk6W&%ijJjkz8*9C8V7vX` z9g)M39M5|FiY+2FeIC3oKhMbN^{;2vlBP`G1HwBbHR3jM##-~rI%BQR;rlAbpVRA~ zFA^rb{vEg;{vO~ZYadktKha@^jzr?}-n0H zjucj$@8SP7hok0*Rt>RIp%^Lq6(sSjuFY3z<#iZc)XC0>r^apstvGW}H@5%5pW!*S-GDzzz~IIG~_nup`eJ#EbI+%vxq z|3`T^&fHU3A7^gmD*U(daGbfPmm{6IC%-MY+VPir-<`Rq^l`DOJSE_-kcZ>UJ*E3O zLMqP?_{ZnrICD=8k0mP3a`-pq;W%?oRW3R6i3j07k%!~VJ=txX z`Mb;T-^jyp=AP0@IP+yXnlB4~x%b_fdzxd_SLG=Nf6DV5@7z=R(nK}))DZsWUQW*3 z(}Xi2ss+h{YhaGsJNMLew7;5rnhDqZXKrWi>9|!%%{^^{>$M!WckbyyaH?F1C*b+g z>y=AD%{|F{65DG0Oa~gl^bEPj`o{=xzmjv0^`Re8_rFHSH=o8}4h$Y+4O%DF5Nh$& z&o#!^YXahdr~kOFp6XlAY(xC}JD%%X56!Es>CHgv!W12H=AdXk>svY4&gom{AH;}V zp8~@CoR|jB%|X2r#e43p@V=SD==uB{)U9SsQCN6C&tXW8XMO88`LUTB@c#KcBd2d| zhgC$pfz$J0zasEw^m}d&>gxre>W;cJeC2ceIeqIkyr9cDs3vgPIqv_nZ)J_->056b z%_8tb1anY!dyF^MBFT3A;dUDDoH(^fP0c~gO~{Aa3iwZOW31|2_td<*!jJS|8xeOt zkI>wgjes9OOMiGS0XHoJjDQ0;Zb=-UYx-87iyl1R7-{&ui=Xe)^?chAg+F5p_di6O zqYS?jE5bCr1n??epBgshy0aN(1oO5>mj`9;F-GY9n&Cp%yCJ3z9a65KG+E626Q#>8*T*G#K2gWCu#0+M!wJSmdI+b^et*8-2>FBpu-E>GeBFRKH}H-xrEgNhX3YJBK7Y z?=PLs6vNcw6BTI-2yZ-3lChW^LzDYi<-xBwLKutn!@O{wvDioP#Wh|KzaqnJW@*af z;2n$oQ_8IdTY)=qcZNUZ_Kw8{%#Bp;;&7GDaeK#NA7)r;oK+vLmR|SsW3jqjnkh~{ z_($Y$o*Rob2+60$VhiA3mBW!~4E3zAjk~l}L$#!P;eS7e^W0eM!f1~gi(Q2OJ1@sO z7ONbgZsYX(@I3W;^;S|J&RDF={CsL~i%NdK!^Yu~g%5c^8x}6lxSZvL2 z>egEC1b^>59A_-n2ER=&!@^1M&&k7a#$s<2a~@l6f&Yy>9A_*xW=|uP=M?JamHdJzIQ9mV))nQ;W%TlvKyT-=6?8(me&3z3*x@MW*=`Z|lk;$#u~?JZYGzQc2Y-t^9A_+cCcL!TufFgP&%<%X zV)rg4E6#lQSLWe3W3d)>QWR$o{O>)_@s7nJch*;9u?z5D@^W&d%V~AgW>Vo#&%<$A?6{e_ z;TVKKdV;+vv zVkhl(%KR_<{(GO_cc;Zp-RQLZh2bxmhvT%^CFVH2Oda@}<>5FjwojO;_Nx#4L-TN) z7JKZj1jU&P|B5^ur^UXw)wvUW6aGVaI8KY5u*m6&zJUMxJRGOR-nYQ%h5x4it>^dM zX|c~Ya`vkL{E2xuPK&+ntCFhBHQ{fRhvT%^HOrJ&oF4EGe4gWNv8$Y~pjzyi@Xz;h za$4-sX~k6wvJI}+a@^h)`(gQJs>MDD*QfuFwl|O0sd~eQ*M5d`*4}4!=JObi42NT$ zj)P;!JRU!h{hXoi_xJwZ zKi>2C?E9?ezOHqzaj(7B+H0>>0J|-Ar)AYli~S>@U&FB1VqctV=DK`nflJ^jxakYE z*nF3dW4h2-Ca%8UR;fSvBCCU&S8{17423b83a&x>&ADi0RrLl_wZ^o;V5%p0eQ0+d zo2&3(s@7AO>3tabXFcq{4W>AKUoe=8eHWXAf^IE^SVe!|(`I-u^}$VT4gMvt10HVC z^}hyF^Dr$?Fqk^`s#Yh!w>!|A)&25FSi1%{KMn!PJ`rQq(C#!nzUKHuNFM3=gL2 z^iB{%)?L7cczChF)atf1jFD$S_>V^_7);&TvW79T2-tfbUTiQ`Zjm-dc7U)yOp?Th z2UF8qCL7W@5Uvy@$zV!~&}KRl51y*5Qh&rO35qtaXIpGf!-D>GEee|NW(p6RPuHl3 zl5PoNdtAP&MM`?yCd`QRf&Kvx`)?)97hnY?z4y%JnJKd;lDpD*{rCB0}@MT4&c zw#maSy8c&5kCLjy1*RM}pZq;jeGRsC8REKUpkUa{M28Ex?%pOi=?aBh5s9yg%cL+| z$T@qOhz+YXu&y3ntdM^>Zy6(lK^W$d3JUqHGfj<==Yh@e@M490Z>KC{WH|^M!X!y| zxRCFEx0)gC1L6CkBq?MmKATA_x9p#`Dz!@F6|nMtsGnx$wfd|_84rgj^v>x17Wqj3 zI(%b*%U7GZA+L4$vtV9}FQ~)wTA3{x^8N>*f7nkZi!uDpXvzKf-oY9T^tmu2myCjU zM)Pi~sjSz5E)FxqM!~$+3x5Wc^)b*LMH%6Ft<~A?c7kFP{dBxPej!+jcKI_W&Or<`n4y z-;`;YF7WffU-UR$7r6IHcNOnqAgesosw(;6PFi$)&F$lU0ep|g33q`{t*LIhlcxdw z<-x8Aw+meLm^;}JzoDcG;gTG8Ic^tt#|ZP0g;f`L(_$RA3tR(dJ4x5I2k^U#aojF& z>N`n>GaC4F#W-#k_`P-R3$3}p-zmm%yTHLMZYOOs@a-Na+y(C0yM^h090zo=0Cu~; zeXymg4DNj&;Qb%2!YOvUz_0CSWH=Rp*DA(wyTHHJuWmT)f!|z=<930otgc`<_XB^l z7{~1bzdY2Pgq{lgwPGB%3p}s3SK6|yW0iMTI+U`ZvbAc7{~1bFR$R{VQb)Bi*ej8aPz8eH}GEIBZ_g{F7WN6(oB3N z1D{ol<931X3b_*!D}Zk-#&NsA>(bnpL;HarE5>oVz@@Quuax`Wz=IzaO}pC#{%)z8 zhoym6F2-@Yz(ZbgUuZQ4-l-VJ>jL+_!Gtvc_^_fJuM0e`T=|jToz#!qzxiJ+cKz-x71xh;ftg!q>UOqJ5%Rf7oruh51}M`e zKLG6EFy2JIH>n$u*}RiGZ7~Jdj4*EcJ@q0gjQcd*G}_C7ZSZi@Xd6;)M1|G|s~FNg z5WWwSq_vHZSNe@2D*w?T$~4=5gAn}Kjj7xGyYKzq{@HzNR~mTbFsDef{h!BgG|hH% z;B7sQ*KGGH=eD_hfDHCf>#oAh_Q1*crX3s$e4@t*H`_1vbieVK4``_eyC&RbyKO^v zZg(5--NiU=vz_pBwh87G@IQ-j+-5tbL$cw-eNs|o;F98V!*ZMLiFdRyoI1dp6yvzf zc3R_#hI1?MzQs6hv;D$4cjn?T;7@s+aI@XwR(Eo8HlVo$u-j~}Tr>4C;wqeCx7mJqj5~=^0eH<~9JkqC_m?|bZU?-3 zF^=18e{s~C{{=p>7{_h46U(}DyDtKNwHU{3wqLy-YnqeQz&|d=ahvV0H@fR04gmkL z7{_h4$G`06&sE@hQ_-}$&GxZz?xU|P@T$c)ZnJ$C&PbJ}v=#6!#W-%WeaDs(hBFBG zL&Z35v+b{4-*6@Yf4LaPZMH8A^wvKC|DYJhZMJJSac4>P0sp=j$8EN|WVzp^Tm~LM zJ65>d-DZ2!4P{JxO0k_Q%JG`*FMHNB&GwDJ+jyLCvpuSFw{DW+wxRvUCH26Kv)gXhp5eCJb$~bUIN^4CUQ@T-?gpqw0qnNh zS>0=ydukY<(H`uE;I`ZU+*ZYKrURc_jN`W3QyjNlS`U14F^=1AkNU-(1wIV?mtq{Z z-G0|`yJpsBX#a5)PO;l=KR(Irx>f*Qvlz#1xBqJBwmI#9cQ3|q+wH?IyCaeNfR8N3 zaog=+SGRq85%{acIBvUrsF6EszZ&?*#W-%e-SuO4joAU9(g*F8?S-QJFE(WC`Q+m84D zxJ-lSwc8yZZe!Z*`hXf2z;3(UwPHQfZr=i^R~Yu%?XLLbig)5eK*oBg+yd@w*YQ_7 zsO(i*k9om-k7BI{YcsCDd=FyljJu!1yxp(9nq?X*0x@}Q(Zx3PR-6%@|CQKcQZnK) zXaThP|ME{DMm{<;{tLc8Ooz>efR{ulHu{e(s-nKNlKbdBJMo(I?XlQJ;I2@AxK(bAX_iXv+x!YvXy@i$ z9@R`G-<|XA`Y3ER&|(_4F6h^7)!r!emf41&f5Tz$^Zjh!(Wi4%@_-hf)Wpdq)@ZO4W>w#+HDyeO3iWCEr{A#NcT4 z7Jj}xzOhOkEITYL$Ipa!vE9KC`MC~1|Hy2plJAp+=pWKOcJ$ zyBa(odjsyWD$d{4P9+Zu{)(^In<1JVaQP;JX_jHVVlflyeHy23^8jQ|Yh+KkF%QS6 zz|?C;z1_lZNX=^W~=O$_s18YY{2 zBwLBHyN7?6RW4O!r&#Hax5maP{$T?SV}p-WD}%dL#QS6EH9Ot1EgUuEZ}0^^{m12d z23)hR2Yo5SDr?yn&(%`?JHfa&%#sg*GR|o1yHU=HefF3cP&@;CGC7lJX%>xSy!2-o zm0jM7JvPM5C%p-LMKKQ7b!Atu5;wdTt>(bj4&Yyt!x*UyHuTw5t)=FLo@7iCI8J(SA1s^Y+r+R>XH!Z#cOtMF1a4?)*(~4btYZ=85MgxD^jzU?c#tv!IPwH8EKt%xH5T;N&=jL%lCRAs8n39 z^PN-;s2h5awSf3eA8XA5t3t=7aZ&08tb;C*uX>?H2-NTXAS%Apy0T62g>rVkK&ckH zV7mdf5Xl}8aH7tXlLm1dEKaA@w_#AKt=Xq0X`NM1We*M{ZtiS4#P5S-&3Ya}+Cr(| z!}!%m_Wgk%_4okD&(i_;o2d-u552#?%E-BnZEK*Kl3Y?PM;Lzs`o{}UWqk2^C4paU z6sy(%PDUg{xFkN`^1@UR1NIrb+(-3PJL1=Y1rb|`6ZSzFT&bfNn9{Yq zdM^QlOn?=`gz~dSBPS5R=cf>86gL23bXNODmZ^`TX5kM1a!&;nU@iLy-omwPeJOmw z7_MbY-ma`h0e+H}_aHkws8n2cyn>q!^;{3K4iGP>WgA;{C|t|@rk1q>J^Adu2H556 zok;z{?vkn`e`PO;493lDs7?U>ix$({p@x;Gkjj(89gee?MV6#W<=omGch8E*^n2Gv zE1mNV_Rh>+9hq9EHohdzab7`tvOY4J57xFh+Wbo4uP!-Tu=jNK2a)o%d5@e{zBrZr zk@?l}cH@61WYOpuZN1yhvlFOP9nkFX#sS6f&JqXQ|Upn{UPoq?5w)}|CW$Hq&-Ei)6 zJWF%myIHBw+}NjJp=R#yi)|H}AH{93a&reaE~P??qF(~iC71n<(BgF7Zav=c#UD&m zp(WmKKexT%`#!t03N7gl^lERnA1F%ii&LR>uEGz+-`I95 zw82&Qpvb@he4!6rg$2bo_aH`}xC#YD+4W^qXtS$`?3r8c%?uUVmRQNBRIl8_-_%l} z?egP}+&+EMROk!&(Fb|ce*kW5*};5Z?zj)(X^)iI1Gx=)#H!GK`7tt=8v}<9q%e1% ze8uYaNVIxr36?^k6eP#JGMa`Ws5YPqHW&{*r9T2{O5$m=i(D?7)X+2XL+7%Q3XK)} z@o)IPL(K|}^WvZOhVRsVtwPVbiiotS7R`pn2T$!Uq2@5$4y>{q$Qr@QYg!0t4wRh> zQnh1~vHM==uuk8P`obTfBdINe5$Y#^r)XuC)E7zkafyo?<)7VH@`etblzpe-W#6gL zY1wxwP4=A%{U$#$b7yS=?Y#V`B>PT<{*ZmAYRbM-p-ZywRIcnh6}lq(PBqIdiH!wA z*PO?qN~yLxO(nG~3ARg4I=W(^fF&{NvBh_MN*fi5^mapw?6bvpFM3)b$J0bW)92}Y z6^ijRL1?(oR4Crl1fbzYQ=vppkDEetT8YmwWRxTU)qS2r7S(Yjt z#)?M&baYi!k66*@cN&(cx_atx^kv>9dDAQCtcH46Nh?99&NR5uzs*Y6Q@MkhKz~xR z3q_-+m2hkQY_%|qNi>i&61}9=_z5q-Su%CMEtIqqr*qX|`dj$#l6h}{qb`TBqDkAZ z1s^KIP_d*vzpR9+>Z!v?8!ILA`9N%<66$9q4F{n*)8{6wzm+g{W=GYJ{tBXj=+mDU z{cX?KCi)jZ3s$TMWa7sU55IKSkaRCwSQ%G%2S7ZO#>#^y$xy{_rE;l~z%F&c6v-_&J~36;(F; zoZmJwPW1&m5SOoQA*fW`-Lu-N$DtnYK~_HTg6I68U+Ncr&O7Eg&!|?rqk3bd+mzA! z4`M!rEo++_{OhNup`16wMSR;N{ldSO^dT>^fM zL3d}URji@)iFSIolN!=4q*7qGBrb!#AISPi!stD(rJ4=T^#C;r!&TVRYg1ki*=Ih~ zya~wN9x7RGdb3&V=e4Qs-+HK_%o02Yj@1XD73$@cV+RWnx%w+!CV&1!6MW*(`7@{&RxYbHK(*CVn`e^r$e0suA=yii#N?D`@|4cr5y%_(9lF2PPe~ZH90-(p z6Qg)-`r?NI&g`vb<}=|?NmUk?ZxJ54g<@~|;(=h|me)&5^jm{ujjV@;vUp!SIvC_G zAmP3^ezPwa;C4+J{m!>h+0YCdQ$_WH@26nqIWIhwQPzd8mBY~+fad~Q6ozM5H*8hn z-M-RZjRF2Ku&rUdUz-*Il^qvM{PcVsiKPDoFw?ro*Gdug&UPw0Gw9^aC?j20zl}Ue z!Zq1yp?JE2N%xWSaai3Wyh7uC z(&&;OBm1E=ZaBfr*ejK{=(AcVj5x%vP|!rrJJ|yS{<18>&V{@V+u<8v$ihr_SiZqVj#-%H*o%- zooM`05z~Aa7dhJ5aiWHLC=z%ad?n*D_;^bUMC+)x7dBQcf*{lcm_tIU7}z@or9g8bL?Vfa*{R`@?2O=-6NuMoW`GeHqtH7%I~3&S}xO zC39LGK0A*|2hg8>t&Xxj0A-s?r`w&v61q%b%IuR>48tCcNUYr^OX^M*loq#8{8t4Z9TDVyW8n&p-zD$E^iKtq z(Yj?wW#jjQy&9>HvP!8wFfss_uPg{Euy{u1?LFJUYWIFO&DSTqd*H4DQttQ+rWD)@#b0W}_NM^9jV!nnDY;C&KC zjv2MzPE^@LBb)_~V2ptQF8Aj*NbO zMV9&o)YG`WvR-{IT}4&9v7MTE;RYNP*zQ4w6B@vl>&(@a-_CzSz3i+1+^+gm8CfVxc}w$+%@FTB@#qZ^h+% z_I_lw+Y>4kS0fOuhCuy@2U)X=c6S!NpQd(w6YWi&#dJoC=;o8nh~@?EX`lkxm;8Fd zac_3%n%`;I(cL`ab8rsO_|ykb+nB+o*I}l|?g|8DPhH<`Gd+girs@v5{~;Ce*4ff3 z8>-=v=}d5G8SmF%>b}#jjq(x|l@ZIl%J5_;@R{7WY03G<-i-t>uW~R)D?3 zg!0v=rK{{K{-~c;V<=-Lk-+{0og6cX#JG9)?mQ+oH>QJ0?8qMrCAK40X0y4l*1+^u znt$zoBsQCOLGi>!3>Y%>E+n_>`YefU`3aF~wOXnYo2tkY|Nk~s zjUFrL6F*)xlW)cN54C{*IEl#rl23&d#0y%ue|ouZiyQxhsW4o=qr?Bz!nOF_?5H~s z_Q~UOq#AKRNIvCKbs)qy54us&C+*sVX~vvpkICXotb%l-F% z6>34wNML9@C!WTh&+jur zxCAsEocwfvDUE6>>l*-vJ|=&W>A=E1{s;ku$FCDv1WS7pP6d#|DqOcx|dE&R!@^!wCNsovAlZSQ=6vSJT*8S z62l>A42RJ%()Oxq&-KTd^W9#kb>J7=q#TNMo9e6>Yt~L2ftjNwo9mZ$zdRLOdL=BwnG>yF7~Hu>QmM)WpE za(M&|am;LPd6tr!%ZV@Y^-7#Sa>A)NmDFr&EWSPPTaONpRz2}>Sp|P6ciys=Q}PCk zPrah$GbHSHJ^jA;eQoi(ANTuUpy!NV-sINJMf~z{8}@sWe!p*)BYw*tOjewusEQst z!euEKaDXq^G}Zj@Ks< z&8khUucC&PQXB`sDqA6wdW%{av=qeW3PMP#8AOYTe3(AJp$sTKFh1pO-u-Ohd|?BWYrm zlsE+)cqF6WkBw9#G%DQJNh(KX3%TK zyW!}ctA*(QUJ3msME~uYZDo>TZ9!<(7yU3p5Uv6o{3rdm6k}yX|JudUiqxMA(K(P+ zNwwy;lo&NTmnAXEIPsO0~;NdkA z%!=s`iiL0rn?Yx^itO01K{NH|$SCyGP*rX*DH)3Zte6#euT`o*eE_Rae8a7nwJkcSx?%NO)|k?bl<#gUmG{*x zp1$BrJS;ouJimhS^}+${Fh7!vqNnDmq@Fll(bvu@F>HB5iQr2fNhQtt%2vLnR?K^+ z%{Sxg2+0X5d_I{s{z;u)Y@&SCte9yf%vUwvk|emS>8RW=xrOqjSuyMX(@I?nV{#UZ zm>RPWJ%aAhcmv}*lTvqXdBq6s3|`+OyHdVJ<+bU@`aMJ>=x?7yOj!_^N{IQh5N8X zW1p3Jb){eN_^IjEpMPLtV767Ud!k?1S;5*E9eh}A;65CUi?d4F()TQEr|M4fUj3mJ zUK)pPInzwwz8|g@23M8a@|>G%n$9<4o_>90S9s3wA&_m1?TWnPpDTs;#P! zRizxBTD|AxM5@YZG56sZvDTi}ipaqW{-jUe$97?V_+nn)Yr4^Q5tmyl>5a+w0`(Uw zE_;27%<+s4LzVY$!|_LJd@<=EoLouuxH2&IMes1rw@G^5%H+BsbK;Lm{`lh>I0NRo zPp6;9Oe9}eCMmxx+fuTBD1N0dK%HZy*u+ZnSA0qOOS1gI%xUe_9|+?LF5f^libs4O z5%c+j)lQaBF)%B7)1_7wc1~)ik}CUC>v1c4HT>nW<^OY3$O`-d&GstK0a z?ViYZ{HmV8b8RYhd$L~%{Hh_pnsw`eJzy7FHRU~VkAoKDve2p>?5pvO>UKnY7b8EC ziF&|bb%XET5~Y5EI`)DKnZ`l3)?%5-mb(_ul`(=3WY2f=U`rejQVu?w{~vvepC^4L z$WF02UnHp>@Gvm!sn%ea7uI&aVci})Q!S*%fH5J=8pDV$4DfV;g;uW;&lh|$^P8{oO zMgDvbAr=?MfSbGO%f3e6^p)d!>FAd(M5fSxbSf3gc{E^)*p)~k< z1mBfYnR=5yI0Yw8S!3XHG6Q-SuL=r{Ps6N>2QzVBg0ZS7s~?SpS&tnmf%_kfgGE^n zkQE`RLx}AzyHkZ1!TOubl?YY!8|=fb%fK^kK?PEK$rKcX@CCl&w5md1AD4;k zY1V*n6t=wH495Tf>=7oMFG9HZ7WTckAK*hC!QE%zW`CU8j}^_#fyuz86~lY_V;^ne zn9od>0$cClX3@0SS-p#2c2;*cQt0E4Ter2mFs);tpP&~>tl3Pkt3QhM*WLeOpg*R> z2knJq#UaO%aT)eIzBLIrP~s7PvUH{>so9r3>1^;V9A9Z20O7|lNeav0=YsYL;Fo~~F1z+E zDXWxVmkgD9KDZ}U|q-M-Lb72V30E2dhQo2+KY`$ziXqrOJG@) zSQajK{AbGHtDx8euTtM^W3QFr$lWhxrEK2nyMlb(tyAChM@~og#l8K<`{c}`#r|NO zr_kh%f#1ove0)_{s3Z$$6>8#4nuVaOE=rRuD4@N9CAij3P`)clE7SrY%zxIjGI#i& zVEsep_JS4wc*}i}22Z=$8Z`hk#APBZMOwhftAsY7E6`qHhLoqvc*h^S zc^I}CeGup~VTP>yk%i$d?LB{Rep)R0d!Q^VLQ^hdjX${aORT*40O+|mgeDz*k>o|h3n+*tt+EWbUXTq*N;;80+raLHhwKioImv}8v7 zGp)t{v{t+J;GC2Rz$GGX-_z#1-o*3`=$&@4B0oU~)MifQ-Pg1nYSyp*$R5|iH7l0N z)YJapi90#}3)%@|@PL zwdn-bO=KP`s8+x)|LK0oHWZlqVDUou<1VCo9Q0G@Phx8xEFNLzN3E2#2;h5Rf@vg# z@S8u_7ms0U8^GORf-F2UNxkR~Zai2K`~L#GRD@s#6o&P;Klo}1ysL>tex~Cx5s@0z zLBjdRADey&r!Lh1*f2~u!DO!UOMSWK558SJ#nhLsWJn$;dkOM*t$$-rRHpX#BY?RX=!J0XchuqMp`Ss25?hmj z9^r@YF~P6|;My=jGH`A{7VE>k zGdZ}sEeo$RaQPg>w$LLe)?>xskWMQLjQT}cCXYPUL!Ho@=>$gaqO3xBgb7feolG9h0_NtC7sAwfsO-z2f1my&wkD4}!t@mw1@8iQC`>Sw%pU%{d#q90@w}Q0#_L5{?!D~VIykfw`hQ?- zF3KvD%?R<_F?R#WBVheVX2pVR2EHMwqseB4Qv$qu*$v^H5H_z$fxav*6I=IQb_o@$ z<|(T_z~*6sdoR0$zZat_^Zk!CB(%Poh8^Jo&aTdQSk(sDI80!?R0O}kNIe-ySU9nntWdUY z1)(=dlJOBwREt-s-vxBrp%}+=UGErRV{xH#$*%}p+Rrw|hM=|ny1a^OhY9qBpuA&f z@_nJPc3>yE>H7nrr#_D})^>uj`(M`juEOz=pdG21VXR#M<#KUrvESHLrJ4@wj{#a5 z>Tfw*QTu+XDaRR_Y#71nGt+tnhcAr!61BNDD1vx*0VnhGerwtScfsjET+BXf11w~orF7Q}WrlE$t561SQtU`AILfkOe{qX%q zuzn>oui!2K-r#w6p;kmh_%855sG5Nd)k{L3h0Da&+yx$C<{xvgAjfT z#9n(D`E&=sL1BXQawGVKUh4TkboL7^U!Pr-nWh#uv8vH(P+*4p&DatBTVG!c}`^|PpR)b z-IHzV$}(W9iA&5(U5O7m%YS$4%2&Yl72vY{QbI6P;m10v>?)M-Z@~TtwZd4E+K3ZQGlD6qQl5;0Is!7IAgLI2w7vr`upfjGNlIJkk7QfPR~{^O~P<-w1;>VgUgWfN^HOcernlZAe-~hG#qdJdGYReZ0O*I&pTyQw43E&GGlnP60h}5p zn2O;M{5Xi%dJEuEkKk5}^1N zu*Ior!)vG*8X294OZ*pZM5+cuwFkMoAyxw+H%uzjh*S%fNx)lrXALZM1EGgUQeOvU zi?8azGWF9t8p6;psBEsb$(!37Ut6Zu3PwH~Q&8Gerq&4tVXzrOepl+s2_~p z`9P#8@(mz-LXt$!6nSpY`Sz4HmGuy?V+FVrd4pi+ZES$^+gcRV6<`s_aq(Zc$QuSj zW1e^STgwEYN|;or$QuPir;asH{nx`%D-b$(BxPEaykO`KtGvPc0~;L1XIiyxx>02% zO6HPx_w)0>yjDx`y-I4&U}Pj7h2GsyWooZrumScrwAR4iR{CvOtUh?G{>?C6 z`VNfKMOmgkc&vH1<|)e;TS}F{RmfJM`hXBOHNa*QoENVOR&6pn7t{yfZ*_9Ms zw?O`zFIP*aO3_cXc^WE=db8~q0jeSe=%<=2a45h68htE&7Q`C+LPddg1<4&I9QpiX=# z6_3mJUVGf_KchnfEHxOl-BDIW=xceHwU&g_7P~+BX|Wga+};R{)d7TCijq>=(nu&+ zVIuIsz#jH+li$zz8Z9iL^2^3veP0_RQ$U#Ek?;mBs(nM1UoQ6ZhszjzIj}W_@D=ej zRepuo7aoc<_%2}k3*q1WR6*ru#ZI``#Ng+E{Tar$`XqCzBuwq^4$QauB!{Xc9K|Ld zQtNjImv88#@@q!*F4GBb9~D-ABa2OY(S3WHRF&T#@$=l$Y7_Dx_toFqtNbQWUm)Kx z1@SQphLu%kG>N>NjONI{kwd$T5z)S>Ff!HRrZO2LV?VsRqIwnz-@uMAE#HAEdjJ%_ z`1P_hm6mZf%2%mGDU}k5Um54@!9&m=*#LYiJq_ptA?sJu#NNfY4%h+eBS!To@LxTS zscR{Gi;;0Tw(_%Rs;>hJ!m!v-xf(mhWLzyZ=vXv6DGYk$0KE7r8)m3Vyg~3%m8r^T z794Xg29{PU;2p>rhM1@n!EOxJsgD5qKt0%ltUs{kS*7tl$QT~I_I$v6Gc^|YL~=&c zdug8&qcTQCeR(Dt3wN%cNl_V(SI@cbH|xf^ zFlN4d8&8)!M4tCAxV4VTxL_q@;fsG?Akocj(|79k$@g(8qqHyOK1|bd)r2pb1iA`@ zN@?2xW$=u%59@AyuY&L~uJ9`D@WN4<@E!vw=Ko^dacbvsd5X4TimmaL1(lMmmaN#W zJ`fR-Z=DMkA#=s2O39B1GFEvDu-vJ367ED?|07mOfc~#zMO(#VB`c5rUt%Snccd&B zi@Medi_lSLK7yqqjFq>vU9xE2|Ch+nMDfTJ%A)_BJDYF>IpT5i+y;1EjT>p+<^k}IS zeWY~y-pqfYpp ziHD$?2mPC#1r=NC2yUo9{3VayjYsxI=y!XViAO=f^gD&sWEplKX+YUr6T~G@3u0^K zFiXo{daJ$4xaN;qGZG~l@zXA}r=R#^4`LQ{YXAeYT-1k7xTKP@UQ1I&ZH&3`B!iiH zYbW{A7)2axcN;d9SUGn!HEz$plzIFjNp5Z;GQwJTz`}elc-{fH?k~!Q% zEn|{S;zCN5Bahw5S{5(qoa*!QRmKbEm#t1(Zy!!nl~P+;k#|&&RqaB!3CF>KH*y40 z+hT(wEKRkRB7{{8<&7ths;Yu@w1PV);t6H#0I=T!t#=C;ovh%INQ@7yGe9qe87s+H zYAr?p8N0Rh%VOM_%t$LK163cFgksVXoAwk|FLt-^e8H?#J%Eiof{NYudkTgomUFT^ zCKI|s-^;@+c`IcIeJp21Ukv_+1AIJ8ko(5Wvkb7D*H=U<>m`7*Jc5eY+SsR32LpS- zZ#rZvfqh7PJrZDU{s^vvOuf%a%xs*gc(+^Mf^ftmsR%h){eCO4MS7yaF9Ew&fOD;6 z>QF0j5R9cy zT0FV}W2$ukKlBLdNlTAR6!5@#yyNl3`}2WWharxW^fe6}ux7x*t;^B43Hq1k-Mgo# zJ!$)ELBig1U|*tTvis8wQ>2r zV_}_;!XlXaB<}r|06K@kGg4RtlVI`9!$D>B2RPUxc&P?>{>2nY^%caznh3%ak}lA| z6zh|o$!hpG8)H}MeBAI3fPsfU;+2u>T3bcNo|Kv5nr=^9_UAYBCR z-HMOsuEV1Pz;})53P@2BJpK|AT@pZ67<3565~RfzZ)9NpA7BfQ;H3-Tf8wi2y7s`j z)fa?;B*oLf0V@L*%6^367Id`d-K&Jup0r-X0VwfF-_a5z zMIhISG6G&*Vj$lS#v>p$(~Fv{|4=sloU|_7*b(X4Z_>3F9Irtmux}`AcYyL1dhYoo zqRpykQPJtU`#v#;0qqXOBcY#74b+wy2=2vmO`u|X~>f> zaKJhb3#(V7)e-bFA`ILs5o%9bgEML)=KD;{CzmaVIf0lSt?>Obl<~NHkHE9_IE2b4 zxEO;^>jnVT!r(ImV+qn?{TI;vZUL~JNAO|}u-RuQY{dKkVqpyg;eL`P(7*xfURXHt zF)~Kbr#iearYzUF9El1V7MeLqJaa}+pzG`11N7n^E~ff z%&9$T`Et-ROWI>%?kiUia{|r#;vRYw9>>!2dyM&7i8;YX&sdoM1u!oRZj_i4q{UzT zI5~7Jz>OZki#fo3>8&N^oq$>2LHt0{CKGcD7P2vFw*>vu^X|o*+N8=U_~@`sWVt&B zwU-TF8X@k*K3S^=RpC$8l0NwjyT-!LDNG;mz24A&`x!((vO|meQCZ8~*jw>*ME$=v zJ8D~Uym}D?ccrCSjG811GQW!O`@S}d6;>8vn2*^GO~mdPu+gn`b9Ja@yxM0NvL;(b zG{%NM@|Kt1t03@@Wf&;{G=6erm5L?C9;B~ zwvdg?lu8oQj=x6B0>;XQ!G(yfjZs6f!V016U6Co%R~-z4y+1FMKaGvou9!XF@B0^{ z)D^tk$&83j+#VyLj0eMbuKWykUP;|z#SF-7p%#1Ug0Y=SI{j0ON?m8gmF#H_;Q8`@ zk``NW&E}hft@&;QQ94FH`v-Ptq>_&Ks;+I{NL3^# z4}(2H(I56zZ95Woq3J|9nzN#gPW37E5S{$4Zgelq^!cDYgb#gfY8*WcUo$xREqr)n z^TVUE{DR5R{D?!#*F;*}=^`Ch!!JM5H9yaBT)7Beb!g2GJRCiFDr~KSEiFz*GYK|7 z?Qn!)bD6u=oWXN6OUUL5Wv#gyJzW%9q@#J5ynW4hd<9NBi@ba|9nE9&96jMxhHGTi zIn*3kKn)Nw0fEa&5h%`CWz9#?2aPa~ZXxx!&650$Z)QK-SS*gNbCotp#*67ZBQ9 zsOV@ezjQQL%Q>1W2_4NPT{ah(IhxDn9L@W~=E5yUbAgx5WjK!JePwgCgQL0N#pd!2 zt-r)V3P<0&lwnainx{V7w*gESry~^mUSYSNMg|Xo)*>Cv6G!dmgiWWTNwHss!qMNu zwmn~*>)wKMso41JP14ww*nGwAXs)KPxj?|ta&YH9aq^^6o9BjV&67MG&3CVk<{Ng@ zU4V|}8JAk~R7gkjO`OeFbB^X~S*>|EqN90OpBCw8o(z>P(s5Bvj@x{lOT)B5A- z^u+U`v`9zuT#|H=j-XnPk`m?NBsNbU(V9+_1-#l*phy>|BRqQ^6nleR?o3hB)lRtE z{cqVU@ z+vZbthjcXcc6>I&H&B%(fp9UbSviHxx;=BRlVFJ){TSUqjk+K}3GSsr3jP6OCLIa3}6?d!rbC z3L7Ihno^5&^qG^`D@@;fDgpXK@Z^YFyY8($$)t4j4srLoof^6wz5R6y?#^>4=Z^A2 zQsd~W^KhzX^(u@_cYsL8T*}dq$UxC_d4ukKBJY_%uhcmDpvX5Cz`#&3@XUM?jrzff zD0O~Uix%xovY&+1`fJE^y;FB$V5q6>|%J*WPP~qr*V9?GM^=nHQ0u_#KG>DXg zBI!K$EkDdqDb$B29vv2}DUM8e)J2JY=m#UoJL$0f|mP^60j_rx6xlx(e=_n{t=V;({XWc9!NJ2F0jj2V-9@}r`G3128PgGVFXQ*f*Jtnh(_xdhf+TlQtKrTP$mqg zW{$W^!;Mx5e(lF$*wOotM_M=jjb*f%>S>~;LW=@L_Xw>}qMZ@<%OdH9fsyd^xkPft z0JbFQ$R#a7X#Kjlb6{7C-q9UbAe%2q81n(z{_2dyf#uB89gtd6?dUb&p*q93){R!u zJz3n;JGvP(j@GMTprsf%DHN);zPOg)&yZTzU&o@%hSb`Ebac#mx-%giT|(TG#l36; zH7DMu?EL{b7evz>IbknGAZS8|$MC1Y<811~>E#X_j zpx%_AZpK>|Jc^MuctG@^r;Z=f{i3+3$5Z@M-(X zTJIM3i9a0ar9TL zl~ibbOypTeNA$?oFX>()a*oK*+cYo-k+8dpfp10e6N|1d6pp^<9Y&A}t?yn!Nw(I@ z-=#GArORohE4*r58>pjsT97?oEZ+M*)sKsuBJwJZI!G6a7KuIzG^kV`QlAc~b&HQ^ zraH1g>lxQr3`UwBR_rEn=7^-*(Hk~XCLsCR?TgJ2H_tl4V+?K-;cnzpZe6D86;XV> zi|S2~+SM!sMT;@5v-i;b5Deq)gWl0Kph4ezFXgv_Qls_WZ>ZS{iO2G{bVnXwIot#3 z=pPSKmJ~Vl5IL0~5eFp8-iGLCT~(EUqMZanw|3og8RK7tLMH~6NUVKD6s;vS?IoA5r|!_@77bo5Sww~CQ_k1+aFXx$V? z3*ml+#9#?}++yN(#C=uVzO6Ly>~2KvWijxPD0XgvVzC65zl~)}1sdqjD9J`dKc_VM z={sp97hdti2a1t3&Y7}*6^qUGQQb{sKauYrrb$vTD2H@3HCmthmipt6T1OqEJ_U)# zSzSJ6!47cj#yQLrDv)?69KHJp>6DKCM%)9BQbV_+(~co)hKr}6lI(>?nGrJ~wKkf; zpg6kDaS&e-V$u%;sm3VmXMzckTK9!a*Efs1s_1i1P~QO(16Xmd7x%%F)RYi8{#SBJ zLpu5y#M@T&S=vbfrm2*KO}BR40~q0QLZQ>q_aQZQT~Wk+&xBEdktGb;9YkH?6holG z(QPi1^0P>~(KCZ_`ynAd6=axG;I_v>pd!=B->YPy#d5f;qxpEzn_mJuHKB_S4_Z9n zE(VA@8q(3HVA=jm6r=xUfK+Jxfw(^x*(8Eq+e2cq;0nu?QtKxFP}1$_&T!jtxwKpP zDk)T?>r!wdJ0LO62GOn!3OdHuseV+5RHMbhH(2OaAy@31#mY%2w7&fZ$}y*Diy%r4 z617G&odj(PX1ZI0e$>dJ84|{38_s#+OhvBROT@X6&b1Q4A&JBMupz=M1X` z6AZ{h@3*ugnsG!O)8{--<{STj^;u)yzOeh>3Jj)1Dfj;bSlw@PN z=p3cdcf3GPzX@m8CAxXinSGa7>}_Giw;n3;DUsuxXqY4g&vHmdQ-isHAoUg`${~Vs zawIh+#r+E07!HGJzb`TS9C?JHOcYDDd=%7Qi<$~83e0px(+VXooWUqtBJ-=bXNa46N56ay*<25FBvlj@exo`SQoEW{plGp+PS@{r&p;Tc zme4zT1T@I(3otWU%#=P)3e`yPp9DjYm;``K*LR9LNA%7AqW)$`+*;z^B<>$CQ&Upp z)W6B81nKA;WP*K6lC~|H8T6XOfNt%&e`az5jhK2m@$^Ba*vX<8D6ysjO%x2;RYhI$ z3PYd*Bb+QIY+NNuy3zW>xP7w_XU8&3N_zklm@xp)9wu0*GQCiNJ4Ds*p}K261rP8^u<|s;nPWr0Wd0(Z`2GeWL&GCFHKv;XEbmhlNc{ zYZXTxPY<+p9`j%=6zGA+Q?ISdHEfCXnrch!)Pj_os{UoF_A0R42GMu-G zvn(>p?kCP|bPk0R&x#r}`Yj|fjOu4aeeYFyTuS~n{fD{S5fYC}=<&D|w7u~crW6AuTkA8WDM>`NEK5(5#XVHYkY_L2>5UoF zPPIt6Q{*s_BO1~qDOx0IG&EXYt4sawka+6mP>!ijO|-b5hZ|KC8Kd>s6qfE-*mv{` zkT#u9`(A-GyOIF}1w^4tghW*l_p=gIUL%^P8%^Vei5j}my^!}0L#2i+yv8~+6B5m?XdV^xMjzi%++g{jr$*~1{nQMHL<J_}p+E24f-(>N7)|Im|(ef)Mxx-q3B*%Bimoe?<_d`FkIDa%3H zKZu!69Hxv4G{++63x#?|M~J*jWOtEW@)?i45@1Ycx-+1$Ycyl6eH99{HBe~%au-TN z{HZJ5Mt{6JJv}R&dte4LOo+Gr01Rl|zZZ#*h@2pDQh%BxMT3x zC+hz_^h4?B4+&NUi0l_d-$A5MjV68|B~M+oI}2sR2!d3k>j^_>g>JMI&}*Id0A+hf zG&_%RzK%zw+Wg+x(RU)!_H?nbtu)iJMAT~`Z8}#;mvARsrf)a}6MkTki)t#|736*k zp*5X&kAPt9lcI=vm|;_ap0c=mh#Xsn{-;2q8-e5Bkin2ze<6}?)ILbNW^>wIH>H3b?}^u`l+aA1q}$O$;I?bEUVnCvE38LK{6ufbpMD+|IQVma&NqHE0 z`w_8n6$;d_7bvIBqAh~xe?sE76lCQJX=Q}#gA1McT0?JNsOO4@bxXAzXalstT4naQvsEL0xf~G3R^|CmNs;{ z$UCI1G@8eEq8jtuUJUUoqo{hpsrA)t<|*r`eV<@d*dqm_7K0g)N040m1yK*3&0L^D z>t*8pK;-*XSV&tT(e-)-^bbK#7bl(puaZtRW-Z~i4~eZ&uQ6Y#z=RImS|nN{`o9JW z3T+OYb}L49yReDjE}2VHZxgI2V6Aynza_H95*j3^MIy=2*quR4*RzH4ocQk)#Rj%1 zq&)@~u`YasCOXqd!6=vTqSD#4u-ux@YseVa%}cUDt4NW`4<%=b^y% zGos+>%r*~1)|yVV0>IGE5FU?-w%-w6g{6$}W|4Gbz9*VF+z8VCQYhbWUh#l9D`M6X zBNa&dSFueDFGANdjZ4-sjg)p|8wP9?=PI(69hGJ!8}%4dA<*E3^D>5l?&wCI4!@cWdsv`3_2u7Cy&9Z+Z<+J#$>rw_|AiU4f$ z6k&|oWFf;HUoo?Iybb21A=6#=zNK8|z>U0Z9U>RR0Hs|^6kK+Jbr0{M$~PCqMmRA6C-SOD z2TQy0q_4#!4uf>`qLrWz13g`wc+0wq6si##-1cO#)o?ZMEh^G=b+|DifkXz-|9hY~ zy8l`@KM?jrVG~0mypB8`W@{f1@OLP*9=M*e{AL=IlWs-gNe7MXYsU!n6iOTM&!cE< zo|~)XAX-%6I6P7L;U|imPJYLUPCA|ZEEAJ$d*Op8mO47BE;8vBIJJJT9#c(e-y;gr zaEFjQQWSJL`e{aEf+!NVFs4-CWrDc-h^(2zY|Mj1u5Sf>9q8%e#49=|koS;i776YX zTOV&@da1zl9^9DTgTxChQZ9jlsiV)~^tWeZ<3A&h7~Z&UCy&SG+L;1AvV-cgA|KsP zB0=03ka)KMjol8!bp5?hz7YT1=f&oyjjLp#5f2zJrwrisMGiZ9Ff>p5c#4?40Jt#&EE?+3)BK63(Qiic<>FdW6T~Y^^g8Sb;^#AIbxxtZv1f~d7$$JH z!pJ*f#J`Om*`nb!n79d|hbmE9&ZzAbVslQh92Mt_aH2na5sGtSm6)S%sm{`CSA+6S zf-xPLrloLdVPgRG2i*i6y&kg?c1=*wLU(4vb0GnW8%04(+OWR(DQy`2EnGwzKWnx5 zovWkc8^One;$w9l6Gmx|73|YPs6{UbcGGvPTvTZNlDOxJ9Nw6B!5Bz9v=4*+A?SG5 z2Pfv@j!@E#)WB^Y5?c+9vbs`{uB*YVMPee0{;x^M{f@zDcVfaO2%DJJ4St|0ej006 z6!07rTHkY=@}n~}NKlK!Z4Hgx8AKd6FO=5ee7SIB#U1>E2B^1_^BFrsx$!R=$O93tlp*cf0MQhQqUkv(T8Ls5 zoOplDRCE^wF|FAMU_BD^X0TYL6AxfWM>nMRr-gm&EY)X4QjH!birJncV2hf3?-|<+ zishmph7K~CL;Dj^bd~08XD$@GML`S$1bR6xil1+!g3`VS1?CUfJY5%zPIRX(AvW=y znYxv5Vtz>E1(E79HS?RZ0xp9@KkjePn}UwT4RC6`MkqGycb4Fr0*<&&^=Xk0;-l$w0g-K>;6Ve8y$eK) z%!IN-{0Cbg$tT2r8Of!lt)RFh3SxNvG(`w@7sg~+GdPpr#2YX;9nCepHb40TmPW~M z3Nd>$n_Z{6uTb(!Rt&lZQ?3HX<`;Nao+GcG#>hLdT2V2uxJ#sQZ^J9Bbp4&YZluJ_ z9AvfhcN?loU8D7On8(kjlMqLq*j!uFMBDQ&fCz%-lArYl#?+U^3Nt z>)}_^RoCQ1{ycAWt`1)*@?+q#&8gU*@2L_*_6IAEZLi+6V&Y0*-KzP^WIRuY5EDvF zke$UgJmMZ0Te%@lvoLnP2|n%cg**y<2> z1i$8Ei)*D$!?nQYSLHt4f-fSf#9R`D8wW&mbCK5fRaf!V;!5E6eBX2+nSw;Ba?cG6 zs`whQRRv)b8wwJq%6V#QBN`PE zu@s~>?K;>>wW&G;3sKg{DV`gB0ib{$PPQ+*CYPI#&kcKJ{mHPiscINRmmGAps zdvC|t&)LU0W{&ZgGh{l3gbbw+ilmfCWk@L;5}C`Cgv=R|2n`}rgQ6rA%8)V>Wk`dP zR4Tvsb+5g*?fdzB{`j5OYhCAF*L^+r^Q^Vk@Jws1JYM^#s&Ja}T^K@@5z#hy$Aj55 zx_FVye76#qm71)Simha^QaZL$i7pW0l8F&&;R&?dMUsZ)dpo=TiuN+^H*4-iiX`brnjU&`T2n7_C=j0f zKXUTZObOKFw~Z7?T1}$h^w8TQGQCKnV9LJ#-^?e5eoJ#+XYV09uI`){BOy@#ijZA1K~~oL*nnUwZZSY^OLX4(7AV7(mwq5 z{|CIa!8h`vy2sgeaj*xpE1(-5pr{@5YORpy=zeK9hP<1k*5f8oRkUJy?}uJHmB))T zEFwB!U>oecAB^M<=w$9M@?H|yRXpTP3+m)^p<5Xu&>vr(w)MvG^&+-jJ-%MS)-&Vl zwQN0ie7%XSd-3%)wtkl32tWSawth6e-qY51#McMd`o{SBC|h3_3*zf@ZGCcl zeVMI~h_A1;^*-_SO}2h-biKd!n-<$2>D(K&gQf*{m}&uowffr?XnHE-MM{POs#my> z$2f=n`qz2D^IjqKNj11ebcdNZf_It6n6F29+_aNh>7M)KP1I*T!NWX`7OKzq*n6Im z=1Zee^qo0B#iaiRWWL(Y+{}M8`gy+A#u&nGh+G{NL_TV=Ncw;5;@tzS#I+|qGr_>E9zUC!0C8j#Bp7t|%W|BU=nT1J}v^^)d1P z^5sH$5@6`9Y?c#T)H)~pH-%M>ni;~uMO1p(;c&mADyz>(Lkr8i1L2vLTTFV{OcDnf zeT1P}$9t%mL|t6G&#W`Xw_k$ikSZzs#q58;l(6^=!*JMT%C3cm3bQGFyRMb$H7n-; zD~Znxtd@cPcy>-={@!008j#cLe=TRl2Y2)0)Sw=;niM8lh8^v{8Gd!I_Qo>mue7r7 zmo0{dCSLa)=@>S>Sto~(_E*mg>;?~JU<*0+D5gv6Bw@Ew+8z3e7V_= znJu0+Kpx>LvrnhXJ^<1tH0C|L;qV`(Z%RXgv$_$y=wEP!`%%4g1uUea5i+ zHL|}k>{L(o*M{w`mA&1td41WN46kb-`)$Ko4Q0P!Sm`?1iw!dy$)00au(8VUE6rg} zbZll+H4)0z3=0Q8@qLDXE?B{9mex+#vJoPDuQpIV9;95atuJWKaWM3)!O;KCXDxWb zFk9#1?muLU;d;YvaV;!Bt`6EmKCW-d9=g})e%GrMA}E&~4?{pV6lY3!a}T;6$C;VG zr#c!Bfo`@XlrJTGOvEdusPUjW7NuZDLNDkKOeY0HmE`Vr8IgXc3gY9JN19+$!rOSR}XW#d^+CBzXI{N5%O{4zfHRNpqm6u zm-`ZS=ksPpZiiiLVhPBxgSJC7v$Krx?JztQk2@4fw%bpl3^SSH2JOyRlm&*>?o<@f zTyTu>?OO0$7&{6_nf)xtjT>}T?pH)#^4VPquo)o-0_a-c3`dDiv?+HP(XhP~IkgoJ z86h7J(6NO~UAq-}ltqnR)ny3vKB@q75i-#VbT3(SpLju=vf7CJ6qIYFwqmmp@^Q!i8KZ}G{VYla!)pgBifEqxGRC(X!BYc$TsYC}*&x#>K)1>w`jS0^ z6krEJo;3yC4vXRwyR|9b7|}diiagqiUyYED*&=KqGq>2|&h=0^Wind}ObOqniOaC|1_-T`9ifmsa5r>!FV@OvbEm7%HFp=}~^!GTKsu^eZj>*7C|Zpt0r;jgXI_Bz5iY zHX?}*erDOLZTzhf^6?A@cRF_-f}0j5vyYNx|9psCQhUnOce`C7%uzvmt);!LgDW7+ z>VP_f?!)psgY2at)~EL0I)gZf|5Rl#bGUX}$r1XceA~yeFjT}wYIn+J&KYP8ouy=9 z{RXndQM&;tx1r#H%t{a<8u@lhG9d_p4sBRnI&BNyu3mlZ3I)o3+D*mcvrlT;}_&| z`kwG+!{6VRSI8X|q-wK4czCOLA=92fIw-aa2OF0ALS7+<8OVXWL-7Ve+Q{IKP6me_{g-7MsybBD@Y>9o$k0Uc$Fp=`O6Q3|;e zf+^w7d4-vV+4ca1?(}7&8 zk?O8T%`Va^eSe3FRn`MikD`=ed>Xh+9-(Egb+x5&0 zZHv&=eORZXPdxmN!ztvU4UqY)>xE&%lJCna(l zvdVx9t||ln*F=>g`C09qi@JKPvP1{&*5#~N6^M_lFxfXKRtHU5+GB)#yo08fuEFRt8q?u`U_Jd+3nKu`hVN(>3hmuw{<^I=Ov3}#C$TPF{>KLJo>0OV@E=_9l4tp^= zr>bL&)n=}-X%D)%b3HiB)wG+18$U%Gvi*pHj(bkmJD*uXFk=bPlUd6>&EIY42lvR;&KVm_VHbN-Pnh6xs1WLJAa{8V~Z zscFGb>*H8)-&=s@_voNl_9784T5u6mY9N^4_LBCjkY|xWSK?7+`oJR5BDi9jy%MAk zbg>vAXQ0_X@&;FCg@cUV8;YktMEN~LFm4}D=|sCTq^&MGR<;;Mj>pS>J%{)c6J(3w zMF>S?zw(y0X=X9mK9jywfyb}a1wqK%6_7%`B`jyS;ca<^OjrS#-tms?O0(7V)xmI- z;ha}=eD!ZGRI7>^dQ_mfBU3ZXh}I0pyv9aU@1u>_yd1MWCXb=OW`s4ZGQ+09g~J!# zP|239K+@UNr{2VT(9~k-8eAnTyIOeIP&Eze@*E59IZN~Gam;yU7F|p7rIvg=EDsp0 zQ}9({B;wkGN!J{-`^lo*V6!A|d|OdO^O!VKD5;8=eKS1c z?C%I?o4pw14i36%eHBqA)6tN)#t?L2eyj8w-c>}=JhW#e_3e+~sqDQc%)4G#0;D$^ z-i@?~GU+5o;-0gJP2X265tPpGCKy`!8-&x%ei3BDK)1!B$mHxs;`UmUFE%QQ2u^gv zFJb6w{7_i>17TH=zCC!CtDm+`CJ(G5aU~!a{-8!kb9j?tiRR92jBk&H=bpM*xX0{o zK<4s+u8T#KNd*Im>thkWqEgG;`i~V+GzYs)qVVl|;n_!@2$!4v4#*Ty(5bS!bjs6^x9Lw^FQ|>^pSY5bR#aPY7 zpLCmMl6+|>o4NjmVp76G?qDYKPW0g@J#D&5RKJdHM4!);Eru@fsj|x&Kgl#=m`*lF z{Cmc~Jx{h6e!1DQUp9WQX~a;;<2BSA9v~I33BqT#@+-9OubsYiJ7yP5>Itx11U0_X z$A~)__2F$n45jgcu{;V=p}T6FwLn@i#%%puiF$B(c8e|17oBzK7P_@Y_$j64Lij+E z=e8StvYSGQ;Kd@Ri{M;#QrFE^{UkhY?}idXrQa?DX#?2RdRJTt47;&uSFhSl@x{>F zk4Xy}q&Le@3>);dcrh6^Tc0yENhgqyOebllePy+P<5OgdVRrIF*&B>sWEwHt{PvOI zBr>G6oTk7ctTnVYR;~)jzO?O!p$*zaJ7ms&(rq!~Z3wpulh$-s`Nqn3+ zVd|snbt4qPEU9mxA2CIFisA-ga+$nO#Vju0)8-SQr|*?y=P_j_WLuTN>z3vb;?XM$ zLC5WJVF?g9=yo8XqGYs?&dG11yYU}f_VZSQooX5PzF*mo{Emcfpb~1(i?@hD@JJv| zmkG^#_a2b0(Ft@ljJN~BHNfPT>?U7j`C6M#1U-=u^cS5}JRxTb1#x{W%@8u+#DSnv zd`egaL=L*CNa$WNT1W%-v~be*I9HfcV{1I|0>NE`?A#X1c~N&2rS%zfdyQBI;f`an zr+Ug)Prl#GCxS_?5WLRj2gMU|`yew{=y9canhdycKrry~qwrb~Iq0rM!ceo(LT1wa zB%Jp%&O6MhAv~V6hM?nt?A#;JbUav5?mmldxDglSdm5A4R4*0tCi&)=PXxDS2&(_T zD4vieGnu*fEzK-4pdtamq4}$@Ifxu|-yq@LJw^*@9-kBL_zma3=G4dyn+aj3Te(5e z<75f4o@DRD&@KCrigxlm>FOBK8p1Wfq{{!We7ogqXFd^huR(AYUQm1?Eq7%*z|w5R zC^Hp=&f`9!$=<`c`TiyUgJ2stV!iMwFVHRQ|j1VP0tNmvp@ z4!R;wXd7gb(&#=p?SVGmx6Oxw?JP$Ai4K*l^6lH&n{!eah5>{di} zKakPYQQYKjAEMnbteEJb#g)jneH;bDMEw$y%Vv^0$dh(3ZZb(}ww9Dc1YMKZZi&&= zN{JRhyB^!UZ}gX?MT_93gzfeiol#b_2u2;S-48~0E+<+9_Xcbi8mcVcC@)$BHydnM z#OU8Dh!(+I3T#)y=z0}Ji{J;tcFm0*R!Ou7rX^v!E=F&vELsE=Hf;Bp(c!B_i{Mbl zb`yM##bQN(<4=$vq{j#fag> zy{a<2KH~2Qj0sh5Uf#P8WlxsiX#6qPxX{5-t-Ybp<6^GBgarT8(DGg-ym|PR#{E;| zUz+1Zi#;!N&+4jtiURK@md|4<|7G4Q1E(G;K?z>YoqQzO{~~oTcTnQ9EWb-U({fvT z>&?hs3Ey@6Ixhl~hb3Cj`K=n@+L6AzvgehDu5#H`cw{E7gXu$FZR0h8zA5f%rw@_b z6F6aW(ZlWZ*^zq!Cv7cvy8$JyXYG`&6>j%N!IG?nQm z%rK-ks)QJ!(>ELFdcvgZ38Xz;QjE}DZBBh2!0Af@w4O0(k7r6Dbo%@N4I50_-5`w> zkXkF~^bUW{P)ufyfK>lVi4i)zCZC%PCN~a{n|Nt4LZ>&;Q?bFMdje!Qu8bI=)63qu zZ)uD5g%HMkZHx6K6Y3g<)fN`e-n zTT#W!g-PE&8dn*UzPZCPr}w^@>sGCP|>CJ%Tol0xVPm|~{P z5vGvIq^Osr5W1O0e_B*Tx`?Gl$mE8@w`sU^f^LJ+QwxX~D`JNcGP%Al6)kipjoy$) z#3d1b7$K7@Hy`Mrx(mAWQQF*}m+Q(<(}w(ouDsDj`xfTaRMien(4?86zEYygpc9YOOYLWs{ z4>t4hPnCQ*J{qeazizhFLO0JOi#|cU)Xe=`q|PRdd)tU*_%g!j4GNKPKtnl&?kjWF z;r-C=8#8M)!~DU_ZZ~3HFmn&#T<|GvdBh^u;_H?!A4EoY{=CMb}~8Y-YC`FguvJyAftrGjl4Zhs|v8v5I`6g~BguDf_FcvR!p$C3MqG zI90n~0VWp&%y9U0f9)BeTW9pcS1aL8lGFMF-RG9@Cn@%sLM8{@V@fD=XN|tGvJ&wo0nKx&O$c2Plevyx>ZlNAgtNBD-ur1m_%b`i^x7&$w!70( z7TF=gjXuR6BS z9kDbIVP=GdRQv|WDRh6EvkWG0#sTTAF{jXFjMerWFsG24)niI2baimf3vTSolV&-% zam))A?^DFvh8dU_yzj-9-fc<7zA>brcQPlRMffY!x5z4`1!7Z3MAB}ivM(F7FkhfsOWoBNmMDg<8uXOG;o)`RwLFqE{b9>}G z!A$qnwb($s?@}5X;$r*t9v8N z`l?}dZ<1L@M@+J(TpXPSAR znE#0nuJyh)E3_u)jh?MEy%~C*hBc2b3N~kIRnTj2A#fl4T-QG@ck^zgy>?j3Lpib= zc$;yb3S9_nNG{-wHUIWut-P6LJ&;|>YdJ@0{;iYlb;PWyW`_^!pfww1Ja*cO?YO3#$gY?DBPb*xSa=w&{UVVObwEjcv>fcPqY?M z4T4r^mSTyZItFrytSJ=l+L>fwZ4paAcfS!Sb)^=;BglpqVVFpCjf9gx_l)J{6K{){ zXDO@Imrg{6p(f2Ljd-`wBxwyrY7tD&H|)_!_F$0OG3frX+JNCeB6uEJ!SN>Hz|z>X0)g1=~yAxOOU?DRAIpIbS`;?+!I0G0G}r8PGGmwg4G(V zJbXMvV0X&ssm+y#2>NnbrVvF)Hx5V-Q7hq{hK<|EE9ABKAbl3M2C`ohmZ3J z>{b|kTTA63f)4CWnqVNL?;7ORy;<1Ru*DX6g|wkS<`I4(+%`_Ukk{sdOssrb*wb+M zGx7@g6+sU0iNY_3ix<+c0jar-6!tM3`=q==?oA*ywWov&Y>0Z14Oj2Kq~3oqbOms_ zWEj#5fX$g^XUzkK>2|^h!(kzZtzj(=d)kV2QV6MU+tI(Hf$S&jz!%ar2vT{oLw*5= zZ$?;1?jUoE>?m$vc+!qqA)O5%cU(ITbGKoamw8s#4;`4iB?Q8H>RLhiJ#}5wQ+I^l z$%#t|Zyctdtnw|SyvK;m5UwXCoya5PD=6O}^NDaznNKW7cw1#7CJ< z1A?lKN!pks&<@E*NO)x`CMiLV0K=8H;rzy&I^=0JKv?NhCdEqsf=XXgr4YvvI8r6t z)}E9_jVKJ^%45<>A0^*9`RbZagllR(vHX4=l#P&CG%8vbbG}BtOr@~`9y7_$CRum~ zl2J&wJY$j)h9H-pha+FkcclKTN+ zal^a2$?j?Ra(CIQ41eh%`(Kb=Pqm$Io}yqf>A-wdO=_XL51J;m%-*Ju$q@YOQV88s zrsyGak||^|A9=YHLie)K?{pAxM8rBHWb#1N3eiIMxkWm@5_6v^WO5-`C56zPHN`^` zUo?eG22apY7f`~y&uKHpv=K2vNy-=@lcCVnqJ{2SXhtp9Vm32{O!{8ll0xV@nId%! zW)D-yWcVI zL^C5~(j56fw9s|3NRwsuFojGy>o!Rtbi<79U0p<2#3UnRa+&@}w9vh5kygoEX$qN~ z6dy|=bf1`_vdrD4kjaCKTci-W?@jTN%=4y@$zwjBN+EQq)3oJ9WfsDumj{z}&u3By zT~$*|m08aeGQ;7V+oTY>TTKy>dABKKGGeh^3Zd&~iYH|bHHA#3!+tJ>&`mSNMVT*{ zLMHF++aZO}tu;lq%nwWxG7{Z_iwipLid*`?w6S~U0W-YCg7J+ z2wgE~nt(DZm_jBGLwzNM&|PPW+hn#dg-qU1__Y*5cdse7%6!NaG8rP+Cxy_BGR5^W zCz?Vgt>^tx2;Cx6yesn!Q^;iE&Os@J?qgGA$=qoQnatQYB!$qOGP>Vz(Qjspe)3bg zy)GFoGaR0~Rru29>JyXAY%!%QJwrQ3CXMGKN-1>pjGjMK^gDe;FD0eB)o7VK8M#A2 zd@Q3hz0E0;fvBTOBXmzev%65cNv4p=sQE4>_i>JNubERO&4y!2E_9oW?mJj?>E6O; z2gvy?>D^It%48bPSHfXzEOVi;x!Ws!Xr?w$CPQ4ul~U-67`&MmO|(vvlRM# zW-0cDkkS59rC@ygqa9sHRZTLsq$EP>&LH(A!-lsZX={@0B_t6tvT9BrpO>UR5?+#t zNmp%#ghB@PK<<==u06te7VBJLNrb$@&73~Akz}Ju>X}5y?+bFNF?>~$R?Y!V@}F+pZ&8m^b59TM7AnA|8+ z6-&s@%q5|Zl{+AO!eU)#5+SuKkQ39ek0kR=@>E1ogyHa~Adev$ZkOZ(%T6W}Fp?z{ z(l-y%H*eVB4us!Ztl~K(5mE&Kse%}elqA>l+5z_`Ng`w}EXbK{cvzC^NO)2KlQAJr zLLrltKxzYqt?xwG(PE9el!-)0cRxrUgW+?M3^B>)|4JgHscudmf0pD0%T6YD@xLV$ z(kKRLCL4C{i0~7O_2yra2pQM}nMrQ=k|bwL^2eW&2>FPEIeql*LXtK|TPKruI-ZwM zNVfq}Tdbu&OCn^h4#-3t!_AU>YLcIRl0?Y6E6wR+=}t&~uB$5 zGxZYwOGx*Q1>5$t^6>F%0=u0?PaL2;L{R9v6h+8pfOI@`5D1_D89M_q(F z3G7PD*LJ+{jPmgDAc0+Nqn~?Bd5E9`s-2<;nFI{-;6r<1U&D?ajL_5o$;h1ilDxcrQIgvVONl$;M&5SwS;fi zkyps+3GzOGy0W{^QB7hX;oTt%*=(|M^0CEea{A5aOPKVDfi%}DE0&N397sjHitsJN z1y$u0(kKC$=37nJlEALX0`0QG&nXWdtBfU&JB>bu$=pVe22UA95z+(#Ils#aUoae7 zUS1)6XCQrN6@=s+FGgFm^kLlydW4+OvG3g-#sY50yijYb;Nb@OKc%NZQC$Es1 zEg(+}rU=gx*tr+A9q&(99zOp1Gx=|5J8RQANBGHd0Ycz;Qf-Yg<*|=@(|L| z0I~%^;U)sR-WF`iOy%L@o@dD8DWfM$R30Lj+;m=f2$|Oe((Jh)tZrENPkDvxc#wDH z|0SG3V7JqPJuq8&_}JoE@;GJmB}{&EkU5s$D~gcml^`wfAB4xx2)F(yuaIXBK#uR9 zgbxzf)n2H5Rd*bH1b3NZiXx=B0@BL;R`}U*;hW#dE95)_X{eqM zHXyK@ZNXCJDGwiSBd}Xxw40(lMDWnkenk=TNE1j6@_=xq;p{{53h7V)sqh^ZrW4rx zW5Ev1R~|mzNnn@%C2d~XbIL;m_u0=CMaZ2OI>;w2kA+xC49+nLLGU9yq^-}46iF(LtwYrf<3uV zdHC3HDtR0<`Wz-z9gtf=B}EZ(D*(9_R1rRH*srR*LT&{hw}NWIaRhc17ip(-Sfo6B zY&wlR8XNrwCNqUVZUv}%Muyu3ng1t1ru3c|Yy>?T^UiZ3e&<9=npZO(+Tn| zy}xut5>h1unQQpB@QaJWb^pjKWO4(@@%FE934vYlm$g#{yskWa{6TrtGJ4@`6{|~|sf09?ooOh5>lb?ly3G7B%u;yGdLj z^Cdxs8utlj8;(36uaKrcNH5_*VJ3lHizV7uKfS3ueEf>Q?p~vxnX5cR&@J(Wq6oPd zg4}L)3P%|B+%2z=K4*|OLVYPbM_~851^ZyN^6>Ehf!#+&PnoAYL~!-qq9{VH-XL{@ zPlY`UZ`mrZkY}Vp9thbc{FuNl&r)s2w6)5^$D;&xR~bEHzVZ;kak5@fgmfB#bb7uo zY-Ctrqr5`uN+5kL9|&g?*bT5?J>F6tK7L1F_l(hVUQiw)xGt?$6d?~IfZWp73d3GCV%z4%4tA%ZK%B1I8$M*$gfdRh3}V&TCh@(QU= zgItxF2g0m10=pM2*wOcthmRQx$>S}fzrf^kV<07(t0+P~HVX27gZaV(^MspTkXJ}k z738YCK=>VjUD7Msj#V3!hmR#*B9CH5e~C%YAIRw3G({0I(gqTJx^SN1=$Y~g>0ANX zx3h$A5!gLw!Nz=`JbWy(h&)CZy$_Qn9mqh&AVm=}v;p$$z!SoAgM|l%$SdSn2e}gu z6;2?q+it-+Zc-jTUbUD!zBT$VCSL{wxnA{E6d~6ukn2^paDPAHhyCRha=ik%UJa1_ zJnz+_UTJUFdd8owRE?QW*T1TEiT}v>HWBS1sHMy7VrEu#%wA?*BXfY6&&wQPW}_jR z<lE9ltY9iqnz$7I|4c>6&Dx9~N zQ|(ShTOs=EwS|vE`XheB1ISmZ6ylf-`>{&6Z8a&!8!;2YO~a%=VT*k0Q>Y{nFbvMPE@V#Bp|SR|#+1 zOv;r;OonjlF&UElT)yk%+ipG)Zm;>oQkC4IY=nHPgfh7E=FB2rUdIT*i$qNFm`N&r zf~4>pD!oinf?Q_|_kD`9wmH?T<@pi_D}853zw}#ll>ehrh+|mw3zhJ<&q(>G5yv6i zU`#s0cFK1|z6s_N;bxdmEN!K2%0@`tQ@O7)=a-mFJ2y<(F1rwj9CTkH$uN_YM)&+& z&cR>cJZDZdnCaMr;PWWTGi8O!T4R^WQ*Q^lqDIt#aOE)>Ox!JBS^4UkPlRh~KC!f_ zb}6WkH*}Jt>tfCUNg*oG{7kLrC@`QOnm-f;80+ z3tJq)dBL3O2%tLv0$O>dy{WQR+Nbh_kD@DSL^TLk36o3Be)-NRMt$>%a5tJyEJyn{ z3M!=jMvm@&bKZcHN*n~0iDSabAac-+K|;r_(L$bOKQ4UsTbwVMQymJ7>_OPswhhuh z+aA@~Hdmz(M=kV#O1SJhQXVp56ofm4NsaHIeADFn(|jUaj#cuBWm>{XWh118L>XKu zbB-t9un`=GCfR0^;ir(?fP~p%n4|>h^EoZ7_C3z7=2TY&Zzh4DdqR0W4owU1kjnFu zbmNWK1L3A&(qca%-#+;kn@@yWVLq|UEB`@3h4kK&quXxIN;v7%hu}E;QJ4or4!W~Q z7>_VoNZ;5`!u~(w%(Yrhb$RgMAB64cI*|UJ?x8*1RHYEd6w#w9;iR*qyvc~GAY404 zJ|+2$e1qk?-+Ut6Bjyv!;LWefMo2FPWpESB`6&7FJS_w_CX>8rl5XdaEJwn;NK8_K z^eX-)EN~v@R&%N|g^nNyEB$Yfe(49Q^wU)eaSW#(Qwgj8PRgV;BA$S7xiQ1xZW1rAA(K*<+%Ww@xE_Wo&guptu*3u2)7=SckP{!?_T+~n@@z>Yd*2GXZ}%8 zA>&~x$a!K-%ePRz7UmP-+M7=-0~KKvO-L6bWpI7XIg@-D=7L~c z)Fk^&GSMNKhJ+CfOj3f>?^A>gBRE%@Q(aVCA|PmxD$h@#X}O$HdH$AepAlOi+_#uq ze14SgqgOe^W2*&c$ge5`bplghT#);8Fu0-j= zQMqy6V@`E-@mLcC6Z4d3e`wCxpH!YEdC-kEq6mbWjEVSX`KrkGqWMI)*UTqYWAZA9 zkh>~5x=+nHlw4@wLeO|L$pa=Slpo0tNcb2nCMiL#JBC{d;B;@xsV*~~I)k8FUwLLi zGb{M4$}^@Qx*A6GhHwoq>ErxGzP|FcHlGN0r}@P4M)AVRQOI3Yc?~euGD%~L?HbS^ln6!R> zlW(njf0|E(%khqUVwqV|TsaEqJ1c^&lsS`eGUFeD$Bs?%lSy7If#e1x8DW{E1nG+? zDQugGv#UAPg~$9|2;MEMJRgT<{?2)oCwD1y1mWNR zApPswO}c}OP$}xbF~j_VO1QE-DQ_~OJA`Y8$@8{<$TwHM`^_i9Jz_qw99~zcXhLd- zl)+6f=M?f~QNC~FOJ6Us2v-D?AKL@YQiDaaSk-6IxBgNI|PkG<@qc$BO!mQJoRg!d%=iQ2=@vm z)slbYD=*)><`dyQHlJ8p=vgX=kg1{M=#HCnAh|GO6oTW>B%Mu?rxuc&@2RXZNeR+B zX81vEoMp_ZPEV$^LC}a)p4UP%Z2YgvGlHjq+|5RGfp8r#`S~u%_mF%&%qPP2GoM&$ z4%aA0AjVNP|PQpJGaw@{uZ zplMfmnyj!{x(i0U3E_h8tE@$W@~x0B9|TQJ2v@>h;5c6cqi`sNJaWB?I@ z4mp#YGRd6lkaR|p5td0xkd|d5;mwV4K4DIE!cxV6u(NFuq<^+8(b-l~r4UCOEu<25 zZ$ip7MkGPFjhMU^Fo%3?<=bgK5$>S*#8Pv(US$w6dQBPJpXR)ge0e_=1jnIC#+#(U z4M>V@Q0Zlo667*#_(LHDb^;^@0bQVEl9B4xG_cS5*f znBnloobvs8Bfd%I6XBjWpIB-R%~b{=mrv!s&YUMPd8op$cnjH;K;)p?k0irPQX2h4 zOF1XrjPrsy)p1M}1Hww5xiMDyODg?ll|md-NRw5L!enklL}lG8-=F3a;c|Q+pIGV=w<$*QdORs zcc2?@L|X_q4U;(?x#TM@-(vHLa4XCwmbyeo6-3DOogCeEb3Q^Yys`v>C+badyGg=# zAvueLHzZ(^66CiveCuwUxjvLrUE~afLg=fq}deTBChPVFaqf#Kn7*=GzZ4v<~Pu>WJSM;dMzD0`b> z${^W=3~wJS`#!@3Pso1V@XQd|e;C#oCi{BBk;7#_Yq)KM?0trXM#(N~c;Az*62lU5osc-Vte&L;~y~wWI8yG04eX30WyO_SjwlC@0 z2KBtj&tUS`E8zbR%>6gmj$mPb)%zoGIqHLOJy`w-B!r&?F_GoZKteuiAXXWfkl_J{ zeJs-g3yOQ*xzCC#xEI-zrjjL*|0Vvnm2c4FU-3b?uL|69TXApD(@pr;-q*f;`Z2yJ zx7W8jujcc5`^=sd$mR=(2h#LKN$h^C9JG}`0{3~5LuR%ogIiht2z2!#hmCrU)n|RG zCrOT{>sy<8iR9-jvjX>t{MJYkVSi_toyU2OWl;Z5`KaFszcA~+!aiyCpMm9T(!I#3 zbbU0h7|&)GX6ezULj|w8_7i^vz3L_SM&-dWSASgGOYVlp>v8dp3f^^5KcBgi|4`}M z@rUx>S;1>hh;}Sj>T-CKr8``DXiPT@%@ET~Ayq71*xM-G(Za7NzcqM5OMzDqF8vR2 zs&`i6Zwh~Xg?Kl@of3EcvF+u^x#M|gyYv@1UN-j@N}b-Wg7+`_zgdz)toRAQCF7XYlB-aHzOVK-; zBH-QXr+7VF?e_5R5t?f=MmZw>xG5mUX~nI>*4@#Tn}6C>u1mo~lwPgNQf zi=M3VdY$)xv5m?Xw-OS(s{|KNh0Xv+Dav6hPOgN61--^9&Z$a2N?adLay6wW1MuMh zFT~I*+{`+qrneL`mLJUo=yqAMu2a0v#wu zdlkhZeiVsZU_rMjD=@ph8LCfU)t-2Y8cv;9lkc)$s+3(dhkiG-ttt^X!mg(>H zb$eU_@e<33P}{Sa-iW9Dd{T4?(4GL>915mdKF7{-zC?2ky8E^ z+y>E_rIPOJ>nGST;iY+J3vqJf31sH{oO91^?3KdYF8f3Ud`!JCf?R<*&+g<*uY-#j zL>_aAyig@pKEe1{DOt`5~!5xVSXOU=zCDh)D<(MaK#eav73*A9N z+pZ7MZ)CAOb`s+|3!ow|h;CmXZpDlIaiLiXpdzQ{^1KulD|!u#YAjyoU)X1{0LgJH z{)2v6W}gBS;O2Ta`i=?^W@RMCa0_rZ0TxF~kRxuz1AJDdh>q^zs9r|DNCEn>vJK-i z1?X`*;a-Ub2zp2L6KyMWixLgO!`p{FUO4Q99%iLrn&;(b@v3XP*T=UzX2rLATOr%# z`e^I1*f!sb(OHQ(QO6SO-+1!e+{!qYE4)Fb$@6h}l~eSzM?lEQesS+Pf>OH zm;XCvZZESKF-r!jHIyn|&icfwXE9$(C}%$J)~Nb^sdwN?-ekwD>glnAI@uc`6$kZv zR3TN=?z54N?n1FvN6A3@;6SLhf$Aj#?E=-kRx6|O@6@*HUyoMFU=rEK7m{geU2~Q- zzo~V;Sz=SmZ-|7q1!Cf`Y8_eQ|~ADFuY-@m~+hl+d5cwsK})^*iZi1am6gLwLHeWY3PY@gj4AVeOUm z5UW{c2fBNa7lWyt_<5?4vJ8ub7!it$$>|*(@9&#A{y#4;cm4tGqfzXmACZ2?a>YJs zO{?c-mMbzbRlJy---04VUI-Qo)rKB{pUM6FwfZ%>DR$=m8C|h6H?s#l=~ogTI9AfAt){Rf4`C&_FW_|;7md8Jc?GX9zY!;IZ~)jo*dOc~ zI-jQ!=WWuIX~U=G@LFw-s_&#TLxsF6nzt3QYjj10>~K1|qKiUbKduYKdh8W`XyQeo zL;YxA$!AB4G0ci_g|5|+sLn1PS&4_6*DUK&%SJT8{G&*{Hdj^WXWte(CG1+VEq*Sj zOC?+<@g&GyC-LadUnlXfFI*?_crR8bF=xF3Kv#;%LFGqh(UAYh_KL{iI*G?sYMsOb zN-qFxK=R2!1xfFzt5Dt(yoz4u^`XEumoGkt0@rI3{EN?FzZ0aW74mSPuud8O5_33E zMkm#XP=&4~ytBjgt8olR+iZck1HRMgagH3+e!Y>!KRu#ZZGqTqq+LRydHx7s&7_yzCwW(s_l3x=wYj){=aUy8&Wp^f$LVy!CyP^^Ey|KT`>Ja6VEg3t z5KTnpY?Yf1NuPWWqKC)^hbz-X=#!HmCW@@GxQ-Y3)hE|LtP=U!ylcG3xqw}>4?yf= znHCt4>UnpuPot;)?*SVja#SfU;cqMV>_AWYv;w&&^V>d$V=zm0U};q^@<$+-ez;dS zx8}2CclnSZql2 zH^sf|6MfTY00qN3NXW;#o+az~Mq(TI=3_iBcedOkfMcv37Bi8R z%=@=pE_iyvCit6@CnZqeS9W-EZ)D9&aNoHyMZy0}Q8iGYESsH>qS72PY82l&>aWOi zZJ_3@QZ2^SR!s8_? zkmd64bAq*FV&cC~rOJF6gvEZJhcSAxT>gDlu=bi5{XVuoVg<_Mt)LA0O`bU(-NS9g zMEA!P+QQyZY56tYifhHhwAy#ExFx*YoC(@@-%4x0iJ0jAtEJy0rY++&mX_bde&M!S zi^kM`{xGS$cW#9LljQUo>zt_;|#OY$C8+DhBx53Jbf z-a4C;D;6u+kM&$e@=)07<7kJ)TrO^LRky!L=Yq-eQ8a%AXtEAyN;ZvkeDPN+DJPkX z^jDI+J}Tp1fy_{0=!ejBV`(e4Ng0^KtbNSV`f2iwrON52da_t(8th(LC%af$7bB;b zBm+VZ;W~-2i`7Zoyv#a@+gG|y;s%zj zlemTD>m+XCRdo`#u~MDHjjU29aVx9VN!-lpRJc^ZnjF;jrO#J|tKpgd+w5zU*z9W@ zEOWWp*Ceso*EF%&cYR{B?}lJ5-PQa$*$u(sTFJ87TC>Dv-;J?m-_;ahMCgvcih3ia z`3KCBp#MKGw1kydAmmi9xKjVykhEn4bR5GO=h#Wz~U z{M9`#mBnxNMaw!IVyMW(GQSG(l1Qr!xAtoVg==urb;-2=-Av`d9|}`(}>+k7nQW z)7nRWv5y91d0uapEB4VIh}|q#Z1_cpKSf&gJX)+$wLGsR%N50H4bhBcMsRX8=h#JR zF!@I7es}r%<85@SSghF>lY8D#>K>81qs_i>Z8npI7FD}wK)XJ4jIP-Aq0%^71h1UdsFmL`)}>4))%R_3y@9OD(OEMw(+j1yW}%gR-C8M*pv24M_fVGk6VZrkTwY*N8%HE)K$^8GbdC2RG{ zG=JT8&&ueEH8^`mlc@!HjD2FgP_Y9;4Nf&D{SxdK z_5IAwXJ5Ne%=d37(3ZSX$B)bF%_L#ybs$HV+xvJEX)z|_?mUApw)J>cm_ z8qs!2|6qBV7ut-l=$RUoz3K}CC04&m^C`8{(|Nsv3i4|#NNOy|#)Ke+El5trxDpIMMS^claZARokn=snyPpfVvrmw6r|1cyxw31nH&qEmu6d#^cNFKQddFp?;*(2#2_my zNGaZGwn0Jm#DeI3*%l-_A;?#C6r>{ql%XFJgLGP{lKeI+ub1Ozf)oMlCoWkj8szJQ zAhT;L$owM&sfjSYPr`31$d>2xdMy;BODstF^3fpm7hc|+3$+yFI8O|R1||ksZ9$%& zo7a0*K^Db=yizV2I50u2qn?7w7d(D#$;vAU8i34Kg($$mzBU@=Fpy3jC6=PjakNkY|?U^{POJ zngX^@wmlsUav>qe{96>{xd=h-N(}O$1?j^3x%w)|*jSJ+$47(Qz4-FxjBBGHU*{&s z3yDDs?fkJikUa zr!gIHcJ7pU`Ep6FZKWVxixQ+F!uVnTu?5+?Jg?UnI@BQ+dv) z85u3fElVzM&OH|7naTwDBQeN93(|IVUN8L|LCOL46W<&QvMM3St`;iE&gukdh%mlS z#;sRL%C61pwO5c{u^@W2y6xCPOD~rsVnGVlA;_r2Ahq68kPD39%~p^%VnOsebqn%D zLXZ)BWXnsxyFNj-BnCNZK|W(B@UVjX77OyikZ6!22|>=^q#z3#6D0Mwgw1(|$Dq8_ zdGF@+%0P$e0k%){I&~|_^~)}APA?16pr`&e0%`CIqR`L_t33K#)v?@t?Tg2P(LDFJDb|(aR zxv_%e;vtt%x5OYNJ|swbDIQD-O^pRm1t(9JE?vFw&w(OuBVeVER?1o}C-wd}(!3Xy z<*R^g^G?-r(jUH$DEnejRKtl!*_;@~syV4S)}(nS73G&$lu4@Qq_5+_m{7^zqkBYE zoOqO-iBYVcliK2)H1BG}p}K&TrT*)R667(OP`6kV)p6oc4kbphicaeH8`8XNMHw24 z@~1^<-JK|}#-gZ_6OVE_F^bi6QYUOq^WIjJO|dBJma8nYdJyH8SQOQA;!(~gMzN|+ zYL%^N-X%qG7ouA<;0;AN&I3T9x^Q;jtC|y!64;dR8(Cc^weOBJuNmUdEwL!2RwznU z9v2D?jYUyCCmtnNVwBCQ?4+)xvNJ(ZrpKaewy3b0?g+?$G$&cjWiTVhdE z)rm)`kr>75JE`CB`jmSVrAI8vaf@;n4@ZTj$D*jN6OYm`F^W}qQp=x7^IlSv*JDv8 ztx{PQ_aVxGSQJ%u;!#>8MzI=C>K3|*Pb-Gjz+Xg( z0JcS0>lCFJkDP_-!NqGl@hGDcqgb^kwN`4ncax&DjYT7J%DluVR`E%FpNh|JML86U^8e^M?*OUE z=l{=U!H1H&>>bj3F9K2p7U>{e1r+rV6j8BZ!-9y2C>BIH1w;`P3)n?b6bm*`u%Mu# zsMrwsSU|B?^!J*{CYLY2^^bW!PbTk~%(FAQm)lJyU7k`ed%0OIUHPxviu9x>_Htj+ zi=FxS={kYbgq&^+*z59zc^Sm5b?Fh27d`8Vy{t-lu~Q#^P{TluQ!l4RUgkcnvCO*! zFAF0tdfF3vc`@n5&VBr^n+38=y{w4541Y$we8TmL>5n5XdfpRz*_iZVCqI7M)`5Jh zUVe$Z{9#^N&cI9KzjLpPp7_LGK1h18vmbwQ`#{=5PIm|Fby=ooKf!sI;pNoGi=O$! zUcO9vvC|)aHK#ups+Y?mFC+BKC3v125z{LoFM8?|d-);h#m;~Hy`2B7Q7`KwFMG{P zC2m_x{}Or8bD!AD-lP{h0rGF_1A8xC3IJQtGkO9P9La5u>F(&_lb_g2#dV3l0d@xD z_Zk?;VD&O8@^X=$0R>C1#LH!o7d`uly;Mzlu~Q)bvtfbERWG+hUb^VnOYki>T&CAY zUi9=Q_R=uv#m<5JrXvH{qFz3YynJC^4&wIAbiqHl73uj;?4?c8i=71dr*T7ZZOG}S zfW0nr^&}{`Y&Kp-MPBp-DE87V>BY{1{1u!9jaM(HM_#(~XIQcW9x0S5s1`@+-JI6c ze*c)t(yo`+dP?6Aa6_>9H(ba?K*i??qF|6*OB`aJ9|yE z1*K&F;>_to$UBk-;=hafM^s+m%g z+MwzftFt>fv+B*}iHN;O4{&0Ye;sxzmVFU}b09BB8pP}9%k|gsIe3-o)n5h^?b4r-k{~VW*?ndoW0)>Pb^1Hi*~Jg65U+ zI<#e_WM^W~0rJ5~1M$b89)ijno75-Pq5S1{_K-6f&qJKDC8O%Ud-juAjMv8M?4(x; zxuLh7Jd0RZ@BHUF?1VTw2!mH4?@k)T>v)yD@jA45rDPkU`V(?FFL4C%|K`NH7?syF zsZXp!`4>IXPYz-{7;(y$jH>s)YKRPH92Kjxg>Mwf@oZj!*g9_D1Ur8|?36tFHK*h^ zKwgzJh}Ur%?&5W58%xQq!{8;zZzK)){}uIzsJz`tePSKTulC2m{9eRskY*jqmW-;4 zzUe1bQPqys*-@Jdr3IVA5L-tNPR{eM!~QAAw#DEW$d@Dy;&rTGZ@dm|aVgn~sAfZ+ zpEMAE`zGH0sJvB4ePSKTpIX0?yu$bf;*>2JRnO#)^2dx{#OmzL?{Uu;oBNl`t>f$e zT!;NzlAVk}L&yV?2Jt$6W^cR>ZGS1*Ls5-{d~DJ{{N|`nL*-qP)F;-V{7p0Se^a&qe5BD<~GFgv4DSb^5?_;G0LvS;A_YQG!sf0fq&amtpAs&}p_lu3+}V|BI>O)+P% zc{gI~=)=EA`PX6pnq~KXRw$1`-jXzk*D)P;@jA3^rewEbupRQQq=ER4qW%+=SFm5= z$cc3*fBy@rNj>&AL7cKBqv`^#eCv#=d#ui`-B~Du*gONVb^QCE>#%U%YUqa<=N$L~pQ2yO(s>x2qU5Hb*WK_Lp%n;ej zn0ApmTY=V@88$m1wvGq*XET33?BCPu7yNtL2l7ctgLoY$;x1l?HrJHwa~PZpd0Nsy z{Dr7zqw#;g}-w%bdmCe5qTgR#VyPJO< z_78IQ3JfY%s3&cZ*dSg(Dlvl06QCEad)41M$0}9*N33DXCAaL-~_vq<;b9 zG{h-eGOC_mx1n6eI4@Rbhx}G3OW9n9SXj;ZS2_PW>|gY3Qw%mk{v~M;uVXoT<8^4m zP00!>uVOv%k@yJWfAA}Be^g$Zq&~3@<=_57Kk3QXA92c-jH+kUX(&fBj)~RTxqlbR zDQwO`>?rU2=Q`|5g6wG++yZ$`(jZ>Px9pAAq0Klsqek@_;@3mH9hLV>QlD6d z@~^&YpzK$P*8p+KmW--@`(JaZi>h&~&bIRcX~*Ufh^^xWzI4c+5Br)TTN#7#kY^J#fw{?CV%lQ$XPMVzuFqw03Q50Wn# zzm3(|%`TAN*{oAJw~oX3!Xp1V?2D4@gBY}gJUnR-ucJ+c#5%M=r)00fpak-XNdxhZ zM}00TZ&p&DScmeDekCh6F)l`&vL&PH2i~kM4>3L#tFzS#0(pVW9f;-Qum4|`7Hi(_?mOmQGrv-u!m>$r!ne)8wTzAVZfguzpgKTaCN>v)pA@jA4zr(}yz zeFORDq=ER~@THSiz&fe|@sShjQ2v*f_myVsZ;d!*OGedC{E?O(Z1s!P*`?J3Ih@V& z5c4*QuZ{Ar!@jP{UW~zIke4M5;&uGZ-gq6_^i#5YJi--yavmeeQKq5KzlQ0)_p&mvCQTsjPHy}z<-WPB%9 zXInK21vWy@nw1i5q4AYR9n?2YrG%|a#nS;Ij3LLQbh;J0Vg zC8)efNqu4+${+A_Rc<=wH9(xQC8O$2H3rK9##>``_Vwn0+{@-0h~?u{zP!u74*SwD zyBvcLAn#2Y#Or9@BC!r_8!Fk$FsPK_oDYeQAbttzMyR~bNqu4+%73Cxp&Y_E3~|bq zjH>rGZX#nDC&cP(hFiMNW^+DbN2&Vkf33s5w#acznnCP*Kr2!;(TaJQOUl7 z!P}6xB@M*C9rZV;yuC?%VjaqFxwfkm7V#P&PT7)Cb-lMLOH)(_#Omyk9Rlgf=CO#a z2B`WXbq&~3@<-ho@ zllK_6Ax_znQS}#%yU1?FpJH`(Q5TL?%o#pn>zK6Wzw%*U-Dc0npgrUxlLqlRer0dG z4sAs$*~3wt0Qrohf%q*^PetWjm((ZLq5KIOYRPSkcOy>Ol2LWHZVhA=<1?{3`{}`f zY+!R2V(X~++<&dZzC6x8jlmy~>sCn^#Ovrz!_s&i+NM;p3ovK}xl__W{4-JaMdclt z)F;-V{Fj^dl#?0HK%BBAqv{pA8^|S$m&fYtjJ|=CviUe->v;b^*I{2@XUAahBIGZV z2Jt$!u{U0aHY}CwL8yL&{CCno`~Y>O{aHs{AU<+p9m-!`y}Pt!e<#E#TQaIX{C7Ue z*cuwEvkeCYax|OM5c9U+`Txp?eSx0+dte~fKz=A`5U=AB+{NqA)}@ktAA_eMuS*(; z{|M^0QF&h`^@(*Te-@9<`<3x;#3@@cs;<(qrW90VL_&4;rQv}zV6#7B>lnpX>G?l* z?JNB35)4K_K0j#?ucO|fiFIfbQ^`)jU?$`_Ndxgmpk9Q^dnl<-tV8*J<;LIeogGGZk4M;i7qNUi zz>gC7^I<<@WSe8~xst0V4B~a%%iee$+Wb_q9;#x|+e{>XYe^g$Nq&~3@<)2=o zqa4b31mct}8C9RWx0am5czUeP-Y_KLp2 zn?5h%&NjY684B1!Hucx58NS(3B*&aaRh_?cwX#xsrnup;~NZVm@=*S zL%%4Jck~D!M@I;cd<*(-RsTz*AF2^msG#jhgCe@qraRQ+9QgD3f&7cl%Dv{dlkDY{ zr1qoNH;@CaP4`yA+OuzMCkM<)che0CE}33Q(nnzy{s#6;s=+Rn!^W9X;hEHdE7&b3 ze7K}Q?o!vMelb+E18U{J*0dFkyYmU%yGxH@%~aXB=2r(Y?`iXO>sv~X;kA@=(Y&n{ zq{m^lPVraItR#=(E4=}**Lw)pJ=p>49FeN^VtIbT3ewl{+EuOf-$$I>yzGdRdzBsW zyxMle#oxA)9#+rfY#+R1`?1=ihal}*QfM)4dhwF=&(St3ZLBLSP(TGzE z*BTH%#cOyXv-HKlJ|Ul&Y7mCp>QY|Fc@XRcuHbq7e$+Y&y6IBBMF0Q4cofFjRv5MI zRcxO+Ko?TmUdHyRezsw+V~11?T~=zRRpz{_D#=e8&fX}TTlLB@e-!%4wb54vEX+{< z;boPiF)D9hq%V8LTc>Jk_?G-Sty2e;sHcoNc&9@nIhyel#HqP!ttuCpph`8uqy4q(SUH>{~CQm2_$=W<(=+T6IUyR?DR z6>)0e3IV>mq%HU%%+iMg=3ps*?Zxj;v-5GPKPA#{=WIs&A5vB10@crm^i52E8;$9& zSN+YAehO!!yrpTbzg+c?MEW;4+ZO-#Ka1r#)o+OO4NSj`ha9}8`p+W$d3vdXKavME ze5d-~BmEP4Ifno5PsOrdUGy2i;;&@-2YG}>J=M2}^yBm*5dRn+19Fh+dq?_}dZCKH zACEFQRP{$k`u$A*7>{E)S@mZ}`Vzf7#y_4%%1l@N)scR&Uc%!S@SvQVRKGORf1(!% z`A^gA<6+gWj`ST&Kaocmy`uU}k^VfrxXIuDQ!WHm{a2Cx3B7d6f04&c{i=GY7p>Ry zXY$CbLR9G*k$#w7u;tg@S}e^}-!9TG)r-pf4Lr=Or|Jhq`k(cZHUE4b$TnK_$4B}> zrf;;lSk6@alt{l=FSqmGdb3!rRDEfr|3ok5^QZB+zuQ!QZ=`Qv`j)S8A++jWi1a7u zC4~Na>x<=0)o+dTPuSfYm-BeZovPmx>3_FdMe5S7{2$d?xq?{dIw9)uVF zy)FOtUA^V+8QgwoRvA6wx-ElbKI83(Q(Jclz+bAK1+{Ry&95bWQxByoJcZWV0GQj7 zrgbExCX!hO_J5|ae89K^amsWr0aDWAt6FjtH~jv}mRG$&&iK!TiEALrVk&<#FKb3! zl)H0~w1C|eF2)kV5!yXU@Wz&E;wB;0JaFqE(p9i_r&PVIr1(@4D%}A}9NM;5Y z^yny$Gd_ z6uKchysm)xo2Vz;QEIQX)T-t6bvwym#v>4?7OoW_{^|UYI)PdG62KfR!aCMarr((ORaKpslAYgH$1My=OW_U_Gj!2t`py(`f`nPsnrXmEVZgGXs=h9 zTLfj7jh7bbe>;|1+bt#UW79n~v8WnOye@!Q?1hylv1XIe<9)rXrVL^ng*df!?Exv- z@Ig)7)harPE$>Re+|A6Nlj|%e>f*86+sQ)4dl09#?m~qbfRqgXGR*T+YSy%G%JMDaUx-s%_qxKn02kSh;OPoY@Yet^f6wSYh*6%KTb|Xj^Sd3TJ)DCP zrxval5Wfa*xFO8aCjsW*I~+uLKDiRt%J8XH^%})LRlPg0g!IN{jTrbIY^smXh*QG|%&k{=vkn-!vEV zm$35W*=#a;ypD4lOFP)z5vR7UCNn9y@Vzk4!`brA1T26q`E%0Oa-uHYI=!J>$~YHs zYU_?uI0cZBEnkFrzJo3ACBS?hk>BTW=2QKgIJl#{!}tZ_)Yd(w@GKzyo!n&c12eBe zvz))Z`o&C?=e69GAsMy2WOYVr!fApywQzr$gO@pp?0_ykJaW)Ck>@JxO|1sRp9_vr z;syHVmuDJeS>&Qd*?nCRv_s zx0F1_T`S#FuNLjZ#B}hJRfIyR!eVXJxR6Tiy_3Rh1(8@e=g_rwa}$I z1Loig97ONu7kE@r>Le)Q9}M#<&V`YU^exTo3S1MDIY>GV?wJ%-?0atnoYgFYfq~ zj9Qjf?=Rmn{)#xYaBl$O*XR9Lt`)C;eZU;Nf`cg2lX#?7su&8lKQV2i#6kNIEz`=$ zW%>XfA~sry$4A6n_J3T1)tu+6emP*7)=RuC)2c3bI?i<2&2FVd`hC4++IH(%5=Pxq zJyJzaW8{4VSkza-%k!Q#tBf9Tc=P_Ui}5$asjb@rNXa*EhMBH-0Pb1==5BQUtcj3k&MSrY`KD!rMrq%M@Bun*BPt(;t@xALEPZzT?ih`lY-!?_vqY}JXhla5vF zCaNy*^|oD$pzJ=}(jxsq)yl+nOUaSg9G==)^eQIaH-K4e&0i#DlhNZ%zNL@+#kk*r ztXyr~4nRsS*%s!c270d>VD8>ukoQmgcGQwl7eN&+0B0;goZ7lp3LOFQZ{VikQ%PTZ#*$IX5vO#LQpQ^mrxxxUK>TLB(e7iGUI&I45N{ zSeF**kM))l+s#{dvn8ccs^~b3ycvK+Jpo>xx2{=b^oX||Jy>pFT#PuibyES(*FOrg z@-SQ8TY$M+l|L&lSXR_U{c#QDGsZoLQ(O0f!mEJz!?+*(A7)^{R>}E2mpl z)NJ@;i(ZN%cc=hJPr#jG-V#8<99SenD`fH<{v8D>&)-3MV-`m^Po1Xu|3@@M5X%Zj=< z?%Jku0ppd3Q(HG$;W&W40Nm<7pPBanVE(Ekv*Ii(YI(~Kz2#ZP^@vjow*(M>24DYf zVV2$ln1iEn5WQF4^(uII-YaI6(Iej5 zsH?1D+<-W>b;|)Msk$}H%7<)ue*xz1ocvjtW?4}e7xil`ekXqQL!8>WT?)Sd;-B1r z-#^*m^#RP^M7@$Kdav9?#|D1vWZ&O)4v~?JV-cqot|K6R8!G#BX6Y*dbFc^pQC2SI zDbJ|}Q0RfcbiNW->_fDyC?}VdZa)^w8YQleh;Q51^KEHyv0L>8opV{yyFKl_qUwTK z+S|e+DC>ApTBMh;SypT}Ss8-S;i;kNqK3Hf`T}Ot8(yBQm{mrPc>H(mWEA5##Hp=2 z7?6@B?}b@8n=S7Kz}y{{KPx4c6?L&|LuWyN+& z$w%0XOw})1g^BkbU=}~Z%99hb$>{Np?Z3bLkMT#usjb@t@V)E%sajGp1L&P9Dz4N8 zr!@g{I;k<=x~hR>eg|i*Zy~MN*BNojbX5Q;X}Tp$>p-@=Qvfq=lRvFpEv@RJ$G08i zBF1YFr?&1Gg%berw{gSELT28>fcXoOY5j<9FOpHq?7{8jdB)cerxtE0ApTYSaSZMmMEJdJV6o)pkqCC)kWk9aA(16Yn9wLb)DRp0t`xMvqs%s7Rh; zd=+tO>+S)ho4^YYztt|ZoT%l3axG*O<2b~rh3f?H$;un!3})%80CR8?4x+c(%8V5{~jvIxpL z7?l?3_2HHi+bt#UVKXvyXHf&3czpn~xEofUoS02UkM}LV8I5Edi#WA)-2o}-{C1d= zv)J-V0dx2D{Jf{tKckk6y7*;Cf!x9PDB{%CU8XP_;0N8xfjrO5dk-*wPx1o9IjLef zQOlore%JpPe?**GxD9~#pYP%Y@6Ib#8!!jo;2_G$f$UAas}%nMu%#0F??belC?}Vb z-TbCIQi;b##735r?ewCWsrowsd!J})2Fr=63p!~hD~q74HC1Vm-g{sJvzwvnBTKOjzR-CKZ^oOpdL8I`J1oIV(L#elgx5qGK2 zHJ*$b)St`+5bSGZ%QzD;TQ^c? zV*u_-E6?SR%)EO5^Y_OSiG;4WTgyT%8;(0rRx_?eoLabB0P(Nqr}oXv(z^h2&@++H zZC{h8%bT6L`$f0atU3;kMWJ+q}xQq6A6h6Chz9&oa!e7mQd~3VF^`r z!KOH&Wev+pi}a=#OQ`LZlC9Vrk(yLA3lncSU@>0{D^EhrCZose@?&dxlJO*R+6ookVp?*6eXtHpeTmBZ88H7Fn4~rWl6eO_r%e|r-;>VKh*PHfPWM%) zFh*VcdUY*1Bvs3`V6WE&Fc%H@Y)+|xWU7JVH+GUijH3`^y7s!y@AG#h8PCi+8!*GY z{H;!T`d1;Np)$z_al5T{Ic8o-wjzg3b&%)I*mGki`B*8*JRi0;JA)HUPFh*PG! z3*aTVy@Z!QwG}E9jv@Y4Y!6QjN_~aO`wcK}uQN$)0wnVuIFn}o6?*X_DB_gqe$agt zDlEdfq%)u|!2#^`vVggGhsKR5HIPhAa8ng0Lm5XS#&n%@pZ`^s>z|l;lL0e)L=9JI z+sBNO_rIutT){XOamsX)0P*K{tt3mBc@F|+c(EGJFheCTysWLPVcdW?WxC}6FTqI8 zk}9gULWP%@u1K8<9-cZUo!X7k`x`KSZ!$@}14!lraBi*PQqY_GZ4swT_p9#XQR|qO ze8s~W^hIdLUauEmF3#k1IHd-XsRzFGd_OsyaSUQi*G>2N!**AaQ<-^F05cq_hKB=^ zQF7f;#c~bfe8ef!oePM6#aET&PG;VtfEoH~SY(Du-Z{Ufyu`Q>amsWL08;Y(4f@&E zuJ}{7yzc-rzD_^eGTM4DqXsulI8gp&tkefPWxB5bem*(7hSWgkH3!VFj~X5VNJhyE zgO1Xbu^-};=^6r3GJt1P=!fEwYOpFz&;eU8IpY{s=DAQ zedTQlWvw7fi}a31`%tvqQql*T(W!<-i!kw?0nB0+tUMozW|Pt5?KhRD>@mKNIJI@F z0Lu6ITPfdDfFn~?_x~Q9_YYuBxA94vQUl3+4E8;~odo^rNiD=F)BUddN;Yl`)7l!n zHykkI%k!u8dP}Ri7~#Hp?8sW1=_|Mfo$Wg0VY9$^0F@*2h`FaNts|JJEx z=T1H3ZpIaeQwui>;DhB${{Ll`-U^t5mv9iJ^&s}9J_5vF3GP&4s{j8bT3VHpOY6^E zNEX0J*NljNKCWp!mA;_8RDT>`Y1K|ymR404^w)-<7C~8?(9$BkA=lDsySelTo6)H; zMdx7R-3XXP23DT5noUNJ_xplwau?&nh*MiP2OzB-D3g+pmS|er6u*Mb+X9%=%G63q z4Y<)89P%Rfe>47oIAyxG08Xdg2-BJ#fN>MRjK}9s>lv0-b@Aoy`qCNp0K}=StEx~B z5Wm$gg))kncN$>+HuDsX?TD6UurB&4hpLUGW zTGm3gv`Ft4x3t=BDR~f^A*oJ9gD~+<1uW)8u=1qUY%+ShO6MIcQyFI=PHo);KuSJ- zEzHS6w!GDVx$BibCqpbJ>Z0xBgXLAm4-lue?mmT;0F9;Tcl$qP-aml(tCGw~eanej zUfh%~@(1$%N1R%?J%IRqd1JIdm+l9cgQIZ}z12!T2?j~TABghulm;j%ZWC~vz(~9V0ot~CuO~XON+D@f#t+@^P?0tLsHik zZNtRd3s}tO!pf5qv&rc3Hm&U>g@fuzO~k3K`wif&_WR#8rQ{BvSE@^Kd-i(00drbQ zza~)w$D zp>Per^~Ll;evX;91u%aTc@5*Y+B0|RC#YIB^{dKG#vc%;7H&Nt{`a30%fHOhwFc)L zEW$yQ)*kFlZ3noU6Knw|-ESYFrByk(wEp!W`B&nYi1@Z$Lc3>Cv0SG5I{-_oc2Bgl zs=A=PwwbmibSLUC0xnLA(rUY1Axr`2pSdb}0Yhsl?Wdl09# z?oB{ScCQa};tj!FW5C?~H81bKp>7_MfD&s7~sf8N@h`*SNljku@KMa_I?{E<1?}tUPZuv&rc3eyd+!E@Hd_acb*M062GFQj2qU)mEtB z*X9r0V{8vd{gb)}mG=Z-{%12u%>^X00L+Z*F6$ZJMw~L;BLFUL#=4{>b6~L_y=o(M$CjA%> zMVvBSYk&*=W>u79nR%xJW_XSoUSNhwzNBssnZ|e(;*{x50(c49^XXMlwG}EXWm+=O zKFy9y-H*z93NUZiF-hGBNakj+^MX3^8siqkDbqa$;G(sF zX{FRaGNoXrT8GF>j2jWBO!ojFCFj2qzPUeT%acQM#vkN=bMHXSe?(B<;J`(&8zN3^ z-46jy|Q09L0D7;?%-*1;iiB$K%<|($@mE zns$jd_rec&&mIg#{3_rrN_>1DqP@A5lY4U?`ChTSq{KHO;wg4-!VsPj@T2OB56iu| zwRNt&xm8_|(Wbi=LD?e%N{h5tl)brax0I~FW>{)vQ4^ea0|2ww1}o2-+iWs=yyvd$ zBcmC|BTj8yFF;C$tW|fl-DI}B`GC1wkUu9&Ehp+?%qeYU8RII%Y~3uKT@Ogf-gRM~ z*R$n)1(?q#^ZR_se5#-CyH=4u7%Lu*C$)7SC~ODBuSpl->g@2^0_JZ-GS71@&uaPD z89k&I;}FEDg=+wa|Lgn3ax}B_WWXF;frIEnygz$WUjyR50A8lV#rqH~&&tW=*?X7s zA|*Z^5udetN-pGq5+ADmPryFJwb8QWS=9yq#(6G#zC>w}_BOLT+iodYjm@ysJ4OBo z7TpH0m|usLC(mY+(c?`!CM`W-4@R8Yy5@kCG=Cw?^BA_gX@I#Knm;F_Ehp+?^2_Dq zI>sf4Q(HGl;Q~NPK729E^9r`SO@R5Fl;7tR^QnG%_9>Ll8Gk^W+PW7NUIS=``7Zsh znR(Sl=KL+wc5u=A`2fqaT3#`Ipfrcm0dZ>KDlp@(J$C^0W|kfUSU_8G5asy?a+bOY z3Xi7%PgUX-`w%V9%E{$<#pYtUQHhHqV#e}(_dMR`s^1P+p0zEv7t>|Y=ysD#eF+U9}PoB*tqsQBGN?mCNyFKF6)&|3#PX2bhD_i9Fx-HtFjJMf_Ue2qm7e57F|hoLrudeT)Bplz2r%Jkjn2+k6A> zbJf2FSe~_syX9Hc1^?d;itCm4DE0%r3%yYpn) zoM!Zh*9>ncnIpLk6>)0o{s3@z;%apmq}+k*@eT*f;es3Uo-8j#Eg5w(ty6s&&v*`E zwyuxPh5%AB@tLrGE@#WT8!(?w=J)xM`BXoLHciPY#+MPNwr+vKZGe>g_EcCu@3ZCo z4w%mm^84IjKGo08BkN1W(fsR&IJI?OEBpwE|J|#F(wrS$Pr&?jPu9-_s~@$TKcT7| z##n+lwQ%hK@%L`xdqHOD%K>vR0|(LP#8CF8_5eIk2fRUv5AH*>`cY1-`q4)TSpBHF;9-4Gl0{JV_@vSz?V)P*W4lTBs~8PSotrAE zha0aKU^Y|X<*6UD%IFbiFKsJ_Gmb%=+PZE44%a@GtDm#j_DTVB*fM`!I$2)S$vx?c zatGt1h*Mj4nZj&9N?v&)te>@Pc{>2}IW)h|(dJYAtoy!;{G#lm@uaqHi^3-WnOhas zPjz;BT>e?NDK-pee#7BB}d;UKD?->HSv5-7Bf0yish*FHq6ALZog z=ZV+&{ZEPIj>&oa%I+aO;CileK$RW>*yoQv(8B6R)dgeq*B6VR>=|CAMcPZ)>c@7I z?!6cdOWl+z8i$d0DPT4&;pM3xv&!fZo4q_pN*QlOoZ7mJ030@2k*l8-YIbR=ZrrfPHo+b3ar zVE&%h_S8}REU@}f%NyQF$yPH`|V)diyg7x~yI9o#5gXBt! zw5PV!kL{L{Mc6!;s#&xc6K@S*7V}``sUEY*=<#+993*ctetdI@eZ8EEWHpg2Z!JwdQXphm6!TAK>RnsWlDTuAEM<%Ik}u;s%%UEwJUKC&j2>^unYHBv#z}}%TQ?Mtl5P)%Ik}WAZz*8zCg;z|bjyjl_bvG0q1_*Xj;t~50Ehoy!{$&2@gIjMr)>jap)5&3g+tmQ;q%o{5*fbmGgsjX|O&=wH?GM-a8 zftfcIFn{;yqYk2+OtGA(W$L%~axLRch*Jx92EdmtFBi+*%+k*T=3pBRqMSUqj+c4_ z6z<>yH!AU~eTbG5<>Yd541EuNRiYf9^Jsk!g5~;Xhe7BpA92rvMrgvG2iEAM!uC|1 zvL`)kPc_rV2OZFM$aOyOuRk`B(HMD?BLDi(2QzM0#jDX`8U1QAdYmT594gZpuR)wz zxZ?ovAA6DeLYR5W0b8;*T1aiRyO5Q<-^A+jB;!knQ>I%ANHAU7#cXG4&0$U;j?VfR ziK`b^A3+2Xd@_#x`;U+M^=wHsDh(s!nOj>@2OiIE!lZf6mQ>H%*`WU+&6FBp;PnSA zlsmED)gLd{bgDm_xgFt}POYbF+og;;|LP@Aj$}Luacb-iQ0M|kaEFFCi>Zb92R)~7 zcCIu(u$nrPV3LM-Kl{TMo0dYk5W|*2VN|=w;A%MD>wrb24HM$eA<3wNb8fF9+ZeYa zPT9P_vD?GU`xA)mB%_p_BZf-l6F4s7lte=lZqKjJX1XO{bnBuJjJbvw&p~H}5_eD- ziYDx!5En|>gGDG5ZKDttO4vuiA52rNOEL0piu~&{M&d$ge76>g9%shZ@^TO3O2ny! zyBZMxv*(ND1!mq`fCZ>+8&c<4p(y#X(+`x-7{5iFGTo~H&%@FzRGyH_LS5pB3S}7) zNN~$B?C-(;a3vJVjxe&YP-?TAJz=4gupxd`GC33jZxUdk{E9_XD4#Ho3Z5SJSPL16W3daKy{7*xCn5h-Y<42YiiaL{^g@*VJ`@;|^lw62mp@dQ0LIyvF zvV?Ifd=YHRCDC&nz|l*AjBo(zMS zrAGorx1J0^=4xVGgw6^j?#FN*ny?>37(1URWzVspP_!vSSSVpv2EQC#b4XdCs572&%wf|_*&l{Tq2xjg3nh%|Co!|rru-uXam$3@<3j(fMVT*){GaZ2J1ONW6) z%+hxQMz^jI!Bl-L*DiEcDEdrNJ7;_aP1q44jGaO$drB9DqHQ6uj5uYw{(uBGX%?XXz0``cU?VAyO#05W_+VqgqM^kA>r%1z1$tK_XEo>frI}!(|5Jm55We z_$OgEpP6?X5Zg&cDb?p5Dk~UQAx=p&MWe06T4w2ufYGh3L~x)!1+MDpxk8CsN_-D1 zY$*}OPN9@N6pljCz7k=fgpDQqOZo)T7&l%PF#q~|I5XBl`Q~;l6g|%9i`&Rh9TBG% zt~J1AysL}lL}uP(z#M2Fi_{Lg4_nFA3oFVMjB^pEOg9OTV4G%P1yjqy2;zteWit^- zu(2QeKWBfq5(*{53(Q7XD2K6|jj&LzXM>+#=>Ysa242w_xlkHm5fw@$=24+c(q6$C zbzY^RNF(SS5U0lOcZG6nO7O3SIFhM_IHy-xp{O$nj@1w^WPcbUg^~*~ER-;+;q;Z5 z1;@J?uop-hP$UXP9UOoCP`QtB1>%$~eld1yn0f1g*iJG^dA?#V*~<6{;*`W2maY`L zn5BOOjBZ^if*mu7vFD^*p~M|2YM==_QiQQnC}qzeq)@aiMOY|dUkZO4?HPt)6NN@qq{pioAJB^(yY!#xs(vYQR@@5Q19 z23{7hP)@`mDwJW&qe6KDCYMA*dwgzH8Ok^Yacb-uDzpJ47@#3u%+x|0m@O+5btb_x z95%g_{c)k>LJSKfjOtG^_%IyrMZluc{uPNrQ3v1b*TKG+{%FxKPR-c}byY zSBtPv!nPLvLY{tkK1SZ0$iF`QGA@+QmS~~qaaPRdw_wJ55vLYzI>65{v`u(|nfEea z0csbGR2?f6C0}<&19^|}bHpjrJr785ie@2ob}kF26Gv1i6N!L-Il8jHBm2XZP$LyaI7*?1m_e z1|+yvL%f@*6^chM3!K@%$LdUi3pB)6*&l|;CrU2Fuu#IN?jeKk!|}cWEGq3_kth^( zaNxvd@)u+391gB*9`1!*z@}Fdi0ve!lp8i4BCQzPAx=rWVQFj8lUaHYV03G15ga;$ z7)#Mvp~NjMPD2y6v1u=CNY=l$rM(U;%0$jnrzpg}HnOkjiO5OoUVGz`2sfQ8~G z3gu_4qeAHmljpucJMPefWC`O#h*M)XS>aMZg3TJ@My3|xngh!UMV(2|UPJtm{b7g{ zN-o5(P{OE=_?s`zCKFW^z@pN&8i_(t2UR{hNSd&}72=f51v1!mW#;t-VmrwwC}j`yrBJllMp!6e zw++95hg+=2$onMnug~^1V=a{TZq-83<2-U{L;0TZPsFK(+XRR|=+R=Sa2^h-0~Vk* z_edRKg`(uuPxO}-?CXd)Wx66jf}b@DqnKJ2KBu8ZR49REVQgFW&tiYL5(?!|^lXHM z@@bnyq11!RgSWtW7q*YSy(8GS|keP12#CX#^PBFyiWjotsciBDwM^{qe58?Q!?s&+O2Km zd&Yker^fDOh0TBjV>QJ37v@5Ize!o4s51%f;jroc><>evP;w!Lg%U>fFc};P$2$qI zsI>V=qEOU9i#y881&q@Wr)=>jV>g?bHy?=Yc)lN-%iH_1jPU`)DM|bdbR>C-S$ZvC zbn8eGOqxcFm8awiWnI*byyjl zLV4pBEfhV@)lYVlK8(WL2$ zNH9aQu#Bk{%3lr23T2u0TnR30%>MV;9~a7wFtV^v{%Djal&)~a{|WpO1J7NQ3+30a z^N4+g{|WP`P^y!2$*A)yFFZtQKyQIKHFn=A{0T^KmxefqJr?4P^~(xHok_5thB%4+ zaiQcw3=1WUYRoUiaxon5dca;F?T3;m6m{@kl>)h)aT(&2%>x{;dyJX428iuw`po8G zPxhBL8Q(>ml6b??BxO6Z^e({Y)+8lpbTKh@nwl$=xIs!Gny^7iTqyUgjXtm`6z!4{ z7E0JA#otZ`lD-&u#{lO43p!22g|h2rEfhUYos)XWX^iJ1PA%L}fac%Y7lfI2BVZ1+ zD@&?{6^fF_t}m3k7#~KQGTj_Nf>$&PZ!)zkJV6}MC(0k%a3ttZkNt9Slm-6ubmavm zvanDdubU{8pV$!pVJr$U@Y(?u$`CA~Pm~VKqe6KbrexH4wskYI*zm1u<6Nv33qm-&oHk98O_aaV7;y2n;ES0BGC{+QYTi=%8{fmfkEIRws z6ZdTCf+pidr{>-6}u={{~j@8&k`|Im8ha%5W=` z-8IC>7E|gq|VWEUkts;XP z;dq|`7L_)ANfe4Yc=o1Y@&n^WqaiQFhEEH`86BbI?3&uaXi2h#~c@ISXucc3l z8SBZcTd1EXdYl{AHIio;*CS3X+!BCCi``5AFJ|5jzyj3PGpXaPP?Y>|X@B`esh47- zO!tXy1tT>J4NzJZS`$Z9D0dKn1ou{9|483}v|3Hq% zz?%tJC_iEm70QRqqe2-3Q!?uOz*7dwJjS~ar^fD7h4TRk&essvGPMxb24#h!&Lrrf zA%4aFFhtH#av_F=5=M2xcRc?Aj#qI;E-LLrlPDB*5KL<)wbvqFhm(_DilY)uo!&VN0# zl7*uEX~IGY8`St!{6N-VjD|CC-IW~_zc-K2%0$Jyv*OJR48e3E*W)xTH_k>EaMx9 zQ)73d!fgOwsA-71m|BR_E0qVn`Rwj6$SPav_F=5=Is5Dwg_iybgdxrR{MN zg`y4~oIY6kF%CwYvbnE-%pb+f8xO>Gl2OWm6*|bdj8hP&B;K%J(-)Xox)d>9 zKngGCPY~kN!hH{je>uHgnxpd$0xUpn(v!N`3Ps7wuBk2q8Al>cnXVnc^(LBybD3He z?j(+=PzGC})T_w;yV)NX%GF_HVWBLokSLT-*${sb7ArCE-UTd_MOZ|IGLv~!D9^#< zxsT9pd5Gs9GyZ}&HFnP^tOq1W-AD*kugHb?ak;WWQD+i7#9`A}_J<)-D7g^BLJ6aK zmJAMpl+DFa*v(+(%?4sS$tdN@z?WMY??9ZAXl=W* zSXMAgKLr@wdhrAgoJWkQS-C=q`|x}OE9}D)7s{k$p=dLnuu#HoJbs&0AOUW?W`OxW zpGFyGte+@-=WC(paek;>C|wx)B2Fz_4S=T)(*NRcX5MjtInd@msZLfXO5Qr6uAI$y zG2)czjsYZSqFK0=sTE3!IHE$iR=W)am!#Rhk^OO@ycR|l7RtX~qELFm75@+LBMiK~ zfQ7OxZ2DpU^{ita6-r%l&dq<&zL?!p3a{jWr-)Nyw?knMz-{pwVo&y1hzI{$7@wo4 zGfuNK#0l(=3ndp~SSVpsr~a>4&Vl1y0a#SpiX>4e>Y&=kzARw86>-WIzb$t6G4mb+ zVmmHzX7izG{be2FYlu@4Z&>=1Y-N_-4jA3~lLW_4CdO7*_}$!vORDBt079Lg_J23q_A}$0Z`iGoFq(wQzj^@js@c#zoA$ zs{wPM4N+2?0Ldu%;lsPjEsS>~PMK~dAi-&xg>_6V3&#^jR45yXfFBL_7V>Qu`{P3K zd4bso3uPR;*&rpH%r0SrPFOrXzTDM>(hRUr>R}OmqU^^!DwJvZp!tkCAGV^qbb&q) zacb-e6lwwzT&E#UWNIP)_D5Nvs51#p*AVBhKMavilw62mp@dOY`?6RT!SNmfEGlho zk|-2)AXkVy$G8r0%H}zZWd3bt-Zmh%+_DC$gtW*Xuu_Q!>i3o$H| zFsij=a4j6~J;0*Ub}5NMQ3r3$947x`+>JP8i+?qCe=+mQUz@O#j8Zo8_{D0d>mW`^ zykY5~(u&P=r-*JnRD##eW)0KOS)s)JQ%0i+`=`W(GB{Z%+C(KRl(376zkE-j%*M#O zBl16-4kKo)e|nC-L0?fl&i5C0l$DHY5T_Py9w2^~CB^bOGw(ydUI%U3lB!^ZqU3s? z)R%7=e?^=!-MfGUziJk$&Bq{y1;_gru&A_$OQKNJ!4tty$y~>uAjBz~ z2UB6!giWt45ZlrAj?LMJw3A+p0}!Vq@n_T6WfZgYSitDk*d@61EY`3LofS&l)a6n% zVN;hdc0N&hBnw45yM%=jws!H$e_JSzVdT9Q`Jek(qEJSa>MN?p+3`jd`G|2R;?%-D z3y6Q>t;O;)GcSF8EP&(i#E^D#qYx>ST!>+zgi+P~gx~++cr5^n zO1r`&3Pl}U{Y`o4#@GvS%BJB3nIF!~I|_*HB%_oUMr7m^#xoJ8B;K&JGMUCKeI;OY zYZVi8I+HcLiOvcoZWD7any^hwTqy06g`&M;!a@m~#rQ9LSt#2u@_vf^f3Y%AD2L9~ zS5%L)-?*%lE9F0a#Hoe*65yc)^hc?S&T9==fZCfT)!Pb1$$Og}AUzldAx@dD86d$- z&B7^6EerFBBPx_Xh(LnOS7c!k`{P3C$_va!SSUCCFHtC)*${sY7WZP{y#`n)L$HVn zr33S*P#%HFrNGesIirSr#P}`Z)Yv_$um+G|nTA;ThFpkmeNk2@>P&*$Ic&Ne`@;|^ zlw62mp@dPrMh1Jr@kRg^mA0x$6pA|N%L9Q@HyD%>ZIM$tb1fH|698 z#)XJelK4C5U$cx^dIex~>t7RmeFkgz4V@KA+`DEony`0GTqsSGg`y2?!a@l<*!cB7 zE0pp#5>6e!{5MS&%HZqu71iT>a9UMq1E(wE)WQ`4;y<^rSPo(49SN8NZJ3kV2JqZR zbTz)JBPTPSgE(cn!vG03Xclf@YK8I>-9@58>28H`)^@V6mi=*|ObsIo3+0E;6NSo!z|h{=-9*aGS|VJ9HLw;Ez6_E?Cs zKPf8|btb_-#E?FQ{c)k>LJSKfjOs!fn@oh`O$96}ZF`a^6m?K%M_sv=@dm^xn|6xW z-OkLr4~Xp~qm<8zo_{9K{LeNg^{74|`i z3#DeVP_!9JSSVpP6#qlog0#hrHvlmI7d@KzMA_xb%Fj&~PeFOc?4Nfe4Y zIP5B49%pz&FQjjbZ?WPhIO4v@tU$CW6j>5=0GxEQYb`fT*C$m4!)>l-I z^Yy#FT*^2bacbd?2Y8y~{9?J8nRgFh>(TBlsr{``l>G0M-m;o;E#j2v?f|$QL$mM& zQ_I4s#1R$B7urT7`0{p9Ys8D*tV+Dv3{bozDi$FJ+!7W_*tx|&?d3vw z8zb-Q$p4dcdWZ{U@GN~r^*FuX_2f_GEFm_va32BU-%bOND(JjMfCZ?HVp3lLl2P(! zpJt^aV^72>)71g^wo0=wj;Up#CvildC>x1Df~VJ$g;Ms%h2rx9vk?}`!R%%u{L^zW z8{!|dp-^tazk`RNo{Eir!C^t z!j)sj(=_SyaWFG)C}06*gdJMd;0So1LETTdg#5^jLr(kj^Ftn4F zwUs{^D=j53HFld6wgD2%)eu{;$3on?rmRrZ8CODZ*z{rSj|(LiVpu3)RI_NSG8T?^ zHegX{)0IS_sDsqX0=bOwD#R(9$FgC!fSGqY5Zg&cDbI}^AdfIUfjA}chNTnBI%erN z0i#5O+KMGCl&}|z-{R>)X@VQCJ7E61CktiU z`TC0Laaz?YFT)s*LY!K-HUPiG(SGD4X5M*#IndTGsg{6bl)SoARhh+Dia2GuGXV)! zY8D=0YFYUEsj@;DK?D*^d6q0}XMbEMrD0@Yq5SzwqENcQ75`^&4+dWGow-o%#3Cw` z+03ItsX@-U6d3M+j{T%D^p1#AWA}$bc{U~3q9KlAY9Y>BT~;XSOaiANPGNssD7g^B zLJ6bl@G8Im!|`qb>;=+JFo{A@2h&b0FZVM(f;eUKq&PDF95ZhN5Zlr6na#H*_mU47 zKSi99cm@USOLj9${{|S{+QS4loWL4--IXhpxHU{oG+}F)xKLhB7K-+V2@54`5aWOG zc%cl($U6}*|8r@cV#fN3^8R`Hit2I3mun#BGtNMqTDZ{w5Ba>NSgvR0Edk7d_M`cK zbe(sA6vfuXYiiJ1hGusNlCuh;ND@#a2ne_cq8KoPam5@yMMYFh$b06TcxJ_bh{{7$ z%!-PNPcgoS2#S~xb3hTl->K^9>E6{p>QvS3dvD!7eS2nhr}|FrMZgrLe9e%?W)A&h zG$&1W9bl?{)m8X_p}z``5=ZG3 zL|N`be2!)Fw{bj4lsLqYC?TqkxPlL&W1jP&nlQutIhuvq4 z>~}z7N6AB`XU%PB>|N9vra38b!&0-R8Pj|RAY2F4tf?A#Fy}A{nSbj^>eUQG66)0? ziSk%FQB<@hBuc1S6AgQ?(oDw4UQ+V^CjOH|dF%hgsMgu-u9oIz`g>_k7Vd1oM0;G= z*gVR}E&_ajDw2~s*e8mV7ftVF-lPAR=A`MK0Zi3kT?Ko0d=;7zM=4Q0^JjSFLtKSE z91mwgqI8B}CLvLpa+nF1gg=k>U}B<5uH-Nb>|uZpWjicNuP7ZDmlCBnnrsC|ZNadu z%_;O3(436jXu;lqsXAC8-o?--$^#FS6GhHURY!%mnByTtKBL4ThC~TbT})M+_tCLm z0X{0#%t{X}iu?ggzdOxI+th3! z+nKrOFlP2tz}q&-+DiZ5<2Erf=r5%?X;ZU}@Mkiz4+GxrOR9DF%qm>c=^J0U!>f-k zn&(kjLUS^3cLChRFu9CTel_6j#iWVuMgAiq>+Vg-Q|BW60gsL_EjZqe=A><^&L=T* zgPAv&{dsE>7w+NZaAF5vxF6__0GTr5RGGqk zm^KBx#r@BUt@9{W(@FAPCArGl#Q8WMqp`^C34qTg%|Z#F#7B~kD(4gOF_ix)%H_Of z{mtd{cb7b95STGWw5cPTHpC zlkp)Re`IFe{ju#4SzGDPJ!ebPg5&LIPTKs~AmMjoYKN8Vrl<1pBA<^^pZfb4vj_bI znv;R+3z%pgOdi81KN;{bi%GtZMSeaadtF+dIv440{@+c_J@j*EPTD+A(`>n+Wmr)=A7H_({bd~V_&VY#yB_Ag-%3oa4PzfL(OC}#x@26Hi_SvJ#+$hsK0dMi=Gh%D- zUhIFA{MnLR)pO!}JQ1Vi$gO!G&PUBcrJ=+}l8>qy6!I}t87j)X_`nErIsLUXCv6ku;phQIb|K*NF-amh$b~bN!Pxa1n|J9~rWKuR z-F$#A$jHZE8RgqN7`sxXAs!W6&pi$x@$ww6@3i%j{85QMn&z0TGT>3>N51IrIPR`t^K*+~` zFe0ZK5vg7u-)CyS0{luUgOs|4a`)t8+nV)Knu{B0o0>mHhkR^`$aVp|?H^BPY^DG0 z1zpT`^n+`)FF8Iv42=y{3_QnSKe)N!wI?t$bX|(C1@Z?a1e&uXYp-<6PHs z*j7ChuXje4Bia6bRTI;J~V_ zkpP7$$j2Fs^0P}CDrn@-J(G{BUQ{a|$5vOF$7T8g;4QAjL4}I zsI;VfoW#_g0r-_vZ7FpPucSF?o0reQD|xq)yc5ky+tfTUB;;csM0Tf=?KxRn>9>m7n|8;FF^f%L-4BROIKlccevl->*1KwUt^3@yi z7ZBO^((=@~NPjOjz|RyY5>Y1d|YAfVU&NZ zq@nUmK9)>AsuokNe2fNFng7ak72qvevbc$w?NDXDmpu1qELUZtI3Gu0)DF2F0{DE? zEL4F?d?fj(icTRPLj|a!TwZw9rsg2}$t4e(1P@NmTo)kZV}C~ERQG`)A7?PNHvxVn z)s9MCL%BPtXG1fa{$ZMvwyBvsAmrl<%Ob%p}9|`z%#U%fgr6Saxi0lz*dFouG z|IgpHGN;j>MRU?NRpXV9H#79Fk8z18pO3yoRP@C5qp;8#+K zsub69_u!h2W+44Anv=Gv=_OnK-?Jw(dj#NZyJl^rzqD5~a~l0wG$(C-P@UW9Vn+6M zz}t0C<>NL!AEo~B#k-hC=%1oF8Mtcz6IrhFYmD;m0^VLs^6R%Syb6*1B`r^#i}c5j zp#aVkdQ z)u#&i80uLSrZpi=E)mwr#4J&Z@}9vrW{h~ z{rU}G8cFIy&fV7RNI#b5WZ*UjO!Op79>6GH19*Ee$xmS+KN*p|FfC7=i}Wi$ZeXsb zznSKwZK`G}AD?3A^D!=X<@3=Oyo!2puCH;}egydSzQ~@ZY-hYW%zR7#GtEhx|9NBA z@G0!t06rg+B$9(%I8zyX_TV7X52c}LMJHR=3^37M5G1s%hn8$>gI7Sk=nM zL0eauf6DY~z*}s>;=IfBs4}-p{!mG-a!_$TPQhpia{Ddd^HH--cQ;Gvm(!dK+`j=6od%O@80CKkyuFy@X9JMedj`MFfkd7<7wI2RU17H1cz2qU zwyAnr`8bxL&&Rl~md{6DS1VexCHc=`dpzL%4r0BOU)g@T-Ogqj{dAg>Hb0>ZNB?4E zX9GSTlO&RZTsTu19B{!-W)c0HX+Rki zs!S`a@|ywP;z$*NiY#ZSjt?f(G3lInV;xR$#;&SM)S{r5B{ zZBw&x=a7%}p2M~|;BA{^ZKeO#yn@++w?!Xw zDE)CXCj&PE;Kxy6aw?_ zL!Xav`7WQ2zI<0SdUNuh!}d?WuXjcEJZ1aQcD;BgKsE01D{T`UjosEv?LffiW0FL2 zkPBxjgQJ)3Y{t_coK|$Qb^QPnt2iI8#%L{aTYMqTN6kVZu*64_k1FaF@-Y+vE6U}7%LbXA zSdT1u&?IWvFttYmekGL;OI<^`8`*PfGnM{anv=GvdA38y z$E%syS%9}aB5N!C)(7=B3+SJsIcf9dDB-`x$bJlXyPK&YReHa^!`G9N`kmdkG(XU9 zpg9@1cK{Pz36s%_Fxd?7_F|IXen-9;BD-x`o;nxl|8J*3rkZ{%%}Lu-ZFD{PcpO8Y zk8zbSpO3yuShO+cI+errGQh95G<%-1{pF2~%&qiy)10)~C4$|PjO=TG&&MQ*xRxIL_y8+&n3W51pGWn>gU$yeFaoZ}hy-bG#-r`SH zDvH{)uQIz!esD>y8d`Ba?uyY_$nC9we|^*})Ei5DB>AY?VIdzwt+Aq9o;G(2^9=pF zB@dbe4^Gb9g+R#18yJyOecFV4{G6%%9q=ouN?Ga}%H5SucjE^L+5b*+(l#|ysVb(w zXEsA*w*|cI6Iom7U*D-m(!fI`MnV3;|7L4ALC+V zJ|BHCvgp}XmNl_(4O7)qKIK6Yh9PJPoX2IewX`7mDWlKKJV`g6hylqw1 zR{GP&^)~O*FQ++an`i*xuVG~E;@Iw8Dq@vhAOGj8V@Z7;zp2v%m3A~I1NW1+ui!d& zLzeGXvKN#5umcPaM`ZU-%Twnf{r4X0Y9`SiM|09PRil-UmooJE7}q@W`RHq&MN>B+ z|2b?Q1pIlPkv&h@wjC9jXXqEvoU~0e6T1%?*)@RA$0Uj5AQ#S525?hO10BXe@51p4VE51IrIPR`u6K*+~YjL50ODeI=c zXRc>z?+5%!D(#lKhH^LbyGCXq{c|)YZBz3=gOHDJGqaxo-uBU~t@ICH-@*J!zk%kY zZK7U;AH9KHC&1fPrt-0g&qt~6d}hIHgUVorZK|$NKF(q2UmxQNZayD<1-Iz?0{PEj`v&0G+ccYxvYq)I zzkf!*n&za<&t|~UAB=4DpEw_rB$9(%I8zzyRd0ytj8d<(qLZ!L7?FuSBOj|7<&P<8 zs7RZSC6kY;$W|*KH#qh`%Jh7|TkPudv4XR@Qu5nMa@9AB^Kn20`yY|p6@br2%|b1> z#7B~ks_Pc=G1P%8%4MV1xF&D%2OiCP&?I4b z9|pYLeUuR^yAv+%3y!y=Icb}!dCJGZ%zZw_h2nfZ`a*HhxQ)nv4%>-f`N3tC_Br+$^ zPo_C(o9IOBE@5P60{-rXXHy(b`M7|oeHHL4sZd?& z8p>V0FLpG`=s%`8X`7m(O~}U|nOXOCY&#%pEB)s6yP6gpZ%1>|<~k96H>P%2$!=sS zA4mIql=?kwo0&c6C(xV>Twj3isbTUMM)}Erk6BEzZwUGMi0pM~dFouGKmYR0%supT zXinOu>OJM-5{5n><9c*HAALQ#Xzriv|KPCw2Jq`WmOW3|wpy{NG4Js87|ltW-{-@w zEmPYK@cEb|ksRd0naW`B%N1rM{kXKEldbCv@DMHeIEhjI;*y3+&iPm}`KbDGweoS{ z`bu-7Oz#A|#ZuM8iWY6CG!IMuY)P(aZE-%X!)Q5jYu=6XQL|9lF7c7%qpH+}d<<3X zigMYdMWyM0b-$7aO@ap}XYMl~*R6W2Ifniu znv=Gv=`36Ro;jbHy$%5VH$Y%eBFGz)oCM7B#>o;nxl|GsBGvmO0lnv=Gv z`c3&bfuYaGxa^(JM_=|XI{G*Ee{k5I4fuF}&z`4jw|{%Ixt#u5nv*tvgkbjoBfAjr z`Isb;9OS~8%3!;rwlVM0uS_dC*}C}vw=ntmE2Dgy_hVP8Z0BRii)yl_Hf2lNE zVwLX$c#AG9P6f|(JpUp2xRP8&;o^LphtYA!?d5=febg)z&`W$I`KV%eAs<8WyrNt# zezmKaOTVb(L6hLY$(cJ92>JLAMDKZksLpQ-%{@GGh8Ug{dkU9*PG&3Y+)fE#I> znhSmk`PdYZ?E-k)H?wuZq(AEByxER^FwIGucM8HE!^j>2c)PPw`FNhsN2&L?sgwCX z`m<8yaPDJ+6v^;e#(tqN}p5|rxB{U~(Q`Pnc@^LLg zpO0~MJ)e)hx?Z&SNAjP;w(7s}db?&hlI`hTb}$_{-kIj4&94n&H;Abn1NeMQl1L76 z;Y_{3qz$wyVotCf%MephK8m+1?Dw>VOz z;Gz{j@cg&rD@$_Ko{RIb$@i6J19IEpe{nu)7V7jRK9YP?&ApJ1p+;X(E>GEggc*YM z_>u=rf(IvO&M^=9*pd-B_3O7GA15)jX8?XBRrgC>L%D1CEBhbmucSF?o0?VMgnYb{ znSB)SwoS6O(*IB8j^<_hB{U~(6J1F7%NW`30dLnTm5=RxK1%)9g`@d{lCPX;P6qB% zz(kF>&TWw8I|JTcOtRk)hWjA0JEi5RbCLduH*aP3p`So=(l%8?l#gdI^!XSU@$>oU zi}*!dIoFFhY;OVldRJu6Q?^wH3@~%)=hK|D*_4Cb>x}F&z~^IKe-3@aOuNHXQFnbJ8|7w@~y?@7H}0*_}$Z zM`UfKzr)kpntkXe(44eQbS~i^!^oZs__LdTd-?tP3C7d{;L!KmZfmZhznSJ_;7$Qd zbOKDyW|W@~czZF)enaFhAhPeJ<*9R#{vo$FGoRD1r8#L+q+R(~`B9vYajig~kG@u5 zw2JD1Z8&Uu0Dirt+4Gd`roCF2A@n0@PTD4_2S*1ovc~{EACn}KgIqXM88o?XOEaDR zinOAWt*Zg}N{f8Fhf)5ul7`9w`dBjgsM>$E^6}f%mFB-PT?KfHKUJ?R`i=T{-%Flb z5zAFcFV4rUKdUtDklP`E&qvKd)xpF^l8>q&81gYx94yLZy<+GYNk^BindoY`1$VANTh8DD@YwY-=_}r8~{Zz-`3HL|c&`0~zH< z0)AaF$$mo^-V>2MA}vpyi}d?G-^83oe-_P2+f+?gKHkjGzdpt#34K2Ll7##*LH={t zJ`MQwx@7ZFwoN(?G5?`|m*%8xqC>I!nvq=(_o2q-2kKLL3e2gm}`h4`252Fps$$t*p zT>-z|ky(yp+iaT(a}a$E%}JZ*GT`V;M)qRBr&*Fja*zvWDucE2cQJRv>wOwWe~qfIMNK)Y#)$GAOLEl(jPr3P zj0PjO2LV1GH48Nq{i^bjUD$VuuH`AQ7P1S44$EO(je2fb$`h4^S7Nf^LApbdRKLY%EPiD_k zwx9gg#e7TuGtEhxADqOl;b+*j0en6tNhAlkaHcXi{JK7-A4)^hicYq!8DOFd$j5yc z<)@Z3R726nlF3IEOstiUFTP9tZ<$^Vc#AJoOX9AEkcXoxRLb`sFky1NU!${c2q2HH`8<1KwUt@|ztnT<;6~ zHU|=U>RhBhVZ#8k1;@M7oU~2VH_FGc41GSvbsT*@`Z|u$$#0YY9Ja>;-fyF9KFap2 z#aozZ^wVih+Wf91cK>2zX9GSTlO&RZTsTu1ET|e|7SX?%R&=s;cLOGRgnay*Q9k-I zcBN8{K9)>AsxD)#e4O@XrD=s#elx&Z4D$K-&!r@hnxRXL3F@kxv(Ah**1pG}&D zijawqBp+4XG2~;Y23eHLX8-MFZl#}J@}NoZ;N;950fc-!oe?>8`+q_{zRcAA4+!U% z%_+I-f6r!S4gL2tCv8*nwSf7bHLlKqZnH1`b&TO<^9YS9PdtZ(&kBY!XL=U z?hbgn=WZ>(U%$lI6s104*${Im{c$uW12+ON(GxH^l~H~=;O)gE`wfv_jmXYQ%Tw`4 zf5xp_n+5bw(VVnRRpT4U$K?!tKE~xDeLnhfkx>uMbq$B@pMYQQ@GM8N{c~j%wQji# zG$(Bn?TX#jOzl9x=VOvYa*zvWDuX-T%A4`@2d5RCY+XNqtwQAEX^iqSOB$*O>0`;{ zqsl+l%EtlJFuYHu^8jygg-X^%BbHQ}=Oll#Bv&!TI3G{LXf1ME{5sA@%|hYH#7B~k zDk2%bKKi1QMY-H#+nr5MtVfnSXc9a)Idi`NAs_2=gq%8%QkDArxDQi%G~icKxysZv zl)KLdv@}!c&!stOo0?IJLq1;3%+3P5?SQPU^iMjqu~|U>6wOK7MEeu|YmDs2fVZ2F z%Ev=}K1%(ocj}oR=r_=u4BR__i6+Bjv=%0t0p4CrvfmK-W{B*zX?f~gr2qS%2Bw;R zEX_&VRP|Lp9>>tnd`yx^4szj4WiV;pre-z$cWFf@Tek?{k)ze@fBc41?gn^QDm>|9$>gJ|OxDWB z8y8iY?PWR~@D@u|885n%N{qWpesD>ynvHQjzK79S$nC9we|^*})VEB0B>AXXl_4KP zZOfuu&e*SDo}quY}H z$^J*0leVe3=Y^1un<28>0^at_tgZAbPvH8~kEJj)B zrM}~5&CGf9m(rXJ+~I(U-hs)PjPiE_-d;@dAq)9DME0e$JasP8@Ad31=6(9*G$(CS zwU6>~14EyWaq&x^kG}Y2^!4-P|9ALp5BT+d&z`4j`w!g3^yGLynv=GP8p6>ojO+nG zHi_gQ7tT}$%fD`ACexpjR&=s;;{g*5A|J13lwVlVP<=}uOC}#x%(7NKu6w4^ESBlp zfVbF$#i^dbS$!<|cO|(>L&o{o@7YRI`8}cc1bjYf7RqQQK9YP?Nz9Osp@e2pE-xI> z(d>p%P052M!Gn`C*Bl7>*p(4E)tX3^kEb!U7Xf}H72ix4vo8VO_UWvx^tTz@&%95+oaUs>&O5?i!^qknV!O*z`FM@bN2#BC_x7dtK3>Yu=VM%B z)90hFu^Ekdiu~uWeGu^HIf(UAer0>-4c?@=>KZYvp5;`IY8Nnf?fPi~ChhFlxJ?lK+e# zZv=R`3SGwexEn^@k=xw>pO2b_qM(V7Bp+4aGvs3^2wIfOuQsc1PN1J&@}NoZ;N;A0 z3xs?e#fY5h`|psC*E6;E1AZly5=~u0xm&sYXtR+1IhvEUsp&W`}P6D z^5X$-FD6a2;n7NS7$SRWTAn%=>EHV5wq^$Xr8FmPejHu-IESHseT*xH`h4^iL!+8U z$bSyoHvqrh((HN4_U``m%?kR}G$(BnO~>vJMmAa(=VOvYa*zvWDuZ(t^M4VPdZiVe zY~99)cvysdtY(xyrlg^wpgxvNKB@v}t$aLbZl#$b)AIpu@wF;&MpLPhd8OpHmE@|Y z8Rz4J7(IpDt^j;KY8GmbCO(pUR2|WfkD=~pQ7-S>Vk={Q;}1NV_n=Ae;N;A`1cZG2 zfDt)$^@AZFw?NSj0Q^d-Sem+qa@Y8%(Pj+&?ldQDQ?qPN$j8H&*;4^;dvDfO`aO;x zVP?=@N^{cYVL5K6nT+hifVca=sO!NRuE@PBm4S0Jo z$wMZ{e?(;6?|$J^1IwE68Y<>O%HJ|E-4q&^>gVbZ80=Q@VNb|T={ z+a#Nhvi-K_&gKOA$uuWz6Ai`g5=M3=;9nn;B$9(%I8zz?`hMVpIxW;9J;C%mY@;qJOgdtlH@9eRde&^zur)PH zjj5=RH2S%dH3M;DANRiY48HuitFWlg9UV_qAAO$6req(kZ?(-8LP*71ZU1{`SDN#4 zJ#Vv(jz=8ok&Za?!{@g$J@2sn4xvhD#WrnsY-xJlX}9|Nz$&v%_t9fb&%12>T_!U5 zr0u4r=iRo^UDy^XP5`%={58t6Z2fW1aY;?FA+oM3w=-+#yZoVZ|I2uFIIeZ7(A@~Z z?m&pG0haDarskHM;;cSPaFO78!TkU%-*Q^P?1XCmSicf1t3aY>IjNMG4;3~8%PO~7 zmt{$#VVhcWJ8o=;4RO?R;aOI>OwY28r)p|uSXKxL%jYF5zk@jEt8A88Zo7Ay<$Lx? zv0SlVisc4GQmA+r4++b6J?mNiKV*}y+QRIP#wC9ruMWo*JQcdTU$6iOET70>_Xbl~ zR-DzV1m6iZ2pVx5mPcJxFb^=yF9m!qtL&d=IjR5mN~jwdSXL3qx-3f?T?)$;f5P%~ zz~_csc$QTJ(zC4NsY;OK$)goNd76P9m7obweC%Pe0tCe3m?$}A=<@c~HRNM|GEbkIn-T_(f zM~rEgYj5_Yd6w(OtHY&gi&(f*1?K~SFH!+1}#aTU9@T6d|;6nhG51m#pN1$rY z1pKRT7t+#iZqJsE*d24ei{S7oH19t*oqO(^tGP4-j`GB{7dUzX?djT*- zDWCD=F6Jfrw`fk9ZZ2S=Tkj~C?-}J=*!c3QHk`k_Ns+jLp_p5^ys9)<_wq^_b*8+^ zF1WF;dS6^#x$u`)l>+_c)$vq0!OZ1V2+8Gblw96vh;zQcVEOWXPuainO>`zz3zN(H z;i%N*osVsy;v6u!yq7)gZ=!a{epxZn{6W86KEAwl$ctcRdGtY8T`s}4N%c`GOmt{$#lVJG~+}K|A z;;7}qv#k1eo@E_R70Ju6tPm2Gzq-Hlw!0c}&X>(AvpkMMcx9HC4NI{+aCnO4HbhdW z_zVvT%WqORi_|+G*{OX-m_yOH`S0V^;kYE9LU#)VuKWm(ebDOhf~5iDN^_*-8tJj<%; z=2_P9ROz=2%L*Z3`Nf3gF_`3hF}O0zb1BhQW_bV==8{~#iArh7hnsV;EmZ6RCM-7z zEPu&jU1yCouW+p1#(x{H4#$P&6uRpq=m7+lKi7050LzNAdVj$r!6||n04#s?M8Ryv zBKe*DO60PtQ+bw?GF2a@SyrvHx-3f?6*jl#Y~0vCN>R&&XIZt%Jj*(sDrlBrSs^4W z-;%KWFyfpqaaLyeXbO*&S^jz36wBlLr&#_R+d{=7V8ZedPx@Tm4cUy12AdwPy*Y~J zv!ZUiIvkfzQ|NAn;2I#XJeI?5Hd9zuoYl_?mI(eQSPQ^%la~tSpQz^N`juc=HJ3cg zN%5sSL(#5~%c|g2mt{$#@v!_UZtS5Iany3*Sylxu&$5oE%5-H|RtO2przb48z+282 z?kcm~nNnP3mLH-TSdz<)`lMJMi*2EzC77`MglY|$;(Qj{b7Vhr4afdN^DNhmSBK-G zU<%!>75okamgjNUH3nc=aaMN`^brgd>JvrTZZ{&^|cZ#tCEmsIVmaBBb5FL zEUPY6U6v({ZinU3xUsKzU%dOtg=blHr98_zo+>_-VOb$0Ebo!9d=}!IFGp2oc_jsq z$}As4jjM#^*Se=zo`-Fr;%qQsxhk;S2HBymxA~pEUm<3>ZoE1im%38uZco9%Kw!BQ zmhLpBu&g+%FA>}*m?d}|faMYI70h^4^Hcpwu&nwxp5>%a&f{s8Rc)y*%aTTGVfi84 z*d7hysO7@5tZGS~WgSnIRLZcd5E7QRPFP-rIOmHjm06xi>7z2sU8vlYuzX?H6w9rM zq)_oG9uk%hRh=SJJQvw-w>CEiqcQXEt1mJ83aD%*ILbv#x6Cd0Bq zNLXHZU+I(FT*NtF5U0%Ye9F|6SsqH2orLAvs1}oa##oJQq2eJhVR@Zuc<@CmvTOQx zH(OV>Hs;YV+zZPv-)YlYl061YXDe2V_m__Ks7(d zuLR4g(BWB5N_9LHO1uP?Rqv%P%aTTK!SW}#u?JVhQOkvAS@l{x%Q~Ja>XKnuAtWq6 zk+9qhZ#iG)rOfh)6kjQ`yq;HxJFo#_cfMvy5y}6*bU-iN zFDbL!jdCPqmLH`eO2Tq;Dtjck{4lnKiZj84<=a$m!4z8|yXC_HW*vQ>Mls8E6y%2=IG!$j+r6!$dT0{QG!yI4<|0(A@&Ti$Gxc zEDpP6Okr7ZR(~hhAZWm$>UIDu*K5_lJkBt`0`R%48Udc=q*%Zcp~ysFSrsShvMgyd zAC@<10?Stc;qxcHJj<#W;aS%4R9T4(%L*Z3c~ru3H6}S@Oy{S34vwRQK-wZq>D_M*;+!v^P-ZztL4-2P7gLiVVR=nH#qug_3l$H73CkDh ziEmRJgY1ddHZxmo(%#h2d{)$rSBK-$3<}+yFSr5-Ebq)=cMnroR-DyO3Kk3A6MPB4 za=$GbnCYnIXZw|4S0TtmCN%+cPXH zgoNcu3CmLu=lrAYWtLn2oM!oMo}W*0dE<2{mgizys5k>mSne2DZh9*z<`ra*K zmg~l=!|`ME3f=7{I1mUdH^b7M%oLUtXZ1e>*9z_s%m-k(?~?jvcU1GI`ITT<&)#~L zlZS9?w((znhg{Z^+;v%&G+GVI58%db)iRD+EKs2uZ`*?LYemGpA zyZM6Wfxz-q4!i#{g=NKA{f%I~pgxDH+W@e9;0N{10*3kJfX`(;QR-Pv9xWXl9svz3 z>-prmEK3^o?rKewRZ0MSJEu&soT0NOB$U8%ctYU z{^EV{iXj)CWj!_PS=RB?qqP~96+*)D&-awx{bnQ1`G;-GEMNU;n&tL9!JE8doc(!< z<&Uu~R6GDCEU(k^fu^`KvO^!-$#iYg-W*2rEZ2=!hvNr%6}mfDa2XI-uI8}2n<*?S z&gv%wFALrgd=9|!Wna`c=cAgx&#%NQhMxZOEGLfv9vvPZ4Y{mmN$avKX|x!YKf;YY zux%W*TzHoCEU0H$$5RiCW>{7T3Cq(HmK)R^AARrS)Rk=ie;Aj^R#8c@(mxRSe}Dzq2hEfVfivW z%f;`oBOAR_PxBLf&-O9Pb>r3H_@PUM?#2rC2Lj7Yv2-Ufg=NKAeWBoL!EJ(h04)Es zzP{NF)%;|?61l7=wLHtoBU{&nNBsiJdhV|-%aTTi!1DdLv0XdFQOkvASBr*ZzIHIvhVrsLtbzYJe0q3*EwHSo+3K<^X*36x8+C-`%K(1|%7tfHPosI3bv*URT83qXkg(h| zVR<+vIsXt_ndQy?lV~Fek+TI*O^DNhmSBK-rz7)Ef zCb$F$ERW={yOSv_E6(c21d9Yq1*-vAK5WZ^ITzLZJ$@x#G4#NZXE}NDXk>WUD6p(2 zkLt24X*6;xYd*w{-G8$&YR{vW}-7FUqj25E7POPFODDE$1INDziNJ#Wc&W z@Dx*$%Uxeiu{;vnLPaAmVfl4EBS9fH7TbJ87jp^6UZ#1L>&C0Y@xx3C-K`e<00fq2 zbJ!ICSXP|X?FHQg{RLwHZD4%q+NP#uMgJ*HP1TIwo11)t%~^4zk)V0c_Jhoi=rvRr zrcsUg@71>R62`q5^vHQB#X))(%vM{C_R!>_d-b;UCD)- zjvHIAb9~}*;V+aPsqq&|$5YSUWG<9KNG{av$%T3han3)9Q@&9D;`x~J8|W51-jiIY z89b(v{E_|*wuOqv!Q?_!g$uPevUVqJYWflDiGLrj4#!XVD0Fw3U?vbQ)LtBRk1*vz zDbDJb1@8zx7W@S0Lb**^n<{q`Kfw7A77g~rQ4;e7P449v5Ab}lYRJKrdig*9CcZiG}^M8 zH7jsqCv=UYmJ83a9>(x2>v-ykjSR~QAz^usgykl9%lT(G$}B(1lNDu_hw)%Y!t(7r z;E>!|qp&ShGzAlu@6xt&Q+$ZU&OD)`xtwFK(mcy`o1-)pOw<+}0eaQxJRLU+3g4gdnnn_%hwk0~rG&gvP0 zYXr9o{tdwL0`}YQjjBBz@Ps5=<0Wpmf{pay1FCl2w?3h5&h*@2qi)&_-2UxE&b}uEzMl|MRia_g9lWcRVesolrg z*qnr+4aXsj_}AC9^w($ZVz$N2E_GZ<8gC4iSN9VT!f+hny=jWxo2HfQUZZGy>#w?R zZYnz3ZmZe1+j`fz{WKY%Nsj+O8a}aH`#`0sx>JUCW9VNXbeMO(lTDXHiyHI$SZ5c^ zJPzj<1OBg{LfA}vnrU;?7Mo}@adaO#?{nCGoYpZ#nRI<)Yx4vBI+~Lb|KL8SU|e_9 z+X3F?&baaSY@1HGE%%epqV(4^Y+!nzvpvnp!W97u)p2^&jO+x!JNRFS`(RVJq|*w; zU6f7!=bq+x`YALg12+z!Mm_<|V3fZS@b-K8^LUH>%L@?M{{sF7OST$IoZ4!vUCha~ z)b8SNE#qG>zG8##4k+1!OSZ{gT!~YA zakb4i2}9d&Lm2T7R%_{{D_WR;(%)Lgm89`DS$TEUKnP<$Mubt1A7NX>(0&g1Q%E+e zN}SrPs(r9Y7}}p2!iayTSWADiaSPL^XM59ymUSg*ypL2~-5Wp%VoiWzx77udiptaTuB;lw3Jsj9tdGfVni4t*hBd? zL;Ew}Pa)avC~<1LqjpjzVQ7bA2qXSgQZ2pR!Ih?culA-dE$d3scqgK~x(|U6#y5-z zV*;BD_eaT|S+Y&`14^9Q52&q%Nf_D!7{Z8uj8aQKe8ezwH~qpot|X1O>dC7+32a}uXE%xN!P5{CB3g)rjZAk@-*J9jW!<7US?t|X23vdOFa z3J78R!H6(!VVl~KDA|ijw#kk%iBmhuv{@|)Lz~D#81YZfY3T23Zl#{O+yd)fT+KwainX-gX8Lrj(ReS095vAkW^H}n(t&1h^8kf9F2k%|v{d8GhQnHRM zymyWH0-wz&k8s$&1Nh67>>H6dwQodQQj*K0ts>zv#Xqv4CC+Hj(`=w`x@~P&lEzy) z;Wa)WMhTIsf`udvyp_My%Zsg__r0bv|TvTTtI(Q9aoaZ zdmiM~?FfW0_F_aB=dhjOS%&r#z_XF;JdilG^FW&#k}$N{AcPT@)YsC#&8{}lcI{0k zTGo}M@umQIb&G)z#xh2PaRlG@SEFQ)D%mDq%}bp6YF^*^Ct>K@_z*^1NnK0-F*Gt) z(cfFgm89{v=JM*s0U?Y-84<>1d?Ee@L;D@zPa*m4TH@4q*ZNvK2}563hcMy-+FJUd z{&~}?Ki?M7vaTeJzgCu4_dXE9_>vJ}+`~7<<59Awm28tQcO_1JxvTGulQ8ssZU`f; zo2;eZo=`Bi(LY+pm89|auJY<80wIhO8R=fXr91cfOx;gEA@Q##5|!>oxh&m{8kg=y z>2Nn%-Hp%4y7l(4b!_49MvYC?Cpx}I>A3c5l)P)(MajFyCGXO~`-glKuX6vw`t*`@ zY~j6YOi6jZ6uyPS_F2GRgyj2MiBsR->Z{>|NPT4+h>Q#VYKeglw>MwXn*p_5Ng98( zDzEMyAQ1Vu^dWNpyY+wLpQrPGZV|DwQ4P1Tn&J35#RhZJwNr_8q1KcwOd{tn~>GcRT3-$Z_!KQO+h z)|ywySRPBp+{n$u+5W^ndAH)7sU9_y~@r6zjv zz$Ru(wCoOmzu^BYU+|4r>w?Q^i^YS?ZuI-pob=r`05{dI=d`+z291`76HfM>ULncePRid{Kf zb6E#7hyF2|lfJtGpv2N2_030&Y;H(=J0*JtBu?!a&~}96cG5P1a682Xf#vtG+eer! zaWkZjD@o(8^X1ij4usq3H%7Y1>&}CM*`nY~bh*e9_^VYT+w_ej_n8}MxyLSuE2 zPnPk-V4SR3En|5sUF7nbUTUJ7_HSa2MZum6_=`N7vx#3bn|-H?ET=35{^YNcDK@l%}PaWa0g)|z+ASRPC4me=%B z%HQwT#Jq}vT>*G@2bI~~?kBM;rzcL|(tJn1p5~xg3$1#RGn^#j+iR_Pk&NZBbdk$zdZ~#{-?xdGi-LU-@E3W> zjma(4VZAQ0oc?}O3-d1hN}7|tdklcx_0w4UMy=g$gg;*iJWY;y<`$A|<`#NQW_ty* zZ}y4V9W1l-Ep#981he=S+90!wYpr&Q%+l{d<<;U_sHes3`C6;ZmRb51Dz6r9p{he< z_FJvhK9O1a<6L>Q_~YCp&ZlA|ytV~AuM^9>4l^C$RZc&eG{SU8qd(0_-!%f*Ey3oY ziHz(yfWLW?eLfPW_W5WFQF8NWYft?Cqg_3wD8CQyH{9Gu|4bcMlEzzb< zkuL8aXLEV476pGG;pUORU*6TsGdGXiXKtQFGV8OGH(S1Ue0ke4&)hr%lFJ+4Jip59 z=wOztwx`U}??2_$;+tp7JZ87oTJ0j4rEi|{YT@RoIzVP0)LQL1nWaCxl~;>DyzR~T ze1?`?5BSUb?G4FobCJ{KmDAVn>|h#J^C6Drr0>23;B~`U@Or7%-it%sX6f7JHR8$KHs8za`Cyi;Hd|)t+orr)eA{eR53}EDt@eq` z(zi`{wfMFfFSA~wvgg){dFEqVdA0as+epr52ej;-famp`GOxcz;#E$2iLPZg*tLx80wDa^TiUoKacbk1_Jk$3jrMAV+a|6HEx(st)6uM?|D}#AN#i|L z^6G8@!fo>qBVFD<&*bvX5Cwg9_7q6qFK=7sncGJ0Gq=s-GCMk$E!-=@2fM!7N#AoXpZ6)yk{IAJsPFd?ur1F9JNTcW@r@2jK@=idQ+^ zV|fcRlYSP>N#C6fn5x|xvg&e%Fq&-Xdx?2@OW#b}DQxMRX}2nE>3fhZef4%>)y{yw z+mhXP5~p_GX=7h^1-_Lhg1V`1xWcE-J#t1Rz~do$O%yfg02To=DH zu4P4TDw#Z0S9`oXjWiPYOx}Tc<`qIqWnLlX$n26}cH zA($nrJs`96D@1v<_!XkD!|eB3tF4k*`W2$QTKozzTxPv@&7NCp=9$l<<<;WPqfI!U zA!ymX0Dleplr#CK4$5RXJ#6L}Q$v3u%}L*l0{D)D9f#L3vX22iC6mp85~nr?YTsd! zlG+a#QZg=lEx!{gw=*Bp|60eDr18Ezd3Cn}AtfJXL>QCUlGkoG!WaVhFp}MK5~p_0 zX=7dzhBnNFFyea0TKbwdHaACNbaowAlExd`S~Lb9SZnu zpmS%p3>P=9oqCe#`OkV=4`^C?uJNLJ%`VeUucB6av|cl_KmE~ZYbS@UvtS#*L}wpg zFsC!hp9lEFxk(JL>Nn-ZERX#X14--cuVig)ccjsAFkih45S}MbJFHF z-m%+_k=+kS>`YNAYnzNVN7J83b5b%<)#+^eVPt0mS*qg#s((RsEB3INqP!e+dv~*l z{@t{-lS6lf;AViBPA!;KjPh#%A5&*}h=xxsn5KKMct5~Xo$Q*EIJIkTSZUXssP?<& zJk{~8Ik|n{s%B;)MyJ$qC272CPF`JGAW+?d5n-5T8k_SN+G_zH-tk$g<+K0k{mm@; z2WU>(CVKJd#^wn|_C+AEGexP~aSng~(l4VqDX}Y#&3QjCvd#CbO?6yF^)IMi;WvKC z%Z$SZm|j>8O*$F#<8<63)d!JLiA-UfK8lbwMQr*;Nb zl(yuFYQHnkQyuRNl-r8Qjm={E6?I%m8t)90S2qI)RNu&m@V0)cv02N|=JtxiYmlW{ zKKFZc2h*72Eoe^KCi?!##-=k<+cRxvic;xw?-ph#{Z2F|C4N|*4TuvM*>izxR>!4O z|AOk5IdM~zmv4?AWMV#)5gAQT_$MQ~kC)MBQ#Fm@g68Xnag{ zviniu)b7V`(^UK2kDlsy_oLi4n=;h&#m(qCt|X0jKgz570SHuY#1X>SZ((DzFGG6_ z;KQ4BZ{`I=K2O}FlR1d zs`4hwuC{A=lVw-CUD!Ekrxnavl=H>CW9Q|alP7g*O{)4YFWXFRHktjCJh*W_?~AC~ zz9s#W&~JPo>1EJ(XvaHZ$sG^m{6OhQxa+<#y12nwh;ZI<}50N#i}0^6EALLWXo< zL>L!7-q@VM&|V7o@OI2*hV`IKapB zrKftrgY5r8WN!dG)yck5iBtPVH!kgz6xDv;sHZyKH!8PxJ=NJfNB@2uSCYp2M&;E_ z0|M1oG9tWr3mTi%4DC9=hxgGvnVV5QH>tFyuwQ#qq&aEx-&X9}F|}I&i5-9RGQI4A z{-!_u4m2kvUj5mgI-Zf80%Wr~uFLutR3FKSv*jC$9{Y?kSJU5_wsvypCJK%LO!Vr* z1^z_{)e8Y1(<$;0o%$&I9}(Gl`^Qu#dt)U|?Tx)6O|{<}>#2_S#>(xDY;fHgH#^pG zC271jR$kp#K%n{$Muf3)eq%EZC3_g)!#g8OwS4X~aI`sv{tTLvwu$b>?jMZoRX}2A zic(qrSSNEQ{e3hiC4TFT2%lwSKLxT>$HiIyg6bbV)$-D5Q77{!eboUeYbS?pvEV&` zA2xcN55j2Vy8u2W!$w*@r$5d9UqtqJz-M)N|1Xl%{@*sGjkn5bzyH@$UE2SP@{QN? zi&yk>>bR1$wEvf(8xI7kCov+7%6~UD3mMus03TlSEYNQ-4aia*S7`kUs-F;1MS0or@vY2J^i$H- zP7YmP!7#u?(IP$wGs<5H_?TYuR4-V}{&z(7zksJY+2bs6YLD|LX{!AmXHRv!$60P) zySSyvPhkH$E$d3sc#pHZx@Unv^-@NJH{g-RrVC1Td%%bH$gIo@h#T&DFes2<67W~z~4ar>~L z<{kP^)7DN7-K~NL026J#gb%`u@_zt6rrqTsdh@M<*>oa{j{rPB$v$q0Q~S7&D{Y`w zR{MS2p6YlXx7?ap1I%$4onOb5r13s(d3D`@K=oiogfZ!%#^y?f_AbDOw`-Pa`8@vd zv1UH~LYk8{pKY;Q%*ehCBzC4KmCu?FH=ojfMRQUz(ZAU)ZVy7eGmxb^uGRV%RKMw| zmX~cG>S%^wxm()W$)Rf?XaVpGsqgYZm{EQb;A8sGQ~l+C3+7Tpb}rzlPWGxxoZ73t zE={%HtL~|e_o~b7oRjLC59q(C<4V$aue!Xt>wrM@y^IL&ng<)3KN#BL!EtzB-JQv5 z`TUkrK^-~Xh32HqXIt$0GPOIT?MzWBZ95dqZuH}6PD&>FkS+1YGO||yS*qiLt$#uF z_3Vx}MR{r0tHRt*|3uo_$)TGpI1eyU&B}s#g;9Pf;Hkb-9-_vp3uYZ6+xn20>SRB@ z#Hs!G&y@D#E35r}d{1?}A75_&;7^_5xY@6cD@o)1`10y3^FVb|MuhR)1C7ljhV~4= zhj(F?YWbY=dw+8={gpH)Z9XJocPk@%ACTCYqEr?Q9A=)Nf0pK?#J_IX_WwR3yB^3= z9anAr3#vEOrsSf$Y)e3f2Pr>*kAs>4QdoFyLeAArH}3YYXNu zM0PshvpRX^K;qOh2cyyV|GpB{{+R<$b^OeM+~aJe`#4)lE%+f z$g6uF2vmQ`h%nmR*Vqh3$&Lejcu(Jzc>$46H==<#gnp97(&j_ry^YOeMs^yI*zu(Y z(+@hfHCNDIM{`o*u`k=e%wc5T0J!I`YI(VF$WXI}etp{7$)S5x@B(0> zOH2b(sDbLHfRE`+d5HRN#Qt|g_5i^1lRP^jaq8KT8_-Xv_Ro%Zs^e!zzZRA0!`DyAqeqdIpnd($70wsvyp zIth9JCVINq!2CaDR|0QS_5Ig7_p#r*&U^QI?-`!)nKDn$JQJBS6p98?87hU8QdC4j z5{1Z6Dzi#7Xi_RE6lDlS6AhFoCF1}6?sM+B=UlhnXaC>lv)5jGo$p#}5BKh~?|x^m za|gK4eIU9{vuMG-xkw>-4kBwai0*3dd<@0D^D!^P&c|>9(dT1ySG(t9Xm$;@E{-EF zI3&>&rSAC{T2*U6oUYyo4%N8x`7FtSnpGJ@SMRT91D|TzoKU`kG)CSMNfC`jBJ{d~ zv-*H|dJ-k%AYPO=82K=|b2_f%AM2-PPV1Ac}owfc^pA`&8>QK)S2l zGe9(Zf9DEv6_iF?V2V=r3=plVTo9+LtAj%|mOPs!_kp!$f;gXqU{}*-b*qgmLB1SG z5sgJ6^j-yLy$j;$NtBRn2dm0X zO~X>?CVJZc+2GZx}VxVIBM)(L$?p%xO)_9QM$hy&E zlZ@=GIK+efPiT~C*jS!IzRItyXiyC%83lrT3xx&=g6uXb=`X)`Y`Bx8|JKvG0wO$xooz*+M_JU#4xqS$y!Q(1+4Es`P#+XgtD zv;~}X3?$gqZV1^VT|ERRgbT&1(Wn`~dtV}bG)i~1do+q>cOAxi-jTm@fhkJeqfxY~ zt_5+rdN?>#awFQd)jQ{wtCq8lM#CodstV9Ol!LC?TgZ z@pepzOCu?QFl*xkR(%w$D?z-@(*E2yvPru7r0!~3S$?#(j6^=(udQfMwIJyXB6i2R zg=7Y}&^!=r`@HVzKN=L0jSyMig6OXHj(<_?JN{L&m#?e!@h{!g?(r|0ov^!#6uPyn zltT)dqSQV9MXPEHh||^k!J&H97G+6ofUJu_boB-YyP7uZjjSY>Bkzl(h{l?9VU`RA zXN>^y^zhCz6wCivL&hVYjHCz>`#GF4TLjM91QP6OH<)aat{z*DpK4lZ_ijP?3i&a= zwxU6`jAT6sLJn+8mowl(1#WZe^e`>h`*x&Db%?A>L3ICkN8Bj(9dUaOc<)oKkGScs zc8|ExZ0G4MrvIAUbKZvf=QQg%ge@K^$ZXm%ybEDKI8MN`} zLW$CBz5IqU1o>TlZAF8s9!YBuu_ynSE|b88W`OAGPNxOC{;_m<1tM!Nh#oZWSR}>1 zW05QTgQkx~>OpglMbd1E{9=+a3b%Tspeah-W0AC~-T`q2Z3j41Bn7E^B$ChnsefkYVKqCKU1y39YJq4%JBy! z1K!*RB}-*Ot)HXjd235G@A=|@H}^rwZd7|L!P^+Bd5`4>yt$9%m%|CtML<~_L39t6 z;w!Lu$}%ha`mgh(!Q!7ffX{VKT++d%Lp#V0Z)ZVFN28P6~`|B24|JJGx5{x zMgvarX?_PMFeOTRmFH$jLuj`5YbzR5DG;el5V3ciOvgLo@oBymL|6I~TCgAZ8#PD(ZRVae~Q z_IrZ2eN^)vmJE1vAC@c{fm)TjbR|7+g;9<_EE(|TJ}lXhYW*EGuWcQv=AA+acymu7 zL~%TJ7*N)95dFbe9XsThl|A?S_@~)9PD%%j!{uZV@)wX4b=93D4}jpTC9825EV|e` zP5LRyfeFJoYw4RXR#C0=-CE5%-&qjlz=YwP@AOR=UsJ7@qvm;QK{fBmYBW0T&7Cl! zQK(H#@HUES{s|-Q&7CkVrrLW6-d0jgA3fl);(#~zK*D((ocs(Z>vs^{#VzAqJnE-- z7dvN4>0r}R!GG+_N*0o$uKJSXdl0d2#KF?4xL{oh61@GnfvuBF7;A9GRHC%^!0?tb z82M|f8M%S3RYM?rL@H_(DTPr|YuB5Mzbo-n*qtrYuCweF6cYUPBX zPqpd^!#&kXvwuHVP~;vw(u5Q=MX7tLl~&c;AkKvG88}p<@#9%i5g@A(h^}5FIAPG{ zkpoqvBl2!YifH^Uos}h5gR=&KczSr%8j8y&wwKY!$08|$aHy0cz78xt&EBEX1t>fNqPVsH?&xML_W4 z3ml{E2hJJ?5}YvH80|?Wj0HGRD^c2;-Ug3@AYbO!Ry3%tCm9YRwn{}rcyOVOAiC1a zX~BLu4f~uBS!wsV-*WE^FU7tyydT8Q@N&Y?XL$95;hy28*)1hfr7p~LxWE*p?ipTM zRlk8a6GjTIpc*&M%#z+rF;q52~CJg5YEkBm-3S%DCO5Ly3ykolsP!3EO&aqvp zxhssVRJ+_!^Sm{untz26_vTI*=cqO%!Q1Ur^G_IYZ|;OqCmU+-CU{#xHSc&Z=9#!R z_c+E@91s2!DC;EP7;%_QXk1gJ$))Kz;)z5|iSKvtt8 zuAqy(lfrdT4on!%X)WyS@qIL#n-!;B67r{1Zmpn>%6r zO|@eQ-gZ;XKVihZxf4dU9H>1-jHEq?*yU=a%Xo01X&~C(MJ4%N{=R0q%%l9-p!^LSHGTsk z>mW#Qf^lC@kYs{c!{fzKTI!bHK~5u2pWxG0G^pMu*#Uy10p-F{42V!U5M8Mav|t~v zgy%dVvW9}_WrTOmnPT5L=iRY$&YWQMIcH}@;+%7)*}*4lnTdSa1*Rx<&pFeo>H*?R zFayD%8Z&2P$tz&3Eg-sj9fA`KZKl3dUG^bAgrtbZo+9*q1!w&Y;^|40kl&_Pl=KJj zNFb6T2oDGH6Ve`KeEPkk&9zYuOfb%2W8VZbooZtpHSZwxAgcLSB+k)h-vqOfYU>ib zEuflzf{A-`Czumd`z68K7gY04FmZ401XHmH)XGl`wyhA#fuAsOZ|+Z+6*x#;2Pms2 zi0;Ac_@}>g!E)XYa_3Mr9eh-(r3^$q5=l{4H6dvSB9T6<#&WReV(*mmOq2r?jC1bT zH^EG%+95~HJ5#-dYW@kvIpyq|U^Y;#*rcF0K{+tN#J#x_%u%X!bJRR)2sS9IhUG4Bq46NPQbTFMPOlY~We`~(gXjsy zdzA;pz7ydSV`o%3!RQm=dV+CJgwyQnk7vtChBkMets~8lw?R@wV<#PY-N9LXK|DQ)67uoMT)7GPa3n<#4)XF7 zvJ@;n{oaxD87K!P80T2FZ-N<5wXYpD@5uRPs`)1v=McDWf>}qkA`b_>ouit6f{A-` zCz!)j>*}a^-Ws4Bm|)`G+zF<5F{n*Q@OBH;{6Bf(-rS!&lX2wyF`%rqAi4(|#6SHz zv*MqA=lD4ttZQ&BpwO9&1@5PUhoWgj)`jlXUMxT_&}KEZbzf?E;% zydn6zVE}(_2!4oQnPJwz>IBy`1b=!9!2JxtqY0jNA(-1zzouqzZY(+SXi3@kTWV=N zJ*K7gnUf;}*YC=fSE$xWPpzTUKElg|@05LWQtN4C)|Ihn zZ(sJ01+&povClS6mxVxumVxNS(Bv{)CBFmtE0k{x%8%uTWe-Hw8Ia)8-F;VElBK)N z2OFcblzK%(ip;`$>X1BbMT6=P$uST-icmK!^?(Sq0MV5yt{;so-!LrKLu5?^(MxwP zybQ&@@G|v*_a0`^;brvF-3>28v+HY@kTuBPzrYlwZg?45RkwgROZRc$P`#tm@ca*0 z>jw~By}N@;ciODGJ4;R>KZ~S@#*-P)3qKCM91u@WqJ+qp&Qb~D>PU(pJj`-$rt}0C z8Uo@)u_<=zn{~7}gD+`w%>3mAwaHIu)|bw+czyjhU0N(NLJUNTpej=u-Xi9xdvIhibm(aixxNJ2C(fz@OG*Cw}rOeK1G}U0y zi;cjybj;FJrv)A5QsmblDJrW9B(*`re&+5>xfz@_9z+||;eNEiD^j#YiBj{qQjKI5 z@`Xr>8fp{>9yj_sM>c`8z5;QlqOgv^Lup2X!uT9(yDh6>#+!CeaV$40s1_;uf<$R< zR2d}&W@D~J3Rs}fv7l&7Z3WTVYBxAI-+U@ZS_5SDOH`%Q=e%6pdA3T4Ij6%BOHyP! z^0|o?D0H3EnCb+QzYp5#a&T~7`_CMC3#_#V#I1L_ZaJks=c#nw>Rz_o=bR27nBPv$ zBQG*1!2*S@a~e}0f;i4U0|)0%oy?J@(6V~PRnxR{N`20cRE6^qiO%Wpg&!}Gdy&sf zv_PTjoW@jJ5XbqY;NU#-M2@@)*4hE$wp?lFl=_^1Oy^4yozvlre38@0BXbihQ0O|R zF|`rIasDwlIG=btN9se%>K0cmsGU>lbH19+cO^R44*QjpI{}`SXn{i4IgP0nAdd46 z;NU#vw;Xv1tn~?q+wub1Ii)`5GwA&O8xqEx4(}deoq3fK+)CLg8`TOAD zeEhK-sRb>oV_Y>=JEzp={B}BDo9LVliyf^l!;w!)v_PTjoW@im5XX5NaB!Y>G)Go~ zwLSoGTOQKRDfKz;L+4!CX!Y1vjSjok?*BborJYmibKZo`xu6L+r^B74Ex8H#{fQPRbe+?fste*cZw3y|EBurr&w;hx0dZTN zqMcLfb6$$hxsC}qr^8l=tZ@YSzljzobe+?fS_9%Ze+?X*Z#odE8iXn{h71&}tXcWWn`IBkErer-v1;h%_zM%aVRz|3)fw=B-*f zW}m(-@8DWw{bOaM=}s%}t1g!FVp@un?Nboj^YI?`jeNrLrO-m)Rk7{2eczcU#{(tnuy}A3ab}rNkEef{n532e9 zlN|Tv{wH}50{i7c%W4XuhwqB`CgDS63v}GOgk1sGqsT)d>R@F!lr>h?XhibIPhV@>s)-NEsdK>ZJu&=9Wv&6X^ zIZMzpFhMjPj)h(nMXLmer-#F>D30peK&m6Ji=+q=dpjaFb_8eL0uuaGyDt|`($yd7 zuBMg3Lrcpf5%%z1ttklC}U@eL!^e zt`2rJZ9ac-iVQ|R3`r4<#|oi$4>)Tgh^L3cttcM&ps~zG{uGiT2rpJdz{nNgLhC>@ z+zA+2_x7;72a&ZkCTCzsi4u0|h63^x^6!uoK{z>jYgm2AhZ$ujikOx}j-zIW1@_YQc zisn>fl6D|CwzMftW`eVpgJ>6-*OqW9*brtBa}A%Q#i>;u*0rUr9MwuTA>Zy-P_&`u zgCH2hFvLFt7di~0n>v`vk(U`WuG({SgI8J8jUI6?&Sqg;a)M*(NN@RXSQUOln8186 zwcYTV_EXPgN@L)xE+E?W&%sE7w12E!2e}&gKqN)9*v+9g9Go=<#M6_ghD=V8sSrPg zqzJ-mED^eL5xCHD5Dj-iS9Zq3W3NJFy&ID=L?zDuqWI|-4P+#^hSt`X+Ho`5kjjyNN~{UEZLybhD0+!KMJSF0OZ5{ zx{BshO_C-cn58i24}r55f@l|0k`8)~t}ShyFIP@pL%zwcplCzQ06|1iG<65K(0&lz z)DwEp|KOmPUFr@x-RMF0B7PRcB_~X$H|SHnL3aXu+H;=Gl$yX1krFOh~^(pgk=>( z)+;eNgJoi42gL)o)R)c3w;?Hl5RT-ruG^fgw)B+KEI0pS*aMmmk?c#|$ zl6}+n>DtoP#37f;YUHo^6%=i#i6Dq=Is@-T1{eAaL^t&{Dn}YD#GvOacL$wr^q_mO zN6!H61Pt{Cy|OpxPRLPv=8{Y)4--}c5N*4BaL{T0*IyC;5qVc6MYP!Y(CY`zx&g$~ zlPDn%y;?)=Kz&R4R^wM<}bu+0U@$h#^em!DN(|{?O9RYME)L< zB1r6^Pln|)aMnH$ZH5s?B}z!ov!&%1I}aH@WAivJc}!tltb@k1v2NRaB)noxrf9uOqPMdaAt>Q*#1)c6mN5 z+aR*`#N-U%DN(|%&gdvVBL5Xh5rj>yr^9jz%PjasC~$lkYT;Vi1=`QUu|% z(wK2W$AdJ5(7GBV_*tQ|WSnfU4ZArR=i2Xi3D+|C`>p--N z5o3~lRxZ`GrLB*ORF&<>zw|38+ECAfh#f^!kAMsP4x&4{I+Y`@^0U%lwfkA28~s`F z;-QwpB`1if_gU%heO8<>sP^?wXG&Y(tUe&xcE8|fh4w!i*gys&ABLoe7CQvJd%#%} zK|DRI|4@A4k@hkh`BO-WAhCBM#OMlep>-e{?t~b64%`gfo)_*8oHKG_!#bW)3qzJ;G_e2C!dp<~Q2(6wV!9l09 zWP?r{SpR{R8(LLvMt+Z9SJ9kmOwtYnM?Em;Gr?KQL9~lS_a+|#HX+~c zS5UN}=7S*eE1G%$T<9=}Zt5#~(EsM3S9`%7bh^=l?nP!Tj7v^9Rd3MGAh4t}K{x?f z?Q0if{Rf=Y1w`9E6^uPe`&D!BKoIhQNQ!8&&n?3G51cgy#M8t255>t->c~{&k0B|7 z#QtS1)_>qa%Rw~U33fX66|Datvfho!8PpW(KNQQnlPf!s??qAsA+*;^SpR{uj)Q13 zjK_-gABuer6_At{0c9cxgMQPi81!`@6(O|Rg9Hbi&XNr}ZD9QeTBUz;8GwAaUsutb zs!7rWMC{`4WBmutS_qZRIuOoBRrjHq;Cd#Gpk}cYq7+2hmM! zL*+=5MHuw5FS&zGH+s;$7_kL$$qDT04f-5!(4Ek*c3%u$P2jB7AlmlK;Golf!GDTN zPvlo1DWWmWFU0x}oOK(Br-$_)ie>Ah$^FPDAt{2eUHJyqf8at3K{VV6yE^-Atp6ag zUWv&WUKQ&<6lZ>rFPo8XLsA4Gvcv0G|ADi90MTX`85Zk56k9z~Lrx(-i=+s`pie+l zS9=>s?pi1{0|^c~oh2J|+Q9k`_*)OPk>1E}^y?~`Q{_o&fnbC%==XxNW`SrIBkxN# z=)daP($=9<^<_2k*Zc~KHq=BAM8ic>KLi*03`93|s~+?*9Q2%*-9e`tJ?LJv+B1MV zA!fZnujdWA6QtJeg~2Ni6IKHdZM#-*&}sjb?ggbi@~%jVXvADvfb}0Z>jn@{59>b^ zPrcDX?m&JIk|GFU+YtcvVQ`^YAR6ujfPH%l)_)LLD`Rp7dd2z=#k&hMkvEaQholG+ z`_K2W{sU+21JPy}qZaEw6yN`&p8SIRcO*p+27M*s%i4c}WUhl!9gyIl(^;}Xrwy$C zK*yD=Bb}kw&#$X!P8B1m03!A+81xa~tjQqSMVAMX4f=XrTiWVBBSW4+zS^&#XhV$w zL0n!m^$l>L4?uKNXHhxw1_wQTy*uc1qX*rKn|mB^C)lkw=$|6Osx!Si;d1Tn7`$RI zVO0aswm%LAVWs_#p0cGG@-|3{XvDdB3hO^`R$mZL59>b^@5Ib>6Y}9miXcQ{-;VVk zxX@$}4R=D(?%9F&f#Ev7Qt;$%%50MUPAsVk|GHIB;FR555QTUf@m|0mMc+0 z9v)Cs4k15`qzJ;GzrPcMp80uLLa#un5=d~+=`7ix(}qNkf-cD`B(0#=-LI=?PT3^I zLGXGr4Ei8&)>shjVsp|#Z>?)fTk}6@Epw1R<5y6$p>6}gKjXKC6g(RpPb8L8r51gH9WW{s6i-r=!$|UOT_8qB)fUk;(wUzfm#hSAnxe#$av7C;O&< zqH9ZAKZRS$G~{#q3W_$=4IqeB`~}_%1TORfh;Hg~Do0u@#-N{sObRcL=!9NDvKo!t)mY0WX4p$eI$9GYqdp37b)^rOZeEG?F3+ z?~6Kwu(05)*Fm%yMi7=LA!90bkdKjnhNK9>pcg>&U%NZVF_f*yhQvXqvt)x#8xq|O zI^{@XsS3R&eqBX#>J*6uIlclI^d8`>!721Df1e$|aNr39|$At|E8KJ;X!)I!l}?AMbhA;lgoDxHvbM^Xe~%l@aZTmvpN z7(_Qg2MWff`mY$4As`Rpg7rK|Z~|t0;AF!>JNSVJTB=AV*@*lzzpkP=^$5w6AY$iy zkS0gKS?55si`Em8eHTkJY-5zRmRD~k#ot6ERU}VA(S|w>f=%@!VX2FYp=KbuO=tA5 zyu)F+PitXQoo@87coC2X;gS=!*c+Bs-mo}~I-baA57RPKi4Gs(K7o8iDmyo}T zq=-h8dFXur&iWKY*NJ+#|DpKlKV{?)^211qAh91uG|WH1h0cL!xDF~DPRmAAVfz}8 zl5fFS3y|QjFqUw#VWAy~rUJkC-$rr;^1*&xMRTemNgWV8T8Lp82hMr|M7!8ADcP`m z%D9|S+WNNWCGrCD4Soeh8|ont{7^g3way) z`$&o)Jf3?#EIYwj2SBtL#z;o|I}}S=4dfW|lSqmnOp!GZ-`KtrB>E1N@`UASQwAK8%I`Sm)XZ;F_Hq;0ZM5snn*MbYZ38I_YpURQ( z9CUfl9dx?UgYHF;J_5KC_SqZsY;Vw=z|r=dvoj?VCaiKG+ID7e&}n}|_dLl*-WW*{ zjpt6F*B+eJ1H{vlC?Rjn=pff2AB?0361#LR;$MRc-3OxKIw*5E?I}cMwhx0mg$vf( zAi-f_EaqgxLOT+j2zp<)`mz`KkA7W6bLu&gmq5gR6T@;IoK!QodWW%x*{v=9U z1Ba$c9e|tp6%=i#G>8z*x>zCU1TJ(rh;GvgDn~YPSmtOgOu%%bhsBFFJr0+g5YgVS zynsN#&IIfPnYL#?ktxpsXT9dx4#pCu{mm~{lr6|VMN&lLqXNCJz*$E?oH{{IqJ$jy zD_8zTehx_ygcqaWzG%OXCR6~?@OU8q-%|7R-6%ILe?_K5QlCMqIsv9{D~unFVHqWr zua7#hq+2%$%M~IipF2j1PWHK@9lRg`w8Uj4qzCi{_;nS{snR4>LGaPW=WaAOYdVN_vHgi;pSvMC z7B+1)*_;6k5(=s{~w<;XDhMdU+wgy=?p?z~vorvP^X zSbLwlF5c(P34v|D@_43{h6$?5roa; zDuv_|aMmsmZH5uBB}z!qE%oFu@}o$KAlw7CBYL%6yjCGe-wdVdAi+VWvt)x#8xlPM zI=5d#xdeK>{kn?gR2E4Y5d81y!!)@CoHYSNyJ)m9*`N>9wWY1jRqM+F`7t>IS|k2Bio<5tl~u^sA}NBzu3HCjvB6m%gJ?61#f{+R zD2{GXLG~m621yZwK|g{x+IHQBh2$K{R^FDxL8r51gH9U~Jwm0EU8O1XI{S4M%_#+u z$_5d;DF*#IaMoQhSiMEb2K|_>Ep45>OJpYU1%3rZ8|o$y9D1u=NLGRitq0Lf{X`G? zZyfZ~kmvAaUA6*#LCh^Hq}LJnPYvGhYe07(%fb~i-ay%k*OE)WgZA-BV6+YocxF19mW zp1=j`RgmC!k&(8O4GZl^bR6ge#Kzx&{2RZnqB*sYWCe)WH(*%)24_V+aa}x9KH0F0 zhCgh20eSb&HKi)R_5BKpHdG2kh*kbcx?Bt{)D=XxsW+7)b(UaQW@s(k_~}Luix*3L zG%h&-z`bE9>kW$&Lfqc^Sf(rl&Uz_kyF_qUXg}}qM)Dr=kB}76*lma2ZgAEi5M3wg zNtBT1PBfC=k^hCH2*R(Qz3Eb5E1D1i(Qqd`ciMq;DGiZTIVNWqZix~$^;`S{6nRS| zMGyk2>`9ld;H*9%+6*Izf|=W! zXJpDO;H>9kwqFT`$fo^^FKHsLB7X}>5iRy$=zRjt+6AKPL_OU9P(1li2RV%VD3T&b z>_-o!%Rk^ksoPz+6DYjg_vw-akyR=tXFzalZ=iVD7a}#0*F#bS;XSek)1?hKs|$$s z%2?(SC1iD>e7PF=KqN&F20asT$n7gZ?nBvn8YDRAbe3$;X#+1v0^Ql7sl0*w6Thya zIW?K&aS%K?i9!DsoOK#RySV70WP`p>*Os=vysw*Ne~SN~BY6snHqEI8=2 zzXivuS0P`Eq=?3|g3#Lt&iWWc*NJ+#|Dl+-qN40a{tc2M2w#Os>GB)6P{Ge!IOC$D zTF$zBJPKHsA6$mT^hvg=24MQO!ieb@7QMuGqNg9+nl4RoIn*;|D*kE)tjqNbnwz$P zriwlbQ=?<17*{=J%89YQ4B^iw($u3dQ%?Bvrbp9d0Yui)n4DqGB}!PgeI?{I zg2WziBwerQi4w9krI8##ehf(wgpXonL|3;*fP{BIsRl^!qey4TK8mz~ zw;zCh*B-I3q1VT+t7uMTlavJ!`&)bzhk>&m1ko zq3#61ZZ3BJ-0d}s*36Ai)F5WLn9yl*hT2Mxhj5Nv;8*1#_Y0o=q8 z+>_uzhT!ct0zAtQ+=bwchTxBG0Qi_8_!5Fk>@w?(_pS%HgCV#T!J`bpZw>_bX+v-m zg0~xjU%3w8Glt-L1XtN@)*CMl0JygyxF*5l4Z$n=1N?#^xH7>948hM{3vl`#vj&zU zxS=6<;WYqXZwM|)@C-xn+^Ye8%Me_e;NLC;yTNTZHF_a-^jsQgC(x+27bl(=&_%y; zg8^L%L*hh-^8&g!voHg?2+Z*2xf1{MNT;ry1MV^~p}`=dvVsj_ z@wCTIhu>+CI`*F`n9a9yl-xQjPxdV+SXhU!j68D{KyL%&-XCJK@qahxE;i>hG=j}J z1c?)VE=hBo0CaX+w4vyiXhU@n-3FE%t%NqXA?q5m4Ltx#-Uc0^E>f9o7!QzX8wRdI z8{ULMun9|886AH&NfVs7ypdH5v2zkY+z-7MlN_Mp{8G4)vfCcFE^=Oj@;DAMX(=8| zz#B$V_TjEw3q&`yRs0Pj%fE_;^S%CS@Xz_*HbQ5EZtWzk1oHVviXgGyK9MfV!C5bZXfu3a zi9`u`?Dc~3KJt%|6hZiWj>5}E?6*OVplqe@Ph8y5S+d0)ZQx(wpnvXDQUQAP{JM(f z)NzutMB4V|H0cb4bzKZLtWL7U-Rruxw6&;IW4Q@o$H9dbg6O8s zp>kvl7kA%6rdQ;2qZfDHOH{VtlJn*hZ*ljix43iOonm)-BvXz9XN3;9wx7W5>BuX!IOFeJlqf*B6xuz`2Jo1 ze{2Z;_i}*$G6atyxZ**x-uUY>fO{H(M-e>E5d3>jfL9uVZzXuIA^4Xb02er9*1#bI z=Np25=nimyL-0U?rx}6|b_4izL-5rEA2tMkc`3lz-)VM#hs1}66we0%m>J~ zFk%(|yE!>2(AW(Py)yEe^A;pd*pMX6aRP_f?a+pkz=hJjbKAg@qm|GGH`K@)wxK4_ z$=jeKi$o4~2B<$krfnG55p8%93c)7a4~Y|WK1matu=A0xxw!ibK->?#_jJ8Z#ihQ* zo%2>OF7C8c>^tznt*?Qy3VrY1?LUjZ4eIOf;)^@|@-269=e&H2&e~m4NJ>GoE|Q`h zb%f*u$vF^-{K5PDP+a!i=lO0d-si7`IHOx>tXxu@mtyfgugiF|0Q+usT#xOdp1#i$ zt%2@+{>x4PZ!!e;A^5l<`16hcm;S-4{{O5-4}v=zg12=5c(ft7Gr`Xqf;YDZc!wdl z9l_@e!S7xIaP`Axz0s23D-FSKv;%meA-FNYYYoBc+X8&h5S&l2eZ;JRFI)_86GLzf zf(IFbm$w0UmLd2ef;Sq1m$U}>m?1ci;1WNY^~Qo$0CzA1mmqkQA$U$pfS)!5=McQz z5d2sRfX^6$GYPKplUZ*}Z4Pj6LvR|w;|;-&Gy{0ug<$u_$f?o0&+EA~axZU;zd=La zRrHGqd7s}6Nx~+MGYk7SaaQA&n@(N3D7a!j;~rT9MBnFG@@?Y$2~=PcXDIKHJpuCW zk$i`tcb|7QakAK)2>|)-k*wl>H|Iq*=WS>NoAVrNqqknXn`FE>?k3J9Xv2QsLdQXL z8(8vf;>6qFZsJsA8;bn$Uu|$Uapp7yr~^QzZ5a4H+VB7rf=w6#iSy=@Bu#MMeG-YL z;XeNwfVdxeucmyOijB(PM#^r(CXVy6NZ#kQ6puArlPw|M2#^g})XNf9LWo`3NC2RQ345N(Dpk(4MQS5>Sp4Ku?s@)7&z-M z5M3wgNtBS=(yL3_uVp0@Nf9KGn!jRk*B)ix;*Kw-y7X00^HURd>z5V4Z&a60eFES_$q=wHU#gj4e(!v;NApRJYm)wJ8J>l z(-7R9;BkiFtu+B&X$bB_@Logkhcy5$aMG-SZ3)gd1aGVkaDPMaB?LcqA=q6FaccAy zcX}?36t|$U6&iXeq2CJ3#of!0By8e1v#@^?XARa_>D0B0f%^kYDE$w2VZ@Se6K5xi zflZu))8Mxj(B9&Xuk`g6cg`lx$JGED0FZBC#47%GbGouQ^Pv%J&Ui=?HgV$3aW`?= zqYbYC7up7*7k4cAHgV!@a5r)0u?;5xO5O%%6Q?`dQ1MUmHVnKUZMYT+!6sY+iSuGF zZ*doILiskhol4|lF7BQH5cfmx-O1yq*w(kWbKafI#hsRleK}s~y#y%hGZ1~(J`sOI z@2|hc7kB!F$nN6Kc_A{LJ$kghe1rTHlA;~8isV(2O&}5(#QS`$Q(hnBp?-WjGVk-H zAp6D;I;&RYY5(0 z8Q{oYUR(cXom&vx+z`CsB7lb)f*TP$+Yr3262O}b!F35fZU|me5#Z8)oApL@f;$_6 zmsJ3Gv>~_>!Ot3kpD7RU4nuHRg3lR(pDG7%_0wj(kxTHEhTz$G08caoXA`{E5ImzS zzy}S%83fz^m^E-p8GxG@f)&Ao48aeU26&bsIECPihTw6f06u02K2s9l5@*bM<1T_b z7=r&T0r1@yg54VprP+7`qlKj&(DV>VH3xhh5eg2FW{D&PF=e= zxC3B9zk}#|BulDXA@^Uo6`{>-+i7{{O{)c$mUFd zMzA?IL6WeE6K{^YiPHgXSPWcfJ&0}tOTJB4_xW*ufOqfnjsMKU4|qBo_gM|ug8XYFMLX(o zl0_sdK@j{UH7rNLg-(HJ2g@6yKKAP>EUD+wK;>^*trLD&+caZ|;%Ssfj!PN866$T@Y{m<^oyV-T z9O0_~9s`{9NX+(o!FPDm{^ZrwWFhirkrdH*M+)@Tg0tQP(RHF8UT%)!$!MBvL;g9E zB1j^OxVS4Md9g1q7k6hKa29v}0Fkj*FYYF^<;=jvoi5`o;JNt1G>+@>!%RKBxFcEv z-NjuB!8aO$#}x;7rXl!DF@WDO1m8vQQA6;bMFGxD@jB*zHgF`t?G3@d6#@8mL+~vG zFE#}KlmqZKL+~Ji|1kuAmkn@b%d9sB5ZucU{8bjf4;X^`5xm9_yr(e0`whXp2u=%` zHE>51;0A`^ZUhfB1b-3%_)$Y}M}prp1b>hT@K1)|iwQ28YStU?WB}aO5Zs*L;fCPX zZGaaTf*TV2u_5^7LID3|2(CkL#R6u%u|5oNp9{h6YKT*#x46@DX`}~M5~0VSp_dYz zMD@DnE=UqKahzG$zlrlA)>-M)wR6F}4kol6L>pkqw}~?c#lR-cCsX0~B&6QLDA&KZ zb2f3BvN@FsUeNacZq9>j&H!ixo6`Z3giV}ybKFgwj%dSJ;6gJ&bQ@UmZQ{h+;BMkf zVjErsD0v&4O`Lbq0s09b-vX80{GV+Y_!`9JkZ)AxCz{l}^1-URTahTtLuk1+(#918GKLvV!PFATxch5($BZq~qbf@>LqCk+Pp zDnoFH;K_#I`v(EM-VprnjR1dZ2p&UlBy83jf879Zb3^bbf`=M{f4?5!*@oa-3EpG~ z{$(J*#|^HUu9W0PwSh;HwGVVF>=RKfvbp>ZkzSS z=hp&!r6ITn!4nO^+pYn4ts%HG!3Pb&n|VUY&M<3WM}k{k2zGBOoEp9RymOrDC~u5+ zLqp#fIf?39+yF=tHgTL;*uRPM5^h)N)U`{1dm2pW6%cKJCEq5_eJBPtaoX@6xgS#R z9=T%P1)DhAt^z0{^MbbjcXRSmq0tcMI$>-zu z`HQ6-?(=l^*zy809r+R@MLVi5$qgjKK@juyIcz~kMATTcCNm1HUUJ#rCFJs6w~K{h1`8SU@J9BG4?%>liEkHf zFpduur=@j+R^)Q3b_O=~wovM}{14sQ;gVW6vZziG3Ek$?!prJ;g$_!DYiW?(9-y(n zSkpjsXWg^1w4}U)wn&u9Z?8iD0OZS%6p?BYh@1}%f+lfa3?(Wf9vfVw}K!8Bd#Vhh; zD)RY$T}5;1N|GBu#GZoPy|v)1Eg;&(W$lyg?k&@`rLB*K)R6Cx|LRvzw4vSr5&Mm@ zc>fK!P{Hij2Z+j%+xQjh44M9l(T(2StMQP#yH^*NGR_m~?e1Og?e1wgUgrexmB3jy z$8294+})%7i^nySdy$VvQbdd08hSIpS@S@2_ft=zgxEtX%1Y!fAt{0+QusG~c7Fov zTlN2YpR?-U3$c67Ze#n{s^2|l_jy0eq(yT)J{paj@5Zu-fysa<5!wtcm z30`0b-pr$O9~*+(5&V}S_}wc2u2{sZH(CaCcY`B(kU?%p+A+nfT(T>x2F0I}L!C1Gvj zw4lXyfbvSx|tV?cT zm(&9&*d;0cF4^CST@q%O+(>hH*O%bWy_5cDmkd0PE?ESHV2@0KB;gcbyhq$qfZ=p3 zLU*%P?d9t$+}|YcP|<%1kRJo*H;H=+FoK^i-veb8F6FK%8peN!Mp!h zo~rnpL_52xnv{oTBP2yT>KMu2B&m=g?k-m$w}JJoLe7nIRw36y?4ANF-Z8ccaZdqG z?To&ADz3+NQBSWzi1wcXOe1)UA$VdZfKM5MErKhQ_NxCst8rgPfO{B%&vpQKtReVr zf>#)VPqhblk0JPWf>X5QYhVk4YZ-z!@L=CnhTuj7Pc{U<)(qec7lPfJ3a3VI z72=!%tXKdVe?mjwRXB<2JLUmM680XQS=hh#_%iO`>D0ALfh%1ex7>UXZGa`;-s2gd z0(*}&dCR>5An%sD^~nqN9$TGi0OMA+D#NK0({+K z50YxR*lz*+1#s4pnC*W6DbVST4zgMB7p48i14_%^$j>1uq7lFqdUh4)6$R0CqMk$v z39Euq8RD8qiXe$h;Hv*du)bCQ`LWKbzZb;r@zB3I#a8|9@zA!MNoU6OILlE_ulkAh z9}nHw1mJfJ!7T_rY6#xY7~tHhUiJTHH5w7z-VnU55x}<_g6k5z*buy?A;8-V!PN== z#}K@%0l<~3ne|2`f_oW)pQ#V<1BT$T1g|j!KUELl{f6LNg43#-HE?!5zzqz+*#r+X z1kb1o@S}#{41(V@1W&00@K1(dMR3s?X1(!XZGhVvf>Q_{ZU`P%3*ZHY;4?J={@4(F z7r}oSg8!@maK)Nty)lyDo`&GxsslXE5ImgV7cT_6%VbWC-m2d@9=egspkqxA*uvP=US2C0sFI3Xr$zzsA4n zclI9lb8RycpxAyQ7C@}FyZ7j{U_Lk6o&pFLtK(t}>u$7Bni-y_LU7E&~vE>o*=vw)YGdFqWy#@Yzp4oEK0|Onf>k}U2JWcf%2$OHIdL-54}-)0DYr!2rv8G@S=yu}dwdKrLE z8G;)UT%mzkZ@gR@;2wtHIs}h31g|ay@Crk4HG=mTf|r&AIJKc!1FI0+@ItVApXJo( ztwNl`t5@Tu5*h&wy$azZs_&RrLXxoe=*+_Yy~p)Hr&HH11MW#Mp*0}d0875T$I&PT z_8xP2%l#Zu@0R=P+za*|SC;^&K%)!V{@=}+!REAvMzA@RAxYSKj5o*Kd%P5FxCXe; zND$oymVA4U@iw@7k6%oNvpE1IZ-cYd0^L%Osue9GM1u`VkG8gU2YNFe90*^WttAlfzXJ9X54>vie`;=Aw z%l>TFQ&#m9EtN=x=dmBtk!pN@_V*(3$1Ov#EzgD6SBq_Vl9$L-?%0i`Rxg>3Ju`0D zjX*Jjg$uVlU*yS=w6vYv^4vq!*MW^~8&Y@mf9PV%Q|m@fa?A5;pBB1>SKKX64YKdT z#^q69tTP~bj5@`)JgNNZk12R_AwIcCib(Y*2(~=uLX!hs4HrVKK(uzh_?D+ri_f^_ zc>^xT>h7xrwmcnBZLWbI4uTfCRmGMkS&3xbhhBKc2VyV0jB>Ib#FnSiz?(Z3mu&!r z_JHUrene|_xLLaVK>3-Nd@_b9w&hu&X*w`V&z6w%X24{FXx$ZhyYvX;`mv!qO8W>#CT%KIKuuqK)weB;+HJ6hUJ5uM?L0!C6y4wA(-RCL$ph zy;e}>BYzr65ri$zA`P+S*&pO>l&wP`-nOOIrn6*Qp0pv+rJ(Z~b(MdRhnxF!70s#5 zB%gzb{WZ2c%K>3Ej=?tcNw(!#P&baYj$Kh$E<=8eUqR7^ssWN5M9~t z^_FLEe#KsgOn=4bMsImuKi=K)T#QQ@*U*o*<@p4@vd)&LmLoWod;*7w{Yu3J&6)>Nuzuz0dXFZB1r7c<1(c+ilHk&Qv3I= zt7il-`I^y<%7KBB?=Q)fGM}fGxsKLy)21%PThNxUony5krjlW}f`QU6R3k5u4|#nR zt`~cKW&xQGCWG^hCX!#KOTp}v(nU&&CaBB0F`2RfAnP*_?fJch+T;)^Gqx&`5~Vdc z*h0QReiTVjL2aVc?tfRN`~%L4wsiF;U!YH&>kr@y07wlaMFkZCCy}0aXCf*%Sd7Xj zDbp>b)FS9(2OD-nS3!=}4U{sAOFpT^EAN4Lq$37=Q>ibtxDmw<`_P|j@o|b5d+03= z+U`5Jxa@$``Vl0x&B(#hpl~UIR>U(ggz@qaxvhR+xl*<6LZb?+9 z)K`HM)w~{nN~{8O`e*az@;LGpi54hyD?nrFDv*3SR5yY{1uC3F1+vmYpQC6U0}0fh zQeOr3$14DpSOw_xyLVHiNSi#Vjufy!p<4kOQwKou=};X8hYD2v7Zo_CQhEVs-3k&= zr_@)0*Wwj`N~{8O+ThtDG8_4-L<Z`!~cm<#ms{oyLd>IqY#rVe#QosU*ZUtyeeGQUNhw4XgsKC^JP=N(1)CWLoc%nL` zz6v}TuK-kH6`<2%ox?I0`IBy}BonC5RDitotlO{+33lzE)pfU9`h%+7i0S*;dbr2Q6>@W~O z>)u3lN_`dR5w8HJBew!{TJ){P@+|TVi54hyD?nrFW)NpOx&s_4u=ZdhAC9Fgpwa(7HcS zol;)~cE>Bg>By}Bold=`iY!C^cA^Ce-3ri{x(&pcj_wAB>8M99OLl<`73zpz{zJd( zg7>hMbS9l0U`h^qy1JCQr(l_{x>!Fc z!jNxYJ>_2Kzx(+RyK#eZyT$g^-MB$7=VIq!cwCSB2-MU2>O}kF2Gt>WfgyNxaezNI z1Xm;YFGKLsVgOg{ZgN zuK=A+UAru}-@$|mcEe(uQ6*XO{dfOU6a)X=f0TRD)q(c*r0>ICt``(i|93wvt=SL7 zB$5jf>INt_Iw>zaO7<;GX+ULcXx>CfM?yw85WXVJg-}T%eYqJp8?1Y)<@qEw-hD>E z>4($&hxkTX=p*mqVhrm(BS4Y+j6ju7f&AG)jxOmTWB)p*D8>2L`4vRk-L=jB5YyIO zskO-J)iG-SBx^sKwXaQFFVQ%sfz*`9jU3YJ!20&2IZXdOq%{;Yo`ObjNGCv&@JK>@ zNZm&gy20@4z=gJh=y7Dp_eer~9Nk9}I&&OP0+f6l^&<&sX&s($w6~R zM~@$WyP!@N9Y&o}K|m?fJj1Dy=@^dh(tCwO2?f5tlnzu-l$|Jd|ALF&OoCY z?-me1{^7RcHJ72)R;%KA#aQTY&$fCWO>fA)1CU-VHal^VxR9GmbKulH$k+S{u)lm<%1Ca68YyyiXiNmukVc= z^Yb9bQMR%!#~AhRJ6v8|<^CS0GxW$v$^W_Hm>kUy2j7S`B>E+2(OylY2K1WybrsF2 z(b1<_6I zM&(G;r5LhbAkzaxH+skXz#Z<6`A%HQ*i5LmW8NCus?LtNmgC35otbhPILq$k+HMAj zZ^xYWXSHo3rE$Fik|G*`T;ZfHidHkfot2!s+30wyZpK?G7I^WNQxjtHoh_}OTk$$foL;7>uw}uRE4O#hx{WX zMGyvk`!yK!;{Qk8cR)!|bZy_B24-e=hM6VLl5@^^keqW6P!S}o0Ys1_LGVRZKmiev zoD~rX5+z9xMNtt@V38!Ch+qPgSHXb5|2)-Qy*-Qme0vUub6(Er>grq1y>;u}TcL;E zs)yp_9G*SiPHCOa%Kpjelp*2IkRJD!mxAD{=;rlOoc9uwzlaP}g43G;;pwhnfo}ig z^lM66%6k39+A;$7v2F%0Wq8{ok-*aX;^YAxta6E<~JW_bbk zmvDIr!d@6jaWWq{&r&2x%sK@pq#tI5&A9Kt* zHZ^(k0SJkIz@^FBDb(cgXR=5kkj)xMRC9{tw>4Sh67oqU3Tf6ILfW~7^nV#bBED^~ zVdy%YI8=$jfBc<*w-ANuKmM4Pz@~n2GL}`nghUzoG15Z?mVs}B=F9w0qTREn8O65) zU|b1`NU!EVU2ubxvYg&94q#ThW>n;~*0_83?wHqZ7zs-$93${PA2zatdlQkUU}p`g z`lo=mn7u~+U4(~OLxobsqzb@w-3(sJ@CJ}b;0PMh8aeYmB&iLF_;M_T8HTbs&!9xb z@+T0h&A_~b=CFLFE~heTFYQI31ivPtQl(ElxeS%X0`T%1khzaM2Q?3iESW88P!Jq2 zlMeWqsNXf}k@^~y6lPumpFb4`b)k0%iPHPON~0{x!B-VDrSU|!1Db--c0etl`nf^3 zb3jE_DuMX@J{b!>&pgG9H3@6D)d8&u53`2DmvYNS+~0FEcqzj>8HogbMngVB&OC-B zwILB-0*%$1&AEXR)tq}FRy%-se6_;zGF@HGRSqZsO0X{xmIG844!E!cGT$Xn2hGDG z?Engb0|wFoXNVfEQKg$`R8p9E35<9-PR66yd>M(-drM<#lwk$YPq}(P5RA41 z$^v!C4Z58Jey4B=RD}aFc7p@*AW=PlHQefetkf~g8cKiILaG5=-_78q3~y#65}1RA zv_;N*07+^?BEA5P)rHL&jS}U6h|3JjOXwMvuhQjIMmgX$P=XbSupFSWaKLx3L*`oY zq-Y)%X$Mdc9MFmmNFnM^jmp|oqmshROJMWtIQ$VF^iDvc^nO)obd`M7K~oxkFwEtE zD*)II$iFfqPr5<3bHG;=j!zKafbrn-EKtno0j%Lx2V5gO%o?s1XegU;-{ofTQigX5 z5_|}Ph8#f7Jb@&&ArYUK#%cvcP3dko$8ec}dF=ef@^ZSI$|wgE0wowB!g7Gh!U5kb zhRns}>8N>Fq#Zy(5*R-|PI@6{K959cGzj5`;dQv=+2u)`fQPl7X>AiF&xf9b`$vcA zGYPwRzihm1y2|a96 z41C(<*=GT;JzE8+=5EmKJo`9>OJD*#+XH-_VTu`^WevA_c0A!>){v<}S(%Fad^dxa zGQ53|;Kc(CS%I9n0f~~N8WQoHdJSgi$>y9ui3-#gAyylUc^qfJa$iqZbCqZR21jrS z5te6F7M|TgnVZQ|PV=xxdzOOeCOY5*Q5`j^WHXIQ3NtT(AC|%YC^nx!qV(RRG`dc{ zS2W+EQ7#9>gVAT^4`3d% zKg;7Db~RTy;08E?TZyn7pt5klYRa5Po(h_WMcM%r1P5%P1KuU7vqsfzu2D&0<|UB( zoj7?I#b$pbO7DG2qch~2uK5Oyb~&IR7;Oi{tqI9KH|TZ_h-ak|@P6i#v*7dmrI_IW z)^Mu>iW44Y4gH!nmu!z<0|Z>E2av%_8Qu#>Bv18Fl$)5u%XPr{Z%)EmomKlkVqg24OxwxxdlmTLn6LC3t)zl zYtfw3C{aD&H;B~^U>>tS%L9+OnyVZjpacUnl;r@Gg#(sQ=4kR%(mX8E4xk`7pg$e3 zn5g?TYN~@u3Nw#i4EDvzqbN29AW?efBR#a6d^0uQ$8HA<2EcZ}C7?cYgKp=5ZWJzo z2jPHo;Pd>gnBf4{aH|7G5FTa?-h(wHNAChs5SQu!Wbjgk_Yx8bj6y@oqtL90B()(C z--q*IhTQAWobHMjJzzA%Y6mcnZ}3?@S(j6`a=;i+g0qOQ9H6prz#_^VNS?Pf4~w(| zCbOSbX{k|3Vdf?9)ZsY!4#nnWBuek)N~6u>%h|^ciH6Cp9&iHy+W}t# z)z%HVoddq7a0!fq10DsR=NZKe2e5`)9gu-KhFQbUqjJeC+!wnUyp-X65{U$6pdo9K zGq)m1ZAiqoa~{kPT#x2_jS|%Z;#_86UP2XEp80WCbCm-;pacsMVL3o$;eff6nM9r{ znukT&0TcuW6rlsg6V*kd&N`^1F!K^v@Kv1jMzJ{viPC!j>7nK1o2~gibUR=-0JZ}T z0d>F)x}5{oP`CuvzyaTZ&vQdD!vU<}RtM}QJj@!N8kbFS^~EpVxKs}ygO@VAzaYWp z8fZvG6q>b=q&6ht+d3C!xXI@9RJ`Z`2O(DX0OloB=xvlw(dCq_9Pk_{!5@gQ9H6pr zzzoXlM4pYBheg@}6a)v{90=5KqQ2Cq(XBKpDa<^6Wcn^nen7GLD-xymHKox!@uw&2gea9`qP z@KT0%AQJq@godm~&fJc~)sV|<$d90@*k?7$w5LG{j-zeJZkz+771;oua6g-e zMcT9!1k;|NX`2#NO`|Hd(Ws;_^VpI5YMeAevDpTR(r8s{&_j=tZ;0kQINoJi4;XFJ zP6O&KH|Tbzy+Gj-FhBRn2jKIZP|PqbYq-_4KI#}|4Qs|#kl%2>>1Oa!hIbzl3G78f zg8k8uoJd>^d5;Zg44N`6t5K${2}-a!#apITS(x^V*)ZC#--we%C^nZPQ5rQTJyer?pKHFu6I`ZU2!L(cUO;)DwACi)cBXxS z!m*t>Oq&mUo~nu&rX};Ord>jKm^GYOA<`E2E^Y=dWq6Atk-)cTNN?oKK}b>?67gM` z1v5-xb7rGN8F&T6>S2R<39V=ON4lKKCCAf}VzU$yrFVDIL#4^rPV?=Cd)&La8N8I? zEy?0QM>OPd!h(y(O9fPLzUmC5Ad}>Gx45$?+*9gio*aoe|ADKCzZTyR~z~PH=QUK*< zbq#8XS;7ISc`M>;LQn3YQr!Sk4f)hjDk;po1X|~hm%bGBlor*5^w0(J&C`6J&2V{g z9ssr{vu_DWiW_u0PmZH-Oiu9RkKpqdL#P=%$r^6;;mLS<@+tDXqIp=PJxM|6 zBn4^O*+gy8sJHLcsH8CS5_q(DynKXW^8ga1Q9aT_+sJoS^HpEyGVKfiY}0-VRJo^Y zzY%mh(~hQa348(5HU^)kt73*}$$YD6=Mf%e4S&5IkfFGbaWi-+!`lW4ek?{qrXXj| zL6X{#h_Cc?m?7aEG-oGDRD`_-v3itZUP4t^eo>cG8RdX8pakC`!g7Gh!U35m^D**d z9cs&Dk#+zD!2y3`E4$D{qH1bXosJrn6lPumqszxjQxu!+ktn_QDvef=ZskYa`d$b72<))F3O4WkDZli|2O z=VtIyhPNFOyfdRAQ;{?0A#pY23LCNqG!^@-Mwxa8D8a8N-ZHJq!n6fnfze8A1JBQz zheg`76a>?*r)gUfm3@RQ`PC{Kl@w+kUtt~bOJNk7WsxY2%90-HPriFK--@Cx(dv@Z}IW(}uKmXZ|QPrDhsl;K^8L;|bO zkPFC}SCOPPB;u<-6=wL5&B^n$ZHAX1R>wB;68etiEp$1RQ4XjBN^mI=mIG844#-29 ziQB<5K=ZIjJAi`VfXQ?~ZK7ssROzZ3h5sE!9$!SA_RA6!o2!v1y|*ZhdXVq1<||gr z<$xdO; zXf{BiB&mi(eATDG49nP@-ijCV#5Rc4eTjJq?Pd8)T~1|`D<^^yOd-N@rOLvUSt#=c zdA4gF7HL;f5Dua^g>R<+M%MMY^VRWcgA^W5=-MD9Q!QtM6qYnOp*Bbv!VOX?{|5$Y zBAvS1pD^dabLxS2*dV15zb;c7q*w>ug<4~$4&!maxI;%L!k3>yl5`qL-mjoo&z*^- z4AKnJ3`IlFuk;eyRPbH2>18mfaH)eQZ71e|3KG?Vs7)I6m^d=jYnM0rNrHO5ufR!j;rFM+^+{Srp8nTSOBbyhXdLp#Y=PxJMw;0n+S0N4Tg zJy65lpxXuL5ek>UK?LX&@OhRhW&|i}xHUlkOL&+yv_G0f-ot&bo54#N-q}bbPzDV- zj+}WGNoqqPz7J@u>^spM|FgCk{(x9*c;+QkgXPtAIh9cXS{jsKn1-?fRAmvM3n{ZN zdAeyH7U=+`Ac^k!AW*W513cRp72u5?vIG1& z!2kSWYVTo;@W3*3|WM!Yq-E8~h#dGSe^RUmOj(x0M=R><`pT2<6KAj2Q`aO~% zKOiahV*%Sfq#2|ciiS3|Wt;r(p-od+C+$-VPuf1r167KsNOrDT zY1D8Rig}>waR3-gl=+-pj*lGKJod|yw38LqQAabs;Wd10{nM%bUU{Iv$ik|8P6iXTY-%MULE(~rAtVb z{fs2XWgXz88KfDChH(3GB5X30HW|b^8Q|0Kr0vW+P@9NaMAU4J+U`Oz57bnmJ|XH| zjXLK-F^`>hXrCJ>ciHE!WMv<2)aSC#!upPVSYq4f9@?kscx9i22W|W0!*lA^Z?Mk` zj(x1H-$J!%pCI!0p6bho@0UvaOq+P_)W-5{VP2Lc*F^C>#y`e zBN*VV!NvfufqU?eKas5e3rT^$b%2v*kY*?v!tIY2V3UVvlfkT$0X`B>+Rn@al|t0> zM9tBtB`y^6K&>WfEm1o)YQGD`yoB1&J~vVBvJcP4#tJFTD*%tXwpU25SAbV(pK248 zeVj9|^W!<@6~H?4nggVL=5^ln5orXJ*#?R7YZ3iQ0B7muaS*btGbq}Pq53+Sy$zb}@41O|G9^e1EJp^nt+ zP0&;uzZFT{T++y`&hj=4b(nW=3{bx{N_C$~JgL|NcZoT;q+ z$2@+w!4b`i$h};1k0(8}f_$HAJ|5AGpCfuoCH(i}mjKvy_t)R{aNG!KuZ zwq&-bK|vDzYFncGiCmeNY=t%>R{rm=V}a)A98vI^jkJK=?DYXpIlpuuy{o31cn8C| zLvKgqangrqy26LXa?siUnrqPFnWDG8*6}>dTVLxAanSk+{P^9%ITZ6So?~9?tV1z5 zXj!jyfqT#={{+f3r`YYc)Va(-D+(G1t)jRGc|vB8M`H%nnV4$OBF!MpP&5oQ*=h3; zw0RFODL>A{llCL?K>acvsKrE$*QlK?6!SpsC2B8GZ)w!`E)?_l&kuCCttd7>M4}?f zvi}t{>=5ZEW9ic|2>eL;Z?W{pA;ne9i{>t}}C@Z;Oyhaa6o&*tMf=JnG$^o)MA4?WBFPDI`Z%G`@Y zsb5yTz?2;WKYFHGe!NIAJOM4pqtAls%roUj(hSlJMT6~^()Pd6_9eiq{Fv}jsvnsL zY79|%kvH3DRMdrH9;mKFH6dzDy!Jzi*Amr=%Z_rSFGu}t z+7crx&;az0NIwuuU$rSBr%C@lh7Qj>0naR^1u{>wEwHSX9wC#$%u8T8&gja6VzU?$ z)l%z>uHEEorTKV9moxcH1fxCqTmovW8+5zLX9z3B-@;+?nFl`4dc};%$2n{)X7YKN z@GxtrUM92b!~Lk6!Alw5B}lMUEE;kSIrC>EsSUx8)eO+wAEPVQyDe+R0qX6iOiaOti#9fV>M+CBu^jB!y-NTP>@9HY)+J+$SudE6l$5i$)_GV z$7b-F?`r{h*y{t`D*NSg(vNHU=XppUgU<0C>6bNK;X}W0(5g5iX3*lHG8nYd{G`AG zQS3p>^^-z14qCmzk3q{h6{S6%V}4SwPDM%mNg+@feR3#J=Cept`z>|)aM0Qe8YiE{ z6vHD@f;<}~s18L@gBEE9X@;UW9dC1XEf;( zV(BX}$ju`CwOIN-bm%ptzY{|b75fT){1Y_SejE)N{rEV=a0~n(_r4FRP4AT- zNi#?@6b-ijfVL08!sZ)bQhqGsvOV)aEhMT2QF}G2lMBT>P$P)yNz`SH`i~36JpR04 zaYT&pY}D0RGB?U4Pi|=%lFG)$ynK*eXV8x3C{?=^ao7BNvCV!35Q0Y%(XFg~h04WO zw6@JJ7hlIvbbPS*687GCE-mNdX*fU5l;7g~I8zzT`2#t3Xio0%&pMZ+;oOobFSq%x zbYe@UQZ{F=(?he#)wGPK$oN*Ovh6d~wiw%IYGEik=JATyJU=i8KdJPR6BWJXd1?T* zlPEh9TWFCSxDQ%XLB?#3L~Xsqr{E{h!18()JX%|5DVn618L{`$yGepqe-t_UFR``G zK3eL80iUlEH!`Zsjf`djib;@kdvhf+H zo54uOruwQyp~Ku&N#-hNe`yzD&nss2o>y z&}4PI%9vonW_xOn}rQBp(uUnFV1`*$^^e zy9`9d;j7=4rNi|3MbaXaKb?js!E@46#dkNerg9Onxr6>x5TNA z#md=Y%RyK&E1_!)%UP7rSw2qnbyi8>vA+|=vk>wENR;l@ z!56QsNt8s=i)i}csd=GqJIr~NNWVwZRXajExSk&Xn!TPk5fIekpMOkJ6KG)DyAk;h zWxWt1D^LvdnWVoOOP{&}>wnU>#Ly$YA0NkLw}TcqjS^*n`|$m)UH~zV@1Jl0#RYO- z)!fzr6kn4s&mxcK8bho$ z1M~RckmXrlbv0M5=e?j<$81>ZdFx~ipRWvMzDS;$nukStJx@UrU9dh;njkjKqfWrSUS-&m;MVEsZx$%_n(!WXu)cN^fc0BV*;%Esc9- z%*BVwT@JUFJV_a|`5_K}nq9*$zftH7UD+e(7Wjq|JudJX=sA$6RwyMxUAegI44P|k z$!#7nUZweO2kr}KFD_mG?XaDr{$%juCBfMoa44Q*{@cOY9FU7kdvm~?=;-r+GT%Vr zQs*dU3F9PaoE<*KJ;H!Ni4?ndP~=A^uAp4Jr#+W_2X422H^b-k9{ZvGlbV zuDX%_SPVVl+tml=dz`9GMv3aD@7K|NDk;o7{$2pP|IH)!8=BkN{cke)_G&)v{^tx= zuY=JZt~L?1#1>A_?S`w#tQ2DlhO0#Ic`7Pq3|G!JhB3p{a>B!`;mhllq$Td1-3(sJ z@D@gbzZXD59!AdWk3^YGH6-HubSTVlmWobCi88|)h}BWVyo56CMfomWPG!_^wFwk! zk3(y?vbHt!`QD(+n&kOj^RUR0*`fvoA$}Lo0bPm8{Dxg~D{Q!?Yi1sQRe+uN@}Sr( zhD7OoTxm3#e62Jecj9v#P!5c?1C{|b)(yIy19Gua3H;9Af57KiubANgXG_5t2UHt2r3~*9B>1ZWG~^s|=Fdn{8xryD#(sdIFW8*yOKmemAy%7#c?tc& z@`k#c%2>bufMV@8XgR>zV$kP%l`>2315Y2#!y@ef3W5WM(*bRWnxauzo!z^zD-rkK z3cS<{`(L8id=rV%+u9XxAo=!dKJJR=IN%DZvK=rNsElvg%_8V_4)}`Al|aEmKFI?< zPi4gn2RPgN#W>&^;bGRWxNSyhjr;v>1}|lJiz1Q0YBc0g?fZ;L&^VrCirSF*l}eJPZj9st1(Mog|3`q$3P_axN!)2KWIhTh4{0gfs?U*9 z+$}}z*B8jpDYHC5DMPiCg4}8^)aEfrS)!$IuRcdgWw#WyVPD{lE}3OLrEJ$yzEV== z^oEolwUi?HT~g}0rKlbI0!?~kmOm)Pv)pdc0`AxsJ@!v9URB_=Js{ZW~hsFh0)BD z&pIMmt^mp3+`_J&9Mnl?5b>%S*Hb_VtgG ze0cO!M55w;cq|{W>n>H02DmrF@EWE5Qw*->;5RP$>MtX~X3vLT8_o<|gY0PPq>e-P$<%cY`m zN<~p!0J4FedzD@GGu5}^{{Iw3+y~H!BDN*jcmwk8P!#<_QFl9v){Ih7R0byT)YW!r zfOfi~i1xa_T5D;CduLo;q9t$+q53d#o?%E-IGC|~#5R34qr8CoOSrs5OZ3XQM45-& z7q}GLqI5>l*E8Zox-1%&h)ua4OZX0A%&S_~b?hN$44}JxzVXK+l487hG6Gy;Lgv-j z1Pqs|G8qWtc?)JjpDryCK@6U%K$wj*jG7rs2#bCURD00xUt^2d1af+UhkF=0!NVpb z8{a_Q9SWYZQ0uM-&jYhm@bp35o>AH^KD5&nJhWHs4_nG)+^6C45-ou#Q{&_{cl&%*k1%*TBZL?{EdlAGVby^!?y+E#&LVhHpcI2=5T%XLFluHzKv=Xb z055?)W1X#aUy#!iJlucL2_7~f*~kxBcPMxsr>MI<^2}PKf@dk}_PnF*@=I*+&|Xjb z8px-(AHd}$8ZUEm;^Z`Pp36v7@T`mFBew3V&GE8PKzz8oL`yUyoRuAqZE-0!Ki!ay z;Hkcp!839cgJ(}dPlz!GXjuc`BIC1}DtNv<#Nf$@5MuDW0LUy2I}e2MlLgEA4T9%E zD8=AeLuv177&S98tKc~b)LzgJy=`k> zfKaMsTuRsv;zp0#!wHLi2UJ_oJ8!W?OawVS;loWfo$z5Jl8vt+?+%4eX{dGA!{?jLDtsP8-JYkl zU8cr{5AD_N8$%}HJ{6ajXdL0O9^d~W=UIiI!+lII9{S2WYmV2!`~;*NQ!A zj^mD~T}B~kXPi%X0m95#TH-Or808~2DxU;wBTb%$c(O^UkPu~B%hwf;ffo~zm}XJcc7_WEVKA>JKW|KReT z1wjH2?8f>BkDhW!)Y!8mmXFxVor+3b+#BQa5-rhHUt;}(+>5vr+s1Up#s>#gY;b>6 z{4}V7BM$2yh%+~8nMvH8lr{>!_zTKm!& z9o&1>i4N+RY&;EFcPKh`Q`Fs#j(bv6bQn9~DilV$UE-%p% z7_>i5Y9i-pg+%d{kL4parE($ZhIdE6fN@-qk}adIvO5FbWB36jEohZg|vdj%x(`?gvu(it7x z7}kjn>X>ZIhP*oz9cL-(ZbwI_FI9BphexOSA+geSu%6 zk@Gx`M7eouEFZC1o{Pvx+{fVZ5-rgk$ME|Ha-ZN*Yy;C79U0E4==e}Y#}ySFMa_h7 zAkMt3Wp0G44Ay|?*n1Swkrg4r=WF zW3sUc^6pS{7*Ol3N5?bgRdm!w-JW*ZF7Kh8uCasm+VNf{>52PexV%K;aE7nr9Jfr86piS=*#+Iz5#IJ6OU~+-o<*@H~$dg^@pD8KiC2 z690=GEe*;CY|2f(gUmM)rJj+H<#P~EYgoU+fRPFQ{PHmTSr|^GKd*vjX8g$Z8TB(N zDSs}4?pZ*Od~BCBN~b@$*{|bIHYC}29rEswKQB?#U7iY}onX0QutyJVllXM{vvh`d zI*MEOcF8!A?tCC&IE0zcYl$zxY6j)QQAsI)twf!4)FQ%GYZxmw!pc#_0oVrmu1{=7 z6-uY0xGAvXC~BB&w1TWVh_%1cA0~Ax<+N%E7UZT{Equ| zTwbE_x!TP*iQfahd`Q&r+9j5c*n}?&OL>%6#pNYhqNfgF{6lUOF2(l4^z^+8cFU$> zgS!f+7}r&BEKZmXX|HREOR#ycLHUS{pAI56DkDG`8{0rLKh-c+Y+OPj(KS%&DCnm@ zwUv@|#s>E*c4C7XCL6sW>kh@n+Z1)TW8+A66&pXJZjWcLZI?A@rz-Fd zaCwQwwqW6SDUL@^RPzmp3z?-Q?(K1TiI(UiaMnY}&Bmo-T27eoo-N%uWcgA<4!{$h$)!bd{p+atK8)ex4}h zV2|F~CU2zEpBsuOe{yr=6yqk+jz+(PXCZBhmN*GkGbkU9>VfGZqS7wYknLbJHCCa- z+^)f&)Xpcy3=7;EtAvP;X_u@3H!%{uCkXyQg9~tm*=2zjbs?B(mqSQe8_f*wX8`9w zq5^eq9Z3G!0>5}a!A!e!0GN`i5Zq9MQ}98!kq1xExdWCg5tXhn#v|X(7-$+vgb&j2 z&+0;Ua|>Q~Ea6PMWCWVBUnY2w2KVQJbU4A-D+Qt5B_6%tUZX;m&4hiVVUro(WP%FU zJ_D>Ff}HVvny`x+Mt>PARDAD*FaHAFIAAw;1|U(fa56T8h#eMgAY*WU9+#JBiH_TwDASSq43}c7 zo6Zp0cfils;4a}QMhemvMwNu~kak5&oLd>OLHUS{X`doCzC@jjjo=rEjXX%4ij77p zHtqwUB_A$Bw@vQrD=SWfO?IH7t=obC3!C%)JNx zOoD3k=PA(43mQiKj6%wvUjg+e8IIZhtddTDayxU!pKM67G4bdf^XC^7b+`L^zf*p` zJtx2V&!C;IzD|2B$yZFO;$91vmuUQ%<59mfN6ym~iRwFbV)=;eQbnX6?t^i8 ziI(V$Pw-nLa$9jJwrkiEFg^d18vIs;5cfk*G44a!!q|{-2-41IiPsrIl#keW{bR(& zPpFfz@h51e@3^aCLsV=$20(Vu^PI5DvZOOMxYN258`Ln_SPFS}sIOn9sJk5-S??9&`wuu&|ZT*&80o=_u=vqErIP{`Q;JhJj0PFC+~@k4Pu`d6_E+JPr>CS zTB3R3tog_t!KK)ar!zK+T~M*X4c1ePc1Twmr4lYd+I1~)5gK4nK4N1*B40(Ab~%8g zv2nx5{Usu#6cSba#Flt(G4TZO#9Rb3?a~flzPB_Z_&yEJSss!rSm25n*&;rsT^?40 zTqolxg2!s`w5uOfb_f-xjeXFD7`M27ki2Ya+pz;#AmwRWeap8sRK^Ks=^l zwb?W>LEF+FLE2!LlVSf0Xy#1~qfL#|NF@3*P=S*LB3HMkuJlgW)A-58 z0?4~VVgEfv-R-bnbWw$U5tzhNUE5^^+UW{=+RJEDT$IZEF(xA4g8rLRRu*Oc<#?3!2Dv{;@k=I%3gA-Q z(kUCr2Xoetydag(WY12)zYZ&uE<@v~iySpdSHphP;mpz)FjFA4Qi?fhpLNr+r$^8X&kiv73*@$=!Tx&U?u@Nptcz3f4I&4JUT< z;h7ZHZax-guk6FJ$%knfSCa9GRAsT7kHyH{d=woYW;~xw&H!g#MWQzA;ZyJuDzG|X zg=4I0S*EInqG7`taG1Hz;JeRf*4L5~^D|{-JDh!b)V`-apZ|MoF!CKY7^wp&Hn(7F z>|JTdj5BcA*zU-x?UHC_aq#b$^16(xn1~oypYSEbm_KS+C$T}NF@Zxpw(-FrlVbGykht7l&o}t54P2^9 zp37j&wP0@p(B&GG;>1xB2(yNUQ8QyZVbNg#GzPuZSzE+sAg5>I;As)g#K9&c8%H4T z4ow^dq1IiWI4=C5CXQ~X+cQAh?zM7d;H zEFZB0&z6;UaNmi`OEkWEg-`b(ck#T-lj=`Y`3IkR(kSicITdOQtMs(Ubujoq@T}(; zw#6C4wt4DG1oV-dNdw|H0e{=UIU4H~PQ=6>>i;V*SnP)bH`*67rsUS~-G)|LatXNwoiY}p*9Orwa zCj^2 zbg1YO7>rr!7vRi{h&44G@+tULZ9N?ZztPj7qDkm|aG0flcBVtN!eVyW1Pww zb1G{MD5gWU#-7T^jMFM#hMfM$s_BphcBVrL#2hq}GKNqFI|5`7WXm8U-fzyf$|2G1 z?u%!6^1 zb`^SOlWD*&WF-a2Q_Ru)_@C8e4d^?avNF`1t+q7}kMqkxJRa2uOY@dh4f!6VUo|;K z^Oyzk#r8!d3}r0MBgpwHQHxs^!~?ffaV?0?t<5ei*alpl>o`QgFdem+kEoZYdoEQKs zP7VMzXCHFT)0{k=iFGEY;lu#IBbuxMz~byR5(B{Dw2a%x=u1@=1AxWI0YK3uFa-Vb zGH_<53-$oOr(o^6dH}fcgB}1BO+vfCVU`Bk835Q0X8=%JG5dVG*?C`Q=WPQhIxkye zcV05%tTi_FNo3UkKm$7i00m+I_>D3SQN}Bl41#PKWRz&~u0&ac+;LpWvHjTf)3F|Y z@unZ^;Vm_|&P_@931ZB@w5+{YbsBZK9P#l8y6_cZb%)6`|H$ zUk_(A)c3OeP`Bqfxr)xb-do5d6P^RKO3zwH@3Dmmgmp72}Y(b)2GAovk*y7Rp zvIqCiae0ZB=+l@tP9k^xvdfeIupZ{IO={%b{6RJe;vp9<&tVQ>Cy-cW1t;Z`lAzag z%DzUK{}OAm);R}nwvv`8ZLbm55H`JD73l?1KTVEV=ah+Oc-iaWxL_ff2$(hUPD9Rr zo?1BTVTEnfDceEK*=k$!+2t!s#$V7; zaVc&~^X#vOBo9c%H91D}iiKU88`COCBfu=pTO;THfLb`~VPf6u;Wq*KA6XCE(_B6O zacymH`MgpU5A|?)USd0^JJ#+svZ=HOy{A()m)2INf?07gb4g7ZjM5PrVOe|p-&JHH zNHa7!#@hAWaq`6C8nPZR%i7zJ^S_Ep^jufZn(Me8wgM*ExCqQ`*28aMI;{3zd$i{H zXq=kzf)Fg0I>TT(v<^t)bg1Z9Fmjw9ke2hlG@O_YEly5{Hs^2TT%kF6mK*DwnuZh8 zArEx3rbCOf*J@0MXVWr%O2)r&WPv@=U^=uIIUOoG&ceYgCBMRS*^#LAFrR{ja3Y#D zv>f_XPlt*op%1`eMgelBL$<@24%I1fK40MHP zUBcyg1XUS*S)VE^xZ04Lpoajq$_Ap$&(qrMvEld|d8Hsqi)(}xtr=Taml_~7)MPta z|KX?Mlpiul55TNw?Sq_u7`6DNEIMMYjPIi5pb2K^M(40J`kafP8h7(fuZP;*MKv_zuR4(Pr+v!=&rx@n(q3FhBMK@VO9m&>H2Jk z)AiLR#y;N|cH8dkw)X>yuFuxk-ImPQ4xf!3fvoEKG_ce4DG*(MEoHn!8S^a}1lcmk zh>a<*KwXa9SzOAoh1vDfu@Zj68;6zfuDV>|{+@6JV$2MG*s``_w_W23&uj7dc5UZM zxcIx)N;nT7r8SJ_^%PWiLQJltQyvJmuQLJhk4^XAFtbN^yELRgn53U8hKZ?sx3M3 zPzaaj>RkBqZzNV(fd|4;8T5uunTJCd4?whsup9Wit`kZh)Cg+`oBvw|c><(inrshY z`j=rUWv_(0*TJV4fLSB&0_6NYRFb+9R#?|cn1?PZ&BuP8Qw{=ul9i-T^W%=@r`D8} zi=bb3%DzIGpRKkv@7=UEcBC&L*>NduOY`S`Dk;T4s;J2^npZ6B(tKKl;?f2%OY{4X z^Pi&@&PteA_eyvjApavPVSAdZ=Rd2hU2{@Y>f)gVF3(!FbCYB3x^a2rLD2g;W!q_O zbv&lE5}x%%6L|)u&uN5Z?flh4G99EvnjB;8`tdGnzmhw@YzNG;_HN|-yHJUq>*`r^ z9aq9uz$6>L19O{|aF(&Tq#k;T8m)OyD5s|4AOyFk&M=q`t&>7I9V$A`JK#9~bXv}y zX*e+*TAZ8?ZO*t?z_~$l@=#OOIWG+-rbC`=YE6e0XRmC-v&qG@jEBkiN~*G$4lPDb zhl(zNVVI@jZoqW;k*Jk0pMsAx($nFrjQUg)MU&8OaF~q%a;8JJ!4D3vY6o|q?kMqN-7C)0{T) zCh+T7$;0G%+|hjd+t~jG^c1IT05xZ;ZOzYbDK6*m_>)FhnlHJMUH%5ii&~V>7|ml= z!V5BGmm+{!nwLe+Ka^VBvJ#H{NTTsaA`b!Fhc(nELsLiXhp)Gkr$L|Sl(nYX>J(UO zc<8vUfy_tgVvVrWu9rVv-UjJiO^#7J<|Bz-D@48o%u@SX^g+XwC>VUp)or250Z-460i$Og!EoZkhoap-&C;Pt5IfI;QH75`3Wu3Fq zaH8+?)LyIa+aF0ZMc@A+E#m<)#(pGWF|zO5A4v>Hm(1jWU2-E)eVFu? zlK2Q5W<7wMzRz|zec%2_qBi?&ZX{+Bpy>N-jooj_j6=-X*s;hu26p;B1)}diK^ZM5 z<4sEjyn;&6356Dan<7uqMb_=!&lDb@h{#kciomr~T-{(A4hVrr@V4eIt>(RMk4cDk0@v{!*$ zwImbHbPeG05-owL1>z(TkDjtfRFut% zBX?jsprastqscK#s50?X%3f+0ZBtKf0%nboKB(m{MJ=2kofX!#)aH@5O1o^M%gH^! zH^t?-G6y`@kXU7ApB!>O=)IjXgPOC|w&t(g+d!ViqEp0z#UPFs$>NIo;ZyQe5Ob zaxN|BzBHVeZY@quw>D>iY2eK7x9j9l)2uTS2<|#D-SWI?Yr3^R!gw0f?fq#P>yk0{ zBMgg?)2;OpMrbw<9qSKZf_1{+(%gA4!i>P&I0;1aRU~Q>=9BBwjHe=c61G3h_!u1K zZh)Lgn5}arVf)jJd7MB;a02}vP)x$4+Y>06F~@#N$TC9W;}nDy*bp+|r|e@m^r!*~ zT}QxC?F8#YLDhm#p#uPQpoGpc#V*8&g3Rip+x}!tM_JQTR!s_Loy$f}D6@10&SiU> zCYlNYO@tE#RRum@iJ!op35mJEVplp?>_mh8K0>Mza@2y*=yqGk1ju+o9wy{33qr5k z5HjKTJ_bc@oCfU#qO&Zai6LVnWlW=tI)K5d3<_HY83Fm0kX?lIv>*&<8$u?DKDaqi z`Xjd`XipBkI7k2U@Ao{GT6KJ1{#Nojm{#NREN91G<8=H;pqT6g{d1>mBRl?BJleyo zzo98-@c5lZSRH@;D4f^;l8IWB(3nMZ%v;O8zcNb!z^sm65;^}iYVogcEj&wGXWTQljlW8^Z$17$q3LVIAyO;bGF*n{O7ifWF8)0(+ErRQR^zodXRQ#a*XCN zZ!MR8sw*b|vot@8oPRd8xaF-S_DhybO^Zl2cql(EPZAmWIBLH(wW?GAy`EF{B-K{O zhg&1<@~n-e14{4L2utlNj|Qa=NJBI^M(vp2?~6?f$jgAKRmU7;{7<0{J<$Hcmn77*N=7)32kj-oc_N^(A;IeE}K z>ui;V6Qd|kd$&eW`%9Kp7)2MRWgJJw*e_WuMvkKPmn=`C({2OKJb*-vmwa-4$#Sir z9xv@LSv~=Wc?BS6ykt9^@zVa1CF=!r+so{>xk9REvo&_NC9{N1v#~W%=or`;9Vif8 zzX4?wKw@^aWY7z?3^GbI=Y~Y-gWObH%CS5i{oh|&ly1=*8xv(A7>;SCejG2-@zQdw zjCyG))sC+igN%d=5M*A{(w?q`fjTclq6N>s&6k$l&{Dm$WDf&a1c}nMBrrxf3pjQY zUs`^^NXD0zdjM&vVLZj#XlKC!P`yh>VENLLL}`OGjEyh`5f(iF)M(JhC)hQO1vx!0 zEj(M?d1+y5lZ`NB-JzG32Px`qzqAY~r(RlSp>EF#ZI^Lqr|YGK_DWfrSGM8)J}xiO z5?E0pPCiG@^9>T^xhAoE#O`~ys{D%kpSZk4OEdy+X2OZ$#k1RN^yELhwD8n%HL^c% zR*~v>Xn@NTMISKgu_LImNwb>Ey`U#KWld1#=dtDX$bJrgd;1g~M{0yMM8+MgDldXG zOOs<3qxe#WA!Gfy(f;RZ$p*lzk$nep{#Mk&d1+Bt*GmhJN>|#oXjDlq1OGQG$w{8P zj^@7&C@sOP1tl*otE@0JXRB?^vzwKqJW8u-ggu*P${jE>walD2+04)ON;ZRjEOi>)~@?tJ?Re9KaR`u+ibM+FCOp9Aj;KDU)jL66@Q@Nx&>?pF_@{ z36X*rGQwg+HpHrU?Dnr@qdkmgI7Zm6l!2~M~6_b%pi%p-Ht za*j#EiRsqj=&^!U}8%nQ-hNUs}!~>w0OS+V)F} zYC)(uUs}95)k{lhyAUre%<7kx&sozd)>MkPj-qhZTL(Fz%sm_N(((gM)C2_krA1ZX z^W7l(NwW8`*p&_zJJDb-dl8U$B<712ghsd9LMA}E5)vh3jRm3eYzUbo)Q3TF09hvx zon;A43>lLsqYq{LYROVcL8s&Q8r(vLfIikKJBl)YO+4De?D7`)DHMOp)0+^-wY~=h~ zsl_cXEwLXRwaHgR_5ggCHH;&}BuDL;-SW#h(62aUQ>iwaY^yzBNK^cPiH?j*aa(FH zKNOODAeGeQ7`0-KQRPe6sXpN$iK}hq_ z(ItqnMp66Ip_dr;U#I2#I1ML8QHztKsLlC4ITLf+b&5TT8X&mq#3Z}=(*Ee^Q*f9o0CL7lw!;}O z?T?OrV7L94-F82q7^2u3yW5f(U-YrDmymS~?2HZ+h_0XO5@cMZjD$Q&2EAa*AfrTw zu1S=9C~JaCIhJR@|NBde(kZRqO`}vA7(@1y?g3PsA zTFHAbP_Kg?Xu(&j`O;Ekt^LySDS^i|a0f8PE(>@TI;9xR@Q%Zmmdk($z7AAP;!)+s zw-&4~Bz1WkSiZF60>Uh=VQhqviHn42c>t<_UJGYzD-pTaz;x`Bzys8smln1*+1L$v zcj%=h47Kk1OUvYH>ZPR>>h^TkcBz7Px?WmnujjM3kpZ|5#pNYh0v}X}ld;HoW+74b zyb}A;LTuzuzbwOjH7+mF5*?0tX&Z7U3%We{4=*h|1zpWaAAB8_3t;#im!~WGfYFN` zL6x=annQd5=E7x_^+uVW$Ee#wN=V)YuP|{fquv-YfR1A zYFqPF&sCLg@%W=gSeidxGf}RAlnJ#cp)s0QEbP*J!GbJO7%)roGRXN`QVZv$g;@7X z%XmQkM_yW-kD9h1j+C{VhAK)nJmkma$-#E!ajZR~G5!n|^mWCcvJu@9KrytFjJ zbep}9JsZru*P3qs249+wnnt50Yr3^RO^VFGJkmBTXEhMqoS1GcPENNrXDf0}(4597 z>!oEt8cs~Nkv4j|wLfZVg6VcwTE^vMjQyy|V&rsdebf}nPzfvYV*uJOEywt$3xQ`) z^%W4!xWe`%%qQ2UPM^2dld%1%(_V0x<$-o4VYbehgzZnAnsNdyip1;yC?;Xj?Fp33 zn4yLe(vgs-EC?&GA!L$JC%&}IMb`DwLbdIe7S)2#0KT+rr-aV(zFmly7H0KJ%Nwle zIo9-*Rg=P5Zyn@>GKY5Jr6oU2bPWXirA1ZX^BpGpda~y$qS~T#u-J(PyZJL9X9%fp zL1=WlEo1_u5+PwE=ED|*Ubi7+lF(=d#W-ZuvV?)?EK6u&$mmHKbtz++C4)g>%OE2l zFB8(Aki8a!a%>2hB>DhW2FH-gQq-OtW?+O$&mv?)M=U}jzL&dl;qX4vw~R+7Y^+~Gka(lD>mkG7KoMxpdaW%irGrNMWxcQaQGNiItvGC zm~7mDygRgTXb!dR`of`XXSHw`hq^tpwOtmXovwug?bYORNS5Qi2A7v;2_&_PlO4!; z4j@qrhwia_#76xYY((n0VuKnc8`U7| z4#mb8in`me(Z7$1jUA}lb3ohWm)O{#y?W$GkTbY{hs#T}1ePSl$ra>0zS6cm*TwP? z+o)<~iJ&|`E-%qIf2Un$DTBv$nr-w;SQ8w5AYLle^^_T2J*!kcaCjIcGLzbAHXHt z3=+rzay5`V&E0fYJBT0hrXgK$?}5upw1i1sS-7+e#C@1Uj+W?|S;pdV87|e1cV7NC ze|%Dj?J&Wx5#Zx)2B#fkQR=kg+g@zP=^mNoJix!W86>b78x+smz54(u z&?5Vjc<5Q6ML|PRB{Z0f>ojAtaH7RHmyFMpb}>GR2gRtU7(XX$^h;(PTReRN{CAC`>Be81a`|`-wQTY!)E-AHJClWJa+A*KhZmXJ$ z{e~cHxt5iD*p|f>X;~y{St@VKB1X$%UIMjOB*l0$NmNYbzmHagTv1a=)7eX8^X$P?(GoK84Zc{dK{8q{hksxxzLl!z+?_pDKC%WJ`k6e4Blo)Bs6H57jKBj z=AJmOg7`e1h9}7r+Cz^X0sY?$Ke%|R;Rn}EHT;m=%JB1lDkZhYXSsiL7)p+N+ArTi zmU#_{QnF52yij!S-e2-OY!rAuBmUnn)NZI>%&Zk%nw&2Ki-vAelOCWeP5!}cgNvuS zZE)RGw++dy+!muriRpgn4_W3oBubMb*d;cNCcjRBCJWq}{0ATjy!WF|R)Xx=sF~*| zpA-7ZQyi&3fYQPo$OoI|aoDx^}Z^X>f*%8U~K!(Dd`Ecm}@%9~1QWV|yuWJ@+ zL+{Sa5_ic<&P#?>0VQWp5ExJivZ8#FC8(f)B7y-4qDT}06%Zu|s2IQq3J9WrA|fhE z7BhnW@2w8oE4%;S@tp5`d(OL6{p!8jufoh$*Hm|JtH*<8$*`@ruYD6K*e5QUfs$7B z+Pqd$3TDR@YKfzI|0$3x1#fW=HoVzd3f>-9XsEG^Ex?NiM#Uv(KB3VoVs{*A$#bAn zZ}R-Bzb}z9X1~WsLDTgY{f92;j9CUbZF3_zg!J2>gC>Vw2SB~Tx29mj;4buQQa$rs zHQbPT-x+k2@YUm^63Rfi8O*c$*E54T+>36bp}u@z;!AJ1`O+(4z8dOS)j2FnnD2%< zRts}rKG1co;vA&bAr3J6#_@Q%3kcN;z&}Q34XXr4r>SMh8dk&37`GDP>CX<$kWEP6 zgJ`&4Cv?hqX1*3`uKt zf;0u88sH6zobmk6*uSGkgr|=?-$F(qodVHx*KG*h0TS-}F=C&OtQY%#%$aUF886R5 ztJVQbE4^$+GcDUTYbD!}egM%l68r7>T0UcLAja@NM(eJ)435@To7CQFttGzTp}3HF z*zN9R%_I@#MIefw3}=C^A@5FwQCt%^$t3frss1mNl13PQSHLus;C~`o;xNI&@G+lg z_>|H(XvVt_`C;K-ghARjWXfpR-Vf0XQ-1|`J?=CkwvR*-WHw6F^8gcjw)1s0PD)l% zCC8dEK-V=#%gL)q--T#;=mh|=Z+<*MK0!{M1Q>$9(1wV~AqZ!(Bj~ihbc$R=svAP+ zp$`MZzBDyeia-}C126<{V?u)6MfasjP3jwY`hrH^b8@P*qP~l#Z;OeT*zM1!%MDP; z)1H1}@sjxYi2K)iTOq5-6dbnNZrymK5{|W1_^D!eFu}VY*N`qKFb_s7Rv^K7Fd{Pu zn`dYlsV(b~9)xJ}>ukVc>keSsF@*aGj}Ybo5Xtb9>2jRe8jai)Fbo+j6JjmoP?--2iwi5+@g z%WmYr{uVSQ-#5uWVT7wG_qrjgicN4yM2PTdJG=I
A5O-6L;qpe*oitLN+_@HF zF!o*gmEu>)a7maM{S68CN^A& z8=nZz_hL&W^?(I}o}9@Yv?wsj?Kg&|3gg(GNCwG)+=`lFyc?`!y0o=! z(EH$`eg-&A1FJkLYJ4I*-=bJEDR2em{}AJ=&p<1&-YcAnWqFNR_9l-v%%ZP@#<508}SyJoVqP zU;I!`1Mqs#^t36)``I$mrJZH#PvD|{0hlxFf>j<3V0*&X0&rT)`~!RQu2gAA{q+FTC&l}?eL{8eOI%ej z_E!Y@gli3^PfW4IGT_tMDis|J7xf^(cuas*9(`hbB0OLFVpU~6(&Z3MUp)nY`D^0W zY1pK2G!6GDM^m>#r#=K2(r?S)OPdZ7Sp__ODJ+MPeh<;K)b9c$cIz^yTQA^9m1ycZ z-<)^14l~_Kie)#omKw0W5~As=ZRUyq#J;6>inK>g^#_<>mUA`6>(-v8Tj@EXa0?lU zbOJ=vLth8LWg5DB26CYl07GyB1YWo9n}XgN2NSWo05{NZ_dg6xx6;P#*2nPQ<1h`6 zdxjCyt*6JQN^&#kYXVHS&hfgH>f|fA=w%}KY=#%X%qW+_Ot+e1i7hhSIw;xD$uQ5sWshdiX|3<&$ZTF(V`$xjQ}R*G+5>FX*E6(o-cD#vb05d9YoVt*8xau z{1T^62H{9e1sJ*?F{8+XPMba<#h|p-G7srWh^DU|OSl(+p%VWBhRQ3e)1;x*Jo*N7 zY8SvT-^MFTdP!sraDS^Bas=rKh^D1}A0T$c&MERcaw>d{>sU3|t&f^+rBmH0^`$Iq zszWqAw1ph5QtnBWCdh@l1I(F+g24N<)fVEq>3+W zhXvpW&l8z%$w{QYLNq<}hXAqb-<>LPEkRHSU;^X}!+OSiW7D|u?d7Bb%4$P2Ep;+L z!e4Vip$oER`QQ)yUvI$QKc+1xww_a(@Eurt4oyP2QX|V)fix+ z;)KKh{7Qrb-DX#pc1XKGG_A$niJrd!Idu!b_rpaj4jUzQlyOKWLNtxUE{lJ04QlZu^{}J%} z17&MlwkXeYZ>@?_7&c`fnjZQWfY|--#PvUP>PnA*(;ln287MSf{d}@?LfQ+WX{j3m zBwU5PFb-MM3%wABH&A{tHT>L9 zW8G7v2xwFdfZ;C}94Lp|utj;EO+}OCDx_^8njX3|K$!y72Wk@p75KK^nUA^EY982>P@0MT^TCkVd+B>WgdY>5&R z;(7dR>wTk;4EKNIw4qy3?hh0gbbx6fPKp8pg#=C8hUF=wOCg%p z__qzewaBS00N+m{H2EnjL3SYB1JN`R`%cW1zC_p}al)@-RiSw`LIvj|df}Tw5fKX}xOei~?361&JvkrN0pw#70 z^%0VHpWaHYf_W#1rn4?YCR0GOFBb*Cb2b%rdH`JLQUyY1yWl(hE*QJq8{)44P!R z48VBlI>_Oqz2j132C`}mz?_svfhDc|!IOT{!aJ!mKa220ZEvn1A0a*FG4Q})&?M7y z0M1EQAcvE_lYq|agp+0f%t?6^SYDcRSD>aFT6!mSCUX&<=*I^u$u)55|D$C6L2OfB7z6h9Mi84M>_&qJwa{XS$^vR#-6rhD+VWsu<^XW?15&{;nt%ZS5YI zBrT=uDuD6e<^cS>hxyMI>fiVDoGDqqB=dZ;;Pfr@N_TcE$)ms$5;@GULob%&4AJRM zXGR$z<*p`5D#QCKkAVjcgC?0i18^!SC=tmXi+N~fic0Q+Lp1*bb(^|3zV(yr^z_MrHNG$>wU#<%Ud}+Vs$wJIhHl-&?JuoIv^uMAJck2oQTFS~AoHEtw23{%lFHn?qj$Dpf0}55z!iv&qzfUM z)?&X8zYWN#?SR~ahy<2XQ##!CepC)4{XS@@>8rN_#NL4YE59NalCEwD+&Lrm>zFI0 zKou(G={c)%2i1>o%7R|?1(;6`kDM8(sRlXxLB-jmGpL-|XM~gw4N8%_k3oQ$wS-IU!Jv8tSu?2I>8Tl1W_oIC z)b)K7tIq-EeC2ZwD*7%o`+~!i(MIh3D?4}IKT`lu5kJUmQzzY9N%`e)P>Pi zK|@VnoeB`U5GFnykPGzy7zS=D6T1}V(gUd<>FGHGa|cydI6VrzdI?|#6^{ZdaOCg@ z6=$B#pmJua5mF9(y@`B`^ti{s1BXGAOg{l|2G#S(VPG`CHZ%^ZKX9n*>)dEL#|?A@ z={`PPOG=@H#=7XCa!_5^l~x74Gg1Dt9Vt29=r0+Syq9D}`c}2{7jyl)H`SJFi2!T#d91MAKUA zsqniIIdwb03@ToW_ym?yQ#wo^7L^B(&I%f8`s!N%VwXQ2FV7(t`pC23Y}HWZ`p?O! z=|BE+j>PuL&*}Otz<6-GCUyp6`t}p`s)wuRM9&?MYvEJ|dUZ9xj7J^?mJZ3`k4H{$ zo$=^Qh$E!D=U83ok94fZzypUtlT23uIO8!3IgF_}n7(p6K8QoL0AQl!R65Xyq$}}6 zMAjf(57D%i?0d10$MN_Mj?@8w@#SJ(z?b$1U#TNMAUzAwv=&?A<8%o*Rp@%xZ+c)n zI%|M5uRrT*sQ{DO5KRYN5IM2epe3(DF4P8K&dZj>_6F#CK&1u;^?`V3KQFPCj7NG8 zMAI7A*c^|Gku_g(?u^-tM>Au#-@zPu4T{y90CT>Tx$8;aLe{V4Q8AB&;B;C%g+lQ)m7a zFI{1(vH_-xe!}(!^A%U}{))ewtqu(0rOL@N7U^V&ri1Pc5L?zoWhQcJF~IoW3Lk5z zaisC0)79lgq_09WE%lQCT)x#*asoLuxVKwHuKb!V;gy5w5|#%pFLHIsrY5*8%k0cX zm+X4b>ylDnv&;oYPM5r?QUYDl0ar18m+UQo1FuW2gKuJdX`tUFyPYm!@Uy2Dk@+aw z4>4VGdAuxkFJ>6J{)r$^I$g2_Ib95ZE@7N)YY>_cS`&H$&@#uzCde^l)gJ)Uq$Mzh zy)G$>yhIp8v6+=5F^2t55KRYt3VU3)_HB<6^*+GF%N23czPv~>?aT7Ol~k_w zC9B`QKl|-F0cy)!TjjKGjrd^uZo&nd-@Z+g0`0pVzKQV@fqwhmg^M>-rv!9Ezm3WT zqz(Gy8!wC9>o$h2uOsvWIPIGTPdyfgXkW(J_94QPgcXD>0JQJ)#}cFkOjU1yY2Wp= zbJ}+^@)BVX4fn^(ZAkBcXgcVQ0I?_Tj>;6|)GUDU?*JdGt8t|9rh%1Z8PXRZnwI(@ zfY`~~6J!^1p`QV9H}xKFt}6RHfi2Pg<>h?miV|NtHy5bP%ga{vKCV(sGP4ULm*CzW zbT75`3UqJPz6fF^z;rJuO!xBg&~z`$1J{tbx|bpN-8-)%_OqLBjlOWP%vEJh_wsjw zoNqL5|a<7RdL<1-7siSai8{q8kCD@f!5PIjtLh8#p$;T!Uem&NWCD?`_H3C#db zY`^pHdH}kYakd>o7)Q9D@C1PE!*z7~JldxnPNm-S&c}JOX&+wcnD$|L;BqKe`!Gzu zeMb20b2rqMxj@QkA1;sQw2!$d!D}Dhl;F3|I`}5WPXPMuV^-58k`7%gmMniDt$(B2 zK9|SKV)x>Uq3df2HvydXDGpCP28U=L#@TiT;R(WW!e#*5hs(!y(^^Vyg@ZcoosUyK z(>|QEoAzOOU}m4IeHfnVqneU-6E3tNX1re>FN@vl2!^gl67B&w?b8^=`f(hheHdrkRfNricL)an z_>`_mNRrVo4b3-a!l#sz5A!MYXCdEqb{B#8Ddkl5@=qzX_H?xJZqTSs{oJTYVOp6} zQPavS56o9{wK7BSTNw{)bZer(W|i?^MoQ-NxSA{O_U3al;)+%=>(!**@+PkPkq!=e zYZB`6giHWd7qU|19^}+AK<<+)o1WiEUO|aONZjOLOWBI_qoB7Yp`J#V3lRI>dMWY^ zaw>knKTpyi-;;DSCn53gk;ak%&$>ZxO+tN|Z~=gtOiYkouuyjb%s})fz*Gk&z#m|u zQDS&03-`(usKGgJU1YMn6=a^}#22Zu4C$83sHk=4rexLq0mfaALJr-(oZWv4c@x4L z7viPFN&JWzUryCnhpA(UE-yCZ!4*8P&SM1__rzr3svx|Zj;RCPal>LvjBxAFzzFkV zzj!HL%Az>b5Mcf}f9lv^FZFFzjvAX={*ITHC{}$uo1)mhY(8ftGpuTra3B%-Z|R&N z|3NwyqUoVK0mMFoc<)C}Jq0i)S?QdlKkuSh&67mvQ}JAqtVX&CqUoR?0Z8~fD{%r@ zQ+wXj{9n@&jYXaMOgciAHy+nvs-%>d$ngC`{Eu*kcKE$y#R*XpG#96m`jhq z?g_tlIem6|*IlaTDW0s*NhTqkcNrD6?lhOIx*5Rf-EPRCe$5!>4&+S;=SL^x)Q|U2 z`1Rv~SHE-kb_?H1$8&Hr-wG`D>gU{haSNu#f1x;3c#zxhe>pa$cM~`@woBvwj0_a3 zCZ3IZ_fUUf?>e_+(EqV;L1~Y)2Sn3DR|klF-h|`G9 zy&?4PIwejXMLHj%>7j=Ma6<>;eHJ4Od7Uv}9BVz6miO}cN-)XWN=^==wgMJAh zb~DVCmUXhU=p;7unE8fZ5zK?VP(|kLqOJ4&rn+7XdKp%r5@X$eRs&rdVRTeoV*R zBX5*_4=(CcfEgZl!pbrh5waH#x0K^Z&qFjV^&SAWv0-*r{8l(-0ZdEy6Dg_#6DiKo z{0`yV$!Q69;g7r*kGGY9a2kIZ6}9dhhpf6Bz-ftk$l=4;^>N&2x*U0Pe<`Od=AO{9 zw^lHD-a}UQ)~04E6FxIB5qG=7*8GRT-l>dJn9AV5`DuGNig7U1?5bLe1N`WK>CLxr zP`?9A>u~ZDSdSyyo{gm?;Wpf=3DL9`yW7=R|HWabGQjxN=R_wl{@zw8Vm}OMi2{`k z$UTYa>MSFUl)5J>G{;2AbY0yOjmRUBn!x416_b0B&J6l$(&Zb5#sx5(&=4e@G&18jKAGb? zcWfvn#}sV=OZ=9t66AiEss)~XQ-1<=#;VZ5OrXfM zR$j!`TVudUEk10{CBYTgdTTz0nYbkQ%B-uAsa{tlLy+DT^wuQQ4GAp(VsDv|AWM-` z+W{tAf2u`wV5;?Eu%pdX%XG9m)#6!JuTPUxp0Lk=xg9uNEs zWqN)&Q!TVu&QuF6=FFd_V*b<=#-V-y(;zwXr;!<9G=bOA7jSzucADbW#-T7(_j~s4 zt&NxlprePFHY3;c>?C;{=~9TMhaL+M``EYyS&Lj~3&5PP9ewPKX$i6iD)loUcNcMH zns*mXLy1Jlw4g%;Nf}m7N<(~aO+tN;@ErjARkkKbOIWCEfaxNCB1d&#B6rN2$nmpi zCUT~W+=(2|QdYN>N0BbSjEY)!B1cx;3E*_mO~|2(+OmuGA#Xy+naH7wawc-71 z6|=LWFb;)=yDgS8I~$p?5|b#eix$pg7j2oFAW@jADxST2^QhBBA#?L6x!%C@?3y5L z3DNY>=>YurhpFDR$c6d>OcxzBQ^0MQ29JVD%>($2Yz#Q>&AW?UMTtbnwC(5e@)FXw zg5H{h`fkF506gdo)4T)7sf&QzCvjIo^Z2N~gc%?1j)cc==UYnA5#^*3#P`-D)JF+t z0Ad&1mLTPUax+wFD!_D(KgXmxFvmRT%`rJX%pB8<4|k49+Af3I z%37rFTt-E$JI5rez8&Cd|(n}CcYq8Vd zmoyT7MF75^L}+qsN;Rnjbq$E7k=RwQNt71IsapVkV;MirUGwhGekhR$S@wTs$t0wY z1-&&1bw@&P0QTmzN|a5=se=I1r2c%F>cD(?nm18plbZRmX;OE-%(E0~&_>iKOa&pi zsi<}5%VgDW0-Pq@iyXq3gpJRYV5FJ=OclBDInbo!TyVa)v`5+%qG>I5Yxv!WoEi-9 z{Ukz@d*hR3Jkon0nnq$@gK7EW$f;KVxto;p@VuM!q-j#JEIgDYpCJ7v=&ebp7ZX+k za6d+;L`fWtlhgp1cJSx^R0rn%g|M(|X1r`tGxs-5>dyUnmQF{SNDnv-xr~ZhckWMC zT?pVbX+`7^#tYaYJRVu~5WqAIw+IKCl$=#6rprR4%OIN8VvmB~OUS8L0luF^Xi~pL z3HcD|UWlfVWEDwHkTb{)yuRn%l&y{OUcccC*c6Idfes0iymj|t#j35mr6{MZE#ZG87?d< z`;!uPH93!Ox{ICp@#EP)Zj&i82KR(O{E? zoFtv0Qo{je4Ea}0R0pn_D!|?!L*}Z9@4W5|V015EhBt2p~?%M|$&DwR0S)$_8=*#{G$?WpNhqy)+;K{PFO9CBjMhGTu? zR8xQyd0YL5M1$f?@_#+(c35}`@=ew}3^()%EqMl#?rG~ycM z%xcrs)r&bF6fVM>@4vbytxR50iru$yio6QL&^rLbJB!z5_LQqqWFJ)OYfsOsLAPGC z9q~#{`3dP?5KT+{1wib|aJ0w6u_(Y0(orI`OFA5t%23yYXc~#V3+KNAIn@ea%(-@t z%Uc}w>D*4PN7@&nX(R()N4@Sx&eZGUTKU)OTxN>QhGA$S!0=Wu^(uy|%~epTS3Es0 zGTnO7c76RivJ>g25KT+H0f0^R4e&z%a_S7g5Yka1v^(6ifc#DMoiL}7*aPaMNFf}m zQUGJlp9>^HlfS>wQVVH)h^CPYIEi}QfSjpULT0}83K!ppPsdWv^70Kc(@^AGgqp9w zy05D8L-0(4y2#XaFJtYUHs=4Z3mpKM*xV~ctTh(K?kaqno3+b&=K8XzPSE(G##b?( zXY?&Rth-#VoM#u!k1N+6S9#8gr(FO)XM|xCO7kpFnd^)9CM8H^s8k(K&&zFz(Cqd> ziE<6nwh&Dt42`=J(KY$677m5<0$@-<`Wdzc(5KSX2XKcV#zWovS=?PL7cA-816J5^ZY0w1|P|@67@I?x*D=tkS^Zg#~b2~rDop`HL!_nar|pzcL*ag@96{qevru@`PmkU?}C=DEGc>l*u} z*Aip`RO$gw&kIM1(CqWm*T@q{7eO?Q#LnK7Aghs6F9XckdEq1xn%tOHRNh1S5k%7n zE!z>-tM=o76F64(#DBIdPnK89k^?tF0pri&+aG=ngT9({`X|EQw6rVXdZ;BVLY)C7 zF6Vk^^E(OB7b=zQ>3Q8F5t_aKOi>w&bOJ=v2>*}XOpqDKsks0{!%H>{ZyYxKrMj#@ zx*DQsgz7EF<(7RG@Bxn1NkHy_!;|GzJ#t879q{@KaZ=zO?0ACszM6D;FX2mC+Ed?; zmm088Ej%-ScTGcQ#RoI3C2>_&nP|yXC{fD|NCyWUG&ytw04`He)iKD0?gf~tK1k>A z=NzMNLubBBFW-T)3PsMfpFc*q>)-i`bgmxlX!{h|3p@3#$IaUz0+S;0@90)denol_ zqG>JmOW5Xw7f^n;$ zgC>Vg1&G}TRlN?m&Fb_r24(&ULI^2c5JQ zc50i)&D#H;w4yj+zCO@95rl72|DA(}?$ z+y%G4vG8+cvX>#@&?k~L0?Te{Se^^fY^u7x!)nDE&->ExBsXEhxpNVE-?~chKa}X8~f5MpZLW7|H^es%|%(djmUnqVd90059K5 zGzdk`^|0T$+y(D+u5(3gcS9#V1Ut3Z67F-YbEZ&B;>x$6Jxh1z&j&gEe*J4t##rEc={ybP2G%_i3`FT;_JfoK|u{isfo z`;b$Q08IS6g2kmN4u77RA&ZeNhiDpM__V^+r~N2k8;;dCfZW4}C(CR2kONo0z|wyg zmcNju+#m4Oq|-YIp8@bI=?C#r5f(20TE~f2{7zR6Gr@r*Kxtl*Q ze8}IvNqaeq^cRSxHMa0z=U8Ye{E`7CCi+Q)CI@?#m(oyIfM^=w$HbLs(gZox6%f}$ z62_Epyt-kwG;ICQHqyWvnntZe=vMVaYZ;7mRF1i(zrF?__VQ=bWEyg6u~Xi!olCc5 zK!A=~iIAec)j~EQ-Iimn>8~FHh`s8$H2DlUb;c<_yxOJvXkdVjT8WTi#k4Fb_yB$e zgrG7k(Dc_|0mOdr`8277B6XEhKDjl1H|6sjFWelUqgEoM*mY;7^g!A#$6V82*9KrY z^Y-F06ghQ2!0=3fYjot`Q4-frm2`OMd-t<={~gk$5KRv~7J$dW-Bw)IBB!r^{)VQQmK$sf=)F67{|_ZY-k*5ob+N-X^pfSMAK5&2H-*{JtQ|Fr-lNIV>}%% zj^-I@yzyLpxd-WVh^D0;48Y21T1cKiPOStO$0P8uzA=t8F1MkTY(}~rqG_p@0`R@C zDoqX`r_KV$*>G%$tNHyys7lWCuy!H6&~BOvec%neCDAO5Bj- zJj8Kcm=7e9a2Pk6gjNMfrt=;Wdo7-qxQ!$SJQ7~g8j_p`B+d)--VHp8K!fC5kfbed zlnAT+__tH-p}>Nr`Tt|K&+;M0Wp@TvE_a%hOY1u`=zC zW>=`xz@R?R;Ix17^A<81>75WwYZ>wl);mj1NST2HwFqD+D}yrNO7lg}WXf8k8zGw3 zVn2)Z&F#pAJ^~m&XMJ-7cCUO+{kNXp>|V*5JG!X+LH)z%!tT>uVgIs zlRUlISCVxEY;&Pn3<2AxSSBx1~pOEzyG|s|V-MTwead)jOKK_vo zn8V#dpPxJ*KE%|U!;uL4o$;H8^#~%V1Tg2ynGSAK>0D8JQO)j>E+yYqQS&_YwD?;| zJe|ZH;bem813QO!V1#*$kOf*|B)-QZ&RKLjXqD!on?qdiORKm`+>*p=NW9!59soO! zn0aj6oq$FS@1fOB&#J9ouYnQf#V&~@xPv_XG4J%7&4=HMid9*A(Q0?Fs8s<6jV{P{24}`{9 zS95yESxSqKFT?uYO`o}*4~x7WB9YiNQxjwva-sJDriXGC+|WZ@aJ#IBE?^NLyqLsi zJmM*^^BR(QG-NL}dqB z^aCTzYLR|DqpJ_#&H=~-+VfazpsL8>eL8bN1j&IKtywl7gBH@5Mx8WuGAzhdJp zXE>iXV0kUCoYgfS-wbG+WjANkIV)}P@!90ONT2SW z4~x7}MY|^J164qR})Ihh_mx+|HuWHT2u;`Ek+6?|_Hl z=yt#mSe*+R{a+nmUWX(09q0f(2OUrvU@oHl)vsI~Fq<((c!tzUmNbTOi=cxhhmHWm zz5*5LhFqu*AW)GfSdj;zGb4$oF-9f}JVUgFgBtp}^ z&89s=tNor;Luc*C+`_>;Hl1O~<`{}YrvWA+XUQgd2!4g;x)E{7#&6ovlZ4zZi=9 zLd>Zq;-9+ry!;fac&;b+N39et4*h?h+;5M<6XtINTWGvTne&ACwD`sB$*l}(F^Ww^ z+BnNNrYD&f`*6V|Swo6V9>ph}p$~sWzr&s%S5N$&tO`fBClfvoOObglIgS3Wp5!fe zIO_-K$r|ueEsV2!1D|P>Ts>Kv<`JIZo9wc31Jc1k2Tcy$03de9H{)dta-n+xhN&XH z@37mqdl8unm0AVB{I3Qc8eodox5bWdz9+Kzf(lehgq{s5l$GsBcjZ`Xy6Z&%u?w6? zmoJbDeFrc!y?M*8y=_fYey08}PhTjF`p64Wv7dmxD8Se&qrZe^uRwh*Pj8w%E&g4s zIE5RcSwr1XVmhq>8o~Q!V;;XbvV5{v&UQDWa||@jik34vokgtp_@(50h(6DHJ}mM^ zCy6BdCtk&D?Pqy+`E^B_6wAbGrrsx`S$FzaRkKPpv`ijrnO4V;gx8YtPmhvC9wm8X zK+U8QQu#^4co8dHhA|z-$!6)vWi)via1(5=b8HpXn~ZI1+M0DG*Vg2*odLG@IJRGT zwj*h47JFP(PX)q`k|hFG#6*Wb5p^-06N3Yg&Cjy&mg;5ow3J!mn_edWrN4khzl#> zm4(3awa3Dh8J|V_&m~5TW_iZG?_{!^Bg+MkgXY5($ zlBFSxL(KuEz3%7wOgLN&EJHmOuFd!?Wr8ebfyOTVYqH!;mZ=`gm0TYQ$BKhxgU7|Ph0b<;?ni_?n?giC=-0B;k9tPFypgPYQ zt;DV^)aV#?eW6Crz+P?goVb505E;oHzDWi%@AskkNI1PcZV}7JK)i0qG2a;qp7+G8zp}ezabI0X2+wne} zJ}4is)a5CHci#M)w_S6!%73HrdTha(Y@Gt%f9b8YjS$xV`mMEDKcY>B;l$5*ZNk?B z{hQa4vl(QcH4N)5e09;Q;Mw7E@y$inc@WAM1zyGkK3N|^(T`p+r|{-fy{#%$9_1q4 z{?LY1TC$VW)A-Jc0!(+CS4J)5T3dYl-_PRx#BH8;-eFLcW{o{FMs9`p8l$WZuxbT; z$7L?XY1Zmryw{woUVcxnn$Dj76ny`so-T*5{^vbC9&OSaC!X!K3Ez+OZ+e=|fS%4S z?!3EcIe0dBTzqwtwSzq!AO8T}#uVF)v+)(uAHpK^xo5^xTA$M_Yad=^^gZ-vpK-hB z9ftESx`^v(eiwDX&b7(bcyRrfy6A0E{f}QEon@irF5<)um%DAkH#_~CE@CsFiv$Zr zeCt$O@bvV!4sh-6X|_GyIEluO^>p8wH4GM^iJloxX|1PO)`QE_Wd`)`cx8Og(!UAk z8Em-2stDJuV&mNxK^^AVPI-)c5tK!K1T_!uc8VRrIXQ0PR=}(fzUkPwb<9al_0~BQ&Rw@t+~lD;78kmORttWNasT|ttBgp&RU7wQHXKl z%bNbp*qDs{dKepAc-v>S1M-bje}FCiS(lM7q_W5lvjccOe{8q=im9@&2-Wt?%qyno zgdtQ0?~#h#;=UuS4b86i%ov|_s~JMaU^Ni>+n#gJbl1NeLR|UthY%|=+4>oL|D_?+ z3}OARhfvm&Pp8XRocJZLP573kf772&VT&F*imQG5tanU*4vKyNwgVm`-$Z4RA7){^ zGHjof>Aqa*9F%{1X6D^ebV7fw#0#WiXAv5|5{f;K{;UKrEQgkBgA-5k+JtX-`ZxXA5_{&*QCuI3O#<*0 zR8N8|^sL9oS5R5xM@QAhTcu)S-Pe7+NweLa8Bc6k?5KU#XYMPaj-v3KXOQ!TsH~qr za|!ymHEvJV`IkM(r8&PRS%t~gBdh;wJ$aB+|LZUJt=L$IFZUv75>>@(7j7vGd`Yvt zemkb53DRZ|O>41#!tTCK$fP|KA{$r?;*1Al0=P*@4S#jty zf@P8wg1R(Q_tK&%^Q1$TCgL0UTRZEcp(Ele`CRwjTJszqlxz*dNrpQdF7JKT> zIZTUCwi^1EgJqI69_sB-eUw{0u}X>@g6aoP-6)5tH_Cp6US4!-&N9iW4|OtB#d52g zR8GN)37%x(smtat6-HTe=-UU&B@IE2hX$s3v&ogdC>bD4P!boM4$` zJrDI#s8;1xZ}>b}HbeD=r~UzJtU3URdBK zSJjl;k&g1!S<5P<$h|mx6k93F-7`k;f~ODj<7eCME-K0d}?5~$vU@s6N_ z*axxZegGr`BYrG!^*kD#cDFjuTug>aqJXtE?usy`|(sfuh z&$}LGwLYsJ9D2g|uAqZoj}0jG>v80J)*}U~X)vA@bil)#vFg7BDz!eS57dM9<8Le_ zZzJ6S(X^Hb)%PtaJ!t>Ln^((2NM}Jbt#N-4aLAsHvJB~SzB+64 zkIC{f4i7<0e=NoC4|(^8`Nblu3mlHaIB8?R!LP?2l=}7f^E~TO4XWZW&Imf-1?x~} zLZw;+^?`cOe%I;d(j93GqG>G=s@opXG6d-eU!5htCCfw{J`FMT=!M@d@~(&Z%_Qq{ z{ARKi#`}T}em(kv&acM5tFL#2{l4xBkq584kY)>ulSJOiR> zEfK2UU2aJ|sG9idtQRjOOKTk71~K*E`=au$hxyeeYX%&~!uYA6gI^Exn0qs~z~`Kg zxtFYLI6MpEbwLL_*AMD9p;G&T`anHs-|=E6`3C6;h^94e|Dzgf$$uiXHhJo-v-lY( z1&4JZrXG9+Ro?Y5zZ+%kfI~AF4-7i^^)Szd_v*oC!%Nn3I1GpJok0gY1RUyVP^qVa z`anHsf7`HhS&eiZMAKR#R39u}SGFVlz*lGWNl1}Tad-h@>cO{KI0uGH~d`-|n>=XC}tSeM%U{D{Z2kmEn)mla)y%VBoEfK2! z=-yFgAbreNXN@k42f*QQ4a9`CCg(p<7nHpM{oBDZd~l)u7^=^6t814@k?*1U%~L;~ z!<2}!xGk6rL(ExNCRyK>#tWdJs+db1yMJ-ITyOP>HiBAR=aoge2WI;Id!Hpj=Nq4I zE;l2+4WjAhsK-4Smb-8?$ydj!f8<$US17Mi$zb@}6>Aj+%C5f-#ksuIFH;GVv_UDk4gQ~KxjxF2eMA-x? z)dOI-r{ohQ9glv{M6!{NfM~ku4giUjeLP*dTenB2;7Bb77^WH5<{c%y7k-KR-;llx z(R7mt)l25b%X>&a^3|~tdz>hbK&Ac!81CiyL`g^Mhf;W20;azZO*j1`Kw=ALG0Kc+ zRp?bKfMGh@E$=Al-FHbzxt=LR(+%&41s*&Uk=v1u^3}1~A3IUr50zR9Fx>Cu6D1vM z{EW9`Al(Ylbkp+y5-UA}s|HU;cjHK%1Q@1gf>*AAu|w~r>4oGXtzQEP-6TTwV}Ep# zLQtjo>R8e~C(7zjsWt$^{X;%c((z2Y($Wj*K!~QBz6v0*r`@%UJ8+~P0~n^fU#~3@ zde?lWtvrqNS%{{aquzf*YuSRMH+*%hlHu+h{TM2B8eq5|30{K-T7Zt*%h>W4(zvbQ zrkg~l&hBGL5vWS|>e!S|oEE4Fm1+kt+(~ou{$$cI+NpuWkPd=qy6I*BiIsnV{ZuA8 z7Ds9}z%ZTfl=o<&_id93$TLWvhiJM=Zn-}2S5KleLP?t@C51sLv{`9w*_ z>z{8T@;W9)5KTAzEkM>+GBfS98bdw0x$KNh2NpGMU5Jw$5dBBOYJXER~z;KVqCr&!fDb_@~ zBJBgwbkhw0h;!~_wr4FXIt<5Z2EcHgX`gqT^zS^TqbxxBG(^))B2?Fy+)mab-Qufb z0}nZI?u1Gm2N>?z`NT=bAD=HPzatI332wUS!vMsYX5vh=qQy|6>H!Sbz=rcwDY(ff%9YDhI0Hh^fl=`eDr@sD@2#>K2?I~=P)029@pZSt-${qN}3LdGDy z8=~okR|Emau4^g}Bc1K5V-M|hYP=jO^%lTz&&?-JI&S{5wCqKC7^3N>Hvz;>{4h;U zBB%ZW7?!u@1xAWQ=rHZtL0VzQJG`IGmmd8FDkbl@+~4IBB^~GVY$8z@mV{`!>0bd7t9Q()Z!PFmdygq0*o1+} z1-;L9sUv-mWKExEtjYB^_^R*;1@`%VE11xap?914wMuH%@&^L$9&` zhN*i#O!OXdTOsLy^m>SAh`rvdl&L6hzYv zKL!A=8Xb|7SicQyxJ2 zFhtW$B2>pZHl;*%V@S_HG~M+30L0nP#MxFw z)%)No4KQ4*o97)T{jaQELTbaX0YuYHB2=HgwXU?rQAb}LEA)dC=Z#RQI{}9KLOyZQ zapaTN$PA>9Lp0s=Z~*qG+>kD7tfSFqairb`7^ZKVe#$hPLy#UfV(WfaCgHCp!51H(y{6dEu;<%uY_p2>7oFM-H0exTg{@Kaineo z7^c$sL`m;re^-*bkWPYVx=DoUs++3HETm8P>R5P<6Xo+zsdoW}dsIGA(y`sWwdFIU zUqdwA^lJc#echK)rbU0ok&5hanP%e$`Ml~&?}8mg%E7QQMAJZ@Z9 zyx>IH6)N=~fQfQ$K2g%~v1t*RfOHx}(@hTmU}QGzl`b8v$F1lB9IMR$!&UXlyyK+* z;58lO1Ejklnr;%I`oQ&_=RZ>dK%-TG2x|R=)xa*PU17-Fo!@?RX~MhmK#zAewFxq57eh z>Pcy+D){QP*1`>L>otZ_bq6?olTV~{E&N;q8H98=MAJ>T2T1I#O-}1g#*tbCFiait zVWRhAC9agUNMC_yy6HIpiPd=7X}u3|q`n6jrrYyjqIcmUMdS~paUX$%Zu(0AT5ney z2G!Y$7DtJy4=`Liv+`~|`ft6pwzNXp0ix+95vsd2*U}g10AIb0K>E?A0~P)!BWa;q`yNn-Sl37#6Eh( zX}!c5>46&2Pq-ev5X;ks1y#Os_P~yY1+GrLHEE zkUjv>bi-mWu=$%!rn@nCP9g5)S}{VO@x(o3@dY*v!U^X$>t0(gP4pH@yxZu@|>F z^*wCs~;7EN3Fie-~=N%=zf4i=_{D$-rMAJ3CaeLHPq|+&*yAO@9fH*v}b^vVOEE^r|kv zFwMs<=DcP|^gfPTbXp*72hnuHw?FW@E3056;#pw6I<|hN6Xi&#)FS}H-7KFd>3C<) zuq;OUJVet?-v^NF)CEN()Kx`Z!;#tzFigoEb3NRMPU{9>{R8QD5KU`|Q2qGR)#Nv% zmwff%0ykAFAXfQLu@4n)k9j1D=u;K^i*inlK{i`pInTODB9olXFb`9 zwCgg2g}~`|hrtV}?_NJhO^e^DYP3Wess{%2++r7#RMj z#=Dw7de;OpnM=CT&)Mqi!%;)vpMXhlfB2bDSm zFqDmOV~#&yB|`h6+Y8G{q(4H8wWEIcSOVUghK=A5M;&Y9-d&Uqm1+nul)Syjrwm-N z({XAOC2f#)f@r$oVWGgsw`jQ$>CL`6*3`Y9Weikm7Qk@xhL*gz>3CJo>aq;!a}Z58 ziBNs)uduv=^i5wKJ5buW5o0e@>I}ee^G=MsxaoL)T|N1mhM$3(ZW5uoV6{3@6snTG zI#vze3+*gh*Mds52N-VNP>~lm9Y-y@Qu-k64>4{Mp*rcRGBOhBIA5LpMVV4k_%mfq zhe|C381AOn$?J1t^wA-d8ZU1low?E(b`|~6*X{sHNw5SX?+HjX+Rt(AU{+qtIcj=# zp)8TVVOPKU*VTPH%eM<%1@%?1X#&}0t#paJ4+=gh8(mdipA??^%)HKWJI*%TwZr}{ z^Jgu1-HTx?xJN-U7{h2kRxcNJJ=iuTo(-^l5i&V6B_|%|lO@?ROQ97HDQged)iHo+ z#jo$l-HJT-drx)}eGrpOmn5eZAB8qVf?sf!(~xSkZ|O(2Cj+gBbByxrdR-yX9b#J1 zv$MprGf~e%`6G~BR?nQM&DRZDageLZs}*_fM@z-YdYrASYnRiCafjSkKLq#t5Yvhk z{8+u5--?La#FKEPNCC*?P`#XZyjCoVt-P7Tt)!x0$4kH5R^%SroUs_;sZV^;L9&oG zhiF>k-yZPyTZDP7;d%=P||+U=1h4Q>1>FmwM3{sGd)qB zLHfL}&VC3RZb|tXRBAWCP;%F8fRgr0o=lRjkbVczw8sBG;4A+rD!(DUAM-tR0>PMwLj=K8lY=xby(v)NRL1? zt?`Z=U>cT5&LREPS7(pIUR_ege*wx807J>0xdBSr_ro-^2FmL~G_55FkTdyw7_(X^Hb)!lDwE{`K!;H$I0F7Hsj0F`EnvXUZe*gn${Abdg;=#@&nSdzB+p$_PH`j>o6z_0}LfMy9Ov}e{Co# z6;WOtqG>G=s;{Z8 zsUnXeo$IT!*JF1QDOWD+Le-|W^*#-+y}OlcNlBD|iQ0pyWz7J@FszOKHYR_tz%EwELlj`H+GRcmE@lyTHxT-vmaEAkYo++%&b4iYL#_P2CWXvnZ zT19{50bk9EgKFmB?f0tkpb;CT-k(q7=Ye~z!drMBzuhk(Rc-}MXcWLyb!B4&W&h`N zR3=eB!_(hZgZjh2N99TCmwEc@GNBJ|Wb{8mcU8Rxankjkh*)tqwQDDA(72nrv=g4& z3GrpCxM_0};>%V=b6`Sz*{VY`3IEEzO#RAj3u9oq@Jw9wsrU^>!nlD5?}MblpId9m z{pcOb2MH(ti6^HHK&kV94m=(A2rO{mJRSGZ3rRe1o{qZ<2Vy_6Dk{_95PBA1I`Rel zPGFu0M0GNcoaU*4XBFXjGUk=PjgU95{QYJ$J2E_jFVQ;ZG5AAa2N}#zaN^1TC0fpa zu+KdemA~MmqDS3MnUfTl&*710C~Uc?q-3DH3PjTy&rN$WDh+X{+IfEG5_0*`w)MlM zB!;vfMAI6(haZc|P~_AkfEfgd@wxnHyKh8MnT7NTh^DpJv5cs!KrZwm!1$%Uf$p<^ z-kB_~Q~$oFAH)W-cRrpZpF*WhdHVUa&F2cQ$r)H>G4c|j%QJ;r%O#}oU*lAC&_@7b z_s1u%D0HeS!2etZzHaEyc?aH_hO{|E(;B;RUrLtF$c1_VjNh=~jOStKv#B5M>2Ib! zoc{kvdk^?3imreB>~501AscQIItdU01PB2_FOeolM+9jC(n0A+FG^8BK}9TxAS#NF zpn{?xDx#u*j}4e9pNuXU;iuW_H@nPWc*z z`C-ts1&UV11+i6eVNeylRA^aGEnAEhU;C!wX|j{_-G)B!GHP#4?WaLg?JV$CerGI$ z?K-ESd^|d$jYOq-ev1bNd6KlP0us+ zR-w*c3(d}TK$;zXonMDKKZW!;hVIw-GUrm|PSTefx?krv(8Mo6)0)nTd68isSw%_p z4g1JSSC1O@kyVw3iT4Y_k5Tt;Mt3WH1c#`*!l9T7?%zB;rvmGF1HE~AP7}JtJO59T z)C8;34M@||7;TtV_YB-w*7~oYy2In&Sr-207%J)ta2KrVlR)vSdZePxLKRm{yzP$l z-M{V5ukhZ2Nolf{F+XQwzC#bzse413yiWQ%hJJuey|?y0SB`;XCm+_h>&Q;b7Ql$Q z(jN_o+CJ>hbjCPcD)h8U&BjbIJTvIYI`9N?EQKo9H^p*)sAzdfH7=uYPZ3=JlD@}J z`d5*62fIBONF&m#$hE~up`PRy+Q>xw-vn^U>GlJXaOfL6iFe`Do`%g>q@znZJBN`et)JhJuUnzn74g(&<~JH2U59b3bxGT{i1gV$whcwfJ<)i z*0c;sOStT-fmH7Hf^y_7FP&u={zm~^a^o)H%R(|8F8g*Mm8)1#j=TwXd$J1uj{#hA zk1Dd7GK2%D*|%fO*tp+-;>9?O{j1X(NBUO4G3GLrJOCf2(Jrp$t~ff)H?@G)w}bf3_pwL&`O(i7NcTQ zY0F@U9jUbHF_pGrS9Vx8K8&&E`%soN@?a-f*>;8|7x5vq{uOxmoAt>Sin7 ztV>z)%(!=7NSbvmLAQk8r)AEMP}XrkF=5}DwBWZ!>d;oN## z2RX=YP{*z80{t8unoXjggF~CETXS$&S?wl6p{)SHFM8ZZ zK(e2YrNz6!dAgEHw$-uOg(@Lcx5 zYrVO+nAF{WAfcXh^;@j4DSJUpJ=aNUK&q9Iq{o+Zsl@#gNTu93gz)3ZkA4SuHN5r^ zAWcg%bmzFb0|`o@s@qR>kV*KT4RFcn_6HKLMvX+d11|ePAQikE=Z{(qc?f9=nSUvg zDeLk7BETi5y9`LeLzwSzKL?eNuMk=g>Cj5p>1Zfg2|2ciR6_NbN;m}};=PU`MH+ND z89-`94;^ak!-We?NbjK$^_(C)4RDcF8#$_M(Q6l$rgBr&El%Q6>?f(4QC5aP#GY;R z{ipe8Dn3}UtlWkFWdN6A;w^WJ${M)rtw4T$ax>R!X=NqvVJjzVF!PoCV}8w*_wfH2 zz@>z{9tb4>gQr7qITwIb!2~Gq->s&VeP$xwvabe3aYbC!WRncl^H$a49Bz*9OO5y$7HD1CUzJWO{wM9RFPaF2&rcK;rGOib_AY?1@0CWezea^4xf_rle4) z`0w@Q7W^*(xRh{*0g0E4lDG;k=S3h@uoDW*bL%$&k3+yB-sf=cBIALJ81=c4$A4~3 zUtHtL&*15NB2dUH8PJH0lZ5P0xx5El93O|-ih3iiFC_`pLM5{hj7@+OB%yk!cpG@? zeYvKIB-98MdjqY<74=rm>MRMhLa7?mcU8P3)DERE2u6*amMYT_kFyF$%fg;}UB6=> zIhh0UWDdstq~)HE4ju+zWUhk;j>;_^jpH&hR&wWL(6n>NtF5d$BEw_0vdz{xm2^bR zcNci_)uXNalTIo1XpMi;smZZ&J?%9&JEa*S;e(8(GNa+7m~wN-lwaD)PzJ9QrlRU1 z+_?-$pUMIiv^giy12p94UF7thLC@Z;XkCe!0g3UGK)xm97ywrC!NOB zmFR^sOZ9q*psrLyy?-7e_HLu^KYf>@;(vZjlW*~V7~oP2mr$KgmfzsAi<}A)@Cr8_ zoA!iv?n(;zH({PT9X#0pmlEz@s$eVLE$V~jbTtYjrd);)YHa`#Z$9v~WSo8xqn1nZ z1ntRNqtav{8J8Ny8qA~7?~MuZR+vXYwKtofzYD%VP^4R@F0u>%?*UwL%jn)CLlQqC z=_`2bUxD;V#M*MKMh3lVOuQW(@4Iq_WSMWW8{B>4m$z}Mz z9N>~$M)bc6vRA};J>avi1M=f7C`6mn-XNBux z$!)0p=P*>;19xDkI4iOqqTvn+dAFkbuE%=g4h^O6hZgiJ-9mC1)HqigUB@7 zxnIMB8}Ek4%P54~Gk}tKZ1aQ#mT%%GP-#^sR|+*u7?mma;{V}TMwhbgBp}?qczv3z zhs)Ulq=}q79i_x8fzq>+^luG)Dd;BgLq{Ui(^0Jlypu*dVP}aanwGGGm^x zZ#7`efaCI$oqzzh*X*HTZu9;8MU{00iULKZLRR zt;~_{LC^-9@14j9e>$H6@c7gD^hkI&b~8IUe*`UVGtuyt|=a-($jwSe#J24*XW=xU#ugrA^rpX>^`NC+~Hiaep zlR;_n1L^kJAl7}Ln-X3Tq1IbK7=s_mkQDGZWewvRP7`{U9n6qwq}Mg{2b8{QYLavZ z$({;?UaXGX(xx08b*1CXBtL;GYw9GSKhefueI}e}E5nzz6sr6Q-x*fm|H(_LaT$f% zkLY?J@y6YlBrn2czXhbZXhW6d>!XhAr{~J&`2QZ@lGEJ|BwhkO2cLk;z5t|x`!SDT zozTeyG7mfDO3FEW^a8l#bk6}vc)=%lH?IpKGHtSS%})<$#-4l4l9r%avw>9KTo{4` zj>FtusI@ks+G_R)rC?e2mm`!S=g1l|K9 zV_3Wff&S{_n21m-^^=_3WT@mhF1EksU=am#3e$Er;2GIQKg~3rx z&UkpY)k>2&Alpj}KW7Q78bH|B589qusqzT^p8~k#bZ-Na@Dy6|0h~$ekEYAbJ3X;h z!bBf6_TsUB9gcXLf+PoaH2C+90$^dyKh4j z9$r`aGAJmx)A6cZnnYav_dOC+{x5v?xRBLUC&m5V`N5fvB;AH?ddPt!X|Qum8R>sG z+ZP$)VsfP&M6??($XIqnRq6jyU1ju^9k-+>_?&@28o$mJg&Q-@2S9^yb)@m$V4i3m zD9Q9i8cpWEHk#chz*Y&r{yd7$0Tsl1cVvQTD+P6e$uCxDwXZ?RfyZyPIm_g?+QBrF z^>?cL0wL!tkQ%1bOrg-Q&B?ft8u?e^Z-mwW(){};=MwoxqJ&2?Ed#*#pYm^E4b1uk z`R9u)VgB6;k%IZBj6wb>qnEkCl4UgH5g;{FXT8Gn9<%T=2wJ4<^hPA&&06HhF7h8U z{M&HLWu)w^n5`ba?9hXke%^kaFmRZf|)=$iBs8a!3iqwV5D zP7*!VYM^?(8di{Ik2t9t_)$Ncc@SE=L94chDq`A0deBehq{%>pJ0pQKVY5jOS9}l! zdLC$g!9{zxUg>S9@Np_+v$nVznzftr5w{}0SsO=|lVo92e~Bs1NmZj$oReBar#L64 zbBc3DfoAP-D$9~Azgbh>pjnIPK(j`7(5$^(DoyghkHv#oeu92z#sqytMSVfjV)W1n zi**_I%q%Yc4;!oSX@<`%9A3MmA71JZ!dIH`a^FSe%Qd2WQ-*IX9A3MmhI=CqCdu0f zw!Z|@jP6&k=}8Atxew|`2SyG;t#xDxUrPh>!K4Z7 z@Nv|@LlBVgT!!~69A3MmhI=2JPnD63XugSP7-qQ52rM0;)^;EXuc4yHjiL*0MkAWV zDe>@I4BuHeymm=H{20Rznec9?LFP$($hs5u4kJ9y4iXX#uU*m)ug&nvCfr7oW90v0 zV-xPq@HU0RYnRmUkWM{^S5?9K^asI})`)`#~fD5--% z+27x-K&>+3K7{Sxmk>GvpM4HU89C`4GpwSfPu^}UMNk=104}-FOnzOigm@PaUri95 zHb5F5$%W&iAY`UerW&jrOquCMhUCIBREdxC@8TjvaxVo^rmsq%I}A@@UjG4_)^$!+ z>#$B89Mlh;8t&FC09HQZ{PPt7prVQSAtlG&& zxRVE@_O(~RCH%bK9<)d^Hji}gJf<9bk$;fk*E#g?Xx^r?0yJ%Wqx0v!IJ)oXojQ2w z;v{-oynWThEwwmSY%S#M*Nj45tfOLi#q83X+hDCy3KgHcs)_8u|9*f=33nq9-e3CQ zB5b(qKY%m^oUXT$bn=|cA)M|IiqDc_0GFKZNgxS-!@Q{nqSpId*r69&-L|UKELb4n zJw=eD;RyHBfF)2{earkMP~8xW5Pu2OQh0FdCC*Qq1Oa;)kec#7Bv8rp>a5Lho2#?N z^Q5&DD&J8xU7o=I7Jy4(_eP@IfFx`kfr+2NsV447C>qq*+9jz>!UJgHd4&5WzG7tt zCc>&$tlGX+OYj*WF#%R(0I5~{NDxy+s30q)l+;9cJ%CGY9B2=@R&d#!fFg2|LY4+= zu9W`x9|Ca6B;GoFIhX*KGZRSJ^~*u{3~y`N1e%tN=vM>ZWR#5PolPhi{B#hMjGu5n z_8XA1_Zfw}xJkuWGWND+$)Mu??_|j_{GSH6lyKh#60d2mG_hSMNCi@h_>sc;TBkwD zJgQMhvJh4i;F8l#0+MiLW_uSnHQTqt4pTDrzyb*m#d&|u9EAI6K*=bI9EHa(8HeLy zN=9X#4d{Ifd?y6#XMxm|Oh}lL5rW&43 zoND5EiIxm1!#(S;!6^|kCZc2nCi*4Aw<;fB7b=0rZV04S@iRnB$)JLtJEuxJ{C5Gk zWEGnAfslZIM|!g}15pTL}w7!!pG zB!oj{G>W~O;{=}ilQoKA%U#I>v)uy-C9{s~y3V4buCBaPGni%;5OE6D0E4^zr01r=ouJ(2nFar-^#BBYE8^p&so>UUzEv z3yrotpDL}O*y#_X84|3()j9sd2o2`=&;Nt_Kgl@RFjnLhMB!FmqBu{0rs>J}IidPQ z(CeKqnJXEm62c#F68egJa1iNf--ER})f43{2-rUY=~Ia;?)v0C%=+_%_at;c;UyM# z%`J{tQyAfdC-NmZDND)$3afD$g}aC7Dc;sx{9D=m zk2sO%6TP$!O&5ya6b)bx;pxes^4@y@lP^C{7&sKJmRf}M0*I>qkz>KddCxO%|eyNISx~Sld{q?X;5F-G9OK$NtK&~Xbb|z3n4qMvbeYR$Ese}JU0GCX- zWRteehf{538ANfKAav;ZgHw|eqw5dGqx?nIAE-ww6jtu-&*j+ZqiTtA4|Lg&0%_bV z%hq95mJ~)Z`IAI>9{;ZbTynY(0*PniE6;myIiCWlpngOPm&Q53j$72^6dvXselhbO z=1QCy9C?^uz$4y$kSGBGyAhCTdf5M(q$8AtaGQtu5;!G=%9}q{McU)P55T3cTb?Kf zNWyh~Kn6{RQ!^;_bo60TnRv@!`a%TTD}XdYe!hx%PLzFTN=dxwY|oka`-2qOoiGZ$ zkFOTH6V#(>u{+@>Tz-4e?u3o_wC1KTXMi*#_0wAT4xEMUv+?G^F_{J3cEKF-3BD#SD8bf!Kpd);jw5A zQ5n7^!t_H3w*LUq9OCEAm>i<)@l|;eO3RX>0GHh2ZGl_{ymmHFL{3u3a^VTIfcS3) zaLFW|htHgy;BtBbDZ73g4Nt)N_Rg)KMRMp?L?6kaQ#klOl0$WP7QXir@G=P4FB(lN z{LiUchUUR-awr9fl@uz!`AJ*e#{WKmOJVn6qNjjxK{#{hES#D{Zyt)~5S8Jwc9@<} z3=x(AQq%eUHYSHC8$MNAauA*eaLFy+FvvBB%We-8k&_g%>}t?auEGC6fJ-KH-I#P&;0MjniH?VZF0g!q9ldNchu-C zALIXffJQge5*fry$s_1f&u219ePJPs#j6Lo_HfzVfg*C~=iwcGWhEJk|B(QfOqe9O2RAEw%`yMH5PoMBkUpBTyMqhQ zV?S7sDB)~cz6XpK%|AbJh4}N&;|7!r=AXaJ^hJ(v{`o9KbpE*@qs~7+lkGF^9EhE8 zmzRJMJ#@B8XQ|4;pF9Wa-I70UfKC48CQXti&8wfm{CT;_a5bEV26k|h=54ITa9B@o zg^}VtF(6H@f}qnENb_Nve#Y%j>t`|sUh2K(yngG@`Oyd8Vd~~Ed}ZPA+9fsIt9e*t z9rZqM^!n=oe%+5Gz6)AphWz7Kq0g&)WxVVo|6hi`Ip`=|{^z7%iGV*t?k^D-QZZGE zl!QbTAho>NYi3E-w6YsBF_7;G770H_Q?CG%CS?K)FnwYV2uOGr!-p3RuU%5Zz00tK zeG((O$wcJp(uRTGpmyvsa$F+d_hqf@^hza6U)IX5tsd>mTG^Mm2=V)}j<&zkE_@X6 zTSuTrXS2xQZ%f3UHV4Zfa2G{gTTat}e~bZpXVGx)7z}Sv3WBq16xU@CNa1+lxBv7y z*>RFz(JIYV5B>(CidJQYNVqw~tbr8c>=|jgjq;XrPINyc#qiCD5iAF%<{TFvXscg! zOPEXW;T|GYx0IoK#+98LO4T=#$Rk7Uf?So-h~R%%p7AAi1NzG|KC2#Gp7Ggbx+G)m za!VF}9*!>0_`Ga%dB*3tl%1hVIKC+L6Y_q>%MX>1#rt)s#*8Wz%gN%OY%c$h@YhVm z0GNYR)Y07qx*4DyDE<-(BxHhM=>xMuzvB68COl_VvWx+jGZjcnn7_lI8|d>%Ut;Kw zbLT-gz2 zJ`}o~u|S&aNvl|HZrg*UC<{PCs{93m%alHa3HcawEhWPjmw*Fr6Le@odOQ-7ko!8I zaGHdu$0Vf23P$oAl~pJkC=1?`C+Kb{Rd!$-s+q|ZUcnPlrXHj0suZSV99Eaqgf6EM zkl(m*5k&a?uTXPGg65|Lqr`IPG~n?^iB}>R9{+XmPI$y?|23}vfPlTiXsQhftqqIn zS11Ly>DR`9Q&OmW_uU!t6#lmYTnf8$iS7WB@Xt)=emFJLc45;ll%ajCU6RVgv#`45 zD1z$A2q;OC}WA z<1k|;oLV=|gvUoHh=5XzIq{u|eHXDmui=Gm2uzBE?tx;uC~H;GZm2J2s|H z36Eg-b`wtZCY2tQdjmJ7$ZiBX`+(G1m-O%lHPK5H$~&hN`zT{YKR;eEj#*KQx7IaD8zUs)`(ks$IUS zmzR}}TD96&wSuaCr>f&d)twWgs^~_Bm-#6y;VM8*%W`<+0|$=Ou0NsF%lE`fU67L3 ze~}@{*M9}{==VUaKLlEWbx^*GH0Ii5DPJwV{1;RjoOi;M~i>Nt^AmA&y(#r7ZyS9M78g!(8h?HNQ{vO#wp{Hr|i;U$=Y%K9=^ba8Y zAB;}?8R}}6kI930HJa%yMVl)kOtTivKpUC zZf_N`ZZxRQ`xS!1$)#nflO_(8oPV5>T*s0A)uh(+WO847y;jE@$+_U9cE*~t5_Zdm zV&%lgWn2a6o+UQ5lk7KBWGnQ21h8KRDiV4S*m@VJj5>Y35vxkc_n@5xD4+EcT%ph@ z;6zHu^A`18KMMmOC{F<>pY$vaa#okhQW2zFL$+Cb<@vZ84KrobTa|StxH$t`0~JxV z*6~iLc2e(PLTQ1`MHx%I?$_o^PVKm&Oxw-YY%te5-y~jgO2!q}H7r*QoIG}dFb4wNBN->cB4`tVt0U~H6*rr=Y?^uMp7ZGM5esug20#3@+X%OLF1)f~xqR+BOKHQ3u6 zd?pz_`R7+b?&*^?8Bc1te-@R0D5%cStf1U;>E!=IxxW^f8ZGyIAg$&8-;Eu49{Sz^ z*e?PV3Ec~9y$MuCoj%`VJL<@O&`to9&pHHGDD)$6A|--yU;DaDv7tN!P(Huhw=Rwb z7o^ID?3ep_4$J+TWB%C_mm|TNZUe2P>tWr!-u(f2> z2I>D$vW`*R!UF0djlb57>D2q5dl3|`Gg`cU*^YAb5SUHz^2swv-t3dLc>U_vk_Nvb zEvPd~Zg2I8R9OhB^E8kS=yy&@@<)dxCv#nNGMCW#b?8~F z)q!5!OMM?!u<3Rv^7w0<#cT6Ro1b1U#?R=0-9RYwUMqANAyd9#dQX?+ zU5>LZI*@*~p|5HTrQrvd@f$$7K4!%5of8LWrWcpy)Qpl zirksiMdc~8gV*X8ynKom>%&~ zyyeMPV0KOcX>6~7jC^8NuRC??^kDkE-PxBfb8Erd8yDT{CE0-=EeFRzWg%`lyyT)iL!AaLMWZ1cdGQG;$b-niG6u(t1mWpCwo`+AI&R z%vxfxW>h_zC0H{`kF^@-&0Mr*^d)Lt0IfQ!p(19n6+Pm0y(~@cMY!`2klMQ#S|l9* zH8S=E&@?Z(!Ein4IN$);c^Bi3Zw@A(gZKK7cc%MQNmx*hV-05bE!M9v{~2k z@-Q#6me&`-xsxa;5Yf5Rf{ejhUS;$i!NT|M;B)!|Y5e}e_X?*F=kuVcaXd&%b2Dr<+G?6;OIpzbG)?&%GK`4hKze9kSt2g_QD z$LDJk9txG?bio`Zeh*|cOS$%58+}X{P?3104p{Ov_5Kj6SG>!=x8w}z7YzMT$V&LB z!$?c(+?ceyfpz4UP7BwO7fefPBj!`b25Hg_Dx7{mntx+i(ZaiSA-^Yq7Ad_?DE%>} zVLRxW2EGVsv7kdIsB+O{Mkc7Zpg%G}r5-aubv4^gFa85db~EyQCh`k)N&n%?4Ecrh z;x*B@TCL?n)L}JlJ?Y@C6SJWHZqLSoPjxWPm5y*C5ZAliyoVGwJQy=SK zk%|d?9yu+1Pl1PX7RwyN$CRbt=In)76;yt?2lK!O6HS~Y{h53HY6~(y0uL#j%{RwlfH?4dEGMaPTAz*Tj zim-t4IZHNE?`yGoQQS9M@;>RG8G1(!O~P|tM9!719g}m}pkKOJbGJajV!c&$%%n|| zp+ko;T->kyq&m0~_nQ?crn(Xr!x-HNV1_a3(JaI;Mm=U2!wm)E6~h*Ro1xfw2uO2p z28l%=)v;m98W5N*=Yi#20j5g4Af!7bL0shqQ={)vCyUjeTdOk9>WN3Cx{Fq z)T7yhVML^rmvDzNNYzGY)FQ%R#P{KtVMHVzX9Pq1MGar0nNF>$iIi359T@Dk5K

trX{{;mMd%V`Jdt#;w&ZJfg^bep4+6e z?BKWu2^AS`#9_ydqFl-O8e%;hLp)B2CmN!H=k;cpCuyCKP?2Fx94UQ_hUCLuQhJbjbWi1L=?%8OT#8iwvYg=6^Mi4w?UMAfHm$ZCcQ8k_u}LBr2bj zW@AJ(NpF~pk-_)fp$1Z?c3YzT9xm-8+9&CTr=0%Bww0qf9*=~I#y|$o#6UD+@K}5U zz2m(L3HXz7Oq&hGLa!4CGkYgLm8-8OYPb`LEy@6@_}b zlpuNc<6eYPij0Q1i8JssKx)eVpD%gf& zxaLx9s+VGR_Up3>nAuQw@Cj&GrBdC z0UVD*LPg8q$E@-*2>F-e#LAPnihtwOwla<5El8+n8NFZgTl~{FQu{^|a2@)z=2Di1!Um)bahhvJW%9FT?Z^q5*YmQryP|-5Fb|cfU zdkOe?KUk_>3rSRkpLcpgX+nAnBviB{`f@3BCg}HZJiQc>c#8Ed*hfy_crp?yT1xw1 zX=mX}Y_-{4Pf5+Ka-jEc5cj1Th;Nk|_=j7iMn|{G&rqnhN{#t%w@O9+`>paKy*Man zNpfM$ty1|~J}I3K<{;90Cu3xAo0|*d7_X$}2}Jv6xwL7iZKRyrPirFMIbMZ?ik8wf zaA)8v`O&t~c4)SZI#zTi1HFzF-H8VNv5h*~?j#dsQ5$vqkKKu?{P#B6MPUy@_rT=B z+BS;HC#4O!3WkwBIvFE_@1Cx0RIBEAqWu{z?X=W3QqEA`oVc6g14yW78T{n61#-04 z+k1wP|0a$x`&SQ2xuYunzS9qrk2!vUgo>8Io#5*AxSNolz0Goqj!dy%#ec9hU+RHzT2kmvRui`%&URyG@v%(QEXc_%%nW<_6V*Iu^hPot`s_dm^EtCGk<$Dsl`y ze*})OC;EfVcI}4ZNEO`wMUk`|WNil~u-hXLXjP9G4@Z zqGfP%$XeOA2>DxZ4EOBvl*%gpnLCSQC&ynAgO<^=wwcOi?||9>$58vFQWc&vzg*go z-T?^}Er~b3WndMbe+-V{B~p2c9d>L-IgR6)NT_K3e@q)&Pxj3TFDDdSk7Kw!Fn_Nc z2f&iHb!16)I8OmjIzTaLW_WJX((K0V6gZ-8yE?K7;LBkA_mY900puz9$0|MLYmPe+ zgACunCyCAb)sesP`PFvXB7|Cmglf)H@K<-%l*XhLBcUS0YB&JW6QQF_d*2KSV-B%ZQG)jlL)3@5M1(ZOkOD;*V;ZFV%ix z3;+og%~Ubxp(#PXGmhbD6D9E!yK!1S>BX@e2^B4gH)x^3`1~EgGCe<$fi5J&We-4N8Mc@$9eU_y9xO# zaE$t$$zv>;>+bC?f9i-#o@(*gOzdrr8xcbbpTH-BS9k@|-K$aa2gUq_V>FH7-IJt7 zRZ_!)n@RPbdH(?k6)mODUR5AX3BHLKT6eDSot_Q>^!Ek@=DFoXPX_@^P@fE5^mG`& z1XKFuX}*pFnE2@7Wj;Em`Da8?|N8XuqNhUvhW_t;^LG?`BkF%jE_S7IN@=ah1@bqD zoZU&4qs|ixq%KjbTy%}YU7gFkQC?D=O|YWQCio9^Ht|uNZ$YQl*+l)vIvec2*LeoT zehc+qCl@=av(XSewT%F^3&h@Jl$3rwp+Ktq!pE}_bDpR6si$DY@9L&S;40d^4zTK)F7w{4_!pu9)ni8VG>}r;Xhct=qsop%nLc4_%)i#vjOZ!y z-)njug`JP?r;-b6p0Jf4)pHn_7d8K6jFe_xQXrd%y7YHj&#!vL^_-Vn&w0u9ocGW5 zoM-EK6w0D{&il{xocAB<`O(D%ay7c|OfIagXH-5(Y#38d7Lo2h??B_JQPK@u$&XxE zCTlsqkA#ZmgroZld_N-Q!q$dhyX}h#skUFOsTzcnJmx6%D(7Rm*Ncq+g0-+Hml`DQsoD*dtH+aU6hzie`}ENb_Vp zoPd8Jj-l2%IEAY4dj__YsT^-WLPayk(5DK^>EJEKF}w{2rSKG6J*+P)IlhF1ik8x9 zCl<)N_|E;)(z40UT;4}|msFfq4aAqX2L9pl*68T+?u9Zvts3*+E^kHt`{jKgg`I%z znaPDUr&amU<$XGsdq`iDjFG_uzO63@dZ*W1Nwoj2OY7bwrHzym?k$(kIex2tMa$?t zJORP-r=b{#eSMuD93Uff`p1@ z(DpMeZjG8_2>BCnqARg+N^uoG{g^ztiQ{cZsAwq-k1vn~_>TO`(pvF{GI|0l_wGsV zrA$zpEqVeo!ITplJ%O3{=*%`jr}k1N>R(T7^aN(G|9)~mKCVE9p#Fm7Vn@A{(GZ>9 zCqPXCaeXpMO4lD%AhU@&PICP+tnEjQJt>l=l2Ht&^)8Sbi0UG3BJ+5Ii7t1Z2dCDVU~-)a{$rhq zkLt`0wYJVg{pUJ^{nt9z&%Stcj#Qta?*^PmF6P~U+O-Ts2~}AByjfGe}jPk z365dh3L~j|RQS`&nNj7q8wnLHr5DobzSl+?Acm%`?oJsEU{2ZSzPVMT@!CWiKz=ue zY{!gbQUYP#><$QSQe)1{DuaHpDGI+|q1=5$5CBG8# ztN2#$wDOcuCdF?Z(o!0b-UJC1ErT=R>L^o3qWoTt`)bFOd=)?L^)_-G$CD6)meD=e zvA5jF+_Q=CC*hdt+?PsK_*Fae{#!ouf2gJXCjQhAEqzPO9L&T%ag zDq2QA*n_R7$5;D|fd4a&q2AaurL+n^U|GIo25lr42^B4aKl!u1bo2JFSxA&$f@8P^ zssET2KX_?9>A|r#5-M6spKi>|6TX>kd8aG%{1cpLhVh9g=KYFu|{ zT|24WE8k9Px7=0`O=_zLZMUH99yBd-+TKQ+3|@9$G&r!5Xn&JSn>s0J0Y=i`fRei% z-%ozv_zMy$EnJHur2`5JB*@^hLQHW#Y)^6Z9fdyLUyW9tDss-dPaL9wwtaY%dePIj z@cVe;<>`GyO54Fd6z<>@enk1Bp;CAl2Q!lN2@ccOHAx?Dx@i(A-O{i?t|xt_!)Q6A z^l_r^A!>(Ls>m z&z;SOZFOaVdaw`upgJb$WzvMTaZK&MAlIaeTC-g}qp9pmdTS(9H1nv-cwY~nUyfsV z3sQNCoi?*Oi;mjJsYs}3Nz9$eHZ1u3%Ww?uFJ?3O{sX}#H@A@+INpqe3gX)pH}iTX zKK~&c1ApJZ6|$mjwk+qk8VMC7qjP~{@aqWpU*QCdzZANrEqQ<*uLset!=^b8S zj6J=+RUW?18DmecU)@V6#@zfl78K1uq$27sYsXWEDSX?CEbKHT1ik8x*bqZuJzNXohwwJ<9@%4?X z!@Mm`b#bC;M*`in9XTVPP3DZeuV;8OuxrpN=BvfR0GL9q)R%gqN`20I@OhvjVvCI$^{9I0%Jj(G|BvfQL8%IjV*DjD( z@O^?9-l1v(ddoe`n{Q9)@nC)+eYeACZKbrBXg{Zo^hYYFdP)!TZmu|`hk3VDd18_~ zrH6U9*1bdT*t&A*s(SX6`n&C|TRoi8Ay=7GItZ#Fe2TX4wQ#f9ago>6# zhw*HYhRObniM4x|%BdXBMna{9 zeQ;O`voO*$k&r(Z#}xN}X;OCC@42#=pH6jYpA@c*AfbD!7Mr+dA zc|{9|UY49>7S}!aF3SW*&tIRd-Po%n`6 zDyiD*ePtcT4-tcw#I*Oavi4tbzk@AbKUDX=RV7G&fxz6_tY8Rw)recu10fiA~3aRXF!0kMRPtRFz}-Wu8_&=I>`6L#}H3J z5-cE91+;>e7e!a>P@(S3n?X ztMAO_|InzkyZ4NADLqLNHoFa%uzn4dznh(MIHPMhnG3*qkifKTJeiWNq|Wd11PJjn zh^HX-Hr>j}%V^GC0s|kRzW$zruXM!?`--m*690gBiu867@J+FLS@It=$4!qCms9&F zQNew-*Oap4RYp8TdjBYK*o>p)WdXL?oWlo-O zUC3v(xmhqq8Cj_-`kUk!64~XBbBOex{Kcnvp7%(f=##&y@XJZMu3NUGHTf_@+#P@t z`peFo3R0%4Tu!c2`X>2odO|}rg;D`Z+Lekt<|CgfMuwtDHHAL!s>u0@Of7AE-iKKc zlM))L8v3FGh@_+&buRYRxtOQ8eqwO`4O(SN+Gce!$afM`FPI-)QCwWqi76)Pq!GAJ`;c4OSGEu( zu^;)2&Ugjb&ll>NzF*v{5 zl}Tj|y;@e%zR|Wc0itbb=S8AznWyw?lzxd%H+AsoiiW;TAxjl9$%ia2ZmyDm{HTz9 z3R&Vq))zyRB;oE|vSlrryMe%L?R$Vocm{Rbq4s^&{(*LK+CN&|udzkDzE?%l^%ef4 zpwn7VLRr(==GrTMbrnrrx+|nEfm_>$v~Cl0y^Zy zJ@l8#(M-BtRonGbw4?!A(gMFEgYzq+Oe)iD6J4L3Pg|bB5bgSCTh=Q54yC{A(@hUP{<1k`NfAM6*pH&K)lmHzE?m$0 zR%-8T?X9(w)AF^tmt%`|{qwn|>zipoWeMCF)T2(2~NLrZNWS_nk7S%*H3_ z`d-?WIzXabA8pG{rLR44V=|?$xqh^?7#(V;|J`YVazUsbh|r`0m>m=$X?yih-Nt&UhjONuE6*QO>nvX6 zPEJF5fF6gK%IaB&fBn8$sGKuEpYRR-6C*dg%lk8K20HC>3hm~n@3ranetNl0*Q384 zC2_+6SR1Xi8`hVsb+5Iy$NH|d95qG{_$^b-^auZ!P@L(c)z)|c4O^>L z?SvPLt!W9@(y)Iqm~sCC#c$Yss{Ow^V5KkH2Yyt_-T$3ZOpgA8Bn1c1Df?1IlgsN32PeRg_)by|B@;&h>#8Za%IsqR6 zUYR5RMRTi^k2?4!T`vILB>ADW?h%p zez4+iIUB#s?Hk~iMe!rR->dlbXXBT;Jp=qwAD=cx_f99lmL7J!d{(Vw^V6F|It}3f zUA<(gF0#$?QUPXpIphTvoO=0J72N?TJiMEPn@gdo1f~~XNMrF5z5{!%+H1z!*I>_A zd+Rv+o`ji9pHbtl%a+g9xZWjXIzud?IU_IW9Y{V`R9CJgz6miPiQEC@R+9H1ru<%t z&PnPtD@oFZ=@M%LIcA~OXlC&eOLd7n^+1knCws3Y5An&3RymS3Tgj&^xhu`G>o?UY z;y~FFlJa{RF|}q;apZX3GhQ_$k^Sbwv_G3bImew=U2e{i4?^2#;*PY zU76ihCoKNHNHw$D%8+Emr6hedGe?rUp++XbABHjuF}{1F1pNJ>>#wIy{j$DU&lyx? z1>kw)ud!e)Z2b6pGKu(B#8aeqC4q$BtzJSFpgAiDOm`K10P!S&q$~L2kNAWi@m9oBr1vm^L|&vK zZ=!vG1l=s78Zy+^UKc9)J!a<*o41nTjhRHs$;wESj$@Ptp|_f1x?0 zsziA=$M=e2&-+hBsZ4%N#8b3H&ZAeFkZfgXx~)j0B<8M|`dV`%WA1oYjts>#)|$ua z{we%K6IUi-a%Ne3o0U4k3fp&UqTENk81WP&;e_-Ac^u7oj=;3BjDaiUof8T28u1>) zQ;>`rk(MC!4}?m6M#}k-z)*@-#O*m0zpuF?|0PbS3Rlq*8QVEWN|Ve*46TyR*+@Oi zbucxFGZ&WG;OU1rJ79JC+pJsujw?thGq0{KqlvFUJVo=GjJ2LjMRR5on39wy=?ZB) zzO6h+ycF>i#23jQ_T))4XCr}uw=r;q4CA)+ZsL82ryv>i{e%#=vc2$^q?}U(hH~Vu zxSA=xS~g3kYVcGPE#d9ghNK)xr!qmnlXQjjYh8~|Jae-c@f0MZo*Yk6d>Ok7Dd#c* zLzx+0l;W>g-%+k6o`86YmT)7ux1c$52m+p@E98S)YseDfm58SxCfV2=c@}L8V$e^W z(!u%n-7MLS=^bm{Z_N9~5B4^ur)C*US9`(iz&Ws)#yR}d_i-Q4m zJxM#Es3sOQeV`kllnN>z)UeW%(&bVM`n6_>0HjQbd^{pYMxc#DOs#J@p|$=E^HfZ; zk>1 zx-|#h4(j=vIc4~mJaa&PgfaSpggIq6UTjU9r;Gjcn$g96B1)<{6tl!vIu)o*NfpV- zLxR3D^=^yB_o&SbI@v)-Eu_d*)w3CA%E97Fx4{DHBSM9d5`?h zEyi!%6_}4;`Z>;=lXUn}R=`c!e%?#BGww z{b6yMbR+JKc#4+r`uSNh7|pqk!0@h)=P9;%{&_Ns_)f%Av<$kdUyfvk#+7-Pl=BRM z;ePc|T$zgRcUgJaM!XC06wP#bsR}FqOSMHs7|dzI<;(_u<=?G^LP3=u8JHsvw&etvF}u>0eU{ z(z)KktI$3v+fq*R7XE%a^)$OsPA7-TCA^dCpJr<*<}~}OzY_h^Y*qM~eU}z(_Gnp> ztQ9u9xoOr|jTt#ht$JZ+VX-ysYu(L#g~9CRUPbY*WxnD}l=~v5Y_nM0;iZP=-;*ID<6;(QY;f@=fAxWoq zmc67LACcKnr-2NvUth)`&NYJ=i|Yx*6fWXHZUondUod z=`?SnjWaaHIJyD6SJvd{<2_f{i9P4X=)_hfI-%OoIc7pt_b1pzHeH?xb_oeH!A4qZ zb*y?_Gi{{~_h!synk_;}ZNc)upIkaBsp>gqV)r7~pFdheNIga;cG}PCE45+N*U^A% zu6mn^YP2Ehzj#YA9?a%qHWH(a*FyXWxzXL%jA`<8Y9$Ly)N_P5?TJ@F1HI~z0uv+0za3Q9fglAaelH33*__w!!0-(!sozj zeLSgP^MXopdQh(VDKKSSR^|RXd0kf8Rc+<;}>y>8)+3Yupd1h$M4#DRQt3Nkez6M=!{Eh5?lN#wG!v%T~=yXyy z9i7SdmhCL3%OvR~&dKRtv%jv9rn7Wb=mS{nwdHh1Dczgv=j2G#@)I(wJC0lCalKR> zpS&$(JeR(5!e~u&)IrQ2+A<_3=Y*2Qmbzw9C%KwRV zO(gQ>HCeJ0)0#MQxJ6->Y*hHRc>CvLv*azce;jXrZcLVZr}h)p-dx>>ziyr-f2+Mj z{WSh3?oUZ|jWno~6yrk+ed#MfVf1 z<_NGB#oy(hvDY?-0b>jHX8zvhT>I~0Td+%u9;XiD-yElM{NvQ`Hr?WQ*^gx&a$s&^`;-hWM9>>)Eg6ZM-%Q2^LbeA)>x{9)3EPV`-k!NpT=d$ zH)=l~Z~qYcX|*R5#8g4TFJLc&$*mA?UxdA$+MCAPo3it2kIA{j+WW@O7RCQCy0nZU zz8diqErlmOFdNWx63%=A(+jumi<@W)zvr>4vXc05#8b2s-uiB$bi1~p*e1|DDOCzabo7SVr#dmBW1}ik9dlfVO8(RmX##U z5$V(*|A^FkTuV7Ms8b%5o;;!@`sL<8F)MTv$VH*+N8TW(t|I>U(&<=hT7kYx7WmeQ0+(K?4z1(N;G!` z{=)2(Y995~lbXhKg>o8}&zH*N*F-!;%WVl?F3gczwlf#`J!j5vgjdZL4Xw}Xa(`Nz zEcxr4^cNqj%iop)!Y3C4aqRCic1#De@5Va_Cs6?vnqUZ{6>u$ScVGp`n+G zY`U;MpO9NwJuLYfo$w=dGI>H`(TsYVHr1D+?WSA_$-4l_ZvH1#b9a(;Yf=gi)bquI zk)PBGaD1=|T*LLb?li1Vki7$DGR{QL!`>RF(mg~q*kvhX(g~1AZ%1py=P-cd>Mi2;l0&}^8s!p>vKS# zB7HQc-EJ(Bk|O`bQf=&>8bKi?0xdekQjyOjxRZd?1YaGgwtrIf0ObW*fFBd8;C>FI z8`v(0{+iaLz;4mb3(5Lu?c6??$Ah^BaXXQ?A7!B?T9J4WWiv8ZJH)b`Knr^Dp_(6S zGqpdy0{#UeJ@>v|BS0p|AuaKtny$Q2LtOsU7((1%fYd>$rHLXE7o+q>%m~;5+3N(O z?iX3;BQ2v%C^6XOZ3i3ny#exfGvgau#%d#qe9kyvZN0bQSa_8m|q95 z3vsU{aSY|CCPtGe7fz83B(7sp0zbB6Dt%V#q^)KIcnd@Z?~W-kd@jyYEve=ApoW0I z0&%|}aU04kq*^v}!PYo6(1J3r2-R|109gz6aV0k*@ixj^NZu@+7}4%56=(tZ((3Li z5WfKZNx6$jWT&P`Cc7%o3aPT?x0-?5P@qM>8&T4|z;1vrK}pr7#gvo~kV-^ZT9P{l$YQ_`A?|%7 zwxMiQHZ?uXu5xs=rlCON@d@*z*E%L#K%q&~YRR`yzD6Dml_zCM7TPWsV^Ug*lt8Mb zu{X5{C(xo^O|Q8XfiwVn9>TBQ652JN>wEfhy5b2lbC{X z6VmgC7?Nv975_?h|ATci*&7s6?qIRs1Rt`{bDLxR580=XrdP*6I6mwAe8^36r(kZ( zI}I8jgZIW@N(2@4s|)^}ib-121E8)3J{GC=XAH$Vo6;|o=6Z+txD42Z2ph_h7zY2p z)QY@Q(Orvm8-SM#2%MSgYomQ7Vm145MeVmyZK;jU^*x&A!h^@y44pP+oCY-X4)Pqqa!>|LE?xwMi~YWk$P`$tS|ectPuA|eAdMsV0}+kA3zr(?iVE9LD`Lr)}kzlPrX8FFsNRM zcF~gBW~9gkh$(3#%5X#mpL(IVBqyNTPH8ZxZbTipo4Ire{!ZnLelmtrLT>R3+T9-9 z4stewc?NN}kSO3?<>w)B<#=b6({ON&RkoZpz#r9DoDYs|P;$W0|Bvdg;ozFz+Q#hz z|0ZG@cM9c%aqtyA@^R4lR`cmN7*ZxnD}5YP#c~{45gELORO#0(#%d*_1_~XboQzUnL$p!WHk;b%khHW_(XU5nDf(3!Q`>{VM6sIv2oK|*s&htbyMdZMr?ovCh1=TdSAp`h)wUU- zK#TEXd>nxPvHGZSMvK)4Uo^&v^&Sc|>qO`M5?h;p@bZ^>I(bHIrdvlgg9~&~NNFvE zpPmPNZynXjWFLaG{4?H>oP~A1_*W5sr}c8QE+WO6Pt%>&8YSsY>jwzm8Ita_ zEYu0Kpd@x$pDRgsS|zhmBpLB{T9%X$knDyN9H8?6H$==%>k^c{%BH4zfZi4Kn$=7m z*?NF}OnXfa&{LpHLi_{tnNV|peuV60h&e!8gcB4Hw1r+X2k2d3UqkrS(*;)F-998F z+jU3Cfo5lCO4l(v3hXfQv{_VXwi%~V*(ux#Wt?I}}LE`e*C}bkpr`PnEA{ZcKxsdn)}VS3h^biWz*G%Uv*chV-{9<(I#=#S4~!*i6m=efxL1;xi!ujkI+VqbC57Y=B=^lwXLkYS9ptZ5%-BC_1{f|F za6=>XN2rH;C+3&S^U*k@-qSc;7Zl!BG5bU1f9Xo$JIQ)1H$_$>?n@;0qr8dKyMKEz zDJ0iIa-R&fxK~N5$V#jL0U3StjbfZ5E7bX{u4NP^d-$g^kU9$e9awLaUP#k=wZztxb z$X~03mq@&a@(#kU|6|bts&WV|$-L0Xl~hIE4!s@CP=h&^Nj<&9^-WNr^eY&f$3y8a zcc;_;ChIgVe?ywqVWwDING#56hPi1a&MT3oxxd#8FiaqPg{}xCx^*#+BY!MXuNMP; zJScCdVv<7j94&7ESx!hcMn(7s(_}qx^w%sZ2teHm7=d!Cq*- z*1Sxpf!hvqjVeqH#MIhi+yHGa9)Nk#JXbk4%v zKnvi}Q@f@>n;_Nn)Xu^a0?opM)4YLTFGb90p2a5y_~XiVdK`*Jj+N5w4hSaQ@%ZbUpE9e=9I5rFhQ&qobu7n1A9?3Jre&2 z=@{a#p3a1tqjH&QDUyzu)ssbp#A>$kx_Vlt!>32P=HLqu|4=?Ws7tZt({xSqos#rW zemR6ehNOpb3v~i5D2dfmjy6CKR!?gHKZ=;c_&$_3mCc%E z=>xT7OthwsY-fAO-{aK0<{}E(P1Y%3CzQ99M7ipmcOX?Jb7f*&&^)Wz;o%Kt?RIKS zmjjD{Z-GeDBd2Q0w19LiOuiK%PYW6JLu< zkarEP-oJ{|Q|u#PzC+BpVYON*QVH=-Vjm7_ls?3qEsxR)=}GJeFar>C5_=luIAYGQ zEr-iGa5x{ls&}-Y2YDE6Az})87v(L)Z-pf>9So_)d?jfsQtEK^3X#!zvTAXjoKoD7 zGFudf=E-Tg{#ma)SD-t!--E6e)?cTo;Ag z+TKiU%?YY)Mo6q?Ta6#n;wJF()W@G^v{-%k>t*XbBv!MP(HYt5;%MJpikC~&)8;d3 zGaX0uPH=R$86l}M)L&|D=nMI8vTlYr1!+2uC8EVS24@cZ%P?;teXZWY zUfei9S0jBi$J~p%kJJLh+>5i!5czZAIOR%3K`m3h* z;)X^$FBE72hr5R6zMRtlJYw$4)kdkQ3B51Z1Em{cZa-O1PT#kc;&yDX zGqIYzA*VZ*W&^3EE^e=fbsgd_ZqJ08#qA=pA4JUJ)*?b;HCq>5+|Jegl`d{y0sk`M zFK%xL>Qb!vG+lD;P?9cgzkzVbkaTftp-!L$C9$~uO-Z`A^%|#0BH}M@Eh!-&owc~F z54av;7Poy+E>bo%)k}j@g1v~<_)m-Z(Q93|Yp?00!JDAmfcTdN&xD#wgGjF?M< z7U2X1M7B2j8i8+s_~Vzf+k%QU-=gC;Q%O30{UBUyNIHHN>I9l~0x9e{=H$eZ5(1JvbO$FV%Ki8Xj@&n?J z-U`W1^n8FTb49b<8dWb_wEN z+SP0`PSY?=LCmGy+<Jr7pr*RQtPwQNY89d! zy|8&$3)R);ASeTnmW`r#U3L1`O>v~xvDyZXtY$Y8y{2{L73v7s44h0ys_8W?i{$c1 zlpV`4*S8)8yGjf09!2PNuHX`yxjJpv;UTeFa7~%j_LEw&o~gWvmp9Z?=p5ISZ%?q- z*W8Ix;s6}Kar#|&y?fS&OLmn*7bi}aq<_mxe|tSauO9yv1)G~}dS$P_T{Y?T=?Yv& zR##_TpKfV)+x7}{z?Uk^c=&Anc7Ggrq$#Soc3_@&GM3`jeG1qKT67%A`_ISRe~Tg| zM)OGxH-=q@PN-UtBBc?wjfD_)MhT9HKK!=9se4)-uORDtP#+;p^$MQFQ3Y_J6`|H< z;l8{XdjQ1Kg?k^(@>lJ6Pu7U zs&re&V|vje5`7ibf&tm*WGrHCy90nNhqx5UE5+! z0I0d%y{f%bA-@*lDOw^^=jTX1$elGLo%R1x626@Le&o4>V*wH_hy7}`-xOyb zHRuWdVbnV@J4qzq9t9;gTHp=FXGTJYYu;fyEKwcSAxJ@Zq2KCsx={0#p}U)U(Ca-389ESrJZ}* z#FCoRKj-co|B-F;7wmL(*}i@?8AN;?;;B=f74dh7r)UXZza&w9Msxlq zFuYWHTGACVe`Fg;9bR8D5Klo8p0+4as*-fhBQWry-3nL8)LqHainu-EDM$*xdB{A~ z-Is(jioh`5fD!XtvBJxKlP?pArz4)CWmN6X^hJhKavmvX6@lT_i{~o7W>%JLAbt_? z6fNOFo%sI;n)4n(z~iqUB%gb{mwZEf1o0Hat0y`n%5P|njEKTL(~_bV$MQ5 z1@Wd3;5A7)En?v@EmG{JhxwoZ@kNNIXes=74IA10=|f34V+l;fK42S@bVVgE%a^Id zGZ9Zgyf)|Yg}gRL4Ur_Y#VD&0L%c%~NlEqR<#2XJ<_n~^+f1GH{CODjeoUWOv*CoN zWBwkKn?5q?OW_4(A|I=)^SHpw@M9O~1ULUYFk5orH_YnBrIM;bbDCQ2#Q15ZJ}a!@ z{RhO|5l_*Q#!J1#9g14<2FY*KVt~HkHPU2mj$DK7G-BFxXd7QIc&JiIsTG>ubo>C` zvJl!%o_WVktpk++d>9wIQ* z`Mct%3V+~(X7U8_vxui?34b;}QMRHv?+_T?QvtFuhfmGWhi=CX?T3B*$n z|Au#5-EF=|!g-6pG=30_mijbDCGoe2r)UY^*Dg_hLUR(Xit?(&^AvmCvi4Gr z{9MFSv=m;+BLxR@%hbo>w6WA--QtQ=`0R_)r8jYZ#8Wg~(5)?9P+~G&Fb#w0f~_|c zcY*3sxTCR|2={<+RuGsrxi7~Rp!5NsR+Q(6HzS^+C0wdOqU=U<4iFgLoBT;KrY{tG zyw3%4l=vj#DOw^c-r^5{XeGu(X*nFodF=@13{3T6&7~(?b-f(lkX*Tv!MwLGOPZ74 z&SLV6`BBWhFpZ2g`_$!@>VTO*{xpkeXQ&_T&62w?t%x=I)VnP;AIvl4Z?>3=4Ryhr zS+Wb$+tz%GF<;p=OFqZ+vo%kkS8_kE+Kj$PZgj`>*|c>Yn19Jn9vkff#Yki`=1fej zV$FVK8u^v!lz;L*?z-nCm%6pC?m`M*t!KusCk$t(rA?0?2X%K#-@0-g@kGQ^w1ijH zOqAJZ&LRRceiy{^6kBsl6In~V4)GK%ksaS;%2u?m5JP+KMccqrNv45A#>w%Yl1jH+ zq%*AQoo1^!3EFWAqm=t*#?2IU^>WJ^l0kkB;wf6f<8F1OHc2O9d4uD5img8JeCbZy z2k{gwk(1PLFxpi}P(z7ieVZvaV!AulJcJj)9e>G-z6h=jkVRy#iiJpIH0E`feu_2w z1M{S3=4NzYp7Kh4Q8kfwHrzP2r5u0S%g{c^uhqJ|9Dl}7=%8g+H-qL}P1DK|m^ywz zFUD*m#0`c{OXrM2Vm|JqZkNSv*g%<41Rw!NemFPtg*ovL{QfLz{{i z+OuY|{_;ka+==OtShL@-loHX6?j+Qw$$!COUNzL$Fz>|lS*+Qo{*h>@t+M1W`9E3A z7fKDki}_znZt^v@@y6v7*fTLzi8oKh+yGO@ShHW!L0i)E?X#p0`IlMDVdHT&=BqHx zh&B7vy_UMZW0uS%|3QoS*-&rAyc*L>v1XsT#ZtfS#2?_uf6rq6G1QHizryrutl6io zvD7~FyfdyolRVl5Nw?`x<>Z~X= zC22ty-UOnENE@_UQSPzu>rj~Lm>))aF4pXO8D}f-Y|AX!M*bm-sj|WD0fr|Adw>y% z{vIIG9Qcnwe~*PrBn5NAcy{=RnUd{xq)z#(SGJMkeG^N)vY)eEDg5*81U&+` zgyHnEwDX|FY&hy}=YIY?L_8Al6fNN|-b|3|(VW`}OmWAbkK-x!vN~<#A>w6-r)Y_M z)h)8DaXzkw0km5*P@cwEbrM~wTH zf&2Q_{R}V-$ZsBtk;uiEJ7T&l*6h2lY28<9%ls#Qg2i+;Q{|^NSuz9DgRy3xnqjGz zUcme(|0Rp*6Vxol*6c`Y=0EwzVlloSNvw4X|H#eF+U&}4Nj!fp35jfn@CRV$x@dFN zDIZ(IoQA1Ztl9T-Iw86*$!2@ZC%>D;TprZtr-bNnb!~?%xs?2=u^8Xa;e?p_98QSV zXC{O@0WXXt@bXd2k6?Ns*6jQF+SccG`fw-tpIOZ4pgsp|eX_tDCcor_Xmb_g`*}OI zK5yIloTSsU0auG9NaQuljWBhNHT!<{MC+51^dNhkWJN?epbbX3(ZYM6_;dCZTea)J z%)&f379){auV=|(OslPVf*GJoFt5k-My%Pdz+&4W6+>C_A^E>q%m>fey}}&7R9zR` zo#5{kM$PG#CCB%>WsHkOk5=4FdHm$SNC%*~eDO%Fi{O#Y}r30z1fjQEh z8uTYQ6fxek23&iE4bxlk<`Pbpe2VGYIP<9VSu6s%%JM5_ zC+Wthr$O;GP)?o46D5cI%7~|E4v;sy(txDX*7BOi^AtPnR*^o$mm;2`rO;WMpzq4K zii9(jz*PQ)_!oO9+%0S%bBPxso}#%R((<=VNmfH-)vuZIB+5p_cvz%}$PUaq&<@0! zbIbozO^#W>VrN-V1|xiu6trNO;!D z1gVV8X-HrS{c0V=ZwVw_!L#>ul{UoP5Koa_9RdlD&(D$pXwG;7LtG2TOyh*1PKAxe zfs#6ut0uP)-;H?6@UA3~@I`rS{%FoK1jfO&aQu($+-TU!sQjODvW@sn#8bGpia-k2 z9LbcKC6hiQ;T$C}4w|t74w+Mdbj4S_GFeV*-b84M^u8sK$O$^M47Li0;racX$w1V_ zlpkjnlb$2t z>>@CQuA$JFMk?p2nU&=u;?EIJ(UPv_Qa>fiapIo?bL1kb^e@TG$+1l^ThGg`zx5Qh zUSWo~HV`LbajoLJR{68KRh90~)f($`(CJ>m~I`KTk{=TM?+)8{G;whRJ zk5a7%(Vj+vo;4?kFGM>tGO}Toyh!#J7V=@x`Kw|&f3@xWj^K;HI-x01pU*sNJAZ|n z>36<5%RW{^%7Ceec*^j8A&|l{H<`|Fh}mgpeH@JMeB~tFn;{ny4?sLcOS+mbT9qJU zh{p%!NExa$jpRDSv{IYkAL<@(V-`sdxMq>mDgV!1o#gmpw^WVAdfbbA-ylo2!u=qY z@9+8+*uk6guo=AXfH=Qf+!k6MGi8+TzMLpYQ<)=(r)UWuzMtJTNv9Ekso&&yo?>$s zmy@={oe@vb5_yGc^+g+l1kEyoH`fl{-}STPCbAb>NK+jPb7|&=n8CZz4&KM$9|!B~ zvxJwQVRxz@>O4Nq-*R&GO=%>m`Xi~HAFtb3z92r1cuM!yYw!wzMAqW+PqfO@qP5g2 z_?vUj$q!8>63VKF6Qqr zorpJ|IFA>gV=6N}#>gZCZKM=eX)v^yqSg|o}#(Q zb|=Swl1>u>!{zaRN7+R;h?YU~rnE74H&hNLu-vQQH zZV5do$3HSuw?_tbw!LK|Sx5X5;wi(MPauWYOf);EeI%T(2uz`=v*V5o%4s{khnyt- z74Z};>1y77)R7W1xCVmw=EzB^ltVHPDQ2FLRh zd+)-0nM^zb@f6L;HKR2b?LoxQ%wR=|Ft5h+Oq@CVds3FXjLF???Z2?8%wf(B)$sq6fNPR5*+@}+*=3?&#(HAiCJ>5+80^-acxpcQhPFz)fgF> z!2Cyf-y)Wy6qqCDUbE!nZDOoB_eC$GF9d!N#DC-9{`$@=@65j0e8?rT4Q$Cm|rKH0Lz}Q^#KMJjJg4G|aaX>dU8yr)WB`DHTsv zL*x}VOMXK63o#xJDkAb7=A_&B=qVDI3rt5%YiiCS3qPu3>fH#0)84{z;>%b1!PmP< zU*ds?r)UZPUNb>Pqd8LueD{yVwO6sXXYz&-;suDOXo;j#?d51sAVIApZMq%_4`DM$ z!hI;Im-4l&RQ;x#(flLfCVho#i_)CSs75vV+>|LNKs%w?QGYe2##KXoHs8@e%8;Lp zc#4*EHD6IKU+Q9N6qqB&xTe{PWM{0>c zWr^IWVD7cJzb8kZKA5f5Eu5BR`uh>Go#!oP(W^A2FLRh`^@K!d1`=Wil{1wFnHiUHrW+#h;i}O_~$8Mm$Ax zm1Cy4z1E9_b0vYHR*I)8{LZUe$pqp_h^J^y_NTdWJDRhY!0^7iC9au@Z5YXuCyCc1 zo}y(~$9FPi)g2ses;p=wMb+He-Xp9U@*k`$Mb-U8e$T{oDXQgGy7f<2%B}C!QbI)y z+>}L-%uA@Zjc6tR;wuG%({3R1$$fEM`bwd4mFqKOOAk}EK-K7X zb@c`6{(A9w@Mis(&q{F%3A0k%Q*2G!s_!t*zB9UgoFXSxyUAZE1|LK`*T??Ux4(D{ zzKY1hEHCNyHpu#Gi*inX<|WAI^f0VCBX!wbjMQ_Nh!iEd^->>Zq&8NqB1I+KDlvnT z>?Rd}GlTO(N@ZC}VUHtDB^{g^!Qga#vYTwdz9-15r-L({WOQ)yrJH<4@{mRNgR^jF zXZZ=!-_{&EI0H9TF@uvv75u@;MoZ3%4o!1M#3U*D!-;K6bGW80U1&>d5~eLZi>+zxwJpmqn6})AlBxxowxk~9RoVZuRqETc zDs)ocs2y!zQ*GZt>u>k}@@HClP7lM5z0H-qyKS4l_S-yUJnMy`@15*PWBK}l-_*ml zsl)a{*{&2bZ}|g-nYYb`_%6*PX2VCa&Ls z`gxdq)5o7UW+yZ&v>t!?%@a#V(Hl-hb#C5{s=myVTGfkO|4A)XI_3ZIQEMsM`Cw~%z{%CJ^(n58-_4Z5(NnXf#rE`1@a!BTFgw&+Z>J}3y^$d? zZy!#7os2GHtI6NQ;kkMMK=BS~@I8S)VJnbf2D!1Hu5Z(2DXR$yGwAKDwJ}!x1WwvK z9r{I>&Cp+qlB#8->ad=3N!T8&*4$kGz4E;GhMDAi7 z55p^YPqaS+zMyLHh{Asq+kdacf*{RVeTVECv84e*E@H9)9ID;LQAuMys zDU>oh9&nvP!cI!&eMStotclIXL09hXLj`RCoy zVOAyD-*dG5mcN9OTKwq?t()1WpKI4p{0$8CYWC^u$zRi+ ztZ{GDUkb`O{h6LTr-#UebWCG-s}VD^v}0z4#*oavEP;LATqpB$uC$v;-DFv`wZD84 zZLKQWFFO6_$4vh!cKZK%G>TLuI{g>;)1QtBKRwWu=P1kBZp&}P5-!w~K+?5NnYs1k zJ>oAAPm$hB1QO2g=SmTp^Cy9+_^BxnetTg z8xRmN%}rHfiG+8b8xr8G0(xZ zV8v=51yr5S`Tdb83(0m?TEER$)I_hwD{so@mF0QjR}fE;-Vy={fB$Eu>_cl3+r|}0 z(bQ0->>v0*z;pjLl%i=NvuE?0d#~SIvq7w99qVrpD??_3m=zjNDN9|mkQ-J#$wwfID9`!F zw(}ac$(ZiZ_V;UDAf*=YcsAlGno--Xqt-)5ErQjInr_W@)Kmxc-y~4u6*bhNbStnZ z)wx4+G*iZr?aU-Fef=)>m^vu;=B$pgka!W|DO$qKkGQf1&Dlg?crKeus62tBD=u%7 z$m_)KA)X?=bp#Ur@;g@!qB$oB4DoZe$k0y&lCI$DPj!)hiBlHBQKWZ-fc8w#_AJ!) zPQrDWiv86-G~WK=Y1*&$N!C8z;O8>0Zdd#K zc>A#5GiABjpRo492EXmJD=%Pjwp;serYjWo*BiCuJ>pLgPeBrX<9AoSM{`aRm}

X|$t~fw*cIk{38#sErbhprlP0G? zbwYK6=->M~*q@nzhTBBnj2*$~HQyvL`Z8d(bAdGa)(DD;o_c=hjomQ#Zw+wCE#dH9 z90@rEPA!cgd<*E*AZRY(seSh+F-Jmv2#f(^HLG@HR{{~JizxWhK)(NNC^8_LN( z{C@**$t_`F^k?8SNt>(Dzw}e8c=Zu|7LX?G%WHz@52Ir+CF2M)Iu!aH+bP0vMM`6g zPxiL5x>_ZFS4S84ley<2nTjNJgzUf1&TatKx^UFC7-VUXG_8_TLhc0DW`Nltq95H! zj5(X=ldEtuq_w|fDCY?T#BK~p4<%j0lZu*lTpKTMAaEZ*GowAoD16gp!fvZ#7xIKc zEogOHT3w+*L8~43leu#vnF4u9e25VbbuITX=tIf>NZokRZZfx|uqq=wTM^AI+` z8*{pqJ_Jd59@Lvxmy^Usm>UGBBt5>QOC|0;Am8xs;ez3p(eOdwiZ#3qr3yW#sNpfs zsZ8i?Y$Xa$r<#X+H6&={3snuCmU+^QCt7h3`%@I#dy&|y6l2~L7E>zT^(dCqSdEUv zc3NY*q!{gG$-EYQYj_Ad*me+(d>`{?;hKPyTyX@Ztolqu=Zflv&tSFUEAIqJ!Vgkv zUI8hcY@>iqwoRgVjAT}P{-Y{6X&RcS2M~%@2O8hX^2rybg8Zyc#+=%kinowE6gos3 zk1@961!B{g!mk<0>0z+3^7XMcE+0-Cv(7-ON|V@>sf`p+8%rT9RuVQ&R>>A)Va#O+ zH8%R>_r`#HwNF+XdyzX78jXm;6ByfF1!7Yh=P{Ch4~30STpMfSC|DukCn@z#0V%bS z0&3$c6pxXFjqOzOjUh1R078w8KDi6YfB0mzu{yc&DU3G$!Px3Gjm;6Yu?k4ZD~G_w zE?39e*bG+S%)8$qHMW41+DHMlaUjKGBw=G&mAt+WjG2#6W1~+#-5cZweX`p47c@!J z(O68q4%GJn+9K`34!LxmrjYk^Ldjvp;hD2CiJr#0by?0>8dfGB<({9VNBt`}DTNMw zvD~*l=1@4N805_|3iU)@ED?eyL1o>2WpyOKND+VLHXoGaPEEjpH0YX_>hS2A7p00f z60C#3Wse6^yK95P($Ewt{Jd#-nT!A10WLY+kwADR&+t{~E5`dA2$~2=X=dRKpZqxJ z`74}@4wv%|kg9J=YeJ!BPxC5nH6p+#< zN&$VMbr2RSi6{D=O5W84kFFO&%@g&>=_HT!$@)Y$DEC6f^KX##oq-z-^_5B;1vfY1 zl}a}ynpZ0Im{%(Gm{%(Q2*2XkuJ|bwJ12lNSAxUD^_8j_LW5VTHa~^rA2NoT1&oJu zU*qC)*w@$;G#$k9m8t=@%>1`ks`r?fzL3xdwP1L3p?DYU&XMph44+vzymm&eI|8%h+w0)ExA(K&pxR60I{RAy~8wp<1BZ9f4vFDWj-n zQeFkSBDZ0-g`=jVP{)&Rq{(3X-w+XTslsgp zfm}J1NfcY1i1&|FB>U*;MWsAgssmgKxB{1WKO(Oh!R539isY4eD?#rHl077r9+MXG z4{heiB>c|+xa7ur%2})-Tmh%Pkq28ivBq~`rf%W%8X>J~5o~V-()gY%m`=)0K8Ot< z`2Pstl3TpqkUI*O{RdFNVnPL$pHd2x?_EMtTjI4B5MgvF>z)L{QWco!j6|rG@b)Re zA)9n)d%A2%%qEu+HDcOR^_cc_Vr{ff-W(L)xzOc2ZldFsPfb{X;>(Sm7Zu+O>{)!C zqT7H};?tREw!ALb68SFaUl=+!PX@(z350$I-A-&(p!iZWQhd2rG$_8*QFnAn$pm{| zM8u^E_jjP6_%?U^;(Iw*?4Ces4>zF3JaWq4I>nX|_#Xps$t@#tuwj(NcQ$4R4woaB2lC7-@oIry6ePQ1EIq~^@?XELk<`V1V}MI;3~yL``@pHiH`r&Y#aH*4 zdgE~v<_HAaGl10IzYC_5veE*JO!2=I;F4RsyCAm~E_(}*mdZ#mp#sZKDFs@5)hCrES3)uFHBB$o z8XHgl<+Ih31anXsfO_?$LC8vo;r z)0;BUk`W+Tj{xb)VESVAtt#(P^kVZZw?e4deDX&3s@4mqrHflJ-mA^pxWEKS`m6#>HC6F#JyjxRiGr5nTa)fS+$}v zlJx0@&JChL^V}7iIqv}7UI!F2x}YfZ4;GFtxF0rXo~h&49d+e({O^f~xK!aT0}7hw z3xDDE28=Gg1Ia!Oq$%Jg+n6Se@>}n$F7fSd}`0obIb09ctioU+$-k zoOHqv2`^+cb_Ybm&&A4ZJ{K#4m-9|4ZXiR}mj%+S*r~g66JAM`YNXdS^om(*5o-RC zEG}dJpiBmG?3a6 zS5TJxhgXVh!~dHAm)scGVngnSaM=feRE}GBV;&`WcifOIr}6(Mz$LeM3r;6Xs3YXk zfmCi*K{@hn9Z^(r5ndPIk{jdEEDm6Q!JxP zS+_e7=C}B49)eQ`nA`K>rNLwKFO$r{#s1#P#;R2JkUWiG=LH}&k-K-rdvjH)>>&Lc zL+2$<=sWQ}Wm?%3gjz2EN%$Q`@DG?Y0y&uwiJEcu1@3SZa~yH|KS)k}CXK5sa+ zI^!|j(5Z-SK~5_9dFhzF$d&98QT>(dN>Tll>^yq>-N#kz`qYon1@m(e6h^Kh^G(0R zHfP+N6y4^Go0Fp3oN;p!RTsQDDO~+esD75VvYv9ovh{LbewhreZXyMXo5@wAOCbXo zs(D9;jurA@fO<6VV1Rlw?_dBu1r6{XJ((<}q1nj-(uepc_tK8oal)_H^+C4VnXm;2 zi>YIjvCkw(AN&shxa5`*$&^LCp1|N+fK>0b1!c+KuU(SdhyRrTm)v-f!)EDqaM`Z_ zsT{Xe$HYnAw=d+&`}qGD;F24w7k*8a18~{D0jb>1f^y^?{5wu8xDpGA0WP`4n+Umz z@Y;DmTCa;t3hFf#SS$LoUQ?j==@!xv|2<km}u1P?r3qb8!PE{tp9O za*NjiyQ+VK%TDMT$Z`L5OjeTjg?_E16vE2^TykT@CE~0Buif0pl`AMm-aoP`Nmu;$ z1i0iDZy4mpz-7+@(yZ(-F33tMu&&Xpq`)IDUn$G*zdDxDrK~#z2(2{sN3T zE8=7&{#OHBa*NjsaxcPVzXhaOx%>JcE2+RLqghFT4)yZn0RE51GP;y?w*!ec?fYao z2bbe?3rtu#m8EjUcYaou0NJi)!X_guCRvo()6kM8_-_Gl$t@!e9`LiW8+`UKAWhn} z1!c+qW3vu21^=@EF1ay~!2az;aM`PXRE|5nW8x(5?~j+04fx*#aLJ8><`Cx_aM^o- zRBls2Ir3IM*iH`P|2V)UH+JAb&h8GmR3ObtzPv;fSRJ|JFNFfLZc3CIU}_l4=u*~A z0>WD_o8jIJHI21yxaE}zY+11_Ey9-pCJ*0;hr{Wdo z6VU#Dz2f{$U7^r8>@*L*NE6RNqjvq=^ljd)8I6QPgRlX-%hd?)Azb&RpX4gRuQycu zw_V`e{H?pdxgET~R`5>P3f>PYoT)&X>Zh?49P#Qb^kjqv3r5<^$GGE8Fdmc%=SL$ktJCj{22UIliZ$Oa7k8s?lbp(M4Ycs_b#KGR*KgH zhZ=lB`Y}WAjvAra)qHHJ{ge%V`x)Z5dLsVgb#NXxPGKIavh(m%!XP_2K(WtE$C7)l z#WUlB6Yp6#8-vI06w4S>U8$hLi;nci{}6ynZrou2xe0LDGl3!zND5g_oUbIeapc*E=3ase)hUOoyk<_B!|-4VtdApg!0hB3iQgfCBbn>j(g7sM`2G=}*CTP|*C}!bQzN35xnm(?I(4dcNQ)*oBxNu$)ss7`_JCJ(JI}q6-=OrzG zCwgAePR1(Xy(|u8uhIGxb9R9PN^&mk)n+30>;k3YNIq7JA->=QJ@mjYJ~*5rY7Y*l zh}wh0DR{6z!NV!MdN@MiW~gx<2Z~(Js2wLyY|yNjimL73&G=af9tPpx+4xzu9wOlt zhlb}NY`+Z@<=0y?_5`e0$J+|E;vX7+Hq89fqiPi_5pH-2J za1VjIu?l%3kEO`{ka1Q6X&Rq>1P><^;wcs36;xTOA5=!p!{Fs4(bF$})2z~6B#VE$ zf+`!(>cyuks9P^2IbDysEB!jHp)CHxKKDVgRJ!KXG+BJkj^CXuGxP$J^J%BRgJ`$> zh-jyzV2sd%(Q=r>XB$1|<2hcJzS3zytNiQIA9NbiBSSufwclRSSk2_cZ7sb@MN2e1 zZE0et)~ZjbhFYo(xVrwbhk<{Z ze)#aPjy{Y+oZBBM;3V7F2ZPN%Oo}IM#wZ=2#25CG7l-r*N-PrrE(; z57lvcoGp6ZhMBP=Ra40OxYVP1F=MA5Gh@fYHzG5348g&@xPJ6e#Ns>+q(z|zm?S)# z7rE>J4XvGjdd5_xx2IVt*9K|4#L*p-sz#6QnA9SAbjRd$J-Q?Icyw%x&wq<#?gUMf$(0Bl9%7^NpHIiE{$YxINlhn#w5R6DAsTla zjs(#L4(p70RtR#Dxz%StkXxkbw5vM!5cRAP^APoT5w-<KTzD3JV7&lpta6spWg{C0eT};>q z2#Xm3QfAifTp57>VE~uhGUD_X{!8Li`0PbM8sD;lvgBX#d01BBe>K1*H#R5XFqCKD zvfltwIUbu56DN7AmbBz!{O<#}}H| zF6TNR&E?-gM}^bMUK63g3tySTHco~HQ>5)`z++mxw8bfMClxF=3XXJS^_g1>qoQ|w z{s~g?j^uK>2l3=nW$4LKPxe5^rV@M5$a|o!Tw60cNuxJhk%#e zX_R+Jc9eI>6?FS|K!n$_BWz27TZ@;`tyFqg&s`6c_Z7ZLh5Zn2J&IRpMc5bcE^>Mz zh{zAUKk1=K9Pl;xJOi))KlNPR^3`g#gl5b z>L7PuRs0FN!z9T$>o)0t;9dIm8T?tk&-VWu|1Zw}i}(K$Bta=?mkLUVkBJBZNZ6?9K zTkf+qUSrGo`B}2UnkYFZ-C{i4+p=$AZR*xCIg)1@Vo@a^Yko%r)Vn-(E!~j z(`p?kL*cs?a9#VpQJwL!5#jB!%X-71asWa$fn@_A|6CAl%f9}J6E}%Smkpb8#hPQm ziD@x8K-sEQoSmRvfXv?j=M{LeFBho{=yEcK-<1>KulZslC$u?Nx*)g{z_|dH$#6_0 z|DMmXBl_{>XCU~Wso)#|^+5!$Aph6!ya~s4K>f#0RFU|=9+8&dx5it!&IceLL--FA z*a=TcYmuUWZe)4N+7FU*_qsVb-sjm7c}wO}ewtP6l@Z8{N(gEMkp=*#EIb3?=ntrW z6w;}YNH>DGl@XFRB1LgWiT z-aFrCM_BBL3-5pWvtG(@eg=o#23~-30G?`aWYhCGJXgZe8Q}kNemGrO$_Fu8k~s)5 zPPgLYSbWYPXbLzdQS={pR>QFp(2Z=ZlFn3?lkoitK$4$8`Wr?RT}4I6CucLrrE%0` zaX@~45MByt#nB?n-@pzOY9g<}c8y#=1U8J|YA*oCzm&T>|PV+;5Kq-BFz9f%pxOXN#0=JvZ54Xw%i1s834_em*IE; zpxM%e*LI)N7Jod1Y2au816f981WoNk~b11RVzNcNETtr%`7S9O#Nyc6i=_V;7*C3`N21 z0!p|Gf&fohec^NNUOlJPb#SZ!=^;R?$?*IF$4`KqO)ax!UXblJ8e+tII<9E#Q5dA{ zL{R+!U)6|&I~rc9snx@s3$i-}$Qc7v>`E*n0-(@7@lBLpB9aW^kGe-fxAQ94@*&X# z;BJOTGGF;ETON-TfT$~wJRcX{P04q_HUd%u0C}~YsEidO?aLHdT-j*`jt{{49su$G zy+2zvMoOX-G}6F}Y{c2vfWNE?|VE2=v2ux&6{2Lh0N_L-N4=7+zLLq)*&~dOH1MpWM-JwDvETr$V zxM{k(qc;Z>h$gX+jESZ(SQ`THC_8*zL%xcX{_R0!Uo;Midn0HFxCQ{av_*w9e71ZD zzEY>kI6Xjp48aeR<-6}|$afJ{G8`=buIiis^?L+=1E^o-U=2AEd6aozDP7J^bgF@z zaW$$Iz-a)_AUOKdb2&V#;aCOeTyDN!JZ545rpEL92AS&pjN!AzxMJIwU5trs%ntCsLFL-Ul!+)eE@L5U)$j=rVjJ@l1daf-jWL|I z7*|B5KpWG%mq=rPwlRj)DI{dtm^;9_7*L>%F=FvD2#KdnpU@_@F)xDmS%9`NnZ5CL z0MItZ2qhGXOWT+&7q>Bwm^LOmk{iYqNf}~p%xq^}>yLAZrF(!&oL?q4+ z)k2!b+N9}*OkY5OXc7y_m}usM^=6Hx*S?^7-G+MG9_o0^-fU#z$$)-8=!2s4RhFqL zHwQ$XSJ}g!1Kgcs7s)qed?wx>scmX$GVp~=GTEY)yVW!b##O>33B!$}OcGnSjj-=2 zKz>fZ&sI)H>WD;o%?Mhm$6;R@hF!khh$+#z|8YM8{~XQEcLkhGVdqtWhsj!HnW4p% zu?k?SrUhlMml;JcEn@;EKRw8!&I^J<$+HU4B&1<1nK9RoyQnN>lYlYdqY!g55u`|R ze#ycPoq!APg+5p1#*>k3o?2*KBXR-BuFq2q4WFfa<*Ub6c6b6|Hh3xn>hq{V!(;nA z<>MPWOF{00@D6}3JcZEkI6lvgR?b>bCm?tnfIl728@6~k5g6ZZ3e4gw&&%Lj2GH?* z-+m%}0Q$;fgc2gL_!>puI(}!mIJTb){@DN>+dl!v;|Rw5rTdGN1n8K*IUJV(B4hrZ6;aHBCuhl3h=kAA zo{kjGK+pv64x#WEc&@k>gJpm-6rS(k_yo{x5M){{1ZBwpw5WjgJQB3kaGZdRtzamW4Nw2jSx@vA!P^4O(RkV04;|f|`E0l%`(BejaFgH)Mi_v~%fzTMsm? z7I?HMv29EV4*TugHh3zV*h>8?@KC06&!8pVPS4LVO-}wlfoC^Dg6M*pd@4{NvXL2r zVp?SeGK4j`^B|F00IkW-48{-`pf%YDSz=twVE#CNu$|?cHF+=iKL%(`{u7R~fM`uN zvRLpRF(S|s)a1+|BIN;EliR}41`u15jbNM{fc2M{yf6%I;W&btJO(O9F%C{%+ylo_ z#u2=|XR@U>Iuau;-%8{lva2_;E4I}>UL_~O=ixAWKT}g?B2U8EIr1j7Z((#cJd=lF zb_d{`gHXya5gV}H%0Q5eLIY3^j@Loj1jwwBl`VsAu{3Tc;)pfhg zCjhlI2Py_Zmjwe_)qv*~(C09$^Mir1m` zO5{POG!N8T2+jrYr=5~vL#O1Mm3U$n%wkcV1iq^QT9kjhPUI*+i?R_)h{WP+^o#QC zRKTK~ay{BKfEMM3aMTAxi?Y$;7S<9J<^JIBO9kwdZiZtn6~uH(Mg=;h(U&aBhQTWw zM^Kd4LFE&SgGG5a9Pcv@GZDpO!}|x%7o8F!;q%R%jTFv9kc<>LMd8Wt{2QKE01nS^ zA36%93D9j2WLhl&rSxcgRROfW3Zja-xlMF4;~XGQxH|OdoUbMA>K=L5A6`o6@Z|7SRS$P zDkOoruY_*x9&P~FSU~=&fTwWxP&YDOHLlq1;Q^y*BeXn4O)V}Sud<`*W$fo!@Eak+ z_8Rogvx9iGdsq^9%vtbLL#BI}6Fg#e554XRJYU1}SB$QYf~v1Q#o?HCApcC@F+Bxh z4yxDAKo^QG^9E(7>m1|_t5=z^7>XiATD`iBL+=jI>Scs1F)r%kkB#+@v!Q_1YZ~~c z0JM5N1jmDbX!SC(c3~|+_1XdcH>iNs>lhqAP(e)fGAc054eKw_Qy2!fa2!GP3Xc~_ z259wa2uFQ@sb1Ps7#)d)bTGT-A-iImQuY)^!jl6*rZ%06JYi2U0^0jBI`$OhCy1m0 zw5NCpj`alXDYlOddW!Bhh_nT0PmvNlN6aJyED(!%gh2nv3>=bhv)@xljLVM!Nv&+pgMQwt(H-(v0kU}ShLOSHsu5{l zM*D`LV3`KqDdgwH8ivb`xO{F53fND~0?R7!t|b4W7=EWP|CLrN=U%Y90p8aE{As6Q z;)s{4kaj+%DmTM+mg-aBI|R^Dy?zoZEkH}P5lSc&k&b_uc$Vsq!T%vZOZ8beeg{NL zwb9}h))JKJij(nT4A4^D4vs4Tv8CFmNR+v-{t~6yFnERI2uk$?s2s;Q*l8?>V;SR! z9sd{|MGNU*cJ)Sf#gyv$9IzURRM{A0s=oKU3C*uDD!%umO~IF3fWG%EhvRlYx9#B9 z_nt~qk#s=xz2{o^t^w$K&)}c{*-WF8c2$94pf9|FX5 zY=+;CNXB+-#~>N)*bJ{zi1!j5ThcVajz}7zlf(NWc5OrwXjN4tLp!zx;Hm@2|2p6) z+_C*0^fs0lS5-fGs_6!!X)?5oqbBE)9h&nk!;TGE+j)BsuXb#E0#D?<2W)!D zN$2*#Q_=UHhXRkj_tcDi@3|*d+Y^Du^k#gLh;dJlp2OFmk2Fa-4zn^B_-5fD^-2m7 zL3-mDy6M|VzbZ=4Zx_SQq_L!x3H|v3hDtj=vSbEWom+r(nMm{cVMr40d`LghG>)FB&tyZw+dU&mK81jNj#9j&Jo$x-FDb9y zR+laM*I{?TS~ILXioK+q5!SUaSHCbzash?axQxR6hUhquga@p`(VDY;UR)?P;`E?$3}o4O70IG>xyqQ|99 zEBn;1Jmw25xYGjoFd1c0U{%1V3-ibmEX;dm8#bPm22*5%N?4OM8AqgMgY2|23Zf^Y zPAmK7aVWSBELaY|9ty6BRgj;{vUAY06d=yqvJyO66o}Cx1j|1p_;4)%Lrv@^VsKW6{#0+vHk8 zF!?f`zdql}sprsQ*!FNCD;CiLiVxMy9c}t-13`H@$9>= zn)DV{W#+VTlUm1@lNQsl3kx#Dc3e)b4DCp(ntabD2B49oU^}2-*pr?Et3&JZUqLd_eu#_>gIY@HGwx z{TJVFTh0oQ2O_*5z*!2<1~}FNc3bs>DEu~EzZ*>Wmx)z+Wd*@!&>ajw++ed>O0S*) z`2xsGeKL9km45qAUm6`-qLaB63Fa^Vu0SHOU-T-Z`>a8*u>T!dGHLUdVaaLhrkq9f z{R@Wu4$wt+wb8=+i|#Vec6OSD&3S+pTysZ4R%Lu5`8asCcEsgJid<~xCOUP&vIs)A z09xh4^BEj_0eMg0ecT8qL|jI=)L%t)VDlGP{{;908w}AXwmygx?P5GWP@?D!5zTnI+HKem0jU^C1AH?BP=Lg9ACEwfczc8zXkB83v3LZ zD+dBgSHn_=3v5C+VRZ&T7up!Ukn%-_YYrFMv;$LXfG)UM2*=F;=y!IZ3c}0h;meMA8~{QKLwyU*md)sescg1HpeSAQ+8uX5H{( z9;&KFe4Dty zE)#PxRD)+A!uo_$INgBv_iy#vUZT@JQU{-P+MdBlVyGxADh<(SxxGQ?!cU0%gLF6= z-gSVd7VJjs?*sA}vtfykxB{aiYa@OG>q!89^WqrSsKA)A0Vm^q*usaLCDKh#BQg#id7|BU0yQAl>`1^P^x9a&R{i z_OUJx^D-p~-}Mhp*srq`JpxzD^E9n_py+COo~ET9u65^WTI%5j3ZAC*XjhCD{L{2P zISU=Vpu-;rykccmMIb!5>=mbJ#h*i!@@{jIWf*vzaX|kcW!C{FMbWftX6|klZkF96 z!yW0!N0cMyEE$oYU_gnI6b=M)02L!aL4pVv02M?85lJE#Q1}D^MS=nbK*fMcP(bFz%rNm{QV2PBfYC{ zHZl$&AW`akd221%fd8#!A7Cc^h0dG}N>TDWa zS}x$Kc3a?pIYPA z^82udVvTpzZj`46RiJGqosT^==rY$%(_669^Z+ER#UPs9n4PAq^qr7bbI@tyi&lrE zIDdcU$VSTl4x-;H&`^>pR>DNWZO%$bmDNHMxB915N@~S`k)9u~{Ma~YRg(&@*iNu3 zK6J!*5_MG^L^1wGg4&#E=|)BYotZG9e&^GqDrD-y+0?M}gTP+Ndyce*&%7E$0j9bo z)T1=<%Z(nbHQ^ApGP%t+8NQauDNp>+l$@Uo=r$0M$TygMS&^A>bK_>n1G*+| zu%*DZx`AlkX!8l_<6we{b}G#|A%N5MD%#akaIZlCRa?<+J{d7MunPD8YpFiOS7&3u zVDe8$?jG=h!w1T%{Nh!~@;KPWMqmCW7^?`{zTPXzv9=$@7{BcqBhC{$lVv;AFZk-O z)r(V4`yfvJ4aWHDjDfYdPYlWbG;&)LL<`FW2^0xkXG~&7M>l}-jq!|7IYXu;OH&9N zqkIiM>;?Jd!(Px=`TQ0n=voqNI3!u_r1~;n{d+ts1?pR`jZ^=bF@Am-gX4J4B+ELg zpYYWyq1y)Pdwz^luf`Z(oiX_G{dBVY2~8_H%PVdX2x$u@-D_5$hB*7aGNd%(t?s@M z4P`;d?Etiag@D@u?62cGz+yf_IlsX2+;vpH%~z-8Ie+{pjyr?*_2;)RarY1fr}={U zIPOL=Fw$FFz(Xo9Q-krNANc>$=OZh5p;n#kU8ppXr5CDtc%iB}E&qc_wV>sHFsT8( zQ1Ml+2K)~uXFa{0u^u;)G8=fBivoq-w*;PyOaH_YB`*)8%B^$vsR zEqW4SHNmxU$DxubkyHb@E;t8Iuou%C%e9f zm+yd>ICni~$SJD-?W@zGn?%mfM~RC*6IbFi(mYvTJW!nYM4^!s?<7r?ZFZhZd z&ku{+dgbEt!)8R|0JTMqya*XC~Dt3it2Iqfn$#K6F-IowWOR{pWz%M%7mJX?J5u8e=3|!4SO&MY$*cZS@^6a zAexC^spD)yf~&zY|L61Tk5*uJ%6tUT5~WV7wJqgK{C@{$Q^S561ojYNt*pR&0M`W3 z=fnqK%S;bzCOgcKo1kao43CY|Qtkm)?GR6Gj%bPekJm*vMZkjj=S^TTl=e6to>y!y z^&+EQZieyB29b&jv#&qC zhnRJ6y8gJy%QV&!HiCuq$E^!cKCVADy&qfW;dK4+QFtDJ)AdJR$&k@d3f3P#NBkZt zaQ(63M7;lm+hM%tCBX-OZv`=a{c)ui&GpAaNSo^=Sbsd~vAO;jMG4}x|6%>H%n8w7 z_Vvew$l({z&$j+J4jH={?h@;d?}ELFQQ6iX2Y9(L#rL^sq00RAM`M!6c_`}oV>bl4 z!0Gy~mH}JWsa=E7Hb7&?mgUyBB>_9hTuTde8$DXwH z$RVWPf%I$UPlz+DBcNNL@HU*TKl;l4{rT&UbsrF^ z1E=ecojq|wP69Gmf9#9+tKj(4^+#XMma^!n%xT%@Q7*1O-Uqq6h{uIJUoNC_+1DR8 zL1;aku0Q?_&p9}Nas9EEmyaYFO?<3Bc6tH$Tz_o*Ah6+d{qb>l9)i>L$G$;XV?H9f z>yJ~JL00_dkpB!WZvD}hH^uiQxQe>|xC5Fw)*mlG{~VmV{`g~1YrZdj{jnHx&h^J~ z58={+YjW0;&$j;Pr_A-ono!~Og!RW+p6oxaKYqgN9}Vy(bYkm|!#y=i81JwBwMXi0 z;dK4c7fBS~=ao5Fe;fhPVQ~1l>yPI=RDVs8>6XS|sp;zaqc35I@AGB-JWK=N6u2gd z-gWQSg1i3sn3sgEKZZS#Z0nC3FBsAdi5giTI+#C>yQVwOXd1fwH9oNO7GbvtIcoYi zGPX7*01F|lcOhEDbZ-{F7$xs;Rq@977o&oj#dKEvXvlKn@AmOOZ5W4t7J&d?%@Ic? z3Gii+hfkB20sa zZ?r`jq#LtviFDNLC}os7p9r~TkWu@_?^Q4~ZUAbPAc+=A~J$4MF|hEXcr z_JM_w7gLpZL7TdE6hxc_Z{zs~K4TP!R^5I643$NA0fC?hu1VeJ=Aj7BK-yiX#s-3* z2(irqD1zF(bZ$XbS3=m_45C>zpO4RKX+9Z53HRZ{8FCQ+$KnNT>e}l;Fv(kv`9FL{ z@g-hXXF(Z>yS2Rzfgr1{-Q;t_k=6E))~wPzW{}kfT~)Uisf|o$K@m%OLD(AZ<2M+m z=fGh3d3cr9@Z*?8&YfORL|6)0JF~bu`+6Ll@C`h-1oPO@f`jO$j=}k&WjmEKlnX;r zsW9jFEfwQTZ+=L%wkAk=U6e2drZ?xA>8|O`d1f_(QHDlI<7Z*MGT%%1u>580Na(0f z!6vpvtXP~V39Yh*3Qap|`_X)5%8hcMW~d_dQ*J~g0yQZx(P$2c%-4J4@PO8#d<@9M zgy?}rp#r^cy0?Iw`>mk#0dpvvc@0SZ5M|6w9I$#1gM?k*lH0&~#FsE8!dF7YT#3F9 z+AEkE?)mQEca)a{Eki{+9 ztO-1RheDfvDJGF;L9BjI^lCF@oE(^M)MGuv2K7FinR73^L6!HBt*qeja$r_MCcRkh zNkiMdLxi}teG%athB33XZJc3D=OEhl804+OOM{dQWQTT3wQ*_wGH2rwAZT0)Of*to ziECU2vNi6>Y9%ET%(if5Ly-I-%EUG9_=~NjH&{b_38O!JB~;99+@%AX$OQ172qf-@ zFBF;#nom=78@JD84P*g$i+nlpnhmWxX;@R)3cj@$B)_Y=0bVWO7I@rtc6$SwxqhwI zasuo#a3%%+wQ*S!XxzxlY`I49^?8Yw8^V^m148jFcRj*6EQ4lixj4(9k>?4p-USG3 zqApN?oR_t0S~SOH9XVLB2$pb`1)l4Tm?7Wsj^i)R2 z*?Ap{aPICLD^px|j+Cr~G97_}Z)Ml(%37v11Y#rUAcZnf(=F3(M8%uiMVV+U7G)}R z5@kBiGFh)`nF!#Oi3!2RD>I$$Mt%w7vO$>$k6mw!9?0mpGF^^v?uL$)DZZiS5@4;FOACYU(8Vcd2keCBu%g;=W)R4aLBhD^r)6L2<#a6&tb-aZ4L zxd=qXZ&PuK^a(eYH}Ss#&Zdw!*Q1Qv;4?o3QSqTFt|I#y%2BXSz}XZMrz(p6Cw!*0 z!V|Bf;uNX-YXK>wTQ{lBbI`d%bG8G#B%CJj!^4|Y+BkQX!$os zkEo&wr-q~D&DZ04gv)=D{1V{nI%4TDsp%AVZnz~ZO~5tVfw(<_hJxK5!9=tiFH<&3 z?HV_fm*Myy{SV?c!Ni$ADlGTFXFd#~5RWSa)i%{GFLUs}0M4e6IE6>T{u6xW8W0ul ztl|_2%`7V);(rI6O(AhIMug=H_{?J>&gjB)ve$sYl} z-=j`b-1&M?Se^jaoCV_cD4KP4dlVDVa@@tdM5%rAirTUP|6Be++$NYfWe39k6MW_Y z5QXTk5LA2UfiyXR|6k#33c;G-jbXV6pPBCsPrS5>Q{?w%ouwq$W#McJiL(r4sfnQ3 z&=(J@I7O20uOuDu-yP1T5NvGp56gA%nL|M|GyAkrDe^~A%>VI!51dUQaY}W}FOS1# zJq@DbXPe>n?i_?Qx>w2H1)`6BGp@OmH9E6-PHS}ijzO@SF*xNLXqp#&*boGRLsJuX zztB=H+{bz*ez!I)CAu0X`oh^%bJ^q1wU)|Y)d<-Awmx4{t}=?-Ex3w=Nc>NLQ#HE>eBz|e3&|At%x6J#aJ{f3esEnu3)oR=y>erHS%d#A@q#vW z?O7nOH{A>E51(}aL^CvoRUQiM;iYhxox13oxh{neSh-Am<5K8F-AM5SXkDI7OD_c) z_0DlA>|jnKrTU=D_lH7oDY*Rgy}=(3zJDovNO9+peqosnuK7HOdnwR*xO*uu5iLh| z?K;%H2*a#f@&EBZh}#4cXY18rIS8M56htBFCBCB?dACB7ytd>Yzm1}8b!YeJ~InM z#h0o$Mbe5@mr3}41kR=qj*@MaUuMB)JqM!VBlS|q_gzTdBL52zy%dUH@=}-pX;glk z+2=?Xw&I^qGkB^$B4HqUky$mB2Cw@K*XyKIO7dICVr39b^KPJA>N+9HOYx!d%oHyWY0~zzmctN8)QKD3? z+rG05#Q)83HYM#YAmX&au=gN*<_r)W_RcJhANFp%5u-ztT8T57%X0j$jTf}3Yfk|Y zXIjUwY=zI-1)`Zv=j9a&6-vagR|doiug^8?U4(F$X6@sx*<;vCq;BMDYBkTMrNiF1 z4!MTCdzsTn+s+u?`amHV_FVqgoxmRjzCY~UPH|^N_pnR_*PH|54tvatJM1wLEr;kZ zLG4@XwvxB;|K2}{+XNHm=dNMd4WIc1h(dhBC*A$%`^GS-IAYz(5}(^Z@zKNKq_o$=oT&Zdw!?NIdo@R`Fv zRQy^Mr^xi1O3EGhzYor)kVGD5m9K>t7tm$$K(l=As_66Jgd5P_CTdVuMWYDls;F|& z&dp$8HnNH8hh6p2!YIh&3$9KLiHRBR+bQxBeAZbI%}oK=WkpBalp={6!H<9_|1X-Y z428D7lOmCAJnp&+VstpD4pg=K@OK-S9#*>#k6A6L=-PdF+-k%Co|u-V$Uyk4+d;H< zFaO?~wN%HLt0lo+>;@cvBftI(tHc`#;_s#S2Ap6og$De+6dLgNQk;gmw-@^Y?hV%0 zz(DtN_fp)6NWBYiFGbDov6|)NKNXUb%+)y%EoufV7)xZrE9j#sZ)+b_8&nQ<6yVyP z0JEpyTFpVU5Idi;UA^kCAnX_7`g54^k2o~LrtQn_5H$yYDEuHq1@{n%5@+AsBpHkU z+u>|Vi}Tq)Lngy#Ed^0M8bQ{Z6gMBf>;~dDAD$Uw$XW;+XMGKtl-APPSBBc~6R}KI zlHDtm+9J#FR`mOXoXL2xp5#xcPP_LlFyQET{eY1)AgSv85Haq74HDzX zVqBn^$F}3M*da-gwh+T0HA0exCKbF1fz0>TB}mduNd<~NXG^Aj4lp*!sXg0P8&~*z zc3e^+D)FxMB+Q3LhO?DV0A9;?3x**7Uda66q0Y=FR>ph&9cPD;I@dtI)IDuGjOrTD zJI)THsRs0pv(qS2r#J_P2PL00gtpZkM7u;$@OAi2>t&SGFLXnpNo#QbyukbC;}Fx1 zIu5rRy?@$FPVb*zS}lm;-aijn4H=NeXOi>D9Qd}uX#zuo1okJDUW@D_wXu@6KdF}P zBc&Onv_nY+w;~3Ut5qp-8k*Rb!YaV((!(c9y3Im%_yN>;c+Ym+5*2$Ga^>F4rll~4 zI$Tr6xUVpZi|yh=45M`mY5m|1ZqW4!->!k_r? zD|EbHoNCG;u*}mwe-b7<$ea7w@dVC(qSRSXzqBNMfT;tVO%3}vh%`<|Zz~Pn2sllf z+19cp{he5z*+u}p^zX!m?4_yNOD`rywU=&ur@bVZNm?LUVnZ)YN-E*@QfD#l__qVq zdftcT9p8vmKfAm}+M!wiuRQQ~4CT1zW}9`b-ub+z^org(e9WT^s~OgNhwc0Ldo#^X5C z6JE^{uPFa+atp5}7aQW0n8hbi(MvtfPYmD5XIr%rz$)_WYT zHy-0+DumGS+{fW#1c;1!TksuVJpWpnh>HBGwthgum^TPGy5&TbuvneejrdQ3!3H3q*%vB(&`YVRPI8}x4`AjG|`xRuO8AK#fz`<+E*C2$q zs*Ua_?3&-5jd1w!X=U8klrzjqJ1UfsItON$wG8FrftzZ_T(#fkm2lN|QZ0#3FMPWR zwf%UObJdg! z{EjFxo7sB|8k)W4KgP1BelL6Ui_>7bA#)&VYy{EDc?qeycSG!_+m1l6N$or`j>VJy z$j=E#>uh0CS4|#67`9SA?W($))P6zTN#r*NrLuPTrGf?IaXXpr9SKR20=88TMAJ<^y&I`33){;cQCdn`wN3V(o%n7$ne> zC`B&Js4bUN@`j&DClmW{D-ez}p@Dd2SPpy2c3h%C049l)#$4aB-RZru4*HYu)s~_Of6Au}w z$rc|-OrE0E^X6i~{bKmXEN557Px9kw(cJ%NhzoNtPdo$`%U?n~>djrnd;)x}M6O~s zLbMFHfP5$J$)4}M%(rHl^%V2{OXjsIVkMvtIT+0xRwO!0aIAl<1O>LE1kPr3*`|m# zJA-JKec`zC<9QJ6OBJ^HiI9>&dYIVq_r7ys!P;CEqmNce%oMw7rlclP;cm;ZCf5d7t zU;)8p*CBhn%f@+hFO+*irJrXyPlK(E{p?eT!H{v~`^d+L`7;uw&d}r~kTBPQXjixz zQE^jDs_%K-lPXz=dN}Y%M zMC3R8{{?4L!`=gew;+8JC3znPqoyDVJRQKnfeLSRPmCSs@HAum0~Hv9hrO#4r4ux* z8-4ZrAfy8~zucyVII9ne+(My+zR*b&T`xYKsBw0K2Ha(|Zy~Gd{Brv=Ysi_tPh>gO zKljyN0N;qIMH6dZ4;fc2n%K%+H2S;~E`O?B@YQ~E6B;XZJ~yjUU7T0;V;I^Gp~4_q zd}2%FHr|lhg9lLDkTA{nZb%qtG0i`RyMZRhX};c_oWBMoN-rpwLwzLP7vefA(Vms| zavT2dfwL)%t&cG2Zq0-j4X*Ft7VwVfbUr5X0=ULj5Y1Q%zhe1i%>V>~n?#4VndN@S zfOQ(udXvcLtT}X_1Zudd?oA?}R_dN_(Weif;9-!s4s{qEEtUL=KA-7Iepd)P)Rq1K1`w7LGSuQVclw>$9)Ym&niJl zcVwPtRn~w)JZmL0V6+3t>{GvtG+SV$yvi>KBadRQhIJcIHCqeEcmqE_*YP)Nkm+yM zP`_e!ElC-Fvj+VVd6HS2?Pu}OiwVI9*nS1$UuXO-H~unr9#{+Wz;)n+b=HC&qwmFO z&RpM%yK@=m>Z%3gFr>^gAlkOw?)DadYmYP1syqYqD1mq1n;-i?af5)fsbPNyBF<&k zKxD}FTE*nYWmdtJ2$}DLDB5L&?Re|uvK0-3FPp*t zeQ5>Ri~ld+Y)ZTAmF=s`PmB)OPJa|I|CcCWO%N^M$QqX_;F9IJ3b+mhYzHm#TB@>u z6v|ydawSUqaL-H`kN^AQ6>MVI-9Ti>Q>fe@Ct3xcLdaYUqG+!YHda7t|8G$_S%d$L za5kkSO7_eWm1PJ1KMmNCci2(O9YPLy{ak&w*HNq98SAKI`W;pMUPq;j-%;tukp}Na zr53QPP9U1M2ce7Jr%N`65$i3Z{xQzN{0~CzsQIT+=`9H29Mg;fG8$}ivJWsgzC*5l zl~qN(XKEFf+4z3}&ZdTaJBT!X0`0XHzAxdlxw>Rrs=3+&NQY%~&br+2CeF8Lu2ax5 zlMj2%MWNiyMXp4N-!WoKMF`i9SFnj;{{=3fPfPN?^Eq4q;s*>bC@=Pe7z`k-nBJN91leP3th4C6%Al z)?S)av@`?v&P`^Zp#|SyLk~NM4n4}1#&JSu5iUQtllJ=x zdAGgKGZ?EQLbJ~sQ2_qiq}jd*GvC_v zm1DlaSeaue4cTq`xd|?~eaDUY8JXvXWJ#teE_9-k`rR+kC z$$UAriDZ?v%|AcJ8$cfEmsQR#MKOugnS>HAAs}(?ez>~Sqw2J;Y8~Xc;w3_j58**! zIFhwK){JAFVqrh;X$vP)(vLTDj8FL%C}TS3ce7RW^Wy4pYwJa`xMoXh>qWDY2DG+b zG&9|t+<*(x>Oc&?g<@S+ugJTtmc_2|ugH6>Mhr;gJ=Vw^s%5JY=3bM|8fsq_B%O7W zou+}Aqa4W^<^{B_hS@c>!~>*2RDSidL1C>K#?@nnv$|>K(LguC--rV64}iae=N~9p z-Bak-n#}56Msc0O_eCY;C^|qUNb>RT8@ToK^)-dmqZeCK3{dr6EoaNLDPZZYf@5G)#~$YS;=}JDK)^$de94Kl|GnMokizD#xTixI4PZS$l?;f zTOE!ei<~Bmdd8b(v#-1J*|C+$bf*bpTd1E^-)vkeWp1B$B14)zWet4R#~?ahef3>Yw}J=E^9QI|k!`a%x~4}^hh@I=>hSUH zUW+nP$84(V)!6JZqZ*^#I=swiqXDhMI!1BI8;<}T-(3>1BM<}qYlkUu{=f7uyJ zmVsq%@%c;LQ5i=gK3H)Dn>R|8`IBqQ9{eAIv#DUO1`(&#HLI)BwGDup0Y8qTJIeVY1_87zKtFtzwaXBG*b(yDfXZe$hXulM6wz5c6Gw^DX9 z{vJR6vvpPwKlB;uA76+{Tc+{YCNcQYxV7EO=4kD?kL@!$LOK-sZ4XxVp5y@iJ#=({ zW{eKdWf%EVCTy-EWU#r40cjjrVaXZzD*xcs#1S_EoD7W3Rl65kY^zIbt~!DkajyI% zDqW#zjRMhb$js_(lk6}3L~%z+SI-f$+QFK@;n?7p4g zTqF0}iTg2{#a`KXQ;hmi)Yb3M)VktRzUE9ltg9m%{KX86hvWuzr2)ULG~n0O72Ln# zOj>~K{)ob)foNf>g6|Jn%NW}o?}5vp(OU)+5=pbS0UxIldt4t_qEwLc=@<1s4b(Y`gr=p9KFc@R32RAdgAysN$!H zzYlD^GWn3=UwQi28Yz^5_&@yko3D>uneOWOpfyoR`U#Z~1<^|QYb^6!hk7Hxi%mWC zX@QGE`m~@+Mcm8F7tb2d$rPRzG{Ac2(*grIo)*Ru`ZYi+V5g#$I72qCb1}wic(*`1Lk* z49Xu0eFlNZ0ZMg&RM6(`+k@s{h0EW9HlO#W>9;xF9((|nL<$vz)>c%&agcNp{so%&J4bV(7((KA`dZ- zw?RYuC*M!{^~@_0_1x}J)bo2B%U81`oNVLxRd+0>c(75HAi-G901iTaI4bYLXC462 z&iff|M<@VRQ7c&)2Y7Fa5;kOZl@s_s17}mi{vU`$&aoTSIEi#Q{<8N34|m@Dz^p+o zAa9mtmzlI{jdUMc5|t*eiDCiao-OETF;ee1QBx|`HOsP0snhm{ z8OUTns~Ld~&tZ%)cszY(<1QtnVhgk2bs2hAeU?;gZH9)jnj$ZK55q-gfSh_UUo%Nf zG|MgEdH3g5w$;E=28>m1U=af$FVK#&&nyZxy@3ZPKV7wzq$ZhRV&PK^lqTudbh(S= z!b(DsACOE#Fj7Cgr?{l1m{sp2SkEV_Fi>C5vsHwG9zx`AX6(8^BWzZohMP+AEeeB)i@IF$cK<^N}?=JxoPQbi^b_U|B1c__s*-$7&IZ|J4HWWWn1 zZcSMO<;s`S-o%Mj^=R|w%9rcW>`%S-CuP9jmK8DE7%5jSD{yBrPEt+Y6_zVs)q{N7 z-G2HP$Sf~168BC$5^@7TT0GjhgtT2^x+g)>)9&_FH(xGSzE+Uu0<%*MJ(ea*kp=4= zkmPxjOK47#FW%Eu<~)Y&pfy@`l*h5%PV4Vz7AjzfWPEA2ArJR$mS8V5Q$|b1vZN-= zE$nHrCeDNFw&$f=r~J`0$@e=FE(;Q%t478?yFA+y58i*AfHkqbAVxcgp`K{yE3hU; zRS50;98c$)c*&2T5O)#KY@3c|csi#aRttUv$gDUJHR|?$Qghnk%2ac2A0!oKdKexy z4X=zHdPmY^9~7*!AUh4#jaCOiPMRIZ;z(t?(M|;3H`tLedib@b#o2}fQCV4k+NG#r ze2okm8!l&Feqml~p3=Pf*5i8B(X!zJh0!?2ED|#s=a^+*#~cOO>arRIEmsK2y%J|p;CmmX6AfzK44`Ma~f}Ma6 zOgN1Jv=B%DW7Odh%~5Lov8kCL+@H3tb;UfbM;Z{|8kpiTEZYeR|1Ry(Tv6n0+dXuqy_=>3zw2fjol-w53Mld zUI<%H`5HWP+9-^+Hd-J~oRWKBWR(){fatZf2oee&f51DL*bvOV_&Wr1oS**m5DXm; zV{=)jJCcI~-d&X~*qCXhokiE+=CkaA-U06_=#0Se*cp(s>Vjzc`H)MQYHo(jyI$aZ z@*V3Ir+6aU{2B8$|CKK1jBD;q6xPFfoUFMDZoSF-Xi_9pf)!F_vPem|^~R@f;0vEf zsIHA;)^~vO9$W@(?0xD1i!5A=J5`}jLK^_vgArw24btUdk(O{)cLY9%XBS)#%C;H~ zPNhe17Ajn;2?*Q`&lm@j5Y=HPIIX?_c@3U$_0g|p@&koh<)48zs!+B2q}A~;+AuuF!sEBW zn2xS-1TFA0+P~G42y};PN4QoC5SS0oY`D7JYNpGLL3YcoNBh?bHL|V+`zOR7hqJN} zXgC!|Cc;^R5x5SXYv3|&-B~@*wq!K4?c#~Wmb1%ELexW$xsOV75ZDFJP6i%F;0!#c z;3U06vGM_g5%{zaAcV8}K&0$6Tw!ole*{LrGYk&D^unDg%Iu&3zAvniC`_x6ZKXqG zI>e{KrO&yef;yu5rCOh^XR?*g-8^wKK*lt>z!URu7CnM>Zkv^qev zJ)HL|dw556ITc|0KL3JDDiKO}5QDe%6U4HhHX6=4gTNekW)Y{$Nl&{Svfbq`FVLL< z=La{Hz6dla>;?P-m}y%$V$Z`p`*$Ru_iS!fiXEkk%}S3vqbJv55f+=3K12W?!|;|_ z=x^`d%Hx0i{kZNay-!2RYKV9jalKL2Lv&Yi6~xKJ;lJ%6V_gBF_eRd=aK7*!k)h<@ z4Wc*i-!d+7^ClDLr!priAeiIk{mdPfd-Hw>);N6YEUpU9NDym3RP^SZUC_IEyMiz9 z0N><`eko8dv2%M-FBK7u^11ag2mu_{_o^W!pkS2;(IWAXU_79FW5{J-nH_xo5zN1! zjBR$kjPgejkSKL3f6-3*;(rjFO%1y-h&WlOlRM$Frh{mmbc3;=ToW2Hu@8}T@*V{L zsZItQiA`u+m6G{TCqF?Yd!4ADS0}DubsmjfM#{(fO6){z)XDOTadq+&0^&3|VMq-q zSPeil$18E76(-O)mb)GK-F-ff<;FqFhYh)o{D*u#ZOa-lvq{*lmkb%LdOO&zchwO? zW>9T@yc!ld$`_PnqwX&uKN#y}YoE+cTi(F7>-M1TWH3<6cPX za!eY&)s9(e#JAcpOO5zeJ7%d7-)hrRqc|h~a^zcP;=G@U2f^1$kNZ};MZ1C$Im0EW zhGuzp35u!f>x7?y5G+9vC0K%D0Ow+a@Xafw!Q^5O#g%|Z$_e;D_SdJK+ z9S*aV&r*FONKhmCtVnj4_2a_EtXNJCqf>z}M5zu^BIoJ~o4ABZ?3GO_)|f?_ccRc!GbR{gI4ktl^D^@~Uq#AU+Sl(bVo zaGGT_L0Z9QUIC(t1*up?6)C*CVR5+;|0CgSO4^-3#HkQTkUQWr9|2LtAMpOqIIS-M zDO|2u5t)nsMQ}DH?MWcwOj+y5D)`JTAgXv6rnQVSw67JV@DqP@l0Eo81ZPvy-T)#_ ztv4O{5kB)gh$^n3;>W5;;SR4C#nEHsr5Kz|N&7SiwqO>8YPf==e;J zesSucZ#+lf4Iqja>jM)?-q#Y>az%*X6Z<`mY@^^;z95ZZq7UNK_4gACx8cC-)YAr;N%ty`QG1Hlknzb3w%QNjUvrIJ|;F(6&^h(kOFjvPz*+jPuu`!3$9hyv0nhA zIxw`gh)fPm{)Efxmln4Jx@~_Rtk2Ecb`_diYX`|EH zrV$XVCEb9kh>T=?ya}1$O@n(J@JnEYxpEsGl7cjlE%)tGvu+4Q7E$YXHZ6VAK-nB` zm)05hM0^#kF%dbOzQM)|5tG-=&T}w3iV473_*QyDkC8fnPnD?vbNK>$k zM|^(FR%ds+xZ3J`gJ&Rgx2uO?5ecc`xxc&JhK4&c*$7G3+0lJv92`WTd+D#T3#b`S zaS8x>7NI}P4c#ND0;OEE9!U)eovG>cN-E3%u3=aa-3v|YCtv+z-wp@&;xm1G8&SBR zoTpG?$SYPaAr}0*tMPIPCG{-TAUvW!4NKF ztgRs0Fb%W>^D)IbME)rdonn0rbB-zyzh*?HME0}yRL)mE=M?MR4BY&g!Q_otn% zN_$SRu7HY8v9b$#Q!H07jn+*jLeMP*rkydPCYWN)ujMr(r{iBDAWn}9hD?KkH5Wv4 zObaUF^hb-n1eS5y=f_$UpD=kH!u^EFMjr@1Vd@FR;1eeHj@Tzm+6}~c`)7Q4m>)@2 z2hk*Hrqqa8H^nUNG013#i_Owr&X4P#YFGGb@1Q<)pl?2^*hbWzIG_Jw$n})D$(KM!&{GtLa&B-hu7eTM50zhmP5v#l=i1pt%=wcc29q^BYhME@mbgyO`Br ztwb{4$B1(rumEt!&4rjoY2JA29*k&-M2cX@YgGW9oHHQK$+Wkthr3B)Mq;s>gxp{^ zi3T`!(_*#;{M{rp5{vf0G_X5>t>GZr)ZXk*cixObthe*#mX#bCA>`&0J8$|EAo3Eg zu!YPTr-n?y2nY@*0Kjs>f0P4WXF(s<1U%D#xdg%-_hCBo86Y6oMhDBgrwJ2{c)9fK z@+!<&SnV}voL8JB7gD4&WUMN_;2|~Y`oP*`=>V2F5(Kl|%jENS4SMRRM++Xtt2Go3 zZ_j!g#k!2v!0imAGayky_PjSmrs99WC1GqPVfP@p9t4L_Ii|b{pZOk$Ch}=}#dYt@ zA0i-8VjbPzQa;1~5jdL~_M0H${JGhXpW(Cq08s$7_3CVY-;kszD+EOO<<-hmz7KJ0 z=mlQtUc5N5S0$FZi0d=8$ZC8nv?{TFtIfJ803W`;kyw~kdG*Dkb~bR@sHLhI;R26L$h} zcl)@sbnD|1NSt)s4vxTQ)JXAguL|xbxYWGcfqSVD$e;K22d}!GD1v8debq&K zy&qs9a|4|BstXm(LAAvUynlcr*RAn_7?SS?X9Br~e|XqSNaY+jXW+*AUte?dZZZ@5 zL*RN$TPZTfVCUe8VjRn#1$!1;2JHsl=VcEssvww@FDW6lFsg$lgMWm~F*s{G0!^OA zMjSY69|BY0d6a>-5!eRLdvJB<)ri62%^2Y947;MW80-?WL<+%KtKtJ6BXASA0~oU# zf%WjLVc;DE&cgF6T-`N)l?$?o?R&&somkL%AMEzC@j4IAI*7ni@JwZ32LgNG`53Nl zWMTC{+me>6@Md6;QP?U4cKRGFU&C3I5Eui`2)M3yAF3`>gIZY#g$_+hSdW4G0ixfb z{0sz&%*DeMT$fio*@3!rj+U-F5w&t4KsW+~Bqyn%MO)2Xq0t%6o`XQ$!YHOMh#I5d zedc9WI@nnd8VzSPL|_&?GpSV#fsFMFss~!8w0r}6yuWFVUS)lRs0|QYN8#NFbUhm< zj28|N?r>9SI}ru$IvAz!Q@2G{BdTY*Jr>;Sp947*l)(4leyG1HQ%mGSYW8c8l0LtA za-&fGa$xgK*?9C`IqpiojCX@Bk1x*dRc1P0Y54+p2keOk??L!Jim$f5LOBk@>8q`z zXRzf6PG4e0K37EQK4?-EB zT71{2(7Hb%W=b1lJohPR6~vy6I^LQreD!FJ^wlO|`0}5PreQaNIFDgJ!5x%&z?V^T z+fI4RVxA;_p3grx5;YK6598X_LGXe_;F!5>ECR>OZDSEQW^Nmcz%;jw({UC!@;zat zxL1+{x0bQxJ9c1aA0!^BQV@&4%^;?$Qlr{>i@>h{Ay@<^O0WpbfH-qsaHKPQR&Nlk ztLmEaoL3wfNd63;PxIYcW87?h?Q{RMI6NFN;`FXwP@adTxz<FkK>_XnG5{QW3xQS2Ld1z`8g^s(9x#8$L)k>P zmyo_C;+}hzI1eVEPB7#~b*J6ggHqtzM z;&@l`wNZ)Y$?L#V$8hbFJDr+4@&%Nv-$ArI4V&Z9j9<%kzRs}KNVzi;r#~5~5K0D7 zA+Dc?G%G&Fq#O;c=4=eD$+0&ul)v>AWzHfRI zK4hjaD$qI%zxGmAtPv1A;a@Q1Cqx*B0|9fmwk zA#RV*?8U6$LWovyVI&CK3As1I_{A6EtLD`7NF6*Sg*v>30J;ZGitisP%m)ScXp8~w zx$o|Y@OabT8YmcACMU~ZfUyeWv~W$3Hx9*0>SI)2jes~~@NPRD@kS$GC}!9expH?R z4#j)(%Mow&@P)qM<}hc&azpx4`B|UOx5zB3c=u=$jkt4j^o0QJA2wwxNFzYSc zpjx50Mfirg)b}=w-gbn9H3CHIGG_EPE4>4D|LVbiY2%C5fuuO=Uc|ZZ5H+Xy@(Up2 z4q!hU6<$ODhOZEJ*U)P69R8QV+0?Knf{2q^%#wBRS?_^pf)l9|DNz*JKMS6|ta`YR zg)@9=R+LaVvw9&2q;_9XpMfR}xYL>vp>}+fY(vFTWjJtQ$1bDYN6CRv`tfd*2Pq6k zypG5x*5M;mu2Jwy3HL*TE|tji4^mLSE|pyB#E&6N?0G-22d~tVC-uC<_JzwZVLdOA ze*e5g%3yPGS!VT7YJcc!dm4IRBK;D1mGM&6E6VphV(3Q-#?a~a(J2G;^Vq^2*+uP2 zzP6{K==8&A<4Md9Mj^gnYyGULtY`GJXP>chCD`Fsux1xjz zu>ZRrIuqgU7(($lgrRvtm_me2F@$1q2ruUeVHXjM@?t60jv=JP zA=Cr=zw2Qj5uS)4MB@-9l53>-+u?punWsr$R~fWq=hVmr`R{&ca+{lpat-7dR=UIOuc($6FIOop^%I5 zVM4@Uq@j`0xJ0!g@(48YX8>o&v!(G0+}k&_2Xlqgi7gw*fli@{_$hio@EPX>nZmV zf&~q;fS2kij5IV7A8{Q)W{$=(Lt_DKHfH9d|3_w`A!H__TF%Tk;KUhmzf`wnW>{5x zRN6MH1PsTa^Y(_Gbr*ALj|- zK_XPF9n=G7)jmQ5?EkKZ4~Z}|hQL|1kI*Ae2+4l|;pG?tXVpH!Q+Yz@MuZbF1kS2` zgwOJXFr5fBE(_{`vuYoq7})<^51$g@mKXwO)jmSsJRuZ=$sFtT7y@V2KEgA3Lg+_? z-(v`zRr?5E<_Y0hBGjuB)B|VLK0+C=|GOTJ5Me?LfwO8K;l?~6l)ngswJ`+Fs(pkP z@`Nyi2NAra6G7*~Le(T*YIIH#%Dueys^>7X6ClAIDIIH#%hGj?aX4UT9 zGsmnt(v9~pe184t%|;C(_JsZnjQ>20|krGnIP&v3i2!3hPAr*b`+KsGtu7?)>@e?QLWo6FKB3qO)Zt zqVs1an~f9T2hciy_MvqK=}w!keyFmfAP<&iQYXoT)oC+fvvG36>a^L1b?1VdEw@2F zaUR0SViL&!v2JA5_tF|k87Bwk8}*ulgYUWap^rQSuh+}Wgg(OM(3J6VU{*q=u8c-z zeFC8sMCcTo~vw1#;aNV#63a&8?`QzBon4wu5~<%v%*{wOgO?7Y_D zHrC-uC}gk0-GpE#4QuBNo?>Ek$Vl%gCh{W(y($fnnbvs3y#{y`Q)yi$XipU!AN3=IC+1Q3Aqu~iN8vgY{vwaUhfY8^ zU!$O`|C;AVDO^8~!u2UUFpt7t{{Z1>c@&;P;f;9|ZcO1H@+f@ldk7b6oVVI|lEQ8C zDBPUFWAZ5c!*>v#pGV=RDZD+8!mTO%M;?WLJ`UlEP4ZS7b18gf9);Ue_})AU|8@+* z%kn5ZpThg|DBOv{Nlo)sz%xf7oS8@A=P7(c9)-J6_=!9UpZgZV>+&ePn8L^ND10S_ z3pdMKZT$TWgj?lNcsYf~<`(Ad@aKdu2h!9m*!Mf|OF}8G=SU=!599c9Xy`c3K~!gZ zGr`f>UTjs&9fkdg9`D$pUvalN#=Fffh~sX<`IL7DB&0+t??-;l!0XQ;Fr{F+ymeu| z;btL=_qDRk^$u7u_k4paZ@%#yB|Ev2p_rt9uO!QxZ+t?@k*?&)*u9cof4ePY1!!ci&2PZbnZ>1QlamV8<`CwmE%Q^R zh31D5UTqRmB0X4}Z4ntapU734XIb)_iO@&lYSXVHtxb16QE4|Mr@4}>I&VIqwdlnWt55o>b1QBWl8QD(mPdym_vp(X6AgEpyh94UWznE>%aIG$654+!@dkg1SYb z*Sk2rl%{Gb*DCs_9wxdvzaiZaZDoPz^l<~&oHNGT#UCkF42Vxb zTKAUe`K11#`Fb>>^TU#mtU|o`If#C0(Ys(8J$2;jtgxHNJaAHv7D|@VA6jJ?om#3* z2?nZVjK;o}DDhW)TSm^|KYy!B!Prc~-azsZh&U4$g`_q_%_~5XUwN;j5Vu&WEh&TrpgyawS%mS@*LjHlp zYP9A}Qo2uN7m+B#wilHukj;#jw+UezU;%M-SX#knbq3KUucRNxJ~cKheaRo_^N$bF zJw`oFMI_RM&AtFJ!H2%?k)qSV3YUYRTQBH2?uWhysS9K0(Au9(OF#6bf*vWlo$A^& zu4bvkde5k4-wD{ea}u$-rKj|}X)gQdc(Cm@{s+GcQ)g1zse-&00oSYsqGe*9d-yM46a9zN*v!Nb6ico3 ziSM&6B2ku|sVHCI|9HH-O$d862tKtL5|-2OSr?{ z$3FWydtXn)9;b;ud$Vd*9EHyL zB4YhM>#|=T3HExIjgQn9rrxA^KZVBq2wd|pNRW5G&lX0>BuXur-c(NF|2&*c1$#HS z&a@H5C0~1#y%31%cW96KiPH~#ITI}NDv<1bdGU4s*_TUeUnXSwo{Dk@{vW+0jLjtM zrX(FfU<49=9ClI+VkWEZ`Er|Bz zkKjQbTB9#lgqGRVR}D&>tuMQUB>wb1MWhe@Z;V&4iDB0S!3Jsc<=fyhr^HKWb;tK* z7m+Bt`en*X_+J$-Zxh14AB0wK!?GDZYdeSn{SQ40b`Gu&%YO2|@%c}X9}4w9jlNvD zIC^E74!Qbr3RvN-1JReYf`h(%KXoHFP^)h?E$z!x(7xQ6>T#Or%dJ#%$pCcP(TMf? zvdhjSd$P;czT8CfJ|20W39h*mM9ajy`+fOo#A0tdqN=xOEgSH^70#xDJs$*Hc01*l z##Vp4lZVmKAX?Sb+n zdVUQ6m^~B)A9)5dkFGucv*#Drp3e;2S}#M!;s1e4!q`m0u20eiM4bAIL-GoI<~ES* zJ)i6_)Bcy9ujxvZiM=u{zxP^X+gx) z?-r8yQ&zN=EAW3!yn;;(y9x+w2B7C>!Dl`kFOjR~yNE<7IIyBTkN;QWM+?87BG=j!>%V1?%nK+o3-4tjns z>PGfbt7kSX?fF#Dp5Kn@ahmA)g;lfL)#&2G5$pGSmwn(Wuzxw>_J`3gSE?&9_Q;(pJkM-e;5_xuS6 zd`wPm@G-dt{Ex{s5PVEdWapbjDKd}QSOcP2_fpr7$sJ_U`K6#R3B)bE`$L_a1jfXmkviB$lJPfWiFABwCPOi>~UvjBPp zgcP)DqZp|AJy7~3ay9E_G-UO8sY(9)Y)XAzYN7FWGJfO#vGyKNRuoO#cioxYon5%g z-6gZTz!D^fMS{Re62SzDiV;Kv0TmH2fdNr5q9g;NqM~9zM4}M`=71n7Dki{y2~kAF z{P_KjmgjljbKdWsQ+=zu|5Y8P)AZEzs4jiUrypSzmG(HNxdytk>y5VxJGWR9 zsCagLdkm8N&rufE!svu2tUx)gp7O2|(PC}0q zjMVbh4QH4%COS-mGfXUv-rtRb)g8RLS!ks5zeh^PoUMl3Y@?icEv^ zZpAV9alomlQ>4m_#zRe1;RBA%l85+z6be)zxE@E6Z!k{32j~5PQ*niI?!o`Zak{#V z)2ee;)v8i*1hWcMSxulx1e?{_j>D^gjMGN}_0GnrI8JMDzNQ@(XRj+1OIHdsv8vzg zFSql*tdb%P4HX5dGZ=znCs;rwwprf3kO zCSPeeBwsGl@GD(-p@!$?{<~XO+uAjrCuWb;_DX{C4?Bf%+A5f)7v3Ret8D5!mHMlK zl#SCWVB_=-70-}loPL=&Nu4*4Ze9Jz^f(;`Cp5Ta7;NM8bBazLNZ$7&)EkInWzxL6 zar#tZr9h>srJBhw{!fPj6$pCc$cQ!z^5s!D>-RIBEe!j8LE~sJ=`o<02D(>CjE`O= zQH=X0r}EU-O-l>;=5KZOfp)1%{dmNFEMe^AzCj$38Gk!bQw(3)yx#Ao9=@9R9Ybg7 zQP$!o$RpntMWY7?5nhB2sy94{P!&q*lpjH#3_5-X^$XRnV;-jxwlR<96VLwUv3H1h z2lZDKPp;pQE8~#%ufVZd`!Gx+;~mu0^;(eC0RCG+_@Ys|Ae;TJR*>$!{>YWrkov=^ zZdE55{PZPLZ)dKSWiKZX77!!JUltci!=uSrFC035SQYl-q6X&=)BIbG+rZg%<{eS3 z6njV1qUFSP+l%Cc#B-%bBH|6lF^L|a5{4*H?QIifGLip_p+J$ri8wNLRqH%yxFILx%qAF9pN>y3T|Km`gXh~K<^kq12BTg)rC_`8Z6gPB5p={%SCln|$ zSdYVV>)kgs#eHGKKDhtl=H{NXFwib&{2=%$xdU!)N{G7M-zA{SV9qOns!lne1xGYA z4HNA3H?LORKd+wToRHPzSoOT~LPz!VDgxW_-9x)c&XBCa#>DjNc+k@&k~2E1#)0@c zo|&C5Ib$M^9jG%dtHxP|J8^!t12rE9VthCuG!*=Bv+V zJ>+fvKY#*7^Uff0+u*$5m<+kjr#;0nCKH=oiw>^nY48Wkf|mjno%d~ZsSCUX6sU0U zgF1iWNOJzfgmfVyaiUZB&0o>?QS_7C1zIkB%{1tho7|5WNo{yA5dRv`*0|ofF-Z5b z_e0UJpAGvHbn?{&fy@Q$FLCJm5iXwmaFP{rcB7^0 z>Ch}oJ>3_X+W%0GJ5(a;*r*=#vW03-N$g|Qoa&-E?E-*s7E{dk=?pU8KkuCnma z8ea7{7d|*^pX)8W#q2!Z_a@v5b#Ot(z&N>m#=tnbwT^6uZY-380Q*PcSgPIjB|JBG z3K6Mibx4~@j16fjVPo?tH|3>Ys8_-Gg?jZ!_d$2%$?b4&LZ-8MS;ml7V{V>S>|Tbn zM#OL?>xM$vg`{8o_>4l>mC@jQEmElVFv{bGG|ik1X%n?voaXpcp0r1ScNmVnQMu~C zd`?x=SGp}v3p=o)pHM#`WxcZ7{BkYlz`}Hw0!{VbfD<41zo@coph|6=oH_xQ9^h;ZM8=;}`ZG?H;ZG?H;ZN%9WOp@KHh%Zp={|CqN92MAZ zBSsP%-9~Jh#U-zOf@}HQMqHu_QZ01Fe*odOIM)>1ZG`dBieGt2PC(rtp#E5#%*5JptWRGaP4%AR^Q6j&O{56gLpD&Q!3Q|uu(lEvn-zVgKDvQ zI+c)&!p39a6XE(KJiFjrJmFZFE~|7uo@rn{;$K5nztJqfQ6~3i7oSFDceFE;7G>9{ zeRmLk9Vh3Z>>B#gEuYHabfm@EHEojB?vb-JjFDu!*K_4R==2&7)iR$jL6$un-QRbi zU6IAJrEGdu+w#!E9@%wSwFhd8E6}VwBm-oBBnM^*0dhPU^em3MCaZR=Q17YlfR7N?6#Arj$#}OpE=y#%jc=(;6#%f@$elWWX||y zA87+-KPXU%ARmV}mZs2F@BPC~sS6#1K z$7hevqDp+^By=5M62Yn2 zrIYDAoHjExPktlbt1&F<>`xz1VW4=uZN@rGih=>M? zI(GzeBwl|APFxq#y(8s1jiCE8gTrFY%`Qr}Legpab_`fNbE_vC|FQ#zzCU@{+ppdg$--cj~-88X23{HOA)kPpW0i>#8!I*VQB4*VfCEbKn+2RztI>dlKL-h+CDJ z1^;bcpHGaWrliQ|H>y=)OR++RTz7rHEnsL4t1x+2MrJX=Pyj>HX+&zMz{Cnvc`5$g16-pFt*PK*Ko6;33 zb$10R+pbaU9w&EyV7>I;4=C1?2!x6_xjPI$Ps2A5ZZitsg?(G$xw$WC-d@*Kk|Q(n z78)D;XK#7q`=$LYZ(a#f{VPaW-n1~bmDH8E%u@0F(k3Q(`$uHvBw}rS!1qhT;D0|P z_}nm2rEL2kOsMw%T zEKaSPipgG6vrvrN8X|V}9;r=Gv3Yw<%|a#BQ^lU$Bee%q?9cIgE1QMt=Ww}VKLNAF z?ChQlU+(0W!`Uf|PkvF4JNad>E;CM7!8)U&uBy>?<@i>k@*D%(J5Tl@+Aqhku5osV z;n1!#T1+v_Szrz!+B-TEBLyn*jveiGSut)Ctvdm3-|j_#Ale`_X&p&S}255mUPe9{Ywg2aF+>cuvK1XV;3< z>!vdK>mVvtZRJ|;^82fpB$t=x$vS}k2RL?#Mm5muyK6tL?+=7oee#{kLo_(qp-Z0p zt?-=FBKTV+3O{>a;PnY}@Rb%kU$cJ|!BxFpb;Epob!1-|$}OHEnD#PYDY}R!v|2pgkShQG5tDT*11f|J-<|O%1p3{?#f2YfrU#$76 z)0TWanvt)siwMrm{e+CAzR}$MMC2Y5&=&;de~oQh{?OYqC!o8rpyT7E4UlSldZv^O z@`~MK0xC9v-+4UZcSj(efWq+mF|6~CB%C^9aGzLQOX0b>r;q?+HL1+VOeLIvrV>;s zZ%jbbEvdUy>eUKTmNzYoO+Zhoct)w1fX0~QDGSNWW@7CEG699bFW*b>uVJuFK!+$g znfEy9oI!Orz_C{IJs&?u)b>aAB{te033?zu2#^g^x-+@irspzDv(y0xNw?S%ZH{~VO`DC;|lBs=el??5V56jTL8;>N{&Q8eF2>7qy zSjHbjg8RI)a%H22|L(#|laZ#9A&L4`&x|zTX{oo$l9Gnxv+39H4C}9P0^StQ?*wD+oMfRA}D{x6MQ)R;Gc6c7Dcr* zfz;_L^;iWdyY7nJqfIW|t^z;*n!|aC2*hnN44UmNO{^efIaa*2)R8KkAxKND zWrEKgl_OUZ=^APndKjU1grU|>fBRSyX)U-3xlAXNB?anVq^k+X?y1TC+dC{YieidMi80QMODzgn(? z+!2R}GT|WiSfZ|Zv~iR_me|KUI+ft*pdLv*tlI9Vpsnh}!>X8)s+viE376%ktZaV)Iy* z-%d0SbsVfZ`a2!ppI;p6P$n!kI?UsAn8)c*4|@$e3uG`V{9!nDDelsbS|y9lPRN-A zB_49&`qU&o46!8K8TmgQ-Pj8)m@BfU@`R@GmeZwsWAR~#`kYCUXYrWkQ%&#}2TDJt zp{wd+8c8nVIfFMIO-UTf?Fse(VBRU*+;=35Hxu6JT!*irpCjGMTK;+oW8=AST^8H0 zHB~EYG%#E36n>eVUhS+|M~>WagX(dsohrK8S)_v3tA5>d)M)DjEIO6Pu8j})lKZs-XddYMAzYGP6mgH!f-@9=B z4>%?_ljetRTd1!3k@@TdATpvK1<0RRX+UXT+Xj?mkDi56Otilbj_EmjgXU{|y@Yfm z$UD@9Z+_bAljK#rf3Y5|{%=mTzJHO>EkIO`l;F(lvZKLR zV!^oVS~mS9@2Z-R?6VO|;aHATjOSL`1tt+j7noX7BfQ7#i$C1=*C5;c&o@pxN!Uwz z#=WFxm#&2z?%S>BftHDc`VyC^W1Mv%NtKQuo#*#RXTGNMf1@}u8{iIF)1%JI6|=&oS)Ay99q&I(xM7KhoDqO&3gknPo~& zx;#rppv=2UQ_wy3>&~qm=A`ZtyWHU(`>z*9ZAbUm+Y=)Nnx_^ky2}Fomt{%^sxTO< z&SV@(J*^eDQz01_XaoNI|HH+ID*h9_i7YucTKAm6nBsPnjJI{q={=Qa^zO|4A5gy^ zjx~~bgu8BaS|{%R@_}#;c zt6Hib2+u6lQ}{EQUHa-M_kY#w(w9g%GP>ZrEXh8POVkpl;w2ik_`hAE^N5iG&D{?> z`$%6TkIs}1RAEp}onjo8mh-6Msj#*LuC_K^#l}ZftUR9yY*#2Vfcg*OSQYz(yDA>p zmYT~2BFSdC33&`m;ui;O`bNe~dNiJ}i~r9PfAGMi*3C^VA?Hzcs1kU$>ZhL$VicSZ~vW zQ@d2^s0vcH=27e(Yo3!$V8|ocR0aZ}A~vkU@IRIiei`9z&2xgnldrDGmU{{Hmf*}x z=$EB5FUsvJnuwKSt6~dO`cFPd@(%x-{tG-%Wjrxko-JR&c{_1T#fzpw#hMT5CW#BF zH6;b{A%LaXQVXxw5Xaz`8(bm&pJh_Ue-|iFkR<0&#slHJ!*L8g(clUh_-7S4ng3Bx zpdd*OrRe9tc~fusf0bs?xtZ=}5w!BP%bdraDMBy(k)_k9s*hNE;sw(#lI}oa#Cd ztmWj68?!Za+e{s)@yYqk3*{)H{c$)}yn_e_W6RI)nlXsa0wT%%U@ubOl@2)f1y!*9 zH<23&@*Z;GbzjODaFTeWKxH}|(^*#Y{|XeSL@*ymlBXR}M>fKF-{Dve?t76EUA3-E zUKrK%yGvO66sSy-hj|T*|D1~vRe|7B95VK)NKL}L);K0#zQ$#M=i$BaNP!A{zo4b; z&;LPCpb|lS9NzVKTVxoVe=d$y({Ef1({$;^##$GXd9q$t1EqWC6SL(qaO)w1%z4g@ zj(leCi6a;t-Dmb*A;v~?{b=G`Jn?7t8Vt^6AJKh8RfWjLeaC0ZCumKyoD!+NG*-Yu zTTy5W7urv0qV5A!w;~uXv^CN6LR(Ws^>`5bq17L!#GdrRU04S-XwCm zN?htB?u{qA-0V`Ti)4L1J2aE{#C;CbE7ZhIF>~<mGwL;jQ~ZudaFQ)_tMZ%sh7MzR>Ha9!b5bI)Le6zyp znrIy>>(MpQH&)i8TOv%xW`=HwcD8w+N;XCFgHE$wJwKdtRLs_#yy~EV?o|ilZK651 zSTPk}bvQtkc~k!{vdILN&gQdzw{$j{z|7+&u<#i}>jw&?0L6Z59BYn$*RfODE_pev8@QF#-hgPQ#w$E!8~>D!{O624YTDJIIr%ddnR(nlm9J| zRm4bvDt_kr4ze$j2V_bIsxbIRohpRO=wo}-k?%jw&K-=;8-rtYIDIAEKW)aWDjJ;F zMke!r2^1(=3N*OW?_FgU|2N0M$+>H@Pg0rfw@YzHE;*^5h(Xyjd>%MZy^_xW?D{ZC}Ham{^Tbra%Qt+jo=& z{4aw76$>s_=UN;|j#-^0ufloTaVl2LUCW(L47n<_o5zS@R|tuT*J8wZc@wDx#nOQ) z3^u6qDUOW(OQ@Q@S=F0^^19<#S!Uc*p=wmc)|30lVf-Hn1&Wpe4Ssife;L95X>oAs zV4ZJo!f)%&{%jMw`%dGd`Br&JJ)<3M6S4b0 zLf1&uH;s=(pZ{yrU&~z=-S;l9PK(>^BAg|7I<4ozvO+HNXEK#W<3@ zdv2Dz2 z1(Vdd0!LC?wSOe8La|QS$4`kSqLvq>Csqw~6Or-JM5LGs6H#jPxjvtMK^emlw`Hpn zr7Mud6IueyBbwXCvjH9NKx&_LCY$2_%RZ~~tSlLifOjKKJknSN2xSh(zCFjOI>bnU zrn7t7fwF}EXELP&RTxZG=Smz&eoLR+2*x|y=#-BxiAt32lykX2sjJCf zGaS1>3fkiW(J3`l3RM5Q^?OKfz=vfD1gaybizA~Crg?nd&aZwFK5qh!X?18zN|Tg&i1Mwh0hy> zV|s63%PgbNs$$>$I?MU|Ukn9`mI4jFrEa;*;Qz)rIQe?AQ0~U-KaFGRep!?r-s=#X zcp3+@#C_-o-vF{<{lo{{oHDhc@iwPS?WVk>y5jgb*GAYWge78!uOv&M)Fjexk7G&R z1;WiJ2N9e29!I*@IoX$fU`|&}Itz5m+FOI0gc-bbp}4`Nx@%^yLM)-P0+ADB-}}6l zHf|<=J*_G_P9i}Mh%q$)l%HMovi2o!%Trdbp-e0$k@JQ?EAZ`e}~8W z2gj66QLX95sKPr(G?r@9n@CM4P_z_iaO2;6*@vLkaq!sNeKd})zCR(}5jdu?T1AZt zKYK(?8P5M`C{VN%Xz;j>4P_GlQ{vze^O&Ds$7|!*yw{`oIX#XIXr_T~el|XupKTnw zaZ|4JIHH#FybY=KTj#n;&v93Or$#qC$6Z;Nm41$^p|-`P&vDJ8iuF0J!X$OOYU_$N zd!k?f_vBSQ(tT+nOZpK%6targeWzVj9@Q<#YLd81v-F%Qc}$h4K~illd({f&Gk08C zBieM6L~D(W=U)B6)rsi;K!YM3Y?KM<| zVG&n*-OOY5I8S>uRH3ByZHA8G>!TtXK#QnP#?>RKV-;{ZfcVLH!tOIQ)}CdXjNTfS z^DNsucAv?!Z1Y%+@+@0DlDb8;O|O)kKFbaPs}-;~6F_zFsma%~OgE88Qya`quXV~t_~ zTD^IJJoi5|wh!8*HI@Yp(i+Qx_8A&Gg&O;HTvS{xXLk$yKi2WSF*9Hejd7ipenfE;`Np@8!BU*e{)nDM4{M7gK^Znv0Gb#!<+rL4A#p)WzuY zulyehuMS$7j$^+pm9wu2O48b|jXYLs(}R?1m*jh2707(_`73bZt8160FPAbLhi^%R z?PQgvo;8c&=;~+O?a{V^MpoYjn7k-67G}JT`#ro2aY!#2IMkofFmA4vn$= zklJ4zWU9j^_)(U1Pj(IeqzRr4(i((X;@2j`62Q~mJemNWUgmL@_}`IGWQqU%&*-=X z9bt1_ohXf(Yq;QQ@QY84^!V48`@r~1ajYU5+K^I`pC$grOySi$>9*^wpJP6pXb!_k zX36JZyx*PhL*e6aJ6>TaycDR8^1%m7-fV_`C{Tf5I}R@kE%&80Vg3O)b~#2(uNCHE z>#Ynb`kre^Sy1DqL$vmqwV?J!fi!c72cGsNv49(q(6J?cYSUX3`9`Pgvu|ogW#>(c zS@SLX;hPq3%u|JfRN)R)s25nnZ2UAvccl8Qch4&NCic_N@B$y0C^nB>;C_kv=CQ4w z{)xKRk__J@eb<-2Q0!Nq6V>XDF!_wP(Q|Wu_>_kQjcf2k@(5ry0a)5_W|)J?r^Qt4~rlYyIWi5M7XRnpgLjA80g5*q*ual5Dfvm+nOSgK(^LGeeBor`&3oFIYDWBicL9Vb&9FOZdD%lc2@5ZQ98tqD@ydSV5^RQJ3Em!JtW@_KX7ZQ5bRyb2%wawzJTBFC z1r?}#|EJ4kDF0)iKqZ16IFd}F;5;~gDvl-cmCB^f)NPT)gjo}Bd8k{$p5D~>Xj?>i zNwwF4tOaYAr}oXYv>^5vqPNB;t8Xcg4>gTnTpGuaM%YRw6%F6aMrJk;wv9E>(wm}0 zwfVTc7~`Wvm1na2ut3@Y^$)?ZjHjF4q@wo3CN|5dG9GdF3Dt#Pu6~~KIy;ZjEjx-@3+K1vQ1jp-DIx3TnNUS z;ehu6%(x#{1z$eN4iNtfp+M2R1vAH&6>$F3IF`V=xl~ndZW}I2s;gd^--(G$&O`g{ z?+D7jiv;p#&h|6R8lG3#)I^oq_qI$aGiy)*GvIDYT&5()CS7NeXWZ;F8xU&^fP;0z z;N3S7JUk4>gnL2pm4@D=)57%xhZEdIuBS1OYEO<-BrAx+1$1 zo7jRQ$uoGK@*)`TBL_@BPr1ZDK%b{bfhrk#d_CFD|86KyiQr8fN&T(4YJ7W?E46BI zla=vNu9WBCCrnt%fcl5wSgz8BHj|2Oc!pf5m5Zd-)}#hT0ty=ohe z^nLVyd%p+^^Z41N?o7nbE_G)@J(60?%Le`~wAt$T(8bZFiAGu(dW>wQf#Igf$ezaA zlGwfO^mtVi>_0#1$@U7*_v;I#yq9022}+7nD-Tpp3-kOM#_X@2em2XM_wj43(~Wyc z{ip@I7EP8Fh1f+-XXTvc8tC?+jkm1q?E&vTtx>kfWvluv>8WbiWG(l>8vlCb`1Z9e zK6&&JwdDh%{U31bGF^sT#c@B`i&tCzR#2TgB2e_=i&<*Sc#LbQZ@*xldMgd2FCZ(? ztcmFcVtmvI9BFARX8_yER~g-q%~G} zXGRwfZM=Pg@|yt6pZ|xKF))0pvZ-Y%bwC9v8yFP3hrQErCUEB8p7C`j;UB}Dpc&AXY=GI{(pl41xfPye@N&q;3*t~UubZJJUX+xloH$w3KS&C zdnijMIIky;!B021LRO#LPmbXK7${JXB**>b$w)YF9FD;|8C)T)Hs;AC{9geD3gWn$ zWd(8roWBIe;Je4A$GvNo(kj~mk<_JBioX_+waRUP60swxFZIgRUl4W2kBE(7HfjHw zsuyiHvN593;e9h3Bj&N5$;OCjY^g0c`NB7zv__$S0FGsB2H{-ijQi!(NxyKN#{dY! zzszZ}`#Ucj153Wl>G`N?O0{f5-fqmyn+;Y2r`OlCZG;#&y|eOsb3w5$DOUI4EM7Mm z2aaf##*XNxe1cPdwYtCZM?lWF9P@wu+Jy%2i;6Lkbs$KTK&vKQY9VXy8TCIH+1%W_0i>2FYB;EPo(;RxQ3T~5OV7<;MbD0U0k2FH z4AjDWrBZV$NLhDM>>k}|Iyxlv=?hfnA_U^@6owzI;cEzY-RTO2=jIl*C2!4e{J$%t zQs>DVvgHSY^4F1({HN>)Q z67KT$mzDh<9YhxrZ1tLcS#(mOw$26C{K_XKTA4?iAD;bm_UQQ6J%>yArHPtPs!>F; z`fN{DpvM0K$FA>+I@SzNm-bgB>i&G4HMeo$*S=u6;Iw9 ziKiEflZpaj6A$C?B@oSTS3uTpVlx3=ur@v#^lf#SoWH+Ff1v&qPU#31*u~2?50FOV zYt((E4NJ|%9FB*lxp{ti$UL2&!vedw!NW~7rmuP=`7_`fRM&i`>o{GkCpQxQkcO|S z7=FqF1@eN1f9k>`izS=ImflZ+)NDyo3%SbvUx2JYTC<7Rk<^WP`FkzQh>agl*zI1A zL0*}{tzYBmrB5Zo2NB13(SwLv`f}xuLkFPHsP{-DXqGQ)vwT}qxRKaBy3CaX<=;oy z^Y?2_LBrNWmziB%n`M=DNWD}+%9^EO_h^>&OrUL7GVw{6iEsuA!{_ct_|Jrg&2m9; z4TUG4>F8l@$8b>-$C}f27K|Coa1}Xa?m*dx_|{OMXerR(BStor{qY?T2d7>*kV?KB zf7LN%&BQM6h1=UnqrYl2GtcY~^VQQ$JyuJN{;pBuZeFBGOyMPa4v4Q&Uv?nw7q3G{ z+dhZL^&GHJb^g`$WW>E*_>Rh3#OFVGbKU5D2=28cU+-k|Tg9A)Nu48_>~G5Q4^DAj zYEqpm)aiT{PdKcslLm73;bl1K5ahAb)hIYB9|?AVjmB62~=A0<+B zZMF3~k}@xG_Nb9kG`-qz%FyJPPX|cR-Ki;GQh+I&+DoG7p;~vmNdfe8@_!#aP^v!A zL`op90H;ds4o$!cu*+-RrBv#b>@TnUQX)Nx^=~vtH{3kMS%t2|*LmY5b*1Q`<`S_7p;l$DWO}NUFhZsV@%$Ys4 z&v2MkU{*kV&dl^&fM+wIpJ>dLc!K$~GHA?SyBf>lIMqSdkm@6a_dkfn>Ob>kK_G1) zKM&7kFc(4P%Od>UBO6P@#4+9_xP!f^6lcu7zorY}$!t}(xK{s1PZGn`9u$vUtg_tXWOjFoZ8L(U#9>T^e z>?UF!0`j23zRZB-$?OnzsKRy;^A?bG3OnHUbfQ&cO$ggZVS|YI1;{R_)5Hu|)etr~ zk@BY!T#w1Kj!vL5uto+dUv`Gd{6>WK1h9VydNdDYv^hZX$){sjhPylEpfaM|R_6$%hSr&m75&98;P0+w2Mm3F_e(ex)QEmSe zLKAn1{0pSh)ht$1H^i*X+TTBw;3J412$in>IgP0o zqqXw%*XZN??(-MZJzfV^*4|HJiz;EOOm7H4mbFhDSwH41 zVNH8ceq$+}mZ7R)WeoYN$&w_PqY1b2y>tc}$yTaX3-FWzniCZi;rN9(=Tr1^p~m-Q zsBc_ZeSUtizYNF{u(w10lX$LQATj|eo0O6Nbfb0ouQIc7FcXA-CC-|=nU5fTji+T) zdAb6enC94@1LPmz+n}=FGD?*$pTk$VwRw=Or4N5)RWq7bi%KP8IPmL%sB^DKEvT$} z2E1iuIL+=b5IuqJukb^cwuwQlV$LOgP{UmRco3t2j#PNHN7L}umEdiU@%<7I*8`od z@R9GQ%a;~(+S{Gr`x8N|1iBn*-2SI@^|h^}J<&CBNN_Zepg)e^1Lq^8HbL=k%bs(MxZ zbC-t1V-0?m2#L|eZP!RtV+g)G+& z!yF!=vs_ZI)wf*#2J~0Ra$WyH?v)_Rbzhi1koyfB$%T9?DttmVkSMLdKbO$)z{WxY z@63RBF=Q#BONm~rkab=6iM9A4ghsN3&>xB34t2`wmWE^*#3qSig4-_=X$^IHJD!1p9s^JEkq2ig4TkX|h9WSBpFoj+d( z`l_Yf*!wk|NLD2QT1gkbCXk;%{|MQH?h?qZjIWYk%`#N~5!R(pg$(6X!n+Jj1Ueqd z_>H-oyOV+$ZqzH|k`}FL{lznscoi}3UdLaI=PpEVfjWJMovf4L=yPORZ4O2SW4T!g zM&tNHaPpSYGa&ycJl$YAK&^@x1bRowFyUxs8Lz94u`7H#YF7{^MeYVgQxB?k=l@Te3w!Xr|F8`NUKmgem}6~kPQH@!K{L8 z0C2*&m4w{@uyDD^J&>)(-h_Dr;@4^ziibs|WFTPAu2IKo-ek@2nMC~o>_;d#508{S zlu^8_N{La_21``(8DJNz5SawY`MMx>NhD&ylZc@Ap@=w9@8q6!uz%EStPAJcW+79*ofrRcLdb{Ee#8Y#n zNOfr7DAoWDU&Zk4PWAnZ3GGJoevp3-p5ZXVAbzcMNirqQ_v|QNb)Fz!x-PjA*fhx2 zB@UMm#kqCKLqtEQ5M7rz2vaT)w=Q{$=yg!}lO; z{#}#S!MWyRA~}$ag%0V7yvR`xSDW}Om3%*_^MIcYRq)7|VK1`Yi{wH6nRvRuw1Zj=`=qrz8p(LF0GT?SvX91?S=T zXW}dXeJ3QP_is*%NGu#>)5NQ5n$G}x4S`pnU==8YvODf_jw_5rHQ|#~@;G2S zklGHFXLCZR{`#ZAE`yEa0~SYlo;0MblwTR?+`Yvxd47v!ZUy>u;!x^%1La!sGM}?L z3{U-M=wVQw8W-G1H9nG8WepjdNlf2ok%T%u@&wxs z>6!yp($a75zfAC0;!lTyr}40tVAX;QoF2wJ#+BBz;B1f&r@A^DR>%=#Ap1dM{$bP4VLiDN>_P35YhNG_HCcl$2eK6M ztK!)L^C>hi$&%2?_$JeCr0+K&GiV{bM-|PyE})A5Z8uM81H^UT5#gLA;n_1!TX(6D9r* z5Xl#~Rfqhoc=m-UQ_lx@j)yrGihg5sEif!flAXi3Rvn{WqB?Z5ZYszrkS&TG49kCD zZr*=@==&hstaFeogT$M4ZxO!^vdy}YaRG;;MmOsoev#yKv+fTtzd~iYS?6$A1;r9! zuH9x`w->n^fNb0AWSEoGqnmYC!%Tx1nRK)6;#h&>+zLWB>webcb+hhOBwkh(x>>ga zW;>LzSvMnAGh1`+HtXuX#Em3mxpo+q+Yvg5syY#E*7XJ22eMqB4l_nQn(J9GGa&m7 z(=O)flO7wV9Y)h=@6@ZLWARhSJOPE>Y?xkBH*O|k&XpL!+Vl*&yju|Z6tX3r6Y(mE zxJw_tw-Em!G*Aa7mu6L&aEoRz9E|pRg>4IH;AfwwA^EaF zA+I*~cM>{_=xd=)^}c8=??pLhLw}wpb?bY~Yy^K!{AS4SjHlVFtc4)|06e2%MnD5z zj&{{9Jh5$9KpV>si|1a)g^`TqtLkO3Y_YNY9uzEqY%G5R<~7L1awnB7j&t&EEdL4U zk5Dw0mua=ScA$6C2@}Xz(^&P^eT}{WHP-5NAib-u2#>3}H`ty~`TPi>Ri0iPkj9wP z-f1S^Mb+m?Z>o=dsPETO#Pf+c4dKyH>A?)P4(7+4D^c(%ZN3a#)~|I*Y1O4p}?kKlv#I7h0A6`mT- zDpz`K)u0Q!-wLPTbq=0}2JYO!f->GKU8A@&Yxw>Tjb$#;Hz`E#@*ITeT_Ga^yQy1E z^pgtFUhg1u9w6>UZY$B76{0tC4&occ_3Z37C=X=1Tzq_;n3kYHzF-z7r_nUvw_ZpY&d)jW|ew$INS{L5flyLrLh9X1*VAY zUIy_eG#t>50NEhk7p4!CF^IQ~)s$(@-5@>=)L6)J?J(3~g!Wxh(I9>c(3>I4 z^)oQ5)uXxo0_HO)`_9y`*S^dFZ?ewZ~E$tc03U zVH&0eddW4I>&=Jp-=H@Ci@Dw<#Q$HL>#4Fm&Gn`qNrL~=T<;yE_B7YKO{FTE>rGl8 znd`mUyhQ#$t_lI(f0W7n7&pUQujNGtNHMW(93rs~Ts?z@c6jT5=_Ne~J32-j1ecpT z5I0Y?g|HL8%92wFI>W*KU(EFeP}63v_u8pB@)R1^s@CUls^rc$R$EJGuD62Exaf7; zLAZawTMGFGHcxR@p(6;}tGV865PLP()0+A}nd@CQrLHxV6VUFTh7+fi_c?kabG_&8 z=xLc7bamK@_hPQ6MSCTRPRQ2rED0W})shPEcN23deQYwc)YI~D)8&!WvWdD~<7>)l zu9t$@4CWK4PjRNF1)kiu*fECsbi&gCraffldTZlUXP-l=j}#uW0gY;|Hx$URkeTZ( zhPe+a-w@&di@BaF1bYOk4s($AP^ad4yOI6{GIKo#p4grgT&@~M-e|rF>0>oFm*$WGuOL-_(_nN z>p4ul7_F73KeZh@>|GYhvALeZ7FELTVXo(ZC6$0`t~VE@w!`gU^((_-b3KP?SQ$h9 zUex@h&Gj6(k(8d#<0$~A)Bg7`*IPi*?}Y3Lfs<%lS-msY`vC0AkUb%I>|K%DAbUdK zgwxkFtT|@MOmn^OHi~=$@w1(9C$>-JG@ZF#^Y?fI0J5EMhfiP7IGMee>p66bO4^;d zUZ3|lyB;!gJqK@D86KJIod$HY!qr^Q!CS?g%ZoGDn+0@+!qr^Q!CS|igFAD*$AGR< zxSH!Zcv{fGow?qpKtG1eT+hMV#)1yM7jwO=59nY}{4;aC&C&2puc5^-J}>-)(MA`4 z?ZLK$%v{gG5(bMGd{+~HnPSvj&tW{n#O8Wii2o2Wb3KP)oF_R9wa(Z?a%!&Ec@s-) z$jtQ|EK5o!0mBkY&GpV9db~pP=Fvg2WdaaquD6`%B?{4-M+eD~H9*|WSRC^Y}$vi^~+l4&WmnP%gcBTm`<#7QK1=Rg>47#|MzO zPgUp+;5wMKP{z&UvRKVl&AGdI{0-EvkmcH8FsF*p?gqo{05-E4sj1ftl;w3hZXc%=H|E9oz__=6W9z{hmV9T+cz+y;TT}M9uZ; zeZn(0$jtQ|gxy;pvANy}#2=#=eSqsQ*&)W6>)lHH&5%8?b(owO?JApReR+dEq_Xva z?FZoBgTe>4zs03=oT~&C+NnbNz&86+k$(`f;n*QPaa=4@!Cdb=gvLQO3_GY_Nyt5@ z-2(JesDink6U&z2RF3B9zK_Y1K9tSd%zX!B4`mO7=?B@A<%Du7rKJyL&0MeGy0B#? zWSQ!9=6Z8cag)=z2IgtV%=MgrCp%RDbG@0G5jEHQ6U=TEQggkoTexIUXs+j^xVBM( zuBx25UQrz0`$99<>+95>iFZ$2!b@&Rp*WpleiC&Go*8`5Ho6 z&GmZ6MR#0YC>)#X6>P;a0y1;G{a`vnX0A63W+=p(x^tvv|2CD(^|G#UmFqZ_(4Stw zDmr-UqwQx;bpG4A%*PnZm{PR{GR6(>twFq&rteDd%{gFs$u2=jSD|5&u zbcfHbjCYT=4+7f*6 zpNGs`&&gmuip*Zj^*Vme)B~Bh-bk1skeTZZkCJ1$Qc(=U7HuXyHP>4QW~~aTx!$iZ zKdX?M>p7tuDZPm1jJe()pCY8@dPQG|)P>Ak?*N$Ikac_~6vX1L=&`xpc(9{X_(^)` zE||@b4FFD%Zi=C1jQwyJ3ET_?fxhC6N#Vfo2%vxaL|)HP>shjmr#W znCrPDvMME}=6XX_QqA@5Me1%SG}m)UB*bx+JWgz`=OX;QK`NN*IVAU6)GRgv?ye3FVoPGuNy06|)6o=6XG0_J_<|&*7^W-kIx-B>EJ^tGV8E zm}?+@X0CTfT=^TKe7&OiQcI^OL_S1o+$yQ`^N;@9> zu~0arEs4@ih~pv|8-3nYA)V5$M&KgITDC)C)ezTDl#sJQs`U-#9+0(c=dqUUMEpv^ zu4Uhf+$}1rmgvrdelyo|;=I#AWoW?;_>4U26t@PMr&Uhvj{bz%4cQdugmNmSX?90L zqrrJEc1KUe)#JF#>Ultu*1@^wxAa>jeHY0gy_j<{ZY=x`=r>R_I6Hx?N&@lVT>U%V zzJY9TcKGbdcsDp72(~L!Aw%r3?Qy+ahCTwi3DTdL>urvwJ;$ZOSnHhIwEbx(@U{~Q znYo@rdg3@I<2pnupe+<`3-u@ozY^Saj-!AMg35P9(7l@LxkyaIB3R3etWMO~M7zp` zD4GmeS8+&8!y=igk&L#e!@odAyUIP_7eHZG`8|Rr#Bq^~HN*d?kam?f5O@u;uHukh zWf|91c7fdqSy!p^J);+7UBwCem4sba=>@b0WL;$p%qU3ntLA!6JUiwx&Gjz$lKg77 zxgL?}szbZYDwq|JbsMK3Csq&_Sgl<1qxTH<(rsK+LL3)MQooxuY3(*2lhOx}wNi)l zD#^H3T5kttJ5bbZoIqAAkQb#Acbh(-dqLK196q};-gTSP!H$6{WGJT+-eqVG&{>fF z)+gpgoy$E7*_*kZ>wwJb3}38v+We@yIjnZl=6Wxp@p0%#h3sLjH{=Jb1|T!nbIS7K zew_^SqSgbYUs1_wu6Gyk8=(rG&|L48cywDZ#Y_%5e~plu>vjH-M|O~z>kWrF9x`*i z_aYfjwj!feb61nn9#&j$fPM`M&GnoJw^DnIKPvvbW>if?b|J76$}rcf9ly%rxWcG} zDw)*Ws#R8NjXq zbG>Iku67z7j9Y48vANzh;=f3149)d=gGig}z8+5?RO`1hgN-AfV=Y2giDc zhk8Hyo|Yb)>zxO3JY?p24(rJpuoy61_#JsrbG?s&dJ=(ix-taFeogT$M4e-Qs0WSe!<;sP$v09-ZPtwhdJ1IQUf050tsdR>S`M=WVoXzWz4>vHj;k8!(#^ViGnCm%&Cvv%3!-eL0Cy?rKno*rA z9E2A|W5uC5H#AHGdKsiYGuLy#Tq#2<6K;>+TMle7WafGfmS?cohAi(7E)Znq zdJa=nmRSZvbG=iD9--7H&4#&N1GYtbz!lB)#>UHC$AzUgbG?;XSRKpXM8O-7jpe_Av4!=AU3l1fX7vR2H4Y}@~0z2#l?YMPc_%8 z5>*2B)>I!Wv)g~Bh-$8P1H!W*GuLx4KNfJMQggj)Sk0Na-c#V8fI@RUCzLH~(5>cr zTYjR6#^!pTgWL+4xnADi)Cd%s>#0CRb3F|S&GjrU(_F7RV(lO^*J~a>2N2E%a^`w# zh<;Wf`ozUSnBEle2gZqgZM#6^n+{=9|toQ${554#%d;L&fOq>3#gkR%eBLJ z;y9|ye(s6O^FW`2EZ1MdY*UZsI{#mmWKhP>%=M;5=_bT+g%Kn+*V_-V&XAewIg}UU z^Bo>@y&+r*Si|#O7*C@-s3QjLT0XaMOQR3 ze4j^6R7g?8RKqMJ2>> zG(`KAICH&EQ1KCD=6Vk0RmMl=dRYm+v<{iMo`d_9;CiE|=6Y>_wu0J!_VY8IqVYzHHA6lCUl4w$GU5{m`TLR5J+ z+yKK7fc7N||^c?_V+mwwwq+7Pb?kOhjC(Jx1`YRcpt zjuXpDfx-@6TP!E@ucniMf@HK$&h)Xk9(3bP$1#zO6_ZnZ?u|KeJO6r+S)gc1mLjVM zn0e3PJnCKWZlAQY6e#exvk#E>`PY-pLO`GboGbZCzUaAT-Y+_th@uL!b`mX01dB@UtZ#0ZMp&l@nzKK^F!Ax^3&(?MQxP4 zLL7r{o~U_Oh@9M$51=&>H3bV4B*~AS%@?)*@_OSKe4W7+a_zM>8DE^H)A`B#IjKtUKW zKb0?PspYN1F?fH2D`eN#jbtj82k%^D`dx)EbI8! zW99+{NwURb`J%_mdDr3?{1JmIQBEAW=0)F+&s= z&7&HPo}3KS5XF{5E{aDqM9~ET-Q%`pkc;B6AyF)Y0)6$&ZgN~ymN+gu${|akK+lRQ z-xg)b-Q1j08Y0H%Nk=Y{0d=|McZ(KFPdiF^3CV-Ze8Rz4h=nj;%_oKteY`@{e8NHe z5E7YBtR#8~WabkN!u*Gp*nA?_XF7w-e8OQUhGLxg#7V>-2iXR?!{o$?x+z;X&}aQg z33LPf5%3Q~;RgDSs2bR*kGQC!Yy-XhUkK?2`iBU-3)u#`L$Xu9r%VO&i4Iu_X${%N zxr6$ZgxtpYl|ZLK70f4`7*90sTg0)qJ8b zNbo^FT^yMCMAt|rA&zr<^zO8yu7JBhW9e8QVEhEO}~ z$i*bYaZ&E<`W>uYNuM+O5ag;$Md>8!~0=PNsdvR~@Rn zt8pwpQoZ62xt1m0P zRvU8?dh{`;G=EkihlBG+;=Gm6)0wOOd&z-v>vbjJTOhY!C~w(&V_bbH9L;0spr3}3 zlZl#>J2fZQYDy>bDJG3^IZ=(0EV!|%EQU)|m!{!IsMMD~?Lba%KJ-HN&`kL|W#o5P zHSUiSIU{mfu2%ttB=o=Xm)VF6TZ+M(&5Zc&N4H)T=78iGX4~Xt}u1YpJi~6dygKg?hT^3m>6|>XC-j z&2J&G3yFkYp@bf@Z13|(>UY(Zn;`iK((2)ya*g7%7aH1@&@hF3uZ5-%Dy`;}KCjpC z1VDD4dRGeRJ{x7c9S}dYoAQ!;{idoi2B(y2A1CDjIDa{g$v4o`xRZUhmT<^8;qSX}ZE%+BR-1w*_Nr3K+e~SG2T*v< z*(0UZ1v$;MJ!jNHa=1>E%zZo?TTM5R^FIg=>>5widH6XCG}b#nzS< z=5*DzRBTP|zOeNa*Xc|9yDx-wlq-8k&VxDigAP<=yW`tP&Z3;AFV<&elAHVID!Rlr zt-&wIkb6xq1|Adc(f+XoJVJ5=s*TXURd2L7_H@CNDF}FQg>_v=aaaXH! z78TCd3b(VUQt9KI^op(~ebq{&TdVXGCw+LAw6ww{IsMnFa-B*qb<(xESy!XuPV<@{ z5qd_dr|T-PlB~BSACGJ0^LZ$=M(_?l$8 z2XDtCsrlz4(KS;-4#<#bKy=*aPr@Uq!Qa!N#vsu{4>GszRi~_B4;eyG{{9D32m7{K zklQg~tR$D6SY6H|%Abm3wfq;a>J-eG^Z^(vw6GBaZ5JC%~ zha#aj=@%EZ93L>Z6EQ5Rob>R>a=l@0s1%T!_!_ z|K88%%suDKIcLt4otd4TomESOzsm5x)co0<)73^;_A`e65qx~$gTOVNyu%_C6P>%C z<2(lb4*;%cxZ4RhrSZ>X?(Y(FRl{St3sH*_{C{ih@oGT_E77UD_i_3EZe21bQElr} zSuJ7ooDV#0neY9*W2LI}WJ{F-Zm?QDU^VTE-n+JaWo0!B-b6r6yMGnm{CmVIzx^Li zRV`(yOkkS%5a7JOJ+%YI1Hv>P1Qg~e7$=2kJ`5=8)l)206_{o|5GazPsmjSyWR&QG;eTcAjE+K+>? z8EO-dchRz((59u@ZD0%8$4O}yh|Q5i?MLQK;4DJrP5i#5x+(Nzu4o-js&}1eEg`B| zX=Y%6I}d_Wqk_5cuIpm*+GnzJW<7=J|MEVts;kQafdussrdmE>MO(X=`jn_^Ff#*i zzJb^U@-R>&Vg-%xZ&&EVQE_5`Rn0Ag?RF(7*jBSxfZ6j#uJ>oGZ1-u{&N~F1qL%6a zxVs?=UkFT9^WX4wW;RUHF6P-exvJE05cVwU<_h)obRAgTkso{iH^Q0vf6zAeEy z0Jl5QT>)n=#3YajK;A~DIJHD|6#2Xd+{oR@`1=e|E0DQWlpcY28stf!;BcH02T`Jw zbsled{{;Iu@kd4BYl!4D$_prXjS@C_a(#tHyzP7#c0=Op1J3&p13>x%W3p=%D@0Yp zO{B1-(~}XrYEH~`h|Ch5`VbpI)&putndQYwkuW_|7gIdY%2b_rzmC`L$0cbeo_iCY5N{E>><)(2V673Of3pjHi#)6CnxY3FC zIJJa!){?f`+ocFE(F)Yg?{O8Jrsh&IQeju8;AI*`z6-fW0Ve|S3CPDlK@$j5{;V{i zl)st^>kd1i3^(wA(;lKONFAWyiZ}(%+ZTlt6z0RekoXHk;TnjmL1qGZH^tR(d7jgw z@R?=*Q3|*2B5EBncZ<>rh}S?~0SZ2PJXf_0FP6EYw40{>IqYwV{~E}7^}YO1hW<2^ z_KP@wzm?#=he*S#-!GQ4OUfOu@2o|lY=))E0M4>F!d(C%za04lwkfBK5c8ArbL7bP zunVzpo>-ADWDf_q3}`s!^Wx@LwuG`Yh3k-5gy;g%y%ORfknMo}<;~-%=-?oOTtyUz zRP7O|+SNpTh~WF8xfJ3bkiUS2FUBdc`{1406n3kG9^qUd$XU4 zPN5GHy#SrUOpq%9oq|!YLse5KCWU(ty$kTOvWgCPc9_Uy5^4FCm6KeN!FUf*&eH_1 zBmD~CY==lJXDJJC?uBRzQV0}@;!XB9OLLT9S?^lq-6}L7H%)>B34S5&Cc7_vvX*$o z;_FQi?yVBsD8W?(H`!1AR!vK^-tA<$iCl&5uW;P22%N(N*OJ_oCdu~c2)qHc8z^Xe zs+w9Img`X*X}!y-7$OxcFmCR<_9lu^KyQ9$?pe;w<5Dll%!*M66A3}cPvK9R-zPkEJuky zEDF-GxD_px2i1LI(WwEariL9`MftX%6TC-{>cm&^H7AWrz+DT^aBf zfZPZK1HRFYl+ZQ<{&s{P)C#OsN`IYp!gtULu?m65a*bN_C%XTBAD#Vx?!W&6`2*1X zw=uwH@d8tkz<}w$YgA?i2?QfVoiN3A)P+wsoH>c-pD1Yx#C%{OZ5t($BKwJKQ?s!p*ZIs zBziNTou3doe>C3t&xk(`Xy@~)amfSP`O85@13d?x<9v4P(VA*z=)A$w{4DB>?BnBQ zP8G!N!@xQq=hWz8orF-^=;Xg)*KmG8BZoOu=&lEI8jeigg=J!LWXCSv7t=&J$uB!qK-pbb#AK$7JVD_et?XaJ`k#5|B|g~*0D3339+IWsI*-5zGZV5V_q$!@Ga5%FsBl}ErS zH#}E$136!aWQZ9c(}A3IrK^XQtPnGnbTi+M$X&>;5#^U54uTvI;&BMK7BfVk=Ust6 zesfm^&v6u^;qlt8l!|P<+6%ir@_9h7*C-+*!x$klY`1zJ?iECj6_G;nZA9oj5s}8y z4enN=Hv#AV^5iDwRNxI=5-KgDYbd_4it%kfAb>P z|9$hKCH~Kw7m@g1H!saduD(f9-@M!jbvvMMUeu>1YYn%Zd-L)YLZ1Wr=B0KWOXUJZ zBByU&j4FE`M3tMDX;tEHUPd834A3_(_kr9E=$jX#$(9H-&%Sv%g3!mJB{wfwc`T&> zdg;n&u@r@txna4O=>CAdVOa@sE1+kxMm|Y3-0ol27iZD@WhT2HnZ2SUGuiaITycP& z$r>fTyogdblPw~?7ocad%Rp`x1zEZ>3hdYx1v8V~OZ=OnAWK)J^H~oD^eU85NDfs^ z4KOL4mWhWlDeZxHH$V-MNvY9E39(|GPTz};OiHIBKN--I(lsEf06i%+`d$frGbw!m z(dR^0CZ(T)oCJbtw9)1{V%S>z^qG{_kTMmIOV_iM2dG}zaURFiTq%|%o%CG2Iq^*Z zJy)LrGF24#={w9pfpr=bIx>OzZ)F;kEL}0hWLKFiRT&}L4nlHA>B>@-D|eFm4$=s) zsYV32gLGwS%9SN4eaC17T$R1Uzhj&z$;cg}D|d=|>B$I1)MOE2>B*I)Cp}`Cxa80o zzci3LYs55irAo>LBjzJyf2Wj(EIAn>95MBhlQEJ|(g=5ht}HpZvecyS{)}*931RBl zl_e%u?)vmkFDWe!i+ml~lp?Pr*RU)pxw4d`r|(9{Rt7Wq0F!@LCjWZg zY@!lWJ9c4-rxmMG1~Tc_uW^|8BxNv@E9TCw9LSEA+&!4+QfhKwX!ySx@bn;6z`6SQJDWa}(6uL1 zqB|U!hnyQNGZ9y2A~`Z1o7jvHGiGFLc4cgCC}XmbD;r{=-0ae>jL!{aTsC6mLd=LU zM!PabpZm-9s?<>LhAqZmRRXz|KUx{r@J!FGgqYKi;A5`959KPX4pFZ@%iKWD_RhKL zL}=yX04p0$vyAc6JdwjnI># zwH4xPkS~G4uY#1$*>Qd}g%d-9J0}*E1I*;T<>OWp=UYV58t{A?$mvw3Mkr;6n331- ziC%(8b7Y$W?u8KiqP^uIv^=TOkFe!&Cz}was#*=o`FjNp{>HQ4?mE8zWF_(C8}J}k zmMr~;qdaxx)9%NutXe!Dg_tZv&5GSBtHvOWEqM@1I#VQ_tI$e$@QX~9J`Xh$qJ}K5N`2QNS2fX3 zYr8lZl~7t^4IMf~we0Go)oA6YrHw7M0GM|Dx&+m-o0BT3{Q%<|Kt#u>mfd~s@Fte( z2TU8*gJ%#ueP+c1o;d({2Y2zz?~tD-_L#0^K1qVhYSX~}CTamP^8x33i2WdYh4=y@ zwJB3dph$GOrZ7fjFW{Otwpf@At@_sU(-PC#*1+!Vh;>G@Baodvm(#!lNlqz0S7(Tm zy&@N#`iM(g2(CL_5b1}`g+NZlIJe!B@-3VyPD4bl zMs}v8Q|nw$BPEsET8*@)>rl8G(4M{uvRf>eXGpqcwuHGX#)6v3NUZiz&7fw!Lh%%! zYsN^j-7wS+bjV(P(UF=d*PNLwplilR+9h>N&CEphNibX#^*6}2XxI01i1((Su?!xaBd-)nz;^zSz*#n$a zGqIBLre?lF_8Unjyk?AlX(g3R&6I1&GcrJX+6ts4U_4cYsmJ`59&Y-mv(dk2fSvc` zkz9S6o-dMp7u_a3bABC){WYRSVQHvXdk3Om2h(IZdQW~-AHwdcsaF#8H7^5Ep~=2i*P;Jr4z@W#0$Cyi4hePX@fw>H_hNp&+FJ zQ;%DSa^?}-j`nh~FkLv~2|k3{4-|<)`#a%OYsKXOKH?{LP}!3}O$+t3uR)_zvVN zpkPPOV(m0v`w6S6^BwHEZ7fv_aE?Qa02wC4`w)vkZUTzL-}ZmPc@^p=ZirzZmjb@Mui)GTwFW4Z(-Gx&tH1Gv&#=g65|?kD z!#B=~{IlgHI`6p&Y(M!#P~M4E4ezELFG)y0Q;#OHVlP!3L~%z?loU`wecU!|-nkYw zId><*c85vOkDy5S?XLwpf;S%n1x2`L^h)`9lEVeC>$bC0dBD8}!kIvDQF}{u0<@h6 zt$SfrJ+?EOO2dNl44t#~3k?}L2RAyg{Hbcw3`^R3do#6T~^PNIB5 z477(R)rqx2z-*l`9!GfmuVM07y4?Zcl-f>dqQP+T2gMLv|2ju6!d!}WlI zVdU1xC8)W`<@dDGoJp`}5>j%;af%W_|ewpk-^ie>Ui6KlbRQ8ovG)*L~T#ExR+uvoeoBB={6<^t|~2)zTx=-MT%ob=~;b&;m6X zX3YSZ4$x|3)@e*7gqU$y9MZGSbx7YW>N4wm6=XM1Jc1evi6t$V5%en*PKgB>LDSCX zz6uD(C0#J4duO>U=zEMGaxbufu3i^RGZYH|T`+?|1_>brGY@1ApbN;@uuE9W4+`i3 zM7M~pxU>&suMko|KZE=T@Z>?}x5iXLh=qVK=F8O6cbA*=9}@l zD+*lz?ZgC-@nXTBS!Ay+%OixKEZh)_PyB==z6r6Fs4W9>q=^}wNQfDo{CuamF2zrg zJr3wnEY*#!1SoD^#zJyQ3qk2MK)xQJ9q$j)R|s+Z8j!01erFY5N(oysgb!@Q5H7`6 zve>*53SqHr!2W%bpw!(PAbWtI*z&evP0s3-A5)Wc%oTb#sdXDuRJ)1#9?@?^_i2dy z?u-LKk%+b52WK!;KcL`9oS6h|Ccmmx-uW5!UgF;boNpjX^`P{CoC^42(oG7n=Hdpw z{1}heY9i7M*~Y-xL<1=~xHEwEg%HnzJOaosFONR?!LWzAV4a_((gkFxH&HEm;xFK| zhqwvkDj^CW2KC}e5Ack23Dhza?{1Q-+7E^(-`i3tK;C$Wm&3m39c1$FwbK8R@>)yO zUC68foTU&SfxHj6*F!j05|r;lp9~a3MgIadDkw+x0zuD$&~ilGW+^a z*l$SiEJ2Z2YX2!%q%W%lK*8)D#d=X+Pi9|_z-~i)KHzSGaMloX?Z?mzXgd#DZ^6eaC$SK#9OPJLVwW@}!!i@whWrD-*@R5KK7{+e#KaT+9Hd{6OWyh= zH;n(0{bEn19nMAUr4p@pLeD`)1WJ6jB*c7=%p9aMx*Y*M2e}+%ED+2=@?<74Cs4Fi z_I>`es$MhFmD$8{M3(@)WHw>sIMWw7GlTeu=zV~mK^TIX7R#MN%ge+vbCdYXlNP%v=%qqskT0&f{jay<`Sx z6j+SHV*aQqdct=j>CTt5^@Puu&};ek1cgdB6<~Qme9PWWg5JdP6q-)}{w9_UftX(6 zMO1HM$&*biCWf=hC2cmYT)>K_-neoE^-sisaN|nC#q0+M@?_(R(X_OttZa7>P1&$g z2bo$x@rIS&P(1Y%FkU@VTj)dGXsQ{8ucS2MT0TZRFB@IkV_h<5Kn?0mZYy z*TY;FV}9j@v%z5~3UjwpAECj9>$Pj6AZL1f80zJFs|Np!Ul%5uU z5f*0lO0K`QPPq=P$-7L8pCjRC#1okop8z=q6p5Ig7N-womJ8@I@Wo!py~(OY-%Gur0C-3skogKM!z{Af6r0Oa*X%L&3L`4f}-=tP23Tb2D~WE?b3L zHSKbp+E58nk=;jfI|!UjNauD7+{F?TPpG^Elr9TOK&i{bUM;)qJz_7NsCUtL8z|_8 z2_r{KLXHEk$~XgHXN_WCFK`wS%nPOkrfsEGYttDiZ80rVfd^69AZqbMFImSiN|Y~3 zX5N%N+EQ77o(oI>83&kolVroFxqReSr{#{8v?Lq%Bf1{Y*)Vc^(oM{pYHZ>V$4_tc1%Gdlb4#s5zGZtSwMlyl zRWA6BW4WXNeZh|f84VPPn7-gwLfr!B3*MOFO-D)A+y|E%$K4H}FZe=`^8kIp8-*lg zu)?79!V7)^vLi%UF8Gj;3;uh!-vRo9&$^s60Qu?r;6q_$z7u4r4NtB0eeh!BZUFRs z@V6kxg^>H;b>lfb1@P~Kv+v-D#SX!JaQYP(2lOpzrC_;*T>;3*EomZh#kZs%BDf#W zx1?zkc$fg_Thd-2-GCxd)VHJypymVmw(k{?7ln}f;LM2(*Z?__+syNi=-d%3lSH}s zzMe$&?d24tCII^OavR7NA>@AeE0E8Hko(~plNdIEA~B#79|d&@pcDTJ=2r2CZ z{M7@H4~38n{0;J_5R!rFQ^)|IeeVj=MF?3yxC&&35aRm|kSzdr^=BKzI;vI*Jo46 zWR<)u?G?Dgl9yGsbTr3hJk0ixKJga({UG02l`clj-w*Pgl`DG6d%H=xx>V8}L!f=x zBd_X5B341ViuNm0h7UpzlX7fDREb7P+Z7Gf<&S$RbLTDDs_rLFTT#qhd}l*d_psAK z2;~iW+EWjMInNMiqetX`0q^3MJoSd~|1kUrn^N#-qRRRxsY=$9`qYAHLNu6@RmR`; z!RzShdBiABHBMK_Z11tN3AEYA(+QL}i~=4G=3Gy3 zb^!}J^+<23K^y+p0zNI#D!Nkk&B)#)eW-|Rt1*X)XgoMnL~s@DBztv^BJ0-yXR{RV z$>D64SAUH=ry&tp<<)4HuD(UV{*^%ca;i6$Vq)aQp3PNqms}METoF>y9m`UcQwy2| zMyOJWR@sgmgTgk@cRbY;w$qkCo9QF`=%~7k*RHf-i#5>)Q=x7yfFSW4}dEL$>*0ght2kl&N<^6NoQyV13 z?F70=x=Fo6B2P(c7$ugzB<7sfa0ph3)=)dKN^4!GcZ*hTw|K3yTSGbG;#))I#HN4g z`h7*Jy^vVHHN2eVvx|g%q0go@tjqjGSEu^LQ`6zv*Ad)e4ej4ZH#or>D`}=WKP|+k zm}oCL-cqgSf6M<6cg2|U+Wh3H$HD9u3ABk!Z9>#)?MhJZ@_ztuMM!y%{@|(4!0gim zTKu4R7ZZ^a-P$XwJE8(2q`WuIc&aRq`cd|_f#h^ zyElOrU!lcCq{&5<)KLCM0j>zC=p%B<)^boz+1g!)>vYoa%T?ajNuAFy1Pq-zYP!n% zCCQZ-<(*lcrS3u0*+QUAKcz>ODmBSJEbAx3KirCgi6*|CqLK>TN=ezCoOm@+CAD!g zP7vcCa@E!?BiB!pM_=fok~+B=sjcOTd+tYu*&euGi9n}O#_zSRQtm=4t%~lDB%MKv zx~sa#Gy`1Ga6JO${ph8uPGELl0&U=uLb7-nfr^Q6|+ ziiW$3fU7P_Bdyvqu5Ws|bX?!03DY-e!t_nHPXE1no~nno(}_U)`x$vWw{Q9(jlSs# z>6_*uru!zhPp02DHI(E`-y~MTzDWq>Rjuf$rC`nr1UiN0r0l0a-*g=RA^v3$KTVAC z=KaCVrKle;>X(b-kw}?`IDqL3x#s^SM(=$+&Us4u60X&~g@`NfmRyf>Z_(~fpc8)! zZU3~Zo9!x7Gx5FX#h6OE*>396bnt}Jn3mY-eNk-k+`v{UX{jCc8&w@x1I5;?cw1=r zhdC9ibC{E;D{sxW+)j(ey+)(3rB1^?%qa~*dB+}2Q9FpYzA-{$o5Z#I*SDt>xBCgi z{Cm4!MU3*wyvAF?n6cA)2ie`}wcKcicdq;srP6IC6V`hY7MZUUH!rZl2F2Z8E7 z{G|@Xa}|vTeYoK4Sr*xwjO?cN#j+X?WCd5=gm>AVi>Pxqfv$(s1@M>Jd)`S=&%t}> zBN_Agrj7r{sjfHlGrNR$G_9<5GZ}scc-k@@!b5MQs2bCG%^8q=yME8`D6+lbbOomA zL(PI7;w~7gglT$;!$^J!<1--dKnlkb1}-R-eXCMQ=pF2X$Da{Z^Gb_-74#|JK}-jk z3KaZ#JExxnO1#G-N^|W-PN|hS>Mr6}i9#8OA3#nCkqt3w1~;TY!IgXSLv0pI&~~RM zmUrgEzJ>UOfU^eT5XeCxZicXC@=OVWU)~ygIS?ivyN%@O02^L_&-W8m7ttDk^DM;G zATxz{2x1+`TA)aD+P?wk5Yz_%cRXz%zY49k6RVxwy3Qonb*{o9;0%W_Md$Q~7y);v z#Po!?3uFyYG#TZtv*0YY7ZLs%SS=S+@w_7LjTaN~KEnHf9H|8(r9!M!AfGhAEv;+g zcVvGBbZu0G?^(*bZQu+GKhjY8y7o!Q%eO@JMy#jk9fOz#GDV2@Al87a65>sW=RlqT3dCHG zu$bNvbIp?yT`}c~39Wh($zw*f>MToT0+jJ5^c88LaxvzYdeWtCtbbqnxE~R7z9JZa zT7Mv?tXDl$wWFen~@I*BOJzHl()z z&SZ!$Kt2OF<0>aqNeD5c>^6_MPH9AyOaO8^HZHc5SkhAa zNKK~~A}f)-6>!Et>;u^=#2|>!(iz+`xlal)fhDGPPI!#p>WU2tJXy~@U_)}$h zifRLM-2w5GFJbvtE#uEqL(TXj2V^cHZ?l1?Y&o3OUantEVc;L{)|cZ~%g&+o0Mqo* znG@YIa&+c}FnSA9k3*CEnd&ka3x%o2p{Z_$#p#(p!T25s#<7`hhMd*fXf8W~fxIna zbJbIUgDl@Iqlsbd){7{R$B9~i%xu7U0Yc4V!App@APOPQ0}9g0b6|g9fjwMe;Rc4P zs<6irKLl`ULmUJ7M2MCUS=U>t65!N>Xb;jB(Bqad#d{~1a?^rw>k8sWih^_;Z-DF( zLVCpSK)wO^Wq*ugyTaTQV-ks96khrOa&`)_D)V`2061qLwt{Q|a=PM&uq+aMFG%6Mz_0c;%2bm|tD2UA< zn}BR7$)7?uiZS00w?bn1p;9bTnr|WhrWlaYJOgqX@P`zmXNR_U>skNAf}YWoU%-(X zfNmMBKw1hREn^tSP{98cTtr8k8jQFwms2@c(VH%SZ7v7r8VGa6I1?ZiqJM)J90{=- zz*%hHOPCp0HPep{gP4BYM@0HUsuj?cVx;(9U?9(1-}bh>EX19=zFMPm%``qKX@nm>^JO_Zf*vKO<@70^Y~8Kfhii^k~N zp{4kuiHa-IUavs@axox9vm9h8;1`XtkPuohnF)*LafBZeZ9M}AIVgk_jrAX9;1K-s z>NB~zF31kga-^{`1^Pl<7)n$vL~8)fg%CYKx(m@6Vj9R4AzDGK23ZLdh`IlS1-(qn z$rMOTxne@AK9A(nM)fz4p8$Nxs9asO2$UW5f+#Z$>A@i85W;RIyZe57H37gm= z7YOsMJA~b4k?L0tL3l1vlM%c^gqJ`R%xF?9oT!E6ZEa@>>}?2b5y6!Zz1GJGCTT%= zYx_}&njWbcCg|#@s_%I2^>DMlh}8(AA|Yn#RL*L7dobhm&xon)?Qz1TN(ysEOX{U$ z;tj;Uz^~85w+|rP<((oO!Yhrn$<6Jr3>pqM`!wO{5 zL^>g6DoNh!t|K4@>92}1xv7Ad zzj(^j@>{GZ@6Vr8R80(9k}GY!CaY&>TbJ|6L+f%rxxe;wuJ!(ih1TmrD6iFHjK~>M!8hrAE02(V^a&O1D{F0XzLQ>; zlLxEj*^0)4wKl;y4?rI1<_%L!-J!H3|G0i3Bxbm+_| z2<27aLGtq`*sl|4X9kO6mBRH8b_|53mdA?2O;GfjO9=KM;a#Sb?|-1~)f)1(pb zV+5Kn%c?4Rr#OE;JXVYwsSjV~ou{Gw8fyecu2NM!yKK+N+QqAd$QFX-=C7Bs-29d( zPw-pafxT4|SK)@5{S6B`K zbG{+a2C9=A9Am|}hMV~?_lgiBgveiptS&oidzFB{*yc~+*TDiXv{WXr< zAhtHg+tT@v4<8h(vsNzZ%KPI5ZskPdEu(SB&#b>zE;&~fBaib{g6Qq?X@WC|1y7qY z>;Ijr@}6=ki7G;^c2HG+NmW%C6+9UzoNAs7jED)YqPxXNCyDnTS@|D^c_pchY>Q}e zymS?n_i9pxq^C+n#vaO2$s#;@Hkf=aYZ!z*o z`1W;yZeOBvVqZPwjLuc10arBKPYJkM7$UOZIfDqa;R6^B>-e?7adj+Xi*9W4JVcKz z=6Q$=C(64HTVuu6&GELvIxbeHjzwL0f8WFYDACwyGz#jP)*T5xZsEO=m#+4R&|w1o zFnZbOEZ^7Roy^ncRnZCHY^A=IZ`o8HW4dyl%zwhCbiya|nlMklG+~~8{oFvZ@cfz- zRTpii1A#7%gYfnB68B`j^3Q21nw7y=GY2vK^h-VgqsJQga!)wch?Vf^mk`Q3@5U6h z1k8DkK&Mcbl^s6?o__s_{}BK3^sCe?DY4mS($uG-p1=Z#E{vNfG5us-*r`J$8``XV zk;gol7w`0wc^%EQ^=q0+$AFVXpryOw1p8}tvA6`wZg2Rr!%_MMtIqX$lop-3pPsKS zn`;TJI+HYfct8PjFRG8tL*yA1u6BGa5?1x$GlvSKEy zS&pm^g|^V}AFmaw)0{+|=XwwETtzfEbntU>Tgn8ILp!ho>H?Q~94NIA^0{(YGCZqSR9h<--|7Di? z0eMEo~HKIU%zp+JB)k0gV9Gi363Or0X?0nFSF zIAiM66)S+8k9a(4@1Th!gV7kZWPP3keFQHz*dcR67LpH$$$6?W`i0wvwB6 z!2J(IVa>o)@XZkWI^XHMZ)j9agE&jg#fp3~#*bmOlJt}l?dG; zT161Qf&3&yCx|9@u;u_b%^@a%i~|ZPQF_LdFI6WbT zfD96%EyNulD}aK7e^(1l@c|jUk~1}(%KubTUlaeSD8wN8tmT|*z;PfJf!rv>&uG5_ z@;sn(**=HpGIz196DV8}6rnkrVurlP>C9BYF;;##_YTYXsFX7YimsU$l=H`woQM}+zJ+&l0wbWycP+>rfGg#z%k{rduL8PUjS0@o z6BDLfD>Lqv1$4Py2QmwArCfD+Zim_o=<+lsxVSJ8mghgj{{rapylfqh3; zJH#S^1-;mEJ|fOwf)gkl6AKqW#O~wJCZOlmAbsLt8?V_k#7$mG6>m$fO{VVKko~Mq7#&qBfX6tD`m*yh_*Yq+iFTC!+DdQ zDoDrUk8cd8+^urd+a{(@5dKkbGMDm?dXupXJK2qyJWC1dJ3NUtcX%2P?(qC6n36Pa zL)Nz~J48d<+~Hk+N8BCWix4ah)Ja$OqF_HjpuLxgor;N(u{A2GXZU{!a79Q(D{$xK z90Vl+f6HoJy>)3!>2&oyY{#7vWLWm&YJ0L-%byIbipoYMHXN2oS7niNsuF0KnKA+Q z8l|VJda&$HhCe!fkMMXsOBbC(*LG0-`M(TsMZ;}I!2DuUCH6OgwtLmD=i%Y}qUlw^ z{6eDr&BFe7$V%h?f2hydK=J${J>63c*YkiK@aGrrp!cpV+lY6L=r3EEQFzbLbXE*%Kw7fXYg z$iL1nWG7v_K!BBT$sqHKcf^!zlN*4UzJSvVVmZhXph(0D7DuY9ouL!iR|NBm6|miV z2@0mI$O#2F-3b9bznBNxSxHcJD?3>LcQHiauE5l}^9yNIuLN;C$;FDyFD{&pg9C`_ zfq{;IGXvrwkgYjq!39J?yGneBJFvh5T=7HZ7o%Y}LZ%Mj%z;=5vP_6c5TApb z1SoCUf%sAA!3*L6^NXjh#Dg?SrTl}GHsB;cTmv#gh+j~86yzZxzJxdp@+D9t1`4`R zK*lDUpL%pfERRW&d-2A?18HEa%gtF7u1uW(4J&xRbd& zhIBcPh8PLBQqH=3S3)fVibPSDud!t-VTI-UIYLJOUA|o(X44ViO8M$?y$Wh7pv%>m zaC{S{Twf>tML?Hp-Xnau0&t~Vb$Jeg>IdlZG$y#X;>+^^;@1PZJiSLbeH(D4Jasv? zf@%in+;o+4EF2z`-~XCltm?uU7Gd!2?6Oy+qm{ z5oo1Dq8<~I8|j(8$6(E5AZNpd1rA*q47n15*-t_ac5)uE)8gNC!8;u2x@Q=e#J z{KfVO5X$?At>yQiU~eJN3CL_k#l%Qf8O{OV{{_GmAvTV;_SD;8&H(}~E@h^?mF3da zXRz$khX1B677_dSr$l94g9!lm4rPh*3q5rj4jq;k3%n#=gr{{-K?Ph+c@nC-F&%EZomZFv->Z~Eq8Ms^T-THh% zin?F;ZyLV-5OGelu}Yd?Aw)V4#bkIaPnY=e!n`xRyU)u$_Tk#BAuRs=-i}t&t|+q5lIsYow`F()Bwbt zVMa%mPI)pgox+G-EhTanG3UIpvJ_V8^NgFR&Fhwldx5E19=g$oMeD<}wDc=0qM%=3 zdR^4xGeGczc&k?-x8}IK=1;S&%U)6?8I4`tzTgJOr*WaJyTQ}xK3pPT?d^c z{YYf2`2L-EaWOIH_}&9n3Ez3$riJ#sb`!L2J4@?q-)|K4INy2Q=2z`|hWNgfSl{=T zvwZed;`@s}Tfc5|M*IGs3sTfExOU=f-5)hl&t9D=9aCc^#&vF*;8EtuGJM&3k5g5m zGb?wjC`8>K(wY6o#Okru%~fqowD?yut|Z}pDZwuU%Ini7MGZmJzK$Se>+VWA+48MM zwbX1sV>bfR)YeE$HK+F!d0)RQ=33PJzOwW{S=qi>6y z@9swzirF=&=mWwR(wF%Mgx9g_ltINmAiR#6s#P_1drURO_442XX*7{a1#=dmtFo%kL4}?pRe4@m5w# zQQjPKQk_8OWWa@4DznK8g{onGV(xGV7TGJNs74~x)d)RZ6esj5go@@KWBq?40_h$6 z_Y)JksCBjZ;_5}J{RLU7??uzp3}RDzL8#1>eW~i;<%wnG-K4b9>frdq)TX5>sI(dC z;1!9fqR{x@gv3-!OXFePvg%Fo;J3LX&@D)ypL%0eMU~Kg>KIiyEqDa7AJw#B>R|my zjZO}Gc83uOQC|eEXT6cvvv&4Z>R^XRvJmR}gxJCJBBlF}ol{zU^mC%B0J9EYS0*SG z5gGe=8T0#(L*mv=5!Dh_42>lcvw#3qt&-YR{ zz*}tO)I(0)X4&ds%Sh=)yK~ip@SUd#O8Khyg6jIYAH;vgo~7kmM@sj*EvDXq{V`yR zpw*IM<(;j!`RY)+NZGff$eZmL)=M31ABjpyFv*Hu`w`vH!bPzx&^{8l79StaAPv?}1;V`aA0D#jNb=z z*1-qdX3r62>8K{J(%=t1=%$DjifTsXvMO2d==7=;Sa}0uJQuFt5Wcr_#0>9hNxqt0 ze^$Qs+@4`tYjawwG|RsdQZH*xE22pLNzl!*%R!Zxs>biyn4s99bv_pCK?J&LdOpPR zO1CS|S9%^96Z8F+_fX^cp8~idt-Rx@Noo$5vywpTz4MOlJa|EeXVAwWcpdSsh}uTH zy_X>6z42Amz^SRe^tN>U0fWh^^2(|ih|x8T1h1PnnGq`{M(?T6Q=Q`fw|HGw%(=TH zc!_`}`3x4LC#09DIzQ;@BoWiqiMH~tn3AgM5$$v$(B4LwBC?x>^ygTmwoOh&XfA3O ziqKdilr@?@|AP(geR|8jrCOlA4HMK4;bqyl+5+hfOxVRz{{pY@O)!qDW09=t) zF=3CdpQ;|?|LKtJdHD7^q_dwO^i6sFr=_Z6!vD(fKNlCga=2;VK=Ks`E;*}EB~xUa zOzkhL^Ar=^pL&*4jgf5~FYk&8H-$iXkCUkj!R&Da3H|lVQ9B~}n)JxT^U3nWiaXO} z{1GEBKb_6q>0EVlvA8Sx?j;1u`<3ck4`#nYpldyM?Ejr0fFWS&mt>FI-z!hl~6ZWg)bJYX6(s`XA^i6ry#-*wcg@4TOpTcLZH}di;EW7kgL9K^{Dw#61uJaTV-F_eDs@ljl ziI;c9gloZ4-cB;r70ez=K&_9tqJr;;VD|k4IzzKjO`AG)no1aEm-p?6Zf2yTdV>F#0arBKH3Z7ret8A856nJEpbd1z zK;;=K;f>~fwHKnX{?ZoeXa21PD2ljykbud|-4Rs*p52l_H@Z!i{y!RB72W8>&_#E& zQT_NI8L#V#Ik%bw^$2KmrH0Vx>PVwo3`5VVBoWi-L|b`J4Ng_7!JHigIvbBso?%z> z$0gF}-bU?N5!!ErUK&x{=w1Woo#iclqdSR+{i~6Ci@3N(C;H!A%wfq387%-;q*YAV zUq6?vvSHN>+1?<0D}Z#`5QM%dZ#^07Dg28Je~GkvZzOI`fMw4o&|OVfsFEqu=yaZ9 zqFXL4qVD28-cFAUVZ-`)mK4R zpV(=0w2UgVC|6YnLS0wPxkn}Vntgtn3O!bL2TeB`qRRh4Bi3Hlg zg_L+$efJef^<9bDED^fZ2;Dxsxccq@SA*tn_N#9LBKG4(ZXI!P)hGJ@T#%^t@c%a8 zinNLed*W-Q)DiwqhHTHmw;v##zX(F#l-Ipqs!CZ5zbt{~UnVWcOTo>2Sat^jU43Dp zN~TQp={&_mcSGs()nNX|#LK&4!fi^Rytl~IEHL{v0{V<6Zx8wm$yeBC+`+_M#l%SW z{q58P{O>9jcSYa5kwAH$wkxG}gE?;#XveRR)({=PJ))dH;pv5`%QsdG)?MVVnlf}M z`iT~Y)i@B!`(aC(D)k?%WDsa8*GUaTA~m1|9?y-=_R5NJQ8 z_*6`cO!=^~8q5Dgz!f1Cy-l3C15`Ux=e%NPG#)r3xQae5vR{j=+yHlm7b@v4sIXKi zzZzOMvg%5-t_FyXn$Y*cjjak|%o;8hy~|^z)bAK{(ryWyeHFgWQMjEgnwV9JTLT1q z{-;UR?@lk}7f9BpNo~FrOVK~Yx8VqdzUc+NUW+SB+WMZn*W#vvb7$NX+1o^R;J|F1 z4_|h8&0^W%odvhL%Tqn?7TmlSdFK@ZU8aGXlU3FwR^<;;wL4rlmsoY=2AVCarAw*( z2-~NO#(K=rh=-}H8*6M9H)3L}&#W5icM(}ijfjGKh(LMQGl*otv)d5pfna*q|HmMb zt_Kk@w0>)QHIV3rME}LFgMtlypl~7YKil;ipM&pQ`v3uh$k91veMcl;;UE&7A%ln* zc|NI^I?ewd#p15$yKfM1k1UnZb$QT;mpqr|UtqDE;HqSWNtkAZN#nr^lgOO4!sNZu zFi1v|bszuR09O>; z8wgbN+qTJE6`j6 z;8SL!)qfgPlW5g)$nOz#4D6!Yg2zZgqz;i3b=s0f=3~pNsFn{{IgQUmg8dUN;+B^3 zKM2THg%}Pr`JJ6SEg!R@C%%PvV?k}za#ui@?GakcY>UuhW;=usya*r;&%nIC&bObc zBo6CuF7^_qm|;9XZ*W?q)}A9ZU9rw8s;sM7B2}gtQZE@P(Z?22BVyWCwAGgkTBTM5 zqq?M}NA*w9@QZ+I^EdTei!Ox!Fs?a-|1eIMfb!m@EmXja-GD$B+pxxQbv_zG#U!12 z%Xn(Ye-FSF4YwMB@;3bzQ5S5LcVB8EM5{D=Uaw5>`yFQM868(j6c{YAINDwfnL zA-R9XBj>Bbwd$yJJg5xVr~6h9);ycNBMG%iNRfn8v!i06RYHZ*P?33hx$=vQ4AuOP zOjU%Ull-&E_aMmR48-*Hk50W#_|R5d5aqU#0d14PD-t=hm0E5lwY(WU-C`u0el2%z zAw|R0@a;4{)*}Kfp8$IuZBbIk3 zl)aHayVP2Ijfq;VCH2%U{+|I{5mMuCoQ%pz?=1-XLjtYws@4#x%V$(lU-17e;EIs) z#-Q;hn3J?B5Z@=_UY{wP?+VMQV)!Y#EE>Yk7k)Ftm&~h{lh!4x3t_zm=pqs#wQHi0g1Vu1*d*oU4$srqty^3Ok%?E0c>moF7*}m<~r1ro+)zG}C!oFJS;rH}sxoGrS_Ha0Uu%TB-tK3URVP^X zWd!>9&7+-C)gnv26syHfAeK3~GFu@=TX{b&N>yDzTBw-VIWn>%$0_qU)RMZc7VL z6(JS1Z>PKp*Tmh(NI8Z#GMX?qGMX?qGG{2h;6|p)Qfyp?4Zm!kwz7k=^>1XZBAUh6 zTa(p9WSpxAv|q=tz-Rq(YkArV>x|4N9;{g1)#lB-X}LDt16grY7^+Jtf-H> zUhd7wsxvHm7=gY3-s^bQ1@I7J;RWz^@dY4u-rvt>eE45pQr8uuZf}C%0vL1=`#&J< zhY7R;2QE8n)-FbBAFHii=KpoT6=}}VL+?W{`xrr}r(zd4FA6Zt_ZP+`&gPv zT1!4-1X}#13X;#dkEf|B!fzVS*Nvs_2&dZiU1j7*!!sZ$Kih`Lw~8VJ#bp<;swcRP7%8~@M7>$+mjy+(qY z2snoGrHFbH%sEaF`V~IDZZA%H-5B#ui_q^zNDqVus9*n0mO$>P|`GVPi`gf^K|S+#Nh=OOINXOuryPZy2GAbv{qdPF4qmf86lD zrV~~vy&ffXT`}rLhzPFSf&J3e z#gO(D1iC=`oPFhrk!R$K;LOm4|kqK@Y^&tNb1Fi@u?@}+B z4_cDX&j_@*6jG*{X5V4aQH9fNU1Z8TNx}XoIw|V{oiSm-=9q%LW1OWb67N(uLa*UJ z*WFK#MP{NSZ#$+Ud3&$ngr%t zO`zSC>rr_#xlR`d|L%Ccwv@Qm$w$Zt=68kY?dL?!lH* ziah&83MzXS~a2PE*v+l3G-6xAbmZw zx&}}0qu|RE`0WQiS>(pd9G&2;%t%%>M5uugTB?=n&rDXWVcF*!{@*%pBKCdVI=qIG ztA+rs2=T%@rE&$BJ%d1}wO5OaNTs@6)B^sO0ImpelKtyS-3jJwB+%lC^6c1qk*?|y zSoRx+-#30aVxV4*5S_O=Ra77I{~6$lhWiYG^0FUJQ$K>)?)t#+I9gHMQkW>*^>sZ} zf%qzbE7CO5zalCh(r#w-K0r^!L}cEv#_D|jdjYNpDQ`dP5SM`2qX@MB_iJ$x$-SVu zn!*1pz!f1C{ah-2GpMfg>4=(2*LYCrf-7$e+h=wm>Kr7{rpx(_*9!LEuG>f9q zl;&7Z+7MWm+kInQ z3<=japPx11u zm~dYv;P8tB%uhC9u{wda7?$BFEN1H6ii54%ezA#}SC{9j)(CfvS8&CKTb6+K`5}+J z!0apICA5=|hE85o%eN?2O8@GprTpI!FYk&8cPs(J&PImmXkTeBFTl|4Me=B-d!nek zj1MB}Z7}Bmflh9mltU!)L?&BTXNY#j{p@o(`?~>2uu`RA`Q^@)!*>JNx_YVTMt6u- z@3XY@)>To^+iu&5x~_U(ldY?jwdOpk${9thcCyOLSw8#ZN3du4Y`t~0yhwXLy&qAF z;W{e_wC~sWzIT2A-yfDyZyz!KdD>-z^R)jspF`iwdD@~FK7SNKc^mJKsb9dH@|$%F zpR7Km5$Q>~uv6<}$WQFg47~=Uzxv_A)erh~H0!iQWmF9e*o6ezwF`4;-#>kqr=HBH zDi&2t5@^4ruIkDE0KgRmw?2XLE*M~|QDDv_0|q|4v}D;ay2-N%9;g650KIc@@2nz^Q{M zo92X{HeDUO+iEJxi2NWT`$VLtFCu;gBBI2Jtdh9}pl4@pBAnFp<^P^J%~#&DWNrg; z&XWW>b3*xA=VZ>;*zYVbnLB~xzh_SQnaeqY=Ze5d+Nv`r{J?WjCUZq1(m+J&_#%>u z$v+$weM~Yp67=lM^?;MQ`hR3j^Obj+%*{j2SxKNXCzP*sPUd`#{Z0pyxfhW9_sso1 zN!Q$5@qCDQegaWyPWXZ6qD53a|qtQqB6BB{KKIwz~Ch}w&c z{6S>sI(R(!t(AL3jK;*)*stoTJNVxSxT4`+OQ5{!9g@`}V9v7yI>Bc}Cz{p*w@<;- z6NvD^RrZ{!8V`ne!Id|Q2UNd^jnWSWHi{omh10%QdES*pyxqhI{l&$Cr~Z`LmivL2 z7?@GDrs~9hAHWq2w*~>7mOP=F2dZ-?Q|5ktR{baF1w<3$zwa^)LPeWO9(JIii@sk6 zdN)(_o#~Xqq8CDWFL$!lvj{s!2y{t|gKvud8gNte&5=}I-)^@022nd~q?C*G+? zpi6mF6=_2s7bdGh;deItg_>Ws39Es^ztr&eW{Ld&N7;FRSy40#yJlv0XJKbKXO|2M zEK8D{R>?_Nf+SJEoG^YWieLa01OuojVgM0QQNfI202PoVqM{NM6DnW;45)~J3Woo! znlrNp@Voc^=Xt88y1J^mx~HeRr@N=mDMyu&io5m@kmmmhE znuK3n(GehaDYAF|4k-6Ikd3+2u8ky>ovUewWqlp|Xe6mJHj=b-C8>#ZySTr6!u7|) z3@JA>-yPq}`RF7K-e{{;{oz>i1j0+d#|vd`;1A3Q=dkC?L6R#lYX zj0e~VO0-u5vgCBDIod;NG8(&9wNEGJkdfrEuyS|cS6J>;)NmyEu<9ljsMd^wv}`1) z3O16wRn_g*hnkEe&oa$ka~Lu{N31!?NHSzU%^mbshin^3o}$u0SA5?H*V_y--c587wjg|P z@3d3eS&^d|+;?%dEYP%;MQIfVTEKPNV{r?-*bwDq3+)va(-sC3Z{kF4)%z{dD_>Ui zdZ}K^7gL^e-kI`5e#weGQD)?c(oPd{C&3Y2vZV`(ULR1LYY9%clQTkLIb$RK6QpEC zt=|m3L!k7VK@DN+oa3@(vf7#!>3pXV!RdQA{!jVyB7SWgpel~R?Zi#$`**|~QWZag zm3#Xu)me!AH7oj-ke8Z@s+(A>5|^$?RowD}YS#TdOT6~PMpZFn z&;Jefppb1<@l%x!PP;!(&VcLv2V~`QwdHB>4qvW<{rZ$ipy>0TnNDA~fHw6uQ*Lw6B9 zD>9V35q5U=&lJkAnjMnzX{4(hbZx>!rJ-%YMCoY4L}f^Li6(qB%6~`7DFxd*qBLLn z5$&D0kjFsx7-~-&k79E6T;hri$MjF_I(2v2*hs?0Gd-N@7O(+LiBrP@I?LFpr&GlS zYVE&2ju$`-PY0}6_eb5h{x#z~hGpFPf7$N3m>5>`)UQpb+AwJQ^ExhI7^FlvF4*9d zvw*tOVS`iM0yYfV;55*H1d}^>@+~Gj{exz6LH`!h9>^w%%$h-K{+UBvAW-`6kWDIN z=2s&|5?`tdr=gm2S>Ljja!#d>tE9Ow!L3s2!UVT|sS6X_)}af-+Iw;lCcGyj7rNkr z<)Fz48TB)cS>5F(dz@Vow)JE|VrFXXKVmXgz-^ks&Xutrd~28rYJ= zv$+}KeF~4vFiHJ~Q-8&5m4C8ahU`%JUn2QWjbC)1FaIdN&MT>6u?;rS*U-9+#>z;) zhHf##$I=ouL9|5JIU&-}O%NueEt_ugC~etn%LjjLMJnea?p+$`G((I(dg{;rpTTmc zM*NFTJ*cly`7>YbB@c6b9Pm|If{**=%K|9(Es)*J#s}-E><2{|@+sFXfUnXLbUP+r zeuQ%6)s)^3ryP`?$}XsxA$i1C27Hy~P@#_8{Sw0VR+9HkX6akW^7qo-N;DurukD#~ zJi^`;AoK1U_&l5TLbw+?dMjDHjh#us$KPkl4T!okBl&)7twh5Y!a)QiskZtKZXr)| zT?qK9;ok<5V9t(ASq0^-1zCd4RcBZy+|FE-YG-`D$BWtuFKiR7tx!5@D^!Mt=7C!D zD`#n*yPJ{ z$X&$ZBo<#U2-z)_{bb0tex-~`2V1oHhwH5XS(de(=e=I*|F7`E`p+BZV(K}8@QpJ; zv{hBz>M(!+=LGzm6lXD|p93;WQ}oG5s*xRy21o(yBEVN6re9|#r8ks!0f^Ux5%w2` zHle2M+|evShh`ckQymXG=TC)e6KWPM!7TwN-KdrEAWO93dd=1wOOx`L@>he*`X#4V z3a9Ioo%_0eylwL9!tbzcx^5G+S+{5j#&ph;jcR2x$fQS_^>dP(Spv)TmZbRKZAcZ4 z&k3GC5s;+nJU+3!R6wQ{;H!qe7sU6E<+7wHJny2&fPSqRpWu0HW>)E;znP)w@C476 zD8`5W=4d*B4{MCtdLZ2vHzz#7vs9fEJXM|Tsb%>id(~JOY4lxZ?sJ0YIhlo4w28;g zW@4)NHA-)=y+PXNvjY)C7*xNDa{t2=gG5usViR-KL)bGQROKO5Iz1u}{m$E|`@KwS88N|n;=r@Pw^#s{O z!s@npp`pnbvI&E!)tt)8Wq3Y5DT5JoF9KO=CZsM-t>Dz?-lPfjOf8iiTS2bnItlPq z!#@oqi8Cou?`KGEq86Z)>0-j+{j;4VhW>}r)$+8`wOdz;(#kopQqyU^?BL3hAZ+*U zf2k8>bi{-V>4dn~3uGDXFpoM>Hv7O$84_eq%#d@4a3_Ln%g5x)D}`bwAVxc9TRw`f zNwpJf`A|EO><>PFt1Gj)&QI6%)topB7%mIg%-~xWo5}`DDPX4DQbhoC%TmU)vi|3w?3Qfsiw}P7x?~WnNot@kaS&N&H4F?YJ%9z ztaW1oq}7db@08XJqf>Q5Y5re+12^VHWaGNgAAVdnt|mIJ8?O@JB1ltRZ(h;nN(E0Tbkp_r+eIuP8 z-b(r9UzXWuUl_PV2y_U)?x==eC#lAWNaNhtuc1y8^$S-T<;ghW-8&4Pt!ob3o*2 zRbLRPKW_ZX@bPu!?}+%N`Qh&aG(TpR{}O1c`H|ou?&eQ;W>dSC*Y?AgN~@#iZLw3BN;BNGEVxSVf2=2Uk@1>=?A~6B1xeh! z6bH_Xly%qgd+`5<&o2XaSyHZ2OSgmK?3>Bb0qwmB%_^0A%1YitzR`v6HHh1=Lt|{a z*GZbyXyIF}#S7mW;CqG-T=^c#E%&CS)KWSONF4+<`RU9s7uUmy=k+x@p4V@Bq0Z}T zs^j|coM58cznbwi3IA6`f&hzw8(cXb%AEwVId!#b^ZLrp)wEO39Q>qf8y)K zlKU3JJ|I@`)rP+fB*BaCxze0Sw+x|Y3^s)D_gy2Pbx@=`k2G;fkb21p44|#ZSwINVkN0sH#Bl(6xVTpuL}MU5EcHy)zu}jUcFrc!^1Of zfs9|cgt`))ebtzNBr12tcwYppCqM3u(w^MtR8Ouni|ErRqp`?_tr>6Px4|b6zKLH# zG=XsKcOC%=?x>$9lTmP|MK1JGOG!1-s&+MbjO#OiuRI@vw@O4hsFEWlR{e+P&vxnnk!s)OG5HbhyK{EfO` z&jX((s=T!SlA}XUSQ!~Y8;iCMkr_%4DO_qxHd#f;!(}}R|r|&XJ3|@k=G;bEbn@7 zMx1wzj`MDTDDNuoga_$;s}{?xO|xh(YJ97&8aAcCxB5y;BH0DmE-I_v>RY@{KlC3~ zzf^WuiPEWW^@r5Yl>W$jCxEOXoV5r;WwUQ=lqtcTpv#GHr-ST)%^96~A0CSsRm$Fn zyIz}WGkG6YJCZyNJZ*6mna_23x~{M0{IQB|0kN!JGgH<=d7psHf#t87178&3!1tiN zM7WjL9qPd6i2a)bYKJ*7RT@}ETB6r2UDsE0{@;qq!lw;elOY44ypurYK;lJnVCh<3 zw%!3eBKuj_Z_kU0==leHEPml0+x1e~TdPf$urLw7W4n z)!is9iQw@J&c{GDe4OZR#>2NJl)2vKQy$CEK-k^9M}P@#wcc(+>PfI!Hf>jQsy5oQ z6VI}-XUAxJpf0bjG6(J13HGfN>4h!tbdZ(ROG!(fZS?0vAN-%VocAB%-RmQvI}xI! z>|l!N(46x{)PDw7lDoM+2>7btUjgD^rlo9NTnHI|@PDf6w6CzxlMi7Z0PId&xwn#> zGGjp*QNq0WXJ)D}TAY*F;E4P64R^y4m8Hdb95u=F#@(*k(v|%8ajqFC=b?UGMa4|{ z^}I;=8n=>i!y^@wnqIHMrqUaY6VtVPHRM-UbOeZb?FTdDQYd#K$XxuWS>@2U`&?xF zoQw*M2}n|HHU7hq*<7DaSMb$_KN=*#%xM|&DwOv+$lRRuEXrlG>&>KgwgPP>!gW7L zB^b}Wy@FU&JJaXh)D9oWfqVa_z7(L>Bwg26bN+XV_Nrvy-kTxak#J9r#4c`{%F`H< zu&URBJUy;%CDqnlr}UJ|xn7s9;HwS4A4n4SYL(moY0dqKV^ht2L*DnSxgUEHON6!n zOIN4SadU5hsJT~p=NY0UczCoYTQKPD16e60nc=o4A^yB0(Q70xfz>zKvqOF5ke96& z*m5ZKl0#lPp&^OS)wf;<#lA%yho&8M97;me8=^VbX{M~`3n`T z1WE9pTQj8Uhd9s$WJ&aFTseG(lrWcWn-|~J%mxbx9QF+L<{27_0e2+Go*@-8<%6Cf z<))q?O|N!lLT=+aD_zT1L;e|xE(S^P$@LlXB9yxZWG?n=l=7@fsb?7#dJ>SN+FCuc zi+scN=X3>MZTL$-lK5wEhIq9Hylu>51Q$}g05;oa7c|wN#2Sslb$cp3&$Evc{6?Ux}ooeKjmPF0w z_`3mFykg8mBmW+Jd(zb3+0D}Y-9muY;Ga1nWmF|7Ggst$2si8Flmj0ViZ?c8zW-=f z`E$I!_kRxmuWlWsGJ-{buWJ6EAPI`G(gL2_?|-wRnt!yeC#P|}0Pt1KKN=+Me`}l} zlc44Sb}QRuq-vWIRyqUtK-O}mE%sFe4y$cfQQKBwz}*V6+NNTM*EZ#*YMZ8a=||0F zKi9IKq-ptT$bV1KCXfu9#BCX>61o*+#@ng$mbGPxI!9&NeWr6XI@LL<%)y=G1|-kH zmU|7z{JSt7Z*yGPr356YiM^WNQ*PsWFW{?&KODsO<+NgRpcVtBU)qSB6LqfruJq(> z;@vHg(2uk)QRli_+oh!HU%#W4?BaR=@KwQoA0)%dU*Spqr)hobE2`4IH*;LIU>^%u z(krjF`$9X(&Xt@6|83_=RbnLZnpW<6P_>;ai@#4ouyf_bJVsVp%h{|_PPx{O>&_Jm zXh9NapkfjX%g>NUFyTEDxv)^TdIY}{yGZ#P()qR>wz1>3bskcoqfy?XLf@qe@hRJQ zz6*in=59*m>kRI~k@-U_$ka)NTPNxYovM7raWXX#X$bhL;0qMf44SbUA*~&#!~o&o zJ-ZK#FE2SaSMMuIgk9D+r@6{mgVJf7(^(f_35=Fb*EtoM5aV3j3|3}l z$mJOJ#(}I-J$nIw}=i)fR&E{Bo{6zH>ocK#Y<;1pv_hm4;H!dP0VKmJ zomXBaLZu4C7Ol@B94-ZF86Cbdr-f9y2Ozv5iLQGYkG-UpbbpB|@|@7$ac<=ZZl3iD z1nP$M^jwO^E4TUoP*2&lS6WZmwO3kC*|leZXxCo&B;K%y~Aq&yk5t*(w(rNg!?R!!KFIN?*fJ`mG zR}FtJNP>@o3=Y1+KnW=42AX%>5o^gkg|aikEU4Dir{u^PTrW%)^i|jI0%9~fy`qeU z@~#J&g?`6qrI~tvC7BA#ogeY9zmB$k{qS6Iikv!^5xC-+v2A7kARRQQ^3>P!Wd+x@ zfUm0leV`>SXFbgGKUaHHKI)?qCEn9p(D6&H7~BG7 z*&u@XcNK@W`FEwG`FE9J{(b2EOqKqk+QTC4l!nc}E6pha-HD${ylo|XdkN znKPOTNve_e8nu+N+v-XZ@Ks0>Yt)&Jgv}YPfsr#tr<_q*66b5Rz69CWnbPXZ%jtJ( z)0{a1Sq@J6Bq`UT;7$aYGsCFx{M$4^pEGsu<@$iqD#Ur8>df1a=FIs#{E;(8r<_q* z5^ukTGY623op}s?Trz!$j!WiN0=yn?OiG#UI8z>E&YY^2l4|6mhK;2z>_&jELXybu zfipuP<3VxcjP{V>`~P+Al$OM4dvNAvWMgMahcT4Q76#I>Gc}RrfNSc^bQIjVpt#OB zy{+P?rMdr9kcC`V0KQ844}m0jm+oU7l=}n76id60PGR@qoPS$I@}-*h?&~6dat(H1 zK-K(D6m0`Zu1|$NS%hNLP^z;sT{q_^<2PL z4ZnkO+fRMolL=7M0BiX)kbCZ}cxB1FOebZ>UG3I-#UW(7D#@dWc=JFeamh?A)|QV^ ztUKYw!?PFXrVY<55Dm{P5Dm|swqEeTIiBoSvw7d9%-#XtZbUph3l=@)NiE{NhLKR% zg`W1bCvBA9?O^^U`%W!o zMl)7QJB_V?_C7*sqUhBFnFm@x3?C@C*2+I2oln7qU%Hh=EiA?8@Rgwax}=0cyJQ%1RBvoX}fUnXr>_JAm z*Aq;C&pDoOIM^-SSN-q9-h@mDjd^Y`<~C#mckxilD*3Mp`%8 zBeXz;mPJBc+wgXkxIpjqkMOK(-MWM~zGy^nEGs2(*Qx=|Jy|AlK&{mHA(hnS@T0Ev zCg`YZy$DHm8my=wxE6k=@ueC~s;%KIkCx-Po(}k`;TMDWcZwCP|3G{3)x3cVPZw|X{dh5Gc31aqcrm5Sz*p4MGH zcSl+e^%DV>%WkMHZBg)!23Z1gr;)%#-KNlBt`Clf>z)@y-9S{@bEPVuu=;it*Rg=F z(tKHUeRY`(jqudN`k!l`)DgH+X(Vc-Z2v_54* zCu#n=ld!;`x;fH(RTyAjrj~HSpHn?WJk`xggwfH+^BPJf@$uzUm{X7qpHxPkFN{sA zFeedhHJ)XO&NB!|uzyuj{)2)$8f1C62@U5~kR;X8Z9U4$EnM#be3kS^f+RR{s3%WD zxhp}oSP(B|?z-C24EH%zQOA<1z47<9@;TQX=?cEu@LvM4nSGQee?xf{ccq>&U0F1z z&KgHq>VK6LwD8Moz9m?2t|v_pbi14MNh2h7;9=H|$!EF)YH~>Du5udJ3(^&Qwc#HL zlEg+W>U$t<^r)LUqtT<$siIa|5?yq=>}q7=qAuMQ%jj_hJ6PkQ4)?3RR>hNzD0n}B zELT@W-?^7A#@DLk$sfeKiCq5pe3pLYv&#%g@`l;d{yvWDE0t7%l@6y zAk8h^{Q3bp|HJ0j?CepktFpsgvwlB%_8cqA(qwu;jX~S=g3{6Sg38bed@++&pr8cp z6_IvI!&(8QDJ+fuBZ`0JahpG~?}P6SQ!?vHN7hva>#wNwvsJs+?v!sS4YRJaBqnP7 z#l+i`xSkGT_fNN0+5O{~6J#J_N=%t5O{i5GMO)QuR6cQXt^MiNUjkq zO}A(T#da&zu!3T{mD05ED*LR;>Z-ZL>l8WLtx9Ewl_;HBH6MFyC0T;Jw;E*ie9>qO z1#cf;NxoM8?;sm&Ybdv8Zz5x{aIkGW391jOBw2gVNP=P!4*D)BFEwGg#SuS2XA>R# zr9(YQs>=O1Rl@aHz*hyo0f;ZREAZ@I4;eqZsS_u{7xX4X$Iotf{6znSNtud*`w+-H z**S^~{{mvV4_@c@j*dlRY6rk`Gt)?#iO zuF@a9$u4jN-N~VJOId{#)LO}`Kin`|IsH+K{)JehA#xs zH3@%&qOl;h#BfZ(H&AZIK8ivAg;?#{=5uA|YT9AqKf#YSpR0^*KDTrwsfj(!t7TJ+ z9DSG}<%WiQwdwx|vd!mzLEyuOue)**GVX{-HO}!Y&1AIsJj|rpUs$iN+{$%Yx`MAZ z{9zyoCarYk87OyUx`fq@^D#-2X`9bOi)y9uERoN-?nsyS)r7wQBtgyJUD*re{R1*X zM{vve5_`WZxxc}$3^M+Am%`7^emjrh@Mf)H#}jkN=JOI*xpfc>%Y_OZZa)7)brY3) zp>_K~TDJLI6>RhQb*gT+Ce&o}`2(i;z~3~GbBHB}_C<8aZm8_VA=@^e->cHW#rXa< zTz3m7_8nE*e103TtbY^rRQsN?i|YZvR|Wrl5dG3Om~Hxd+N0W>ij7O*qk1*b@uT`G z0SS6Pmy`}DxW|Aj#eLL%=YT!}NwxI+SPnqtdLH1bq~8N1iNRVVcR|{ts*g<6`x~8l zRKrK-O;k>bdHZqdb6pPNRFJrjgm@ev0hsk(_Hd!yC%K)RMas;Xd*Y9r#(HF;FmYsM01Pzv5aVxvbjWWT-&_IV-O z9@RA}9kf{Q%2>GW-5@J7{Q$@^Q5OOmcJP)mC8_GY_cWJzTwh6-_tk_y5k#&xYMG5x z|2M<1GE?=a%v6zP?_F1RLV5c@mfYQ%M0WNx6cZ1tcB?Lxa+*2M>DS(&MVmD%a4o~B8e^)k)Ll!A8yu~C_Y?4S91$(tFnt<1Wqbnwoj z73B%I?pq-9eGDnY>w*^$C|wtv;;MmbE3}cFT+dvt;}nrX!y?YC0e{mtd1}rX^1S!F zUI#C6r@QAY4@Ee8&3-5TMSWerubxB^5>vd=+0UutwDg!fR@BF@bFmUu)n60y!`97eYYq4+%(-+Pr+Z`6EtD<%A4zS~$MhYlTJ&!0ep} zF;oL;Zydz6KxhxWMCCrzjms%(8+cpm^TT3p+Q93Mk`4`=X3CL7SCv|a)FHOeO> z5zyNC4!1a!m22%hfw~sZQ8$4WYAqc3PjONs{|Rb(SGogsQbC6*(bs8#>buMyGnka1OBj`OM6%LpHVH;IPd%y19TI zEj>iGBjuHqdSN%RjOu7k4P&0n3bl5hyjse&LM6~f18khUij9tCv)et!_Q@pL!Ht9) zLU^q~W}*@VeJcEJ%I|wHe--?blz;X?{Ic1@i72DY;2WrGA>4_P$bTSM{{k<{ZzlJ0 zeE{%Ph*Kdn)2krKw7z;AmT8?#mc*6%Ubzpk_?D$UtO*A)zjLN&Jdm-kY=TQ4V_%+w zALT)oH9ZX3akK^`)spX4mb%0>2Yi+Eb3vTFueadU2-9QXVzHoP}L=IN7>Wa-RC-4a(mAm05h5}Jq* zt*p&#OorzFFNvCbVzKO35tj_9g8w;)Zt!>tunkNrK%Lphtn@$5Yy`uPa-{?XZy?Cr zn#CW0(Xk97(}~kG=aY#!q%$51D|arnEw@c$nl@4k$IVk!H!)qc-Z)6hIu=#1j^$-l zPt&AhIoUKPwa5R@iH$mzkX>HczlLnfud22vsD7BxQnYTS|t&Lc1tq0_{*ymQ=rcXh8zu1+J5&kk#_`x2IiTVs> zXV^2Ody@`(hU{>Nve|R`vKp*%!57?z+7Mo2kR@FXLV{DvG;F~jC53@+B{j*O< z%3QAV0beEkDIiIdVIjzS3zC8U2&sKv*ta-cohh|bcuR@Up zrhb~Mfg?3wwVc2)4{LJ4&(M6$L-2jzK zj#8BJr1HQz9nD5=??esB4HVs|=*UZWv$rz5sz3jEdY)7(sospY{j;i-6gv+cQ$w7m z70p#tqOy|IOq_6UiTuX3de)&Heft07k(&5dk9^Iqe+ft(@vqg;1Vs}SO$Jden<&cL_2)uJ^w!DYjX25+PR5trLC9PTxQSoaQm6M-g2`VPx{6VOyiHS)( zmRDKwh;0x{RENsWt_?1u*5n08Eq&19XlVzpOC%@1I@IW!T0cqKAu^`mj{>Q7OS!c_ z2K?FN$aBw@YPXgdu(N?N0s(m&kJ7zV5%4*9i*e#LW z?v({KW!2qT()hhP(hTcfCs3SLnU}FEn=|WF;v26UU30FGLuT}Mm0^A9SmdfN&XE3^ zwYsTjk=Yp>GxAhL?;W^LF?@auc+Sru zJzmuZ5cmn=d!YD%UyEWL?v0k}*M1rQ7uepHpraN$(hTrEAutPK22i|*#wSY24Yl@j z%uc0?8Zb5heS^^Fz#ui)tq9J5C64q28jY@5UpmK5CZVbw>mT)SfR}d=^%64A0^URd znXftG0FBBvsTV6b(iNr5=3dR)RH-&m-I3`8cvT4812Gkl4t-O})EvS_WM{BDs;x+) zo|_v>6yeGUO@EMIrJI)0V4~Ke`7Y4^%dbl1GGsh*mEUaO?SfljslyyMrBL%c6?EZgIGEq8Z zWtYY+Z{P>yxOgBaiprHUb10_@ZW;e{*!C;#P`^$?>>GV7}XDdUZPF*I|Z=hUez)jnx(c~NC~?X!TW@qdI-_qx;H>LJy>u5u#d#wG3& zx8z>; zsUnL%?jmEjUI+Lp#PJRHmgPi@yxT7sj#D3H-ad$T~6 z)qF;Zc_S_#DUA}@)qm*A8=Ve98YQ#KYk+sa+XHz@HP%KNCwEV~YiHgX9D3JA7Uj$5 z2)jQ=8ZRGx&|OnYXLRT#l3!1<0AHo`zm)>1K+tOoGT%-@kN2doe!PFxVJSUcBVT$V z?4A^9+~4h>l+@DeA7;sgTt@)DN;9dnraVDyCArrf1NjQd+X*s5pLfzb z@MM=j_QP_MNf8~KQ`EIP&RW-xScl^zz1|;a`=E`~~ZxhJ8`pCQr;D4+9-3RkqmSy3f z0(YPe$l4JN<*wR5_Dy{r>b4u%y#mMauF0KfG}G$c2gz`G2I;VRAJbOt@Sib>3^@VH zI}2oX^m`-@-uwl<8{r{-mfU?#-35Fy1U(G8rP3s@^Gd$L8KeoUM9>mYmT9m}CZ%l~ zS(J`8vZxGCpRev+q1xX@+9?g&$f7i9=!xu!87fu3Vk!+~MQNz2dBwjEDjlU^2f0_x z{%R^|EK%)#k#gJ^FC$cn&@d)F^j6OF~$(pY_QVAFpeRibG8(~}xG+NT5E@QkfgCZl2 zXD$q1W>lt-wi5LxGW!7U0|Mu*a^y_FTSH(8#0$Wn@#u8>5l)LY7(D}xPR*__^<$gM zi7)spE3d!WypX8#kvSf)z|hr>91TdLOMa;@-D9m?sc0U>p8DKXW*0i2{Vz0o=Yf0f?GWCj+XjcVOBZwMfj?p4adN_fVYCcyAW$M@En0(Abtjl zKkLr_;ifDwP*y*_aWXuI`%wFB{)!mz$`Tk4F-`;32s{Jv1V9nhA?r`X9_&&N-gAoD zR-;)ROqACP^ev)ascw4$ZPrl!fY+43M2PXgAXS{`mZxNtDtOJYNM!;q8gxy z?vC6#p~M?bHNQ1QV&FO zQC$Sv`xNA_b>t7Ey*CLwL*Nmhxb^q7sAl!ljqvqG*bV5r-jP0l-GC7gmjE+8E&Iq;WZ}Qvhnb$< zhr^`*%RQOt={-4A`oG46?bm*fQ{h+5Wo5*? zrl3V=zXEuL1a?4d1B!N&zqzrM2U1ji%1Xf*PI>=NM7&DF_%LaMBV_@v41snKZGht0 z)UU{bE9>=cedLt!pM~wc4mus7Q&sC_0+S(b0XjUDN~vw?(Y@AEES$q*6g4f9Mx`?K zveWu>9#iiH#O7n>S)k}ZnvQdrPC-MjxtCv3mourDB?SC31Zd(Wr#zUI^58mquOAB@ zAhF{QLJ`5t-64A_r12fY*0i@owbWHaNz&03MEzNw5t7O6xVTpuNwYf5GR^`8OUu=-h7ZHn87kpT;j)XKCHw~ z?2s=@5q3938e{q$RN`vs_v;GfE3V%HzDi5*=@)_QgL0)XmC`EoB&j0jwWuU{uqy+; z3bCK{vq0(*bel)w(@b1NZoBPh>B_Yy;HwbdEbj_rFqC&X$ntTr70|Msfn2QoyCQxQ z@@J#fg1@qA{Ymyq!X116WF@PA9nfan;aSr`@fsgI85-@&WU5KS(D%tZxM>+%X;8mH=z<2~`w5=E4f21-hGJa4v zcs;9#GAq-&t{m}^BXt4qPvkC#7@>h*3A_aH3@}J_x>Z89;>V7Z2MU{H)sYutn-Ntu zEKM88R#N-z5$lP3TfhQ8LwpH@1KD-4h2^B61KA#`r32Y1pEzkuLIc| z@O5fRAVYR(1{uh@bue`&QSW1BE#S>05PZsX9xzD7ihcZZQkwL|$E60cd9eMOpyI^c zMX`V*$0A?@*&49DR-l=fngRIr33SLxnL2bJtCgx^Dvs$utmr`YMs@HOqE=&IDd0Up zpz$V0>H*$#0@p%Z4HVba!cSRn&4LYNkAeLR@pDw+Yyx?oIg$f-#}gO^aV9`%_aN&H zV-J?A2MlC)*2RM>h`I}@TUG4>0(&8T(ZDGLnt$$aeh+~G1TKO&7Z{`lieJyH6We5P z4NFH<-X7SmA+$iXb`U81!jS^N+eF}Gh@l#IkHB<@sX+09yNY5{UTBKpmgi8Y-x9x7 z74{(0ce5kC0B;9@n;<4=U=x8A5Q~B0at%vOWtb@)i+O`!m)}C+0A3FQS3-=`KpO({ zA)e7dBLX`iwgZ;SZbRX;-|BEg3B#SsQ$=`JYOryo4mP~2gkgsLtel(0<-AMFnQ?04 z`l)grkCD-UmGhSnTL53n*~&NXOB@6SsiKu{WXlz0#pOE;p_2hC-`62l0KS&5mFusB zegv#sBNJX|A}ZG&UpdkduyUOX@hISHxmtOCNN62kn<_nNq@7pmxcFM=do}~$e<(iR`mjfc z&-#WM4S3TC42L)$7^JdAU3zj9TdW>YdN8g2Y!uuF5$|l!D`+iLGeZe%gZLUKnn@8x zdTvZbn;>mO{DhYBYoM~*Xi5O@c>--9js*JMno3_s#QH0mQnh~EI_WZ$abwh;v9LD_ zkyBAV87O*_=RT6+Q9nf9#=)BrnSktdfd3%@eu*J;zut2MY|yuiLb745HZWm777zL! zqo1__fDUszwP4dWL@UQSJI*NSk}t2JZK~m`zx!?e+)rfbu1>S;8ZRAw}qRN9%XLxQ05mOze0(K{Iw zk74)&m7f^@Dt~0ezsy`+!B}Cu@^6j!eT;t({28#^XVdx2$=T-&FA*zARr%rT4zi5v z8o*Zt|6z~>tsg7k8(Zvt1u{QIFx->GZ*>^>It^_ks9#!F&lQ&0JpY{CI{6hW&tvM9 zSg$I`)OuOAb|>U3y=ie`5)5@puFoq zW_WYQ)E{SM{o_{pq!t+pwl)gncH-SfBcaDS*pxJ9)SP|9)6kMs{ko?Oke9eF1AJBR zr-LMT^@u>;gL1zHSt3`3`e#Ym+l#I>ebu?_N=No^{Ri+>!T$&(LCFD!%}h918D#nw z;-mz9>oER<<(wVy)p_bAXS+|#tmZSH9-i%fOJ@1w>jtS4+;2_)$C$~P6}~oGSqpOH zGEL*=$kqj|(+cALXIZ-nUa6li(^Y78Bs9CPI`Vzpe0dg@`$ojS13r#~x8WWYN~%t? z>3Q-I*DnEIHT;)BlIX4#yL#hP#a_v>hE?n;lY+QnD>JItN;?NYl)>ac8lvd623gtY z@(WYVYH@gE@iCS?%;GZ;(W9cw$fDBDevls8AILRo@utY4F0)8*ZLTYKEB}Fr-^DWj zrV z|HGX*qU-Tq^N6o&iQz5xyc!1xkc0Dcq>~B_j)cBzk|vbjRE6IDEk}l_(3nW5e!DcG zfdnLYd2)_SQlUE|Ax%z#wNrBBAz02A>HN6nsT*<4`{)kN9Mj`|vJS0zAFHjz<9et5 z!j9dbwi|8D!u}n4ByYnk3gqdu{u-6i`K)v)ruC0$tE!T?<`vp*osSJaJ>pP2od<}I zU$B5~N!LU1ED#;P@Bvc}5`4Qb5S_C1CV{NHUxrVH(hkK79)B&6X)5$kB$TZl26f@j zQ~t{b@e@C)&%eO4=2~Y%%z2$ZiRz#-qdKUxbG*6JWkDeSs70OIirryxi#*vpR)=&x zZiY*MgI_xARrHBw`O*hr_tZ$EJ}IQV{Hvvs+ErvY*HM74(h_W90%QV|I~imiT&12R zRiuB51~QB5BY>|${I~h@fy{?;7lBN?w~4FBfVO?)9j+SyUxg%5YXBva(>zr&eJREG zwninR%&26Pc4k^xzwu@uRZ;ZnfGnZEjepnMfwWZqkaT`rGMiQ(Rx-yI=gZj$yH`dU zr(!9sWYp6AQ$!|ly&dpXn)CNDbw89l7i1nRRS%LX^3}VoWFgljfUiOlOkWkq8Yp)o z$i(k3aTR%GS+;EB`UBvr5Jw&{fBXlO=e0IEdND@| zw=*gQ?sM$1T@!W~z2I~QY)^02JDC-Rqd(mpkV?0GD%}fOQ;&Y{7D(2*%sdsj4xO<; z;g4r?221?D8c{l(puep3nGLtLB!dSKdkpzU059X5y7CRgS3uF#$vUwCC#GTp`EyWy5d*KtbDh+(jT#D$ls|3h7$My zVgn#WSFCFiTX1748s@46?*>HnBfL+w?<3ImTV8?y?=}LbL7W2cD;!Ri$HnfHQlUFp z8#AjO+X{DXMC^K0#smH}1d1L?@PSWCk@c6@L@FOcoGRWzL>@lDm1N+$F~^v z79zHm;#>oG%LwFrN7({JTCXBCSh1K%x;0p=L4R9mr9A@eI) zJbVj__1+V;8>ydE?REl9zh_zj@MaPi0&zT0+%%8+pRzzbR|_}tSag7WE%BpOp$CDN zAr@$0D1j{yUud8Yfs7vMJ2;glcPy*2nFr(d)D2^vI4w&QAAXJoc zyGeNCq+3mxj}!eUP^1l6+xRv`RNM{)S#7Nm+k)&DfHh+NPi(IQtPzV;nAS^~iIt;9 ztO>G>0BghsLma08ZN$bvTm@Jo)-iUYlnQ;;Cg?;hYi-0HM1GbU&_--E#3~?c#3DUc zBI>Y4><5IuQ*CX;%Kz+09$>e-EyPhk^egwE3*&Be;7AaEl4pY(=$y&arQFv07!j?}Caxn8SJ#}zr^X+f0$D2V69z_)Bqv}t?i8T%EWpZsp3!v4{19i;j)Jb z|F^a?(#VN5qQcg;b2P(-w6-%6&W(j5Uk++J6IJUnMtcw4)VJ%^ceT%uM(eUF>Du3m zZljG^j(!{==S};(KpM5q%-({myBlPKqGQhGv&=qM^pZE#m!ujRkffHi;|l`7S7|w~ zt?Is$(Xl!#w;9NEpJAtFny$(}^@m6ot|fr4(tM53Z$G<1l@WaMM}eFR;amzz`D>Q? zs$}rx^a2?R_Vyhvr|#W^?4VaJ4OR_0&GZ9 z4b-h!ENi%K0(@2XA5`=-i2wI{hAr+;-foa3bp@|#65MhY?|<##mjxMLU*RN}eGcz` zu$-ak{P@L7U+v--uP4s^Z!caysjcknQ-9CnJbm5Jb~n~+L)+lysSRzqSv5ROZmic; z6N@*rX@DbS?&p1L1*FZDO+AmS+o!l3j`tXQT5aI1hvN+=z$o>^i8-8Xptsx~au zW;nB^cS|i7X>{%xq@NnJy1X3@;>qX^lr9%F8OIh;I6HxGxu|6aT`p3^tA1f}9VlHc zDtm2?blJ_cGY~HqeL>dN!FWfR(OxHImid)78i<#RRJ{a7H)Tf4MJjnGj9Zi$ttKn; zHyA$xjjlVN<)TzZ*~&bSJIP+BsU5kUj}kR%k0X}?-gE-XAQo%jb^`xE{00nCoo@5t zoVnML(}BX0y7lG7*k(jU!x-K7RHF8eN9<1IZvZUNejgu30IB7o*un@>(B-00s-??C zpCa@j5G@xi4sX=IE*I6nv#BY844JJNWVxtiCrmXas>W|@3nC9n`;J}^iP6n|ck<)V~LzOsO2%Ne2Q zdcQX=uKcf#Jd{kKp^w&e(aBv?<@^ptUI(n4oBYGW3HVyhR=%ea8VU?jMJwOPmMhAN%Xc0^a{(*g zaxBoZIfH)~rbw+^I}>UPSh+?fywF5cu6GbW39xeA1@RN$Yq?r^7CNp}1FSqF6Fgja z5trv>#Qz7d@_Zj+J>YA3S~YFN9x#cmWusvPJpA8)>W_Q3ov-6(HgrrO0!|0lbC;T0k@dik{n3 z6kA}qC}k@eg4%M?YlwL7gNC4RyjobTN)Y3Lez&G{;^m@C>#?+`8*S!c)W`QAL)iIS&|SYN-tMEm-JJ$a&Oxz$187yv$Y!ddm7 zR4A!Be=cq)O^|5=n0LPVD?vm0_Yx>?Ajk}SVunxImnWyeaxahgZ>Af5*9@ypwe8*H zCazNfUp4&mKpdd-nj;TFc~5}M@aZ-!{4vwT3zfe-;;%6NuS^%Oh2;d@Q}W6uC*fb) zmrfV+*Y>5;#fSd2y-gPfKR(TbKW3axAS-YU#WyEcE@3u$L%FS=_$Sr4 zgBTez`k6rb9a&jFbs6(?|3=cUO__lj!5^Vap(a13Rj?_xt>&rOs3vQItj&}J(gDgV z0a=3MRJ&?+C$|Rwi9-!{E>P{kqKv#trSjKKRh5wVB63QJPss7y+)uhVr-d;)W zgut zyv+4Az*ixbm~-;w9VquhkcsbTui`55%E~UXo$F4(S0PsZ+PZ6dRfsqFTUn?LZp^45 zjbM5MK-6Ysn(lw!K&mz*c<$~3IaL+U1DVEyN06!DA7%xvf#uu;!eIO-N;{ar;Lue* zihEY0SAq`vC{AN?8)%PX7UPfNjKwKZ`q2J~jP)DZ*4fjm&0*AUC^PCely5y8kPD!^ksvd<$ZqzFk1@Jb{@jTF5XVT<53%KA zrCV(wf%)#;1FK2FDmPC*5#|N+_Db>s(n}(ZY&}Cf*u5kvYt+)lh`(OHVN4v*8d1L= z`PKh1{9#e*&}Kae*Q&-r{~@o@wwA4j8e1Sb)L7-i|Bw&9=jiNW40@eGmY3^sQxb2y zOiDxGQSN-)cUJ3MWe3Y{EhA^cb%#gd^P6dzspR&nyURGPHv+y2(FEUAM(%=gXMxQ6 zWhSm7H9zev^SHhU_$nm9n}s>D0?PXYWa1|rXLCdxjr4Oy;ROWfk~jf`lja8E-`|K? zzK0PLc4%iOm1gM^Q$c@Ht^+#6)QpTeW-6)r-JkC##mIC4eAVzPf+Tp0^W^(OxhH_k zz;rd7RJr*>T)BYjC4jF&v;(zrDijoZxOSct|lhRsnQ%4i>ZA?ySW1nGeu6#R8sX<{#-+9BGVA? zRm1-WMAKG~BW>Wh-68|~)o@bfx?R^%j^lbF;HwZ5nw4|pJScYrD0ZEH`X<<_q@hgU zItlPqh!50%D=X8W+-E>0zO;XoQ`}oefGJPnj(#~qUgP=};H!rJAV`Axf0mVxpxmt> zGw_$`sN6FC2<#WGzX84q@reNGWe!9<0gBU;q>5C%x2hDvt`GPs#NP+DEz7Y8i1!7V z_%$8V+`El{BvpR=F_q+0uIB^3YWST%68!XSS-Aqry$xgr0v=T!dg?6stD(Z_h(R?o+MS| zzR5Jg$Ff%w@Ks2LjTuuxh7!zK{dgCdeMB8Oe&{G^3?ui=V+%OXO7C>_oDI3MX?RxQ zicA7yU+AELmPeMYBCQ}gJ1XgzB*V$gyDRg(EXW&!6?!~1ns>QudeAvi6G9~SS44hL zky)XL8c2z#5+WT{q#DRu7K*6(l!z+Hu*Eo@7?g7Sx7MNK-w>5MLLFZK`Y(?E96PR( zj!E*C_Wu|=4>&1`=I__e>|Ni%9J@z2at299jvgQx$%-PNA`(SG5D^6@6;VV{ zz{f~XR73?MiWxyc5m6Bl5EUOp1Y3i#<9z<_v!72@S6Bb4I?hbbY)?;%a(w*< zhuoU&Y-QT_R|i>H6SN?`wIIDnqdFJ2X5*W%I*JBzk3yy^WI_nh(nJt7lEg!wa942~ zV(H#8OmFk!gGK%4UM4&5inp{`CHij#^ptuSEES`~>G38i#y;y-L34Urb9zK``Y6oF z;=;nHC6LoV;2bL(>C zIN4d6J4?y6Z%K0)#hz-)$PaS#g)62Hzl|3rny0ff3DmtLZ8L#Q#F)%Q*8_(_uGXb35UtBerb$$n(nm1=LCwz!&21S%b46p`MIqH;{?#EwTPLcE8cFawJvJA~ zTLMe%`=_pBcb3(}h@*?3||7dB{tGO8_L9NL-vK`6$1!hZ`NdN1yM@#*}OVwJjtXt2+ zxg^!&N!9vF_F(S$Afv`hV+Y{9zJck)qQP7y7PEC?(FKH^Sg5s|SXfV^j+Cq{6E%5| zXtivcjv#LLNt6$~hOw$`Vo&N;lq?CSCFYUlJ!t*$EzsS|qn4<1{y(+E*JHIM)cwh0 z8_S!7TmRA0s8_QP#-(;_j_gMAydhCb?7>TFP!+bsD0MDL_4w)Xlcfysx{y)hrLiC3 zy?%meOT@J$mTOD&1rbgrv34zCJ&mdtRF-L)Jjhndw#0biIpZ$mydAZ~jXiZJ2Axse zK$^D*X1l8+38_M|2PXOnWmsWAxEeBwW}34qOST|+`(bu0d9DQ?KIflWaQzrRgiQ@yRg_*vsdo3sVzp|LHf)~*Gur{GL#_ZZ3EMvGy5{P9RMKKh^jPqmvjT5G2s z`2ibbr0Wz1k-AU68kbAi6qLh_(=gPyR z`A@)XDMuGDa)N%la^)pd-bYS<8hTD0;n`@2f|6=epNB8|2@gX?EzFxRDR}CoEXg^m zu3QV*XO0@l8UIkl=+yZ|nJR_vNzZ;JEBY(EC8$^_(hA>`9`iJf;@*>PyM~DS$J<8S zi;))Q{^7PM?w@UqNU(=IZo|8O56sq7iy(>H9?#-k4s>>~Y5{F=UqBNUmrC=nSKkZ= zPlVl``47Y=68;BbODl=BQjc#rk6IiFdQ!~os{burZ(BtlnBKBFOPh!?r*`-{`Jr6N8pcSf z46_x>$5ZM!kfeJ0>-ScYPkIZ;C|ZKDxay4L^?_NtX$$S$HoCS4i%&C1nobZIDrrB)V(Myn|?4roo9(%d{_1BGodgbSX}Yh1uOOpnK05qqt{g`4;=?1mYDMi7 zTcuq^DMxx`$S7KZyRmCP)N7e;$MavJy#}|H?u30HqaX=xrlZFrd9z^lm_7P%bn>3s z>z_{ECymmRw|aPOY(IIBaP>c08ue-}f$_3g&m4IX$@>Ck+u%gJq}rhH(zI@*J$b9g zTjx}gAJpbt+^F%=*cN!N_hEYS?yo2B%XHk;2VoCHwRR_O>nSl;Pu^!}`XJGI*^_rK z;yJr#@EPCe>YmUYsU)~>6VJa$^R9r|Ql8h+o}$9L{b@va7)IKmCp*hr!iA7g3v(fi z*TUY9^nwvt@*l#VA)_Ek z+^u!(IfA1>pAXJ^DK|XWlC`3P?fQ<{(Z=-6V~sR9^?cAq-19-j#|FUoSEOsU48ox| z9cByg6#DRllOLBwGVyWo^50g;Q`1fnn#i+Wy?_{FBlowb;Hj3=0!Q{ zJE`zT9W4ag%<4O%KTq|Oi|S{8 zUrDY(6?@p}yMX`6)5U#f#_o(Y*PYW}i|wk=mgvG**6ZU~eG@EuGFP5fZ!bBkDJ{r7 zu}gdM0J@SHey0@ZCnCLaCp&#-q$Yu}2BLIU@QT26CcRFaa z%-s+xt(H=7)zK`eNP07e(W-#q!Su3t?Lh05l~rGlmXZU>cQ7z}v<%O4PaFHRQ(GQf zQu{GT{v_urY^_P?uTuT&qWYu9vg8icFLC;?6~A3~9{SWiy&&-2< z&A~*u)tvCQ&uuL&5tqx27-<;#~?9fZ82r`L}5lAh)3UeeF@E+=h&=SetdKl@de zUk=giKfqSrljtP4tLF{k`Hg)`rF^`rX;G6F>d~_OH@toFvtQy;Jhbe-yOCrb@k?Ap zx4R|<1(!vut!IpZvg*I(OQ zb5^hac{W2b$I$gq%D|`(vkNA{&(CB_brSssFk6QOY8>3OJy+VQzKheB*Xjh%f0!$$ zpz_Xh`pon?m$Y?Oo0_#+$uz=iA)^*%D2$B-bu2xLXvdP4-+s&a>Fzmfjo!3+N3F$3 zY4N;QyB25Vo(N4PsNW+;UdP7U0kfs7N@Ch+KpmFc(oMc0JOmj<^SmCfeq0$UeV%&ulV(ypB3%a$=n-Xxf<_(Cl~QXx0L5|Xoyl0lJQmJ^VD1v+Z4OQ*1WpHI5FynL4}YJB1v6gMfGwVD@NQQ0|`& zziVyI>b2p447q{uwsZrdKFm2Vz8KDbVM~zwM`1S85o#P%qbHtG{VPtt*XmDwFIP6B z^7c9XHuR}F@31+m%}wX@lcR*O^T|ss%*QZhrn9s8j5-cZfY}W7&o>-|bCs~@YSiMD z9@3VuD`eEdRDelhfsWH@h<12t1K8vD!*0cn-=9aMj^AD!hc{s3ErQux50RKQu+(9X zujOI$=(YenDb!>JGH$7izOrnOMcQSYj_C+1cZr2;p}{TJq<9C~1Ktcj6`Eq}{k! z6WyY!1;R-&ofX8o0ShI~kEWk?Lj}HQ>HO{cZ~jJ}c)|xX!bY9!>QQtkWO6Ti5 zby2~NfEK^UnqNV6b(5ya(&DdJ?wD}ozp)`pdSK@dgV{DKyxts&*Em^G6|+9Qk-5T8 zIzFy-d2}Jg&8$`C8g@P#Z>Xz#oU5e_1qps#W#m!SKjri@x!m(%*40K{QT@m1dWt(; zG8Xw&Ua_DNKUb1{q{nh5MCD8G)BiTjdXnS04hCL(m_43f&i$w3c_-%qNvi9nZ&#Bu z2uGwF81-Qq!?-JYB}*nD`O{&xq3YUN54$Q`W~=@-r+-eJ2Dh%ul_jcw)al1mM4!kz zjaK{|ojuC5SoSD;p=+_DY(n8t){LX&J8ZnaoU`+a9%brq^ppXTxR6EdkWsV*@8h*L zQLiz~7Fz8jsgMnSRhN?pyF*4nEOIK5EvF&*!(bL}kFwyS$Bm3v{Zyw9dvG?Lb1kaa z3a4M3cEBx(wN0M`FO6mGVT4F9#$?NL>g`o$HBOJ^;Qq{Pd0+KAi|Tu1WXl27|LF8# zsgA|7MJA#z0ZS|Gh%2(Cg6cb@^z;bhD!e*AI<45%vE^I2(g(12jdp)BCw}&5m@tox*Iec====*OCp2nsOYE(z zm19q-!{9SAnu(!52WB(Mv*UP(WXmE{-pX`+TEP|n<>5B+0^zHWQ8Z_U(`oNN5N%1? zaxZ2}^7uD}B~hg-iOR7{VG`6mm?b49QA!&6Rb8ah!C-IMGY{?H&}iyEGGbw5c_Z8mZBvNdC((+ay{edm*kmBUVCd zBtZ;M?zWuYWZ}|uGunax}c~~*OJB&KwZ*Od~r=Twbvl(awC0GHzrs^9x z{iCW+Owa=NMQ02AJ0B(t5C1u3|5J_&bK(|FpHMu#Xrh7hRvkNO z4008fSqcGcMWy=kF5#z;QDbwzvd3VOcvAZ#dn%*&%R3UHUpKXrq((hrEnDN6$mPC)(?6aFHVb0JAfzF<2`!rqWk15*RW&OoPR%Mc`g zCd}IJfQT>A6+476(y-TA4hWKTaBZa~3aBL}!h*VBFB2rHy5)`z{mQJ|v&w$x}9$U?hlS}24sny0KOVG4{=iiE%=rA{` zKJlxrUAY}yw06bgrfV-L(b^RiC9zo7uDk$<#I#r z=W0*Y22@S8l!`Q`R&75Z604(e*r;l%idt%#x+^(7n?1m zHndb4ze-F0y5a_j#f3|))RGGRsgO? zmUZ$^nerPdKY4YepZy&##zd3lc?=!yiGIwKY9#yh9j3dw3~KzGDQ#5W+39Pl{;U!+ zSu7qaS^6~8-dLCBp zi(jztYa`F0+TzsgJD{@&9@HL+*VsygM{Z?j=g6qdw9n6=Z1Zd)Asl<7lPXKWS%Whr zb`9envi|S*OslwXNT$?6<+X#It*_ZU*grD_Tnzvp92?upHO86?AebgCtI27E=cMC} zx-gAkQjqsqc^Qv16|xy!<7xgIca@jhP(1|MC!re2IZ_wDDO8JiAFpDp)rS;J3Razv zDNliU8?uN7=xxng^p91upn|ToGi5vJyB+3}ZCa^npO%;Zpz>lfBHIGWkT%2=f8{xi zq%`RjAfsq07<<4-U7{VFnKpLimanvzX89R)!gkC6)0_0u9cHgKZ*Y`5l<}ziS+LYi zfi3rMI!n{>R;8IK6V9gwGC90e(QBO-dk3R0aq1TNdXG`KMgC*lvQW4)PV?kd;5BAO zb&BSblcZpbS5rYfTHd$5u@s!|m9@7oXZNZu1s8aw!X*2-rpvtwg?BbrdP$p6%g;8p zkb-$$j=i-ZtDrO=t!hFxemj>}EB&lja2ooasz0@;z6JW>sz2Z9^&41h(=wJ1SKWxN z^E0^R_XQuS(qHV}(f7K|1Qq%>@4+8}qJ3Ge^sDGMt+vKJJ_U}JEe5Ac!IgeveTDAs zOi$)Rcf@x06d8RN8xGjI&GySyYU$fe1}~O^8~xn6d{fIep8O8@l53a6rQjw%NBxdN zITF(S2rrO=IiYq3nzx`!MsojHaC4~Kn8iNoP{Wnk2Z$^}xC0ufHeE_STSq=5wi#;puk%hw`DD)!((`jOa>rhW&pSzK zU&50Up&HubE^k^SRf@1Q`NRDh~t>O*_Au-wen_U?*x`^^PxXR&k?FvF7#@zzZ?B1G1%k0b#v* z2tNTVt&Z@qUsuKwvP?UK#`XV3y&v3eH7pV-Nu9g1ljcc~pB{y&Ev(*v?6`19uec3GzxMiQTpR}f z9c0Iav+;}DxN*^m&51igc3g}_7!4JU3+EwT+=Ckzb1}F{J?OZ25@EG^h+4Xqjtl1` zGbN=4m$tf&iI1`Q2(n|s;Z1S88xwzk{Y~>pjR|MO^Wo#$)QySS{MlR+vL)?}a5Cgd zDn)vxWN|O_>6^2op1BCaiBOTA35wfLy((&>N9-c-w?jpGCa1WK>zVyv_d>R3f`lg} zpyEAKs<;Q&Gc7P^2HE-!MHs9e{;6k@#htjGxdxl7pdvk!TO99t<^izxYd*z#rgU*5 z*E8>6@D^lC`UAp$)I(I#uxEZu^=hFMofpexBuDD`SY4W7PlRmGIHXtH#`Vkyu)`tSGqVw{hYEYfdGL#Sa6PjMgB9vQkD09q@2iLOo^ehx ziaT*V^9wddA=@(!pIIF5dZsGNcPm4MJ>!f_aU<6=-7)A2*^-V&IA1-uo@rk_+7~Rf z+t>C7_%|>7HC*>a9jWkK7(v^cODR|D+BebcZ<2Heo^DZZ?-OaY`qtVhFw-Ew`FUO= ze;w-mfOaeJ4I-^h`npyM$Seb(ZuI5%L*1^lCoLd;l*5F)uN65fUgd>3)Vdu>ZOTwa$Ztes62gTVsYYZ0!hEQe|9-8Mhotq;erL@1 z-B7P5{TVfAL*z$We8I>@)426 z2zNsRwY**aK=TT*m!MWpCJOzP4*lH`tLE=QeSq}6kpBUZ8s$8x2KgI_^g!qah5o)n zGlkgsQ2XvtW4oQX!mq|%qhF1S&wn)R6~;`%mnPPhQzEQkB{A)IcD(fG^Xf{gJGRtH zC1u$pZJ<_7D7LfUsV^4oskxbi+Kf~-2sTjlU9Lom=Xv5mtp-v&XICP$(;EH@bvx4Q zL*@!1{zO>g^87ObSwE{{_oK2N|BXX2C0X3gmfl77YU5vbKjYU^-u838W8Uo6+Z2z9qgoR6U*5nP``)4BZPvP@kMj%D>^ zMEDa(`T&!6)hs>IEWH&>$~-JQd46BYHbA4wo-}}L1DuL5016vGYW#;XxqR8nAgWwm z^ zDdlBfIQbtAewaGY3)b}r*J?yB1IrN}ffy-zE8<+mQ&P$m1+upy+dzMy=6WmgE5b3T z@bc+AWEA({E}zw^dQt_lrRk2)6-r%{Y{R%yI-6J?H)8eX@YY+ShPeR4aq3AM=2nDx z8qtP%8sRC(Hjwk-74w!KHP9}gpQ)`jP`nyv7RWYG1B7}I+l1>4k8>4INv?tPx)a<+ z!Str*bSwr!g%@jQ$cGM(t5LqYSYL_3<&Z7JGK3}SA-q`E(9c#oOOuk)GvBP4>u0Io z#_TnymJTszla-R3O@6-L)()@o)w%XSwiP=fw1*0Zm-CQa+(XoQqri_)2ikf!BFxr^ z7W^@Ul@Pxo%~!IRFKxbV^SNwm^A&HlpqN3b**?bqN1CDL{wKokP}FPQk@z)*Tl9Vkr4>Ppn<0X>(UO*qr{d$_9%C5;-cyKzlBG6eod~M zkUh#>gzDPU-G@=mzAOuKjyJxYob?QM5SzVfmmaaFJ7<(BPD=II_oa-U@9k-Bcr8i` zrJv@UK~~CCAJOTLT9+KBdFyHJI?QINsXfir(5FO)qLOm`I&OIKh^6MeYN=0-R$%og zBn=nx%)~+C^1fCN;HP(TeUEYJcS#vG$zI>@+FUE8&J5Lheg7D>c}Llcuz~8S%U+}{ zWb+=>>gv9Qb|$pb>-!qier->F0%IQ5$lb7C(R>40KdWNzp|Tz~BYanfyKkt5Uf*2- zvr>}2lcYbr)<3;~3hTB0lDd550V*2N%lmz37i&s-B<%JE>q1Xl82g7;x%@k#Zhsex z*VRh9-9>D-w~dTM3Z>AEyUNu)!IRpM?fyOpy`ZT38|t;{+{gukR`carHCyx5>(#|T zCqi~4I~><5aPE4wmgFZPJCa=_8py6jJW?BAYgG0xSED~MIihA-BNwqXaz{0O=d2i7C?NJe4{ZCJ|=qs8(Qz6@$zX3Z6+17Lzjso=-UH!Y( zXR;32)^rh9E~%A!UtO6TIp>*TBz24c1@rv4UmeKPcz6nWFjk+~r4g)M1nboRWsME% z`C3AhA}J{&!sJ);YHwV~D0mmxC%F0;GW&@5J7Cg)$AV~Ty*{~i%9c%dBY)0|Zpi9y z1JVwB8^|0W;(rD^m)O}*``NB^(I+0S(Gt}9gqtgu5=%+-TJP^~u@ zMGnt}Bvr4_fG@+w&4)njb={@<#c_L@*yo5Z;Bq}=uMIAenrbk(AKS)+D0=G6=Q6u+ z8k^zgHCsf3Z6s+Aw!78-c_LX2J;{UyDy+)~Xj&3$47K_q%?(Et>G{?9qUTZ63rL?2 z`QH;+hw!W#RA3mod^2T4R}7M|QvM03zasrhH7F8kN=i;6Pf9`q72l=3j$~rZA^U4s zk%n9qRfZEUAg#=R#hjB*;hgM}IQf#){$|g#s<5S2O3k@d)#JSCW+-aYSJ_FhfvWE^ z7U_Djx)y5ndEY|2tk7<_SJyAYec217KMMJoME2!-vIR1~Vi5YNfx31R)-yr&PRkW5 zTckVrpWx-SxSQtJv+qhU%b}tXJ>hUrhnslU=B#hQBZ zCswU`Mw3bQ=6La_#ZDnRQ+ES9PstzE3gPre^7p{(Af3 zj56d@jNHlmlxFO%4B1oPeF%3$?$oE%aArnkhIyGm_e9nB80a?0R>R>~XszyDHL5qK zi6L7J7m2EIxUd@hHA4$!Zc^7;@&%aOs%GgCdp37Q8RDeq7*18`BkZ=Rt=@*lTQG@+ zqAIo0v$-=g(k#rk^l@i$2gpuAQ+bTdtHwA!{!H!+_=+1I-I=@*`PS3C!!vp4qlVsp zW<-x6+3-J-N<}#o%<|(~$-!3qQj9N#ibQNHlu13uER;eIu4C6=_oUitE9^kn4n?hC z??s)VCr+wZScYW4A8z(?^e1M&t7+lAXnvGA*LP}La4i=Jy|y=O$$1m9S0jhvuOMxa z=xTH?$#+2ZYE-Hf=P`(1&?6C-i#5wDVfJs(nEeZJTY)b(p^Clf^!ooFf9vu$W+T69 zz(07bhu{yq#(r7-1Fxn18&B7XeoHPu93b-qagbS_x>G}?(7*Ea>oy8e{gR; zq*WYzeR;VE;N_6Ttoj)BS*_3FCHmOarRSpdmbkS0NlUAy+RXT_y{sl&3mHYTM=X57 zO}*p|!c8f4LG@YXT_ z;o=(e1epc*56F}&6?22boUJQQva+&B7Rkb|4ckHwlEUv0DYp885;B)?K4cUv z!5xQ^vJ}aC3TEwY;lz?=r`XlW(()SN+mKPT1cQUPe2nCM3A1)9>tHW$$w*RhPj@^?MiHJ583jqfs4XJbBmD~5_3;|X(HFf!b z0+(DOC!W;c$Tck2nu^p$3pw({le)*+u>)K94yct@uBR+gFz#i(4Dujbq(HV;G?Me$ z6$^8{VpCkNXhag<^k<>Y z$GEqbl}W?%!eTv{DpsmjDp0gkuk576VXxRyNl=+y*@uyT2$t3>EokLGRc{`O^tSa2 z2IOQ)SyZQ`t8pKGMIwWq3%`BES0tL3%G7xYA00maD-vI7M}k_6(L?PQ<%?p(N`;_LUxZYbohJD;d6bIAv)?RoZtorYz03a>wBVWXjS`ieX36 z>u@kn9bENRJ$bl?NbBX5+skc7tJ{MqQT+mnf8@C0Zy=sl{KAU-e;TZ?4uw+GOnsn;RZDCFt^W*-R=n zT`M+D9jc#h`T0^=(YMuzRnZ9RGOxd$7c6IK1I`0xFL6tVxrLc3v*Xg#xGW~we+Xtr z?W~mysZq0Nm%K-9ucpocp94u7OtxA6e{G`;aopL!woS1H`d&L?M3`f047VpS+q|?Hz=>AfsrOVdCl_lIN|AN}!Vz4m!3p8eRVzEoD;S zcm32~$^fYb88t9RtbOOb{FjE#>+0;+hV~3|Eb zFUc zy(BIVB6*L)EPS}02$Kq#ba$4#O!zuv6eOdv2Pfq}NcMo7R`0(akcl<`{rZnZW4YA3 z(OBMpZsAx~r5npC$4-VxQ1QbIX@jBP8D>lLu+{JXC_@IQeoDGNHI|!fD>jz9HO`fp zfV~CI!pBqESXNIvW_OYY30FWy(d?I}@?VA75?uBUB`f&*wG7#U>XRbspx=fJ z`C8$BIsI7R_PENN;nlfPZG&p-GuW&PP#(X z2nRq$(d;9m?K=t4_JlSo$C}+Cpt^0XnT<+2TbmHf9q~vQ--t<^N>j_5f-QE z*I|y~Ft{+2&vt_L`#P`2Hrg;sai;1=JAGJ+KhR&S`YVg*6XT!eVcB|gsm-Ny4s7Qf zy1A5%xXq<(#BDA$kEt(v!VUE0Ep_{ab9-?(B~E|Crp!7|zE{vO2YOYF6SW2~&#wD~ zb`<-)=scT|!R-#;SS_4b^XV*|ShM%nCO+W|*c%SBZ%xfznGyEif&uQ$kcHK^y=z|y zx9`L^9dy2~HW0=mfJ1mp%+>MpI0`#{G!yp>K&>TMx*|hf zK=QZ2?D(npR^`ICrgksS0JH|PFBNpmflhd{@MgaeGM}KqQO1*C60npb32t0Pma4Dk z^oK3JH$G27Xh>@k=SF^O+U(f-kkXrrDX}P^otmI@55}l z%Td0xlZSdtUeH0lB>WmOik76R8`tk8zYzYFQU@)EmFLr3xT_Ac?jC=mh&#p3FE~k> zk=_O}ismh;v&&0&ByR}J+N~>Ur`TmT*OT)JCqYKha$e4@RG}uTp@?}4VAl4oqPB|P z`D1B$kZ=WL6fMEJM>zi>`CDMtt`~L;+vzg7wpaN~A`4PUBiPK3$(yz7{}Kdxi54=_768<7%@&CyAfCGml# zUr)N7C!O|sbZxnBK}Kqxq)K;fQ8_jnCPB`k47ptkajz@Hw^qON?hIL>`pxP3)U~D7 zUB#{~@0`H%U%=iWXK^x~(uRk6`gunn@-*9UKt|Dg>T6+!lqKp_gV|q!0Qs64veE~2bJpOO)62`>ipX0VSw0Vf%T zuc~p$Uqm{nzNgbyu-R6=DnkaTez?ja#b&&Z)P3cJ33o>e9s=*|9 zU`0{}Bl#0ywnL6RZ*$@6if8*ZCzH%c{_)3>l1Sz;FFC z3QDTY=6kD3@>%kOj9Qq(FiH4BIG}r=vd#ck$xp_H-NwO6|}*D!s@T3 zS+=PDlcM_b%K7p&D(|qPi-56vVr+t39bs)IS4e&tS_g=FFl0 z>FDRl!iup=wrazQu}ii_itdtK>mi=v1%u1`G8#YL6)@`~dii&x=x*2lvIWtMhP26* zxr7TLqZZ~u7?*H9O7vRhue#U1cxfA;_qO zc^$^VFvgcZk$kf*a=h^>9W0ybF#b{b)t$cade#5bAVV6X@;W&E)##%e9>so8s-pBK z910mlnl>;#MuXvbNZw?at$`Zy{VSrqd(>q03;Q2IMnU|+Q#V8IMDmuxEc_%3S4iJv zRe6H&8OSI|Mm1w-et~31=9j&r8=jKW)1E+dw)ILpoRseY`p00_@BBp+W7KqJAAQ#2 zAnmDg4dQ7}l>YhNP;WSodxR9-8`iCwAwzNDkAm5(ZeFKN*soTGOh)C+aQe^CM@^{M zk-HnnZG?*Au$@)O~&kWr9~I)^5#^?cNX^S{cYEtv`bLlZVR!vY!iAoK0GoUe9MoZY8SslM^%xXeW5-Rks>)}-1)vA_P@RF)C0gp4B1O)zc% z-;2vSB=0qtEr=R2w<9`sbW{0|a3^FGB*Ap@KY--@2($2mntxIuJzI2OdlB|OgN%Y? z)MHeq1yMVsN_6<|L+Wk**6$3^?*p@bzte{U!M0c9G7^ z#Igd}L3jW%YG$5RwgD!=$Ir#(D3UKPMy_soQcL~Ui*YH1%B$-1yU<6YQ?bcM`3gMY ziI7pGDG%d=q!^xz5W&A!nvmRUwb2AhB69&18L9MJ)ekXox2{4r)Hh1bhA#CV^_c!hs)!V z^%5;o8fHuMiPitMA}%#l-#uNQnss)4G#*`=Qo97tE0-&$0rti^i}rX*n{}$E#T7H< zGQz7Mqi6}v9qh{jByR=G&I>O)k{1@`hbz)XOmeX{tQwBl(MUEDZb>&V>ciX%V>1zL zj+!si9QtJ6{zdWoi%mrTwX;pBkJ#+BO!rLEWP2^sJ(Fz2?U|(b*r`cvrY##pn&Qwq z9cG*9Y4qX6`S`8vH>yQav(G)Rgp47a3K=yw?UeO^u?hf}*CP2h!>r4f)iAMV8HIlU z9VZgI<$WJpf(4iQ@+4aCc?W-AKmVCjWT`IQWfS36$S8;}fjr~OH%P2GJHA!_zQ>k) zvNRhLFI(TXsv3^3*;^_BuA6RP)Q9;|U5>hjR_$&R2Thje_vb}tx;bSZUrj+q-B0r9 z;>k_(Y;tfD-FyYf-hAhOdz142(9LDkSyCPS*0!CjBzz{_z^D&%9ZZ5Uyu$w;lJ`|Q z!Zu|Gyx80E7IRBCvL4mTyLEcXpM+T(qBhYY7s+TFaA!&q78)GHN4Dz8lF7>I)nH*C(FA8XEr?K(ExRQOy-? zGbnP5&9S+u%U2TRr9R-6=>|r97(pe$h4} zQ5WWPm;@JG;>*2A{#KYhkqu>pTaWGdD?2s$RBdYVsga_S&;Ojy;+EjLX};{nk9Q1a zeI)V00UJ&}YY^e1U!?sod4OcSR#(bEMlH-YFt!)pn3S66yhbqV;0AlVDekA*1=4}A zD`XTT!9zFmwg!@a7R~}YhRpCh9PMC#@g)8KxSDVRqgoh!cAXeEhaw=_% zM$R}la{gi76HM}@K0v<{%=#S}wj*~Xwc=HF+s6K!;We0aZ=!X{MZagGe;Qu*X_F+? zQ^&1UceE9lAz6O-d{uV#-<~z zOZ}_S9RxkCNA=R}m)1MGQkL~egxUgYAeo=&LIP0Em75i*KqgO_1(sYlc+aCVEx zDea=J*!}fOOJBl)kWsV*N&dPy7sW~PMo9O6z3dUoWb3 z6BU(XZ^C$1+dM8$s>kP@#}gh^eb6#4Z>#?MbbV@~GN@Uxwy4Mc?7su{O1&ACZ7DgV z)k!@KoZD8alU^4xil%EAKOKmAr^D<-<;3_ujh}NL)={p81?ze1gm8Mgfl(i(2aLZG zPNhAvH90s7X+DDY7|gcVH7dB3Vi{W>=q!KC)W_-9q3?ZtTwWr48!~EU9)L+O8!sOt zc|XDaPcLd&Vt;#yzr~jHkWn-94NQUt`EfY`o!1dow59{7TJ_p`qcKSEb(`VrmuT2U{s?FiXmBf38 zyM4&0h4~65qvB^JUryfsyIO#+ zziAs5V9(iQtY91E+%v06>&I&HPSq{d)nDsN%XCL>uCgTV)kQepskN4I_w0OWty$c5 zXLXf%jDP#JJjNeJ9sS&QqYL2W#AsZ$Cm)?unHxolR-I(84b0~HOxTF*gZpi)ewn4( zh-%;aO>O`QPlJqFnEEhD+@KjuL9}D>x?c;Q3s~FWX9_!$WbZDR&ESsoUb@@%lG>j& zshd1XxCSz6VQzv+@JT!_8<703Vb;OXCCrQ5j$wH`$u8V6to85AiTLWHAD5qjFWi)F zvUYwoDflZ^b?l$M4inbh|HaM5%)m=(G58DB?#I<6O0%l~n?(`33b1(yY-v{kwur1G z@dl5l{O#nLzHZ_f;yL%t{6AMLK1ILAKT#uWuae1CadTan3)vN)8p&C65^G6Mo=)2pjwd@fR>o@S;0PDvb^sdd z(3`WSg{Ibi+J1K{a<$)6D{^hbt^KqxxAxPfxV4{ltR&)R)0QjA!?vZi3TxmfbY#LMU#HWaf4ufJ3S?zMI{`udA))O_PLs%49*{vPerT6z)= zgp694dN2vPeoy_8yeTkS&Ot@judBg#XSbK@2ycRnqM4&(cNda(KP+V@NrgOgd3Sk= z@HxmRNP-G?m6JD+{P$rN-mi}~^K6=Vm+B8VeK%bOA3V1^DMwLxIUhv$7U-#0cx`N} zprqQoy_sEFz%+o2TA06KeCOmZTNa{?nj0SlkFUNY3^ubpP#4 zH8=fCb#rv43U)nD{ng#e&fP@nPmhKF_fIDx!JHjQc~3zfJJ24zKnE{uPs%=2-XBi? z0s2&TZ&y%KZGKzXj4yn#^(SQ1!u$s&iDy2dN!z2OC3zM%rp_|*Lmq|PTxFh4zvN4A z5dH+1ZHJ8^hQ)7|GH(_voQ zQ?nbLnV)V}iAYhi3j5XU-+O#nh=sSxnJuO{sH$D5xeQs7>hY7=73D?3w;-biW+{wG z5skYY$vX(MrM)n;w<|R-w)SdMc1?fzlQ8QeOx3`A0ppK8E-rcKyc#fTe+wm*q~bog zyuCChYzr9$NpQ+foI8=cQ(zW;u7xY4S!PukL3kcy6ePibn|+yzqd9|F0Za1mq_B#EWkGFuUC%e=!8 zksAN5WmM@}M&;N|_iD@R|A6Oz82aDBY)0Q({hDpQ{G$3QAE)V4E%W>P#abruNm3dD z_BuL?QFuyg8TGVaQau?!I0!O|mS7gHMj?3@!fXj1(-I^V(x^*ixtj1g$S8=_=bL?* zkK`?eS@<;;u8x|UHn7M`%aec6}a zF!bX)qKx9UWuD#OOKDVon{<7uW&Y;ub=;Wg{vm$|0QQDBi(zDw)-vkpiQ!%4e8NeP zQM3d%;OZ(QZ#K+URZp3cRLCh6`pR8|_d!NMTywbuu0-&MoB)!cCA- z5DPYj(K4BzL@lFJKHD-ngL7j>m9Aw}j*ZvDHW+%$m$Df8)nGQG%~oHRQ(C_2PfgdS zT4v(WVlC5SUs8qx_AYi7-SCvwGV1AFK67;~;f;_{G;6?cwGhc$1+yj4^N&0KEzr^= z6@K)_?(!nxTaZx$vlPaw5}c5>BY6j5)?QCY*xU2p7;CRK-nx?VCt=o3Ox3`A0h0+m z-_4V6|0rD%wU-aGE_%%Qcq~}QJx$q$eMu0 zy9Q=G?%^M(w8l_)x3Buit%P?%M$wW~^}&^Kd6;ljN`2m<4=Lk2e)1($-Ul%2a@5Qs zE)~CGcrDpOcmOhrmZYk;+}>G!CHy<3K5yOIxE$+O&ixdZbzs(Ic2Sp#@3pv-v?jeh zWE3q)RUfF^L;4XrEv25&dMhq>mmZDAy9{PM?kw7p3jc#`fd54}4>F2oAw0Z9-!`&@ z@WGV2;L^o;@+8scA?xj0yvcc=zE4JNVwK-R?d^7H(~9P#HuaujUj)KqkWsWGRb4IK zLb5)i@1Rf})YzJoDn$JgU^c5VJaUm>68cuEKgsDm)h8a-sokaM?9}cK?uG14XpM*5 zO(;{lRo~Q#ZbF&bX@svjEcfL)bvMtsdsZi(K@$C9)jw2JKk+eN)~NoSbUlY5lY=MX z)%IEAMnC1r4h8LTph*li{vSD*CqJnEH>c0C`d_}wldR9tmxkH8T7AKND}AYs>J+Ee zvT-2sbfjDPtq*hEzx~11Pa!FOWd_?=0Y&w+GyZj!DVi$~=E}MSHwJRMZ*ztWH z-w`LTy0BC;w64=SDce7;A1;T?f0TQAlD(b|qivFQ9INfq<4Vd9!gC;_X#UB1(3c5F z-Ze1m`_`g%ik;oIvD`{{Cu9`OSDLv$c>u|K24?MQ6tz=qy>~mvn}qK{M$vqC=-8&86XdYD_ z&XcZ4UO$+%vm-63km-Nbk#h(~Lq3i0@qf zkSEKKywxxZuTH6Af6U>*S5k4WU!IUx2;YT_BF%Cb&p-8PM$6rdy|iadf0LBs@8;tn^4oWS;}7<(E+nRORweQ5CT? zdcTD3blz2V-W7xQv()=kwSLxFM;6w*%98j?`?}noqJ6y|eA>%hsn^N=r#HM(Z;~+t zm$wzEjdTBGm>q(nSP;YccC^&|B3VJdcx)g3)%-G9`W#-F9+9SV=o5- zv9XtfL9ems=E+-XNF!GEs>{^MUX4i5g>^W$BYDeVHoKXe63H%nHgf|JNvhlNzxcAA z@O8+jg;@;aVGx@OY(w(@fLRB+oj`bf;8yh~B1P{xwM2a8_JO37_zJVKFl#nivkJU_ zvosKuccRmumR_DAwmfPx_kudohwu!@sD)_+lSJj0*<0W~M7yR?`@!DtPeWt-M~n7< z?k%8SoW{@KiL~d}>L=}pe${!gPmlkXzNJcydGYE9hJMApQB@0HylNl5ctx2mk(%vu z8ry}R!BXl_53twVSxnW{NJ(|j;e^`Km9Q6N6ePjVFNh3A@q-fcNMv)DEz>gtz;G9Q;<=#BvrjJ-br2~d^@EM*4|!O zJ|^my+84R2h`Ylv{eI$@*oS`E(YI)Hu-B64M}D#o3zZMYrkz)w=yt!<=ZtA7_#RAU z?7OG?=rPZ`U@E9jyz=T}o~NH-9246aMyY$-MZTPdg*VlieY-HNCOul(FAUX+Wz9fM z4b*$badEkU@Gi)x@zU5Bc(1WAJqu*d;4E;Lo&{b+W6!Z_&3VI~1+1rF#;v}*OR~2M zW}EQ6JJT|jq}n$BypkLuJOUX-^WFMVIea({yF8e+`)+8}^y|4|^G_-z^+<0F8AbDv z)%onfhvZ)fvvyCFr@@}w)lZs!9&4~uBP$nG)JVZkbJlmy3gg#*(cXK%!P7M#gQ&V0d+zc5tGxsQa z1SW}%ojG9kBUaY0wLh~sR+UlAaC>h~ML7`U9wWso@pV-8C$1^dR%+J2d~>NxdQHeE zT9T^XRJWP5K-Dg#PTZr1=5*rehvrb?;Q?N3BaYO|@p_)6r0cb7p8_8~`MEN>SJl6+ z-My;*Kke>S3;(MQe!0w-S1|P7f!R*bf1f3IeY!6@RR2f1o-PQV{N!LpB>l-xrPn$C z9iU07!mKA<=#)0CQ%}E5JVBa}-U>2`=6l~PNbHK_4TV`dU69Dow9o23LYAb$-&;{n zE+Cu=88t9{VG``U#FuN4yhSi;e}8Ci?^W%8+)1^$a#ejB5|N}@mpH$r{6T2GNi#6&!+Z#1JDiJrDMzAT9cDf0 zVkZ`No$gCx)wgx}FRnnJIH=t>1f4xH^$bt-ndi?bd%gB=A~BrwX~ol<=GpY%j1j)v zLbCUe#hf`=QdQV}Rk13lqZXUm$jgN9q#GFZVeW#_s{D_z1Igb5vsJMFBTV?}Wd1>~ zko6N@@!Hv}@C@KVUAh3N~ELpP~l8TB<(H~3ZQ2EWR&(%SC9u~S&U zjiLW0%x0}Gy-Ki=mtj9r{jcfz)Zh;=J8tk#c`+&GJ1S8LW<6~oo3u7jPj&Wml}4mD zhm4|`{FnEo6OuO?X6*_;Y@PMni-Zi+eLqdQe}N30gdplxM+s8(=oygO{LBs$G7QuCkSI2V@i^i80#5RSwd` ztyVX(AO9_RVkFp=8|`byllnT0cbha)ba|deBcs51-x!yY9~cY6st%}A7*M2=++*&Owb&9?ea|2;?j z3qSQr6FL6SFHdS>?sbFN9GjD)`w+L@LAwuet1L;)>XmV=Wf0*Q$f&VtsH_c4f<=4_ za|)8b5M~`lr;PB*yt`bnlhZ^F_6Grb&pHb|*we;Sw;tKOP}&3f){CP9dL6Ihx)ZZ1 z-HBP{SQ&eIdN+@Sz!>@?V5vIt!d;_0nW_2*)$5E#Wy4q9^$mA@)!iajOr}HatL|fC zomUe%ck0B+61yN)W@&vM83g-ym_#+E`^Muc^MpDVR?Odg7BG?IyjZ0Z6xY%hNX(eqc-aG zndh=Qf+@PS6v!|ZI_NW>cV8A;$OxC zp7ccW$HQz1=Fxkp2}hIt_5z+h+dhAFQlroMo#XSuB~GxsqpX5jLFA@dmy6*W*v-2(%Q{cevXdmedK8`(nk^%Q^&L_ z-7&3ltbrYg6~AWtYz+NYuv8s+JNF3tN~7|wQm;A(*r_$qcZHw8FnnEOj#ojo;p-Z6 zyk;5+r!;fD>T7MO)$J_{16fG8GToI??`Dv)k;*2)B>46#Pu3xMufb9c!TXOyvqomi zhlD#JqaX<;vW5KtB>!)ig+~vuc*>EVTP1wTab~Y_QsoGs<}iDq&{(+2(Asoi=~?(K z+KRdeO@e{E!rnHRNHe8Md=C9zO_)<0U*r%yawdHBz-_@-`E?XosyRPX;$ z_8#C>RL>vx+}$L5LpJ0F5<)@(A@mR+bO<0Q9fF8TvmjCx6;K2O1r)`CiYSVRh+qM+ zq9O=lFJJ>iMeK+bMa8bD_|Mz|G^UUm=Ip3L6_w3%YyXSnsN|7Bfv@2nz z*h3iWg7bWsE>PkweA$wd6f-WsbRVSp4GdN1ss5rsuLGI5QPzuDsQOibemnZW=j#UZ z#aoso`6Gv`ZXKf@UR?#z{DMA&U%rpW3_CJetLm)rG7wce<$5kcH!;^O1kB1?@5ZmcC2 zalH6Ws*qo>5O`#T`#XNs!A1OQy8tRy{AE6l)lWmov()sy2VrbBfqx7tdHxwz`1g^K zQ1UF8fyBms1IPzp--DR&vKIO9KSs7-KeeIuPm!VZA}3nWVi7OT_JiG}oZJ=rz*OH| zWW^-351R}`hMJ@shM9~@9y&}q(%zDPOww6buc6wm!-qM|E=9;ZiTf&lHDV*Os_e2{ zxs51$EzHz;ZPT?*?n9ALS+50pe~AR|qpr>7_FrwjomgK!oYWF-+`IMmb56q;l$%H4tvjqH;K#k`A z%2s%XX>rI=Gv}bPKVYm#q9aIOV*J~8GqF`3t}t6}R+#16^yfr|g0g49Op5+jvfeFj z{xhOOVdCZR%at*?0qJ0FgUoyRb_BFO!yKwh0k_7V`%q;g=OR& z?Z;=48j-sZo`6hROAuz=J-=M0Y>0hK zq__fq8Ljm}fH_kD2j30xkRN@o$8+Rcjz2+;qCNG-8ucaRFDAwyPwi%qc5Q-IOPEQr ztSz?)-CAYHlqsb%?jXsS!iWAEk?tG^K#m$X^GtN2k>>3YTcFtWAZ(M zW&Q0t{D4qGEqy{TrBMy3Nqj!!D4HcP;k%Y`1B9wPRJ%iQcN9UZAI#)W-APO#*LLbE zBRQT1ISS&-))BdKK9Y46EM=&C%uR%`%H$l(o-Qdq4GqQpX5@Yev+!=BUk|UK3YR^ zj<6CrD#MHr8pA3@IIi4K{J8JE8gOUFy zSCRkvqsjjbiqQP4agcvCV;zZZxw00?+NzkL@@OAF$*bZ?s*XQA06smgK;Gi`aUATZ zwX+^ZlDBG=_0l9$|M8Pl;~+^jtXv zpz#UM>BbZOA9EZ@cUTnpHWmO{0^HsNGu^|#x_P7L=S+9~YxKPI820!AY`q^?RB+uP z*JcXc%#y{_*)2U%=$ zf+OI~i|#R3+Ut1myc3MIA7=9NR}LK}J_@RX9p9J9PaOY%97VG@Ep|=>yKGo!$NU$u zL!Qc&64Xs0M?tKJc@+1*NcOof1K(W3+}`uYk$n<2cC-FJ?vs#K_S!~_IcYjJH$Bw+S64XGEZk$gt=1n0*4{@1=Q3PBDgtj6<@ogqhmX|6blYpFd~U)}=jh$se(wv;D`-ZLnCX z7F&`mO8)LA^&j&r>eN1^j|SSAM`-T&b|Tx=Vs~JnkF>>Y&&vz@RR3F`kN@vm9P|J8 zb-FAjsWxS<4Ko=$gOVlm8Z|!WU0a$F-wJXRE$-iZ71j;O>I)0)B&Lwdvs=mtj-wz) zLE^5Vpl2i5m%_q&py*Sjeu-DL;xa8S*DL7O0CbKPWP#nJ6*uv2i!71##8@u{(JRPm zaH@L^atBfOepf@@;kX}i)WF#cV^0{qRy%}bJ1Kr1)Vvo1NXpDBDuud^#YJr znQJj_q56)=`eD6AjzuN&U}g$HyGdBWInQTU*M{&(bT#VC_LG?`Gq%l-E0Q{yZTd;UvvytmFPuF=#j@A1zAS>&wHYoVz15)Y zU5_=Cg6Hg7BeMy)gGl65BJCAb5>a*L!Xf|P6IrB*ELVf%M556@=I|1Lbvmd1b70J= zuVo1?UYfRKy*so>BB^PVwfcabqHp6AopBmpEtd9Gamw`bkvm_YGuU4p8$s@?kY$kx zV6LsMvSj_EE{;laq0w_%dr);ZH%Hy|o;AIszGN)OQ?TA=sWPjP=I}B#Pn}dt^zERG z!;iO?+QkU>e^p10@W0o*|5LwS?QVMVn)iR|7uU}7CSe|;@m6DQZBZZEVoy(MB3DJU z$xK7d97LgMo67uw4zCZ`1(i*CAqa5iaV7E|$FHjzI_lB67v|xk(Si(WbF}6)=J>U# z=>)Z@K6rP3O@_1~+OC9|w{|p?d12)Jr4T5sirCm(zF*)o*Eg zU8*UZ7o?yzK?>>;)Se4Ltx?pn0Ht06l$wfr{Vf@?2FcnCGnMdOtE5*n)OPx|`tl0L zU67+_R=&R}Lq0;Xet;Rfvy$x;JDnx)t!QKNf}Nt7jlMZUY7w-GVa9F~cD&q8@SVGB zN^6cCAxA+h#`u&gy^-uwU=e7Fpi~NoFNPdNi+jgJJd=%(Fz=zVve;PW${-$a^m|L>|mczi244D^2F5v8xIF zId7+(7cn|IM)5^-(wDW9b^~gL9*qh*sXAhDqz#2K5Xl}2GnxIOsWxzv#|mYv>dy)E z@2fuRqHhaX@)TY6>5{k5YWzSAo_UW^v8{@Hg>`qKtWd)>f#Fo+FslbE)~bGQpx1HV zyM_2Fr;ZLI-p$2VIi-4YA?{n`{07 z0(TFXWIl>y%6F>&Jy}06B~#K|^f8zz#vX(3l9nlTRo_0)s~>9Gd)5AZ1A3y9Ej<-< zQUHp7i~O9OGfY4Gbie9s%NYu~CJ|J}tvp7`8l}|MuaAve0?o{PoI7_-p{!Pm$6@s2 z2y^BpYp2l(%o|;KzeifCmiAhRIup=k0vQJ1Z(8 zZOmDbHBTdJXPTsf$nca|`m`&JiL<-GOeXvRj(S*?(%|cU+J-xJxiU~eV*=3N0l_H9 z5=nQws-xik=Utgcd}ImCTv0TX`LRw{hmOs6pNCW>?qSK4)j+HbFoV!0h!1g#EcG<_ zJRbLR?C~%*{2ucT5?S#}E;&*ME&1or5Gww;h>0K{Ckc{gQn#vLc5Z?V zQOzUP#FPeJ+*KLQ<;eDISaZypK!e&|Um~N!M0cr=vuoOY&`u=!bcH-Wt!W5}h=Gi8 z?6=WwBziqm+W*7mA;c24)05KsoahZ%lyNvwc1I?q)m5L5;xr_@lG*uznn%IZXUP|;ls9pZ2&E-rV5|LFvY8P4Qc2m?(62BR;+Y^X%;vu94 zjv;U~fo@Rc+rH)OlXyj!pUM0bUunxOKl6iQe&#R4XeMMbAK+4DUqa@eAbOKRH1pHK z%*S{j7G(ZUqJM)-<}U~{-vdaP`SzXJ_!lynpM@}81Dg5G2#-OPpZ_cK#jy>IWp0@H zKnk9FXe+#0gM!or_5%*She{4ih})!u)`3mgD{T7v8<5%+mehiZbt})nb80As_$({E zqo$#ocySmw)Qu*4pb2dBtPd;7Hn#L1k=ADv#E7} zvZ~^fHnk3Lc2%6esdb|+>;n&(O|64Kd{gU^?ZsuGZE7fmxy!PO?e~BT#6^FoLLv%I20tzc5 z)>RC|b{?>C;7@_<76f)7yrO|(0_~V;Y6F%0czQ#*DJ(!B1*bvbX>3dV7m$;|4u|Yo zV;agbgryotC-5S|^H9mE>J36qkx&XenQs0dkb_{qQsd_dH0#db4%wRs3_~~(s$8Db z-KE`A4N4`@F|2E?92;@$ov7!4p9KxtMim7Z%frm6ZoiH838FVCqzmN^AdH^4(azEz z?H@${0ww&~>7{b5pAr7HG6!Xv!mia=y7XWVL#Un0RWZO|tr)7`F3^i5@wY*CIf1OB zIjteP0f9RaZiR*@zTJad{_YR!zlXZeI)5B>Iup66kiDBgAB1Bx@FIa_2sc9QmT-X# ztRr$U)~23UqkfzCH`HJwff_w6sSerq5*UXt8hX=uJh`GNzIzlmlUhcr7R=mXEJ(V; zh-tk{9@ANx5Z?lF6wSM@zq`@}$vPHh>}n<3DfY{Ho#Zr*;~_`U%$gl`Wg3z-2WITX z=zU~NA(aKLEaG@0~Q5xb@yT5&MU~i*IoHW^}i(RH}9kU zYis+%Oui#uWel&t0Vn@hu=n(Y#{%0qu`u9SbvdH&rItDfY!SC2|_a@sOiv zmLmU%_D8bjz>M7|*wOw3-5IrI5yu-LM?vB~O9S75WZw@n@GDKNWqsqy7S+EH=*!49 zbMpsW*{%8miF)rp3-{_B6%>u0uah$^^$$ZyQ*bH>0!Mc_EU%Z|9NkW6&)3;%~!89Q5+Rt6NOx-OC+<7G# ziGGFZ*Cgwwf9A>tRMxgceL{mN{;gF-vX|pWkfUh!zoRA(B3ac7{Nm`D6xzkVO!`6V zh`sQ@J~$~;>H%*JIcj`+aiTFCSbHMb9D^lO`G-*^`NcdGwASzn(Wl+)^m}%d00ebg#;UhSmVH1vOkbN8QZ~+*R+Y$rPhsp!(~Q_0P3;_%r0F zf%7ViCQPLXqtQy-WSy@KSzPEh3Ttf)Q#syRgnP%5+S(oYdNk!X7q z%+%(u16BWG2UpHk{iT6ki(}5r)Ma+zNd(0G=BW(1UP1Q+pmlA7GA~c7Rr^V1UEO@1 z&~H@GGXZFoNns-T9jf0O=&w_KH2O{B8j`hJTP?T9Z?)%`fitZ(MzU7IOga}OPZBEjf(7OB5XZ+LN6|c1dOIS|BiXORjNJfJ85`cskoQ&pW1v6O z-ml34Syrvj-qD)80%NHQs2a=|pX}AxeVNRtj4tE`q8~A4i{49yXGpgpzjQM%Ohrt) zIUOUt)snz0aXLkg(g3#_ORewRUzWyB|{irCFQ8O!`Kny)~x|6xI4)DD$LQ zO&Ji(%y1t?=^rlRD?T$kHfDUeGq$**f4r;`tAwKMi{tvLef*HGh$##luQBWa&aQp{y6fN$o z4X$L?$1V?M>~6%4wOxq)v{?gbin=A_D2Qg7wufdqQwx703R8HU$P1cDq2m7ank!?F ztm!c0RBy);3Zb^e%_`(_j@LkrqQ&jOZQgPu>j9XtJHj1cf}LU?Y0*QT;`jpOC|cZK zU*P@+$@&~-?5<0;Q*5*4{pDAVe?g9->DMoD|677x4$RoSkDbI6a-@A-X@t5tg!01pPjNC=&FIWBIK>xAo zqtOj6HM~Sep&i7SQK-qgN zjq6_vCE9?qs0J+YPGXN1$1S_LP?`{9xBgeOxuN*kx}It{Dy>;h^)M0^)lE zsDswfOIkyRh%q(v{>cBVA+-!^NCUL_+gOnBA2TE+CaatxB97lk1MyT{_a3u zX7n@WXULa$X6a|baTow@Jpgg5pp9dfU`UHMEw-vywdraFIH>&joMui3;;**Er{Zk?o{eD4AML)wu#12Y{d z{(ydLOl6kGTw5r8iMEFXt~NAP{d?$7Q~jBNex2&GR_L?Oh3NcepJRK5&puU@tmkzY zT?>FAsb}iHhS808xbhfr*7m@wW<252poaUe%9eLIehfK^=2_B8SH4BEq^V!vlw>=_ zF34~toA^A)Q8c?hEO(_5L90_>cN3*fNKdg}?k|-790x;=qS>T%g)65bSrcKVX+P20 ziYerZI(6g%j&mVLL3~V; zaum(xYE4}^3(1-bGj=nR?G)Q{WnHj9Xt(?7TpQ^+-UX30|= zUw|A1@hS5l?thT14`2qqu)D$)vTa|9e9Q4C$Wf5EOS-xeX%0LCX5e+SF=7gNae2NJ zqAr0P1&RC2_tm5v$*O=E`1O^%R1s6iYx}ZfAjct)qae?VAP&#nz3ONcA_wAE2WB`(ND$Kz1%aXcI*R^&r20=7WWUL-zCQS4wmuu=jC$eGJl%4 zKIexHw-2PGPS-&=rXHqDZY|Lkjin|O!X5SC?1xF#=M5Q%+YsxYYS~l9Op?5dBelJk zX{K;C_8`_C05kbmslK018F53QoQtmf8~$zaTWWj9$clAcjk6}_6v^R;=3H-_S2H_* z8}vkE)W?jyz0!`=_>a)M2bJpGXQl56KTq-Ows%_PH7!ZFlBj&%3D1LUy$|h=&>L!Z zMdwDL5&!blsC!x+`##i{5kFrIRub5VuwDZ<5vax+-BGBt)d$Vx+%U-=YW;nr&qz(O z8&Pe*ltY7_C}BJ?w?3oAUa731nYHU>?FusQ4ioEhuGLD0$f@5hpruuqN^xuzlb^ zf~3m~zrP!Dsn5_>*;Y~e*7*8^sL#QC3fTtF}n+1W6W%Io}G zt>O9kus{94FT7RpJ#MuvqD^H*B<(xTOA0lm${93Tp|AS~u&1NZ((T_5>w_LaH6-2M zb;Pq%UV11Yduj5_GhXG7p&=bd-!C~r{C-Nl54t~+gK8I;ez^#y0$2FC6NyXe7(dZS z-AA<%wF1*wqs)P9t-zKDWx8ltJAG?qweNxOvsh(B3e0OOBLED8%Jt2U0K;x_KDyYX z0p5RwWj@f0p%Q(C7VO=Sc_ZWvgq<4DH$whE_zkMu>U#-ZXQ5+J z>Sui|JLzh6>8Xysh*Pdb3F1;i$xl{a#&Psz9Q~C}%GT9$yeJ-R;zrN&zuHn)mk?EV zpe2QneF1@f2**G}rUE;95t^F`T@RV+U+%jRiA$d2(wfooQgx}-zaRWQ$m_qQ+J6H) zqo-y7^lZ?k&$iP(8^qftJ};@C+S*T@N*?z@^w&jxE*{nX+Su=c+Fx(SW!PU^`w<&_ zCH-|+--Q19MX3Kj`fQg;zC}7|`MsX(>#MSN`nia(oupT#wvM;@ifv2sSqil)d9Hbg zqe>LVTex;R)MpSsPBGfbfiKoy2a|U&#WAm2-9h|H$h@%>U{a<0IWYd~Rv!?(S0UPO z1BfG272@~RTI_RP12S(c-4o_|6_Buh9!Gp7Wcufg2#YnK{qrq^T~Osc2fdaK`=^v> zH3StD38m^D&h#=<9WEQNSA$BlIs&{EN+CX|n$}=jDPF54z}uk|;KOPf2KGd#%XGi7 z^yJXSy2$UF6S(X>;uoA(8?!%VA^JZED`yv8K@%^qsbGdct0w1(3t-8IIKmf9r6KHrGZ@xee+21rOm1i`;EotdT zcO=2A#-(yGQ5Qjcc~mM}5gymTA3zTy{0i~uHBmiY+NM6667@AUUqE&}0-cWMA`7*fOE(XUA`*K^BVI9?wkKZMiuvw0{=3|2$4yWVa#ZR# zd3J9C4fzbO5E`P|e#fKnTU!KR0Bl-*J(znSujThC zj+XDz*>CyAC(?+JY55Bf&eMRl{ObtYAg|@8Zu48d`A|z5L8j&3@GHQUKnjYgEg#cF zwB_f6zZmjceiy#Vh%+t!Ao_bXuCkhc6-#LO`o0RU+sQaB|E&7dmOqHiSCDD>iea?6 zlD2$c#Jw{{Vas1Z{H2g-`L_{vLOUXj{UrWJ%jfvf+VXoz+iUq@eSNjId^LuM1nd89 z{VtpRVJGPtVQR~~k~+e#Np3dDw2nOcb>z{@c*$_iUC0dZ0gn4i#rXsLDMXJ@hz{@p zBvlqG#2?_75`6Z+^~S7frJITlK9IYQ^3~{wrfDIVD3r$OM@zh{;PmetR=j_ zZ@ow;b?}O8n(8S2q?pq7!ox9;xhMrTtU%&hhc-dUr-GdXby@JQi_-P|a9^yKwxHic zLMbyL0pxwJhWXK?b_E?y)TT^QCG*yl|CvPy{rLeWf@fmWiVlE6Y zbHIo@bV!C=h03}w&`&@g<~9Cu&XX)m$>fEapWd*cY~{EEa@6?t$gS{xD_}aG=x8O4 z(n|Unjj1HH4(1b$C%69z8Io}n3Accmmvi)kVv~E#0bpLvIhebT{2`B%qVK9-f=h|1 z&)+`Hk-pdsPqcN^os*-i7{)Ge6(Zx2?5kkLbNmC~lWOGo9{^95`X|*W^FG+=)LrCe zwRj@1&`*54Z+@@NDK?)C>K2#vug+<0EV7(N%+u@z%{V8q-1%PFUobYEXJ)@M z5!R`JOP$&Ozg()DsoupRGnbOl+UlXfdJ48kngnogMNXCiZXS z%{g`AyQmFBTYCeHeiGwNOno28RZzBh`uKgM!6>j2Q)ji(Tgm~Be?X3!J5MTm1xBwN zkRkc}rLnrfSf{+Te9ePwgf>Pn8ClF{VP^8X-q1*%MiZDN1WGi(xAJ(3tZXBraWVBj z-8hfzcv*0vVHOP;=qo%IsE zCkl%N?|e~O6u=@ z@MA-auW}-L30YsljPc00u^LS6UnswGw0mKvNaqunxDzqXMrSpIg@)mEh~*y8&t6|3 z?KyUV97W57%?(Jj$Z-Y{vPQ#%ijx!;H=Ap`3OBN8cuY(!8?PeL# zi_zbv`n5^=XtWz!!eousKJq#-rdRsA>zH?w6waCY?4fFWfUT&X0@SzO>QDxyrs5vF zFQ55t@>~zb0HYrtnk=N(ER-p_aD4kn^ZbW(-Mg5&85t{)jzEuz!yUDEvSEA*eSbc4 z-ng6W#ZNrjU)D6q-IaZ`>#2)3AMYxcbG$wdchue)4WkSz#&b5T(K2jOgnrnlE`tG5 z%~%cktlIJ-lC>9RGT)B`66#%T7x%H{0LSkkN70OZ=hT+Jku0Z=Z#TzeOd*Fx*OA(& z>q3r#_)>dPZ7CyYbqL^^jKmajPez6GN)eX46NU-91T+vZ3 zxg1b6kWWwYV$5<0A~dxAb;K9O0B+)r9R}t9kq83!ASJQ<481X0*Urjgto^m-QtO= zEbfILWy?^a?bBc;wXqede;NHbs-GF?k5_#(I%r?EWG&Zg$_8T0HKkeSLhqV#Fjg&o zO;N+}nxX*~q0Ox(&mq~b!AxQYFlCL#oNDr(>OV`;SA6+Ylt1pg4-AF)gC z162CG`2e*#0)7T>j!Fp-yEDxA|Ck6aIo>N4!#3cqqCTKL#Lq~?&`Hg!hzbc`MI1*U zconhZqu8Te75 zHB9fhIj@@Br~2mty*6~%5G!l&L74k{&1$lXcdwl z0J}1n$&By(Y$=>Yd>wKW&E+4v z_mQm6VJ3aGWB)IL$Ma^&uN?n^90hSh54><7@E$O83(O-q721e)Nv{B2g7t{XJjiX68jg+;4v0coXe7OY4S`0IZ=#?Vk`BNV=Gt*@? z$NM2ikj(S_y_rLN9~AXnZC%=o7W9sks2a6M8pM#(#RleJqFbxwfV*fu^&s?3YA zz0>7FqOBWXrsVIHnEuS`SKJJLLqK9`pRuln+{y7l$WaUDG8p^l_Dq+pNY+l6aqtK> z2^~WXhJMmn_Hq0Iaum%G!Pqxwe4`pEihX{-8ff&RQUiXV)Apz4PP`qNb(jlTIX zSGU*n>UIM$=IV9=DVTpu$9)%#U)|I&yt-+Cm;NdADx}E0Fq1j2k~*S)T=mZ->8nMz z6W$QXzO7jHYP1S0NXj~rUl6qa3eZL9+t>UyU(Pzap}LSUPtjc)4JLn8mMyAt4=c!= zHQXvKW(iEGSn^GQv;jC13Y+tXh<$2Vq6ZiII>2kM5J`_YAX+_Pt zL(o5|`fbU2_ttctr=UNWs1L6cHJ7Gi{NRdoIjo@UUD`4yI;SREqAE?{y?8+x?88Wvt^6wUkLPRUU%ENhwgSnyIW+4 z-`y^4?svD-Z%mJOH#H2qn+8~_jvCFv%Fc(GOnBv;hrWsGTPNxBqjPkm`$k8);l!Bn z@Mp5^jfZ1nDoa-NiNwzc;QGHHwyHcSa?p{Xg~w$`(25*PJo%dza;m+`u-Cw zwOH!u&TreuIUFxYG;q|1(;LQiF$>b=awK~(%oK988oTpe&z6;{zbDXtX7sP4->CX$ z0{xq+=Y7we^!@UD`u^X*z}?>|d6g1)aZZ#rxIc|p9^RAS$+_?`QcD~a6>Wigt@O*yE%fLg z3Soe1|F$8S3ljQak42*kz!OwD12?z&#a!j){QaVmdccrXQOA z;9o-LMmk@t$yJm+kS~KUxD5K=g}B+Ooz|gdScqj_A%eW$66Ag1Yf-rYNA|iP3-J+^ zDLya7oSzGrwQ4K}0*sa;$maib&7Yzf=Jj%RTd|uD;~$`Iz>|&O?JLlhXDFW3XBtl> zPVanXW3zieq$d%cG)yCSZIiR$80@T(Fmo0>o&8^DK~3!iF-_{qvuemRj&l)xVUi?|Kp2KdAnrK>xToaSv_E zmV>JQHPDaCCv*G}%a)Yk=&Qp_x+kbUYmA;pz0sNTXfbJ;TNWpha2~B(mGHHP2DtlO z(3A7xS?%{{VQkKe{DkvDWy$(X;}-_;T6an3#gtzQ|wD4IpWveV^SBx@DS z*y-?_ke*_DR%8+h7J>%fJ=#=Z*pL zAjb`mqaZo!;UiHgoKyWp0@hxbvG|G3now^F@BecfIl%FI$Wb(Zc~jEmZzL;rQbNWO z>=ZlVg$f?bHI^ojqiA;i{2?ms30nOEyGU|+ialv?YZ=M$G{{jjTVL&q$~j2Z#W0i4 zYTO$qq^H<)>|RW^Q|$6FImB_ILyn@^5Bdzw&kDlk{LOqYAEDP&d6%r1`+=f2tg_8` ziu|GvxZMVf+WkOVqhKZ(9ktCoR+}UB%%B}pZ?8PoUZ!xI4LNGy41v)UA4TO#B>Q%l zvA-VYq%>WMT1Vhf99F^aY zth7`7XT4gUFgdR*F}1p@Nre;xYMuym)YcJHEF_Gp&PaAonDKCr$>WdYaR@4FTAl!d)y$+1HjY{*R7?ssF z(62-v7Dw3fnty_wa-0g$6;o&&T`mk~(zWSH@{v(V&tvJBo_cxOAk3te}(KrT|u zyJQrX(f{}{hykqncsvjXUyQ_iBmby5) z$%SrlXl8efiWNK>U!Ik6dYeu8Y-q`fH@Tm&jZkRRehg8 zf4tFeL4Ts^Ck1-No5grGw`hKb!8|jdwp^f?`2l90(cgi7vFev6>9eX`URxqt&}Cd& zSxr8pTX@^hTGpi3ig3PKX{z5GLHIL`pFuSp$PL$5Gb0qtG zd2eN;z1m6ET{p2`U|+-O$x+#E)uo7Kn7>SL1ULqWa| zH5AwWt3WP5vaW-fBHvP%i|W@uT5@uvZGBTOF*UwA(ot4%ybp5J!np*-oXFETvKh&K z9%dZ8i32(Uj^9xI`+@#mqaV-7b^sOYM*4U?+5VNF+8mhGfzN~*OCjW_h4Ukf|GSUn zNGUqIUErXbIa{0kSs=%#zB16~82z^23*;nJ*13UR&(MD*s5YN`+g|2#ycTlQ!Wjz_ zciO!;gQ44aa>v~pl?6oG*TGD}2Q`JPj%xQ%V5g7nurm*a&St|D zal8DGCQlM=eH_^JI+!px$20uc>&PKgoVmN( z=XrndJ|&!~A$h|pjTCi={PwX{NX0!L<;mZ~Iph3d%ry!Agt`u@a)>=werc5E6;flf zXGqF2t5&7n^ew3FOjfsgFHhE^+7hV8DB9g}(2?z^toH-``{={ID>21SII6xJ zr)vtE6YjAhjz0G4ecE{0lM=~P12cF1{`5ND-1QUB)b!@ARiH!;Z`&V@cY7erWbH85 z9^Rqk0yR?gV*|Zjprqn$zPp%;>WW0QSqHjsja^%}bYVG=4z}E+wzmhMc4iUj<*YRQ zfa+fe^zoIZH`}$mb)!G*Xv=N|=}NLDLq2G-lr6LKb){+6Zi=OdE+cDIo9KgTkU~Ot zQR8?iqeBn(#w>uv3|`m zrcyDmo+JH`j)zRjHy+aFezLwJr=dDCN$s}XohNfqSvSH`j+Svbj%T$MtD<4!$I6~5 z?JiOeF|{l@x|Q6|aZ?=asJU|mjGOHRdGaEXwHIa*zyF=2>QcMv=k=5W9KVMgMYGJw zRSbVfR_sjQ?(Jkd#lHG_JE>266Ub4txF21TC+!JZ{Q|qwlkF7yXJ=PNay$)k6fN$n z2UF!7B>M)K$!Fz`gqyap1pal?=6<}QmaI~fdjpf(=TR$JCvsBT`_W|#e6^EIdbO6E zJZ7NC=6*HhYfIP+0aC#1+EbqA z_%h@uh`k7gq{%x-)+aCnKlr|;sF0#g4dq9UzeA3K#J%^pG)X%PcnoIX?;E&64lQK< z19cO~Q4oI?2d7DUB&$2jz}FbKLT>xak>fZHgB%5kdk1wp4#~P4X5d=6;V2NlO|i8| z%nJcKX?MEZ%yAv$sPXNQ8Ss8HVS47iQk!$K4NYP1L}S`Qt%Dl}t( z65f~GPt|t9tWt@bO)5%?X5Md6p5zj=n!=1-x4j7&mY8CHVpHtS9J@n~qUEH!hv!RP zq<3@(A!{7W*e*;?Q}NH=Q6SSf>O#3Y^{J9yL%Z~!%E_Xcl7JC`p5$u zAB7x6i`%<#nmmhSy#q6L&A2t89=OdqQqSC`#1#JUOL_7Y$Dbfa4V+hCO!jLX);!0r z!xJW({%r0s*S$__RNTB%q!rOUA@he?1DR%?)ui@8%>P4eR4lR(i^vcl_Q^1Vklpxk z-Ry{*srqSw{=gfm|K`tBxkU9>2l}Ip{#^98sQ%6*{g|VM)RbRPwf`r*n{@8gSd(pk z_>o>J>rQg*Baa%{=%b!&^w0IGp_u!&)M3A$Y15>0@7fnocBxqK1S|DVW)T6%WLQT8 z@1G>=rQW5(Lo~WLCrz?0)4J*o%=EYC!%qnYbW&S!zoqPhknG`sD^vDp^r`%aWR(^1 z@yboam<575TiiNUT})6dN!wdYPy@_|{*@|sBH0hXOa{FgT6ZK>o>2YnM7_y)DhmW1 zZ>n)G_DRrI6m1i>l-8KLE80b0udA-$y|{*imce5KP=_QuCw* zQPwdqlau$?;DtNl%OZm=q4L@IkeJ!OHj@)Lj)EM;I~`ywOP7`>7b02L!OVjf4P_cb z<82q-PEXm;Bw|z&QwN8OTFFBkpAKv^;Mq88u6XB41Z|~OIx1YRmxfQ!@DD&86`e0B zU0bTXjCt|6nlgN9y!V)%;rAZJ2EE7V{oZ4GlHYrbKGxgxB!hqHhL&N^JAa9htMsrj zlSp;l|47w%_xo&Osz&}_eRj-Rdg4JmW~850Q=)gpydJ3!=gfZN=be<5uUfO;IDytC zpsVeicXC#BO;X0BlBogb{&!XUuT1@~s;E-rtdI0mS%iC&!YatAsDl?np`S!*>7Xh3Vs_`Ddd>Jq;zINN(TKudQ;Mv z<34vNRc2yoFM^rAKi#_%;-%6PQgT{T4aTB^lGu=fX^$JpHzLDbhgoEdsss0QicO z>9yo&R0APSwMu4+ja5UMgmndp9j#WAyy(s;iP8DdbGk8e_=ujW*LZ*zYkH?@|8$kO zIi-r0tAmGO|EeOE3^Y~b=6s(jPb=uNKdh6^sSWWbp9qlj-r`L{32Bf30iGn#x6p7+*l=dKr21!*^?BfTs{Y+%{kLPe466QzK)=hJ3zz>>Oa4)P z`V2qa+U8uC^>;1FLuJ=b*5CVAEor6t&VfEt@niNJTT>oF72K61js#{r`HvAkam+Dz zf6OuZ|G2KGK4}ygvw&28`Tv=!rj+Ex-F$uuw*{0b6K3lE5*-UV?8&x@?=v$+j^@}Gauh8SUQflgcZxfbkaaf9*zSBe zDP6^%^->#|!|_tcQ8YJ6lk((xBqg?jBC6gmFc&=WjknMsu71If`bZFiytvk*q6VCcQ_K?G#%uIVQ_Eu7n&#Gd15` zznQ_ia&OZ9H~2RxA0a|exV?!gBah5O<;eJ zz-%+%xg>l@^@jt!mvD=-Nx0xbe}JAms+RW`vPr92hxB`Eck44TX-c%+24*T?ubGm( zc@y)$svi*O&3uac-iA~efhsa3&`)B@tj^#3X`6b#mA}#ZS5@m&-ljiJlfDUVB-$c3FdTvCKe8Fl0`=Q;IMmIRwl6sI7&ze{!-}&w(LGltR z8fo`_Xb#!G0$6$M*2W=%_hdsW58FDcjok#qDEyuRIdut?^iQy{s@Sw?8|mf@2U31P z0?w|Ao7cs;T;1t&gCOCQQ1UXs0uraC6njq+R8txmQzj`J!pN>X6d!5V{pAMA(}pO! z5o{qjoe$Zy34Dj}g$A++9DfQcP(VYL>YK84K){X9lPDTJ-dERo_%6I@lRiXnfE=8-?$ zms^T;zw^K$3#W$l@43{%yrW><``qk;Zf7Iqp2}YJP{|z$No7{G&djW1>vr8E!48J( z%>>><*b9}o9iYoiWpa$~IbVAC9bm3@fc`{Hz}q-z(4+)hfx)#qIn}-Z?INPD zfp9kaWMS)>M=Yn-qm(}r_93A6 zq0+-IHw=Ni5rx+94yBe6sj3tSAikChF-KMKon_Z9@%L5x$&nAX@{4CE%TOCZhGTg&1m=~YZLTN^c7`+;r6WD8Vo-x9Z} zR>ejuD;vb8KtF+!Dl4N3Jgls0V^{zQ(p-JU#Vn}b%uv!O{b{Z~SwOCS1l9(Va;T(c zLS4(5Uts;^%j)Dt>;=2Q^DQLEQ`d z9;kc{15i+m>`?Lw!Kk~ip?x)o-9UFi3^Qp7Y1FJ@P&mJ)ode=9;NOh_Rdsv95Sv@Y zM5}7%<>c!DU?pR@a6{!g5^QQ!vC#(n0K@>G{h*`<%&P(q8}K}^Qz3rEBNN7*qADhO z5x9$7IcEct(s#H6neU%+568A=-8{UYLwg*$Y;>GDf%%Ein(J8W?==OPq_8PR6xabE}=J=Z|Z|TBxvWyxy zzDyQ}4g)Dz{R-!+G>bz)EyDU5sJ!DVag&tLm`Aqe*S5UCJ{rWMKp%!!n|7=3|NQAS zbuFxP`-MBbRsYO#pS8ol5G#7HUgY?6d#bO(xVR0c*$ZI?1IbH^(q~6fYg|bVdjR{0 z$gVhqrP6$GJs;B+$m{C26<YrGInc1UhP0J0aP6`5Lm7!D#XnHd~(d3$y%xlNp!*d5MgpAooc(kHxHiwnAe=K;Ru?kO6M*b0@UVHxz?MRl2a}t}WV)n1 z4bjJsh=fv*rr9C3Cy*0xFa#=IK}`+_YaoykH|y=7?Rzh;YK-jE-M+qE2 z_!24^NW~8hy#*3aN6gP`96J<9^)p#;5VEHb=z!2p17iq`Ll^^<^rD#p`$#ASp3Ek) z#{;=kM}k-)!SUN=+7X#27#t$ zQ-0_&d*Fw016IlpO~w35hSZs;E@y@Mi*Z?Uto`C5N~vzA<7FCT+^#@a4wbJ-u(3mH zUexrw9$##41@SDVtP)h7do3za~1G@>UB~ZEkkqT_s65h8CY_u!s|ER4% zw?L*V1t7Z$JnTxJg8c-lyv29ucctPb8_FcmMM5bkvG(>6i^<|$z^YB=z8NZ?vN3LB zRk3L?Fx$QuL~Ed}poweV;eAv;S=KYB8CzCnAERD35OWHUVUTpW!^hd&s8{;Ve70@!ZwyEOS? zhtR!#7FfF8mpl{C=-lfu$^{los+bv>0LgjqZ=1t%wMRqQTX8}KL45U0Oq1v;mSQIVkWEX+>0`zB|MRAP(PTRx( zB7qbXM5pq5@zlVn53F2A?Et|A*tZI>t*b{J9k9(n9YoqyU<}mp39kT_QU#_1bU5KP z>?i-RLH(0k$SLGUJ6d#=ROb5~PZu>Fw!;tnLEs7>G{c;oK)dYmV-uG=b_>gD$yT7& zo3M;|&2wbmtvt^g#aq8Udie|X-aD1YK)SqXc@WaE62#v|FU${ ztMej^7uJ{ON){+PA7&l{wO23hwo~(@2`Z~1(Es>S!s8){DSq=OO=Td*A&{eJsW|1d zJQ;;_24viQi5V|KpOPmtP}vtJ>u(*ECyP|SELlGp{2JA-PuBN3IZvKa{q{h=$M|oy zHdXeZvc3)U{m`3N1E;Qy)Hy*xF|`@EuAUs>n9d(DM=hL>Vd6${kcUoM!c4~F4i;H; z2N?%yG_iA2>CSNguiMz%wak_y|y9itXN}p0pxf{}nlk7Wc|q zQbqp(S;Jt)?&}wm>=Zk1c|<01)O&wN(JZ@gQ>y5VzO@Ku>@G~UQ*1#+7rC9I&c8W| zX0OGYQ$^?6tS4c{ZX0$IQ^;LQ>d61Y*qOljRE7Wlx%bYv!+e=9V+_VJ#y*xNM#*H4 z?3qG~LWD$;ElNm2C6N|IQIgU^3n?i?t0bgdr9CZLNJahM&$;*9`F`p5`~SYL*Ylj` zJm>Q~XSwIxd(Zuzd+rWGeNgEuh-|v2iawBZe}EZy#zuuJWYw5za*R-)QThsE4J!vu zA4s}6Fav)~4^Uhon;Y<%F+zQk;wy-)4mzZYJ~8pS!3?~kIaJ%na{N_4*y{Ih!cFYU zTRHx!zuM}%Z&m%{3sPl{>Tk39X+}S_ceX53{VJSreFw@`!I zP`N9t{uS^E-B9s6crAAw;R}$jXo=l$q&u^=6l;cN10_{){$qz$3|i(vi__VpiP zUKYhr|1xkdh8gc_C~?JYXx~l766!y~=D%Tyozy;6^e+Q+q5GXS6)rWro#+Ma)@X;v}y;ESuQ}9aHq+YIcL^Uo^ zZA#OsW46B06>a{h9{{tPf6nWoHx98b5Aoy(LwAL>4!6V(Hl7@j;%SEpHlBRY?2oP4 zd|s}dtRs8{^3~kGN!eX6c2i3AWDkRLdMPCkvFL*uP9-ptx@qmsBfNZ(dV z4iNrrZ8Q?t_-by}j&DUEgNo0^rSyF?LMF_3T(LnLZ7}*;s&8cVx2rxa?d#{+2mKXY zF?uN}X2s~OSN_*N=xP|P7}W?DvXx9-ki6+IlbIfL_9RC?PxXtE^+z9ewR+H6Eik#fi2fJZlL-~O4%R~cTefUqpSWb4GC%+sdzD z7=>S*tiJ&L9M#{JtS>m3`ybV>vie)h2%Y|XimX@t7OUT4vfWQ}?nU((M05VaT7Ofh zpc4#yf;Cm7xX$vhut#|}G2z}q&Cd9Og^{FJhI~cyrSw%PQje&6nzj2N*-o+XDXG$p zus7r@ntekaN|6Cb?v*f8jw_Pw6kDx(Yneef2l5q7eiwMM6v;YW!EzWVSVhDq$)Z6e7pW>%LLH+RUFAI`77Xm%wUM&YaLu|Nn&3&>XkKL%s# zYz|frBzFkRG|b+nVXsR%SZ8YO;%Ys~FqTdph_iIEosP&S42BWe z6LY(CQe_$Z<4M`_8p)^jksMRnHC|m)+}JPUD$8C{yw6~!N*k#XvyTat6JCRspoAzuyrKVTC3WnqdmMd!7&_Ln|ENn+36ks_T{-`nb!sJ@jyv65VZ zYBa=X9MvRjmtYC~Q5x= zh&J3e21CQWj=60(mHDTH*UEl6w)%v|ia8jz5Q}@wPg(Wi;Vr$X5fuH%wx^X^Qzs?n;LcXFU_D7X$8G__agqd>gOU}<=kLAj2(rK!%5{({HdQelJ^KdfqX^tCXk;k2aw#qV8)-m10CK0D{jmB z_Ph2c5 zcDE+yr`X21yZ}OY81fZO{dlcA<#FuF!Hk{0)-7>`oXt|2nyBkSzJl00hS#{8Be@-5 z2CnaIhu^hP+)|!OpF=nR@)hZy4r4ES2KfXecP`8f^5~_|q(Q#WFjG1FuZl?N9Q_cSXcXA^oiQKP~=a1SymD(;6VE zjMW2D;R#CF6lPrhtJtm+&J;d`Dyh4cx09pO8MyHj04aXHv?c5W`DzxZJC1E8{fJ!{ zsbgp3_cEgHEZG0*SB=i?o+Eb=-V6C^#?A`!D#@cr?q-;AFoFwRi+Sg#%ib|U!jlh4c7K5-c8%bMzpJ@RQTv^1&k%PF zCx6IS3;zok=lq_TawfX}dZ*L8ubJ>pXVM#$X=zW?Ns){el+U{y2j-%o&lT)NL%m9@ z>$9$0gXG-+GX)G@o%(G(t0+`|uhnnSJCTg@)L+#ndBE}3D-~qJ_i;(z^?aE?c=-7& zdAfXxq|aT}Q9hf+OZGqPD9^`%v+5`=Cxd-3h_o@0Lz>bztCERq^*U&zAc(x$P{#Jl zs-&KBD)g(Uk%sC`GS+D3MVfir+Y99SKh6E#Bx%&xt(m!0oBQ))o^&PMI|pWJQ`6}0 zM}M*EM<(f8P5sxCNvM`s_2K(?KP=XxgH(IotK8!6Fasi|IO=F{X%5F5fdA`o=rg~h zLlA2!De^vO@8?AS+^_~;ZBqYy*H0_e{3$ARUfoVI`deC5+wa;rc9Vka%v{w5n-pYc zwlxvcFqJc73X{R>dt}OFT>MYP`mY@&PAqP9T4Od|yalj?*1MQ$ELHuwM14>VUV=B( zh<%3fRt4=y1WBt`OQp!qsM4Mew{Fo$*{9oNO4&0f(WE9HZFaLE>gjAL=omel+p>pP zC9J$AFas;uLu{f~&FmpIcsMm5NBV!&{1lx1-)dg(i@02;S>0^28fRWC9FP8P)i1aD z!KzP7JMtyZUGuB*unW^xRb=T)p!Q zemTFD7pipb|B&1r zFq6s1nG zyA1s?)%)v0KXKJ(%+q4EL1&8f29KA`S+D$_gx*zQLG6-aX@m~|z0ZRXth|2KUDmzY z1`hg5R6jadzpIujQ&71JtzP#|WN?p_f=}GWJBTptBkjxGb>u<9)sU|i{tYn6@UP(f z^)urC_V`~v-cu`Hr9eO46MR6pe3esz$N#n2#`75FUd3m_`kU_(SH7^b)M^zsTPxAn zM|l2Ay4%8nPT_oTdIY{l3?G)SX`z|t-a3`z|cXkS$+{p8by} zUlaZS`3jO&$9D2yYduRdASUqdKJNuaS54zObZUb_`F?_1MYYYF4Cy7`PjPFs=u=6W z7mV&M-%oXG8@tTU}4@X46QYf}0N>E#x9-K6s=8 zFWsNZ)6Ac4OPfsA?p&BD?Q{1eHLI2@b9E)TmvA}cD_TbG?^{W}nc_b~#C;WJOn>Ke zo^V7IUSfcd)TH40<8RVOEG|eT&jHcmaX+~4lmW0uy zhS4C`2&H|+lY5Z72Vtg`^_OU7g8nJhzmTk-GuM+hRKLgS4_v4CJKpqUzv}I-+<0kyb`@4x_OHJk;?9$S5k3}aGy^fS*Y&rweIv9&3?$}A65O@Wc{9_p1iF3 zZxi)F+1U!aEM+--E?}k^d{^N`D)-dgK^J(I``~pY2|Q}(Lv8!vr(Hn}Q!Z`ggmS6vBO5bhI^itHSG2?~{mPTuk=&IqWA_own_#Ed^Iv8G4B?BA zuW0s&{M?i6NbaXFV>c?W^^k*vhaq3l606TKN!f^9Ihe82L(BZ1)Vt5DB{fmk zg?t4`tonPL`HM@c zw-Vj~`HGg;c8VR4)>>XAd<*gwEwS8pJoyO8{R(F6-q5igSIAex zX#v8ckgp)_dAE5|=0)JuVFs?}iG<=Q_M)<7r77u!kgsSS9B<~GRwTDC%-E$Q+bQ;z zSQ{BeI12I=EwP-9JpV>=Z-E)RJ2*8afbymMshEK8M}eWc8aaGfQ7b%*Fe6aSrW;ZzY)n@3Nv;( z=I!}q9Wq;BlC-f#M_EI-0rJ(rzYWIuCEJ`|CgUb!jtJEU-fb0$I7OYJDYo>xSC%!sfttsQXlfw!2ieC7qTgB3v_N5Yd>RJQk$wl z{~y`^mT(~CE1C`FAIOx;k=%!2rjAPzulb#r!pio~E3abmWbPmK+#TLOxQ~SQ4=T_Q z>a*p0?d3Vb4VgqFZlsLe?$xZf5&ig-5>U{>9|} zPSq%PMVh;p0|)r4LqFH3f|-H=k88efzU)eCU_&5Z-TS}5mO97aUH(y4nv^aTl;yxE zNSPli$(4xSH89f#X|w@t^Ft-MQS}Rx_5C06WCzAUi%(%bRz!frUP+K`b_zUDKNJhT#fTtf1a<^cS1v3_-ut*pn3V;5#G^s~= zW5`!DkHdehB(0F#o-kuKI`LpiT(P<5HIe~@mq5OvWw6V%c?;5iO`G-3@xP|cW~<5n z*n;xJ2MOcv&y<~b_kM<%Qf*`U(&Dkj4du%BZhrYe)yC>_Xqd#BuS6IDqu3rg#vsTV?#}2^3}kf1>+kS*l$E~cfw3%HIIZtrv|Iu%#_awzkz&3OKk5_ z#y^sqy*0tE#j{%*$eS6J+qGp!*&8aw!;Tp$gAw}gB(_Ww~`Y8DhFJSuGQKWrsP$O zKV+&u*Xm0aQgFNZ8)wqv?;UT$)lfCLBGE)*XQ01U^|O=og9p)^n^FB=nlriC9%wG| zwtD=)dekk#B(|kcK#xh`Rl!_S2P0Qe*(^q_zd2noGPokUr#E@^e{T2Wwr+KqCqkodq*?@nk#2)|ysD?jpPw@)gaFqT`wWAh|EXjGg|` zpOBwo&!6wf+l22zzM>^Ia-t_+Ai00SjNOme**l9P$P!ohJqEZ$Txw9F==LETz4y z9Tz@q(0_33cK^XY5^C9}sJ1L7e5eHNtGPcBCb8E>dGZXB_ae;XqKgY8_7Hy%ZAazq zwR&|gafQ9~cNh7F@JGm35Gx_h@#JqLFZK1%O|ar2|6)%vQMq|m--Akoi@p{6$jm}% zPS^(W73o)l@xBFyJyaS3Gv!i4GpiiKOC!bF6md0b->j;PC!7ZPYT@^VN$k2?D#{QeAio$zhQS2Q1`!`~-J?m<}OFTqZ+fA(cD z%p2G!DVp8>278i4)M*Seb|c6y+PtovRwOO$J-S6k&KQQ;*}zPX(3GG@bZe=$eCKI^ zCxei@%dD#+wT@LMi%F{2U1*{#_&&kOuFOZ}-edI}uhC{x*s2aqYBf0afEBzA~ zjM=!}e6As`=3h_kEcX*W2Kj2>Uk_v6wT`!kk=(anrs`Uhges`P&3r6vAK{mfuV{&l zEoS{ElAF3cw7W9dPO&F-tRR(0uLk*wme_(Gp5zmCTU$Fn*-o*W`ko@a2>U?3qWM?8 zStb5Uq{$}0OgX=q7~X|#(?2TH(k9;TO2)<7T#JF2=F%HG+gxfVu_c+FJc8stZQZ1q zw*l1T8@{=)mGBM7R}c^RJ7vlHNbY`^$yFEaOI#rbuS%8Q2>*n91#$O1JxlzzfR~3E zcnXD-xI$X?Y$$oC8$iB-_-bDFN^%O4+YT1giSOe`Tp?B7Yb(Ws=R&@MBvu9ZP$c&X zn1Sy#aD_D7S6!wP&Vqae@n`&t-6~0GRSf$UDP@flTTaA%9A>QNDlEPLepgnmyhyke z@)abpe|X1Y7n1uO%)m>&3|G&swqL)(6Ic8h$6Cu@irIm!TKFHqn4xrXB@>-n9cCQN zGB%2vo6}XA5*9+fg81XJ8(Ux@xx-+I`2|bk?&E`moC1RvIQAMxYCLB=ubUFihI}>m zyDK{n#v{v~uH1p-Erl7E!8e6AEytOV>G7r$xW@=AMeLL)Fo#_^~rzxJV$(&V5GD&vp!A$PgkxRnBQrk0s z>@24ewuXE~%gARI@ds_p??uGB2xd$d+{2|K6DA{9F3gnI+Qmm}`n5Lw7nVRdhjks) z$ispCJd)gdZF-4)Tx~w8Q%0U3d=Byz>E8vDk*@{5-yq_C05f@)_>L?7K<%#bJ>en9 zSERoi##eXm;$~_mhLvE(@SNoOszK?`>qvdln?Syz`T81mMM!QjEV65%{{p<;x$eEw zbz!~2pS!%i3?{rB^3}jU1IC*rCwp=Yk~dR8X6_BrJiDh8-G?M!| zEVAPmnooAMzz6RFlGO33Z26GzE67&^{{{VLTzlcdks%^vd39(3l*nEH35{W7NcYckfF zS1|Ug-4Sb7^0L@N*|G|GSGOtpYPmdKD+vgf6ePU!2`h@<}g%d_}W{m6LKC zB)1r5?DV9ZV5is}pVpPZgqK3Tq9rz(m%b(FkJo5My@V4^!D{S4ge2kav7{UOOON9>LWx|2DQ z?vT{^UCP;kB<~d143N_yFQ3RtgcZC)KNqI%7Qhqd@t+ zHZ_(lQF#a9#z}MXXYZjk&m*Zjm@}b*yY}-SD#Wp)WMmN>0cUG9Zv%*%fX;!uCy1;; zctRs9h`f)m2coin6WtRzu#(;S!zu5y-J#i6_=0sE6I{uHywXHkBNRbhY7&vcQz_p5 z$jy8$c?GA5*Bj_Zv@eG=(SAa4$l^AY-LdQ#KR1M10Z*N1GuaQ(DTM@QEyWK`%#^D}N^XtZ99)Kq7lS3_a3#i7QMo5ueMt_zoO;GMt~Nb$ zyGaKPAzv;0+AwBOIPQ<+4TU8*?v3MdsN9>ZzQl1mr#Jqsv%_u7m;4zzZrjsbh=Xm+E17g;a&E_^zp}~dMEIscI#FkOt)6+pj!`# z8&6fzW>J}p_?4h0S(Mh#Y+XJaO<4t$HLcZ5cP@9-nPcJ%of?`)Pqn_!S}WcT9d!w+ zqoT$rYO6)51B+5q+WA}V|B>9UV5SLf0TVR=_x|W7+^ zjvkYYh1AbEhAw)loBr4OlcjR51fMGShi$aXbKbhMQmUJ89^AyE>AJI1%saiA>z`;O zb#F>(q2GeaEYr8W)V(>Sxqc0y^iF&>UwM}!b?2s3`9veS(5LP#DOoQQVT*qrIj4U_ zf%0IcKqZfymzT@c*N2%s11eC^nTa579>+f2f!p_haTdUh^DjYEn16O%{xlxAvJY2#YTVqxB`*vy#zBQ zZLU6JjnKcP`uCFb8NaB%KhPcL@BCVUzi!y2evDlPEWzIzUVyKo`qN;%0I&D-CfU3I z&#UWv2)K$)t7EfX&Xit&+)EVMUix$hM|4TWuP6+rp6ctGIXz_x;p`H$ujc-_Fm|C~ zqk%h+yrnRcVKD!zg8p&UKbxcc^%&64Va$i{)};UIbDmKw=XQR zTHEuS@f16{o0}$*-LqjPWBv5Re`H*QnvB&|k5#p02;l_CSL3Bl2Y8o`FwJ<6W;{?c zUW~@Hm0H`3ji*@cz1((_>~4nj-!rnNe8vSODna;oRvlEI)=fSl{5lcntF6BQCb7HT zbmey>_aB(?uuuydSA%oAcaV%vc?b^q3X)hWs(dO@w>>QQqUl>Swo1)P3NgdDP_v1z zHIQM1qy7iDug-Xs_zwF{BDphRrW8Ys3pE>ZQB}Eva53a7h*zxM=H3sDeN>U> zU!>MrStxm^8$iB-xHEV#OHM&@yTS~7a$@)6K^Xc+TDOo32nRvFq9x<7e(!7%qWNm= zuKTlsuhtHy?+0J4ol9ET3v|94e6`kyT=`aO7nz}5)QT+Gj^yoy znNr=Um54oy{!7*Wl&pW^?kqWq%FXyZ#8;-jM}3*aPb8Ke!arc7*ModTOU4Q<#{lA{ z9J<|#8P@#0q@Nx^Hh(XvZ2w*|Te4)lu67R?yRKcOHp4z;;8|VGzAWaeg+=c;1cVx+HsN9dN{v=itC)7pp2dlCG zlJHl^S2TZu;7^#xcx7NE{?gLkugC{KWgI0gAjQ;FzmsWOT;bfE-a1ZX%L99cFS>JBcf#;*nfgK&bOcUqL*by^U!%lJ`2yz@N%C zH;NbY_A_$GEW7VoKrP9=}C$Vs%QW-Tt9&yxhEjb0O&v3NVk z1Go1&=Bdy?J+db)buxlHp6{6@MQYK>THLAE!ok0F%#`7%?Dazbi%vd!Pun6g0falp zT0Rp`YBn`5dMiWjCR_^nik4WPjUtaDxtn07jGdD06uau=PVx@ndyubai4A&1i!8cS^rOd)u`dT7EzAe7621=hkQi`>fv)5$%VuQN9tJ9u30jMsCxsGuR6TL(yLMah_=#-uoL7f(ytBUi_+?KFd|;f%Prg}Ns?Gb`TgzjF&qBT;{X1Y1Yt|-9wjsH1!;ImR7)o4m zt9NpzMEEu2D@bB_ZL{QeB-i^kgg>PD9KA`O*emYgQ5jN^w3?8wNdK4`W-QXepN^86 zt;0mSEj3e7Viy)=NjK8m^I@hCH>%fQgB>+m{drZnjBoRq=YwUW0cp)4Uy*)!7@Iq>uUC5{w>!)j z>b_nQSKLRZwUP@72SdJs_}7*{sm36=(_jX!8+V1JSKN(RRb)QlV#rsdKN%(&A8YB? zBT{<(cQkB!6(!>zP25wG81yRh$7x#HO?vM%LdWV~iqIu7`kI>^t7^uG?xW*=uA*|k z4^jH2oJFZA|93Djo09BygqhOozzok`X4u5ly2+I-r4Qji$XC37DvUSb8K_f`-1#tb z_R?5cV`#kTxHV@lqhg&W$yHzKC2I&bSsRT6Holshv)7FXv_s7~w8O{9G?Kaq+TpQd zY;+YXL%;c+B==vK$!#$GCM>(@1aY-|ty4LPA7I513cTGKz4Dwb4jm*8z{jS3$zEYpkp>G~RUlo3a~~#MQy| zmv@i`;9FW7jRZEnnwzpa=(OEW=P$%=$dsK&Eo6DknOSlk3U4sXeu! zPS(%oBycM#_dyts8qQ;1E@Rd}D;n&zR|ZkyYFMv_FE0?jS_1af+Ft~d*f-6yjid(A|4G8 z&8dMRx|UW+GFIzQeN)qPKpXpk1I;Dcc!fG2v%WWt${uSkC|Ofq(9#kL~;|DONV`Yv%#zwfgv1rOgY=Ks!Jn0w#C zOvWYucdp6}{u3R)wvfvo<70taE zPx5LJbqis}d8_0?DRzAAM$(C}JLD^xmw2!{AITj8i|iz>kVf4LWjx{4kgp&vqpkU0 z5Xrj>X5i1UJTrKz6FZ5-73kep6BDDdmptv#C7pCh^)Y8xCwZUnd&pPgrOvbPF6&`B zz=rBovFv#ITKt!A5QU3F%^nKIQd!)zYQo_+tuvvJNG}=RSNqsO#Oi@|wqkX3yS?YZUt$*yv$+ z&Q;afE;Q9`sxP+Eb%$uA)lOZ=8moO(rtAgmeFifns=z(oRDUE{{}nHUl=>BY z9F~Aj0bfJ)4U+U34{B-j;m(MJ|k@c_BDU|KiI#McmmlK*f+oT`6g0p zPj-b%XvpU8S5B2$?`BsxkB2-mxFeoN{$CtFl6`u~3iMUMnaTXyQ0uYmCi>5W#QrU0 zD_nee`q2Z^ecPqcP1m%9yGLSwAac#?FHX!JX=mxXKc1%>Oj{IT7~&%w(eF zO=w27t@2hmIjW#TvRewV_n4~q(zV5|5R2Wv`8!ZwhvXr5@CLGX;DYtV-v zIWI95oz1B*9S6>#S{#pzIys1MlAM0atj?l+A?oHWrrs5kIzuX}OKLM(>ReJ!&0BnM zXD6vMw1Rqh)m`N1pG|c}WT^Yeefu<#I-{z1)!BtZ|J$l_MR{H-paB+hAV0v7_XW&U z!W`hIybauHq}TKu=N}TvUvA`AQYimtsmr&wHPHjS9kcUdhlMr`E7zCBUI6+ht2+C0 z%%667nf z9=CA6C`jOO%c&_VPbDP9Ko;CXP2l-gY8aPUfBg@?IkHGs0I; zvD&o%8qKX&^JO6@_o01_BzOPnSbV8kC#A{H)WO>b>{amFApaR6lK)z1x?zat zwJ|&^`r{+?=nqILoP16FC_TlbXBRl<`ioKf^I_g>*bn&o9x7b@Ecr4xj|tQK<`tIFen{Y3YqZeLCr_pq|I}mtZ_;rwFFm6 z<_`sz>lHRm12>1H9@n~ZI^>Td;tf;HR5Xxx5s|Z}5rK-;aFsJpK?@d5%is1|qbPS@ zJ{v}l+2U0y{WrA4;DjzRm`UPzV-*x4{TB?{zwS=fv^2Qe|b7cY=P1uDQs#}I9~@yFfb&) zcy}YQ?$V_l?zEmek5c{#Bzc#@UZF^vAiqD6$DA4PH2~XRk5F(OjR2X!d3sbWD@FM} z?-b0{#?!p70Y4cE%8>hXyLw{rULP_X zZxn19`P>I_61WowWgiHB}ryj+jLeMYPy~Tk`Lo+_<990hp|fJ zt+%4>VLV9sPmno`7Q-i@6_a=vZ4%#aCP{~LbQq;yLD-fLF(@`uMl*n>h1VVFVGEkY zD5Gx2>f}rNsP*HQ1&49?FR(k?949qRsek`@5HRpB=ik^G!eb)|b) z8c)WA==B~Wc{xd&K|ZVK`-u2=E4v-$&4LY|>B>crcP)|fHy}X6PkyVm^od&KGj-Ux z^y4o6wv)0B$YYTIDiME+vW+lrHSEG!d}I;wRuFM#^I>6)EFrRp$W2hOx?AO3wF8R< zYo@io9-(NCE6t(yGk&ivD-iC2hTmS9rD9Pd_Z~?pZpwS^AKKsnQaauU9xCjdTQ`C^ zqOj^CoN8Vl)R&Sz3=$d8;qBUTY3Pi$j@$-Euj;DIuW;<24~nIqzvRii5n{S#{6?d> z|B~_3q}+nTn<4KZB2OW#f&9CP_;Zy_gB7b$mlu)pZ_LY&oUwv-lp{H%+}_(KBPkwGEaKdeZncD_7xlPh;ZTBVC{|9V(VCz65+@n-7Br8jAjM}q3j z|DU}e8L+PkkG-R+S*d(zRQ1MkX%0vE2T)VNPk|nE{zpYjKFeJGb5c?)7N{JMhRj=cepRPui0vn0J2xD|=RbJ)^jab1N znMX!BYt^V5u=yCwQ=`1HXEuzWDUsxd<?PNrtIEu2WQ zOV=rs*Ac{npdWzpx?*nOlrd~X3(vkF)%VT?@;ca8p`vCXqBGZ>pt&^H+}2zuY`vUF z(j@)qt;?_47JY#Igqt5Azq?*shHC-r7;t10XQb?3Q=doytohcs)s?DHPtDK5xbrfw zDqbJ714zC=A)j-z{#(h2y&O-k*7pvgy^rL3pu#HLjU5Rg%zl7WuFlPR1Jv)6z6bJJ z5UDobmCBIUfru{5I1lRaRd{QDwYj!)VJxbbFYYo^Md2+|&9Om>KR=bi_ax~~jBbMp zwT>2-65&h*_maK~G8MF#QhWkd+aai6CH88m0GSH`mZFAPB=y3o8`M*Koy9R^ z6xYT#J#7xjGa=Im7D87A67{Yvq`#YRZ@bEN-iFO#9;zUvd34YpnF7U(7a|2$cL1u(n8|Flz-$Wae^M_#f zXwIFF&v_X+ORgTtWnq#NNjwlTKOLbdIyKXy+UO2f8ban=Vr|?LZNekm8)z@cl+?nx ztqwD@Swu&B0@!g-(WA*F<eXhd_%QAlL4G-%H#(}h6Ak1Y!Rq0=h(N_^w909iIJvYQd4O<5(A$TV zrd@P$dG;b#dP8PTWpSmVG6W?NGnadxwoI7MnK_KLNR!+#^jfB)$IW4`#`h%1%waBC z7Wi}cbS!8)hk1bXWssS}Sd1r|6q7iIu}P&wt~`DYV4?_u)OpEu#O3}m- z%k#@ntZ0&H5h*8PPn=}FO+IfyoT_z_S@!|kq%xM~Vv^}=m7__fO~*jfPn=}_fv-c5 zIY1{z-r6bJ9-x}{unQez4v@w0Z3V?79w3`kiWl>aKS2Ekhi$2o%zBj341imrNv17m z!X$HCXu-*MTyQwIV`nCrXHuF@P{}0If*FUQS^a*)r|16AQsdo^|u)0qz%^gI( zm&)yGSdVlW6Xw&b0_dAb+6!{KqGu8DuTwS!=8c9;S>noA$Qw!|?_MtY&~Ux4vkqf& zP#tz^dsuVVd%1-`7C?o1FK1z8C3idr*xtXbCw(2%Q}5p_rkwOvOnA@s3CSNp{D#k1 zQCnt)HE?7)DCWa2lqbXL57m}3_kn`^%0&E3W#wQVAES{q=pTZ-!?@_Tl*0Kc1jTf%0E%$`ZxcQ3QlE#nu@F5)gt3b){r3drmgJmWGwoEgA zqtTLbWc&#!SK#n+$lFQec7)p?|79ZnI%SW*iq)vgw@79_Y@KlnF)^4AkHK5XeE2my zY=VOMuthVu578y_;l>ZRQU~H^=EH}VsmLEHVx=H%=EFO+$T}as35z*uW#+?Pctas# zrShQ>^SH`!Y;+#C9{f5en8zKDH1oJGNd5>i^EiueBCob2I*;q0fm@x&wO{T^TPT>v zm5qv0A_ZnTyPB^?I*%KM!6n8>=W!P5MpBeh9Vcg-8tFXlW(;mrBQuY)Mk$ezA6iB8 zxHW*EfXqDZU4-3GQN^%`N%J_}MY!+2;6Od@?^cCr!L#BcaALb}Rrh+37hC)4799O|${#YDkc@Tek zCEr_-8tP1AIaW&{Gk>%=hy8KDg|mxKNd5pajbI^6)qq5OB>NFIjDgHiwisTapH$Ka z#hR?7jnFSDQGr6jM!1>e*^p_3{!t?g0unXCHqtjkrV&c7;@J~q8lfvfC#c6A$2USS zf3)4#5i8k&On;4FkJ5jkqD9B&Trz)rI!toJN_KLmNA(Q-3s8^hblgmZ%o)KVdG;AvM>gRR zUIBDDWJ+p5?67hiJUZHMfPD=rdNH}AcK&EPz;W|Od;4tXkCNY-KaWR5U12`QS^vo# zo=r8*BI#%H_z^Na&*B_WT-C?D;JH}#W9JbhL*HtK~|su9AY9yt%}T*&mu^$6=UqUW8@5cWYmW*y%nbG7Dn*#VE2 z!~c10pU_s=Z(AY#aXL9^`#4kT9Kre^J+e` zl`p+>?O%y?&cd`hs%-t-9hD`g?Mdm9p5aw^aR_sqrO!8!^te~%B{I*sp$#AHDetB0 z8*(|{_M#6|@M1iel$;M6wvhD9FlGJLZjxRxu#qoPCsqe|%&j@})_jwrQ?&|NPMFa_ zwWpV4`H!2B<-cx1mXesIlbEHOn57qH`DJR&e${2r#oT%>`Mza++2EMW-f}x@W^dBq ztDy~!spyU%*T$<|`RfVQni{_yWZ&&co+3dT98>k3uBrNOXudOA(+2OjrVS=P<;qBC zgJT+FSAflXnu!v0LyBpY-GN_s0n2fgKH*FHyKbfy%3CmJT5h-1jb+4L!BKW1DLqga z=>1h0Z|4D9g3EiLf^ut1Z1`g)w6-=)`g&e95L?f~NwyF5i~8TqIc_x#fNKN{OqC6A-7o<<$570CFRF&({2ZBEm1fQneC0=1PNu zx-G9o-CB|4HG!?g;Bm;yB{F0!o76zXie2TLsh|bRrsW=bqQnt99f!v)J1k>|5l(~U zG??en`%>4r;y^o`+e29N?6HhD=V`~-@H8zq#+yhM!e;_QaOhGxT;>$iuTDYrRV?#T zR^HT%#t|$vlB|L00ds-mJ)VFobs|oCzy=`v@@K3paT#N!lbC$(tx3KQk>uYE^X9_N zphj(={QEYSSd?Scj>s81vJd@9Nxf%Cx(&>H$a|Q`3kc6c#R?nHmyTv@8IL4e{7&t> zSJlZ7P+w#7B_su96KtxSXw##fmjfdHtSjXpez~m^EOJh;(6TnIK{nliwZf#On&qxd zFslBaMnlx-Wnh#0HCILp&_$}Fx_Bt9|uwvYt1?_Nt2z>>^KCJH0a3P(; zW?e!1+3Si{w*_Y;Q{9Pp7!R51TBL69m26|Hdmqpx3fJmdkQYfd3tQb+fNq6~#)SU< zqq;UppU4H7+S_cc?loFat?o~_`2pJDObKDOx<$6x>z>+J+cIV1k*THhuW;=TVNcM% zUGi^!y0O@{Vyeba;KI{je{FJjfTT*#xqP~aGs)dVS|b!eJ>O0+@>q;7(&YJO-q&b{ zk~~E5pAeaga1%6qEf;v3Nvaem{;38R=aI+Dq&x%UDX6DzQe|Pjv;)?3O^&CVRDDG9 z2asnrx_X{5tR%yA7q3!_f)?UYt8v_Li57k{DfI!>g*@Has|P|isBjP)?^zr6;={(d z*g4g^5%m<($Ev|}A_ozEf{GQg%DEGjS+n>J+n$w)^4z5MmpQ35G}A*Qd9l^dwSsBlnq*8C4d^hcS<{v7@Rz(6$QKX0o8vNU~YjQFSTMVNg;1 zA;! z;E@A$BP@T|L?sQ-Wp1kDKMR5mG;T?W)07Cvn=9t0PIouN>0Q9Qn_&GY(b-zW>xe8t zSOn$wU0;$NfA!KF)yB9L&Z&>(D3ZJ(usr~GL4^kn)`>FWURuGmor=EuOC9n4B<)W? ze?S9vut%aTx*thl(WKg}N=-C_uRwVp!)k1D`MeD+KZB8P?b0I2+Lf7?=id#&n-A*< zzC9$;^Mhwgaw{Y8VH9gtCGQ)wmjW1OKzuxbYQGyD1S?tJ)yBH+917*_0d@-}H>+9h z;RGWuGEX$xs75LGvco3^Yt^VQjW)CXQ>C|-2k(9dZO<04W1#+oyaq&yHnS)V@@f*9 zj4%NzR-^W3qIsIwYAAOmN8LIt6-iM^8kH(lk;CmR2KF=f1CTe5Nc}CG%As66Cf0!; zNs)uEOJ#3qfCD`!J;8T}d_6ekpjgAS6AU>ho*tC3;K!(q9vqQ8-Eh`gm5HRp?93Ll z1$QPCEBypqnY-)z4}#!Vc;1wf3Y3-k+N{*t8;1IBGQI=yt{}1-;cci`A%**%tP^=H z7g(L^)bhSXo$)d|t3uu(B3%(WL4}7#)Q$`?j6vE9POj%&R$Fc${W>+MOynhm4H}6N z`5WO6sOY<}ht&$+6>wVG8t3bH9d>sRke+rlo34X^i8KN%G!?tpvCN^0pA!j_`&?))M&*;TLF@+h|&CW8J#BQ`a9z#6L&b znXqE@+GQltQ*NV=@&ef041XZ9e-4cO`|_mPD_r3qgKj?(%>~5zK|L>9nHQCjB_DWe zeNF{$3EJC8o)0m>VZ=HlxlUPpDTBAy zseFw4kzL7|#^TaXz#V_4vB63A!N=p!x4r+ z#R@TJ8f#T9u(D^G<$#u|m7ZxnLHGbNXBumeVGN=(O_euzo(`Ea%{d5XX++O7w<6pO z6m`|l%7j^y^|Q#(Ix$LJiH2-OS%Qqa}=C(N&gAZ50JT}Tc97E z$f8~`tCDC2(Iq|eP2Svx%+p(oTyNUeYfD(BFW9OS)}9 z=Y%7w7v268%3RWCU@~3J%q86#dH-oNUybyV{sacA)aV|wXM0gNRkX_%v{^mfO!`+a zKS8;=nRMF_C6QD%Y^7>!Ed59UzIAiy#@ji&KxT94VF-gYqMJ*vL3j+x(|x7Sh`d=T z&3YT?WO};C^tWKYR%6{`y6jtA(;&0QbSs3GP@e8FZS9>%vYw*-rU!z(NR4&B>6r*K zG@|=WKZ&p!>d_-CPq^K5UTKd1R;*J^+BYPtM#DBPWAR`{XQmpA#=615Lc{p{MGtjpvNF{wXXXPYt0~YweF74 z6|z4wtF-4x=Y%t>Hs)j+vt(9ju}q1gk_OHaW|jJE(+;!LNGcs>s!sD-g#COXg&6~x z?qh9AonT|T&+9;6R=9Q_3-=?*W?{QewOu?7f{OZvgZ@AIxlMA!N^A~ce%jC9v;`fB zn-P#Hs73NeN$A+x*n%zuxdsQ>fp z>T;*i_c4phi&H1gnRa_hP3PiL6K764oRZH|Z&TFsQYOZ4PhnkU*l)Bd8Z<(%`5_GZV5tl92VS3HmsHZ)<1K zqOMLTQZJ+LV~)&!9x{DVAUI@Nq?4uJs|iIK7kUkQQAsO(oRIa+A&4gK@pY&-YZLxV z@U$`6Q%U8jEe;Nfw)eXwo{H=fe!GByipfEzFaPK7ye6dA=wzOCPIYz>oLEix>SPcZ@XTqF{!kdXseNflUl~SP73WC$!(rsli)u*v zBSGYLXAI}?Pblu)&In0=)G4PWXtakjH&ncg#9hv3e{om#m{U<}q=8-;9ym543wA=*%Xw#FRAB1&Bwo9o#!B7-Co#BWLpp!C*unw{na3H zHPA`WxGPt-lJqy6vK=YtauoMMw>!B^hT9ZrD~y%&C!Dyp!at<^02SU}TwlgaLBo#) zen8V#v&Y1vsCVsU`2tj|+V<-KL8y%0hl726u8c*^OFVE+UH75Nb(n8 zW<7En#e%r>^-h&)r*rjQOOp36tS89sC!k|epy=nED&3OM6G4twbncHdiuE0jq^OE7 zIF)B&>1CW(8v~FDnSZt(>FW0 z2eh;~z$7=9%GC*-vu#c&*?%b%+^jVPEC%)?Ssc)8&d^pl6_)uackPhmU)8t16hx&R zq^+{m$r*&He>X|qEwGj#Pci6yeM-=*8VS*}6fFS#-$_sQSz0zJ>J=yFnIxYV;jjM* zK6zatM1T0d_>4;~|8b_j=H#TcNi63K{7pALA4@1_rt}WcO%=V9B>!bCXKojY+&QYd zm7*i{hExAiwYW^%VE}9;zE`OK+}YS$d}<{5p(&%h{=Haw_W|3E!5hZtU0n2xj+m7a zv-?zy{s8tX1_vSU0Fms^c&GtMZrKD6DUmNW&N4sEwq9F26ahOOgI17#5)rR9>|$aA zjT=4Y7e@8AQdE!~PR&8MDLMYLu$pNswF(xW8cB(k_o$^-!4&;btR7Lzd@Z`QOq2O~ zM9AkYMQd?Oin$Z`4wKm}=hT;RQ8BH=qZ#u{f5)kj(T;qc1a=66-yrWnA~~PCQU#LS z+O+?qsKu?6=%H9WoC&NQ25lg(6_H^GmlzM*7}D29Ww26I;oVN{h3eruU^ioMqk8z5 z$dd@GPw-GG@)dd59_!&K!e7S2(1bFSi9AHL*y~hwxjy>W0F!(L&xi`V zN!$57r*b0zo&xOUEjW2Dh=2zb_<@sEte*7b>Fdc;^55d5ZBeN#UJ6jt6!jxX-e<5{ z`#B*)fzJmbpS>0Rp_8>D*=L^mw5Zfb;;T63rxqo?yW_X3`sR-6^U$;2QOWqTQ2mYP zDHkGSa6>eS8=_5DE0V=8!FnrkmTc<6_0Q8AqDf9$y&-b??)A*A&?Kju-U>N=Zb&if z8z(!lmoRmTH#p|@XR=c{pZH{Qp8y57KUasBdvlX>O&BrvLQ}$sxfi-Nj9gx~i=m(dGb`cNXLcZ0|Kk7gbuMr|761RAbIxb?vuoK!u50nJ?jMWAx@2?N5Y>uJC`vAs zB$8EbiIhthp;QV*rHgJZ>8hwyA|;iSE>h{+4Jk?2|MNX(X7;nY{r>xSygvJ$nb-S0 zb7tnu%$b=phmZ0BJ@lC$`t`xxGxU>D&Vq16H5Z>7KsfZdHS`RJJ`24J8MZt|=%XMU z`YiIQ>!F|?`Ybk%27C@S?Kb1k-vviKy#`Cfymmbekp22GTqh_%R2gzWKQ)`z9o6~6IR3J3=j)^=qVi&S&@qx2eO^djXl6lsq_JCHCG>0z#u z$u)PyM)Ov$(dkxmC|c(^&A5y%dK*4R6G#dddM&OIwu~sQgE>>!GNOpeh(c+V0Y-&w z?ndP)tMV}P)n!zqBeF%unvSR}^4ffcN=fIl6P+DE#bxke*?hyzu!GZJml=FmR;r58 zOr|aN+O!dt8_*oZ-jQ-PE2{p6J-D>ZSYkRGxio~ih<&Qtjn;lt223g@9i>f0iIk}E zqG(7*8I?X>#iX~p+Le`+Tr0oSYw$N3y0@!&kO2YEy5U&@w%WRPCiMVf%%!!mBw=2qhgwOgc(y zhJ7uD?=JXmukj{Hh;)=s5btA>5a}r2M|e*XQp!Sdc?%tz+US1MMVTYGD0%A!o)C2Pc5`})`SmQFIyRh%| zYR(gONoMFC&a#q}m0S;pMv3il*c~gj`8(n*qZ6BOgLbI(G&l0+El=TLB<#dmd!0&k zVy(R%VJBvuBLrH@GdYXqPwzOuex+)-Uv4BH{)Q-TrVc9M*m~35WE&i>71m} z+Jw5vuSJ%>ovhN@hR&4Y!%v=~((=4&y;0c){Z&xBEt09y+J~N5-?3*86z?sHsk9EE z=WIN#5zi6rc@Lsv=(z`vcf>PxGC9{N^vwUB)2Ko5-OXC7w39>6g?P*p&*Z|&D(%$J z^C%vN#Pby1FHh?cdba(6Cw5Srv$}&y>lu1Zz~gf9e0)VGm3Eq!CfC0fkHXi3bf( zeEy|{D(x(57vs{3LeIK~>5@V5qVa)BJ3I93 zg-3ySZv2WvD+h+2*W)omJZGQ5aA8pB`4t}fz}?;@bl$~bk_H?h@nDH}q#Cb#pUKeq z;FakrZK#(bMSTrHSA%x={ZRHFSj~nasit>WPagY$U8xi`0DwE zKX4rddVb-`pQ*J#&o89R4GoYfV#3TXyg~3*P%^*p#&-4?%nP}k)OQlx@m`Au`*Kfy zM9?9~-;2#J6L=9d0bbSS&4VBCZF`im21Vjoa15XExGn_+YDbQG4Pmo5D&p{dp*+M< z3&$877lVSvI37h<4f?djQP7Q_?{Ix1!G$>5{K{H2C^!$tX9ypQ;}RV8k5O=-pcuy( zgbT$n3rCmVe04G?Sb$?X!c-6&^>&V`W-d6@Nnr*kRi%Pg{jEYSejK#b@E-tu>g~*- zDNEF>Twb%4S3Qty6V2Ud?Q)v&Wh9`EiBl;)>>&2GYEYX zTWg~eC@Z;ip5oQI(#AF){kb4aXP(41#l?0+V$1j1>X@E}wH6eKuZanhaj69_ni@mg z)Yi1iy%uZwnOacwcTN)mx)xYopn0|ybS1bmC=z8|3#Q?k1avKU7h#7uq!v^mEw?Euma(3P*Mx_;rdvDr52FZleyJc-gLYueM4rbsR`-ho0(04vV4ohJxrbIieMsP2_XA*omC=$;;Yw#&} zmY=KfngsM;UQ#k2%5bSxHBc{PlUpskmqEH)^u@QJAwIity$$-_L<+1HqgK(nJ}X#) zZ{2@sP(a@-qOuf@e-+A*8bMcl`xAT?2--?&5GH{lVe~lzpMt^sJcZZ8!2V^t#xNp~ z?5KNnF;mg&dQLdEuE=dv+IlOuW!l`z?U?D+^3*>crVrkzr}dY5SF>X9td%cUX>aMt zYy6q3-mYbp@f}fla)s?@?rT8*(WJMeB0~+vTjsCZ>yGoh=F)GfjGE#tbJBGu6sxp? zc9O2CpFU87F1}hXE?}ugls&YlGFm|nWFofi6qVMcwS?Cli_}4SH1&c1YtM=*<8-gq zqj^fE?gPn0&K)nnv$iY;AfJxh?9GsV`nHBDBi~#6!kJ3a7|3uD%4v93mDGjw{02go zL&~w!rNXYFy4ul4Wpu5OOf1CPrdI`(k>?c+I73M^fmEt6-n>jn2$XYfY^^eycpJWJ zrN1W=$nY4-06gnRNJ`qYBX~iD;`Zz=}Cb%Y)5veAx|t}97#b|NgPQ*V_iOfd-LX^aoqPR%%~b_Y<3#Q z{REBUzSTJUik>Qco$qyT);ZMp*=elv6Ru;O-%Mxp-`@PB_9}gi@6}u!3pL~^C`{Fw z#5mTtIQHCITct1ez5A}{8fpx58p{(kmOG8gNu5=CSKs^h)cT=@O#g;)boDb!uDq+? z)Z}i*o13Zh1D==KE-BQI328Ja!2@1>lk|+~nJT@4=QV14!8QnL<8f>4(-)#_xJ-e`*=hWxg8#S*l&n-g z$$R?S?C?%6cBJ}AHFb;jxA*UyUaHd3h<8S3V`(U}OLek$v=HCgVyVuVpE3|R8fl^( zy+TJj?Rb8>rghhj4dVDElBOL`gpOL;@z85Jq_K9q@<~gfwbhP?w-EBH(OJ(6usA}U z$3sV`GkY&OzZ#t3~T}?atXvdDK zQkw6byRT~4GRx!$B-GP<+cKXejb>-gLwKEZ{A*2#IbulQDzrn*j_Belg*Pdj>qj%M0X{k)12(pIlRNOrFyr@94_C(95pXXd2jrh#Ya2xA>2xM@C-ltqa8!&KSOq-YQpj4poFE~ z6(cWGy&Gr9hdNZnEVQs`Ia@b<{q*zOmk|ucf3rzr*iQy#BJ@D`w*TQLw4# z3{})FbK`3*m82`EKZJxzL`Cg7>WB=bb*6fvb4WQ=C|8e|sES&57MV=Dqkmm&ZSTUjaSK(^KOt)xp{|GWejPq4Gl?LL%A7*&WSRdxVuuoFF zt1WK|oQWXTiiweNhUYnnDD_|F`zaY6hGJXhPu_z63Q(K8o3TFY2%7>gY|4v=x8Rbq z-qv_`Bl(NF$LjI@h-;lyD>0xCt#VzDi`5)x9X#Gb1NpZZ{t5H-hsE9HZP3MicpKDT z++V3!BhFW{Bu-^Bbk>YQqkB?nH9Q}qZJ{m1;Ps#&A4h#ZP<4SmyZ&C6Dk~X%Rr9ZO zeXL+lsHXyZte~7HV!45;{4+VZCnfu7qKe9K!1{n68_tRs;r4)^u~-=m^Z`E)xQP2h zamT5wCVUmp$EkdW@Qpa+IF;r>pc;e1JG`)*%`r|oAD%GQ0%awa4ggOBv=7Z&uLy`}BQyOk1s*-%L;YmszEvXN8W91H@^WkF`aV`{B zAKoYYU7+)!W^$lT0y-ZCBMbnA8_VRwL2BVw#+vV>#G5Ud+cfwU+Cuafh>hchA3dL? z(dFK?##slVY=xVwcw_rt<7qrpT#hBSo}L;`vNLz5H2q$T%h_koVfbmF&nB`w@=kKV zoK5r_!AFJiC1(@SrJKB7nUG0&QA!5rOiD2nj+Rc$r16Ab26QH^M|e^kQdy2790G-N zN@r4-_;lk*8}>_>_=ks^#OI_0su|FUw>+lz;n~DrM(`y2sBp6h)ywu_&v(J)9n>2T?YDwBDOvw{ z%P+f>JdmERRvL{n*c(Xals>Yl>LdM9AL){od+Gm%3V)f{bOm2%JZ%Nno#EgA7}Ls1 zp_}?bXV?R~x-*wezr;4|F<)= zBwjo*bl3RN8Rm|{sC0(CDhBFgpc}sr5#9&7@oQz&JY^+&t+q4#1M2S}vGKE9ZtWk* z6*|LLh^nOVv*Af<7UA~RXi$k>80gy7&PBWyin}$25bVESr!}SQ)0i6%0yX07@Y(6{!^&ya$50)FK{gMxK zhHnv7Nj}(cmYx&x;a#lk06HJeauLUVO2~)Qj6fv;oezBxP8Wyd!vcgGL1E+n$cJ-X zK3K`+1D#>_i%mYfiT-Oq=R;6AP#)0vaFI)nmCEKrGpJ2KVm??d8&0_h=?vc^s*-%L z;Yn&Z;kG{X!^)XJ=ffBm@uI|hxRLNVKY`B14!pi)6$`_lQ4EuCSpvBol( zkrHn<-4XUew}??Z!IUc&OykzbE)0(3if{P&n`bin|pWRb_Av^rc^laFICV(jP&15ELerzVtBh z_ZoA78jd+h{7s`x;t!zoInar>yd*V2Qf3k#J1J0EKqvlwm-v+li62Jr5TFzPuuJ@| z#KbQpd&L)+?#S<>^N*-89d~a#qrHdPc(%C>K-ty?5 z;n~D5A$XBcB7REK{0`8Gf5s(#0u-0?6{`m-1?a^0M<^7BBz_UXE#mP_ znZz&mwi=)FNw4YV3cH!|(fwG(wmkk#A=JN_pYTbS2S@k#$0MF}Wz*-Kte)ckTGI>F zH#eZS`~bdz7}!M$-T;GMG8_r{07$&WYhw?lq1!g!Dq09-D)j3ZF3E-6X<_pA+8*0)w&q zxC>PkU&VhaUjA)~{=M3L`A`dvA5W<+-g405@VA)~@J0E|vwN#e{9pNZnB~9V&eXHesxo<^(7Va>OE-zVujuhtanAHK)dia0n;%`aer{bxK9bJv>$|FR`9BOq zg#>+Rnz|gxznULS|D?6hg)(Ag1vQWV3qVvzYU~epr77M_y%&eSjvuY@EE*+uNoAq` zwWOwcmH*p7RM_g0x_(_%^QK7j6I}jJ{Ak_E%}VJC|JH4%sDJn`pC{35HTH8p^UCSJ za+0jyj32E!xo;^|q2Dw%S)I)PZa}lu*x1b#CB^;m@kjBaRXg@5r7H9%{;aPi@qZeK z3R{i6ZHiXC8Hax_KU%fJsijnfe(k0X>M8zj1W{qDjNLVvl#|nD>+E?lS7oL~I?2~) zt6X%&C~=$-IxfCM978)$o)?Z&nKdKnU!29<*+Tzew5yI|o$yIrrt;|t{pT&0Q(D9J zpaoieYcUQ5FM2wbV?jGH9IG9uYJk^+m?+CU7Ic?dvKF5;;_F>due3{mK`+C40R(Gs{EqM|h!0{e(kd~CL8-wPHEY(o0;OpL z^~?%XcMy!ju?XQ7aa@GsZG>$=#pT1+zq^Y&5}IhWX96bV>(-IFf$~8h-@9&s&=kbh zBq(txzDss^eGPKc>&=!lyaerH_@jV|%lEG>HKJl|No;O z@C3-dpIov^Nlvo)DBs%-)o!f1?nFh@!ixGt3*YXO!%Q@%_UMV zzawZMydn_n$MFEd3XtE(#FExAOqR^;@^)^qOHq-0Xem$Xo~5$utDUypLf5;am^mVL zrQX$U!1El?_0B5r`HZreN4@()5+?O-0Q@2l);mj$sO}OA_3p8W5T)Ktfpj&{_3l1| zWkAwOT9ZkJ5c8WUGJ=NayeyN@9u=U1n7GABEs{4KV6gBxN@>m zVomyu#Os=51AS#BN0XWqlbU2hB2ID~YEnarKdepJQr|5(*-25umb$O&{Q;7>AB6Ru z^vd_`N+up9N|Pp`R2|wt6b1lYke11(6I?Pwtp?+(Rn`s2HE^c^U69tH3(~4E z5-g)^3-ST|J zlSQ7gQo^;T9u_63ioH?j1$0%kOg`gNCWECaMx`&)Rd75Cmjhh|EtL+kjFPQ_cR;-z zM5PK!Mo87OThMy}=&F~|g!L1ktDZHG>b{A3!}Ej8~uqWA{rnrC%VolauS zdrNAo)Vu~w163b{HSYnFMV_*fiAOhrW2T}cHSaVOdIDYZER)T|Wiq&Q%^Qot7@%vO zr3U4cY|UE)^;S@-=2^8Qr=DmpC2B2$RgJv{(K-WBLAlVsk7brDvphid3&J{lz5&wiEScOQ$Z%xpgRW^`cXJ2YM%ESu9TF&_)QfJwcZnZ+vf6A4=2;|l%7wj{(Potfma9X z9nL3(Gf2xS)KdKk2r7i;dLiq1q%#q!j4K|kNb;F28p_mByX*5SiZfl}T;gpx>ZyG* zdE=z!X$@6oTff#VE$XU*Qv%fu-0e*mqcYq1wJvBp7ZU$=f}V%75tMSM`0IQ{_0TA*rG7>0WWUyn*HU}r1FQNwqBfj2g$y&SL(CfU z$<#K&mG6j#D<1m8s+Py6v9MVB!>ZAZe5Fx;uhX)U)o@|)z0T<6*wNqVv|%Y~f`pmx zbm|EW8yscNOkx>;gUKKG%}45h3A_h z-N3hY3db2o(_iCi&<)11czs-LCRSRlVDf%Fn4R2=gW~0*y!mY>gk}4g-ZR9rq{;tR!wbC!%m&+~I)30?dvs8_{1nL-7XNgz&Pfun4 zNVR-$VbEA(5X~q9!c2Sv^g^sV`Zv(|u->a@Nr};Hx-mGiS zKve|0y(NAnm3gaQV^WkEzsW1T-CpgLF_rm?->_XH5oDLlh@J($GcQwDu&NHg@3ywo3HW5%@-SI5_AvxOT@-G!i$|=zloaQuG>m( z`Su$B?CqjC7}_sreFsW8IIEv~?lxyZ`NTKdKTTbamNZ`zPv^-21W)651>r@Y;q>PBS`$?B3^pKvpb3t_2m?TTRu+e6x~OKMl(jli?^%iJeuD0RbGsdh+3dGg*EEK?NLn2)W|O#4#S>azJ(s zsoPlXA7C@LL<%!I>#;~ATQ1>JXm_Hq1Z4k-3Cr}HWXeZn-eQuPL>W`9?+?ApvCHEPeeWD~Km(&+QASKrlp#{(-XD#pn zTP(PeJDc$Sq97%A4Z;+0NXb2e@H8+bH~joC*`BhJJ=0Q3?o&yPl-%cNdyY#^O>rxi)DkJInYJn3Ok6KXe9{zujc@>%rdT(KR*{!zjj`&X z1>GPHlATYPgWA>JL&eNm4Z!fTgULu8;l=+0; zu<3lRNSe!zMS*G$!cm{rtTxPQmPkf(yhM<&r9#lMu}#%8;SEU zwCm8C39@f_t%YMUE-#}~Q@%U@B$VI*ejY*LArRby<6VTEVqqAAP^%Ylk|{b>Gbt%4 zItxnlDt>~?`8kf(F~DE^{^Dj%C0VV^`4U0(&*q671W)7WjnE6^9Z85W zs+DR6Nzow6aPSJk$BDu}aPC6513c(eBgm#P&m?V)R~=tHOp*R9%U${|Qc*L-tG7$1 zf}qXVcu9;(e|8Yz08sgTjE0OrY+ zRDk~eq-FBrY8f4Ubu!qU26qz3Keu$$Jh@4I(S5{16U6eXlft+&U-0wqlBlJR*nye1 zK-ev>HdL(Cp$InnxqEJgDBbcgNIwJJxVJwiP;G#2++TF|yQ3tH`&ptSje9YiD}ip@ zmm%CO4r$z9LwE(~#@$-*lvr?$`~L_(C<@ZJR~;Crbf6peLWDB_*Q&ea3&Y!lF4jnP zKZG>y&r52gai4_7L=bk%hUu5rv5ot3xc30txc_TZN;mGoa;*-{WA;JQX*t%Oq-GC zo{Jvwm{O?6i||Y6CB_{-g*}LVOdiuZpG#i?c(jXgc<|!z?9j|l{KgXcEg}0r{Hd)i zRQHR+YOB;NsG0lx7MI>idS4-E{CSK!fqvw*JVveXYW~2v>(UC`N%#_wf0nUh8lYfu zOv*8S6L!usL3}mbL@{(BQLG~9WmGqTU@4BT5WWO?4|QvzM!0lsk@$X%6b0MyP8-7h zM-aS(BNw5SIG)Ba7-1lYx2Cn7ZL~SPNVE@or!|u(yAU)J&J57^6_RgxvbqZI`=sDq zd>mkFq!OgC#$mW1Nab^H@50`xHB|f-x-@GyNyQq3xmrLW1gi-0hB82c zTFMb!Lr6Tz<_oKttY)EE^MqIJS4nvlg04ezCdig9!tzp_6edZjn*Z}U=)U|RxDNue zXEiT#cZnJY7#Z`Nhu=c&fc6#$+Ti#N;fUxp!lC!D*vP2X*(7R|U-a}M5+&0*1?Mw+ z1@s;k%S(5OvO_hyhh;3(F+flG%ySjPO6B&jEP{J0;BV0FhQ62E;%(w*eNUfVyvq0= zV=m=u?LScV(=YcPHvuP%=M0RzYW)6%+=4QZ&e9xJ{QDo_t)RCH*d6yOe>Wefm%mi= zeil~92Dx`^i035Rj>vta@1c#PeS=)DZ1c2yzIJ77TE4rVK7+dtI@51ZNTF_SAF3$n_4UN4CB=iuak(SCFy}e z0m#3?@ZFV5*EYyTeG;Xx10BjOGI=i>lJct?{%_3VF^65kjQ+>YaRgvCG)>8uhbv!GP-$4EG&dy(+xML~viUm<)6!Xcg2_&69HPz&U8Pa8EzuH3CAa;oul3j9<$`Is1 zxe^;=fv$L##ep$p`1;Bu``>`O1?1N)9c?(WASG^=%~cyvS;+`&ai#j87< zO({TEyg3LnfUbCV8YQJlq_6@=^;t#VrE}ph{GUKV#j||AUtw${R=oXamZ^BvMg*!V z2rFKnVp|WaWXnb(j!Vkr%1=R|Gsr*J@Lk1wL@J(*S}I-vI#e+)SLHu`WoZ1rYMo@& z%O=HaZ>NXO+~YTyf2YZ+%ZO|w&{?$`VLQ-S^`6VB5~(Ds-j)R(Zi8?-r& zrXajQt%j`xWuCddcX_`^4040^lvGH57h$6x&^PE5gh@c<*D#ONrEk!f+@LndbCSD3 zsR?_;u-v52VrL!DH>uU(s5xWU>bRS<;zfK=6y!H79W_(0QhvG)9UK=HCC9WG$&@7T zSo_^7FE!V60cM7RunxskW=x zf_E^KpC=$a27_PZIv>v7~tX6WUbtsY)eGBjC%lrhgc=Qjq2+X0)A-eQs|HHy7CF z_D2n|>~p(psgBDe(&gqlc;#2JZ@+(X>*d5P6G=m`b1u;JX*R+v5H@O7>;GufCb`T} zjl#NBTZ~HGT7!*8#HdUpy^F9DsQi^iqg36}ZJG`7l$A2fRm(tNjo6i{bsRItfUa7W z$%hqxW@A?n6F8tbTgW2P6-QClV_`WtPl zHyW`{)yc=6nhKMJieAsfcB z=NklBIFc`C-#N&;rdMO9!!iUqS?eOH!9#fWA^cPj+>K)y!d+l=`zei7%tf*ih0ewG zE|W+G6Y?>Xk3{Ec9KksHPH|j>qbrV6KwkIEhRy&*j{$v;U4r*0!bgB$CXNLNH;Us5 z9NQ4y07?zJuucwUgu)LXyLNNF@ok6?R45yrDg=o#w1 z8_9?Zwcku&q%%4+m9k7kNJQVKax57BrAKoH3}50P&O!P1G8n$W`4;DX=D0<-d2jMReKUA%;Q=)^HByA1S#)wiS)M?YVfQ7%pBIG;9DgI60HgivM(Q3D1IGxXKx^~!Lu6SaLh_4=1Oz>Bj6xV8 zjy5!R=sj>8B)iQ^$0wI}iH3HrX2pv}3KXs2?V z%3gd25L^U;cX7->m&P08pP}I z?hUDz5U1gY&!rTs#i-gM#Dkvy_G+SELC_Tt#|e1}j=YbbZKRT1UsO2`c`JK2zQy}4 zNOyqjH)!W9$8(atR(^8c_ae~`pyWz3YIz*H>#T;%oCB?Nle!s`F9F@8I)^l=-{AWd zkiYB?7MA2o#9dElib%Pl;0Gv`CUZmSk-3vuMie7mLlPI5CkM;b?E?ww{zVi@9L=ubwA zRc&z&)~+vQ(Gq>9Qr;p)hiyx1qPkD8?XZUl7=~PGV%Zy!VF;tfTI|{@l3@ma@z)q9bUVwYW35jce6wFY^+_WC zkf47t^e50~4p|<{gYYC1hCKpLW@5Tcp;rX@SOCk5sLBDO{G+vla~M?2BzU?|LRR33TE|xy0u} zaf!c@@NqyVej~yEJ=H8LnxDYAbo4 z=6J?ld7+UDQBJa)oO*u4i`QUo2ej3gdH_Vf!4d4`XD6;5Ap7+R4V|v%Bx^-~#QA+F zN8uh33;)9re9BMeH5_gM3TK}3t@bkph|>Ge zW}q+~1Yh8|A7Q09e!%e-!ka*4FC;=?3L*aoh8pn){?#ro`mu(+;2qrH5?f;c~1-k&*JEUkPiy)GU>73 zGJm;YTvh_bXDtw z&;jVG_O;Ph%1UmxkF9DKz#S^eQq|@lTrUo(Y8w#N0bSLszE{qYt7?1U?h$3FYHBvG zqKiYSS}Pnmpzu4B9#hrA7QvQR1y^$BYL&DIHKgpMMHqmABA~LRMX))e5~NT~S_Iu1 zOou!b=+4_=%@1~1+4?E>_l9=uo^N#(6&_~GDX zGzu33J$SMFuo8Ua9U*wAv9=bw1k#LC_4x>j*E3qdt!AH*!CL_+<>) zIynnlqAYH;_cHfBgM}#s?Sb=waPLmw`AT>>lNhNyjQ2l;|0SFTqnkKe3=)Ln!)c9o zlbe_Y1o2MPb{ij^Pzs#Sy&97wzS9Ug3r-=3$qa$zr8udk$-11Eyw<;OL`P-`u7*De z=-C3RlUiQKjCb#ax*X^#;~Y|Dw&MFLpoN#{`?+kfQeyNv>K|c$4|LR)A7+Ep%THZ?=K9}O zY8WzJ_cEjHNME#x{M448iu}y=U%e|^GrDXIvZCornIQM%=U2RgZ~56wbgONY1$78J zcssASfV|?IY_-i5#Ss$H<(VKex*qRfH9y;zu(cQjOL1Ir2mK!?5?LNE zDEaFv4hsw#W7Ar44<3`@UM0%%;Bgk zRod*!Ssql0zKnqnq4j}(n&`xM@zgRSPI5X|1WB`ALZ=n93Gl~*>>=;xIA)TQoK7wW z(OnB=Dcn0ja3zkH5jKJ7r8u%>?%u|l>?CL56fay4{RV}v#Dc7crY&QOCot=wrIV3T z-Uvz8$!LpCF3`z17h#YXDVL1Y@)m3|W}+}dEJ!j|BRn7$q-*9cZ#vmI-{p1pN{m`o z1d-r2yx)emT@-J`@dv{1phzeM_u{w+$W$g9_Tgq zI}w(ML)P41M0g(Pv6!{ym$PQq;lF|UmFVhvcOUf*1XAF+2(17Kt4&3I#$18tB$qI* zX3`cqgba)<(ohiIH}Ah1_@u z+JmUw53(f(txl4YtWHi&&{5~!+iRaC%g+n zzBnY;Mj{Lc{FTbKlybJD5I^f`>B^3if za~!85^a6QL6_#kHhuY71^@GEB&msIeQ8<9(C4?8n@ga_{5%z;3NpHbl_|$uVGrK|F zx{MNg8KJ!;UcKORyiX(iR1my}<64Ak#IY5}YJ^oFwD$u(U*h@<Mj{MG*b22H=3%vCGTdtpcEY>Ih5CmbX$$o(3?w- zj>kJXLVnKSr${UnOh;<|ARCxLUJ>cDdKE%Fxx{&RUqEWktF-`3v#{6@<`bAr*1pgBB9GquBX-68>^8mj4B_z?I;_vOw zQ8{KQjn7Y3#@=8r<#Dn11wrW#2dVUnVQFrRRwOIi%w#%16Y2 zygOL)5gG*$?l#Jny+&5qcT#9Xn`xJ@=O{cn#L^>bk>toAFJ+&^Bu9sI#c*d(%At=E zu}abMahN?(v(UR%bS0Im5gq`EZ#Y^Lsm?^ENhfKjTAdoI-*DWH+BQ+uZ#d@lB3bQS z8E%mjpX;Rr1Mof$>lnzD^0S;sf)tw1IT{vagGZUR1iC1#!{$`jFWSoIxoT*`lTmfc$65F>LVIxD7SzZ3ovF(7U0CWIHnAuaot2wwoECS+7- z4Md#eOpzcx97|h6cM9}ytPMhIP%<2|7Lv+au*0!ID4Zh}BsJF{OcM)cIHs~?UAVs~ z0k(G<$*!Sw&JCh^T6Gml_knO)b)ccyB`Q6w8kaR;8=}+^xzS`DcpLdDQ>kyE{-(t3 z)`5S%XH=9Rc*@lx6wl`K6{%EiWj~Gm}y`WAa9zv)VZ%tJmTC6tI8#HMNct?Sqw-?9O?a zu{c+rUS!?kGxRi69 zDbdn~?W9R&^5qu@MFs*4q5(g`2=ev zK+hms9k0BOok1Q5cL2~c$d>Py%#}6+}=@qbAJGeMm0ow_Ghs3Ve#K*WCDUm`oSpgd@O0p(?4255S zUIoj1l36OCSHTJpx`A*_+#2zfmF)DXTNAemdQIH2cs0=^q=HGbEK=V1HqpzP_!uG> zEiuZP_+o^G;*d#$%?K|6Jzp^1U5u6NQ@)!o_!{m3QI`3Ficir|13h2R8KD!<^95Gl zQ&zH;+a4!&LnJ<`!Fh?9RU$7Bj11Kyu=`r&K*|H97{X8exYA$QSdOi3PboTZ*^HYXy*NNj$o9S96IUW0vgY`DKNlPo)<~4JI-|B2x ziTDHB*=Y3_O}+Pfzf-c3DLIu{>^Hhal;qWmX(&tq=H082ydiVKX~wFs*fGg(aoPtA z@NXd~n90vt==X!s6YDnS$Qe63606FyXK}ru?_hZeA?L5-@ft)o;0RXnGZ)uvkar5l zI$1@&MS!B-SvwN%Erh=;TGMd^WBK_W*Fn(tA(CsgcrOes*-1X z_!}lK)$}*>T)fucq%bSkY4^IMR^F*ufyOeR->I=gy1g9f$e@2)@CwcmuT&6bY>$4NmK)IZ+4{)WR`mBTXnMXpCbs!i(Z) zi{k{sF>!Rm(d`+A&!C_X$3lb~L3|T$71>0SNhgtx4PMRcU0mjVg7(At1O)HnIOACk zp#{Nq9Lo{z1Vy4#@Dn~qa2*2i^HZ9rb6rMEAUtbda+QzX#pY!M)qRe4A3!OGijV)f ziJIoLR*Kf=$t{y4B?}0;4~=CYdL0fPan0Ke&sNziCpO})l?yoTOg(N`cx|=`Ga1Rs-EnDWg?=gws%wr?j^<^1uROY+0WuDto zIPAC#Ste_uGUu)5x@WvVCjp`|*PXxGaP8zwc;tpe+BmoQILuHHc=S(f5^=?6O(sU%hIdN!J9skrhC0sZi49gocR{svAmfLPruK79Pi*qe%^qz6$D@5IE3({XvzEB zRx286$d#?{;4T^&+58 zk2&3$v{J<6jC;@f7PvQo@GstyYnW`bcqEqNv3B@5=YK?|8ny$g zSvje3FaS#0o8${9ywTV^sLyNCTTeE+#N{?38V;__|CnfG>&bXjE(dz+iRH1x3(svm zxs&iE!jY{fmP0lP$8J4&nea^@Ci^NZheO!i1%_8G+fdrFpOYXA>b63uRxzgDsjRT8C+?gQAnfnEWyERNnWnvQRx z{tN0KKu7Hy5_R@llF7Cf=?C+fXNR{89}_W>eMXa_-n@@1mIw1Q(CP#9gSln$F_AJl z?!kN_+zCK$v9l~zsmk)%V%I|AkS%uiz+Vc&xu6>i)jpb)Y_YQ;N?FNrWL{{_Czz9Y zp;s}r1(eJSeUIyZKyR_Lx?VXe`XNx}hcdRa7aEkz4|T-V9u(eaQWrk_>G>h+V)@DC`F4Wj54gVrJwamm(QF3M#&o$*_{%So?PAtNe@ng@;(OM1KF}RYa@+1$z@Ncpk5&BvVX!o0ZP_q ztx`&PCA&V`aA%+z0G(3nuqiEFGHK%hrQlXB~!V)5nD2UVBmL9s${B^ z=i8FWeur!UxSTt%97`I36`9<(X&CxKG+CbPUwU$c2<1Kf2$*Bk37 zUvKJ^w_xkd$7p{f7Np)-hb@h`mvD#Hs|d&?kff)>DFPBuJ>SMoM)fVOC-~d`jIN@ zr6rd=7*}DW80eH+rl+jrO3LoRSOIsLL@XXn1{PGq|%D;#IKQSOFulycI z?E;(fe5*etTfbJzad`E;bCV`7@|Wmge@xDJu$*)!y<(D?dBSh;%~xERocGWXyX}EK z@8Lp(3xGbU!5YX={iRen@8LXAlJg$s!I>jUa^AyJ2v2|_p(UQwpskEI8SJS=VukY_ z?z2`tLg_ErCAop&vqDP$U$6-os>ER{{R? zsZ7V5mC8AE)!*`Qk8ijlIhAQ8w0nR)mFX9R!$6DSwCXl{n;7rp_Pmqz`!N zl?mjr8pRp952vQhNmEO0xEyH(I|C7Q~V%9y(#V-nEg`LtNhx14`NG3f!CpW70_q- zS{^e^Es0akE0?9sI(q_@0S3)Cb|i)lSkQ}bD?OF}gLy084GHm-c@-XpkC?-EmXT*k46qfHh z$*#zV<{>o89C6qk{jMNfm3?Tcjm}E;by`vHB#D(wxDjm^@2+(V`h_+Tzw)S` zKgUqq!G@ilv~lRepIjgBRW2fmTY)}I{yBtafId^sD*0N;9VWk@ z@Xtj-&Xh~p%l1YP)-bEVH)Qo+nHpBd`_1^|`c(`dL$ z-_L9~*(i0)hIvl1%j8n#YsIqEw>vSk2O5+=(4au z5ogFg0ZP7}^1UgGVOTmJ=(4anN&l&%Z-KdRZvgpyO=h@ypq~Ks1V~MDRw^IyGE$nk zXqf;h{K1sPlbCs2qLu>^ER#xDM#oKn?1B3M&}ETrSWNjiDU3wQqJzXCW$_REza$PR z3(Mq#R%LZ;Sv1+lD_cO9g=Mj9RfccN;w-3zK+3{$Oj$4n;ZGNZ4T(5A?zZ4z{7p%# z+!j}3XcEvxVRe$q>A2hCUbxFaeruDu(nYb{v+K;C5rb}~xAzSshMMM8}B!l5M+iS83UW4HtuF#q1L(8yx+8Q2ES1aIc*|V^N+hXPN*1$^NMht{yo*s71@zf? zmgprYhbr!DyajM?2KH>c$4o5dY&@T@BGa4AGQQ?yymO4RIr%L0;2Jp@?;_*b)p(-$ zj@NTq*c1y(OuZWEYkcf1dBTZ!15J>ptQ00z@!82BcE1#=bom=e?gmM6cp{$V>J#yX z7!&$LJQd$CwVB#xzLUpn)gPYz%H8vavGfW-`>^sc$UBh0!ztO3xAK<<8aQCO9zldOtN zupfdk9qLq|RNt!;;z(6nB(1!)cLTl;Kv<>MuIGp#2N~XOMkYq>*ECBbC+Z8FYv~2_?Fk9~$US z5~5&Rc3C=61pkBb?3WA$K`HZ*^8B_z`$Y39eml=Ajg3`yh;XOLcMW@=Ala)38hwB* zgdqMa4+WOT+3oP;r~~h3cs^PA5}fCOiocLRjHvxWWOLUy#~DQ)A}Bb(&)+alfM5@f zlfUAj5v=!TQ*fdXdnY5h3ORZQKSkoR-X9%l7>3=L7M+V6Oyg%NYFC23?G|UNTElJO zbvRs1p3&k8xaDw1tKp#%fz)(2TvQ5I_1ns`zSrF*M z2__>Hi$e}4coty;DC{yKyxLMnL$xsGcwG~7lK2k4n8g2#(oaAq-tv-Ee|R?WZ4Pp# z0MLnV?GisrDE3$gg2w@!_&k^RwTX#;jPQp+cq|0M58}|rLVU||E+}kSCh>aY)72&3 zO7_Kexsdh8xDXkIj>f>nK#zPZGoq|yb==73R=BqSz4T*QN#*%0{nV2qneaktI&J-9nX zS+=>R|HuOw(A#kG5l#kro2%9Ll$ER{x6O4V++m`uUyDUpAr9G^^ESeEP&mw_$82-W zlWnfnk7iwR#cpl=1^!{6k2-=Qj9%<+_JSSNV$?NDUDA8x+XAM93*Cgbr zeN5qUv#U&Nrfit!B%2YE&(Ulsxh=VF%k!P&cr((Y`7m<3a&eZ%3uEr0LL&JcErg-} z-tiVpbOT{I=uwVqo-o)~P5+eRnlDzFah_LdHC1&EGo1zE-|-!%8;V06o>8D?wg_r? z(1YUyb%)bcc#Y`DEsq=KT>7?iyo;;+0Z%?`JqFGwP$VG*@|Eisa6Jd~cEqJF1}oXC zA$p&p=IP{o4FGGIqv_pudFh=1}9DIT37u|TK@!mvyEEO0fpi^uk}!w-GW-Hy3|$8e-2bb zu+En!Z&8?rYYNa$;JbTP44=S7h@H+u!e&!5Re}`e5hfd-{uecAFtq!zuo7hdSe1>> zM#poK&DAq^`W=FR5&U+z+d%XW_)5jvBt%7T=prqN>S>9p8??iy{ZuxpNI9J|zwl!^ zs+vc6{SMfuyw7tulGl!&tz>6CGkXMGhW<@dwV-uDEgu9KI4(vQ1x7bcFhB!KHrKc& zH7+`*5V8Qu&7#u|$3}z=Kn+@IVjRn42mK8!)jLIS!gdr-wuN|my}P|p zlP0RnVL_G6RJ6Lka55R#?N#5J%GczBs^Rw$S9--epW+sp9#oZY=QjHF|M7JmU{Vz6 z8?T<8ncXED5LiH#B`+W#Aj*h>5flT8tB4AU2{E7`22{*v7Ea6wv*;lhK|K*MivdMX zPwx&yF$WZL{QthM>#Loa#{WLg3tvyaZ^f^xs;j%JSkeW|j{0tuaTm1fP7^(G>gMcY zANTnEgzm28_sl_aepHLWd|9&OoElesNUR&vZ}1~-0hrp_n#T`pE_Z!AWgj zX=Xg-hua6c>ciu9vONF8kmsqOW}A%rHIk?fJN8glJu%h|8aw>Nqa85i`|-2@6XPvJ z^dLXn2Cj`xIBcq`J}RM3{h1XjLETNK@Y=tB%EA)`!bP}QN5@>oQcise=VK5*%t7;C z7&?OZ9~=z9CAw1JrBC#D6_n9fN zy~L@*2ycQ!@B{~!U|b{z4{`7q#v)J{;nk~3V|#2~h%Hm~r$oozB{fey%jXUuI zb9CJ62Z{PMXsZc)6{ziyZP2jnfLEKQ<2KD(#J!<3`jtf-fsf@h&vUN9@{QFoaiJ$w zrLk;|qOYiGHg46`-Qq=2uxuZb>f_?h7mBDSwEYq62jW@|&cc`h3Jbirx1c%0rdo|0Fhj%zNk0jp(>dsl$)tz*RzdoHCl5rhR zZB{io18)|mDt=?P5O}-&m}%g_h9_gvT-?=7fpQ!m(#-b_BnmbOY>U+r*v6cC5Y~J! zpg2-Kr4e4Erp+|Mb)5eiltjq8%|#LR|D6d5ATf$}ST-|md=g_-g?H@6E#d|I-1HB6DG)zebgY_MuZi!Dz{4+n_2LEQ!tmWVXjFsTu@q(gA@@}mtRyWzK zewokEkgs~VpU8^%8&4|=AA5I_-~Dg6_~X>)`Jo8Rl46uUagQnD9xp=c@+Tu%(6B{O>FM_0lN1^EADFtQZVAaCdviN z_6Csjz9#7vcW#`?x+a@r!2{cJ_%X_TU=vEjz*dR9U(u;sH`4 z;#~U^Pd!lQGG->Fu{P`)s8<2*Go=kPK{i-6J3u5`Xj`@f_F_<5oTyxyZ|&MAa6bYi z3CgGVmY^1K)-?n2cSB@X$7-^d4CH9$^-7=P9l5L<=x*J>dEa(*@A~GQulyPERF980 zJ**MMw2)JSiDC;7-^0PF7}G#qH7@;a5bar@wOQF}v+t`A_2bktaF&XwI|rX)d@Q1a zsO}1%n!KAMqW;Y?*U|mOM{{bUY}Pdg?2W=li8c2WMUe@SX?FRiW8Y?*z9z~ipzVrk z1aNf+7m1k7mnix+>vyK4>oiVHhIA~5r*Lp3#uY%X6w6F59h20LH0~e6eO$!SxNlIA zb?@b=EV)uloDY^I)2@``s@M$x=#?@X*&h z#?CoKK~ajnpo@aaRLp`xDb>@QT9q?3N6$~~wc1ju{8UaYCT)*NVvpgVQKPKOgZNMm zcEZ>Wa8Kncf96&?gp`pw-G~euM^(iShoN0EsN4Tt7Lo}g2(xN^(r~xUo|m`}g?28X zpA8C@zN^2M4$5}Vh=wx6sfugvu(9KQx};&IXeO4DQHe6VzYh(eknZGvhype8`l1W~|WKf-qGA-y$2)vvZZUi>P*kcJ2D zjv)RW-fb8+f0Cm1i|%qkuyyeg@X zal)VtS%3#5=>E=VhDgQ<7b7?js5=;wg?`G*I*b!G&bljsy2JIG(3}Bv z2a|AzQzkZdxT|&6-2v1c?0{6&$vDA8Id>+7Azk#Bl13RDypHA-(aBJEgEm>$6sU8U zh;6^i;?%LhVE;O&9~<~B`*d`$Lnt1zii>A2sRtPyTu4ki1MQ5B$$M*Mu~}z)e7meW z4*1c5(XuH)7#&;-`)Z)01LNn?d>bA73HM1*lAwHwZwY$6L)N_p!DpOI`O&pFy030b0*3MdB~)UM-MOcIsFJ*J#I`W9J7Ls!g)p_7LVJb zTh?s=w1SP9ab}suD!6eS<7VJJ&S=?`Ab4DV*!_TdobhvMzIoi0aL0g>1m#nFOHiSA z)@=pkPd(0}sB|rcX}h`EiyM~ZtjbeDk=6x5~N)f&A&c_y#8>gRsb%wW*$D>SBDn#Xhabw|L)w(Yy!L`;3_h znqu$Uyd@g(zT99c2qfOuK15`ek4I~EwTQ&~dL!rw)ccH?ab^XLdEY*8_Xg^HcA(y8 z5_UKAQJdJ>yVKE47p>fj+X18+Y41#w3zof}C!ToVCZtzA5&dlvhIryD7|Vcqq6Lr- z0a*O%iG6&c*3;kGsmPfXltjH$q84BLj+oX0t$JfJ?kUf$_r=W%P=WVFqh(Wq;EVgi z-VdlR8b6ojn=hUX_e@ZdpnQsNzG!sGocf|gQ5hude8%PW`n5LXTYT{zqPQC*<2|3- zs-P+M#l_;k;)`#oz3ox&ZY&{6noy5)o8@?Zias&NIY*}U-FSN%g3Vy{Empk^PWfWEKtugW+tsE@w~4Q zd zZ-IKA@pEaud0xA1v%F_emY{r!Z=SaU)a`)$)nDZ2ce+pL>L%c0cH`>`*B70!Uzsba ziVyX3Cdg+;OSWULz@GAabm~OeNV1JQ2|bn2j{>u!@xwE2qxRi-|6<;Z_O8BmdeYO& zgY4HWwijI9)%W*jBppB5ndCuRRs4mYm~tbx^fY;%J@UJv16Mcy8J4>9zpw(;_KaQ6 zjO(TmdHfWoIuE5lfqp?}yo_4~uhr+gFfOm&j^z9_2eI zxSAc3C?;-76t{3{D$)}`d<_TJVq6Ud+_`O6_na58=Ln)!-(^R|FY)tJoPSaTPjK)N z#)oq75C=_%WnE*S8?dYiQH~e&*U?T}NR+a(%Ro5&fo{NJyx1x4vz%_Lav-P20{riX zwro8dlG0#tjz0n(7FV_F<<5hBjw)Q;Wjqo#V$RL zcjI>?osBbnkM9WY9Zq%GF6-KZx{+`63_Lb`-wECZnf4dfz?;CSE8$!UD(X2HC8vi? z#`>P0z5)6P;yNEqs^xM1xLCNaH5fGb68z-4cly_HVE& z?&6fLiDc)q$@#v~zGZ77?L~Nd0$mfy*m<{HLbElIPUZZ`V7K3SIomjuZk2GBbc>tt zdiAZGo&$6pB%?GAfKb;d^{)gI39oPG7qKZq8_CUTa z;gXP#Y=5)_xfVMo_1RPET|TUhcF4LvNx6RZ*n@*^Ao=XkRGaw-?6b!?nYufvzCV4S z&JC47`Rs9H2~=u!6k+TE^s~ol7*l~ObU7+XcerR<%SEf7YT2obGs~V2iudWYe-mNc z01CZJgGoLTt6fWG>`WA_Ha_j*>v081dyWvE0r4O3zry$e^gW0(LwvANu)^wCU-gfV zhOxGm_Knw1G%u= z|0Wkt62Sj+;V|lw>F9|*6)srz^(zHGph;(nPNt>{Bzhpwsp%^)E(L{YUR*vkEgvxI zq>T}{0_bri#2H-)QEpW}C4eeS$v9kws$V5`_ zax2iL3nCdXjzX{-&;g?{*$28@W4Fz-;9d&!mgRMfSAY&2O~L|CWnvpRHXly?0=;b> zgfS3EVaTnEiFmn$!jLrnP0}cX#)F|9C_1@Sz6Ik(po2yeu_bg_oO-)5vUJiWK4}p} zt@kSzdwS%|O3Egg()vPTmQmw-MD#Y$QKK<4DUA)%+KkA$R=|%MO^{6q!l-c=>}`RL z8jYVz^KH~P5$+M7BtiKU-x4$n>g7QGY9cO3i=r}E_Vu!~B)^FJZ3OL6Gz)>Y!=|Z9 zXwf8*qR4&VE!*Tl3b1i^ zpNq2yW`+dyy=VCXv0TX8EI<1ODGHW6FqS`Q3a+XnC}|4j63ksdn*!5hf~L4Bm@3gr zQ}8kKt2Ru@oTeC?Gxo^F^ZZ6JHxlFWeJyif%%?-^YnGJ&H=}#* zEYbZ8?LvZ?2u8l(`Ry4DxO%&ple^Ai;g;Oz+IzLGV10M%7kg@duP+bSTh2>J@?e$q*m^y?JR*cU}m{EbqRW@l7`)6nCllm3uXR3+$e+EkBa!&G=Ewq7k>xk-*Dd+zHFLmq2_{R zX=?Q%P+A_aF!W=PzI zpZDSX-k|npW+RNl(vTt4Bn7Qr&kTzDaQ}Qd-02{$<=_sC+dxGd4r-?HkLj2nE(@VU zebn}SDb)+1y@F^NNbWX_nqQx2Xj{pS+--c1V4cLK3lEe;$LB@q=n8v^Qle|OFAXPf zwRaMZQ5yv-$s>2d_0dp`Y6r;M0sAP4qGpKmq>m$VtE8wei|d}~OAMnqbv)#WAl{XO zpE14xdMmXoB&@;mW%@uo@F&aMg#EZ{13q(d^JDx>u)@tO%0Wb3b~AJV`tw0D?07i5 zkZP7aRBDC%Y)QEco)#du4-}Sqei(KvVZP4QTPn8Vg1Q9}O-oh$o=wyh?(v_4gte7g z-)W6V-I}Vd?mwZ-YWCDJE5w6|xdra|A+i1o#2F557?X9Efx3R38bnOnA!@Ze9{hVh zL?byheE+Q53B=oTFb!iW81T2Px`dc$l|-~{zHNLjKi|ap8$@su2d`qhBnMY=&}3}Z zH3Bu?vv~7Yp_&>jUrkJxT_@339?3l{{LO)@m2H`S4isHV3q~%UdJp}I`AikD^&IhnO1E||F+c}6b z?o32=4_A&`xjCZooSF&e5-?z`>lAo#!n>t%Xj~n2a?3gYGKiaS(BeSeW&<_tsU=Gd z?>6{!X-Xxr6_vX#kx2SN+a3N$;0nz=t?@;*NvHIVep8Y3sjg;KHLuHs9GYcrE(!16 zpD@3IHjVJ6g2E=Ae;idPH|6p`sqUx^x;Q(*dJ;a@(+40vi&HltyaB`~bMOMjQZQ-> zv|*R=%ccj>J^;O7`-Ge|SV=DHHmdA9SJES`Q9b<8z|Va=NHb%6-r3xXVu$@R?22bN zrL!4!KGyR(_fa~VF`e*E0Ze9e^{IG}`HT9!{Y55|7MqeM$+k{f9OO^*_dm*v-&Fsb zOj;Zke*b?aE(+gyMd^&+KiIxx>rgjCB}ddpR=jU9r*0#iw}1h)^kT;2Ydzt`l@;*- zeqP4;7lk9YLB`>GXgICD$R_uJ>p1@nC=By)h4u$mk#etRx-?G7{Uy@DQuG6=q z$gbejwWw!-_#zIf#?z65y2m*f5qxf~h?wUZvIU}NIW+~&@u2oGMn8LbUM5&cQ^Axt zn)n$g*TKC8&u6g>%}7a z1KJ7%uY=)@xM>)9ho|l=w;g)-|6Ewl(5&ED1XCrxA>Hs#1n~nXtSDX2(BJ*nFYy6O z6VMS+Xabh`6I=Knr3rZ7|0Yeqk>U6MuL-cU6*mEz{OP{>*wM#Z_IiN0_H&l}+C#E# zQ&70rv;4)OPY-H#pIPVZ1k-1p63T&qhHXVTjE8n0fs6&oG?HgBeO9iqM|l^*Jx3zG zrZi&i%jET8joL2Q$eC4~hkshs?-I;A7PT=mX^lmlpMZaX!Um-cf{$8rV6_=Jvv}!P z|D^b|=EoAmJ`&6ar8S?-F(i}9clgXXXOBMHav891$6~$e*%;MgHa0UR4+n8)= zP^JlS9|ZRRiTl0sxb;5D>617+$L7UyrOk{tAG$4hl_zoMqdgalk~8{5?tYHu0Ueco z<3o*{Sz4O>BhT2j-G)%4zx)*b$H0$D<%yi}8F!7QcZIzecEb}nJD*L?r=Q3*JBm^T zI)*hiFUm@2bYW#Z?-s%Y{Ehs?Ct9D#4fc~y(t~|K3PKOIqd)PJ|519dvHmyd!Rja? zA)m&-yq5A8kFC%2isBwDX*~1C*d3`^q4As+ISi>lh)WY{35sy0c||(KpRgJvrk}Igw}ZefbKid$~3$kkS1+B zO_XzUC=5ws>tUo(8qb=exj_Zmc#g+70BGZBqI@b&Z9E(ORpZ$`6pmR*(In3(PM3(K z@w|?Bt^(Rt88hR|GL5xW??C+*@QtU@vME7mJl!!__Xi4XJdK}A^R4mh4!78b$1|t+JIRUl|d33FdCFg#QUW2JQ>Y2pbeO5s#2QJfX#t>6DVx%%XNK?rww&@ zh1(B$LflM;Rc`O&et~eFv$&1PE8Jz85cgWRUrF3;ONUa~laj_$Uz{B0&qdCxc3LPr z{9?%R_Z^8_uDaI8X5A)0uR3F9f~NSYJ6_T&SKUzf+koV%n-n54%g3W0-Ejv*a@8G) z;BcT|CgWYgviS81obG>24PWd;11981ceKHM$4~28LTe} zZ}plL9s$$$Tr-@w^eQk(oD$nr@D{oiqLr(_4y01#DlmC2SjAVt z97(TS1;3;JRl<;~pv&=D*9quVU;*Sq03q@6I`MzXUMCKNcsmg}vx1JRASX2g{G+o)<4pXiP^lN=;^M3O3%f=gCFqT>U2g;eo`F;n{IX z9B<6?W7fhoelAC7C6w2UlH4Gy3=hc7@|7Wvs)vmy0=+@_g%bD)_!|V-f7JNAPP_}{ z3j4raVaxub9j5R)FVOu*jh9tk>HecTLD~VtAMmjF7>r4xlKn?bl}l8))c&Ja!n{J* z^7iCcAr}isOL*7z3C=$b^j%xyFu%iNcYD`%4W~blVD{j}TchONP$=R3+GbO;t`a1F zy3eRFo4Cm^tN#|)E|JX3KBHq{jsXK?pHbsAa|wLmxZ&9i#K50vt zyw>er4)^}X9fJ5R<-0Yn5$HYxubjvPInZmvc&v3Ge#@Pe&JQ)6#?uy1f1D3Tu8l5) zpwTc#UijU+h`(s!lXT*|R<;3Y>wA(YSl04OIy4Qt67~q7y_E4YDSk-L6u8HOLbuXh z%14{5QdIAkWR9F!!5hcMo3z=D{L3muj}y+dKvyX;COac|wP|D-c3GvU@5x!W8K41@ zrI0QPv<54Yum)Y{Gc1x5P}BpB#F$Z$I?-4pb5G&rDM0N>f7{4^9^u^NLEO@u(yZdY zdysnRV^_d>4d@zH#*5q>;nDKHCJB&!b%Rq`IRzy5NbP-qywmE5Lw1KKeNzF_AfSEI zbc~aL_Dv?@1HX`-B-uJb8*zNe1e1R~NtPutdK9cgmfR$*k(i|y|B;C92il99hEE~O zBePz7;2Ah2&|ZAXnOWBZXfJLOK7}k3TQ5HE97X^jd<-ND@O1U$C|Jem+(n99I?=7p zWdwj$`z+J&?oD}A{+pb67Z4Tjov6{c|0)xNPV_C%Dn)`^b0kl_UUk_6>b zeCtG?xRizr$e+5GMZs6tTPE(c^Iqgt-0Nd_9{?S~8Luh?Z5Oh*+TcCWh^uwEi~$Wu zhH#sQLN!Yx<{uHQ7Lio`r3iKfTKUGzIJ1Igy_NsytgM>?wDL_7r^HtI&(G!+Ct#JY z(*`DEQb`JdwCyp9w8ycEn+{b78K_p{ANi0?CktB;$Uiu*K!~^fm7Dr>Z z+dhv~r{Yp?ZvR)_d{`(Hvx*Dl3yD~~c^UDX1hiU>$&w!B`SLyV zDT|>4@6ATzE8j9f@a9`!-wf27jg?FD&70di$%8NWB?)55js$-9|Q!cFBm&VLM&zq&0>@8%}^J2hS% zkvqO?;z85*C-Zx9YWfSj;sxR{9Nd9%8_+eQO~jXRBHAr(Qzs%>Gx`-c%S0q=Mz6#8 zMh;}n=#DQ^M8MTw?BkS0qsJiR?oi))=84%y^AhjUB_(Ag=Yg&;$J70^f>E!z&(Axs zi9O3`8jFm0eyIOz<8})qegB3wf)wm5si={6uNHc06s+*P4L9gH7LmLhb|Qk~fZlhS zCX>>H_pe@p`wS>NdIiC4vd_7KL*1Wj=d zFMrSRV3dmKBV&!aE#*NP37#2qpsUA&kk~S6N|(w;Ha1I}7M-)#DhC z0v+#|BuFn_eCp{qm7qwe+HyrWOCHx+;sZ5H}DCT^gbc!m0qJil72vYjj=tD4*h6g0_UZ1&}|z2yAjBa%OF$Cuz~S z2>KpGzU3k~6wN_EF9Kt-q@tf%F-^D#9)m*eOJ0mXAk=B3nsBauMv0U_YQ2fiW}A zte|0-)UdS8*$B=A>ikBHQzCQzd!XJ8YhQl{`~=kD zO`H$nkWg{>{e9kx9X@G;?g*t~R(G$I$)gPq@x|ZaRi;*fP97OELqnbnC zl!!hCTEV8tq%|cKyuoYSFaxb%6R{66A~ z0=jV$m8?^4%uKLMW7qO6P;UmlaWMhkEtjQ@NRrp5UWENT(8k4BxisGzmv7*H4N4M} zPw_25Mwf(YH)l~)28lfh7Z%2ocUPfYPuuAXNnB4<|6Q7}JC#Nv5< ziKiFPiZ*5@rLl@W7V6Q!d!7lhDM9eOt6UqY`rTON0i{btglq4vh;#-1@E}2u$ zvnVQqq@&NccwWU3f&+kho>AFG^DjK_e5mJw63;VoF35^MIBtG8=@ZX; z0Np=;dS1MONhP43X946x03osBd5ub!&yP=<$>T0{Jv$RQvw{%Un<_Dj>(voaSD-a* zOt!WtTXWxP`bemU1Mhl94@reH@WHlN?vy91nVL5&_GM$Iq~S1Ulo<>FuoR2z187L@d)KB4#|= ziAZKVc7-zn=#0nl7{|$h%y`_4aU*bbQWMXFc-Kl0d&RBiOAy;~>Ul`difRiEzQ}x%(8N>o)fogaWtut`IRXo;&@4j_Q79=Jew8X0K;i6Iu?;MSl1)C z7HETI%#1UuxWPI_B9aE{Y4}fRMDpzT*AS6emO5#$+8l#O8m!L|ds+HCP3xTLEjZYGm|hnx=qQ*UC1Sax9w46kfL>C@AkJ$ULtv)XhMN=b0=QWUYN)bq~7 zI0LBXSpfMEKuD~3-n%~kcHa>^&rU?nte_*}<0WG8yt|0!HlP)4%uGsS6}<-P2f%xt z39><8yXXbaYx*BvRR!vK#?Phs=6M6*_6H>i%BT32paY?f1@d=8WPMOC983lbzV)3s zKk4kH^`(I7tviu^VTV9gs8(}oH%KEu$$_i?U}Tc&(R`A6 zFkE?UY+bZZlChD;`^IR^7``)~>$eA$HD%6EsPX~l$a{-+%B^4N4?gG9oO>Hc^^x~k z`TgmJ-$!obf0uP;>f08PK4~BoRu}y)RV**pHmdCZW;<6|xqI98{JW?}d)KCN(lbq5 z^IJ}A;o^Z)y1VSS(UYAlp3{EZxT&t&gcf4M+DTZ?2y3jd_CB$JB{r-p8#Q&!KNi+m z#+trM150docz@RTk=fTPpv1TGPj2jtoEy}=y^F?3Fo%z-L$dxH~`p!1-@MJdlIIE13od z@pc@XiZKn09C%-nkyG&u`P0Pz=!0YM=M#U$i9YcUAzA=5@y6rHMN#6v62t zYWhjU(E-}Y@TY;>qwX9v_d(lGVR#RIta<9O&f$y;mT`J*n(6f4X#}?n+RX%YBdA!z zLHrK?Jjc;9p!V6xoq{$BmW3ia!N)70d=B?h3E>kC;{Wi^hN~IFfRP*f)NLs@N3QnY z3p+deDGHV^NY|@lCLezcWf0PVp!T^pdj%GIaQVOje?^q7coRzBQ{HOe#cAH^&Egwi z97K=@fOuxnLB-!V(7)OjY8paeAFlG)nrCX_ccGk15NAv3rXS6gVxGm;Tjl(BqGs`h zQ0|307xbNVOP9c^2$tnWdcvDwya9DN(7yru)-=RX<;LU6a@EI&@$;XMegrk&^lk{Z zQ7YVacV^lhDgOBQyOpY z8lT#o7l@$?$xXWE*Cwvr>CI~SB{PtJRB;H3hR&pg(69Kit=ozJM}i7ryS33r8@6`~ z{*hE^wI9kvH~evCclUVagu^Dd>UB~7@$`DnkhewP@#x*ZGqhh94Uy&ScK?JT0He;_ z%=rzme~r2de?Gs?16O4(_ARk31^D%J8D|}Q0$XtBGVM;F&-)j=H$bxR-S((ke9Dtn zu&u0n_r@ztoUVB%M$La)u@RaNl651zWsE-~d9{9B2U=Gn3MU0vro>3@9!@lkt9!&qr}oEy;@(;MvH|ERQOe(u4w> z3-_*usOn@)Qx)oJ0q$7cIMeAl30?*^Z$Nz&^p&weQ_pk33eVL2Ag9^qJa+)KGRWC5 z1Z!3^pM@}H=nuCa&@sb-7-Qu?#ti3SoC8L-_o4U!Pwh0k#`)iv#`l&vic~{p_G_!;Vtpr&!3 zE`byW%PLIziJ4~*MH^@}UocA!T+QpW!6SUST#=Q?q@P$RqA#KCj9^C)|IESB7)QxL zj^Jiv%ml9X5)x#=WrAg2wrQVn9hBveUj;RtP#ZrER^ZcrL{cpMp#2R0dl2u;!N4!M z{RHt~4kln63~Fc48|)r3W0nnQv@5v~$^~%G7x8Tz%*U7~2Ul_MUyS!a?Z@;ArjLSU zA?eWQdnlQ&$TWx>HSgqlV04p%KRB~5#y()=-ab9PL#bWbxz5>dQLsWS?H*OcE1;Z) z_EfOjGZdHcSw4?zgWe*3gJ199^esZ!hcYkf2UC8 zu523jf}8sUO6yf`j`=fpj|-JFI#XHW>c@MM{Qg!DbIX~jyugh-r6`DzXBQt>b1i|> z*|p*R55E;J&(!jdc6Nu;XB2gIa;3zx2fO~9-wIR++g+Y{gwAdZ=A1@JPq(3e_anOD z&>efYUn04UJ88G}uH)CdhXEds+Wp+lRY%blXY=cU{4z#={r8|PmNs_Py`q+jCpdn&`-~}_->Y@iy`vTq+-2YL5)GKr@x{LK zV~zay(RaM%0j5m(lI4d_j%p=}tAAj&986hySrhqjfc&`oPc#tVjp&OtU0n52i{*tg zrZj$?)oh=Rnm$R1U-~m&S%WFPAJ4h!SE6Rp@gMpNyQP3Bw?Ew3Rj-MfO95W|D{UQ^ z9ZhZ>xel|nn{w3~MjJ~f^*d)i1`VCmSHET5YVa4G)On4QT8f-m-U%hhc6*#d9ZN5N z#_tq0&`!#jnUp4UQg^_;4HS1$CSy@nAE{lZ64hf8mGts2BYUYlDi&rf(}YfH9o%mk zqN>|%rylP6kgGl0P(d3<-FG>U;1A_g^&bpmK|Gd&9WjQ0Q9`>XI+mZ}SuivifXyZ#LUOvT!c6K_~`-+UH4tF{QvZW84(@@V;w=t% z-m;q~{=G)mf|TpX&jz}YzxcwRtZj+b$K9bm`jV^aO`;y}p6~19Z-jat(E4});~p?d zXl3>BSO{aXgz>|s%OnV?kG4_HwE|inV=zVmt&bPH#!W)p(FQ9iU8#@xk{YRx z+fm#Kv_6a(IkQsr@iN?(fYryq|EE6YmDY!BOtM|@?X^;tt)tC$zJL~7Hztu^bQz={ z2>)B4Q?Ppmg;|y+S=eZ&zmX4_g5506xn7`X86@Mg3{u-jmqB_KcEd7AcAjOB=UiaARSwqTp6+q(r2VV`qfI=z2_8v;!po0 zoo)T)_8a7W6iw0z(~th;GDwelg)M_bC`=N}WdYR*q39-)WD0vk;ZllgF{fT5q01y| ztJ%WIczlKoudZ1|eY}pJtFo-E59$uvrE}oq-4fwcRP?Bnu%>ZpdpO&IQLbI&hM&VP zHL}_IBp*Ayx|N2cjK8`c^A+PRLfsGAbd;xn3fX*JH(fV6o3(`?5+Qr3Z`%Jtsd2b- z5Zwe^&A~;nv7ky>Y_b9SJ0g+|*q=r4w8S=r1AT~ObUt7&kBvte4X>xJMfC~Lhe$?c z{gpD2Jw)0)mvbY5Jw%#zJx_G*+V&U!JdZznC=1L&LG~$AILap^2KT zSM-92%%&0L#Mkh$UO%)inEFu zoj5cZ|1q0A+;yOI#K!?!fIaeZlxX!Cl!c^QQ!1F(NVdl_hlKc*_@b_1h? zR@UKP>OD4cXG$19Z2C_LLOT3!QG5op!{5JA&g~Ag!@oXw`EoSU;SUs%bogJw`5b76 z-?DMeZ3MK#AB?dj&<@{1h@6BFI{a~*KTrfR0Q(!pd2%2f{y#A80l1C~z$zJ)Cw~oA zax5MG97&CI_^+dQ4QPjNOyUb_udxBx&v1VP*5SAG#Ptlo3TOI4kO3GSS9aDXr{@Vy zKSq8!6p(A5bVi@e#Q){J6@4Mz&hzOk=jHlnc{%6hqEB9gH$sJ*=?P~;(5LR5n&eyu zpikY$VT=L#)ZIjt&a5PRyn!cA-Iw^xlst8p7h#MY2P^1EStCigv@F`97?*)^Lmrhb zU-@VcQt>EPp|&JOp&KeA(~?wNmhIBeg2*BXFM3iGtm1SWM>@1USU`U719};nCX>?G zW%w4<6@bf3+65D^GD%sQq+O`k7>c$Bb~cyhNmnTiQ%M5yDPBW^gwx7aQx;@pu$uS` z%GmJGS>#XF)%u0R{si=+8PimyG+}k24o%ru8x-F18S-C_C2fno*|K{_yI(PjDSQ2s zfzsVqc;D2{$ru@DmTlQ5e)eWZ5?aY?cpKoqET=7rTxz*nHQk0yphCMq$(4 zGIGjwg=s#KC7vtq zmu||pFQ4kmEIhy2EZSJ6sm6%{jGEaYV*ZH}eWKRy;NWjH&$(y8Y?{=FR~tn)l@~9g zZrdQcxJAy*0JEc2obAooDXn(u)4183@N4n}!QM>29 zYv+#Q)B!+m)9z_%y-$Oclm+xb-!ogJw`t8cBuJod$_&8R93)$Sn2y`EPIM5Xqnb~y zL?v5*90chAQEkt`B^Vb0-2%i!+`}OXTYx;m`G-UxTY!9sv04sf3y{VeF<=3@{qAy~ zXl{H&%?=(9OO&$x?&fg%0^NSscx>_tCzM4}0=oUK>?LwG z>@!v2>NX@RM$EZ6NtU;J`8wk7#;N|OB+;uvfU6|_rkOn#3hyvZeFoLjVW|6YP|d^_hz4j0O!b0EAl<#YY0&8#|A*$9K>=zHU?u~ zp!a1aii+e(2IanN*;ND}_ho0np8;H*+?Sb%uULsm?!`{K8lv2b-3jS-5X-&TGK?32 z_NXSx26++>y|xTI7sOp%M4JGwUK_5#UY!7eJh zXVbei23r1Fvpsrb44`dXTYo+A6toM?`rgsdMK@C-(#B0E)RTa=aos(SU7_Jg8+V)V zq>XzJ&iz2!IOB111~0U6?{NNKVAK-%Rob||AwbtPY2!Y=fdHkA`yJx1K-;*^opL;8 z1osq)cx$Xdt~r-BZvOQIAZ^?@_y>ZdjWa6CPZE=~apNT>Y2(g?bT-g7?p}<$fcm(J zcuG?qleKZb!2M9f(#CDrh3Dm+y^ouO!!oh?_)cAOZU>+~J_%zY;7{AQUA-uBW@#4r zC(J|HiFkdK%mVUP)W#V(w_dWgaaR-lEKuCWnXDoulQ!-WseWnWRwG*lw2d=OWm;ox zTtzqLErGUi#;;2At&Qsiw+?6Z}_ z!Bn92Z&de{z@N5pCU23F zOD6wLddu55Q*5|i#gCFhY2zjm$+ke77Sn8$(pb~drFYJC0)8ZKf|e;kXzO-?JsfCT zXZ%gld~55jhId4-x9QUA6gKQ{1xotvbIhdHJ(A$_{+!II%xuR7dvZw+qm&G z1hjQE_s*eYWC(jQp-uxjguN5vcA!I86IH7y3}KgZ{#6mk5cUs@-{e4suwD96YCwmu z*1S45lPF~fd$L3+L)cy53&*fm_emU0+=;v4_s&*~9(O`bws@0HN5Xt9QN5Gi? z^mD9-G3JB9p1#EW*I4na1w8fRkx09_$zY8~fKfkQ50rM_xJ)D^f%nulKfig-Wr4Q& zy)k+KZSzgQ0&HcCskHgGNy?r$+xummqNN2Xq*2K#sZ+% zyop$etvn`c^M8f=rHG}?U(}z6b3m_oldvpXnb@xR1p^q|0lntm!dL;sHxDALBfKc@ z7D$5oH_MzY2|k)rzeD>KByDqzJgG2t6f6ly{-lR^1*J|A7%h{g&23uCRaz7j3kH;> zTN07^Xb3OM<8g!5g|0j@i0GIfcO`ok`AEv77TKMI)gDYX^nLNpTqqas51;2lylvHI)jPhl-QhM9n^1tIz#I%nNGGQ<=f0ym&OjPplQjNJb?pRg8=&(+^D*uPTG=Mbr{dK4pl_MAmOnq28qd0m=2AbN zyEc@LSxV#|iFtrjgLEE06VV!=ZM88o&MebdTm9ZvtN{Uh=V3J7zX@KHNJ8fk56-y= zXy;+9T$*p4$Chxrf|3N~Q+!KMwvclfAb;v|7Dc5SN-5)V+i!I@c@>xI39lPamor{f z2)fwiI^T&#Ty7%#BP4vCX1p~Ns#!^HlR|w{MB;MSBe)!>%NaA{%nBNFxq(~fTpyq= zcLm1ZfV!MX;*{82uE!8E1=Qv2Kng>qBTY_+<6UmpU8GT5Zf`Vuh)$k^U5arbP?s}N zKFC9YW%^zol*nJn^u0`*&ku!TR?y*bcS^+KaVv=DX`mHs%uG<5#wz%dZJ2oi-s4Qb z>VbddareWX57gs~#ZTpY^SCu|-v%WK%BT32po_PqtbqKf$63IYZcgHHL+`zZyo$#? z4DWuR9%sC&5OlG}T_s*B9=8tuHxj;j+@esZW@*IyBbV_`a}kNhbsU;=ZGd{5Fi0_EJ$2}-%6pyyWUH_Ws}G0Xifn-4Q`sMl%|#Kw0Jw*n?T_WpKw179?!a92H~r-+bPg@ghfk>Qv z9D=bxo!yujXI9Xdv)>H&dZ5m}65~Cf&Tf)8B{pYozhlm|1?ucOVr&njFvQtSPG}T{ zT#$1ljpFRbL7OBxarU_wcK~&E6XkG$w(|o!4f*wue0osIwc3>D+R@Ir}8I6G2IW z@+rP0=nAN_fc&YmTfmh;VjHZ>0+Z!K03q>m5j^ihO%H4B zOytZ8LSk+#RU%ixKZ$4&P^U8{TY36$O=C{?Bh>GKzY2`Ta&u*Z(ByX*!CeK=tH4-n zRg*X)%Nv3TO?#B&PJiZ^B^rLl@%1oaW%J<$ZbbzW9D z!4v-r`#qqZXe`DT<$UwR+^&=YC`nL0#kT|*T{5ShXaQFSNs%WGz4c-8EuPq$D0%|* zMAK9SO|d7=67LmH+z0;NAo0X6eGVdLR&fFMdkB$u;^_#c1NB5>X3`owvEjwiI}qFk z)B{Zvr$pv~uRvV}N?tTH8FNQIrNy3ikEBmL?|XFXfO=l_$ee2p)bmW14*`S>N+THg zcZ^e>XD1?OR?u;YHkTR@&l^TOLx5JaG1+L_2WuLu=+mH{3cTkTjkWa31i|xefqgSj z&odTx`{jJ|ycgj<4@wf0Pw_25Mwf&p8;tvaD}yAw`N6Qa{UhXCMgrdu#n+N7S-im# zQxz=Jgpok?ZhUtR3LE(F!$?4fz3TLKiVZnEsalf7pA3PoUPRR5^pgl?B2cF{CL3~k zO|jFTFUb|BzXJX&kT`v{r!uzoRuWCps&rk5NSyw01djrBdSkLsNx7!P=|4d5K2WDO zYMc_8(>LCow>ChD)0>QE8$Pb$6|QfV^oi5=N7oOi)9;5-57g;RmJb1h42skD^r4#5 zhvBcCh@4qLhtp@J2E^%S5YKd=K4#2JN@G6u7}Q0;JG}|mNVB{oZTPzq_J4soy|HNO z%K7H>6?^bt43s1&pW<7Bj4qi|r?-GBgJdh8aT)$@$ooO+^nHk;H&CZHO;t)0oPK|} zqd}qI6CRvihrflB{is+L_LI*c^ey3d)*|`vC!V#0UT}6ao|~R~pX|=pZZ&6Y=Bf{f zIws%L&77gD?tRxXxs!>USw7)H?fZzNTvpwCm{iOKy6T=WGbxR&y0`6~v}%A`uv`Au z%l+gdac+*>j}q`o33xuIj(~NTFm+wawwolWajwBi(!=t6+dt;r=PLLY0j(_~va4GV znm3~lEd$U9$Q|a5I7zT7I{V0qhL18m5E}crBA{mwU zM(i|G={}>;6ND$N&Mf4Y0&R7SmkA1cxl_h(zeyUT)p-T}GLW=7gG0j2O7fYsIyXFy zNLrmA5PS=?)iEY(nU|GLcmr1ldjs~OI0YwhBVp7ySQe4BI+LLu3#`@A8*P&@bzYa; zXjlD{G)YTyE!-=Cwlpg+UIW_Fm?)o$O9yrr{MC*2&Y>{OvNTEw9xD+`OVf03nn$3n zVoWw!Dk}`Tuy=;KBk(Pa(b&+Trc@Tb zCFn7zi-7!TANO{RU9V>KNMg{-YC0FXdfWSGRs!`lV`iLL#ol(5c&K<=_p_u3OYRR6Ny+nZ!PhB z0koox$x4tuSkqWV*X_eD>A-uQ(b!PW3sRo92khN|dR_yo!Ahdg_W~tB)8I}8B?)4+ z$FlG(K{rCZ9?0Jfk+rQ?;b79!OyS>A^3Hg{n$t1&_w;9I3;WOM<^Ja1d*U-qdN1Cb z9rV6<=PsOjhH#gHco+wtVtgzS>5BrB`LI|1`0$lRqUmHO^hJSf zFt!HzqJW8bK^9SXQQ&aSA1VTQQQ#VkzsrHVDDWl5=RjW+xY0++u9l+aMS)u+N_kOW z<9%~(L!d7T7>~V6h38)s*qYOW0Tn0jx?UCnYp^s!67aBC-gP|`_93cBvOGUX=17*u zMBSf%%4hj9XqN!Z@;@;i1Da(MMNUK^%O7%nwFo53o9stZ3^dEfU`zy><(JTYl|>?KBIL_SE7Wb^bGiCw=fzIh0@^AXGn3Yow2D<@`JM=owu&Z- zQzB~>S3rLqSgWXgu*vuqm&{97c$uV0TEaiz{s6QkJmLUeCIQ+Knkb)&OZ(tmd9Pmn zd`npL_tHBc8b zUR4OX*u_2&w-gr}0>1zf7uzuus#ziZROdO*A`%xn7QuL+E@n)Yx+%|diHmJIo)!V9 ziy4&#Z^}gGVoRYe0VOVGX1QWtCZ zPG+xA7-kh0#y(Q&;$f4BXAIB^H72W#m*>kb*xmmSZhV3FFr%?bTA3ht*nY6A}UZ^OKc;Z6r73CgGV=3zr8Fb5ChPd&^6u5`;24;#MAi{w>2>)6sU(8 zHBO1l!~PBRZBXK2CL_JQl16N=7f6$MSbSK{{Rv+^YzPNi0rfBw1m#nFOVCQF{{`}=Qx+C*Wsro&(~Ll#6>pPyx{4#1838(FVVbIxCQMoMg2-LZanQ>+XO^I_4J(95=Q0F#koD!LH{~hW~P~zMsqxJUA z-DeqT66by#?n0o>{XNDvK%Lt}`BYr$+_h!S9p>XM4uxS>$QXq&TOtC_M|Nn1n0gJ_U%BO+gQ8{TF$rT^EJ53KuLn~DZVAB$x-aJ z0_0Df+XAi(lJGd1JBnMy+r+v1qv;3KxlL1*(gf!|2=0EMaE(uRaPEf3(J%SBh@4sB zarBsIRQuP6S{!{o!OR2d=*G+hO|hdleg%y<`kV0I0Ewfo2oag3`R5;T^m|1lj{Y-( zAAvf$F*DArpeb?mO(!vB2Gr4w8mC0&=-Wdb2F%fGhvbdhX6ui`gQUxt~ZM)E(hv*#$-b(uPJuD^Tl1o^&W-45G1bmt*1uLtl|Rhx*U5B+YI^oi?@ zLANhZ*E<8_G@!0$vU~_2WKdj>mGJf0=7z%KW;+o%vw{xK%SsK1=iNa(w*akZV`fqs ztLRlw{|CJ1nSjN!eL9LFNj`kb9ZRl&dY-YE-7Dvt=kqxB8c@$W z0b?>y&ofy*1Q04$JgXLcrXW(6T7Gecq)*SnU8t^``s#>}KNR?}}o zeG7QkGeI^b2#wyKuzv^YddAPC`R01vCUX*$Bq*QaTY`)(nbWS`qNofKdv+`{X_t!Y zNmsuIQS1h^t2a$mN)x*Jli{8K3IlzKhpxWid9$5j1prS9&zl+3R=iEr;`J{Q%=19K z-k6!7DfapwB)Q`CYvF$d60hIJ2Ol}JB>VCwUHw&WArh}|bv&=o0`+=hX40Axuipm2 z)-X@XT6-JDw{{|OW(6GuGgTrMr+}KN=3^V4z$^*yPHzJCBk<`ciX>@o z2f-c))ai}ItNZ1AbNYke9so)bluz+3K}MH&)#)wZ${?}bo@B=SZ1FsC`tyk597&eE zWMP`BV3{U3{Q|i6fWkPR@Zj`1zOBE>Pw7AWP91xXB~JqMnKRqM)Z2`-EZ^`{a#EK$ z;&Zg@thFq6RR6V~C>F-bV!1p>sL%T`wR}6*=$(eg46*0Po?7zUJ9#GI zS5_+Y_iy|Buu>mORK=6$+LUwM#w=;N@y5v^cW81CdT_Mqk?&DzJvkLkVUB@Ph-*3M z%0Xu^pyg_|67V8+y_b8p-CJbhUi`c(=SP6J69HHGHfe;xsszRK+RLnHpI&= z2R<-cl)B^lE!uVaH)(hc+GFq+f!YpS*T$?2miVUpb=;D6KOW;-!LB}R zS7fPx>#0SP10T@G+HcYW5`^$H@h_%dDp5OtU(qKBdNM9@t|lArB^sKN50 z4_kD2&hm2l3 z-gSJY$!4_@$HAdI%(9VC$ETWPH2l2@Zj^<8vV;w4-oCn15J$nPumnEeq;exk;2LN* z!oME4S}Dw6DFnpDSurst))99x_ye9|SU~CPN z5T5cOM9wTt6aQ527!9obj}Ku2f`cuDi!d(G5G3V0Hn)_UWe>Znw~G24Dj`TN9z`%; zLMT2+(r`csp}|U0K?uX|v@{GjiREB`=3*?y7?AA5c&ZP9oiwEMhDMui`CsCdofyxC zbf&0eC&mRB_W}L%*F@~ZD2)p{F>W9t`SkZKI4eZ7ii7xMo-hDcxXedWwi9F5JCYKR zPk%QjMccjG%iR!Mzos`S+9ula0ttL1r|Jl+D=56|Iahz(OPykxG`lQe-XYqnD*p3s zE7rR6c-zRhd`g^z*Zs4Vgtw4W#}d|2Aijr#yD@GDb@im&MC^issN>33t=#_*oyw{2 z;H(wVF&qp%1(yKvp&XosaWY7ju2>!Nu_Os%v&9mxEM4&sqz6PL%k%sj<87dyF`Fpk z%rd8v&zO&rG{|Sn?$n(71C7(=d5p^5N)i*FF}IbNr-3mx*X~ zd_qG^()BNhySkI$9w*{QIaq`79?-#;N%ASN4ZaST&i4pF-?2UmV+P<)KV$yQ$C!6D zn~`9?ARZwJmd}{)hIS_?`i$Awkuxi#l+T!Fqtq79PG-`y@ENmlvT07kXUql5-Avw$ z3rqJ#eYIzgecz0yMg7T+yrvb07t{76J9zAYTdyKJ@>uwFva$^5W8to+QOlrE=>;W^ zg_Ds-@=V1DtdEj}*!{lMgeXJwDFkqw1SEZn@iM`(7DPtd*Gej+ZQZvyoeW3qo!ctI{v$;(;lEj>@CvjyrclQ0eg>MbT=Q>QYqdCOx^ z7l~G;M(jYmMW#qh#5efjEu`@gNuvzY*PvM`I(Z9!%QG1J0QCkF<%2wAQikbI{M9Fb z9Yf)m6?8=0R%$_p=~Id4SfHa~W3nl2dH(!Uyl0>;1%8-rG&a>O6NG!yFJONLbeL|e zT$*pg^p-PnZUaz~pnQsN!}L9%?gr#fr$sD^${=CeHvN+pc<>tXE#5Z`%~YV?XPT;@ zDfYfrAD|KMyB_|v5|=*F>m3T%EFX{7?DHZL?^}XkF;MR_X2zKnH0FJ4;C=wq`pD+6Soj)nkkX{MCr}nVfgWe9Bw-$i)-4`;hdCC!UUEx`ZK~css_e zKt0i9`4B)zym;acK2(3h=vx7uC$UqJGb<>G`YwrDd~q2uJrA_%jhRVltm<=TkxJlw z(FEBbD5}Kdma!M?I-tI2d{!p%DGz+}#Zhqg03`{^r}&njX;7yE`BPt9?q4D1lOb=_ z>Pbdc+wi>aWAZJ&cs-hHf%>8``DDmzLb52U+r)px7nj0cB5|oN{wG9amXAk$aqo{1 zi7$SN;A5b^Xv~Z=D`?CYTb#|?UqF4)4%8P-5~sxf$Jd#_+f=>(f9r9ghV;c12+M)(i>j9J27KnFFYa|#vv1jH47br8$1763hf0%3SNxQOHUZo6%H%eB z1s!eqx;HYq2CgeAjW1JHQ1GY4F0eZT+ZC0?z4MBE?TUlpUI5Bvh+E|q_?jW5OW|x+ zRAR1I>Ea45r^Ui=DYtaRsU&e5uw7AgDtVpEuGoGjI?@%Ngue_#UGX26qrj--(y?9f zT~U&*_%R9_f$fUQjAiSTyKlB1g};F9f~pi|DQOpMGLh9VK<}FsNY_&}#+@#wjDGAU zr$O88jz^~(u-)zwgkivTJ5@<|WnVbD9XC|ep^$F?_n=Z`)O8*E)w7cQm-zkmtwBk> zNz(5_=ZFdECQmZ`a@;6W_AmDPt#TDrE`GwjJX%Oms+RDZ_aWrDgpLe5cFS`K-|X#d z>R%fi)Hx&n^mKZ zo&P=TzC^tT^DR)+`t`I`7K*~FTNs?4Jc-?5SAMGB#Ptn?hZE?9a3UyL#F9w0&Z|^K z>+yI_@(p$;U*IQP$T0{mqxyl*1>dgtN9~ zSe0+4RlAwR>Y(5h0_)s8uE0yWMDpM|d~S9Cl;h!c1JRcH-$AT}aEXVe=3!LZ*r47n zsffKqje>a*DEg>TbI;>rr10u~9B&+c%kBeiVUh@x$=;H^>{&o)J}CK@z-ENkL2Bx{ zHr@nF4aLO&;*DRDq-GN}=T@9tQ2MIN&N%t*OzB51Bwt@HzTe$q^f%~+E>sVbqxV@d zyv@=1beEH~gV!G)G;hC)#5O>CXA%tz!u14xMfe$*)GdqJm>-=UD|V$e=tPf-)-0l$ zPi6)N6v?eoH)BnihC|f%&b?qi@4AqMA&l>H+si50}gQs&yh6{*#iz>@+Qm?n3x_^Z}KB1ejutlq+$?$LEtNdFC?&;K%J@d zGq5_?M`!|yQ-5e>@~?Rqi+tCoE6(NC(|^^M$k-Z21kohI9nV^+f55Pek5%-kElVUGZ2(XOvz*H znMCLoP&9+jJ*Xz9MB*Ec4=RWA*u9kaCBivAzq#jdf)LI{L0;IG-5ZJj5XhD~jIU3O za#`l+lU&>tG`(XFxj2rf{qX(*g-bB5yttQimZ|<<^~2|2G`oYNNicwP#yGRw*Kv}L z81gpfR0@Yf=?Awb2uBgvh43ACIJhB0TXjlqWM{2u%#WW`Hl73%O z4)pKDlAKJBy^|Yuz;on+Yrul#I^|I{vzs z|Bnkb4pC0{QanMk|-3 z$&hSY+h;eXOY-vk#s}~>08?-X*ZwYENhR$v@+65p8SIDr7btv*T31%!CC_&!hGutB zlfd?}*ZCyXNxmuuMpe-@@e6s`L{uMmJwg5+O!xL#QArPwx>aI@dwxbI$rbEb_^Uwp z4?0^AJ_m&|bm{LCQb}i3>i*Etj$#rscvmDSlu?Y5LN9r`8_~3f+7{Rm%{kt>O8Rw8 zKcYDmZW)lRSPS7^CTg#^Cbfq;KhIg>Q`xRU=17GkDbEA8CgZe;BYw+C>QGOF)jawFs+$y>hN`Ng6Xyv^h}fNRsmW(LOkT0(<3DUJ&uz z0n+qd?%4s3KsjuedShvcI5z*t)k&~Vv>GNQN0i#714TtXrvLAxFGtk1(5?XXhO*HKG;3Xj{9p@5-<^;4gTNk9TM#}6(Gj%)ZiORi zv#&98baX_0EQaKWO3dQi0qhag4xu$L#irtDHhn~C6xRTZ(KC-}^9RP{h#HFSU{IV} zUOjz8X#`(zauk0=(Zv^vAvvO^V&pbpk0@nw>%W4IKcb$6yFwB^uzbSqh_cHwzI47# zU{unss^gRH1h)_r@BPo^8ToP&(kSjJIg(%2&$WM; zBrVl<8)j}2Q~y>p<*)r36Bw0h*j~g9L$3Wh#FAY5w_)W|U<;wlSeA~z_8&Hz`@*2O zk{f*cGifx4m6Kt07ZrF(zX31ffb;NhY}4J0nTfzAtvazR9bbP>!d)UsH>#L)w7li7 zn$c*a$yx2pH6$XpE^_lt=~g#BdPMxz&U|M?Ht!Tn@J_*j^Z(bu1Pr5(c#t9XX?0ed zVE!I@jCn~sG7x)>pUS!7pv7Jk6Enj$o$}2^mvuCI*cZz1y>jmNh?RtsGwb?;O-uIG z8gb=F8Ju^uBfGt^-G3f$-iUQGhqtM1WGo-H zV4ei*-5uq{EwB9Dohu=Y1orOEY=l{&C3kmJiyMbn|Bt&nYhk`1Y#Abr_odh-Y`wd) zo%pYSJ&P+R^z4df@yhpe7aP2RFL>_ zebMRztP6K7!Zn~kTsSol_foF2@7=k%a32u9xO2*9v4u<8^VPLm4|g44OD({$zt@+n zW;Yt{>7Z5Bf5_JgqV_`D1;W(?IzPZc2}3TGt3r65|Ca;wW5OA8{cHm6BavMgmWq(?+cmZiLp@D8xc zQdB8sRI(j}A6b?X1Q?KIDOKljUm8TqQk2R$PEul7%9BErWhosX9RlpKl>P{Pft}M( zC3?B9=qN4!)u`EU?-XTOmQrs%BYj}!HdG~%rL1!s_d}fx?DX2p2x|eGU6%5=OEF=# zNrqVwvQ089%Tm6Bwi9G5OHp=UREqLRov=k#(!JbTgy^8u;V5h#67c(m92}x#&T8HS!)YpFae}QP~Yn{)mO1>gd z`b3mu>Z|1f?m+|FO_a%7mz}oP(OKAA;NA#qH(7_U2H0+*Dq)tgc9W(Haod3HCK`~+ zkf|_L=TQhMLkf06oIJ|R>&fUGFM2Zbx(Zx^n;@jho5=1V_SI( z`Uk#tRPrSyzORkWZPIUHL#4!i(OXcICN{f3bBavtkz&O}r zft`6(J`ed-NWj0iU=iGhK)DPhviO>zy-@c6*=(0n!?|W|)a5!XPf}Lta`hfz+8Ed_ zr@TtubY_>EBi&TGTp9dRK-A^7_)1kNqr8<$6DUcSyBdWnf$eh2j2V@@j&`{x;VuQX z%k4tg32c{Bl`u(S#HNh?T?` zI91@&D=rZoBKW!f64+gU?QzOtwyq-I&-KH-0F=v6B8#sXx((_iAX~mn|NiK^vP!zT zqW*_Lvs-g1w=9cYhR#x8m&I;G_z)C)z*Se(0xzinSt4sM%zwfCLzHECY~#mxOcSK) zu<%)xW5$$lR=KNY$KjGgaV2gD-!W3YOrD{Qz0bi^YaF>g+6l*iWCSZ%D%G|MY#X3$> zlw$o2XCEl8aoSdIVhVx{b4y^D{Tff%m8F=+gg7NuYb=LTSl(i`DqV24aR z{WnEPhD@X2Tp>zQg!u>$0aM)Zzp_y&yddMEi%&WFhJuti$x%&)WP6vX|8dZN^tDj> zI|_qLPj-|~gTrkvJpVxTJyr6}wKZs3y^3qD*U8;$l3iJx{R6@-V4F+TVn!u3;G3)V z<17{gwyZ}Y91d)AsY+<3x#rqi*)`VyIA??6VJ?T|n(G?3^FMl_ExE?Wj7m|tajCZj zootIf<4e9)T69Cu{uM3lMAE$h*wQ|QungE1Ri&5_B`M(?QIZz@5zcN=vd7v|CeeW@ ze*eF6n%$z(Tz%_lTd3qSy(Q>m7YJRLA5n}^$vY_@1jUzECogizI+LXP1AEC*X5e)) zFIlHlMMo}KkYW;fxTpD zz+SRcIm}YlOV$tQ?GjyCtEK^Rl`oepRnPH~?}p^}nrh@%E?;$@peq7<`5KBa7}(2~ z8c28pzQA(%y2aH;_J~Pi0;7^wB;_rVvRuFJAfelUy<8|WmZhVYi#MUZ4&3!i72;V6 z{`$2S_8wrbU&?1<+GX5lO0QpamocmXmxT%*Qv)UGiZ`P$5!kM%%viQg zxwUGKq3|fMT~L+6EG6xN8=$@mbgi20daA~Jsmp0Smq9sHev$l1*ZUK_pMmXqJ(u%r z1h8FC)e_!-FQ9b2CC;kvda|g+4xKb6Fe-UPQa(ePK)T-LByw2mX&rMnv=mxM7$rWgp1jGJki}i(OVgmq^0D29VEj~*nKJS!-aDp_Z~I5+=$Z<&K2yw zhxi#_$j-)mzuRT0su@U@@@}YITXOUpF)u(_4NPI({4}qU*U*+gCu|O52i&hg;pN9S z^Q6jNQo6jncsz{iD|mMkh_;m6E7iONW|>q^A}ClVnUN(gonRITm>D|JLXPG4UBO2R}t4g758>LA^?Sk{2C_O@; z&Pv{T3Br2_9F1@kC~DTDnYVx!gN59$gLv4J-RBX1t|**OU=qSD5;&E>a)c)Uhw|up zJUdulul*`%<>x&XkEf&v6QONF<0DY8o(w3H=i;2U*C`GQ!!1z$fx90hHxghp5_7jd zr_&^Mt=@)l+$)S05R}$%wJ?Hv%y@#*Af>)Cmq}&T3JSkDkgQ09w0N55e}Qd~GK3RB znb0aU$N+C+kJ$LF(Pl9r4e|mSPm2j@kT$D$b`01C8SWdTEB==>NN-V+2AK|L8n6xW z9KtgakOuh@VH>awq87M^gazLq$!C~@2DU*CL1-rdX^_(qP6Mt%Huwfn$yY87^1S3m z8e}XQV}NZCW%5+O|J5K5!<`4TLB4dv|7?)5@(t3D#|E2J(|)pn`Q-X-g9E>=D-9w~ z9%A-QVBcP@yqMR}>^J$W+EK#~LPy?Sp8G7T+0n6YFISx~ONSFcvNP8(>PEiwsYLEt zx$I;ZZ?X#G_PVB#JQq}vCif!q zlBi^Te!^P({7sSe0Xq&*jVfN(=d@b22KGU+^PZ#M0XrnUcn!zPYU)GY_N~fQy`;*F z2a>nLutU-9FEPUf*z5)HFyBNfshZ^2UXd5P%z8y&FM!J8#TON-fD2$U$5y+^e~bHO z>;2QSDyilAmj|ut9ZaKs0`05USx^e1Ls+S-v6P4VATF*aJ%&TL${Q@J2KEp>0-+F; z39Z5*yu~M+e~3MVXNw6rgcqQ3znG9i_<%RLArI^!yvynE@?do25UwpsatL1o=PFA(_7GMJEK&HXNwXS23(C|M}+m* z{S)!qL4N<7rk)dcNsCmua*%kgK4}kzcH28VH46%grWbf-%uA}1S3f8UPlD3mUB1}} zlE)G#+}J9u6lN*O_<0wMW8rrdov#S6<*%dAS_a$^Uz#X2$zSsZ0NmYO6rfuxdCpcP!1Q3aK++7*xYU*{__gCW&C4HMJM@>ifNyh znOkM>5tRzutlWMPZ|X^Lv)1MI`zDxcK;e_s}{0xxJrj(?Yd}H2xP(3j&{VmkEj)+RDt0T(A0_$v09xT`w5l?A>c0Y4F01F;b}v59EmeDZTum7d^Nd4!CB-Ug zq0kq8tt))pd=gnlRQN7Go!-we?LnDD7rb1psX5Xorjjl*%R4=2l6)IV_z6D`pmiT8 zIOW~~uM~Jmm2y`H`N?yjgd_R+6ot(o984hg0Z+#fC@yhH%IYj}het_n& zB)+5?aZj;-8R#{h?xsoNmWck88sQ4|?jd@Ys4pTA&gQ56hpcA;k(n3S^8le4px3$k zo0x$<=Y@x2=F|hiJK06S1l6S3Fprv~&`AeGdxCOm%Y1TuJ7d zc(|RtEjRKCI}mi!{P)|Z$?@0%ki*^3t|kMcLGn)m$sAEHl3t@Ja5C&j{vU}YZBI!_Ptmxf zDU>*RMb}6J6x9`Jmg|rj8_VWpc1Wd#S+sH;(w>%; zo0H9@bG|Q?cqbPNAe&Gt&W3V0-}XhO63VRNlD)SBPq#3tD%qv1G~`b4uso)b;MOECp#s%E4&OZ zcLUtlMfq6*#h>uF1h9^{Dkrj(#jy}ad=k`KfOW{1A}j&YBm=N@pqp1qm=Tg;dH;`O zSQc7sg7y)}SZJke+$EP)S>v@<6G$4P!=CMNeYUhW6%Ow1Vwrvuxls+RBu zeCA~)W|Fh&XJY)aDUAt?N?wtg`btuknV1PAG#1$VKFW+`>2QV(<-X7JP@e^ECPo$F zSqgq8<}27=0DB`!`P{Ck*c3VwllwW38-sEgN@VdhLrRyz*{-ODbG=GyS8zG=nRb+0 zx?($$Xaj6lRGms*C$lSVmF_ED@ih3Sf~YIr?Me_Bm0UV@Mcf^2P?D~A9SYY1+ZB}= z%hu7ZxB%{aV7sCQY*$p}FiTmx;ydWQCA!iTH6Rru^Ug%yzOD;wS5&oxH{c5_UGeeqT`|3!N+Vb&TCvgTiI+&i(i2Nar!%k(uS^z= zR?yLg9}9I1a6M6JtRQs?_8=tM_r!-_KL~73OtaEbR2;v6?+v)?LAeYivg~VyeuDZV zkS!%Q#`S?&7U;+mQ7;Dvj%iO><;Hk}Ejgwxu=kLaSH&!YCwF9*(n~AcLp}@k>7ZyH z&7vCg(#~k;edGIyzZWFsa@9-j7b_9p4Wjt3{vPojh`WH~_IavbsXQboHp#{Jk_#Dr z^g1LX&A5$E-+oWY?RQm9n1PXUlg1SFlW>5`>j0j z3JT=qsH(#UOq{mrG6DTzA7 zl^w{6oQ@uW_As!gBjp7~C4DAErtdG6?91tB2mG(ayj>nM&zD&xNmn*es$PhaoQ@9K z#)K)brz2&?j7pIPI_1_RJb=OsU{6Cz4YQQ=G}Pz|W}$(uNw6z8RgE$pOaCHIVDdteW7 zWyXw3s-uVa)Grx70(Z_)T0Bd^pL15iUIFYmNBK;%`koLa;mOnH^O>C7(nzI03JV(-9zOYGY&_KdGo zm7?57UF`BhQIan9Ckj6U+r^X_Gb(wVa$T(KYi^tX+r^X`W+`bGTL5)FDA&bQjU}?K zi*1!WNf-MN?pwfiv7@)=m?MDgVycwLO3QY!EB?8j009ohorf)};GE%BVp6!&}RcMXUl$6LffSj_bKxAKQwxkG^CoX@YWJ zEWZ;e_J;Nr{5K@U{M-zsxL5Z|+#kD4l$t^N8HJxfSV*A$w+u)@cr=0I5V`cA&fiL)r)m<9JN0EJ|qnel7a$yw?!DBFS zG$?FGttpESB3I;JS3N9-F&OFvLSII;C@J)kHmjUcUVw2s)F~j^>;uOWS%(vCX(rwt1KnY5qERb>D;J{Ol#H_q2t`Ca6fqj-(d7;kgg z?kcFGfPI$uL~mB5tY?Ymz`YN!*_S}Ic2^_rv&&<&Yl04+NKWM8;g_Jj2+AbJK0&;b z(AU7e1WGkyW-jsmso{d%91Xy(h*M4&ar8;y6Nv8)Y%b&_P)GS(Y>-^M5VV|e6uFRx zd56Qh7}$q-mB+W~oreGBihBvv45-tAeVDh4BN?NTe)PgW%)1)ybHF~#`!&LsAbOa$ zqKfOovYdm!DCW9k=OxtEfl{#)sxN^) zz>cR>E#VFL0?J*21rkX>T}qmoyoW@bprGM>7YgsuX1_@c~MmW~czo`t#!xbc)K z#IqEr{^&WsFJNy4c08qg?rXb@`%LK#t4e#QLQpP4i7dWmNa3ganNuAG%Ih7rF@|CaI8E)>55 zc08pjRkBq4c&g4%tUd*q<0)0E=GF8%OiJe+LrSv5>1ec$0_9Swo~7c)Qy0J;l#!B* zr=E0867!Na4k^u(lw@o*1+B@zj;)?XcnU;gtLJ^2rlrUP1~!QS8Cz|Gzs2gv*y?qs z6L=}+QbngSO=8DZmG|<}2w+N;pB`K3+cX2CQd70y##TjW9t!N(N=Y$K*HX&ZYOJIv zW2>{_pD8KYvDLiBdOb}`ky;*OtG7f+##R$ixB=L))uRZDBp_p}4G8aoXl$iMxKZE? zH*%TqW2-+P{|4;XN?F_^aD07JiOf~*?n?49w#xsRd;7qSt$H9FF98`_jYhZ<*s;|j zUl^6hm)-o>YBtQwSNnHRP1LO6?qNZw0)VQ?tW1Di(jY~zyMeTj`-UIfc_9w#cz+Ti;E#VFL%*sXW17|h+ zqNXu{QOPTk@`f@_n&kF4}(gNZtl$KSOeazR9Uyq2u=A#jrPeuIcxy8|lfz zG_!pYre8cS>a4PVt>16oByqGZwdz+pgS5Ro=@ccEp5`tE+s|mn13vCBmN($k;)Gz= zZoTev`IL+P9_ONU6wOD_cr&i6QBmumYI8M-;#=HG1OHx~u8--7y>jklSLK;u2OQAQ z)c>K1JpB{9i68rNso}lo;5Fr8p|=~fH}~)_4+JHJWU4;=WmO|j0>z%<$I{zcx=^0^ zuD)5vzdRh26q2d>_|~dM9tw(m#gC=OEnO(#Z_Ujw{L2|IDI_zjHuSvfs*4vHYs8P$ z_@cC-$vJV<31+?YkeWh&@v{TWA;gPAkrcM6`dzJtMx2OHp7ThTqv~tymuGB3x&7q# zH@w2s|E0zFyxL}Fz^iJ0J>W&DnW_sR%h)OQN+#JS-hF0(-cNOf+~u|R+nN(8OqaP__j zZE&<1!v7_}vQ3STck`O)=Lk2b_jt<0o)0>mFh0*Li)~t3Xkx2_18<^^I?k+_XJTuD z%JNR>jzdRQp@M?SqSR4d>lk}Gs61_Op5ditZARpoJ;7dVc0DQTYbQvfu?2Ik>(Sag zA9QINGqL%>QQx1`z$AX>@dS?iE3B1;k(KM&c`ZnZgFeVv;pwM6B=l5}b1GJrTdzE^ zH=px@SrZqp{HehD;~P+X6MPuE#c7H|Zu*_-LXwAJz}?k&H_!z1HpiYI@k5?{gQ#Ts z0rzUYV9r0U`pn-Lw2;g<|APhN;EHdybvLnzR8EVA=4o~;mYwt3Rx+_!LCqUS<(V#j za)SWyf}{rnbJ!E|f~3*s4mGh`f|}nS*U;QW+$dnXlq9q+E`4g}+wubM3FB?Ql0fR( zk5lHoqoR7{p`3@Qkz0Ze>ju+1rW5tgUkt!OauR`}7wfd~WFD4S5SUc0;Pmhfc5m62 zW8Ma3+ZN*PZ)DH?`xyd(A@^O}(rfYf(IT|&ZTpC%@GvpIK=~4sNkot3>`B#3@?{N| zr(Cn=5xkI|$)k=tiw@SEsIV(PFaFJ}1qcr%@X$YuWI)m48Z3Emnw5;4wsIy0oswJG z9e&EsxPMvn3&OVv3^WOIGANq#Z&R;T+4SZdnb;lco_w9%;Vb-f4)|0INIoTd9^UveV09nFkyDFr{psNE#pLklAQ$3%t@FYpyXEq4HK-#2dTcl6!kSBrwS) zx$Q1F3BwJE`Y|_QJ^^8hz>Z1@^ED{?_|~*iHLE16YQyd9&aa$c0VPvDvA+oMa&~^} zR-qK2_zPFpx~?|26+JvPC&n<>_o?2Jh32`5t&)YMM0FNTQ1o5~FEo}{H>k?n0h*aX z#190$7QNTj+loh=?N!5-?7fBPn*du;+nWnKix-SYiY;=Rg{QLn1t^O_IFi8JstNM| zNM1!C9LUd`gkA?lXEjfoNLmy2ahjXi9ac-2AK`=#64*}QE0BCxV(#FlMs->n6dw9V z6Q7I9Ug9KXw@P=#lS5&IXYtbseh~=!5EzIs0OVhtY~j_ac**IT#{Gf@pOsRHw?VrF z{!PH-_aXGFt8J4ZMQSK@&F#gJb0IX|XG3yIj(JHJUI;n4R`R%tpXHc;0u+3ip%i8* zi^AVvz^55XIbKOCAa82khfirJS?yO^HqK8k_$L8U z%gkBjEUmn$T`rN-Y|aW_{~}%Vs-CNroHw;wax08*6F*mCeWXpnWgzwF&}OC@S3uc# zQ5#)3aYdr))u+(ztB86W@+=UpB`~{Y!b}C>I|RDc;tR;2s8-#k-U5rx#lpl`hp;ib zuOj|@5Vj(4;Q3sN5_pxs`v~uWl3xh?f$$r6I0z_mTUW%R zT-Hls>}=d1VH$vfS+fgFsbk;}IZ4-)C8{Ql2p2&)9qwr$Tt;9V!dOv0xqVtW$X1>} z*E$=@e7FyZ@<;-2BD^kv3km#)up8t{TlV(ZNJ~+klLzP2U0+6ROMBMIOPB+IZO>wa zBY`Q9{-}DqWzpHv`6#hq`{qE%14LW;<}C;}NkIDM5`@Jd+Vbl(Y-$EOD}k4M#axw0 zymA^jDuDJrTJMROHU$1a_zmP=varCb#k}Oqn`EclrtkE{zyxRw8`2seoJyb@!m*%W zbjP%TFv~#hPdp_*70NKU7m4yc1g0ZQ1BLT5l;c^-m1f1mr(iq_b(PSs%w=lKm6q8) z-x5)F8*iu@-T-48)GZ)?XU*oGl<<<4O>XvGbSh>04O+5M!sGyxzcfQBH(QCDeb0!} zo6y>$&=!Q72=qnhBY`goj6t{>n1Y#HwKSuZz2x&C6R8WKJOueckiUa*o${-A$@96{ zceu32FVJ3x{~8D@lj4sEyCv{9G4&cJcss8oKXq73Z={-+yg8Ykd;m%p$elsqhK%f1 z&*pO{ZHq9VULf>?Nm+AS!%J#jZf3NAF%{}}?5-|CE_=`ZnOMj<(-vQZ{N^)C-Z$W7ezbOc}6DURKDS^!dMk0&=1xpzg zYEf!?N%LgyX0C>EFWkFD`F#Q}Agq?aIs#uKdjr`EozyC~eT-tFjDP8O*^BFoMYx7gE z@}@MI^)wX^rCPBb$?u^gx5&>%e)5Y@Qy$m+iW-dVo{;RH0F5tryxZ1Oa=oN^VVG8& zTnI(BR85A<(~ge@3_Kknm;14ay0@N5H$R!E1Dh~b1z`z+H3-WkP(+|tLBezaVM_u# z5WbW^BLbb8CQK_(wg+n^XR&8Dp-(`Mp#)N0ek(9*T$0B0ew-wqNyH!QPm<>o^=UK4 zfFRkAK-i0)I?Z|M3$RzB1wEq78l#flj8C*n)U;Qk-f&L__DXaW!YEOeE0L-&E9cC3 zWxW#3fqS1Q%a!OAgqI{BSEBC_b^?1PQhny?e1@VLmg`8F6h^K@HCiyb2lh(T384r? zS0dG8@~%Q&{FUfz$Y+YST#0T(xLyKsC0dN|2$0S8?XP@=rllx9iRTjaYn(xTq;GFP z=UtHQ<4O;VN|74FBl>tKef(Dxeio&uujk7kbK42p3@YiXtrJfsst3}4WWZU!Wx^Z` z?0~Zbp^F4$z&R9QFes1#r`n(gI=kMQ4nC*Cy-k#5@c9(NlM;}@=Vu6?NI(Xku~t+H zuortZ#uB9pX8l04CEVt~R`JOQr4o=T9))lPVC(m!tKF6xdDAmx-Gu&}Un>8^RB*exYVM^;b!YT4- z)?{VdN3-lnMq4S&YA-o(E1YDM9M2**{y$DKs=+Hae212VWl7FXWulpHR+T*ez}lkc~BFE+B7$Ge@9 z2SFJyp39)d@~PH}Jl^jlNl%4U*v&Nagt(U?gWOK!H7$K$u&k_ihD|=dnr$;V0ETQS z893<~yRwFg2IDM=oWwlaf)6#?>@kB~Bi>l>ErZCMhEtJCZrOb;9%q zrbtFIs&arCAu96Bql{j9$*4u+YDaNA=O`nUUNS<_s5&-^QAtupC3agKyub~+xmcUU z{k+t~1g}rlh@nJtlKG;kUa%so-0IFmi zF($Jp>Cz-Cgso3X~|YR zZ|<$V6~oE)u*`s2+~7FEbJLRJeqO2exnyDytgRUTpCm9~OT8ARv)55cr*Z2~s(#i$ zbauh2);3|PfQN&BGIW$#@-I7!h%N;ABblpLTRfBQ?MFFjSP2nY zC8qIYBw%>B76kfd|x<=b?I!yRe1O2(Y z(@1+b^XRc3cu5P%d7!PF2SQGHyxD{VHH+L7a+&o@6-qmXGAY)IlC%K!`x;8(9!~|j z%%Rz@CiQ}Ql9axYW9aua59oVBUfPD{9_rkne*pmo^6ZmE0GIwQ@vx1nj)y**xh()n1K2<=b!zk~8h zoE-gO8>ykxW`Zg;JtSe8fT(omd1_knN+&)iO0uT57YZiP~i_FYZnJ$)Hf`KslAY zJjIpfD<)8*9>zit`&~K?qCuNdelV8jr2$gwY%*`VZ77Gr2x#`FpcRw0I(lp5B~?fj z@}9C%DF2edzd_iAK&wNUCjnt60s|2Sfc$fs@zE!L&q5_#AD+q!8XZp3;V_}WzgcvC zXr9&yGPI-Y=Iss-{&)yF`=LDn|8Y=o%CXHolW*Jj6rE1Duv2m{l;k=5(8!4vtf%y( zw3JRFtOI2eraw~i%3{T{*L3_t!^1MooBT7HwlLE@D=kHQihcMsQtTnn{>J`35RM?w ztRrOsrl8yAl)2kmP^maFU{Z~O#^H5PdP432!pk!Q7ZMo3-b*FsEU|=emjwC{cn;wi zz?R>WHs9cq3B07IIq4d2T}0RD4=uktjr5Vvl1e%TN_7YhJsECzDzqJ#-zMp026!C{ zWQc@dQoO|%ljUcK+6wJ6VB2nwFJ@Y@iYC=MIPoyik+y4i7`Ntu zZM$v=$AZYhgFYpdB7HK_GU#}fTDS;>A!=b7!tG+ANUHixZ(#y?CEX6THlKyCQZ%G$ zH9(@%DMJmXVlt#CSt;sPNpvo>Us3;A(#hY&v8qa8hUAoP4QkdIhSFcq8g`Z}*WnoruajtyD25+1aX^;&5BAzGCP9C<{-GzQ^*l+HYo$u&?WIwehszVN7oIS-^d z5KxK0wfp>LX1FVaF&o5amt^f(my+>)L|_-`;rJ(0JhXCm7AIgz5u;3km#1I^P?pDK2Z-r znV^??@280P3G$DiK-y7Rdld1YT3~R0JdsW4F}Pg@&>eOcEJ5wl%;B_ zpO`RJfvuX32pxc}8r6?^OPN)3ieyE){Tc927Xwl?*CUJvQPrpg9LEZI@l~@3@&eJ8 zL*PAxcO)QH^9#aWz?K?!cY!(6nPBW9o%PgYtkp=e@B~rydolq5!g&O`Aas_%-2^T~ zI9~!&2uwz}6%>iNi=8>lKa07;6FEsSl@t?J^(mOkRP_gh?*OfI39a;!>#`<8l8mxm z<(y>AndB=u5n6aHKW$Iq%mwV3R273PMLXFjXVOs4p#4zj3#iC(c}+bz^iqyZjr0a2 zli?I76JU=6;oSrtL0AY1u3k{!RpMS!WjhaiGn9AWz9q_c5cn10XHYVSK>bpx78JaH zXxb9bs<;a9mSi6IGbml)b_U_E1co47D1n^>yrr{(v?XTVDp<1fz!yQ8isEgeJdVIh zgcTAPN#HYtPeAEwPS-uv;;O`yHgNkRkEv)Dm)31=7Z*vIJ9cc)D14LMb$W5q7rSc+ zTzWFM!a;Hwfp9)Q7oL(Zr+_8FMnYZD?6C=<^YR;E{}9hfz9Bb2UXU9gEBW~p=4N5b^TI#7Tyw8k*!tb! zSRc9|u+IuBC-iJRv*@$JDX6W0eOCAnZ&D@o1rM$hs}v4k`SdApdx2<6b$Ydp+3&P? zT_qWjO1Nw^Ro0ED(J)7XqBdRl9HryM&0KhOe~TX-p2F^@iGLC#k0p>ilAoeC-cMW2 ziP%@gx`n&hy%**W!rn$8`7uBCewW%#_*7gDP2KsiBa0&|OANEqt6K`)>j_5Iq>I4k~YIZPpSsQ=)1Tn0H3P%mF10 z2-G{16+r}gG$YWX9Y6W)8n!T9yJ)Mbq#Zw&7o?V5O+kyHbvm2Y1qE{%rz?|>O*?JX zu@A1Uf-(T^*&w-;0Cjp`!xpBSPf8_KlcXA4LsCtl-Hh5qV3Sg2%xkNTeR5TjS_*dw za7ir*I#IT5FENNb9AOtR9|7xK-RyI)=T(l^{Gk33*ORJvSO1*D3OHcBs|yCO$^n!K z&3aeg68a3Jy{mc7mN8w|VGHl-Jt={BSJTgBJsYsz)w2jILGdD|@PB$&Be2t&{8;ZQ zHjn`YYSz1Y+vj_&q*czldIjS1z425Ij~>G3Q@?_I4&VVx*tcvq^%vYFQyz?Ac@_QL%Rlu4BJt_BUF{(<$bUPpLA z0^(hDIX__z2iCjV;G1Be7{$B#{aP|E-qn0Kvw9n8ZCfX#YWU;ECmO$vZ_)oU!tig$G!v|?brtH%)@lYn?v zA0cc2WujxftC~Z24gxsuYM-}Sb3InoyE+fX8Nhm1TKU$yTF2fs5+mMK>!D1HfHG0e z@UAMk>x_C=+peQ~i+6Pws|fp?wuQaxacJ9s|D}suW_!cABD~l zz*fy5gn__Tjj9FN2DEB!g?qCoi+8mQVW|Y9YBnN#2yE4;rI@#rSv8MKR;1hSga4-( zkg91qj6MOPs!s{r#Wk1T()=B4`v@#orW+4&`RQ6 zHSv8VEk%_h-qmm9EA3sqi`v`3dRI#4Wqsa6q*&g&`U8dEYx|9n7u-=uoB;J+k(_1Q7^4`@UD7F)2@vcrsI86fLU0sK8Ehuf_bSrvSx4H9L z=5WO`Z} zAKLi;@U%8#ZxgVdma-W}VeLOWEi(!yQ`q8Zb#=Ms*-~Mvr_~Bh3t&Ah<%FKCN49!e zy`i2Atf$r8n^Z|hbl%e%1@{UNZPwE|(P{CnS62z2+(?y)r*$99nZP<9%EO(I46CR0 zGVyCc1?NNAIT71CA24?ayS($E?1WLt< zcM=6Q>uC+}RkX|%_O*!^7fD^!v-H-Yn{ZS?l zCs)w%o>mHOE0F1FeeKABQ7JR2nr#0_(QWZ~Y#dyZ0zjIzeZIV_w zPwVO%xcv>Rr=>(%6JmK!YbT^_z-SxctknvqL{H|i2Hna@8CXwi62eWOOmwWLwT{q>ztDRXwePCM8UDU_C9ZeCuhAVed$Z5l?Fi!X{8A${C*4L(ZytT0L)bo>q^^-1!C8 z(^8f(D(SV_ds>&k9R_TB-i>gVD2u11DuI{0vbOE>aGw=rY1^+5zL0>p6S=n~Oaj=p zRZB5%DYI?&Pj+qF9)4S3+qNGV;$voBtQ(Ra75uM#)MCP&TPQfh)wrW%@Fe-TiS~Z>F76V%~7b2W50jZkF z2)6=THL4%;mNKiRu4F~J{WADV#ej7C%?O)7R5fZL?k#9ueAWB|dB12&)wGz($QIbD zISJuJz-B$Icf6-HK{_j*)*i`%cv_c3yG-=N)4B&?h6KdZT8;3W1Y{oT3xut}dRqVU z6}?Q%WqVpxZs%+UtZH|J5UabI5pGLx)HVOf%UYM9%Pt^6w7;B zkE8IIP0xB-sup^+EKloW*c*ZMwEji-8(2?EmE&2;-qUJ!2V7u1tx|-Zz%F&4%~1?twc4SWoK(gw+y|c`R>9JT2Ad7_DH*ds<(k_@yX|r&VjQz5Q`r?t@~YD`*+T3dXlvOCCtcv=VE&8ioW;b|$K z``Vw8UD*AmWqMi~kEcWYe|TC)VebfFJuPL&BX;(j)*zSzg)N@e|6Hzl^Q_obPir#q zw*u>FDJS%7J+jr)dJ^g~U_Grby-AgH6zDyzkKt|v(PllZ9ZriE*}6(N?oO&qJgxmO z{{q(eP#*7ekPNG*)$ATp2Nj$TW%K?9VXO1e6J`%#mv=stjT`10Gs>bkA2lRD;(Uxi z^HLBwAHREQT1qQfh5I1S6a_ZxY3=h>w8s^;a5~1t)7lL0Wnc@dJm$1MC;RGYZMv7q z1yG@|%Ek#5wib3j%)f+PzOc&X+j_nPHiy>J+OLHzn8_6aM1@sq@;{YqVM`(R6a_Zx zX&t~z(W5ObMPcdcZ%e_&(;AJ=mB4yh$_$K3&O~}n>p8d&2wyxcWrbeyeD$>Q?&A}& zpuDHGjG7v}*m+uK5`8Rib6Q08mnfOj8a$hW23SvPJ;HKO!P6S*3d+#YN!szn)0%h} z1r<+g#~fCc0PATfljUX=biAik{eB#Kkm+d+cjUmRl$q3zl9YH_olq+RHYsJYLaSm@ z@^*VWeFpdJbW)M0mA}P#TK5n$5m--ak&*w5rY_Rq?bgdVrOMzD`_mZr5T1%lV6g}~@4qlWn z2LbD8bwMZwWujv}t!oL50?yNF;H^#*tLkaJ4dYc{JuR(#>uD7}!o3t=J*^oCQ$U$0 zXLwqzomKU;o|@@At$!fz2iDV4mN6>nwc2}HO&?8|Ccw7mNeCwb>rSW&^T94LudHo* zIo!)cS=x3M!o3m@cj5(v)xfr`>c_k#ThIJR+YTn{wr#(L|D_m^wynCDVIPRx3AGUS z7Bny3oj4S7A+YUyCPIG+h^I9k;W{9j^|TtDsMjEs)S7r&@6K{veE~Z2#fW%X?;yMd zY}Ke*U{vx3v}%5a`;#b3)iijF#mB%_O;?0tfUO$Uk9kWroBT+@u9vJxw;v4u0x=*} zb34Km5LJy@hM6f@wDzmxEokc>u_JuHNV7Mwx{(H%r&aI58-z}D~YF7 z;(DFQkR+qgiorL@S9(tC$j6zH2iDV4MUF>rB2p~xX`PS4d4P(P6@y9+y`(OVpVOKO z`!--bt(6EXfc3OgIi98LJ*`jSZWd*k(~2!6U0^+}mI%#(^|aIyZ+dsR*KBxCs}yce zU_GrX5k^WtJS}fYJT28Hl?s-;r*$8SGeucEt(OtjNwt2Qe<5R+W{UeBj!`Dvy?z?5n5sJDi_Dg~BSE>fVu2*s`Zt zwGJv2R@pg5C7VBfq_Be}KT_C5Xf6OzVU?QvPbI7HKIHdAfz5ha_xQR@OE%Va^-gn1 zNIb25===$+r=`rms1#{rtRC#Jik11mdRoc~v-#?2eGm6rP~Ow(N==nL>O8F`&v3^a zM4nb}qK=a&@w9$m_fB9vt&(SXWgV#CX?^HhRVD5C;%WUL1r<+g6`Cu7b$^r@Gb*W$ z_q4u*yA5P|S|9tQR8lob>Wqg-N<6J<&vCW`HYsJsj7qBG=d_N1+X=X&B2Q}>^uL3V9={5d3C((1?-5!L(wn)9$Fp4O48xu62;X$?d;8x(hU3eM9?-1+*rMAH%FAJ3YAsqe^=oz_}{)*N7+ zt*d?3H%M~joUI{iSoaRBv!z7Vw?QoLY)ywW4OnMOiJ?(RyDK?Q?Uurbv$Y2P^C0bP zDLtO0=$)-^P}nX?8P1lfk^D}^Gv%DEsxL5L1!WRtovqG<4g=QNnvXC`0y3>t@FI5~ zfOWRU`c@brMsc=|c$kcfvo!(EIABMhk0Cq?th1#`+^G{K9d&LX{#{X!QRhB{KP4ao z^8B?4(+Jp6r)tNHN?LC+X8e6V$;+s-JN)B-9d(XCxD-Urmg;fijr9FU^BxE@L_N4S!e5R-x>Bu0dTf%rqoGswycv?^(CqkSSPC^!XcncxYo(KoX{}f zoUF&Zwft|frcTz&FrEX}$8vtQ%5bipb=LGlRzEFp&eaq&CIjnS zDT}9J9A7Wb;#?)A{cO7}N8t%klIbg@G7C^qi8f5iBX+j!K1N|9ux+;=;V)p@PW58m zLT1}NMpl#3cDBt5UO@>&o`PyJ8&Dx_X}_erf7Z5J4>XPkw%sm47$zpztn+l1AJePk zi;eU2{X)|BMZXKhJAo~_(gUNCS1ecbXHi%MY|*zMd=6~URWIf(WEOp)B<+hHe>GuZ zAS$|Q#=WMdUB2ikG+F^$^ivRei3#T%UEoU2F(;i2=V-_xjudf@MnStm^kf?9K7^SP z5a(zu!V4fN&XIMBuJIK(SIlHPMc+c-t}4zM+TvQwn6wlXUZ$NsCLd{MsQEgs(7-xF zN)NIW%R57-pwJ6YVd4y_TIeP9Lc|$L%Hvbk3AzTAt0XOXY)X~lUQ#7-f|Bysly!a{ zKw%EB&d)0dFNp@;%wBWGt$kZ|Fej8k}{MEuTW4KN}K!V$2yrf^&vNy;{Ff4+0 z4%s~&*mq{APLQQjcb9wj$dhoFfZ~^&egA5YJ@VF!0!g~9d!0t$CHGkcJ7ToSCfD=G zH?YYnKbFP!$({|jA1FS${Hr?T6EC*^z8lPPj--3{i!r5*-5y&hDaqL}ZV!(zbIeFZ z43{}Psffq1{ur?DHBlZal({>7JV%<(>H zmGqtCb=S8H_rPfWI>#F*kPl&;;^={wYB=i$H{=;E`ATI^xIKXVO68RZBPAeTshopw zA1EE*tk{oYh`wzft;4{ml+`}#(0&ov_EA>GpZZ?IEY%MeS2nRpa zR_M|`kNTuls@OhXLHz>Q_Ic9Ld3%kMvf3y2O>S`m+dhXP6iPta=S+nDp!6lDTfTj? z4tb%stIe$Txe@K_fNdXT#j^OmW;Vn902H5Hv3<5_`&`{6-9CJ5h~wul_uN>zeNJ_I zcoCOtpIeu@_G$AL?{x>ZeU!&i+syVEM*I-aOHLZfVFFq>?wEOu=tV-2HaW{DH$o`x zIQgFFZ$Ob8Cj%WNFms{!W25=o+>Qpe?k_^PPy$l-k0LAprH`KwmA8D|YyAX9r4~{b zY;v}+J#jnQTY;^6WyP}izV45DhrEK~;XR`i{V8*ERQHDOTU~W@6jzV8Zg7Nzck*@m z;+I|F{OhZF@N&v%n-OfFybr;?qWfQbk7jmINj_3Dm2@UadQvz0echPvIhvr@^TvAO zUjepms=UjX0@%7a3E>1#x~^Q^q~hyZn}gj&iDzWR*aJb`qE~79>O|dv%Jra3VoDmY z=g{|9h674k5qKYAJ?L>5fgZ>5lNylMf)86oWufK7;HaSOO)D6Fj3O!=&QG@uES&&F zb$KpM$xP%!&Y2Z9Otxot*qonI>)J&tBi(S1~; z8jJCI9E{1g7T#j_8!(>-$rlKO%lR4fF>fXU;Ufg@LzoGyv69W~d702!@Nlr5(D5#d z_T7)d9o1 zOeU{niIP?%ADr<&VRPHd!9=0iyD3$g#4tlU&K zJuL57{}#6X&Hg`ue;tA5A90of^2^*%zlFIqDgp1{qe)!kHM#B?`ow$L<=@4hNsz~b zqLHYT%(9Y@Q;q!Q!Fg=<&*0DVke&vaRehT|Vqo?|B<)N~iFFgZ{KxoHeT_M#$pim zcN#?n7n!YS{@6oA4zQOYKMMjmz@!nbA=gWyM&Ox>l#W|U+a&2*+0}R#4`4y?UjqIz z{!Alu9vCd#k|)`+n9wbNZnHhXwJx^&eZ(fWlulvu9vEK$e+U7sT)!uQ9=mB@;2%lg z6^!S=;IW8%pUswQ6EpeyLKLe$w?F%FZY|AXM}tqfB?xS9r!3?6db`3tx0m)Rf!iI} z-tJ6{i6WMEQ!&$kqr{w8+wgk0{}HkDc8_ALlYq404vcNU_I9dg8gLYo)V2JiAsUf& z+uNDXm;xygNMp1j&;o?Lo$8ndTrMxJw;KStAFy6=KE`wjh*#W+aXXM-zPN9n-a3j> z%ax2x$?7UHcBMo%HLpo1#H!*%yUt>2()Ude$JN(fUTNCF}g@Vs^(;jk-%1s z#*}a|*=+Jh3U;_;Mf~(K_;V!!shZUoD?wN_8bi{>pm}jsvjy^IQOnrkcZ^>pAXU@w z3s#i^{4$r3w;@r@FeVdc<6as9>{5Pb}-#@BSF!2g!O z9T>Ms;8y}KVLSs&LGw&$mMg`hlg|uwhSKGm9Mc)(-`cA*&z0kR7UN$c$=?I*H25Qd z{}h3@G2W8Eg9QG>H~>t+kmYS$BvqZ|sz5$7G!9CWZ#gT0!qHjT&5QH-%+Mqly`lCJ z`m0&FtrnxpXNEq6aW2%!Kz{{2vYf7llc~qr$sz-w+=6Bi@P`w424e##Ii0{~7`s8i zAjZhr#D_S`#Uw94kA#xm!|)UM(+PCNXfJ_@1YAs;7nS;P%~(u!vdBkJZbNvBh<_mP z62>zU_<}&I@96YFzul3`W#2z7_=RDJp3{@g(i$O(j9-bqpg$2D4+5Wn?GRLyiizYY zT_7DoARU7IHh2RWukuY>Q+*Ss)IkI;C$br!WHf83mBOF7-iUV zmgXxl+iRTJD91mvTTm_n1$A3Dcg%#dRAZkV7ed(t_ZbkhB*2k%N>)kn;jXhu;3G98aIuV0=FOlWe zwY?;|-*5@s^MSqJ@GHjm5|Dc_WA|}!2l_RQw9&noeCegyx|r3{^`7Oz8iFN_z(orRoO1;Afdk` z8rwx_G`!{&rLFmGpk0&-3Z#otR^Tkhr;D=PQ>*O5!ZRir&y>q2=6?h6t_djfs| zfA$ml9u#+r(wC#u-PIL0mFCkA?Rd`8`>E1=&E?*aG+*r>IOzl1e5y$}O?LAg@m$n= zC&3>M!sb&^GA^R|E)|h9-#i4Dg3{*ONa!hv#x|cu!+q@NHl60zHJ|?@vm6oFGmNqV z$Jd**^0}8ApC?&4!yJd^7*N`T^9fxAY!lXv(lSOZJ$zgfJ`eXKpQL8hg$=l_*XD_ME3IKI$1o z>iE$)Z$8v(gf8biCGn}xa&$fCt%tf6*k8YfQ4bqFIvchny0>}lf{f{hGl~CQ%$ERm zMce1{C4kejwtpRMzmDw}mu~O(d7tnY`r;(M9@RF=$sWETzC0qGPo&ecwm*@z{l={A zZ)a_PF>CubrQ3^J93Lho@k@q@+tf{O=TqNR8}b!${~CJN`u2nAQ^RkLcedXepB8>= zyz}mR+nB_3ucFXSnR4zlxoKMA=$|~(`IF1K%I%s8g;gqd6kUZe8C?p{IVF~U;Ht7LcaQmR_0g!doQ68W*~Qp zIN26a`avsgLz}iyr{*oreb0S^eVo@-GPxUf*a@CYNXQ!+{U02DQ2unUhJ4Gu(a&50 zgHt9?n`jc_y=u>l~wyw_>+%N(b@*>w_v6Tz6Lern-(NoR#w zL({PO4`kkd71j#q3qWB9zf{5}%%en9!dI>QZZIzYg$@Yt!IR5T&v&#^E0mDNiIS~_ zuwF#43Kag9D3y3w;!q-oOo}vq{FOJkfnhxqjW%?&@VPPdaK-nf5&0&+m92;3S6?N; zN7>czH|_-ie-(lIFzy1Gcd}CFL*Iy+?5SOQtK{+K>G+QjFJ{;6kZuM3Gy>0HYybs!Q+b-zl(Sq)6)IQp*FgCW?w2Bd zg21ppxG@h3FGx0b5>BpBiB&B36t9QT@=s=G1AiWYei(fLzswloS`~@7#VHx?TqV&? zs^)BVoeym~2xbs4`FCcqS%kSuc~|0Jd2>{f!XVW>pEVO_vjEv+jrbJ+RGM^B|)cz%TPDRi~nq$)lf> zJRb8pc|5?b5@_9kzn{QG7#9HBY%1asyj+@1KCRTgIK3(?bGPr?(iTJItY`A2M|s9o zj#?oxOn9XvEM3a;r1LPaU5YXjPOTd4Qf~N%&zJ*-Nr_)d6G_Gdl&-6P2$XfO9{~l~ znM%d^a!^TzzJmK9D3c+sd&(7FGqlVEW(km=^|&pr+&!pddiCP}ZlN{&uOx~uhv^=#QR=@|B9q#`@1etLR_D@ByBdV_q`+{@xEUX{0MA+ zr_4lLquzH!GB90%^*#+)?^CfK6RY=KjBb`_#rrfM6(ioK@|3gmqDYRGo?FPTc;ZU* zD42Ttngr6Ey;^Nc1{JR z1F+4n%tTD1%|8X|IUw>!6(pS?9i`&@@n+aJ0qc*-PsRD_j~n4W4a#JQ#h{|nYW$j^ zZ=ik!xWA9Vwg*nvkjx7y3C6IofU12oqKhMXVh{! z5^g3fAfC67bgl!op_Q44X|$nVhx#grJWmBlC#W8!;yiCZ?C*i~Jmsh2eD%EA>A=(k zWirIdV^KLZe$CKPP>%%i8>VN`o!tI;ixSGA$i%UJjU%zWq*gbROWvO?$zG=yPmM^~ zm%{tKVO$5!jXt{8I4YaBc$2*j99zs2@sqvoa)~#QJd6Y7uLFyOdPNk0XErTGGTLkC zzsBYj@RxzXvbE$xwtPltCn))xK&M=8#DT@$^^txRTN&*7KY1roG_TMU-yZFGi<=HI z43?o6pMkniuHEG$O7(3X`*@uskDcv>{1C6IT&i~7Q{^a=I3c`{H&t&Eg?u6@yxQ;X zwHeN}di_T$9cL2Vy+iL|SY5w%FFsk_-5Ycodrp};e!AfUN)-mpp+l|uxX3R-UTw@V--RCwU=il*l&XA)PTSM|f^W%OPl zeFYJ%^CV`OaO_D`oUgb7b5_UeCpilXP{$iBifR)~qOO-G zIqP}K88Tz8+Dkt-2=76hG$XnP@$S8B+1|9C5u<14JSk>2NXx_fcQF_H3wh0)y$F?x1BY)5C4XynzGJT3ORb(PW>Gm;pj)h;kB2An)0+L~Or za!$^ya2H?f6IbQKk;N0D-O)_zXwUiVAIyGcSz8^k_Itid7`>efN!1*bs@YG3wq_^Xc*uv{=Z-OC{^Oz)6O!97Y`Gsxg2-A2psUN{H2N_|C$Y-+Ku*h(98C2Q1=Q9 zKf%oB?+r}ijS7|5duEZB*UK|^{?Hn|^`*C~N*#X$Z97=&7gfqLiFc|Dnd#Fxt(kMc zd*o0m^rNaZx;8X^#lZXnYmv8qfM;IY-rD@`iQDX~v5yr3OW1ojIN;U)x&s?q%Enb} zxCcypvtbLfsYO^iW(yAF82qH!OIzszKWEQZP`&{E4gysw1*Q@xxT{Tbr}3SocQX6t zC7S!|pmc`Y5%~8L7>O|gHU?hnG>Nv8KlEd0n>VUBYScuoDEq136&@D>Dn zw2(?=N=rGBJQDY%klhdzizqB41?o*!mCmZH<%Efyhn<02_`_hGj&3X{*iM0zm~>W{ zJ!AR~@ap7)9c!_*lpby}_2jT_dc2~K4xdQ!#;13aLfeP2-aXC@j zF&!bY{3v^g-8KB@VH8xsr9ij3#zUH$p!9>=7x0_dd0AU?eY4O9j9JzT z*MFz$KT=J7*>gUu>A)``a3{v?BAfUeHF#?z;@VV1@3vX^HKJ|oc@@sHBHBXW@T&Mf zC>n}esK&RNR0qG2KZ(t&*uNb3rxB=|$B+*c)fvmf(5joNOlhu%J`2<xjTVA8-&7wU-70r1OA ztlZSR6BU{1LQy>?DFUHX9V1#BEE#c9TZ2n)|tJ3(Onre?bJbvI0RN0ly1>4y%tG_)Q5+#W+Vc zJoa}Jm#st?-}+Q_|3x-G%KisMu!BHBgTOQb{$>JWF-`_WRqK(`7toB#5 z=avF9BI^36M`ExupAWEkKZ@@`kf6qjE}*8ANKFY*Cb!LFbDc(^h+!kb~DA zv?sd~sHI~C4%$)TgmTcDVdq6WIap_~nS(XC@U}TT`MfmR6CJL0s|@?XS27CZ_)v=H zEH^8LGTCG0I?6I1*kffq###x;)Vl2$TR~LqmGI@ZQDbmnLV0p7MMzIz}NnQ9U^HxT%;{nL4fo<}UrKib$ zq%9wMlqP?h{cizZn*0F9Zz7PUR{ztt$ z7wAg1tP3oHbEAmF1vX$j39Kuqh@}P*X@C9|yT1VaPKk;j{W_CtrG<189Z-~vZI$Xx z0+R=9s~mxGm;|I%PQe%r%iBRh11&rr_ts52b zdZ383ZuYSMD-pC4A{CEfYA@wx=}UXKh%%nt*XSkZO8> zx@Zx_@0qbAFB3!Cd`_Nsvg-zD*NAQlfln~DN#J<`?VC|z;6F*=Y>YENkwo{gtMvsE zolFdsr~-+?N*{o^Ql;Nwdtq+p#;vsI9me85x5iMc2F?o0*Tmt1e zxX+0ACIX4p)By0WBG3V&9q4y+q>bhy7F=?GqD0#~XSr5leoSx!l)^v0F6AZSyxCQP z^f%DbpO(`67>%26K`j{iD_Dk~YU(Lx>0M$w_wzw0L&^CNDavC6F2}FS1u-7*DIkMk_k+8 z;6F{^NQ@&S@GybXFvfy@cbBbeQ_z$g=z(Rjtgt?0-Gblb=X#a$EDEuys8lIuX&OOq767;m z%1v+=g2G+Bn>!XS{JHuF6*BAPdl=6`-3a8DxsNmOR7adkNhRymrAF4XYahJ5!YwEw zbb}+Rr7ydP$Uh!R!?uB`5B#GE9EZ^p@Ef)+(k^DU*Lg(GurJr25^%_dX_l z*gvWG?a;7<_7;;CC zGEKaY)Rf0~oGMRjYx}We>IU{)L^Lyjeo>YHh;qYog%11 zAg4Vy8$q$&xoUKxy&>k$SujWLu-e<>W8}7Gm1wY?a8{$R!cF4yyf61l6~Q`&xJtm7 zo^oq2lK9Tj7wl`!P0#)w$qBH|fIb1_*XK@*N|Mfs60zsanctyl4Q&DZ>jCS#@?H0I4hJj;Jo>Yi2jE5ID$ukpGyooF?L9xCV^xJY5|ynuhT7DBnfBf`Np0%e}&Q( zat7otu2!17ALn!4{B|$NuZA`h{t)1AB5*OrED5Y7uo`0}D0q)M1$A5z)Y2;!d*1vQ z%3E;X6!Gr_e#7`h0(%HFFAPjmP_TrKUG>cCj^fmqY+wEWl)i9#1OHh9Q!%DUU_F6l z7)wCEhLKj!o88ZDX>=RoQkUDP!ew6RASf@QeI69%=kl;FVkMmAT9tXF?P2VNx<}}n zsSl<5&QiKO#QF$E&5nVo0qn1EHeRHfl(VvY&7TjW8~kFBKbaO+dXBR+2^pX57T>%G z+IaZmB%S=ugbG|?ORZ2Qepa{-(LiX|A(#*RQ3M{rSR;Xn1h!#(2#8zP6`A<$Z*X$kxcfx#F9CD4$-9E{nZASc<(v~-10OTEMT zSss)XaPJmzhQJ#buSuXGfj=;Q2L&(VDysLKr7>AQdml>U&P-JS{O)D~ zp)W@lKkEhKTc}@y@MljIyCQ1BsHIO>6I(mZ9T>NO z!9pon&z9|k-UB7i5op_$YdO&SEdrTUXEbw>=9pb5WZi}ze^crpf1+gHKzkm^ zdWqvl0%sk;MKmbrO?Fh4be2Y7@0|~WavR(mfqx2tZ!o?B^dO^&>ws%kwYo)mySBsi zP^uSm+Y$Je5*UjyMgr3b{0HM|2}~mJ0>*Qo_nia^9<9~X#hq}Li(8g@_zKEC1m8#u z+X-|#5+4Qrn*`3p7%zbr2rS1~3QB$^@CwFDV1?&Vo}B1Ds9WtJ!Nz^;-UAA5zmsco zcc)Y>w_ro1oWuP4ptS53nC8G=Ltp?#KTt5BeW{oMe-w=q%MG8Cp-h8&o`|OsxC7%h z35+MO5#wo)FAdeqWuw#zt6FZ1HIk!A8tZfTpGpMMSQU@rIvs>B`Kpe0E26mEHJaXu zEQH)1SjQWVaiRp|p4Y_~vw-`Jsl?&hHHs+VEZv`Y^YEh?jrsLp-H&c5D45AXs1n~< zDoKsv$=+2^j_XcCf`apTB&7VLvy^Y2>|FLi@l+d^k!HL@nu3;zxJ)d zTM#mf#Y~ztzky-gDWlZH%lxr#H~fdk>8U<5jx#SyCh}yU`}ZBy%~t+@3<6>Ew7&^Y z`=_)HW9#mo_P1=DG4~|wgZ&4Nkf;4EdocV0_auf7ZD$e>^p%aLvtg{=xF}JXjU#2_ zayHxr4tRFP$3u38$){}CZa3O@M%Gu52^qagxHk`Yj6cLPw>Qf082@}GW;|X?l3&f9 zyMcX2 zL!fBmA#Kb<5!aX=CweAR%^U8&%H~%gJq-rS){^(xGVmC#V!*^Z&T8pY36nPr)vayY zOA0I4b2pSlV6g1y{V-cHc|i-aAxehBV7O=M^$m_#o0~B~Pj>k|_|x;)z#IzvP6R6V z#9s&$t*+YKDRa#-l&LGcuE8g4_P6urU5GCN{|y3n_F^Okimv&GCu&i86-q zdIqns*?*otryj=yG7zkjE%)=Mb#LaEfx){VmuzOsFNAi3k{tv-?8ERDlzdHKV_#{<$EJ_xQS;9tX^OZx?84j3%l zl0|HJn$RjxvOFu0sd2D0;!0*0iJ%PO#dqDBCQ8}9M3`UNwbSsXQ%;H2d^LDg{Re(m4 zk>=@~iGPrh2ia9)5D%_F(OX$OKjhW+D)FXCGjkaGyMq35UqRzahMX@d*;@*u**yyI zD{6Zs&pV{r}pTOoDq09yTR087$^Naul7ZC8r^5<%K za)axR4L*V4f54wBi2X8~LX-hJ8s6JFJf-lC6Rw;yEvJ$>NW05u_$kP%fE^9rcp_6a zfE^7N48wzf9Sv&?sW1i|4G(93PhdyG?T6D$z>bEW!Pp=H84Z7iu^ZUYutvq)4LPoK zDnXSIfyn`OP%<53ssvS7)9vPS$fE{3ciSap*SN<7RnN8&T_*BO9=7m&~kpc3-sDha^B^5k9 zr^Z!=vyu#qtBzuH2JFCiB*q8{$iVn=j5!jJf$>uqPk_M^fgK$e$>>-k<~2l#SVzQP z!rUcN84-6pnW+}Qj)+HM43mJ2h;PJL0BodoSo{Q`$3QqNu9^w`f>*WduSAB$yV(5^ zu*2fE5yKd@+{jjk#dSvWgdf;pad(Vvzz&O5!ew<7jT7sz_;k3ZiCBik^D(ZGfDDV* zU_1!yu(;4=qtptsEW_e1Qr9vpejon35`heh|HAkKgu`OhG0QNDD>f`{GA1yMf%U-N z7`-GQ9yl3e63}1%Aq`nUE{exXohPhdO-@+aS0;A9DB zY5elcv71i>bD@0<|09XuIs&OvFhIfiol7I|V-eWrkJmwI3pXR0TL_$hF&Gs7UzRu- z6U(#6bueZ@y-?`Wxf7|SO*u<5E6*h_gRvayQjq_3UUNrc1d1^Z2d3aQ&Q6+g1!rlN?funz zp_~SJEXez#K$SxFsvQRpNYZDt!#2CG_T?uns9}ZmT*=qnDdXEnp0Q+7IZjpg7Cv zsyUf@t$i5S2g(dI(||vOz+#L$K*=ctp2gS*3ZDJD)cdN(W0Hq~uS59~?&rY&f9%jKo+^PCm#WXkV_IrVSiSPkS7GjNxdDoDfx_IaROT1J?12(|EPvkTPks?HjXuX&8oi9+8gnWNj)P_jZra?| zQPR%JA{PhuLXlr4k74oEZNd`(H!enTdbv2K_U~!jd;q)NqfTFuN-iNVVH~s2fL}yl zH^%!CXi4DA)46jE{6+*`!*~V^E+n>+Q6e1A2@mu>gFwFAnR_Zq(l|>WVdXrMQ)NvK z$;-H;>I8;xz>Z70ViZY0?$V9M7zGOCE}h2aIm^YSW0cF_&K0rTv0H_4p9JKN-4=|^ z5|A; zi?IiU^{Fu=T?|?fiNQ`Vs&OV)alqE+5g3O_Khp?ln12=F~e+ zDp!u?tI*5?fjg>|&eBURicsio7(IYswTQwSM#Julh%MhY<_2Qss?@(^EB}Ug9nCQH zlcL`W%!A=x-fLoRpwCMrtX0%5Y`>md&JJr61y1^wFXBbq^Uc!<86+ewIn4Acj0gi^ z_nWJ10#u@OGMZ=KSz+Gk#5i!vbuvn3>8El<{kKj=c`W=?UO^{wh1@eK?PQd~i*<2w zX(vMl^E+^&)B6*QpretpW}W*n4IivP+})rRjLG79F>Zs;xyU+TzMK?RwEb# zLw*?_*rWK^Vt$|mn*2#kaZ zh`^47cH^jTiL=T`D3B45DUy+oN~)S3B4JIbKt?+m8Rh7{JTpc1F|s*N@*^XgKt?on zB(v%)Zq)!&BqJG>R5!~+B2$)R^wM8OExNBpxQ}T{GE(U;BNW|rh}|VqlVnt4e;d4K z&T4745L;vJ-es=HVLg(_k7}DKqp7EXL9^D?$k%Fi`OEq971`MhCO-CT8%M6gWdh`W zou^l#N?v47r%8;DKv9Q}N@;b?E@-t+PNn>AY#z@36G8uHcb0PMnY=M@zD@aWvGrnh z&jS1=4iB1}MAQ?r0zIsH{d2F%BXc*h$6w5!`=Blb{xt+DpG}K^!9rc_^-Duyibcr? zj1g=f21-t&zUN?E1ja0?!`)`r*Q=EZ>&n!-wxGB&#k(8U6VM+61-mSsnlCtV0E2UN_YuL=IHcCWiKwFPstrZrojA7po*FKdHxasz&vA z6a0lBtj8OptQb>jMIuV~ep}}RwVVP)P%D_iLH|QTSX-GxD&w+5!rfFQPPQY zkl^YuaV}(F>#-$9b5OiE%3_&%9Hz&_-Y?slepH#@>wcR2^=`WLPZj}-g)oVf=yFoLr_ZM2^bNxK9db2wKEOG)>zR=NkDp3oF_Bt6Zc z=PX@*kG{)421Y*$(ihk-MJPQHqt_mp3jE=8Q0KutR|2|tbKj5_#H`8zBSo!3e)jGSrOMaaZ#HmRe!`lTx1& z5y@9r&qHu72uqhghW=mq31?})DayGhgSpUx8T|3j=FctY7ukgSQ1yFU!fNRYB_e-S z5|$Qv3c(Yg_}7T9EhLlehG`zAGF5Ol5^+E#!4>a_$nHBHYwp)BJHHvRo*wP@oE4VY zn0iBkn$v5LcqeGPNcdw=QcNIq9!Cr)lxoumQqIx{YX4Qm)>sB=D-eA?GxG}PdQ}Bx zSL6^sN>ba|H4tilV87;~ytL)jzBc+4)O7Yw1@>z@%BkRDbvcN>>~cHQTLHgJ^6b*o ztC+k~!m+`Hhmd+rc0B`a1L!Y@nZ{f-Oa7s_*?lUMpw_Ia| zhf{&%Lfg@E7N<)4lhKOJO*^7!1MESs{6w4|yItk84Y68KaOUTC^!0%4rBaE#Q=t`j0G$w9?{Lwr_S9QL#^!Zh$f!An7Et+u> zv=Q)!iRSy}EzA&i%dpf6E4sm6zs|9u`3u@L@MnX9fk!uY8qZl#+n5H&`dxzwP=e9? z;aro`f(6-J5^CuQR+8`!fzot(U}^&!y|O%~bo@?!m+a_G{z*;wLWPUJ)Cx01u@C)H ziZBt{TH=2g_)`dM$Jh!?(Xm@I=6n~!GO6}PUSoeYn-ep*WhjdCvIFA@9L1g^fIo`B zOpF;47(!qz#zTPL#J;qxM$H?E(}aFz@R|ZFx%i-t<s==fC2}J-#Tqn~W(KALkfj=x!{IC!74jOQbWvkoXEin5$CYlm zm=aIc?JtRMgVm9kI!J62Wqqn`j@cuuy9evTlO=#g!5I$h5F?x`r|nMlIE-YjnB@Dxh1FCHIt;be0O- zvOmLNp9t)-Kgv(V`MT`SrEq71G8xK=@ijvmp*{`dH&MPUrb*5-AGkM9n z@IAz_99S1t2~Pl{vuCyopDno(7yb**xS0)`gXs zh-=EYaDN1SfOTON`7x2Y@O-G(0Ci!zij~SzPNwH2ap!rGKXK>v=pF{vo%dnv0oI*0 z0zPvd71b3`RZhZ;YXux`pGFeBgoJPIpbf~9+$eopz zbb>Zf?wmVc1N$mq-C6mmIA7iQVYm;0G8y6nPf><6e$CJ}s2>9P6>BH>mYto%>pxrF z*nF|KG#Yw6jq^$894-aLzoBwF_$S^E>i1UU&33h`+$cp^)#}qwBa;b-otU-#kF4!_ z@h<%8%*d0&0eMYhAnpE0}pgGiOci87*O9d zy^J{l;DC2-+dd|7McxS3zB-i+qrtEpQKqCs50IU({-YB0`z+ckgYjVni7@{?Xuc04 zYU$!q9kcdIB(sUiE<0ta&Y|#r`z9G(J*r#Q>QP5!tsZr3*6LCHb@iwl5S!?QwR~4| zj}zNH2|FRQsEgFcT@>ba5P3yWPuaMg4IhAe61L`whRVj;m-DzC{O*lAn?=RD*qR%{ zhC$#mKi|C^7aahh=i`cN$jPp|!z|QxIcX`Qw=gFiiY5M;L^A{EYHeE*Hk}S*Wap#o zc*LgjTNG0nUm2_inyLD>RGJJ8yVzN?EemCpHB)tuY(CL#u76axxob66vAT%OFDGR} z?0_R`ZQ|dXRU4XIe{4?lCg40jEwp*eTi{>SU>i}bMpJE?sei85^!~oAX*N~NL&sv8 zs=c>Dcs(MF_nO}m5ypFmhlI%=LikPyr)GzYVjp0yT zmj0Nf3+4DNEH1+TXFwn%GwBfM4Q5w)i$LO2{#eB#6j@WFg)R%h{KWqQAP}~hFsX7w zQ@dj-sMri`BmP)z&DLeK!moZqvFX76F2J%)rRDE5Fvp0lzv?DP6Xgk|#-MIy4F69D zfsjn4Hm_GQQ?U{k^T#IB!qSEE_4^ggb^N~x1VS>Eie9N?mSZLE=Z~cyq}olMP?nxl zpQUp$<~a}u$xOQAWm^aD5=eZ>AFJ4bBC8IB-i3Acf8+n(AP}~xv;>c=*c@FG{#acv zbY0}|tlzThgR%}^p3vJ~-q>`4(hUS62oB+osdO6pbv1{aUW)1(J0)fzjMf@?9Qh4jD zSpK{GvFgpok>g5VY-wP2%l18Gw-09f0oh*MB8y*b+SG9-b9b_qICTAwrX3@Jd|8pO zCo3OLgh52T4loK=MLhhJt@=7+gq2AKmQ}_YS zUQo2+u;z}(^GM`lRWV@>WL6ZL#gsb|S z{5NScmGSaGlp&STTD6u-^M3+u^Yttr2nuo!FV&>un$D@TUmwa(aDNcKTw`g>xz2KF zyq-MLUSQR^fw`Z+USKJ|LX0ohB5#TZ$puz7_{AW+z*14A|0$ZCK_nMgcOw`JZ2POG za$KX!Rv7g{1sialT$5jb9R&KjD|jNbMM3uaNG)n+fC~{1xlrRcCy-;H4eO`U}I)muFu5 zkyRnw5tKiZUSOkFb6W@M;5T-?M?9N>|1E)*|Ho?(V6f0il2p-3LU)6bY6Sko_zh&F z!5 z*kP%NCu5Ql9aeiuNjj_zke&p#!}<*46JQ-yMag(d+F|v$hqZx#b=cqUWhND{4y)o+ zOso#uW)$Z4#sI+2egNdssA@c;mR6N?STo7Iy$_%}6LFe+0K|4|%1OpKd;sKL zqRQ&nl+2fGq8zA?kZ!_uY?qMD^Fh?HWuASp)H@Pplhp5fwObw9p89=@T^~W+4hrO( zAnHZFv-BBWlcTEU*^VZ;n$;EnEhjNAcYe1&ig}*K+!j^_6iLj=OPGAtX>qX)X8%AN zTmF*?e9;}trB+z6P2R8F^ht^N6=>(9m@X}Yc9*~K z@k`_?*RG;xzcqBVbJaoAJUv7EzMUwYd~0Y6alauc%eRIaJrI~Wze-tqJ@;M@ea#nS-u5-RGhx(nG}qVYo@AUu_fZ|c$m(m9%(_%j z4s4!?)%7+1M>g*QQD0;1sc5;&@0QTvx=!dy1qc?|P~r`=(D_O}I5U-W1cn=z$U*bLHvT_iC}2R01FP*K?qY=aY( zT9JqjY>kR8L2$8%?8*6kCt4=?;p9B43$e?|`CdpXfIWEL#CR3hgGWW4Q7cS`2sn7^ z6%m0PJn8j3w+G?LS*a{@EGcnveqD%iz_f$Z2G|2;D8?XQ4;U5kQlZOeR3<%OmcqSG z#B#vg@F+9hfjwYU!pnu_#CpI?d7PIqfjwYu#<&Uavj@yKk;pS@xl(e#)S{Z~0i!z; zahf||l#`5eIAD$;s;mP>$vKW}^F-qAfZ0ekp8|T-A*Z4qTQqxW)rmGV;Z)RHjvYA_ z{f6upU{6JjpI{gV?5U^^#xcO2iZl+-#p`0zQ_&r8ZxgYciZ)_AEdeBK$>?{f6ge5Co?@aTuqUGvF#3byKcZCJ$>{6wWOVwf z(vuN0&Whw@q+He~3?mOuKr4vnZi!gFhxaPR%fKE(Do8k~HlG|rV-6)}atwVB|2q&K zLn=y|O_B=7P*o|n978oWFhw8OW9Tr9uD~8cDoVvu(qm{m+)*NyW9To8_kcZyRGbqN z>oK(M8E*Undkk&E_z>{3$52Ig_d_jLN{*qCWZoV_x-((aax`}g>7Hbq!!dLfQDq%N zO3rcQ*fDgFZ2kq@NoU`qElhQn-~FLu^{RR}6`b=JmpmH-(**c`!#f{i3Mj4}2_jz& z7v!;zLhU!rbhl^nXNQh@Suyd+BmC!)Zv^)8TzLtjmcETC*P&-i9^~@;PxuEwczNE` zWmYX$Yls>gfk-aTkKV*mzrZ?!G80BEr_syvthju`8iLW%iIbBZR zF7dm=?gs27zOqtrz8s5^p=U|mine8{YvSY2+-o4n--tjlf1_y9;{$Q>aSrOcYh z<^Ga9ip%{2?N8B(%N@R%)i!~3ITi8ovT|u!m)lg<T#XlrWQb%4CTc5nxRAA zrLKVdtjB2*%qtB&uKk1_lvOV8Py0j$TJfH4qQk5f^Oleb)bglumftATHs9fb}17Ej|#bZF^aZ2NZ!Q}+b;~s&%23U_%7TtV#zIt3@ zJF_X_m&p)oxs>B;hF;!5;{y3vk5gjW^q`XEHs#i%DXVzg5AgN^>v773~#*c$|uuDHnO%F3F>KTpwsXf%Uk{G3Eg4aVlcP z-*Rc%K5p5+dE9BPa@5MMoSxDO;&Gpl&J(~kurd=yE!AiPU$~oWfym>OmW&CU$K3_{ zPGCJw`KdTxJ?<;G&x0}<%8BtcLz6#aQ3)VF>v5Vy+AI@~d&V2sw@56sr;V*=-KOJFYs*5j0)iu2XuK85=ND3c*R0bcHK(+su#mf2fCe%9lZ$Sp=*O2Ol5 zZ8(myipQM*Zy>N9r@UMjb+*UdCtfNZw-Ejf62JAh*{)L6vI+A?JZ^X|MB;HTBX|s0 zk5gvCsO2>3aaZl(6I8%@+%AkQzT&gcW}QP&CPO(fzGmn%sJnsu ztjB2*X|qW@Zll+HSwG4u9#{DnW^DuOamveeQD=MHeDPB8xSsHj0ink&ca^G^O_)F8 zamB|Y5|5jL;0$0rPMHa=(v%(kch{<0;l2Z`$F=^I;{sTZQwdKE%8AwEu7f%sSdV)I zV-4VEJx)b@r6%&Ym6AvCxNXqh7oB)qjo)}=2&~7cC?}St^|-PB=5ecC<*22_mCC6s ztsovZoOJpC+rY|97`0TR4g53IA3@}CN=wEB&g1I-&Wl{YdYtl8alU$7Z@5Q-G8xK= z@ijy5K-~i5XFX1nNSpk*Y+^dr2LmXpc-#SazX0oT%FA_8XM5b=Qa|ExhaTWs7=#|T z)>W!nHevoqrT(`+BJsFW5sU`bs6RXG7{e$`i z*5gVrx&waJ<5bR-+T1XY^`tVyAvI|3$S z-@bB{GdD_Xq9i6>wu`v8g8sj8FMdPBOBms4L-zVJFh>J!x68XUN=lkFVPswXL)bhQ z#>Jp0J3;RK2uD3=6ZxF5g3PYDlnOm+uOPsI(5x1G9x|RWKo^Yu4^G(35xThs=IYxTRU4_9?uW5%NJnAu_jc>|i34r-5ffJPX(-88>0v0PK?t6?s{5TXptnV5>n=IYh7-{u>~C zlA$8s%q1mxlCe#Q@+9LYNIwAkB%|IzCb0o~tf(j%Pf1#L4X(Nn?g_viD`$I5$^`aU zQE@6J)?=kzGHu!bd#nt>7zFs)xrHNLU8@zAk<9ZX!%o4`9llX3i;`$fO=dlzP)h25 zlI;|nt4aPspilW^ZlTHoXK5NTw=gBmFLMh!r6@J9a|>0Ij%##o;c&Po3SZ_HDnHj* zE)AVqcrn~r!1k5OuMp!)x7T_Id6K?zIsBy{>?>7N>3@p0iAefNlan?tAhPYRn#ysF z_La}2`Pv$=?SECpwCM?K`>VK0OswtSy9$dN0^9zRF(v_i`Er*<<$2E1N|8cVJb}iQ zelu%sp(?8XPsJ1zw)5LhA`5^$qp7AwOrw3%d)3qC9T1&jRZue~aHrS!XZ zqq`pBb&4JctAGdSVRG$lH`>2$+8hsTH#!|-3b5U%3KB*w8z+C{f$+JKSLsF{g8yF- zcB4w=Q=5_s-Dtklgmk0tL3#(+ZuC!#-+(=gRFsORq}}LQ_0l}oM*IXm{cVG^c@Wsc zNX0oZu^vX}Hcp$#z#c}6Fm43=>;vIxt{YX$m6C3B5}CJy4BeS9YB`#FAgp_maSjiJ zZzHO#?pDb;jvSueZTB6{czlU$J_+0d;VYO1x;U`;sFps73S50F-4tD|EKEOKMl4fZ$0a%AqaZXIE4);fk zwD}!ahilw2&4*G;9qvD_0o8J8<8U95dFybxGhx(nH0N--CmHA9aJl$JmcuDI$B|w@(%X;TMS z7gQ=&9pyyof=5H`0m{6{uVfb3k7RjK2E_pej3$5LdS{_K16bF)0pmJgT~B2!t6wf% zd(FHc+L(RKY-bec4rY~=*OBmANmxAZQPOz?*oIanbFRy2w4r~1x(`I2r?g~D;5@HZ zo3uFuSkF^FcSg%4;I6Tcf_o$=lcAg#Uo)h1sZ{HE8Zjf|D2Xi3YrknMv_t|b((C?dq})hJnwq={{f-r-5y2HM|H}Vg$f>c3L^2mM-i+8*7KB^h-=Dt-p2?& z0@m|Xv^l9 zQpY?`cX&oEr=w!pNehVQ&60G0ZD?gCVj6Ae^-$M>$n#W?j0v3QZHK)TSkF^_D$dtF zFVT+JL75EY#Q2&arAy(g=V=mYC&}`>nn#SI+~Ro|l4u32=cy*wX|g@9^{HsY^M=45 z1VYby#Ce`t*%kb%h{W?|A-E7&&r@c?sO2Ul50eF2ooP)>}m8B)3w&U&6E zk#-X2d0cg@8&A2#^S&pE?|}6@)#N%&w&(qyc&m6`tqyeGAoRS~o#&~QUBSnkjz~Q3 zCv_sd7`2?HjOR^2a5}J_r&K>CQqQ{{>VH5P&r?~-$+A6fgXB*v^ilb((C?Yd8Uoc;0sSTS4e~pF7V}E4zZ97Lj;fq9X!eJx`elqn6W@@w^OzR=|3m zQvH}nJ#PrqL7^OWkxMCy69I@1}0GM=Zhl#^w9 z-WtiDc-~Ryjs({8#$lWatmkP2IW7X1S@FEe?}WkFI8S$YMlGkq^J+>9i055PI#&YQ z(8^52G}_QFLVX@YeVz)EF@f{EZ()B8tmi2|73ZtxRWC~OZRK(q%8BtcLrRyzSv<~jV^E_rFEIyZVi|2hw5}yO> zd8)~EnrzRTFWxGiSEXy(R05&rb#R`iR(1uqIR}w=URMM~zUo#J zoeRojC@03(3@Kd-SNwJ~gJC19;e{*V@tmsFxE7|wy-K-&wTv=PdUgEnY+g+|D}mpY zz*dY8K+!uqv`_)}mk=~LHZR{k*!Y|Me~LgJY_vHnZCZokK4o8_3}27S_3LIm*I>Z` z9JbLXZOr}A;!U2j@?{@CS-NTA6xyXNv>`-1Kq7xV;vcX}KAy;81+rv;eD!{C_)^X= zo&hR}E0>6_UM8zQEtyKBr$f7r2~bi;uuoh(kIM6OEWkv17mq}Hi*XAq~k2NBB2jdRG~jgmjRh*7>zTPV4rv$O)F zFvn5>K}%>sJ^uLB__LmBSPN{2pz2Ca?W&@3<@$EYMaRRraDZ=bw-$Cwo6Sc?9fOK1 z$Ha6D!S&qaw|Dxxmf%`|pFI?IMD@dUSiGi=u*c>u=qCevC@70-uyTA3g=j&Za30S$ zE@ri~G?^7%ovq>w^60tKal{Owk1lj<$}(P&(l>e2@y8WjhaJ;k_hZ*^m?sMR7y?CC zu~L(gla^e&drp1-em2j8bUE;s5Ll1#umo-<@H57KkpIg>-e7TsD7C_D*52opTqwz= z&)}t%-aKdlra-=?lYT=>qLw}aLe$HCS?46k!$F~ZSw~r{4;m>QKl;kfrBLTc{QNT2 zDcT_sktf<=74+Y9A$e)ZuK&Wj4`c>Wm?ja=n7Qz3S56FCExhsUItb@4U}q&M&oi4O zzRJNMyW&~XK0G-DgUt-K%B-Z;F2WwNiCIa%&LBdWmDC$zFJPZ^o{KRF*e9JT^0MS% z70KiDXQmT@Jn38wen>}Q%FgkZk{VC$>a35AiW9f)6HKoegyUqQBg9UlJwm2 zB%-))+H?T+5ZQ|H2C(N&6{liiJ$H^jK5fPUd#GHCaW&v)XC?J0Z4*~XQ7Ls+l2Z6O zQk;_5Ncpm!S}G;=Kgo7h(&Hq*24v4lQdtm_$*iQCr1@o5QmuYm`T#pCNi{rgDVMK| z&PpnVd${oB!KkuwofQ_8rqfwTW8sbgcFd%FKJ{KMeonx-8_y?CGG>|w|8fwHnUq@T ze~Qk%0FjKD)*)C4Z2POGa$Lih$+mx={%O-4*!JIo@dB{zui`2(v9|x{0ckS|*!I5+ zV=myAFSC+Vp64vB6e;9#@>_mZk}9hIPsKqgp$vW=CXsu9?O;_?Bc>_sVC@p;N~{?p zxsFTsQKFbLo?0 znc9Ui$3G(r0BDZtnq|`0tnda@n@N?(0sxH%rA=*M7XbJV#ucErXB1VL1pvaeD(r+~ z-S3$#VTAoJzlaEBi2oh>uYq+}N6`@1m3E_FadhO($SSL@MF~MCCIq!~8boZmLNW+lQyaUFE1HZ7)Abc)uhp zo<4hZtB{S}nSP)>}m z9-n_g+B5?4%O9{H1(F@~PV2Uq{au7|I(yNLVSyvUEs0{g{(q+yG)QalDWLA1DUXz4m z#5aj}&jitk@43yTO5co?vcP`i61vlE?DDtp=N9PK1N&tyjXdcR)MRQu<9BX5n`Go8 zS%1U-6PWy;`8Jn|a-5|TArX~aB%*`RTAr9T%|S_pSNZrCMqkkHs;FScjcykeRktLJ zT3LB0SufE^QLjXO1<3m5mhzKMnVpfJQI`AW*8S*L0h2G^+^YMoCZyKc(FQ5UQPi4s zpbrte4T@VteCI%4GdXnF&ggRe7Zzb3hDNSZ)2+;6@9SvCq-Y}xvkW^JZTkoEU2eNw zD(?3`bTTG0Gq)U!a#E~W%Z)A;iN)kgkN@AOB8{`$;j+@}Qgsfwm)^haFmBiY+xySK zm;s8PjKprKqT6KR9uA4e$e=~{duCbaM0KWKN`%t;A4LBLu)V+X5>8Pm%DD7ZXGvJ|Cge~|DUn*0I#C>{{HOUn|pI1KrkXLNC^Fc7$G1S6;MFbM2!U%CDksA-LLbtp;C5j zY?PFxyT5^i769A3Dl=wOQXTDG--G%NaNWJq;yDVwyEjKOJ&1sx7VMbJ)1=Tu<3d(BoVN2lrSVbtBvgL@UI zc-PgdyIN&m3mUgPm;K_GR+dcX(CVPz^68`|&t;FI`7p50WuG8y1opX1m115!%B+3d z+Ih%@7?9_({Gs_K0ix$JRl)@YDamtLQc{xVvNA~BfPF4I3SkJaOh91GdSlaw11roBV4~Ulm;$=xj&$8SrPH%j&Ib{_|k;`iGSh6E6E$OXXm0jkqDM~6Nj+rP9Di*ks*GrAL_%Yf~S zl)qyxUpu2W;l5rsBTaJn+8O-`^><)vNZy=o_w8OKt)cV{LE%XkQ(k#<+Hn}~cz}I_ zs!XG<5#6HM@bw1dr;p2dKb5iUe)rv zS`(K+$zSCqpgg|~bLDqa)6uiIJO6BE7aI!C_Z7t!-q~gI_F7PkQi~=*$3>XFFD=2S3lZDV2QVy(4HA z%*KK=-h0u$8`#FH%$QLrdkFfyq$-W~E%|L)DHLN&O%8V@9!x53T)$5 zCOcg_ZLd?S@$P&)-=P89c$FIFC~4!}7ixb{tMRH@o>$w!B_@qGbs71S#yb|hk-#?I zClMY8w(+Wggg4+bD~)%dD}isk;-l0yUX2NiN?wtaPm`3T@qSK1n}BUA%Eb3_otn|M z()t9xZUL_GDh)@>brgK#-4FIaU>mQp@^bmwc*nv$1=PwAzMJdV*9<9L3TKC-N@P;q zsq}FLmq+Ny()46FdL>EBm2Alaq%u=pQXM}WeF*LYpkj%$T5C9Jhb=SQV`X4eD%O;5 z3##(wlCm`V53%w-u#H}sF|U)|V}CBWl|~<)m~R3QHTuhaLMpj*Y@?qy2PJ9rT~O!@ zY@=6Z%&6pbbl7qj+-hKlEgGE`l3fedyvMc|x$)fJOvHC9g-qXGy|x4SY{J z-vHa>lo`v>(I%HZiC3V&T?48R_X?p)#a{yl!>$7M8c=>-E?=*KiEzh*S{X{@@HIm> zLA?RUpM3*&vu_|O=`B+x4yQ>=ls9lMqVoc<*MKtdJmR!f$G?I53GP-zz4p+$=6n(AHjF(1tlA^p$nac0s@&GqvvwNsp%W{BqF1F6}d1Yd)Es^;U&JZ1% zZQ%9EpH=~{{M&AeUbXO&M#onjC>Ke^ue@YR!oH4Cj_m&?uh}bmx-pwQVl(ovYuOTtmyc3Yv2py^ZK>vWuC*nUyzflDy|1~5|Fc9EwwW1Uc1f{G zUmmo+hu{@=Mdi?CHo>BG9Q`9WW-9k^_T?HhBH!!--i$4C%q^4{BbSSjnxOSh&=>cL zin_ryu3*uX_hlPdfPtxCT+=~qO?r4xzo@L0d1NHt<$!U=)Wms^4R*NwuVPa)ihC58 z7nIx-nC@50=1~g|GU>B}!i7XHA!-2_H?^7#rzS-QcOT8#5-=|~1f9`WrOkriY?B@z z6kXh@g%a6d`-Kbr ze9YTTw!g%}W?(njRvkV?^4gJ3ah*-JOJ$qzd7hM!GD+Ai!W*8=MXLRiS{*r51Oc2eZB*WQQVudcuGgF z8Jfk{pm}ZAZlZV_F{eQ}2~ivyB-^jN|*GmxmTA3s}B z*#N@s1lEnoH&24{<&D~V&6Fv-HZXY?1-+6RIUIh#&s}3#X$!(v30!_^zBvz+U;1xb zuT{@%LMyQ~HZZw{!{KxMj5&>IO^|#*j@-#l)#-F?p!$mIDD7(;DLEtGv;h4-Byc*y z$rAXAz_SQXN#GX(=1iuJ2&51Hxzwjw-}GBRnj;g97G6iKOdx6+ypbSzIf3v}ey%@@ ziw#r@xBonjyhP|Z(EsMFK)UFkjNt;a7;>^r{w~9=!Eh_0x{k{?JAg1vV862&aDno7 zFUcr1uuAz)1SR2S4$mNd47fM;2O$%F$M53tJg$O@58YX6h^VAy{&k}*<27l$kDvPFb*FGr3VPlB2Yas-y95*6A6UJ@iUpwBv5`#`;3XCHIe^$ z5Dzcr@FT=82H{Nvo;oMrJOPro5eVn<^A({lKw0Q;C;BI}UygFeQf)`rg&H$&$QMzu!3#{M&@Z`4^7t@6voL=NDE%}`Da=tS?h{KUe;1`~{H#RbUQxRI zt4vC4BVH$^;@yH|atV}h5kK#t@D8Z?AWMn8#C@rvvJ}rLe(qX|vK5x8IXYKSm7vs{>A>^plmx=i?aE+w2mI5)MSHSIJ0qnI-JWl27uJEb5<`$t9VY| zVUn87S=pLTGDR=&T&?8dIRla#VTAASQ-$>dZ3`euJCk(%blmi#DJRAy3PvyB zygy9TEXZepa20_YCo``H!nFjBI*+$5puBO@w%!7ZW5mKmvE9Pf9KM?P$sp`R;P&(L z&9xxhgTV1qnAiZ}E(9({xCoSw=Q`8OB&?}qX?(rGv?is>Qsd29gDNh2an= z@4#IT!r=ruUcf9QFlJUPX4;tsF4?1<+AE8hQVxkI8AehX%V!0SSM7L-vlB(hMP%=C zXa`~N01#e4;9`V}Bycu?dlBxGz!(BwB5VeOuP0FY&`xc=RX#zeYkzL6QTQd4!V4K* zfbatXgAoQvU@d_&5l)rB^8}V5+z$HxOkh32Yv8sZpvWDZ>E15u{bL;bmFS;9>4NJ^ z%>c(RUeb3aiGoD$@OCJ>PG^}O2v-mof^aw}J*ZnoImlI>M$ zVMHw0w%HqUFJRl|aD-tJkhXad!i6CEE7`GCTT|n#1YXkU#Kc76{+q~ADYRv1-7aRj z5Ll1!8Ynq+X{lF>dC8eK$zJ(wpS%$R)1dtV|0gjphd}WRh6|u{LXV7rFvmdt58Q(0 zK-m{=e-O?ma05?)ez@zPlF(Kk@G?a(&D-vCU>Jy}Y0=>BX+YQ;-pjqVVom!Tyt<|}d#zC)lZ zLRtc!5;zp$5MWBLUqiM;f`1`Tnc3pz1o~H6k=5uLdo?8_ch66T`$L(H&MXifO5lEkm7xD{ z0`DQb14^IyJJb4h$hFizzeD&6l)vHr3Bt_;+FgPo2;U=c0K#B7a1X;mElQD>G*9{O z1gY=|D5t|cRg_;RFdt!_1YRKU1j1vWVx(&`y|Y$7Q1p@s$F(yW7kEi4C2>xoWfA>) zat^fQMf{YGr}h1{FFX8BSX&1 zA-8TE8=U+GO7dg*d7Gb-a?~_82`{O68Qa|?+20SEDZTo&uAah!K%J`yaz%0}6!}XR zGF;k6CzP0AWW0Zz8Od zz|I8PUPgNc)zh%n{}_&}C$t(29!((KcYCQh!6j+TCQ0&M(#ux!M{)p$erj;gk>PQ6RCx`x1f3P zccRxJuM=$^3^YJEAc$ASFWwtb1O(2Nx2DDhaL#lMmtY1;$Q*%xHmxY7fo zQl!QWkv1Mm8$Shw(V`T!^%5Cm_VlMgrN&M#abKe054Vz@3^?baHV4=N=R*h&NI(Xh z8xTGOr83}D8-bU+H647$=2BW<2cMl0I!ZtWpNAkEBmo(GPDB_F?9E=yvLTPluA21& z(M50023- zB}G~14Na?VOR>Edw9=X^Zji$*{+x%uvBJw|yrDK@UWzOlQ``#PPBLhj%56}2p_e@G z!lF=;o&&3FfNAFmaW5%VK6E-1hB0M_WZ6vQ5Q{G?O4h)TzqAaTv=gpp3a_^>q*@KW zoit^{l$Mc_#`7|Vco`j?E_ssCkx9#_NC^$h8A4zrbPEJK5^8w`Zv#M5MnXv$0hw|c z`KU@GvshFZ?IdNila^7A#ub|N6358q2T4yxHc1)L*pbZR7=0L+av8~}$_|FN;HH?- zOHxKJRWfSPxS}ZT{Dw&xsZ_}bMWc4KQH)BGGAgluPXtf#!mbQ!t$Cl9z9`Ybx1*+J zJ%?KyoNTx3F7maUsPJZf{v{L z%}qaOO^?mXvtJ2KA^tp2C0_|Cr-eB~IPS}!TZz6I@HgtnWP5X+yCirIDG64kl9IWr zhzS?*^BUCEAiR`7ud7)50IG$$JgCfvbmfYv{f|PJ$MLH`{}ZV1XAm9+HS?OWXwP4C zD&<9WWm?QCef$oJcMGiDuV&sIlnyFs?^OaXsY5!M#JuG4V_JNaYPpOI-45#lj8ByWYCdj`Yf^7tCGB=wT-xx& z<>-71>t6VGgWH0CvUHT$|6dM%MD!b=#?=#{kcAz#6@KPsK-0+boH3H zh6f=K)ni?zt;a4zcLuf|l@aq+olzt8_??ta>hTcx2Z5*_JG%@RlaV4dDP3F1i`3(I z6vkO4sYj*8yyR1&9*+_wsmFyVTqjDl9#x5t%v_o}Eb{gEG~_3Nt;f$1J_Qv$9KBXO z4$;eK>la^-MnPcMTjR zK1xl^HaaBj5&|!28R;H+NcRxZ6PmK!NKbPZc!~bf{?T`Z`@%@gryB=$qEzXz9D38C zdC}U{Zg9JTioIRob+G^?<1cdQ?%G#iw58cNe&}9PpOHnoTK+K1R0H>6z*(%vSAJ-= zNp}bP{Ci-}C+ZxS6NG&!fp8i>O9|Zw>{kM+8TU3c&8C;fD-OLIEAo{<{MvjI1GSa# zDdVr&NDZYfGgYY@3SF(zQcP^~)Qsen_J2l{Kmg5an`*g^#M@_Lfs#fOWi1s_b~99&WcB^ z^$g-q17*@ml~d14J_kQG4-bcN4b-avf9d$xOzI8Iv?%p<=d2+0CPX~}?GaEVmzi2_ z6!G>>+DmIY(LaJ(87#CEe>s&cplN}s>QW)s0ytlovaO z+cQSF~~k`AnvDZ3CAc_B>gdQ!=m}-bDQi3YGNE$x4C_vt_FHFlY^~ zr!j(B0aLOIQH8z}$VgF@rAGzrOYXx$1+*Hpj|HW&G(eewmm&=w9LB~|c3Hp-xYLDy z@ILK5i>)PmYLQ0WWdWvgo-47e{1~##1>{<;BrEoM{_QSzJrB`-L>7je6h+&GD<`?i zmE-4Sj#r$YbzEvw7Lv@2KQAjd1_fOngt=mu$o5(BD1X7C4R2?!;CO-}EeDSWH=fu_ zw2nD}#UVE=;tK)rcrf<8UFGPUL#7s*KX2e=CKx_p^f;3~Fev%p)nc@l3f$fQc{vN0i&VRfNF`ztOie+ z0q4?ru_mDT3-s}RO~7VU-UW6|z}>g7(h1l#0jilYC6nu}2^e>4z8L}Rnt+Pi*jEGC zH36zsL_L_iT36C)lGiw4~;>e?*`ndDlnt)4}vQ!?}H33@@HcLR(1RQxgJYd%Z zEJs)j?3w^glUvdGq{*6q2OgnTWKBTD9ejNQGHU|%UzTt71J%N{YXT+_IuqD60UEGt z0@QGUIS#U{37GgGhGk8_RutX_c1^&a%ULH6?3w^oYG9T8nttwp`8cpE1C+<3n(%aGz^BA-0(ND9 za%e)raVrCocjuctz@J?ipsa*hFDzXd&<{!)*p&f$-NPI_$gB*wme4i8t_)BUY|(@X zzcSzr;-3I^Wx%?7^UX^jvofIb3SM#nyD~sc)H5n+w_j^zz#;Gt0(NDU#rG?+~pF*!&QcC@TYQhJO=?RtB_mwQWpBiqvGXe}{*;eljZqloIn~owOEmW@W%D zn12b_l>w?0<|x%(8SoyCH@*z90^Y5yZ=asYqSQ)TEijr9w zpxoqta0eX7C6!qj&=3221G_Rn^;o;)Yc5Kq*2;iU@J9fiA!KEMa_f01le4}PCAG}T z0Ob~Vt}lA_%7DwUKHH|?OxcwIJzS=^F-)O*U}eCCk8I8A-fyGu7E$gGC&RRI8~>zWo1C@F-j&Y15QEZaIqmP15SPz zSAxK<3{W*5u<97_D+AtyyB65@_w61*5g7ACAS(m zeFdqMGygvT?mocIx{N_MS(IhgMO6YXd1ak-xdQIxqAc_O4^$CQtHm|0(pI5JR#8;7f$hKw2ha)j(r*T;>MD z1Z?{!QIbj6)=x6c1MDR1V1xmnT4;7A_Ci8aK)KAss+k=kGdlbF81at)JNx}%>NZgikrW?xmi$ZGo;*Z#!!1$M@DG{Q&;$c*b12(v-O)vnY!<0=of{FTb$2DKXV zQWQyTv$(-7Y*>Sh)nZf@Hhhn;71)Ihs?5g#E(ulk3mbNN8b3S0E^OEfp#s>24XP5) zQPzbGCqX?y=<3lg*CoNsra2eD~fnC_3B)$`plb}Fy`XT!e)*o6&i5LQb-7B>8dunj~D8`NBbTywgxq1iJywE%WugYp~Z^4-FQ zzEJlPx=b4=y^)vnXYHJLPN*XV7@V`>ZI&=I=i{&W%=N!M%<2m&I znB!M+UJm(xO~0u|B3Fayx9Qw$vZmjZ%xjW54Jvnzzj;~o^uv>x+?kjvta+X@E_3|X zQi+!VJICMUSqA68&hbBraJK|xj=%47TwcJ=@y|iH091=LJICMtc?Lbe&hgtR;z2G+ zV^q?Y`!YqmVV5Q^aa4mSfhBNbZGuwEAV{Mt@r zj(o6z(|1rz2??0yF}3RU zB5C?revPlsR_jcmGgLT2sv|SYuY?9{6Apo4!6gvvu;2`u>Iu?Z>&)`2N+UMYMy1v) zzs41sl4-@Z z4V|6k*A~^(lw4rv5!%grm3;YG{vl*%FtD@yN^WM36LMvb52#Z+%l|u+UxA(FS6Xv( z1vEd)-{xiRXTZ+#E2o87B^)=)zdzCY0sibP|D~=YVA@L(tV(Trh0OU`{)tde2X>Z! zGs3%|TBvnq`7e2eH>1GL@;{7lC$O{pSNd*EC7pU=mjC886i;UP`@hQA8`xQXB?d+% zRq?a@BjKI|c&i|@{LlDAGLpLvmlv%ik*ryMWyG?y9QUJC@Ddr4S^mo~eyJp2XQyBB z_EpjwGPBeBzl@H|@;?gyVGzwu*Jaw->92_X2-w+aWdug0C?CeoPS2LC$n5mkwS2q{ zqS@)UeO5A3q$YF5&xn%DPG5<_T&pCr(@Kqb$*08Z^mb8_+3AN+ctDiw?6fMyy^_vV z_}S?XA-@mo?6i3eXAuM{K5|*CH9KwV@g;XF2#iWuv;0@m^lUx0$52}k)ni?zt;Yk1 z9t>TwYo91IQ~iJW##cBUiks^Hj_9v}o$3#KGwBD#PxW_tlZ6z(PW6vP7%c&r>c0!&4p4c7 z)2uVquk9!>Drvvqr}{rf{ZnA4`jy2#KF*-$yPB#0R=!CA+wCZc9enE0eJyo|TLvoX zxkr?`-HxgJ!%g|R$^If|s`h066zhj0UGPrYT<2a@;CY-{Vivj0sc&&=+Y)sQDIE#I z76h(AxDr$gZEzV!2JgvF`UHN|MxHtDB5afg&8NSKjWdY)5S6#ZMs}dQ6LZe0#7aL= zDhtZPN)GS-HedLHurGn}2xm&57lC^b?gS;nSUaHFDKFX9Dh;+c*!hb$NPY~oAK-s4 zI^r5&k$)?bk)n&Ycx%|~V$t~vT8nphO$O{T17!wT+D@mAdjPwNVX;plBW04Xs~C>L z^xCmrd#VsKR-sMMDQwB%---W46m}p`@jes9fWPwP70ge{&bF11?Bl&zcX&S-X;>b0kL2+B$?6B_L;Q z*AH0~0ZfT3L;2MiH(pXdMCt0F%~(n38-7NiFhZ0nk98f2T}yMO3&_p#I%wQPQ(-f| zmPT*XvSwyQ6m8ei?BIGYIld-yyrMMgxZICrElt7XON)(UA@&PD))mXCgIXC=D7k;@ z_ac*^E~O(&3dt1Q@>(%hFxi$CY3VDWx3o)ZdIarcY0Yq1E>j*fl;tu#idx{KqC6;; z0H5umE$uawZp~|(f)hC|Jy>{d?c>|l7Mn}t_*HuRQpsGw>l|M!$E8tOJJIp`I4%t; ze75%S%Q?PYj!R3o{CSe=o~tTMx?RwDfi=26BxyOZ0(wJlwDE}3PBQ5>L3459VhY8; zSqr+FR&b=7wQbF|I#bXV1JaYmhGrXR`Ax&UO}cqdd?a<#(z=mp9&|rnk}MPp`%YQR@f8-CY#fw|NmD4AGq3DrR`UM|kQ9xitU@vc zjb{~`myyD^_^}4gv4wq$)BdR(|F-t=2RZ(S9G6Qxi$9;^a@E8->2YZ$J?p>T01xo3 zqPh{4$A2p_efTd|mc7jE1!oGX(3i_2Hj5u?SM*z*Y|`(SNGWi{+`l+dLB5Y z6nTHK8l~|5kSGPQI`R8G&Gr0$5G2K9Qm(nZL zu;`Va(q?_^>@!UIsY4e`X=xU5)y)D+Vs;(a)BR)_*qD#_h!o5VRy1Irkxle!!RnV; zJhrN#_~ltg)LKyWJI0=ONsv%OO0cSP82+3#zY+Z-sFs8M8**gd4UAU7;1&c%{nfvV z*9^@eqIu7*=YNeiwa?yR0OS$gPT5VX@(Y@Q5q$$F*_uTOeTptYlejTR-uVG$|Ae*{{wtvLsM|Bi z$8&U&KPQvnNGN~8{ayHHckSR=d0x^ql7m?{dm)t0AF~h$lx-T^-m?;3QocA5{SL;F zP>%remlh8wb9}|Eprn%ZwB227C$!uXRnh^ZD1`??xgYXM5bi@@1Hy-* z%~IxdPInQ{H$A9u+clq&bnUESi;MoAC6PAjid#jE)kgvTwWc}VKVeYKGtpQ_UW%>S zc6ddX&u#ln*TzD#i4vvnZFT8J3e=jI`WqSHf^a7SWeD9sdA&XOe#GhE2S(2Kg0NNC zhQl?)A1eyY2uw$~08|U5e>aZYPv~CIzkN5;XzNDImO+K*1+0XaVJq-_+GBC9INNbMg|0D~R6}gl`ad9bp})`ZY^| zC5)spc=XO;-r}w%{wXe*g)#zvO`4JVX1!1jD-FkS^f4%pfSQ{LeBknGyrj!9I?ZkV z-A82mURYm1{}`Ad-#PxNe1w>wRPIh$@sB&cjM05`jLDF^g5=4KUbJU6J(L31bElGJ zwuMy8th}OMq+&)Bb?|3YAPCPSunA$k1f~#}|2bU^D4)}$t+&9!Td^RMeb;mNIpQA$ z;cWyC-^|ho5I!UdU$Afmgv$u*j?fd>B}bYW_O`;5pWr)*_@Sa86H;FyY?gpbNbUF~ z{*C~D>4sUQNti<4`E=5pEo~7g+6EwHkp1*}9ETBBQS`(Jzv)b}&mJkA))$;6WOj@GP-jR?F z6JjQN+%C6~rQVz6cDZ?7l}SHqZn+vMs-mB#;FlCL&{El*FtrJ6CpZ$}1YkQs)nTnQI=&OUk@!WT zAf4a~2+x4(k{w(p_%ES9f$anu6wkEV2jT|Xs9(g{8c=U!kt z!6sXorw6tZRHe98@||Em;(G(z3EqgX2-r?g6*y6(qMhJcqMs3pbb?ArnZlw>Cph#w zJl_G^3I6MwnM%59i%#&TPsp}(f)_!b0?d#dQ_)o?onU=RB^}=R+^X8Sv% z@s%I2o#3+wt0W+u;8zG=0NV+wDMndj#&?2^e_&Anu$|zs2ty?xo#2HC(*S?A6I25{ zWJ+o11Sft*X`~Ze25pJxNhkOT!bS;5C#bE)c7ll?`D77Pug8Gx1iKLG3@R#IRk==( zN3zDg1O5?pZG~WZ>MEteFkseSv%qPI{+5MOJKS8DkR3cMf5PcRae&!Ac zd=@KzaTi(iZjkP<4E{3uulvK2Dc#seqsdbz;uMh2qHDybevPYZd@1=_k7_Tyy8Xja zE^(r!8dcu%UMqAvAULwz1Fa@@!eN@;O3jl5vD(HGd=O3W{iz7e}JMjG#} zsid#`2OrtWb`59#!ps?9HdI{o&+rNvDaw1j1<7%ehByFz423&@^|z};+&Ft>Rp}V! zg&oZK+j$KH$ooYfuyw1W1!jfhea+C$Td?~OF}Fcl48mmuK1TQe^#7DV*I$`v0)xLN zkd}R7)e7TathApKZ?gI;%*fubQz1?Uc5hha<#~-Lb8Jh(cB+HEL`P1?Q}7=F_H?LD zB1cD0N9s2k2w*THd&Q~(FOTZf0g1G^l0>XuW{phc()?br%Bi2rA-_*yDrxtMP0L=f zN^W2Zn*_E%Cs7G?_J=JapIw3VYpog$y{@kg-5+)m+>?RzYrV_wJhTDp*IJbu{5;Rs6x!_FsL;9)pAUCll*FiuzQaR@FGN48I=$wE zi@uRoOU#q79tGht0;xY)jSTEgX{y7sxtyMFnl`;!n!4^41L z`;=BhT3v=oQnF3Yi;!Lrtxg2KMfe)nZF*EGW>oSeV*Av;C0W^~r}1C71p?7FJxb*o zl9bq{=bo=2$~HZ{Ae95VP0uk1M+4h}R4FevCGp2C+oxUxcZw*>Ha)K+tOd3Osd6Gm zSv{kt{^rU8)-!rHgsy-;d!7B_t41YXO1A0woy^D@Xx{c1YrH(D>3lO`u|o-8TLeeiA)S_W*drOa54j`mt#LHz>IX~lkFVU#l+_mV4} z@s98d|K&jz*ltw$dAWRbgx?cx4^S&Zi5$LWNa>QUZHbx$cU_k`b;8bVdgeBYEiG{< zDI5iCOH^XYt7Nyt3DSC{C0+=B8i-or0nU10RLUxPYNzi}l9sp}g*$+4iOP)S>eSkX zsDa=vQ7WfC$-$1QxNK`G)t7Y%NFjw*?Esu{+ObM_pZ$E^Az_vU!knje4 z0c8hQNz=Sz^LT)Xm~$M!qV_gBb`%#J=^nUIz7c;2NGP#B&sU z!+Q<(tH3rq<>%${wc%}t`!lGOp+pW}Go*AWoNaiT1kY=(YW8#mXKZU2YTJfa5~NHq zunkX%DX)^<@V=MEDh=-__(MR{@J{p%PbHU(Tyv?pKcFNH?|c*{1KaSF88a$*omvg= z78Di(+whbc<|t{ydlBjjpjN|EwLGtu-SF~%B!ANIzC-V8U>jbiSjw~mw&AIPgg4+b zD-Cbl|7v&|5g3)c9u;${BrFZ@1k$Mi_B<;ymZPKRc_Gy6fNOZF5cdl5dQSd)!+RR` zlfX7S<>%${wc&jZ_ft?SLx~)|W+)X>;XwWt1^U3!k_IMkLe8^K>0FFMbfkOq$}T*8 zYOERZu)F=wC^zOncU<=8c*Px;yZr=z+}`mQ-{J50vHZSfhz<&63(|SnRFXFpfjdD} zk;@c$Ib$k!bO-Gg9Ftu@W`W&1Y}05klVYJvTeXrYum|^(DX>XF+omS{Q2nZ1TAAyt zuY-r{-F_RHX>@|HJ2GEo*>?}J?60pI%w$y-``(DXEjSN)nVVy7656_B#!Y-4EQ#6a zz1#NRZ)&kAr(U~)--F_v3fN58ZmF?=16PAlLO;S8VWopi$uDVt(GD#6lARNNgz^=r zmI&)5VMIJ-jtACDf+~h){Y(_E?K)Z%#Y@6NP?m#giO6_KIKidFNahk&Ug0I-)C68k zyqAQ*c`4HyST70bM9MS+)=PqFrpy_a@`|C>&I!q6%KQq^dP&$q;9X$7B&b$BQ!6T4}pxAgga6xGapp1hirFFm{~7n#sljm;qdw?vk$Od5;_*7OdDXmBuqva z2dtL_O|!n4b{T2nCE=~#xlF}N!oP4n0U0j|{Tims-k@5z)=L7HwHXbpmjn%1F9~Y6 zz-)pnUJ|a|j$!eVa9E?1=?Sctgn5N2a}}^&5>%;yRq|dE-XZ=GkiBz4(;ZT#38)z5 z&Qh*_go!RDW=dw01^gqVekBXyA7L-i1lB)-^1_Ix{t-?f{y1R$BPb^xah!jInMBV3 z{8|49%3@ALQdIv4tD!6h)<43JMJcllWc(u(@0c7U#3HGm+e}oU#{L>${Ua!w*GzTvxVdL`PWS=-_kh&IKZ0`Wc}Y(VJ11N#so9*_ofDK> z;JLoOvi&18Ynn1U0sm+HBV6q=#mhn3q4-C*KKLjZb6i81ViPChts+f&CKjRfLy;?wqi|X;rye&Wm-)+BrdE1EbPF#}mH^{r@I= z;y2+J41O;u$j%A-wx&k_)^EaCgwYZZzX`V?+yXK?C#cm}u6=c%unvWn#DeUc(5fV5 zS_11nVGzPT5|Et}rXZXLtmlL!2)BaBbHaVDh>s>yWcq> z-iCe%*e^S~Aan-y%T84Zj7nZvzwA5&?m?m~Uv^GJ7%u_woUjOCA+TR|zUs4)k)rCB z?=pY*i`teiJD-97v>1>tJHJBs0z_YSs$R^fWM3~~#DCe@pfqLb1KT#s5xPr2=8%p? zI1)sE)^EbcE=Pe;N$rW>gsp~giug^Kg4TIrM*Jo$LAVvzofA|o<_$RWMw|=$`Y%P2 zofBS$zeWs*gMl9qz6aL9fEoyM40s0v&DzpW0PA3&FT!5HIv7ypc#g6<7#IolB%zCg z0j1}8Ni(Ys2IfG$3|I#PN>Aj_<*UJ~|DkO1)!+m0?*oy80aaq*hCka;D)J@ar=lbd z20ldLePA671npAhUkO+T0|eTG$iaXbsh4X`9Sj@*C5xNIvDsC^4GvR7*Kx0T)uNKP}DwU8UyQKK)TK8p8%+8gMf-8n(k8hbTuwSMP>>`yIB%1xBSPl37i>DpUnLA=KVEL20oZ1(`0)**QV!@f`YpcvVnxUJm(x zeeSFpiChgncYa^8pK)rgWL`Pexr%bVDwOqz9#M^1Ap-`yD%_Vxza?H3o~2G7kV=-F z6He*KAPZQp3V$GcD*^GUaCIlDA6TynpCP;rs#jsndR4fhGtCENc1{@TlPsA-lK2>S zDI`hpG0-SY_X{#U2KFa35ZF7>h;q%WO7+}X^gae=z?}~4o#-xv<)SP;22>^Rl2_I{ z(d%&6iL%^@enZ$U0l5>k?m}w?_D-ahV&0OrCAp7Gmcq!LXaM|ufV~rqK{y#icOtbA z_ZBoS{!VlyO5U0dKA(lVN|a^Wt}O^(N`&Qv@xqq`{IfeJXfoneVRRH5eUTaMoUqJQmG!Bx z(gnn)!hFY_pgtAO$EU(Wj$nN%H0v2P_AH+YdX~bR^YtG-6_gy$A^)#VMm6$sHRxpa zmbWOGPDaVe|03u5RM2+>#$=?Zv58NG(MfE3p9;zda=5j9Dkv|O%d6#6L3v>=@4}+J zBx&zcL86|U^YN(B)4R8W;2 zOg~YnY!f`c#+nVl1q5zEqY+>cU6z_mUVPQ~~bNx=F7c-h-mNqcjAD%?;X9r3AfJ^TeA@&!mF(?#pJ`6M{eMPC?sO zU{rF6+Px53HRLRa!+|trI)N;A3d-ku%`0qYoO-wsY;mV>Fw81pdv^+x35^HVoq}q{ zbJ7%d3bQ3mai_2v{_~)=QsTdqx;&#w+fewvI_?zuqQ4g~CE`w@qZ^t9UedLdqCQl~9~Jg^@&`0IWNOfBc#12gSQnSWNsP zVBINvj<87r;!dGmZ>}LwS>@V59d`=ajsl~S-+o1RPB<3zp}@LRP*yCL?`q}(s8fOM zbd(-r?9k(CC{&0 zrpnpwEZZ;KH@JW-QW!^8+(LyRg|5X&zjjr0JQdz?mwTf#q{Kyw+VYHA#lWoN&Xj31 zUsf?ln~J_Mo?j`85LoDN8e2SETW|Y0bjRC*6_Bn7Bl|P`+JzH$>mpwZoAmGjqLil0 z9l(;9UAZv)_||2u%yFQshqF@Rq{ivpS{c41-fm1$ zaTiuSyfB?lcaLwp`sZTPuOemk1S?~!UuV`}d@D1J(7B-RtOJ{yH-56jCn~bI(>ZKRZTdTa*KWq1FwRpOCB|=V+T)|bqovW{#HP}2b3O6 z;5~0rCAS%fX|^L<(9}SA8}6GR`Ww~jc{~{T$~=UOP$z4elHUQu`~mSdP~N34+p;+t z{xl@N+v9tN2XnafUcBW4;XVWgBJ2y2eF&7l^m@iPPFqB4Vr)RTmBV8pog&0d1d{La z6S8-Io^CVv~4!R1>D*zKZgtmHWMvBx#$tpYn`7u%8Z`96|><6cV zN0g|{4>ZGeS7N?~_BpU+RvKR%OYVP-A7abgd+(I#1(H(cOj(u9%8B2z%31^SSYg*L ztFrTaW^DHOk!(+s{76|ZLi0ipl~t9J|EXjZ?ty%lDDXGxJ9@F*q7!lqW-}i==WT%{%QBY3=cC4o~d`3fS`mNi>>t^C_0CudW9IgxDxUt@c zM87K(ah)~9r&jm}6gSpuvM-Zyz>f8fag@OHgW|_}2NQoFuw%We5#~xj#(HletOJ!# zCZZ;R}eux)Q z!KPSXrnK(D-P3wTIkTBK*a&eo+*j_1MciNQop9)s`q7*CW6~3f#pnd0&H&N9R_@$u2t5mI3SBh?5k{TPps-n9 zRM$8-!d`kpyMdSt&nKp65GNCqZ|~pU(?Uxt?i}tKHeg89oA_QJ87E*&)sa~i^US8m zVmN}M!(bg#*IL3-?ByudE+l?h#+s>G!ocndpTT}&Zc^9q369K3j{_Xi~!|NjcVsD;NAcWwy*k} z!?TH>B?^BNSc!121hx`*2jMNiv#?aaMXogkUW%??I-Wb@@s*MrsgXa?_#Ko={i;sP zOI}AhE2*85!JK*E>bhdCOXe9q?~aCn(pT}(27Q$bZ@0Vaj~&LHq$bm6w(HoM8}p8^ z?8sILOz+j9*4+oKOnjjI;Yk@OvW?}WJZf008X06MM6|k(7_{aq{x+s(2S0MWxVelR z_oui6L1yH58KLPQcjRbqhI?fCh=1q=XO?tpLv72=@OzZL1omcl2=h{8zx<439d1ZVA|%VKbjd*}wK?czX$n$j$H@C|iKN85SML^#kn9us1?4U~h(M zATVOUHZZvv4kvz?D9FulCc+E}$j$IBgyq294Ey^GWTdEKxfwQ=63fl-O%z@S_GYL= zrl7sDI~RI0{1fi)KyQZCO00b|j2f8SsXldATIi+7A{Bq}0iz@vGE8q<$=i95HB47U zMvaso@`5xtZc9=dNVV-EoFkQ8hkyh?7{c)I&3R{Nh zTU;6V{FbcvVfrTGKL%AYOjiybMhx#Y+|WF)iqt{549%57R-yRec$)Z5Ao{bz@t<9S zF|!yZWk0elWtZXj!4Rv&pbW>A#t1(kO&yM(P5fEF4#yWFTn92kRMq0uooMNB{2543 z3$gYPRoMxjnw~!yqPCTM$q@BRG&h52h#F|CFc~RQ^TYAR2c=9SAb)l^p67IMl|Z&A z^SuI#v3pIz1$U_K}8+GSOCo>9qWj~~f) zBgv1H^&2#|fT*lWP5!5nRcLZ>$`k?lvm?arzJ65F;U4AwSkjRZVqaAD0(OL`MBZUJ zWmWMb#M9x9624q_J$(`x$!5U#5#lFsH-cIt#GziVknsrFdAdE>kr874At@6Fc7&)j zZhz4HU}+!X_Xc)^sGQK02*-^O&m;OAp~wia#;10OP}~UdA)@aCc7%9}qu|vGiXS0v zBmP@pM~LOsJS+n{LOdJc3{bh$Y1SSg<~ip@e5v3h{X~+&+JfGN@^WBDh)Rm((0!p_ zhr13`6gq>oM~I{5F#32W(}hrq^z_!lxQ&I^6SE!7pMj}4JL`TN%el#>$F|wIBlqJA z<$inzw|cz`hhB=#uQ7Qi1%nhTgUoD6t@~==4sM$S$TnocB7wE~b7DVLh;m zsnq~q+WR3>?`coNv9 z>1rxwP5Bi;Ul9MPD9DPS(xWg1?24fM5C#JN%4MItCtPXRCr|Qnecq0Ct@mkAP7xhh z>%A1=HVMccd2b-RCIMONUGM0W$p_V>}?m2q#NG_Q<;y z;c5xUBJXtwYk@UtS9)(J^fQQ7dcWYZyjyhAqCN6j4NaNmz^NbJe-)%p?nHV+4wA8+-qo2(Fv!+C>WM9^+4IxjPtb=d0tYwxJ0=bMmf~( z!2Z?T&Mp5af16NAol{W@Yr2uG2VotH&QMVDJg?xC81s@QBTJ&5mzsM6+C}g$l(b4_ zWhupTl~^WKv7u=+$e#chmh+T!6sQ8)(5 zPjG(_J64pBB`_aho&*jd@I1n^ zpmK!EkF2NWuk;xjr5r2GF3a zA(f+~&`a6~WKDZ9j6P6z2lmglHGQ8I7?oN$CE8j!8M1BdM0AD&+t!pA^D3H*w6(Hb zU0b^u{!NmWZEM^77%wA5YO*A6swhcYdj*9Tfo*Gf$EQpHY+KtGp)ZKqni|2&pfjwl zN_|_K4tXlDZB1D$Ky!TGHfd`&m$|m~9Q>z&ZEN8P%;-r#+S|f<6uK$;* z?aCLOiv2EI$JW{3WfyE-TwtWDk4@&sx<~Fhuods=EpUOmlh{!8zl6p|_U?uG&* zy?$&tKb9W1bfHY2*33N4f4KmYLNa3xhhBVsL%H5!oB6RCpA2Ya^7bFUpRw*s3x&RX zMh~-{|9^p`uuZ`d{C3ytj@}@Ctlp8?dIKjd*nki1o{hTfDIbrYarRjz-KXKkgIbz_ zlLr1VkR4xB-yCSY)Lb4DtBIornNzS-)+xH#^^N}VnfQzSX_J!DuhijB;JvVz(DPt$ zEBWcn&wF7ney^9{?);d_eVrYt=87SmBWr#!%q4d^^X9HVj2(9{FL}gR?~^#KpuZSh zim({eY^Q@6jJEX1VL!+dsB((I)BwEm==5_}8mGt7SiN`|-0sKmUosuWL*H$G2vPK8!~Bohyy zw0Y&UCNGAKRZ?go+_2~hg4x%MF+UZsCAPH06*Q>xi0dz{@h6Ud52~8; z+UrbbjzJ|HE`zWWM|T;;*N(tGzn_V4ssv<)=U#+6LFL;{F?xQlxraojxI^q5?j-HZ z(|fmkQnL00tgUEo7IWEws8P@J22|1|r|G@B-`@)Z(x{Id%`z%r8};>GB_l=kAq!^? z6CG*PGvJ>OY@=3YU{vzwNfypzwL06Vl@#XENy0YjUogG3uK9Q_Kc`VkwJr8V$VgF! z(t8j3SxO)^HR5DmErVJCGirX80bDU3!a3agh|asCbQ>{mqp((#vIC}S3Q_mERC(?c zrM@NMY>o~*g|}Lu{5Be^Qh1((Qv6|<3h(3agT&u0oM*XfZt`Xp!`b7(Snu#v4!0e{ zq61L!5BEdm(8hdGqXO2e7bMDili7k6ZOvr(<3Z`uw0UJRfmuhVxCL9pyaQz&+_fON zmOxpr12dVy|63g;aZ*zS<5&2~lN&cjZXg!uP0oz3=Gnu1f z=fdQ&-4E_S;4a^iohYNKFZmPL58kTMC?9)X}l?gHnoX;a>AUO2LO1b=2g9t|F=2QWWNJ z>y{!tGpntzNRjp==8o!%xo3zGclpK^*M@2c-i)Y6gyuAmiBcz^=<)`$c zvP`A!n5)n>*fIGNl;qd^)MPipCb{H`ce+!IOFPVMe>>&y#gBrtKiBTcg8ed0yjc#r z_~MO9n z$3QE4X;7wiN^-dJ$<4bkSnY8@=V;*cP~_`A6|P^VBdLep^qEdC4mt8MH~xgA(4%&$}qRDN5t!m3nIEB~=plG06*{g!B0+ zI*n;X5Y8e{fv`KM_|@fAoyXukiNEqboTI=M)0k;;4!b4VBxlI^JBOc%=#B@;F$9t| z{FLAJZyT?|i+NP`>JhXL-{bHimTKCsVc|P5SWJxyQA1{!B13<{z#KKJ_(c;Wu=yg z^ydXThy6HwGx0Zxeh&i44*W#=RYe!eFkTY)J(^yV&JiE(LNb@_Vj zL?~r)ooQ^`OPXVzt}>UH(ipGSF_*c-P)wgdZ>x2|EGy?h`sJhhaYva%RFgB9Km*A$ z2$a9RsMM2Lkpj8+mAqK^6^HkNR3XHT1ga%!@Xts!(iy7#&`Unc6L$KRTRw_C7x5C0_vEzAPm-$$AISX_Y*iMULl^dF!%S zvi>o&)R|;k@HK(tc79B`G&ju)zMaU6{SZ=`Sh+MYjl*e^#L=wkG^uVF>q-6M0b=%~eA%p1Y? z^P8I{LC`=#^}5bsMy(&zThCw}9yAzvO#sa!`j5%vU7c;K=RXt3T;-IjG z+G%$dw>&WJpM$%aCT)XeH%uuu$8g|aFzO^2^MBChU-rH+O^(Z-QB9*BPs~Or8-ymi zLMe@9tP!pAKxnclRD2xXcYxg#N@+1m(@mi|6WtM{Ws{`VKSl}0%wjCb#z?ykz?5u` zbP$|MU^hoP2Vnw;wvtlAq0fnCTP$oB3$m5eO(@&|>{e2$6VK7{TS+|+_gUe~R#M8( z^OC0Fx02cd_e)^6l2U#mhc8nTvXxZBvs0!3$X}^!C8dRD-zHZLnuOgUO16~h0e`nh zXVkSzSS0C)JS#;tf9oI$yM&mbu#N&%tFvf%mL@g-7Dq22dMdCrznic5vJaw~UpCOy z{2g$X0$cNMAgl+b7X&t^){ZTO9gZd7TKU?)$cw;hBR1jbF zW@5_x6X}dU~7JWuld3c zqk6kh&Z2ZT=fIf&Y|Y<-uo#$<^Ro;md`_YaM;1m%W-f#FG74)z>0Mbm$s8S9^N&IK z5$-nO-%GV>8QE*PPEFdHe-cWIbNE6Al(nQ@mBliyI((`5&M@|Y+7HN|t$8iHF&QZ; zh_Crm;E#@UMxA(T7f+0ve$*F@euo1o?3u(|1?vh>byXHE&(fs3ST)1|w&u0)#$=>O z$=Cc_@ZX4ZM%{i3ixOR4;$}ru^Rp`{?1RMo1?vw`^-dNo&(fskKj&z>b156JH9yqX z{5DtfSIJ3~HeU(n0AOo=0>atAl;lxGntsAdQHGw}x8V5%k)j`x6jzH&5*0YouwFI%4lMg}cB@317I4B{nXAV>$1yG65=aTy zwu*m={~QjV7yDwoekXGskmJD)LP}1iT^+tg>~nZ+`7t0S0-b>9&$PU&yhw#FBE>EL zAo%+u&d|e?YQc{1mCi4Bo(@M*@#8?Ph4D7h^x`B6o5EM3^m~;dim5<;1N)2OcO_Mv zP2nr!dx7^bxuWqb_eYARCsjO5$G7r|{!+`B3# z?2GZ7?Edmpz+;WS4|N|Kr>FO8$kbTEkiK#xSMMY&9BrV(O`x)}6Dh5@PgoL5@Q zm)2StEYcfad#~P$M^l=+NV*^3ok(y?a$qWf?1|ilkYGH4j(A!l#Y1@?V3$?LM{7T3 zwfEy0$A_q^w3)>`(03rw9wCk*G1;|^(tBuW^v#Af5MD&0Mb>5;j8p%Nke)bSBTH50 z^~%(3FusTNEmAht;_b5HVBCkm2mN4+e8Wlm$pdg0KJ-fT8#yIXrf;3;FP{Eid%qjB zE}p=P1KzGEop(#9F0EI$#j)f%hoqX5SRIH2GYA}jXMd!4R`&`LD|@w#x@7JD@-dJe zAn6n^LlN_Cg;UN7eZ}dE)A1|Np}qob-l=e?Ou0)@y!ssKtIy_@XO|Yn1yy6wpL?(3 zYBHOGm{*>&;@UVLFGQGM{Eo#Uun!^TR}}W!SE^eQd;42otti#h?6ZNK{no&OMmX=n zd#E z^a0un!N}XuP)<8z{GUYHm!Um4h@`Qg&PL2mTn-0Oads!J*(A?0%9<9ftx(2Q7gs&X zpJMk2I|5A6y(>NwG%+N0Tm#HCe@QE?r9svi#I4Du^pz`e#HYse{RQ5&l< zh{8WZ(gsSk5=jlbK+4vuxLYBeCakL{M@^$4|92+O?jbsysm?Vxez(WY~zDNftA;S_co zM#>Jhy5eh{cg~|15nPHV&gV9HQcXt7xo8-zTKZJhvcnYi>IX2fxUhMjMmNr)<0!-& zl3swi3Nc3DaLo9F%YWFu0{lJc-)a{9Q?rO(e=a_Ob9jPH5$Ucubnp)S-~>+CSdxN~ zI5nqohe3)LvueVDGvqG7{e}cN=Ku~~C#e)nS0wzFK)4=ffMR-V!#AAxQ;;5i5V!%) zHOLX;FJ`-Tn|mg205cH0e@<{au`iSS5)v*ZFhEm!Jb`ZqQCpGXx0w-lOkU3-z2E6x z?(E|!%10y>U(MP(WPqmh_=*_E>~!RvO=Bi+lPU8guXS)d(MQARj|8U^xD?ODNT-JF zD+t+!V7~lNO!Qr(->HOd1UBG#SpyvjoOccHv>;_eZT0LFp4VOUSG1bSxFo)a8J_`@ zD*Z-z*?YAv?*N;PnpsGRKFv8~5c?tqwYVTO4|6MkE=RaGOr_qAkSQ~^06K$7zXyr_ z9K;qRe1$-;6sO9yEY(Lkeam#QWAWiwSms^EYP3mv(%T_n3j#qt&IyE$L!w{P{dl-= zr-dC4D_09^3Ce&>gfRhWx`X;Tlpp)7HtF|=-#G|*0C^DX{Ydy1cz(q{6&1^MvG%oF zFx!q^pH8Q==6fRqdi+WHM-bjd9`bUUu%R`baF058UcN3zmLNTv5LkLWFB~IggOVF# zg%v){O~gO=@L_{F9r}Dwegl7N5&P!j6ZEDIPN(vj0WSd@zC=nVeO7VFeA$KM{8DdV zrRS7$QFji)YI=Qp@d;ZF?rNW-u21U&hLE>*ws!!!4KbbW({XW~&%KA>I^B&xKR`^U zyE@kDeC~a;xYPX$?4O7$$Qv=GL<`cl-BYe0`%dHaPsF@Q>0mrR0CSn2PVz9sWbPnY z;(Vdz!Y|JJPOx(kej{5nEtYL@=4&yL4@}fGD^kg~f_Xy{S^EHSfOvYa+ zQ+u03V0A~z?zQ;1x7h-G(80FI2k>N>K@I6hbAsUyM!lkx9brlF!)3kr;nLXsQE@9d zpDhC|mN-FMOzr+m@*fqKeq`hI4C7cLC*Cw~VxOTj(xIdTU2tv#Jxv*{2n-r(!9(f# zAB_s>pRW(WIl%?4sDqBTMb$q06${edf4PmIMAf)Vci->EmMcc>Jz$Njfgay^tG=9|J0nJ%!uox{5 ztERE3noJD_xm%=K@XJt)1t0FrVy^dqy%}tY()ou+`>kZI)yqQ(rd_9#cgUk7=v*|=V*S0c94S!v71`iCz$bHBHK1l6X~ zc?8fyi0O1*jH}^%l{=lSK)*-KegDl^tMdi6@bP`W#!Oy|L0m!Jk12UyQII|DS`W7c z=?bb7VmciM3uG)X*XdkL@&$;=+(EL$`O-R_$G|>{@GI6%$KS19?J$;T=HeBNCHIP6 z$EEXbil*CM=Y2G6Fj|VW*EuRyQuH;T>R#L1PNfRk?Rc};jRBjl1Zf6(hok-N}I0~7&WXy@ zYqmSAgORdEsn={UPVZlXhS*&7qP+}GrEce5{U^jPzQ!uL9kXZaS#dKtUxYVJmU!0{ ze-s565f{H$`yGd;--#MHG556Oj%OzD>y=_Up37`L+VQjmr(Hr?$Mc9Stad!zEl4|_ zPi(w)JjdI3)A4*~6UJ14%l@`;|E=RG(vIhSlp*>H>*-U~W=DzKa ze#~~9_`W>^>RTJ)-q@5IwyDA%wEV7` zUQX2%bv^E;saD(rim~D=D!QJ>?fqy-lvdI8^tKJ6UC;S8QM;b5NnOv)PT^f~*Yj;s zvdL?dOX40hrgtlO((0w_!Mf{o_71xJ0%|gjq`+6 zH8#vgVOI1+4jZIl|Dh*ZOg;}I)L;Mpyk|?u>&M@4FL=dkeZVM6rT>3E0JR=5UD2kv zCeBy6E2=Y>Rfve`4ZeugI$v5>R0_5$;!5&kOey-7lDy`%TjokK4A@DC>5CkWwXWb? zUvxLg^AMA}gJg;GrS(OdzuiOKREi641w8=OfaY9Ew2FGu#0kLiP^5txFQK4_99@p1BypTl4sf|Moxu@Ab$7Fn;} zzu{EsgT~k^{(tpB*Tv1`d{N;|lO>MTCey3oEDCUDTzu09IXurFQ3EH6Mkn_{w*bFU zDW(saWAo8IXc1cfu@9Y&^FjQe41?;v3m~8DjX} zsA$_j^KDPk`+znmt0jRwcz#EU9!etlG2IouNlCRvuV@)o0NdwoZubcLJ|>Yeq`&4= zQmy^PV*p-GQXfFQkl=g*WAKbd%!j>Q@iNnM($7mT;X~VVz}}+dX9#S^^8;c&@a@Q1 zX=HtOYI?@LQhX2NtVl40z)^UPK=hOhMOO#gG_14N0(Y$D->(Hei==VDE=0l$2}sc$ zNoe*nahA3zAPqYQl77K`D}?V6bD|y6Pv`IjcXOg^-~yRRJnShU4SwbeNJOKdS>jaZ5u`_xmnArUvwV@f4J({LEsrY z%Mf!?or0`bf#pjR{i6Fy+X@G5V_B;=Fc|l$=6n_3YIiAK?2C!y++9zJe$ssl8GVG9 zFQ_|?pT^PG;WH^)qXq1)gfNlg@2=ty&MVE z?6mS~ZjQ$KW4*6%q7OcgGO39@2zCKtOzc}cUn0iD94RZN#|2aq3vK5Zo0$GjGA8Cy zJaN95NY3|a&T3*!?`Pr|F=x~v{WOj{qo)8JidYkKgp8Qr+gxH3y9V@B#F&`FXQtzw zi7f&92$E8utTenU(1$?ZNAzd9=C@-jaK89&9bNN8H8<^=Gag|13B0n+79aQW+BKW! z3-{pRiSxN9J$=5|qnh+T?y+b(1~JbU4(X?H;^&L;U@t?;PPFGP&e{CO9pwIl=%Zl) z|DepH;k1fBcCXo|`%Y0>G$x>9il^I$4EkGzW_q-Atc`aWuTFJjJ%L;5kt zofn-}uBRM5FYkhXLtCe>)f~5u7Rt1-OYy|{s@UA)3F_t#k}Fa+u(B32L0zO1)Gmn) z8!V`LMJ4KO(gNvZ^<}8J5HVRhq@Tu#C$86meONt?#QmfjZ?@3OU=K3rdus}s$siMd ztk+e__OCpAJR=@HF4h-tx^It);K_{F_zTY)HV;A$^i`ZD3we_ODeq}9OUOU^&(`0@ z__9h_RlRnb#=+A6Yy)5%j+p;!8}Yo0nE&&~T9QbGFUnm1zvZg}{Rh+SAztD~>>n^4 zEp~W6_PPJ_${Yl<(&*!9@P{Lj9Zs^R#S`bN8RcvK`OVji)ef(Ra4k|cJ@qv5W)bb{ z%AnWiN+{+U-EyBBrvJEdJ68-L7T-^;#>AD+dz}`m=9T1KY%BaYc`iYMC&6vR^By8* zG1VReFgAZ9U@!?8$3adt7)9FlJ{ z%FJRa2a;HAoO`9kR1TRd-Bn%6Vk(zbHC;>6VyX`NHILJ4$XK+fjjbF{;k%Rc8l`$3 zDL#^yH=V|svO{yr^-H#04(>#f1}vsIkm4n*oN~B2GWJbyHT=TH;KEl(3ZKOppwT@x z;9C#n9mL!nUWkpu`QmD~^-DTmMpk-v$bE#JZjhwWK8Fv)`I4FQU{;LwIYdS}Hhr-} z3Hh`|SnKnQE!|4J1z%vVbse5~0rk;eG8r__-ZykqWRs0IqaqL5gzXjJvXwT@jli4# zbpLbeSBiIFOs@1jb1cs5U(hgUkR3F{8qwojBWbSaB)VC=jk87X0 z%sW2D2p2M_ZBpj1*~}+kDuYuq560a4rSP^gD(mEV9cE3ULfS)ykF)nLVtUB7c-kPQhjb)%n0|-aapUYf5t8<(UUqVa|`6Hem5Yt1RZMCrDsn&uXawBb0(L>gGf*A_L z^pFl5i1S6|foPW-9|W`vu|1@NWy?gXy}6zy?IBMGbt+RB^V4oJR8XMw#g$ z9Z1r1uhc_2WUeexbtyfhORJi$C8>vefwycOvW9Fyi`v+Iw^R7fNm@;*o0&qFS&JN@UF;)P6}#{#!FH>)pXTuu zG2N72!rO*CWw=&jDGQZGmqoP+zln2hbac zdEKJ=vwTq%F}Ep43gs>(*?Zm*Bp;5TS-bhQw%oj~Z8dL0bEsW?bZt5Si*YT~wwWUHhv+_aW&Kkc*I^%Ngo(IHn+CMwEwNSo2BE z;5ASifxm~8v}0V-0r}!+C5padbr6XBc{r(rR6U0-Z~n0RD0$noD!g49R=X0NcN$ zK%Gmw7tq~^nM->W&&!Az<94LX^qkyW+Am>RXhtI zW{Ra0PgjJWnM<2sn!}+kmrk;5M$<;bOtLtnA9KukgQs&7Gqo-{$r8N4+FK-=WLX|( z)yS&sM}3n^TY#D~JoP6EJVu_OKH$ z{2v3pP6lTS!BnQdL}XInY!Odz10#QgYzh2 zEYWe;Yu`3!q%g9?#?#=aC2j=&f#zjqo8OBq(fMpX#uDFAl3HTs%hU&wu|$XT#ra~6 zv&13bk3)A^J0x=(5O; zDan@j`c2eZE%ALg8xUiOj>8P>KUm^IwO+Nvzrg>gc^OOmF}6hKGljs>Gkd}eNNR~K zU(1nZh_OV6^u_sNjgzCJ5nYEYZ<3 zV|ubBuG8|WC3b<*2{D#96wir>u|%hURe`qZapBbx|NLK;=n_0}zL-Zt?XPD z5d9fTY`oX_#S`Z%j3Ufa!@ygqw_4&YaBf13B|4-Zb0US2QFXi-j#}b!@Xu&o#u8h` z89ATL$5`SQN>WSw0>WpAu|$XT#ra~6v&7st8D~d~C3eKq9x;~aC_x(8S>h0&Cn#4f z@hUu%5PrrI9X&Iq*S4AKm2=`PlvgeBQ78*lhxX2Y;rSCWmgp2@#R}rWt0f+2+as-a zb_slr*+OC-8L!ri)e`r8iwWTX*dOS~5J)rhe~htEvM zJ4<{F?4w9ZfwI!@u0Rg1l{1#;vdE4p4Q$2rukLEpC0pVO zwO+ME?`=9-IFTjxj4jdmY(B;k$IOJJme>kX3&dEWL;B);F~?cr0I)|P#u8n?SfZl@ zX=G=Km%tmRT(v|O(1y|8+0ip&Urn32TH@DQUbV!#;m=ncYKgDmc?B_+=oDnd3gW`6 zB_3DF631)C`kd$zyx14>$he=L5w*k}$v*ZQo&p8?lz?2pMUF*b)bU z9*7uAbok74ytBj!U@u2f3Y3+GcLj2At(;j3=CWWAn16<9eck6!SGB$eQM&*!*5`0M zmHmVDmE8tMt?z5_Um}t9of2E0^CdUxW+kch<*lV-M2z)0ByTcVcFe*0Dwcxvhft0f zt8-9hc`A{d)lCO_1Cp{7%z@dVO#c>hqWjFIOlo0IgMAV)7WNySpAlnWj>O9J!aBUm1Seuns6+Z`9Ct#;0zDhC7Ul>UF~Pm2 z5L?)6(6bO@VGf^}j&~Ne3hYWGr9fF}cvqlrfqsqXࢧq^o0xxdHJyV zI<{*<$|hU98*&}jm7(Qg`z)j5(04)M-#3E+gWR9@$=jOC8&Pz$5Z5ebXFrebZe`7& z+IFZ}2dDpvpt{V;j7~2}-U4Lo1E%Z@{<`wo(dp>DHB+Ze@tCReRk0+k(@-4>)jE|? zpUzsRFD>5IDHyj3xLL~NYRh-QZN1$?plUK&W}x9Z)iOP`Wg;z7d@s`+E(6}b#FCi~ zsdw%j6yj-;zJdBF(y0#97Y@go3%%Sw$Y>lC6Mf8jzIlZV(AXXwiFt<5GNea$0$cEW zgcMI(!~CGH14HY-9b{L4Qy+Ily9?`2`tQ!@;x>*kWiU>Q$1eQHp-nM znxt4)gyrqDZ7JQ!K&N)rYX-UsY%0pHM9j>k!|}ZlEr{E)_)?O)pBEHEyfzm+P{T1$rI;u1=92I zo)$>Ydf)e%JVQ*@4(X?HVhflD_BN#Ksno1}y_H1G=TbazKKB5oInTM%Hs4mXe27vA5^T|#30)HGWOH@;LtLZIPW^mN)0z(B!V$bM==-z<-BRB?rtbr?JNAv3P zJjXRis31waGcqB%HK6}7j)8YDaD{aKsl9CDsM|J%3X;U=>Iu>P8T|x~fmh=Tcao=& zormr(g+%L?jiG`haeJGD==O|$8IFN}U8Zn_H2l4nlylTCO@s=P(E^-LMx08(zX-=L zUW8GxC4HX4i;g*5uI8w_o|?~4NaF7y3DK=i{rhkX@1pLSlR{R!Qca%XsJowr3X;UU z{t3}-PyG!z2Hx4g6*A}OedQ~Tx~FNVAW3x1@kRG8^@DpX{M&A-T_MZP7O73NZf6=Q zND@Idf7A*39US~_16N3!THWPPj=FDXs31u+I3yvuL8w0r#}sDB!Kz&$8$NF#=W*0M zLqi2g;=F?rqWgpTGjI&NtAQ)zxXFn#Akav(fxG&LL39X!N3)= z`pQ<)mZNU58!AW=i+GsQ?R5Qe90TuQ3_>CEi})xVN8P_URFEWo6~6z7&%X@Ez(*dS zWmm{4b85*A9CaV(P(hN&+MX-AfwTV*j)6BfaD|Ln*;bZw)ZLjw1xezAA3V`rnEj17 z2EL)IYF9|#h3(~gj=C{(s31w)^F2$~@cG#bEPTNJ3Rg(OcSPzFt-F4Q3X;U84Y{IQ zef#@6_>RsBSI7ygilq-n-TgaMkR*??SS0Up)Xm~U1xdnt zHdl0qcz+j;fiLW+WmibYOIt|BeSCug2^A!XQHyd#H;eb1;u!cS16Rn@oW9bLqwWMB zDo7Gnz2%AS`0fwDG4Q$uu8{GKYRYhqx(R%!AW0ngrYE}LyFU@fz#r|Pg;B`OoigMm zj{1p#P(hMtGA~#3y8`~BI0jy;y}}i;`1vlfilcsNAXJbfT3(wg`i%kq6C49ySE6u* zOu4(CZ0D#S5C|0{i5FLRe4-Y3RU8AKXy6LDuze5S)8R1>2^A!XanJG2AwK^=90Sj2 zr`i?Ls$Yp5!BIa<5h_R$myXUA{ThWo8pptk_fxn+_M6&6F6Ve95-La%e+p@L+@wnsb(KMNBt67t{2F_hb&B;8*XfA>>|$~PQ;L_$SN zV)$ZD^wSi6CeK_kPx2Jf>54W|m*|E_s2~|}>LRl0nUQEql;6WqN>Z~@e9ebj$aPJ&cOMi4AJsuHM3S}Hqco}wD^R7 zO`4$t=gQb9!;bZ|V4C3yu*;DFnqo5ilF;Xf$*@VBq4QPFFlQ5)sp$G+ekmEUVk641 z{92b`3D`Es08KF&o<-gz)Zod69%kOV#wQL5zxD)$zi%%yO%F|@=BvO8?#i7l^9=3J)zC|`V?t(v8b(AERqhD${W>hLiU#H`T z7S+(P3I6-4V`g$WmZX-W+y=1-^PBE;Y;9f}9koAV1shU6AX&#*sXB@)=~(A<{Mf1h z9Y@0Nt2){x>*$-T<5sUbS8YBgz+B(knziHlz6Q=@xj1?%xUd@xP?aMR~YgjN9F~ zeT_TNxI>LQ3b%UZPQH=Vi>7}i!+h%MlX@@l#Vh@DlM{dKN=9$GAXhR+kt4!`n#-*g$8PE%$s(vj zY`c${%s`3{FPwsc6L{|BPt;dn3FBI;>i|STY?8~$fVR?2&xE4sT5$6yH z-Iet9{Y_)27yDw;{@G#pGmu~}&PWKSBR#4-&{S@~bDe6~_*aDnKV3tQ?B?MYK!UwE zk3o1833d>OHPpMdLPLbvljq`GXkd{83^pla` z4=~H|EYrYF0=x0-M2hFpwvNrBL}j1j`G2Y%JV{cAE%ZxB@Gybn@EnUsrvZe1wt4cI zOl6(t6$d91JrTg=%D9lg^LUnPU=)Er@%(`dP<)RoiD~i~D`1g6HxekigEQuo@`_EY zKC*}K)a(W?E}>YJfMv&r=AW>-*c*vceAdYF-`g9C;5%)z-GspEq~-YF=0N-+!b;S{0SH-lw5_ zfr`z@6T#+J0%_GW=Xr4cP36CN=be7OwAyFCf&sbG=L=+u!1{AKg;2c9s42{oU)}{`$Y${0Sx7DVt8POUFlm4*b%33SEN1sVE)SCLsp zx`NyGU9Fit3yK#*E^ihivSMGH*=_rVzk&$9#~F_9Q#7;B2;6|@I;1E$voK9Tqt}9N z?VfPC71f?lksA{VmD*ClrvVOH0DB7MPa@K^>-Y*%?mv>kaX^BRI9p(Rf&@beRQZZ6 zr;(HfNu+Do)O=lB#G$+Viq^08PuN*cQk_}>XP^pBC-iqfQre+Nirk|OXvV1@GIewGtsucx0fvf}jrjoxU(BvCd zlq1EpH@Ar?Y^&CW$=!A}I_~C54jPek7ML@UvLQBuF{VE{*4M|MvNLStq0AQO!=2gf z(IiC+_+2_n#x2rTN?Nt4d!g1!7x2$P|BZ-Q!0&)T8pSQ(ulX&{zlg}o8G097LYYxF z=oyUaV#48BIE%c=HmS!HJcokc9~t>Q`CVhn!drw&ZCEVJD9LZ6yaHqm(zG#i?GEP0 zK3DiAhgHurbNJi9ZbjJI^*C;Fj=`&R{5eALbx))CGE(Y)$C7g-IGe!vct&gBbOP_; zc?T&z`w_l+V@t+o2T{=E#G3x^TC|%kHBw2Oba?rW;(~UVDKZ8F$mYtYqmN$-72aR)fB7J zEr`~|jk`*%s!iQx%gI%jowPwiOT;WYaWFq7L?)4bn(%*-7p;mh2%Z%T63E zQ+B9fvHIhAEt4)g=?b<4G0RTw#B)1hmYq0KR$5+W*-3#lH@EEMrhYVrE^PWTZVcy( zD@Ml6wAFRl$w%a~0WqsQ9Fk?nl^VmX@;GEWe^n6rZ{6d{!B}=&i4ZS4IRo@a#4I~; zD3%>p#=B)Fw}PFHq!cJC4etu{70@pb{h3#woMM(8SIPooYw(<{cC{UR(h*6?g(Xat zC@z=2K>Hx%aT(8z4wf7Hs+s~|(9H%kUqK$%f;tz%ImiOhq*O}*^}zaw(cJT&N0VOLC!u8s zVyqN5J)q^&>%7@dbbN@p&d@*677p$_?K7t$$ubd05 zUa6^EF4qFR8nM^8gXPAA*tnYw_(k*3>-=#Dk06W8bzVeAZm#WLi24aB)@$2oN=Sdw z^?JTY>3Thv{>=Y7B)Xn2ic41E(>g=qdVW|*dOcqb;Zj4=>)Ap5*cVX@slgT{>Gga+ zgat|}n`CoIxt<60u~+$uR)tcJ=S(wFV?H&f8(Q8}&iLgO7Si1$AM4-L-#pg5p7dCA zQ(q}R!0I&GcV?=PTde2YSK*jq)}fdU=0Cfi9X9VPk&kAzL_p zjf4u4#NBfn%Fp=xzi^D2K`j)nkj;~OOZ697GKYi;lEi!Dx(~r%Kb)joml9v5@jX-H zXA*ym#-EfLe+luUiSjRW@!I5)r?7f^_K~R^Z$Lr?ku5EKJ3jwO9FwiW)d(^(Gt~&Z zvA-Pp`s;m1{wU0F9_BoWgv+@#MdCoZ|UOIxFaLop3xvOQsu0!aiAQ)@h~J*Mp%F& zi3v>J9*xgG3&)gP73g`}mJyw|sJWGQ)Cb*UJV#w;8Y(FqqRt2$NgR|>D7wltxDUrD zYJ(z4G$j5Bjep(6>&$S2-P?;L|I3V~M>S;jFY#VbC?AsU_j$#VD?lqV>MN*0lQ%=j zUzyQj|JJqSIe(e&ckd!z{_`1!9b5Z}-ttye|1K!X!E+C z{QR?h`L=B>xqVbanV~A@Nb--yPL7eq837gzZ2%?&3)F_OuRvz zw74y8vdF4t%znL?k@Kjp<&!&#ZJ2Uil4IZ67h zR{s$kQ`w4>^jLHz8|dJ0{$m-PuZYjsV;SYzN|L9V%ZKL4^Bi@PicochQ`OOL!-jX^ zuy}+_bZZL#SDZK#Ju@u0{;9d%fs|abwTWcE+FEqih*0sNZf+sfzx3&TS`n!h3`6fc zlDL@I{fYDk;8ZBBfOwISEAhwahsbb_=OLkz!tUzy!(lm;S11$l`3rE2qCCA?&KZ8D zfdzyI&b!XBPIXU zpvE)Q(ZA`GV%;Al@1aP)C#Mu+xvPVEhrQ0lpxcnGMeW;&*&}5)o?VF9hOACZaz1^M z)1TWfrQt6*Qh=EKQXD7{=ZoQXzm!9P_CU;jDFv~1=gWwcSA2|a6xb1n{>&zE2U`xG zQPgU@81y^g4H{f`voX8a+z9x3B-$Kq2bey(H=Nlm$Z@m9`D}Kg8`!*{&7m9EyaD|+ z#B5-*1J5=@I_cW&Lv8MSn1K_VpX(I_`sLP!zvf6i{UZI~8v?Tg=(q|z-X9Az#*`$F(2sHYQRz}KSwaiPIeKx<8ib;CK zUJUlTg$6Pe8-#{|NVFEDNWUfG8p4l#ag)Fd-h`=NwYm(}C5ZXDmE#2I9Jd`(pdaYF z2f|&Noqni~)2G|M9cjzYZa3OE6tMyE>FB|Rgv!fa%S9w&KCpU2s%Ed z1=q3TcgX)0#Ecy~q#v^#$Bi9#{GI>j2!9=P#AlMtgpX>(zKD`fCUxH;6a6Q6B8=k^ z^T*2}*^adm$NeRm4Rn^t#59S6@c~6`5-!a3s#q7muLiXOIYMpE;TY>u+)wq5>HSUe z9)+msO^tIbYExl)2mO&F2O!4uuD1}jNdgj^-Z;|FM~vw$$FocWYI=L{{ECz>wM=Jv zW+neGagNSc$F`|fq#<`OQ<44m(5^^X@NkrDUxeWH#Ui9GR>`lrAbf-kWHDUb8$O*M z_cYxW=cZHmyX0D7xJimWwZT8Bqon);OVoczgTAlRmH2K?Ls<*f{|d+4{MT?>*1l_HP(SKG^o~8Lk1xM))b|QP#fKYlWW=T1 z=XJ5YuFdzGX(Ptmp zQ_kghIT9*9)Gdo8Q40M!eExi#3OfDt!oc>^O8Is}Z8nl)b15WLe0U3vRR8l2y0Lfh zYF%MO{q#~|KPA%t9j8KR1;qY|P~xPs>q?citolYmC57LrvkOP+XZPxt%YRW8!34S? zO~%Os$I5R!JQSXM@&(f3)I9aeh`$>b@;->~>HahRFdUQR!7M%V^uMj-DfgB!_2oQ{ zx-U(r$WZsLkr68zLVm?>tozUScj6fGh6=J0PP)0aJj(G|BvfRmd)G+yUi!a!Ena=F z+3j~Q{dS_qyoT}#k-;t;qi&d5!*I!8@9#T8p9&KD0`cAjl8a*?=Feavllb}?zmJPo zsw7tO=dnFe{;@7z@zDdzJAQ#Sdky^qpZ|{ELVI?}Qw25eZy+N$UW9~74@=eQjU$P# zZ*M4*@da~mOcAQvQZDq{);76L1M+}E9&?Zk6F-Ld=QaK{7q3)F96p1^NsN`^qIzQq_^{sPg-*QqS9mD*Fsw4V8Hq39S ze~W5dH0&(NALe&4rdRz9W$1U#BgPQ^L&S_Bez7t-zWhon$v@GLp0cX1n+-$v8H^b1 zqdaPiKK&BOZ|}D>g*JJ#_d7*-Jf@7#V@B&JkBHIQFIG^LM;n*N(aH!}y&C1wESb^F zZ)Yyk>ciBS^wW7Y%-@u6J@ZT`)vwW^m;o_)jTbe7c|>31?Png*&EK#!UTIXoh03@& zW<2dhjbIp0dnUUE(|6R8{HMJGmC-XQ?lWH0BvSotRq#@*U}3U?g^_}Y@{kt|236mt z3|+Y!)$q3DLf#e?QWA{|3PsoO2Hpm1at~gib~C$vp;RZzub&b>-B(=8lbgsv99ON6 zx`(FmL$Gc@9VqX1x{t|-dDr?+i(_BZtLW(Zc!w;0ADj-1-}omFfQT6wcR2RvunhM= zszbRCpw3eh-okjQ;}Rd&CTkJEWh^Nf{WwVJ~}s zB4%LRk%Bam8yK(m7q5XLZeZMecEi#5>;{$BiY(PK>0tPwV9O9Q7`_P4gNPXncciSe zyv(H6dA81W(kr;>`73A*eFmx)w}$g2w?=_ByFLSLBcsm|^9%hK>q!cJC4etunPpZh_i2ls3=Lf{~&YNsZ z?65zyuDYk|Sa4?}X0t?xV=SdfqA!8a1~C`CEl#BjM2qO@4(K}L*MsgSZJwWe5+(mC@!yGL$rX|mY zJ*DN-8^Bxe)*$8v&?TrM9T0N^a0;?w1#ty+HtZo=|MVMxOW`9}@yUp&+A>b}`3Is+vmRneMCipD!dF2NJ$OICEPQ#1o5(^b)1l@wi$qCq=tN2njq zxf_ZOF_JLXb3Np9Hu56#*Ndb|-0}{%q0l@Q;VFExpzHy>ZNg>o;aRQSX1FN0iF(9QE5Np@Jmw z`0ouxM;ZL7I5AxRuM~N;^y$a`8BxmKPg&K=@d3mPO@umFVLq`Ui5pO~l%W3(PKu(& zR#7zisN_|1nE&CZ-!2H16s}a~H5^HdM3H{1Aka68Vnq!)uPT%LEU!pMM|*ZOBYD|g zZSzDYiDq#0Eu%n3N@C7z?}9N5AO2Az`s0xSNbSx)GZ~chA5zJ~I~T}h948@Wa77Yd z(xNvK^zXxom3KA;+EFJOtyTCnv)jv499JWu63iy)yrD@ezC+O8h-3I~m@}fdo(I;J z?>Xu?W2hiWyhL{TRRg~!Jw}{eA%CH^6>K#`1y}2Kfb0XHEfOj`)bU73+`7A=9EdN_ zi^;IxS5D#r;&s3+&<&iE;s+A1-=z(-qe+VIOuTkk!E_gI3Z1A+{9KK{-^EAzf1#54 zW|sfDi&uQ1c~)Oq*PMVnrF1^LyKLt8HDX57LLE5deR3Y}f5zvBj60ZeoDDze5Y(3(URB#uMI)JOEORL6mop+p3tf^GWvn-w9bR(^IwP&1^BL)~1^w9XrGIX!X42mwgYHWzXg6T4 zpikKN!BJdNL0_{eTF~?2R2j6{#%n>ZtB5ai?Ur27(v=MgWf7pK5Odx>&_69V`lAaA z<$dDv7DVmTEIQEGRX=m1^S=WsIEIw&CHiclE_Tr! zHTp(>S!1Rqs9ty$xT2NCV}cWF9W zlXM{|C+nE{GK}L$#GoZlqt6&hD{!0;M~~QS2Y_Y}<dLeD{I_w8f^0a27fHtI1FIUgGEP6eOg`oKJrXKBd=W=}_WN8{ zSNOW?pM-*1U#1n|7QcllLh0i91yV%%Du>A%*ig<%eBE0Hc@r~#snt*p8D6uF32e=) zB8SX*yt@e;MQdERXna?xbABOD{pX#2fr(!5c(z>W%|zK--Wew5G=SY+ez%R>^lg!J zyRKh-d8}tG>9%Q~0^^aMhkWyVwn26~pkai}D#fG7w`*64km}*`+Du{JLid-+g-t1+ zlxEyjAUjFllZ=tU_3kW?W4--qR{e@5Ey6KPntM#cWP4W5W=8GXlfFL^Dw>J$k7pN1 zjz$dLg>Rs@+#5*1KNZKQIS7x}5M7dY2;LS~mdgbkuRuaYh9}|h3gU;b2cM7A%PR

Gk8XyDgJ|z#UhDZuZKBPh+k!VYyDAs*TPAT}7ymSSgJw*<{H$82(?~|=h$)&AN z$>RjiQtR%G5G&Z0BHXg%UrJk)9QV;70B%rmMQq?VnPo{=_bekW+bn)23u;qkA^(hh z;^5l?D73ri|5`DaAKqquPn z{dIG4UTJmHme(b+qN`UbFZpN0I(#UdtUSBi(kcg{G``W5Fsz5tWhRxE{PQ!jd4&XN z_SY7E_X$jNrK*w%UC2@y$v;&b=W{blWc}JQr*XENhLTzOaU59n+Ez|%VGA8&Sw$8+ z2e8L+V6kaUIdPXQ^s(S7S@0ad9>;;je-rD798Ot(oktkf)f_SX8`#<3C&IJj#&;_qjce1B;@q^dZ^7JAON zhb*|y?QtAfEQ&56Ufc3`LKfWT_BakKYK7Jm#WTyc(s>lXy6SU#90wM0%Tq*qw$Sxx zMi%MEabQvPY%cLBTj&_eM6%#Mx5sf{G4^x`agZ(a+;uxyaG%@bIIs}4aC^KhkNaf7 zeQuBAz@o{5TB2}-thvr3H`Z03+v7N}c<*vm(UL86JsOZj`f(gsEFM%tOl1ol#~Do) z+~@W<4lH(d$}WCj3q1$lNEY1Z_BakK=H+&Z$F@9fkp=g;J&ps5BM*~A{zzGKokv!z zt3J2KabOX=F}G;S7P=m_$s+wY4lEj+2oe+6LdRH!lLhy=J&ps5+$&3oZET^>CB7#M z?sIz_2NtD!;I%xqJg$)i_qjce1B+c<>xf)YvgSIE%ve`_Zja-@V)Ht@gOn|FJyOXc z{WuORjHR{2D7MfsmVsoUj_0;G4lH61)DaulLZ1^YCkyU#dmINAW?h%KY0KjhS#Y1r z7>oQfR>pB)@$EKKL}!sT*Lj3uUG=#=jsuIh?Ri8kw$SxRCX4jrII#HPi&Qa;Ep&{f zFIjM(+v7N}sPk_CzQ-g>rO)jalLhy=J&ps5u$IB%H(MURkOlX-J&ps5Px5CL;n8XH z$cS~-=k_=bESCM5LsVr8U61l)k$xNp7FQ0{5d+vl$5^_P1^2l ziY(HP>`mZyz?Kdza@?Xi$7QSiC%1>=M5dmBK-I>G3_Tb=F&@|!omEV9w%Il%+y5`d1KaqAF0sgFJ5%QS|HOD;yCi!_afNMl zjOT>3O&jC!DRV@v_K0ku4Hq8{6+ixoo4s?&Hmr*0l~hzk#($9dltUpRk<@BJr(P86 zzO}N&ek({ z9U*#;tkRDV!79tJ1Q`|ioUOE|TV7qs${r;8%)GT17d|j> zcCB%XV1e9{6%6r89t-8Ew5x2Fa~lQ7G1qe zM$I(NQyyg(|(uWDvOq)BUnb}56|8m~EoSHeBRF9b} z#60w2GHK?IH3&~qUS?h}-Q&IKxHJ=nYt51=NX$(iCQLIQ;Z3cQsYOh?^4@g9HFNl9 zy!?dn>P5_DA11SAc6^1e$`CV?m}V8c=|pH|ZCAYQo|sL zs~a)pD|*w(rkU<*@t#m(rV_Kkhsmy)(RP z#4l=JovHKrpvPR)yDr&=L#Mu20Gv<%VuVyO% z#_oTZsj#UOr;|jC-TyGN`v7iVr@Yz|WA{JI>}{jka2zpq|HI6Ld2yW13S#X3hna{V zTneJRP7q`FKg_%=tm^)l7`y*rrfcRDPA94$7`y*rrhO^AsefV|d zyZ>S4@_yx4M-gN9Kg{GFhd26AUQ3Cw`yXaLo9<@j2r+j5!_2`Jc&8Ml^MDw;|6wL| zygGh`yTRD~4>R|I)UmQGF?RpM%(tga&g)%b?EZ(DY%`TV96^lT|1gtd1RnRHGA|;= z?thp$_l5FN2Z^!!A7++UsLJWwCC2W5n0XIxyq9@73xTowA7%#7I~=GyC5f^7A7&1$ zQGT^4F?RpM%qe=>jM5oGjNSh*Q)sO6QC|>a_dm=GsiMZ}USjP2hn^b{k?W$$a9%fw zvHKrp@|IL%H7FL0-TyH2+<{j?QJLe2vHKrpvi_;YMnhuk{)d^750wueK#bl0FcUOR zjn(SPJu8QDd3F(F_dm=$!V`D0%-4yr`yXaT1SsF)C=AB#f0#)Lt;FdRCdTf6 znECFt8qf8JvHKrpD&H}iz{)d?(3Cc&!A;#{1m|4Arz6F84*hY-q|1dMDfI2Q; zA;#{1n8_EZj>~2dFn0gL%*`t5xLlAJyZ>S4X%^L|wTQ9%A7-{xQFEJK#Mu20Gb3iF za=m5}WA{JIoJm#39O|6wNEMwQoD zV(k8hnct$-{P`s@cK^dn)hBAMp34Kq?thpWTuqIi6k_cDhaMUfk%guuaG5(3WA{JI zOz)%0Gl>|x|6yiLj5Yru#_oTZnf*%v&g&F0cK^dni>&w{DUFS1#Mu20GlO5@^;yJZ zivwf#Kg`sbs*X)b#Mu20GZP!BV?ldj?EZ(DbDIirUgLH}iz{)d@Wg_I8u zFAm1;f0${2PiM+9mnFvTf0!w>Sj`XLCC2W5m^s)|9pgq2WA{JI4Bf8!bP+Lj|HI6< zxoWO{kQlrFVW!*XY9G8yjNSh*6ND!^WWAgvz}Wo{Gyj&c=0C*P{SPy(uc$t4N{rqA zFtcK)+6RXaWA{JIEUSz!x6#=6f*8C1Vdi96bsXGFjNSjxD=$Rk>+@=!aFZCj|6%6A z7R3aW1Y`F<%uN4MF>%D${SPyP@Q|}Cb3JMh6*GqzyZ>RP_zuNvBgXE3n7O!59V@R8WA{JIJjkKu8)j)RcK^f7 z;qfY+g2dSUk79Z$rWP@F|HI7ZjTO_27`y*rW_XA?e$6Dt?thrM^+fsJO~ly!4>RX3 zDu&)s^}qZNGf_v>vFSB2cK^dnojJ;{<}Cxp?thr+Qclgssu5%NKg{gTs+exX*!>SP z4XdkT+*D%h{)d@Ef0pC%vyK?M|6%6WZTKDx`Hi#0*!>SPx&Bgfzn8?={SPzGBGlY3 zS6MK2|HDkobXA@dV(k8hUZf-DMo6f}8{SPxgyVM+V5;1oF!_1g$DxKBD*!>SP z5BwE#iWs~9VW!~$6{mehjNSh*Q~Nu`WGe^8?thp$GgC21#Mu20Gj|6nradus|HI6k zHi{WXjNSh*bMaXe_vs2^?EZ(DZ>}ol1Tl91!%X}Aig`?o-TyF?sio?RsCY1T|HI7v z6vb2^#_oTZ`M8K;S`%aUKg_rz6f=q#yZ>RP@k=$|SW1lD|1i`0Kt^t#BgEMK4>N5m zRAlA>F?RpM%){ks-w7`d#_oTZ32CL~31x|~`yXZ+S5wTp#MtLQ%nT{1m=VO-{SPyp zb1G&LF?RpM%)dXYYb6JXvHKrp&fU($b-zoD-TyH2W6yldI4gj$`yZ7~Q^k}d#_oTZ zX-e;1C;!uw7`y*rX7L48uOYG_9Zf)c>k{SPzEW~y_CIAZMnhncxU6w{CxyZ>S4uU}OA3?Rntf0%hp zuPvl9&nL$2f0)VQR?IG9?EZ(D%-2-h<~lKU|HI7C(P}R0NCac|Kg=Y>s(q(0F?RpM z%&1Z-=2?#zyZ>RPY7WJGK#bl0Fw@3R%p79u{)d_Sf2erNHe&4lhnaeB)Y!N}jNSh* zGjX#@$E*m(?thpu8`R^zC`gRm|1h(@l$!I`BF652n8}<~F};Yf`yXZs9ak~fnZ(%r z4>NymSIj12?EZ(Dq@1cxFA`(-Kg`T8ujak4iLv`1W)AjOdF4$4WA{JI+^JlP+ou{a zcK^eSFM(cQ_NXn?EZ(DYzNfb?Ixs;Rjs|3dGjF`DNLk(Ov zF?MIfOttRn08^V7yE9_u{gEnI)|(i+Gh(LkOLd|?ix|5zVrJe2HBsJ7jNKVA6I4^B zbBP$cGh(JhmKxk&Z-}uwBW7;)SCim;mBHAZ5i>VltCN=M#Mqq?GZ%ZQ9kM$yc4x%Q zyLZ(|-85qC&WM?SvFgCNo*26`V&;e3DoAmT7`roKru$FfT=!SR*qsqG^V_PxT<#Pw zc4x%Q$#ZJLU4S1c(EnA!FxZG`~SE8L-v7mkD0Wx}``YhV;rQ0xlK^%vd%xS`@8< zPoG{)3iR8*3`Y{96!;{v*-tlC&}+;UksRnZu4D>`CQ#dI;-n;=j>7YmUk3Vh9|6Z` z41_w$BKUI*@*4^lzYYuuU0PPYW8E+of>>rzoD)-v0z-^b85Fe-#0gEw+#DTR;aU2d zfmsXAbMkj|ZUcBg&}Ax(BN(siym&V-=59i=*n_4r0#HvU7S_u?JCFA%iJ<7+OYn)c z`i=XE(9XEsGPLkK{4}MXQTV|}M5sj*jVg8V`7AV6>uNrYb-)3yW3)w@-6`Ec^d>2K zos_q+vi%HyYm9^yFE2`L<=lcnsn{i{hMgJDnX~RZ-C>Vm>Tf9+{hNr~Exq zxt%Hks6tRtDvfR@*QOHOPOYGI(7JA?YdfrVdicWDP90FN#CuWYL=3gl>KZ=n6m$kt zd}9rg+fGTYz=Q4vz0%IxPN)B)k7=EP`l}{3Ny2KUGp(H<{)YO-B5dunsCJZmf3aa? zL*@mr_~fM9Y4bnX6jcdCZB5DCbUWpJSAC_V6M*gn?WNM_b{bMSj@xN0w8>i6?KJR^ z)lO4i+1hC`il2CGrom3?Y)zkb`Y9*1(|IKKJ0+PT585d&e!T5eaUr$BOQ;!O$V4GY zSnYIv1i~jq4ybO6u(i|b$&n(Ck~UI6)U_ziNw?Dv5qT8#K8SvrlDX-2n)+=W9#2yM z%p#~Xl}5Ky)sGXoomN3xuXWu{d(T+yG-*C|xlQzQ9Y5Ib)}xZ1Qh}J6BH7o%zZIrP#j&L$cDCs2YmvDG24s6| zd7muW<@4s@Cp?2s6cVnj6$1QfM~i|X=aa?OM8DdL!o*Cli-_-)AI*CMwU9re2z|7A zYsK7yx;ez{2ARb+Fh3AC7@Qbq&_}DcR?6LXj3MOQsH_zgD#o}yf3f36!M{6`o8^er|~cIkV?h5tuJwwI6s za-6Aa2>+2r=*X7^MRjbci^ZHog=sFOF-FwUJs!~>n|oT4A)mHjeY_Dhb3`^ zI8*}fN3eb=hR4g$)N^3_IoZ#l>gbPPEt@OE=h(6l(sq&;d1ZVAE0=$aa3U0a2GT`s z?fnSWzFEcP3qJpb_EPKKk6@kKm03{xN7RiL*|2c(-j86_s#;in%b+y03R>46!TK;h zK-9k` zVg!hzP|s>&wIr;MV1*zwVi@qU#4SY+*OAHiyNFoGc? z9z+#Q$>K<_=ReyFExM(kqFRIKswr6{{SmA?|75^xK_O1k0Ojxf2v!)LPL$t3TmoPP zL2@+ck6;Din*~ze5ABFow?2Y3Au&M?2jd2`yEYvk!Mf2uzxa=4W&Rk5L0GuI^+&K? z?f?=Cy3oOpg)4uxN>REU@WxB zq;m$=N3eQiz~_ZA9*u9Ht54KiO+%XRIhAe!}L9mbrFzM_!Y& zdx}QbXvA#`mCmkdgSR<5qAh1vJIXEcfsDgqh9JAN&Mx#3A{14i)w6W%>|~ya(?j#J zv-`C%5}^`!on2{MC#CT_8nzS3zA#lsJG)AG14VIcSp;br$z{DVoSh5ri*aDf9!Q6@ zwb$9zNG&F){jWj0rFE~f`|yjpg4*9`7%ehl;pDx}?$^XfkqA8xG`H5Zvuppfrbxu* z6fJ3I_jhH4E}CO|YnujVSKt?1ul*3)Ke9BNvs*i-0vy#Gs9$R$1;y|>yGh^1fY<_c zpC)QZ!g6+Vd!t&$MW}yRgw5G4Z(m*1qsqSl5Y$N3-sbEc9?8d$Q2<17P08YDXP2>F zW<}Ko(Nt5iNZQ%0d4yA#9uNm=fb#b`yNMUW*qO}$Fqa@X8nm;E-ByV8wa_+ub<5e6 zy;MsM2jgdG=WIHh-NJpP#7lDTkD&cSx*Sg0*^PsLL-Ft6LJkpiIx~W-ZSsT0-lh_0x3jqn%x)=Xj`KIK&ASU~_hPCl-+% z{x$S(i5W=wYiD;XqO{n7&4(;=?d<4WNd#U;BWyIrcB1c5ugX}Z>D!zg(U!Bj(y^9s zfy|D@>`r}|W@dNcDg1MBXbG0Cot?}x@o{VcG%!AIdZY!M9hJcA?22b9Agj{{wgbq1 zI8{eGJF{OtF&bN@L7GkSM6V2Iw<~iou>=RY)sQx5Yp=5_crdekll=*3XSD8hcJYth zg4+K9w7*H0$$OpM@rohxdq&P?(fIy>rE6!`C#JH9#pYsK($20>&+NiT#rE1Z4bJXG zM#MVXVS5itvpKu+Cn#(&8tPO{OrwhHncWvB5VJ!KI# zXV+*^VX*)<#ytT4Sdh)x9k`c?AtSsw=6_gZ8yT`V+S$20p^AzJQAJa-NZQ#oevWu@ z3y2*xK>2%}-MOoRo!JNgV+fL?K|8y8<@hwf_yXDzuWmWJ+*6%$I2gO2?YHT0cBQT+ ziCHu+x&rM6>9W7IvpaAgfIocj8k#=_E9bAB-4|J_$>Ct+f|j3jY43G*)8dQD@nj@J zOx1vPb_ZM4;|$tB>qNS&pmuhD_~qveK7uw@+k2f|=aS_)gRdYi)qr+(JqIGxz60C$ zY7NegcYWH~BM+<&U}v}E-M2csJ&?T4?vlSLoY)e9#e6~irJdbR!zzeaY%Xm{#v8AW zncaWe-J&M6##+x%5cb;HHBUg~pc};g7GQIBg&*ga17a%lPl*X8d+qE_pDrzyWAg^f zTsyliB0HSjPiTaVMyt>i>FkQX_cmuowB_v1?$09rLwW(A%_p>bq;YoZjw3vh6Ivll z*UnDnnYiJ&A3M9vt>Nsb1YTzs{dq1~omQ~zK=xr&9j~*)rYzVp4F8WJIj2{Kvnw{G zoVbqyEr9ej${dPx68WINf@ZX| zbnWam`bUZ=Y|f!2?d*CUz+F5gu|3|V!P%8tgjrubY;R#{HfI-J9#=b!K2V2hB9oL$X2v7!uYj0*s6Sdh)xy-Z4G$an)H zsFi91Ll#FnyWE?yDyjg8;+m30($22$9$bX02C;z#D1WcBbCii>XV)D-UxH+RYG*fN zj*7ZZhBm{iTh8wDzEN^G7%QQzwdrtngWJ^;*J%76h4wS)vcI*ni>Vzfhl}wiv`5-r zJG%+v9CA1q!5GM4SlHg{>|%Z`Bgd0b6k&0Y!&w5Lp#EUQvtj_{ExNSWlQC0hxtDia~QRh_Tgvx7m+Pd2~stZM|oxV z@Jr3g$%beHsgt($`tYe?#RLtWkD!g!y4Q!->0VvX@c0tiH>Atty*~Ujrg77t?}m0j z>)MCU!K20(vH3SGX&?URj=wOTVf()}4L&^Ea$F1#Z;$@RBD1vl@N&_);D1X)P1eK& zs<`&y)87jO(G+SsO-z-9<-^+-MmHIQppLc(n-6dPds*=*Y>b5fmRXR^hYxI1jUi(n zh!dKU#nC?8FGq1j-2w5preu+{4_{scH)I5Lz_bYq*PHV9`tYtzLfD5F15kz_IU2MN zzxC2&y*{)iUfuHHjS7^J!@=kct-no&51)aL#`i^c8q=W7CSCTo_TeY+NGc5%<9lcu zwY~P?vt}2O!@>9o+ArGP>%)I5U0#kS;~~Un8qhv`OOXi9z=b^_3JX_I`|x*C3ULO- zp_L_F7SQX%%U7(y88m>{Oat17cgCJ&^u+dlTGM+z?dzWB#rXo#I_UZyg?e;n%3^F; zPF9`W=o593DueI@Q3O2WEe_~1{9z}JP^+#+~T%aKTTH4Hc&DSo1) zjea0TSQO{P6z*!6uq#GU^FS=ol*~RI2 z7YyHA#|1-Y&+B-vRO`Tq0*RQx6hVuW#o}N2gD&#hcJIZHFDI6MmQ_46)r)4P8ja-! zZXx>U9p4#2Q0OWAz==~|B(8UdaWh}+;?vCiKcW-WKFDJz7PCqbv_(Dqc$+yR^_?*j z>OxI4mW0*Jb>}34*Z_66McA4-?%V9*T}s-x2;vWm;+%9d|NMI)MF|u!7z>w2=BAtZ z>)~pzE&!kiL2aotx|w%uN#bTsg;rbZx|s(Uk!a?v4RGTNn)%0O*aK#QT8Kq9Gks=I zHuDglW)7lerVU&oVz`-;=AfCo(a#+E*-JkU=qKA;{8Xc#59sF$`Z+>BlnVN!HX0$Z zS5#^4lZ;(``lLJEQf~OaA1yLrF~3hi!)?Tmw@(H)@PH@?wS*?NOTy}t*h<>~?i0UJ zxwua*K)b4S-6!*0=o5KEPYT_h(u;oX(ycvFe51-$bWUQ{?&`*Z!rgottLcp}RG|dg zFAnl4UPgq$#xNFuSZYz6 zlWwfk2PG7>2gET=$=q~fEo;m%MB^radjw^n(&)x||GK&X!-0B+U}0T1)}S0#W9dPK z@f3oDJZ)x1W1hA>;P38lnR35?pr_ouN>>uykbWP^Bc&S3U`@FTbuK2xVCytXG3t8d z!%-i%Dkc^~TCL?4{e7m~#Ztq>UP#Ba+{kNvDr!nXn79V%o|fB?wKe5#c-3FD0Pl~Y zX2K#%#`#-Q?uuLTNyH7gq()>W)|7k9wg6ELQUgoo4&_Vqm~wZW#GTL$dLLriP!aW% zJ1AC&vDiG_GUwQao^tQ(<0m@7!fyqDH5TNZa^G27L3Ahm2(*)=%k=e>`^z(hh2KqV zzo%_+;O)K$ZwCu3hIZL#?ES~Y6|pZfcIol9DL2v9lzS+?C0+((B`jt+WSn-&J-(eo zG=|j1lDRdkEEAu8Sq|d{G2#ckG3BNLc&FU8{@@F~pTc%N+4rTE(o^n-hcd~QSPy9{ z$pgJIrrhgd3(AH#1?jxD_D;FyPAV#B@H~e0T{C)8&a!QG;-{-@3_C#J&2aP>m`AB(KL zZOWbTauh>GSrC;qC5t0n6J7l;e7KOJT7u}JDOn^v<=$8l_dX7XI6(uHzjw-=yOx^c zECR5MAUPWJl>4)zfqWp?3+=F1x2D|F%ZJF}U|ffG+oog6Jt$it(FxsYm>7rwSh&CS zl)Loca&ovBd7%{|UFNT++)Kxlki)@9fmTD?d#BuqMQh6OWORVoRRem;9oVBTXD}Mt zMAA6}>*{~TtqGjLVra{?y?4sp){&Pp*bniD2K1DB@&UXXYMHb#M%Vo{uvKFlrJ9;T>eAl7S2=B5vGM={HhH-a1oaGIdzR2qGl zo0dr(=I%p#qIG?kTM%a*=H#^L8O@#6EX7^HBJc^C?v8O`Yer0-milFE9p@A$(O8A( z4@_T4m{|<-y6V{25Q|x26eiP&_z_Z^{0Mt*M+m*On4-l`ULmeE62f>(6n_}EyXO)Y zm-%ICc+w?4%7HgIm@YFrbYlg6)(}{O61ZSIoRm)lehKd()%{iy+|DA9w%#xfcP}Yo z?zIXGnEYXsSpETAF8v{z6ueqo>_3!Iya5%2&1S37D0c_^u(oMXIguUP3u>(^X}681 z-_;QJS_fvCHX%ynnr7S&$b;xgpkwVDzBL1-HGV;w&0yD_YxyaT@dEV<3Xv|4W z!R^oxM_7rAF=hdXCDtJPK-RM%d}WDTq43&b`FLb00tvWbYz-{#t=Oo3X)4zBLaaCs z^WQ0n$y9(DQ~+7&sBC$}-`M^}Yja6UTd|zpisgP|@Yq7dR>P(^vP!^WE~IqRR;8LO6{vLD9cLxh$f^t_Z1T7vHf?gttIWSpX|IUS*hLQL4-8}(#i*uYGlP+sW z#T0`k|0#To&xo#N9Ug^a{{TTXjI1 zrhRm@fXLE6TI9uIUZAGEp__Jg#sEnbov}tpl>f2UAVkzO=V#3A+H~hJ~eV+7)JHT;YS}kBV~* z-`ce0+92Xb>BWLAZSiTEmfm{W!$VC=+--AsyBxY{zZ{sRX?uQ>Q+$9#hEqZr$D(P& z@x!Bb%k3g!Dz?wlS~O{CMyaC-N=e*`&@ZU>NE()q=kG;qV0+IaH@I< zN#3MnJC8#{_r;H>M`D*!xbgBi1V31@IFQ8g2C4q@LjG-D1)tD^m|p{qvbb+r0SiJp z4l;yM0b;5K=um}&(nh2Zz8KNeHAvrRVaZ%65lCAvn>#D#5;fv;WTJ!U=^@^uDQ*9v z66|Kf?klqWluG`kEIEm7s$yEP4#G|?enVpXx3hype6CFNrLutdKNpxHK3|T;`-E`C z_dHcg#21oEq4@RZDvJ1Ga;12DTT+V4mH7C6ct37@xlHv@i%L{nz4Dw_(RtA^xJyFAXX4EYx|ogsfHb!Aw?@A*#L2{!^pv$P>4 zEK@g_44GeE#McW5rU1(_Q0vo>Vrb8PT}1plQ8i1H6f*r|&`%Mwk{Tu2Mdpq3{j)b% z+;?H3`zNGFUis7RIYfN(sCEwAI7USbLQlA`u=N``5s7b;De%0%o`-Zpvl}JB#A_}( zYIZph-yu_rm?DBoZ8U(?%qx$lT)JiYxM4|7r4OV5S}vGpRR$5?J7D&#NHGapr;|dR zE{gnyc~E?xOb0)SkuTI<24%fv$EU~_jQmZ)#hjn>;@$!(*dfc<8f~=oviam*EJoXI zIoie##b~3=n7h&Tb-3@x@WVzY7IQc4e^x1SdhM zYN=wn(Q|Zm@%b4X2G64`MoTDNER_p2SK5Ak8+-l$OEJ8oN(7Rj=lPH5aXhfG3y6+| z-~Wc<|JbsT3@g(2<5es?NJNg#28)Zo`32m8n7M5NzNC1=FOVK+ww~#OC!vRCMC7I| z=p{lUhWViMxSes(JtXx(3XSoeL?CXZ`zFGa(irfAB#EJ zr^tcAGrB2TEcUhdv2cnoKE;;#WHH0XB0zYK)?|xcd@Xi9!KCpZwj3jixjq&dg{O7} zw#Yc#r&P;3V#GWL{hBNm`dBzb#F;ht7-7|*>~X>61$|MXGAz+_^gvr9JO-G6)ll2h zZTzfV=Fm6zH0jTN8S^65Kr)MXumqpZjSEV7lU!9eT`^)RHd<{=iA{KGVi6nP!Sm)B zgR)Ey;+q40w3+UEyXoh+vyf`AzJ92@yWDSttN}g3L7DzXmbT~4_$fGDgMToF0?90* z=v1VI8r0ZRR!qttBi_MA-x}1ydise)4Bu2)*5H#JS>?fVvdwht+f7@4pQ1k)a-9$CmO_sFc^MoNk38!mtjA}q;5uYr_$6Qf^no)(s@bWPt{G+rr2*G;# ziA9|G$s=p<>ua8;)wP*cf2*l`Rk12!ceb2O2KWj3pvy#%RPw=OvHaH<5!&GjesW%o z!Tv?-2CW-%-MuHK0NG|(Ur5lWT@EPQjbwY}cbI*413wPD@C^62Vcnp0L$14rj;}$s z4Q2^JA9*o8=6oNK?Y7%6`|u8a;_i|ytsAs%$aVLFqq59xN2Ut;+)G7e>%sq_4<5j5 z*o(XA^aAWf9FlVP~Yrp=n4rAfxo2jW*rgJwoGU@ zCX-%dGKn6kHsl9G<)5*CoyWZz=$An|{KR*b3GKaP@->;rr=j^v+w#xYrPJ_`KTLWQ z4i{G~6WVFX-*fAUMF=&`wUeAg_u3RafW++Bd=r=OP-O}lAm zsh&C2UnyM+pt%Q?fn%?Njr>Qxb2+eg2tTSw3nu^uE7Ji1Phd`Olb#f=!ethP977@3wO*A-7kWEYLWz$wUw zb!pu=yc?%F5nnhfVh|d!C*lG{|z~fL4O*%dlv+7=x$-*UnB9qfGE+eUqNx2 zfdiQV(2sEEi?2l$nKh)6_`Ob)ct*KXMOq4ELO&v6_8BxLO1NQbVKHoRq$o9BmGIQp zkzB$qRjm?wpoP9W5Gfn-H=^a1EW&?{f60@J%8R>=BSjMcXoagaPy+f95wT@!OJYcH zoETIqQVe8bdnOR{BixsMC_%M+r%8aI4WIb{<8z~y5rIDykV%}LRZh%JjuhLN*i;=! z(2t1tW@8>nwCJ8iG_DXSE;F&d8VLFk5o^|=U?}LD#W^vtiWIMzC`xTfKf*oeAZ{N_ zh}wK6i=Yj;C&-#ep#L#y?5!pq1Wt&0bt^&EuMz{Ys^9&G5$N8MCocg@>N*8w+KvHP zh^u&Sj&L8_iF@)Aq9&IpD1uP#Q4BPnj1K5>pR)L%a3$)`Eq`Yc6R}}21Er{<^n=dI z6(#V5v1@qS2ze|x!h{U{FQ7)xhQx$>PuC>c4L*EROzu({C#v1xAK;ufwX!)aVhxVXwM+tuH53imR1B}!`0le+S}d5e|QI!fwbe#-qU z{klJGCwKajM>0w9ngw1WaBKkrXVb6y&))>4=vp2v3s-WIYL;mH!7b9GFbXGq4ta;r zQ;2`4wiyP6TZvqRv2q)YnN0I)Q-W(Q1jq!xwh|mf36if0rF|^)O4?Mt+c9~Z(wz6F%K${Oh24{Jt$oC`q*f6q90DszezDl zaLT9oWP)`ks|Ja~ADn`>E#zNqC=*UqCL=wuY!LZZZ_0_2zI81xB|W+~e+_b-mGo3f zTBfx%XJ+-6T$MkQyy?pXFI2aRE?4Exd~YUW&IVE@)$?(O{$*uyl`_d*#7e(m9rE;h zbz1Fd-CcgDJnnR@$;XPKpQy^`z#rV2vZL5owxR3~Hh%IlM!XD&n!X@WdjHz6F{Im< zwT~eAFTDNW!$$#9@2>T6$7$gFJ|~z8exfjcEA66{_gczZcAabvISA!n-fgE zccXG-43w3s1Z%Ez@uO6!no+r_Ql`P5jT>&|(vxy&Kit~0KEPkX7=>TXU;?eX8`eu9 znDE?F?9dYUmlV|HoOOv^l$sjV3JwU|{ErXY8u}_qL zaaCkr$OCOEWcGyeFRs)#WlK@gCkNJ)N#C}TK0!&pM+wS3WbHgvu29}%yHi?n1n+^u zfAPMX<-<&r;N>+rWrF3Wt5(X5KYS8=gAYh?()I*L%}<-))GjqB!G?YLpx47nupK4% z!?+w)V^^|{w8FjeObD6$_kj1v<(A1jGFe4A%MO2vA%c+ScgRonb%q*PU+hm;WrjC3 zBj~t)5Ee%5O}LkYeuR71P8X#db|{Oi-3u$_2b6L^Nu(?GIh`x_34SKAwe4o^;73P1;x{jme}Y&DYta1 z&f{CS^DmUnWl9G=PWa7K75tAt-Vtb|z=h!)8#A;T*hgt*nI9mgvvQ`{gQjqs%*S}1 zxJ=GMVR)n|0jfdMO>BNsGoL6Em5t_d4QX?>@-+J=hgoj95!DIN7&hL+Yj0(Lnh_y?C%g#dMnQOm^YK*P;V|~;L`~f}d-~uAGNAT9!cz7V&7<|ki z^=w%9P0(A+v#X1X)Qe$$V^QKQ=5boPWUb8zE+bMe^IF6i1S(U14U5cuF;!4P-^1Kk z{5}RXHMBvqeuJj%2Ax6}bfzIbGZX6M6{%OkBCBmokVZzUt(Zyw-&P{B&St!@BDF=) z@QI)5BDN~(Tg2AZnk-@)UgILREgCuFV1&%AARwDB3 z$9S`2YL2*Ye>8<^#h5tXBIdEyWD)c78W%BNTx7ZP8DwtPbP;`9EPq^N#AUa=IFb;X z&BE8BkoWr~!&SaLO8Vy8(^`}H_Tn|pw|B|N0<-Z7D-?MkmbW*JJXw)HzNP9@GICeV zczdbRYg(H1hpY9)_^Qulq^D9 z*M@SqC4CFDeOmvO%Kyqt?43t=I)=R#S-2^$0^s6BSKVAyc)AA7zg3>4=aBs7q7lBL zMwmnp@Rs1tKQ&ucMayqSWJWO+tqkNUl!*ijq7hS~w>L6B^1nbCNYp84~v3cmh#8PS7wqHzFbXh_ypL2H92?JmpEN&uTQB&)2T zO~GwHsld=l0GBi*TV6q1gYz$oW#}mY@tIX7+4@JgOgn?m_pZuNHUI@Q)M71KNI|=T z1Ny`>lnkJjhGcdM+8vy~PY#AU0_d$FnO!tpqV(*|m1}x&(WgFIU;^~1#87{T=s9O{ z3D4mmzhQXg8+FCEkk(l;A1I=)gq9JW<3auv3}%i(KdYI(yQoX|R|*xLQvo}ls%LEf zg8tNEMATs1Was%UVA8L6z9(oNu76`uC3uJ)OXfjvwy!fYGa33E&Bz9iUeXom{1tqs zNEkEUL*J$uPDe5igD)*C#ms5wS2V-vEUtre9tAHf=EuxGq|Z-V9?3iot~*mP*`XKI z45xEq7}9wXT&S1T|Ilk`hRY+Fzk{pexePhDJ3#NH8BXVLex&m>cyEnr%#4RVQ!|`S z<~CV`=UMO){3p{{34N1|$%0o*dj1J+)FdY}KS94}V+N;yc^(v2K#h|p&|ecHcYQ98 zWL^YajH}JvnPmZXeJspS4~uBYybKz@H;kDC=+!jizzrWFn%*nx`8TK#;+t|z* zZv>TVtm;x9R1+fQZtS4hlPAy!8vTuv>)IPyf2}9I-?fkMgc?Dg9Vp8BG-$KE`hZTt z6J`Xpd!ZcqYDgQj>?pR_Ej*D%P;5*=Ha`LFj934zYO?TTHG*TTOeux(|Kt~Tr@m(^`LR85eMnF<49C-f%7Wjo#$zL&LFgmBHsc@db zn~gm?jez5b?Hjql7S_DnFC7)8CJE0j!~_<4`rc_4J_Oiu$TBm!cMcCK_7U4zqN}&pUGp&mqG(eu9Ua zsyBckHfZenx;WMu&I5_%82S{zmm0c*PQoQBf_^ld5#6#ev=zXQX&_awE3YxtJ7PHZ zf6DV<<9ARGG|7E{3XTy2oNF`4HIJDDRcTC5t**vEV*99Yi)4vbx ziC5o1^S#rCbILk3kVC&lBVbYE;9e}+{EXoY`d1wlJkUycb$no2cz!XQGbY7x{&k@> z^6G)4pEaCC+murEht|)lKmQEw>zv_Qy4k8fv{_z#6_xM2;e2yVo#L&6w%)6skH(I4 z!Eml!t!jluI2^r$Ae1>3j~w$F6R#aR~p-aK*MzN2JZ54y8kKn%4|xrQ5jnM|HXe>X9ZX zIm0K->xSzp-dZl*Nbq7j`h z(L;D{8O}4UYH&UomLPh8g=;Q-vOJbCdE0Ofi>t_FF;K~xd`ia;KI#>x^Q${XK;8iL z&hHkW`e{-&Pe~lP4Edu{?=@I>?i$XPwZl2}Dc~1s8{cve`KRHGTb-TB&7clx(x)6} z%HrsC&v1s$P~4r<-4jVa=nm~KctC#MJ{otYDygowBa{5CX>ec%ROoksBuEk%a zu>EXk3%vSc+*|2+Y&dU?ti$20 zvhoz#E3bYaH~i33!&L+EOj-WOrDzE($`2$wDTj%;;kJ!_c_1_q-^LVq3Y~6m) z27C2UAD0jwVLDGeD9iS#dT3j{`iV)HZCkiS~jc=aY_;@VmfmV&&1F_1gy5I$SHmE0S+WZP1m9lDwLE1fLlYH zl7u{_t4akGxvLDIj)pj;Re#nJo;cIFH%J|lIs+J>A)mIFWHHm3Bd3~LOb7LqCVkqz zC~hY46gOQf|4@gqZJ>^5(kD$xmM~p2_NwUm4N!k;(kIRGG-)kqIunnmL$C9D9Qv_v z&&y!FL%f4iH{Niq%Uha4@X3DFD-<&*sH%!^UQLlt3;H^CD{pWpI0wh6^DHLilKzJ&cPXBc(Zs!J&s3GYNeA@Ac z;t|4=fGNb!douAG&GDw^(htCeSH0f40% zqB*{ByjK<>g(TBG_kIjRKL9wYA$BJcN;cguaNbU@E;0T9@K8gXQmd_q4OTLpAJ0hS zb_iOFuYO?h@pyCrPB^+Q#`L8kykI{7ROs%mx~&5RRp)J!#9jgP9+?aQFzemkO*n*5R#hSh(&!<&B88`h=`3}uFt_Kspn7Xk zmQ()-#|hm0>#1ov<6kLPIuYonX>j4l`ER`N)H0op6-k`tY5@DvK)#1NiPtur&FJes zG?V`o=sy~l)$;!ESeeigF_t=}>spp#T){A$2IRrQO+ijioiDz4iM_9`=^PlVoJ9ht zCYscBOw zC8$lB7D`*8`IWPrqsf8wYgf8Q_2RQ zfDMuT}=?cT^jpUBg71ZGLNSS6!)49K8Dvyy_pcZRV zjzFI@TbZtePt|qYouH1VM~$i`JukL6V(J=bU}HIk~?; zIX7EXlw>-0SqMkTjqqljO=qvCDl}RMMitW<`s5?=E~azP1~n~h26SjTc)FPlJ)BE; z-Z!1s|58`cX2WttT5CBBxKa}DYC6}~QajszptsY()6MM1$VB1kW;$D5sLoAdZov~- zSh&}HDkaJ8rgQ!QbyP13s&-m3-7NmdTw8d0n9isFsy(+8jQXcFI@{M!LGs6-o?E1F3^|W8 z#6G60&K7kf4&MqVh()%Ca7@gJ8Ao5!x$c?@mX?GT@74cyWA5{T>2efM*MS;BYp(UA zPJ?iXv!Cf4Yz6T@fHp|$j#+M8Wcbi@y`Jr4M==xHXFfXKZQ|)~I{Pgy$^Eqs+7_?g z{cJYj8DP2!O^9auQ_#+P^}jM1!ZXlx9{WfI{~kko?$t+!2Mf<2({&V2C`v~az76%q zqUxV@FQ&hPP3PDuYP^($mSX7~eB*%X$tYYM9Rf!-PdUeypt@+1ckuVS#R$((94Qi1 zv}`1R85+_@QQ?r}Fw^;K2{%`KC8!_%H)+rzUE#w`=aMJt9P%R2zclVW?)%^)Oy{fp z$}&#tsFxdoPkA}Eqdp|-%k||9vT`eD~v&~5WX4{~Sn^2~S2*@R?o$q@9 z8!bpUo(CduFv)a%azIU;euDOk)|0w)=qNms(aX4fTF$@zg7(tVRh@oFtt333n6Atd zgSbu+J8|n979I++b(4B@K=@~h=}f7quD_NCP&F;&D3^pI>QtO6CaXYFYXDtski11? z8rexIQz*G%QmGOa>_0d_=P1lAImAOpI0PNI|9CTDLRfWbCvFpu1aB50suID+R z_uh)j)MuKm@9?55xz9UxA>faNYb?j8J@r|pGcH-h@5=zK^;TTolJY6S7N4uxKu4g1 zHBJ?j#a8p`WLy&R%r>2_bt+ge3+S@9;vG)nBzF$ZHH#{IFVMplE*wV!s|n9s)7k$M z(?9^UMIp1N9=Hvd|K} z`nGep*f-w{*u28SHxx95*3zr5e29~b1!lnGrgd2#0Bx98-}Vmj|J)3?JXl>kn*;50 zuYPl7C!z9(`oEh=dx#^UH0mynpY5>FU^49CF&z%&!GM5 z)fWcg81|JJuxhw^*;mvLI0?eyvj;eue2M_f*Ji-yM}xT#2>>cv5bjW@fXfDp%z(}n zR3`61YvUBQKX6bBXcvu3KjN4{hx_GtnOw3u+>QkBNm?kWNAE7U1qfFx zCaH2R1@N5(3CG7c)_Yc(t~>+ONz_4T$GrNpo`@iShe*D4+IbV&J+J=tCG__yoHboj z7mXcz@%~>dYP`p+K+Iva>HI3Yx;vyGw4z=;eiZ!N_oi!N6E)|k2Ca@)Z(1(9@T@VN zyU(f{Gde@-;nlxuj5z9A(>2enE@_Q}_K8>jeI)jWb*8i8I(1{fQfO-}T@96n>oM_J zZ#t(eR?)Vj0M6Q=eYbG%-e5Y*6;~7UhX7us2W>Q+%Y_PEMD9cU9}5qEK3>HyMGR|` z>H4#UI@7EGAXP)$yAs-Lx-ylj!H%^JfNmP%l&0a0MV>9Dv$wyxD{TybNfsm=-R@!> zZ8e?c6V&}cOQ3DAbPjmXq*#&LaCPH76+}4!>WU`i$Y*lg(EP%)-E__hP_y9Yp#1h* zRiyn&I3%>gbdLN-#Se1=D5N1u4)f)#2!rf2ol8fl17Qk)h8mJn8{XR_xyy9E*Iiu? z>H%t)Cgl{1$;r<`aK{)Xg}9<257To%eXU7u+di}4gnuxdHSu;b32g^(P(xgWkMIU= z&mPRShpL0?ZvbvukZ^qY11{$7HQmJuPC^hC1sOzt<`ldG3vvN5QRn)LBM*SFzB?*Oh|OfJS%7y)XkCVkrO=N&kpAH=MG zu$px*1GQe0+^G^eWICJ7Ra4&M0Iq9DCnulLz|Cls=3yLYg9YdQFVL`qR^4@fD17|@ zyrk(z)46AnnmZK%T1n&f)SVx=aoZWCiU#v3pb^i?pIgJF_{9Yp(cID1(83S?!jMFW3mUR zVVd-&>39*G8#k&U7QNEsddvZ}RFk@?MG{?vJZ(C|7pXf*b^|)1LEcf4aV6Mu#&j2X z9>)3J0Pwelyd`u<@)y(na$s2|orj}EE-e4M*8O|nRyNaFB8zgi@t~?`k~8j^ge$V= zOjoB!bob_y_QN8pt<3US|SkaMw+zJCn-zKcL}9tZK@E zp?`!!;=h~Dq7HTPRTOCDv^dpBW+5D--^Zc$25yO{tj_x01KL^RG>C=c-fdj>`~xSc zzxZ*9KL#*84WzD9N%AK4ZG^F9-By6wrb(Z29({pxrCX-+&FAW>%NbBNG%5EU&RvqX zP3Qmcbsq3JUH>0H=eajeZr07slN~~=Bt#x!g(3uzNbJ#|R_!e|r6{qfRZ8vJ)!KX2 zo~1^uYD*Qh_oxc~pEK@#lKk}V^*VWZ?&p2J_w0S|{kx<*9QF>Be&T-;fY>{f#$ff1 zO~0BcS6!k(IZcw-e@Q{hbJwPwm>~zdDbO03I)7S^*x$@2IJJiE`QJ1Vg&qI~<${`g zf>M0nrmv|ed&%hl=35Y(r4ai)oIQ$tVAG!-k`3Kf0J}|y*!PXbsPZ9>&CHjBlZ()P zcj*Q0V8HbV!=N3qT)u+#)}OV6W@sbjsQVSP zw=R9kTjbO~Xi8?v0eA389DRWk4S*;;QJDS3zQz(j1-Twt2|xo2;#GkUAjB~=j{j@Z zi{+PL>IQU#h4TWO4PRLaQGA2C{6RcW z(SI2!3w3vBLrtBpF3P5plOBqmFjMwqvp_Ak$ZTz0odks5Q_;V88}FsSoZ^!u7ToI+0H(FZsYdp7xnk)l`<_fd>T!yF>piU25? z6H4gIMmJbXyo@Kpc>A0RpuPzadrfw1(pNENm6dBuU7=;T^vUDUY8#5)zKuN8IRV;K zm!6V>H4Q(-h$$eC7=8zBol9?%ibwYxitc?urtWcQr(OCnb|}_gF&fnI5s|nH?XgRb zE{UoXpy)gB!4e*R&oh__z{#>7h^8twP%++S$V4axt)xp|xC2dUkmA*2Wl8arN-DG_ zrY_fx-#0{Z2P?YU3)waG0x;f!a<&1oFnr0ThapN-d!q8GjEC1Q6Qz&{qmZ!9JeoXn%>o2tvBL5{O%`rt%~;Grb^IVlc)walNrF0%6!3vLUv`HIaw@1r z7Rmc=W|-K$8JS-R|G9ZFLGA!`{6nPCIh4dkDB;c)5rVt{>XAhXa~vw5gqN!;Bja%n zO&^?ywrFnm-NOj$BLx!k9#Q~HuA*mNrjM;1CYIkjD z>8394#Erx#M*jM;=k5()s0Fd&MeN2ytoIaG^sE80ADIPUodxkW*=m^O;~Nj%&}GLe z`i8GPMHG&Me`fh%DN3#GRh-FS(@XJ8AA6Aku|HrYoQi(>syrO(a~?BXIFSnfX~JcP z-$2Y-dkjK*QbO^n{H%g#51PYJ2g?|bXhPrqn7@@&^uP1Uanf)A6D)|=GXcdZdc;y$ zBbNf$Zb7_G@E~NvODTGqP0k6=0ljVExu^Bvn}TqhP0^G8m7BA+3poD;Cn7DXI)}j{cJrM28j`~m3xoVd%n{jVB6B({p8muew* z0e%Dew}p$G``n4vp)fMAqWVC2chC&*dP5u-kiW-XkYN0&sH4oaimTq6f9)w6z zj6;p(!|=PH?RV*KmO4pnO+{}vI8=o7H)z*g`eyc+N-{}8pxv54b%{ebP3uYE`8urjEkEnx@VxwjDMghSIlBB zrbDSI=pIperYpueo198k15hU?lrZ4aJ|woOq9400+l5X5 z`dE;t-W+PC=(q6%3vZAn0a$E7x!W3!H&^uD!{P+K6X+y(mH z!gIGZ9B-lMHS)-%kZ{11uQ7J$hS-42gZsV~Y*a`*h zjvR{g0PuMZNDfzvY(&%5TG4<0SawQZgIZyc{BSpaJCvQ1Z4`ZJeVa&yeV~4_NHKHc z(5H%in9Aw=EdbxiWI z3ED+d7w?I(|56HTew`HkNaLykdII2;2@(6mQD_W0D?tI_!Qvp5-yc}|gcJGBM_v3N zmHoptV40#<|4Js{C(yc?I?rf1bXztOt3}-uBdMBP zZ5j<~<_Ab&vpbri6>@=MEvO$YlI3=Gn>`d`FW!#Dk2G8cb>~Cm(I*OyjwyO5mVbD) zQ?6nY7%p3kMKLn)IrP|Gig9@?6*gl*RsImkZDyc(Y9+hLbWoivGFv#g&EAU9?}mKX z&4Yb^<+Y;UYb@dZHg0j3(z$5dmET zaLYGt3)?DqctY z(uDscXlX9JUwiofT=804sF2V*KsE?(b+cF(0>uZl(Zk&0d~ zS&jkSt|R_%GX6myA$>+E#?}-$P%8qhq^ZlJ7isJ!nV3hclB0-J0QF6X*#BZT8Lj9? z=g8izE3^!kzA_r4u`!CiCQF_%m;i06ORrW8qnNQ+8+;^3A>Tn;=hEA?#`@AYC46I@ z0%Ed!9NKwPm$Od}jaR}~PgVu=2*AGP;`1rj+X-d!uc;a(IAL!pF!^d!gWb zsd&D5nL_ySqy^BHy7bEXQ1GTGUO#k|Z%)|*?VwBlb~w%`OjY#9vtp4`3IEd-ucrO!w$`YEV7&H6IwayCw9_N5~g~&d3=Z5DiMwegZgM6O1@Dd0( zS2rjPFHns7hvl5Q1c0(8MC>E~L?vISXyJH(j>oJXwC1KR2X}SZ5!XeEab;>%5ruvL zGIKz(6HomJ<`!cJFU#%#f;ZQJVt6l0Eu%;=EY7#G61+CVXa6N$?4)fjf9%0utTKtKHtTxRav zXBdfoujrAr$oDiy>I$+@=`4SIO<5?LbfcC(f;JSM-n#ax3sQ(6=AJMcBFZ9g6YuN;#`3(G5|Gyu8pWTY?u}ZG1 zt^t4a1AbU6Xtw+8QH<{1az=0+eAfT(hu`N%MXy#{=0(tb40z#0^m7*sj_+0UFSg2) zVM#!ne*hPWn=hgk&g5a?cc6UHPCuaIEj)Klw`K2o+OHU^FUY273D9l-iF4}*Fj&TC z?|55$7UWA(@X%1EJ#@5 z&@sg*_@J7Ab^thJLBi78#Td#S$0+o2nh4WX0Cz1&SmMwLMXzgaLZ}Zh9fT7}AS?}I zuR8ciG46%Q*TEJAP{x9UB@UfbjF4FQ;-k6%npu#rl+Y1}2u`6H!DGoho%#Y8WH_T&CqYaoZw`6w%IBG${5{J$xhId~1(zojX9$1jDH1_}| z4`&s9=0&+U=KcsRAzZdnVCT>|OzN+j?H_=63lf$h-eSn|3({$*9Edak(87X5m^gG^ zF-EtLTN?uad|^StQgb{h8hZid$zM*G<^ou5LBbM;E@FwYiQE#{3*dwW2}|=sl5xaN zF^05|$HQ*|cx*w!5{EA1Y(|*3C{Lb`G5&+gUJ6Qe!Gz#fY$h%%D)O!bfR8LlSmMwX ztOY)nJ#}LMtu07|New~X{f5I5=jHHaFn~-85|%jhyJEx*lus8e0IUbg`}vmjxq*AJMpT~mzkadLyx=LsHtgUenDIP@oW zIp52?D+Qpk1qn;VXQDh^#|Bt}OoDU(ZF51F&ta(lmtyqAERZ)nLjjD<1#wF^6g~D! zdFXC2fbT6x#Ex6KsTg`cIs7^d;7l&4Hol%2drQ$r50X0-_W->75cIdAC)Jb-Y{pYO z2?UqD6nqztA<}Kd_;jJ1DwG9K)q+HrICMubzU>z!V%Ho%2MZGElvNH}GAuKHgur?ROE#1cw?*h4-cnrWV79_&NEj>_-%9(YArH24s z=Yrhvd6n3QigDjAM@9b6a0nJoBmwK5hq@# zwgoK#v@;=MZ*dpX`zMO=Y?Yi~d=720sb^my<>XVvn4VV_x)q?d<|IXm&SP)ldWI+FDAH|4jAm3@Q z6~Ipx#MesrAA!&D!tTUt#i-gKRwT+Tpl|*YpZ6#7iHG9S{?@=-h>EFZsG)=s;`+Qmptl2`oxq)+2M?$>dI_c zRr^67ZZW(ZgTc5=RgKiya$S55fTb41>$8B|RO9hA`A(-j0FGIZs1axIMZH+Ns*jl? z50?D};GqSHmVra=s$MWuLfR{A0>HVd0I>(xu7(4As*$I-+;ENsP~L)sB@TJ0#EOE$RHHJH7JG>Xb2@4Xj>y8u%Q1#a%l4%4Cpg?8+y*k5~ZZEyy$kC|EUYc()Pn)f)n6X+a`xPb0A76{6~2u9pkH0|AV* zAQ3kXg{peqc$vHN0IaYeVX4U35;)kW8ZZ8kL+*V5ezG88i9>l*-7b zrA}BN%B$+p>*e~r=f4>L!R06dV8T_*(d9;j6M6-UG3_7$@~Otirn3KR1fZ1#iB#&u z&VJ-qjh$8GI>;aZqbx|oZK53qJR(p`Lgaq&d;lvgNW_gp1yH`r%CoZj0i3iTVQE~? znmD+p8n-;;!Q;OHJhdQUi9>}{qr!06$$Gy*`v;e!3?Kyxt9s;Kd8DNz^okZEBG(;E z5mo=wSME7AhThs@L@IS>2@t6o$NrXG(qI6Y7GxR%6r~!Mh|Jst09IL$h}-pc*d2;i z^*?sW5#j*=rz}XsjYCCMJ+-*pkh~4xnFR?;W0tw$)p4pZWT`xN;**8(FI=`FVCPVb zYHWQaJI7K0DqE1SG@>NtF2z+nys12!l@6e7E~tJg`kPo)FVj^noec#rDF-CGuZAgD zWOJ(e^2Kt^b~&gm7Rd)2;>MgTq3Uz+>RaB?od$KyBCXj!HtgAH?2@W}>yq5C{}+_o zTQjo4+18l7U~M|%2$ z8vP-1`*fVKsH*BgS7Z(^1hw`<K>WiUeHC1;+$i=;9 zpx#-eNX0$$id3uys!k3PujxC4kyDbY zFB~PyQYQd?EJ!4EZ=8{;i4w6vKG!i3`mCIcy=ieQ9wn<$)8Y+rw{-xvm=LjlI|QXL zMb-PPmM1DtL%U+?ym%A)p7UsUYN>kHWAeG?7XW-|j>uWHQbu$20lX?86?JWzoFf$l zU*UiK5PMuC5+qI450}d?>P;h1oh_1=16~TnoZu{bu&=gi9Bv@%%}Ai*Oq|%Cv!?^= zpf@Qa+k|hSt#s*|XQ6AWtLlYtu!pyJ`=K3m=|Q3Sae7bHPvk8t^y|$(k9qj?*-99J7! zLrYKSf_|zI#^HzM!fZE4y|c^c>>6Vvv#gc~!$fG)EZx3Q#|mTL584x#9(n~esHv)d{gWKrcqySI4_vk) zFB;N`{cz;3nQBxEH`hNvRn0*Xd((YMB(^z5Z`cz&%oxb4! zrsRO+e4CRkP^xa(MC@0A+MbhSWtP}~Ylj2fEmeJQXIWs*fx2yxy!^3vNW&G<5zba< z1HOi)jl3=)O(dMuA6b%E%TdPqm*Mtc?4rrQ%bL*GKU>81{pBhuJLf1ysn-`He z*8zNn^xg8Gv`jsruWMa=7~%fQnP6_15h*s3cno>ZBTd%gB#%bq3Hk7xe8)tUGj8b?+uYB9A8nn4Jr{+W=#sE~-)QTe+^b z0l@BDP@z(&s$Et6hZ^$4&qV;&b3wO%!!sz|RAYHfxv`Q3z}=oR2@d@OOWjp{`bs%j zF94vp1&O>1n}RU)P>q;da^P4KK!aQmx6~8q6fF<NZ_80#gCZ z%?XJ|?lV;VAQqE&ORyQhP7}g=E1IFa^;Y#E8I^^89@;fi7oC83OG6)3@0lTIeE$L< z?lShoPJ%JG>Z|Gx+R6q$3|bLWmpw_+h*X@UQ}wE&nROq zL45Tp+sR#y4_5V65pqUj(@_86ToobPBMbKZ%=%}l{$rSY+AI!eO$*O$o#R6g;>L10 zpbgOe7S4Ow+<_UY>MyFuz)S(U%EEJ6mm8cs0-vjT;jiT_4uJpN@{{ux(hm>TFje0@ zPxeKx!25Zak^bO)xS!z|%kL>9awrCTlI7=v_bC*QanA@fDm+cTkEk{HUX~x;T8eg+ zlV7M&Gq2VW?do_?3oMd<=`}Y#QjI!sHbvmufF8AQK2ykMc$6A7dTFpAuY-DJk=$@L zeCu|19OF~G)Ha_ZU_IP>4zkI%DxaY zUe)JUmJ5-WpYfo9&4&FbqHu3*uq2_i2JBtf#B`f_Jh(ssbSOL(mLW4_qLJ+wB4LvLF$rQtV|}U#UiI zoa*Jb8xP>C4?#1L3juO%Z4H3!79_&7h^^kwQjIC?2_+hwEby(DWBGVC`iW(7d|eP)v`b%55BeNcpF2i2+|{6^Si1i8SR9=C zM%7}57RC8TT<>CPbR0jdSNk-^-g8yGV|!T=M*$dbLc~5H664-^s(z}VoI5Rnw#udJ zY=A#s)n8R{iU=HlcFNSn06(F92dqggP<5v&SCel8cxXbzUbQO9_d->_x>zoBXnH8| zhjaA;#QrtD78<(<8JSXD+$k17`J7Nfrw*9LEyhafF_|6>0kq5o@uy2xsGf)K$VKUa z07hESkEj%QfPg)qw_5eQGf{3e%mc6@7sSqkZc_aw;$cAE$n1rF-eTB?$9aPxTMb5U zRs9{mkXXzN!LNO^&L%_EIK)LGaP0A3TRlF|D|$>ZR~0bRL3g z&{ACblT{x0NATYh-k`!#($xF)GkFon>cn8H31YU7DYJXjbDeNjLoNgdtyG!G<&zMqP2 ztdo1u4%qzOi9_7l8*fKPl+36iZEBM@J`S~w7h#QcUIPbs)9{7{()eI+8+oHUo=vRG zm8i>ms*yLozID#j=BGYuGSXfS-`kC0pY?s?=l!%>9nv^Bl(d4<9?te_4@^~J8vCY_ zHxAEEpZgJJcqo*SE_K-9V&qML)&QS!XR-PQ$|9Evd5(!u8a@{ zYl*!=xq1Q0pQb9F#gAEo6J&2f!kRBCN%|KFBCjE5a9%Ar#hV=~C6R0??$;mK1@8Pq++PTmyoW|FE950*R*>Tpy7}Tx#Wg;~0g_cg^TvKB5BWt%*G-x3=8>QAGC#?xtc`UG5X>9sc7GXp$+Ru$ zOR^Gu*IrP`o_2ULBR}*aCPVlqr?b-X)t?k8zFM6Csg^18e84BC%Ttn7KVP@&iGpbh zy@$o{!@NV-cXP9v#9>Ttwo7FPkgHQa0 zd=S!cOAe{!j+AWcw>SVt7}=5fzaTv{W%`h!trbjHzq>1H2*w>H%mK&CiFkccNaI)j zB&(ltJ5Cc!4D>jYAw#;fCs~8MlJnFOj|$X+QqNLDZsL_PS;IUB$0mtZvNNPXrc9S1 z5Tc0=+0Z4BWDWP)I4X&-@=vD$TKpmKZ7|L;j_~YyUJmef0Q%Vi`EZkW+8@^y0>ZC-#8P( zf|mh#NG_zw$F<4ZdfMYv`8YBI@*EQ(F$d>XA#WQ6U)Lka5LjFfWvfe7ixwkq(}UY! z@Wa%fp`2$b4}Q$MH2D1_uFG_HG8yUvD9@ae z$)9>>)vQ9;M~90+F3AuUITG{gbPRdh&U^3|rMQ{`rLJj!M8^DKBX2ut6`K?yZE>}S zsfc-W%&adVchMiD)#asQG~|g45vE%NCy=+@e1GsOLiEr)WC@g2rU4T3cR@e$wwEK% zK__X6F97d@a>V7g(vA}3ZC}qn+SVehd%p_hrlm&K!qIU$O97H-C#rBBJu<>{S zuERY1o52$DB0ewtuDI{mHIy*h9iVix)HAfnNlM<8@Om~eT(liFLt#cH0hw-LkyC5? zkawlE`ER|+a$No1rTlO)n!GEoRo?GT_TuVkrtpMt#U_$730mca8o2@ec}|AMrhL%~ zq2ygsNR2x-qV1*qkVedgzazaWHT<%3y^+u`SvJVfjHO3cWCqxVTFUf zfb_=YoAp1=jap24brHpYyrCovPNYE4@@aL6bE|e?e*r>E;Br|e@%xUot@er`&Yixu z?v@}A*^TQ#Yi{bI7)$6!->w-c0_q1~xCQaTc1+=x4*ACREhV5i0G3)1FR~Ik?E4%a zJ?6LD1K^kiaZ4fRheZ+RQSHL)C=r|M&>t{{6-^POkV^4kc$=f=g{9%-E~+xMhldhx zIGN{L7sV6jDbLrvJ%wBZQjAMxgGA?9-Td^B3Q4^%TM%d2R=( zi%TAXdB5|b@228~ggg?`IG22US2^OmlCQF(Henx>po<}`aLKG?bzb$I@idvYtaLA= z!!G&xr)7!ry4S&``9x8>3hAaxPP~ke-_q(YDI?P69VCx@X4;7Avb#t!ao*KZm&wvl z5PG!9kdSs$3KQqUfTBJLgz+>HQf*TfV>w4Yo$8zMOK zQ%B9LXg)@ux1im3>7{3T5obZFd*DPQpKRFkBmdz<{@Y&$p-Cu0wV?i0L^?%6EAG<& zIbWGLqp4#RKA`=am8zQ1YPIFa39xpuUda`~sQ? zV44XLebb~$#2HV0Z4dGhx&qfXn3_!GOZX~)vm*6ftH~MFPXNxE5Yh1pY~rj=eY<0= zgx<#WXQn0%u-Tw91ym!9~#JaM+9T90D#vj>x*&2Z^##^`KEwe1VU z*qQzSZKF%~yNzt>LbdPXii#9G3GJLqU;iWYUR3k4dkg&mv}Z1TO6?Tl>_@fUKgbEK zPXWXqPR4(Hm72skm}+`Ppzt3Ht+Y%3d4CXb4yRhgD<7fPf!4^Sv+1;RG}V608zl4| z(E7S`HlKD*q}qX}ROpkT&2Z^8J9(pLqFPd;7@_|FZKF%Crn?d6*Hrs*l~@y@C!w8l z=~>%R4d+u$X;)MDe*o>7OE10zE#(rby*nhPFVv?X;twa|zp4g$l~q)G_OQJ09}BIt zOFuXe>ARL{H|&*#UI$tumtN^Cy7?_s+lf5{UJZIc>+8~|euVO|lWNJ;!-PH=+6}(?KyhQz8=dC1Wz_HLuGo6oe?0Gw*vHgR3)Iy)d zXS8*oHL`Sl@M#0BE!8qf8lhcqy_cogHNWb_+0zynU)oMs=SIgv`^wa11$A6QJJ-ji zo!Tv%h&2E$(iv>wFf z7Qp=AGa75$iSuWhmU2TrB;F6eXC_4Khc}|NKV$P+6;@ZY=3hbk#?tkxb5IP<*|cq~ zF?3mn>pM+Nwxn#y)p;H-fcB~>qIn*`6%!)%*9XwjUbK0o7L~8+d;#r^rR%F+Vu*Ly zro9e8f96*N`(JP}2^yV6Q-8&#CEzeIFT-)r;$1pho^}3i3*0iLlE}vf&{~+f%*Sh^ zN)zW_Htq1@L}6qAfT1RY7e&8BLHpYlxDWj$FATGw&2#A&)|4d9`?jFJM&rEJI3L-xn*F_q?JTZeVj3T!^=9)@=VMzupB31`*e7qD&+t{n{{B3z2 zBP9k$?U9VWX!J>j?U543e>?c^$@H<3ayuEie?SsC?yKvxIgb^OayZCG1Naqq8u3{1 znO+oE0{NBq@81uP-PfQnF{*#+xR208vLDskDsO(yPkzMjLo9@u12GORzX&gRsNONs zfzKe3xpXf$&mn|dGXXi5o9%YkD;blXqQRr{`jh9h@oW!bqZ*&pxt*}xZja?(s(Q_^ z{S{S?sI<3N@u*O8!nVE%!LL4E`0tP9YY;c;7B<7SeO4V}re*LC2q1&JyFbNT_Z`|@E7lrwn9+z4K5pAy9`x#02+1Oy5MjB4a zDuJKP^6V-ZL{b_a+liOc(3bpH(}t`ox6=-u^RZ^qx#7z0aro7dFPL?0_;7Y%{OZbo zb&IQm$%37B*YdRy^QA<@D;dtVo<$>$wV8I>!*BEyKi+1xfz;8G8?vfur+qzGbD7cb zn;Q)7ud#|XsNvDS3*y&6{%dfd&u^F#ipIqWn51 zjku|7+eD**-X9f3yb{uCVzp-6^n!#YkE-oSY{_d!MiM2tahKXeO}-sno4An$Zj91s za99oYP7uAz{WIlFp33ALv_CAJ2U*^K zs8aQEDe?^3J4@%=WKtiaT>B$G@s9~5`QSwOSY%B56BPC1f}}LG1WRZ8Ya}i)2+PTI zlCLpE;VpbR^p=ci#Z$i(UWBgEslGE4G%gQ-G}I-#&t(btm2XwNsEXBvnULnXWKS;7 zcGTRL%rj>bq#Z8#L0c8R=lRatBEAlG4$@_p{Fuv&d@ru^;KrXq`o|?_3{QpcrM?52 z=Hap*vMU%)hMc{@TB9q26L(Z2?GcxfkjlB_`|Vl8zW4QAEry`=AT@ExtjW>nx{&kL zi}B6No{;)ka`V%p%*Oiqh3p-phNR(qOUgja{$NZ9Vcd(n3d6@b#QY$xicVUuwgnF_xGs><0w;;(p zN~=v9U?(Bv^ui#IQC^1gm7CEcm1G_lzw(zGv;)6Rh+h%JqX3NEwjIUsXD#z%Z6NM_ z2JWbBKdyv*fmQp3Zg2l7N;1>E)@JC0GGGndb~xKsq_pVuhcN@0&6Ffklh}@%8X^7@ z7O5mN#O7doH)9nxczM^$Ymr;yo{7YRrqh5duwtiYvD>84Wr-*CF{Z?pBR|<@Tq%HG z{=U^vwpsJIT&Y_z23HK@La7K+rX>9FR`Vy@{=LO;eVp$wohZIdlozL8EGnT=8!8n-Zb#W&B9j$og zkNEiMu1eh79U4(HA2Blb+5E}?I|5giT{$`nD>lvWt38};!N_nzm*NlEL&Ko*tfgrT z%sgQ8?v|1t>t6t;Sunrd8v5pIWd2c`*P4uSgt8mZAMj_JiHYnl*QR$Nl6lf58cpKC zxI1)NzcL8dX%Vi-QW(Xv>zLm$A@i(FxyaUpvY07|w%IIn-?}Yj;yIgFES{96KS88n zrYc9y6Y1+=HHa7W@F+7okOb7;>LfKz28WkBC{DUI1I~B&#Jk-1``2dM5`Du-jk8*K z%}w&LKE5i-x28EkG;Q-@LOit}@mJG&f6M-||4*bdBTEoWV)L{>DV#{t@`n*0S~6@M zZsV!#z_s(#c^0p4p$olPfjq}<-#Q$Fu&?kxh_lAlhKG1xcI~kxek;V;TrA^YKef{E z$ww0BX-RHzt{<}zVO`g>q$Mnj=kX~3k4Riz;#LuV6X9&^52TGN2wCdJ{>I^gr?w6M z^VGsuvcGll*P|uBY+Hr@cIN*bjK76(H6QT}HR`(HueNCaj<;}!}eRq-bU8nHj0 zvOm4ppU>H!iR{lD_GbnAvzh%l!2XDr|qQObC2oPe3#a4an;W!;-?v=NV>GNGjUZQWo(pPdI2 zaFU8PcYIV4v-1aEAO%L)*hOFZiT9YU(WDvZ_HeeHK6we2!ob=#4CNFTYYjwN&C$QAQU5S#Dycg;vU!cVfs+Tt00|{0@|x;nhdptF7YMxD>ew z>aj&O8Hd3BNL=pw9l&OpMAnxJT zKz(Nb+bA3L*%&^aFAJ{w}7 zf8*PxLJ%2%H2ws7N5-)9kZIPPGjjuPJT61VgBcEeLN2Dtr++HPFrFV$F;kXt*vOfsAqHN{e;!PA=yh{3ACF{dvBD8IE0Wv>(+PT? zTugn=tPDARr6PF)W+L=iIhhO(d?ebJey3gE<4aaU+nYnrmW6R!AEtp^>l@#=F5i{9 z4D#WB(HZkP)&{*<>oE{(9p!Nn8ct-ms67Wllb~Lkv+YB5N=G)G-= zT%{VM6ejaxze2&8OJ90eyYMWYw8Z5;rccV&7F@5|B$7O(y)0FTj0N?zMe;ov9uNs# z(5{VjBi{qqWkKAW2<_docHHxlmhp2eIRoHN2Juq&Vi-#MWnO#*^qZEl%!|APV6R|8 zyo?LzUj4$bbWMA7xB$rmpri#^mSmL?ANcHqL7KJ|P(+s5Vfkan4}yn|p~drpn^;mmpH zcXKfGDyxI+56O6-v6Avm%iFmsc@N6>BQx(=Jr@T#(_`>O3mj9Vo^fl7^X984s4^DG znlF+b_$_NTu<)`yq7aAb0%*k`p7<4KBk@_ULDJblDPQVAeG|L5@y2oxs8JTl8<=)e z9_9n2Q_pFys_}Rz3qb6)sGBT!EXV2izG%N(Q5sD}oC2w4zgmna=i%ip^Tzug9;zU4k7~{5MH8hklexLtDTq~t zWc=ZUk9e>=?bGk6MphcMq=TKTexk(wVy-0_G*0ga&&gvim&ZKxJM5*=oIqUFHI-nblF#Y*x@7XUJfs=)_KZo~fatKIK;mxu7Xeel=k2^2r2PFn0x4-KqzR$fz%OaZ#e!g&SymIcjwG)9FY0gmHn zok55D6#gxOC-i>c7cDQWJku9N>?U0&q9o|aVJ{A*@bEka`I6BrTS$g~Tmi_-w9{Ap ziB|%iiGt&)O)~n;bwK(-d%nVtMB;LsDbdF_f{~fa{i{IV5%_hzs+em8lT_$U7?Ul- zr!1}wW`F1x!2x9OsbJC*`rus5%_`Nw91mPtzbNT)GMG$v%M0drZSwm<JxB4~Jc@mNBh*ol$=qLW4O(zTLhmzBG8jajg09aKBgbsCFQay$mH3t;o{*|#XvO1@sP5)ttv^#pi4c<#rpPk>O@Oq>ihSxL9JFh4#Xk(qB{qzsxlem6NY^r(X9&9qew%uh5w8sSKSKD# zPa|D$p&#?a8VAanW1-Q031l>`Pc${GfP4f$e44zTu3Q;F{AhuEQ?KO}TVI&NGuzwX z_AoC`*xYqJ?RT^cxJYC1L@_GC!9>cqYjC$L?u62(e@}4H#_f3yet3~FY9cy8I1#il zM^R);Zj{ z`H-o=HVXY{xqK%)6V23eXWT6?_bqN4om`TFt6)4Sf!+w8MP-akflJjI-b1-czZX5lpbzRu2VmzsiVTH}` zV$3k;nT%nD0{8r3QV}qTVcxsY4j^N}c{6S~nD4W3ldAcEs~%b5bRn{O2U6B%_%QAi zm|qwt!xp)q0JtQ_$X}zxK_Z)>GwvmrEQ{MlpKPuKF4?#cgu#W)@MTOGN^b<58Mey* zM1rXm{M(**aqP=x7>uh1CWUeGo<)31fJ+Np=jP5UC+zt#t}B@SragMLERt=R+s5&% z@YM0jxP(&1s7WSa)+=_gzJvOBvyxKZpE`8*WXW?H@iY(drvZFh)e|oj%uwa#l|K#g zj;?|Bk#;2PPhUd!(RLWy&~|-^vX6ElD*p5(Zu@9oV#lAp#C;#_M>PEDOFZ_`PQ(*` z`Vuc(7)-p`-@e2r`CMz2=&0iB-O;A|)5;pVi6aS_<5s_4AB3ff$ASnuX~Oi;4Uzwl zJHXlgIExIr$*Sp5+iHlsW@j+WpC%}O_9{h21Dk50{K=VytSI}_s{A<`Z9Q!e=7*n& zp}nhCA$yQUNL9iDuovzy^LkIT18gYlPpkQEEbc)n;yO6WxZ7|KEpCO|tA;7ys(XdC zEkHOx?T88wtC{l5+j4d9Ke41d4WJ!Vt$lr*GE>-v0NPPilG3V@D)5xbJcj&&NM2$K z7(elG87EdPE=xXz+C3*RpQicvz($5zXk!(3-ZWpWS^*b&tA)2TMX5o<8|+^m7y77$ zPi4513|x3wD+w3+s-En?;!0d7h@pD`?XSjoPH-pdaH0Oxnz%4njj6lJ9WQ`*sw7mX zjtfIn?}YNNnQnvkhaQuCefYLHH6gGqa6U77v|OP} z#!L5iLcam+uFHSdRnhRjK=tV8S)8zFtI25n;CLiSy3iM^zQxOyC4DeVEex#~(?$5F z7AOn9=qvGMvZ?plek}QQuPJ0Muh9fvMK1RA}o}?-4}{ z5SC+qLAzt=3+NpVY*Mx11SbSjaQ*`>dj@ZAT@n_ysh;EZrjc8h@{l3`N?H&PLwfBo zQGj-;o*y;$3gL63=mN19&;0l}wzZH__9e|E&5aKB~<1 zkD;}2>65m5KtHYe4~ehJ^#RbnFm*b6EEY%8r@W$o&Zz#k-{t4fTmZ{0h_{OA;nP?e zoK^iD?M1-%0yu0!Bz9`hPX18iz^%sF4GrjHGN;k)+qcX@vlKMfG32+sW}(KnLZ- z*?zouZ2;eyPtUWJz|S?+zy70W?q@pq`8oWMbfbwkpg&dr%X4dSXe)sIxgbABB|z6z z|933~^c#Tt7Q`1mSptysrkM#KZ>s*SR%3RC{IsXxsV_KJ++{)OI5t1XTdKeDSj4C# zs8oyOHZ8+Ec}e>BD@(!8J=NboT$G0P;CoqqM9L2RinZSNRsX&rq7fPo;42H_OX2As zm8k*ff$G2OL`@#fH2@A+5KnnNn-n(Z9xey`3398M566E8`rv{1_fc2%s-i|DT^Eahq|V5-l>@-Lvp8Rseme`hR)20w+HK)h#C} zipd#lUIx%tNT9WeoE!~mrbY5-HT{y+i8D9|9zb8K0f({O&aSTo^y<=MdJ#s<|=uX1wY6Um#*b>YDpiZ%fWf+>EwHHKHH?4hRJ6R5O zy+w%7L{4YJh#GDlzl7J|K|TTCj0utSTklvmTg%OJm|G#jjNXIxB)4AM&FC`1NqnB< zAzt;c{sAYt`33YnUQ!-F8@M@&j!h)d5S-aWwlE(+8@V~Il(CZ}2x-~GF7wz3sfnA< zxQ+NY8H658WF5E|Tc)!oNCIfOTj=E@`S@gFEP&~z1)1&a_;3Ji=9Xsznk%;a{5^nO z7Q`2%MFC+e%mK8uTj7p)xdoepT>$kPBiTKXm`jSN0NS`Yo;M65&mg=rMOqZ2b26H? zEX!(ddpAduugeo92qWDBaM^NyG;N13-D4c==Ga$-J^v25hK2A3XsPo2&w9(V!z(X2s7q9LWl##rRKJD~m>vKDX*rCMiKQP-~BQ z#0VSv8H8EQ#zP;Gb+EaY*-br7+c^MCY0nmy^YT$iaprr zqF-Y@Hi$OY77UFbiUVmL0>=q_&10?&+NZvJRa|M-VnuoUY@3gzC(dZC-No+So=-|b zlxDwe@Wszg!Tm1UNQBa?zzmkiU5c_KFOM@?laMEUTB@~3L=f%iSkf2=1KF=2+Dlos zn374j>tc52ODqq*z{z-#V>~w`*<1M@&(v-KwaX+)oJhgGfva9P$Z@RRpNDjX$$zo< z#j$7fg6IIxf%7ZzBzg|%UrS!;wzyYy=!1jVQ=Lji{yQ3eeWq=^tCM_1zc;iqbPB2g0&1dq3vQ?vv0fS#f>8P>=|X7z1Jg{ z7T^mK5z4^nJMgk-TG0D;-$*{oQc~f)zG;GFu)7A+BFe3xDAEC!yXBUml%x1i`v_c~ zYD$z%*TlO<;_|u&!GBTj;rSD}{Z)WASsy1ulgZ-Vr&An|<9$B~iV*!Dr0J$C zYR+DI0pF9|fkz|gUgX*`rm|a`rGZ`SGZ%#+B$y^?Kc4M6!`^#lGm}LpLuj>?& z4`;B;l>t>ZK~t_DSQ%?mOl|?GEt9A46pNdZ1mBH(m324@hE{3_q%T}DtBb)jU7K^n zo#)m!kd~P;oyknhrCUO3kf-huWFr$;$j#K)f~c%e`{B6GAk=}+c^zon+^hr5h82#3 z@!^rf#|LBO|D>&B&(`=o!iSA^ahP^Bo<<6$<9SMzRGMvCmWH1bd|8DnZx;25!SmvR z$l39sI1tn{lvISX&1Io#_U*%D;FF3}pAt-(Dj6&g!E~y!pA;gkLG?08%E#8sD0Y`P zG(&m42}k$<%=ur4GJ7UDN7=YJoU8$Lz#>^#{@3OL<;Nkhv;7 zR=J--Sh{#ML-~iZeaEZEZ__*>uPk!Fo8=@S9MX45~dM-H$ z+o8eq2cKC_OY>d2&X9Uqa>m;dtZJ^+qSw?Vqj334Cb86%C5@J3`g-57ud4I1vI6=K zmiG&EK!bWV+iyhVG1hx@PN@&i+%ETZ_7882(ww7~Dh zlZAGa6hAl~b-IqnpSAnJbia0YjV~{c#h{nX$?THTX20k$(N8(;QDh8&4XSv7#KAqbLeUHVw0!ZRma~n#l1`VBL<73E+(0?}> za)Ea5)ed@6&|es{vLt&2?X9K9vDbEm(p2BfRY8Q!r-P8_dEt04NVD|7c*u3~v92BM zD(eYDXk}r z!5=sgE52rt@uoXlpWo;B1Rp>j2W^I>^EVqxeZQkb$8fR|+D1zc8^JQ2S8?S1@(~u1gN<&DQA8c_#d*@X= zt5=9EqsCINZ544apEW>FP1WqF*v0x`^rR6qga7B0;lzJt+I+sRNLJ>dK75CK;s4Qg z-r-Rd@7tf*ld>tKkw8cw1xN@90hWa%Bq0e2X*5GH{Z^Qj#f)$oAgvk!pJc-)|W|zsSWenjm z$NKkl?4^li48(j1{TpJa6dcPK!exF~+ObrS83Qr7&|4AX$di^agv*kk#|{^o%NU3o2xeFXj+QZm z%cmpapRaE&V<2t;n5D!y%7d0Mgv%b@yU*ckuIO@*uOD&y!F*_PY8gW~maYZNN+QOe zm~+s7A;w`FEn^6m{rw+)?by8r%NU6BLgn^{(|OV|hHyE$(l6-$X{{d83nb14CY?C0 zWxR}Gqnj@`hY7|HJJ7(0>I|a0rc@*2X+yZ2T!mT)2H#HOn<0a#kT(MfO`&Dd->ovBJGIstGSZJ?;ZGDVj#z5>3~zM> z_ktPfS@JtFbk!>{{H*x~jQWENWDveh`um zmUUG=!7`rnK}da`t>RCED*cXPeMYdfRf+qO|1_2xvN%>_`MYrZNk8QHGqR53&yePh zKO^oqVzrKUgybbS{xr>X{MnS@_>+d^YRIdx9FXYvbF#kU&zNS8KTQf8f12hy{`AXp z{PFqN@#ou2&A2)K_@_Jm^lRbxGa%dXXGN;x&))M6Q%~u1HXU|IoeLd*mTH?vec}iS zJL35Bj{Y<2s3YWwPaS{e8vG|k*2ObQ9AJuUqW@&;Kac4@PwGGPK$GDS`p;DTXO8}} zT>sgu|LoO&KGJ`_)PK(CKUeghyZVni9w2gFLHbXm{*$2pr074n`cJX`L$?Ju`qMgO zfc``4BUvibf9C5ytMs33`cJw3^RfPOLj6heoAn%@opiq+RY)^Gnr$NGEWd@SN;LP| zgpm1`_4xQ*u*~)G>4OGk1k1e8iRHLqHs4YzVj&dx)OHAkp-n5{l4ic8=TMrLweo4^ z5L$=&eaJCr$I@Wg)~Az0C<^U>mx{y|_m0DQiGW}^$odJC4F-9Gal3bs2oYluG7HW! zn4WhzXX(Yiynijg3uP$u+bL=LCHU#}Yiv29>2u^8%FS7L=!X`C}Ez4mpkcpK_bli=kc z>dBbu(|$5SpY-wKKJ5aiE1J{`bS@eZTXrY#0vq>!SOfxRp~qml)4Xhiu31!z&$no9#H$&3I??C=?!Swo@>Tce@d)DL$MwJ1W_`s>^sixU) zT%+f$QEuL!?o4DT4L~as;tU;E32LQIZ8xiXbX(CK$Uqki)Bl+lGm%&wH|xN%3^5Jr zD=vh-N_ZDv$G~XoW=(#wwb%mWO%wZ6J=y$n2DRKoH|xd&x#AeqZ(Rw;lWm)uSNXGn zTp4ZxcxXbZ#i%FSNp4l=Hg^k3#aP-6NPiRKhpmnamK?YEmE+t+`}&?@I)GUk65HGdHKXUpE!}*!ufP$y_*vvo5uv9Sknf#U!_i3GWobS^N&^Pm;MrY)USweK0pyyn_-6KnaG!v8-P6w~cw0lIR%o zTjw^ML}Tcwnh_59iQ4Bow&5h&LwZWfDv2RJ{$5C;r?u(*6fq3iXsrv|vn4T*`qut9 zCXu7hgY>E?k8JU7YsiB<8*FUFTMQUGp}l45$1FXDmjW2-!LPe^4YG8y` z)r1bMH2GC;RZZN7Rj4K(bnp=#Lva%r&Qd@{YF3k(*q=HY!8R>PzJ>8fHF?+?TYwY* zX>Vd}DH&%q+4u#P-1LJw+$1^>p({Z1nECtN#1jaTb3iOLDNaXIIWeB?;}rEKh!0GP z(=w~cTl)fdP&oztM`F5BR%SI>zCM`cdyxKd%DS3-xvvpt5n75(aNtylnAPO@5BHYLYiJm80*2 z^qwi3)#SHnA*z~u2koq>JF5wuLGi!Ugsv8a7adyZ@RE9}0~>{E;_Sep&te-pgfxY- zY@~C=vjMXM`}pztqAj!zT6cC}4E|3CM!lu*@&;ada~1mB5IMs7!f@2K9hIm>jIQ&{ zd^c_pYk_R0RKMjjspN5qdyYhNNqh+D6Ozx6td^e(u zCxBvigVfs@Tgfjc_UseMu_r>B?u@Paf_dRj*LM@+Ve$${Yn`!`yde7I&NhNdYCoh8 zov{*Iuke_VtmX+}kfKOd z$F1bmzH2)~aO|d#GR@dSez>2F*lUfy?|X|52=1w4Rv1}Id=jmIHn?k9FpQVo`iAwBMltt-Sc*vy!k*HB0!ow0R= zC^!+v@*GGDow0R=NY028FTvzDAnkF+))ivW*CB#R>N7|uoUwIx9%hkIVZo;I`O1+H#(PpT}lHvL?Q?1*w~i6vljmw(2ZPa)d-sNS)nPMOlgpu?AAeft`&-SyM-d z7lqJszOqzDh@_BLciBW)nhFUBps)H|rxJC)k|7Gp@~jueV2eufloc39tA>g0DB)5# z%do#OX7FVa{fO9aE7cTbk6E9^+N(0?b4`X8^orQXhRsD;M{D&BSOx-VkBh99h{kSc zTw9ctTJvwEiBF)PFqwUP=ICBY**NRVztrLL`Wxgw2^me6OvI1+JW1Is>rpJKqsMsu zFg^^fB2t8zd)Z6g{36pe_4?^foBvbj}DU$qL#v}3BjFF%&&M4 zeGtS641+vcBO>AB!EHs^B5TJK54K<~lx13VBv9HjOq4D0K6@pb6W9&8Tq9ysrw>pA zSMvQKVT?MQ6)MVB)hH|X#QbkmyI)bpt3&LK3|z6`0tH9E+3g?1-7BLW)E8yzs?nmq zN#b2g*;;uS;YAU$ZZO@-oo|UM$q^AU!Y5`dUf=1{rjih)BVb5Cgsd0z%y@LMlDrWz z(fILZ3n5zqD}u9l|AT^^WKm*$-z4{jqBp{aYE53GlGi{Z-}#xr;(3V6OrSol$<*3b zt|>!aD3r)2$r$>$A$EXzN0aJG&^ELdS(*~-Hy1AkBloYNf3F!~+jk)fa?{uzRj`RF z+_)(c(=Un#z)o) zdOI_E$?x|QA-9STI6|-Z2`{Qfv*f2IYauk=^O>q1qIB60`0PuhtZa3kK-G!7kuun5 zb^;TV#Ykx-rPhQ_!Vsl}7~!q3{||yc(UN-EpVpH(%%ZayDXSa5z8uUA>O9CxM5j?g z#fQq>AV*or98pX!`9{jC#(g|D7H?KH0cMZccgaOKenUu_A9lud&kOr&EXZ=_5$o_If4$bm?37@Q^4LK`0_ zEzP)js*;$2;8(RI%?5y5#`MN^Bwja-Z5=3H2eHqj_z|a#-oK8N84+v#OyvicUqL_b z!l(z9wzcD$18fz#Bq>GIf&mWz`M^lFM)C{V+}u=yh?H%8H$A3y)Qka@KxAFEn?8sU zi7hBRzVQ`V2+q@q2%9R;4q@4Wq5Kr_X-K`yn6}p{c_U`WD&42$32$US0n#)x_NSJd z>&ehN2OcTSOF*40W4W+8G$$G&rF6rmafe`MVF9T*vobjZUe&AgR}gF z{RmVH^`3T=>=$&ie+%;gC&6vuI%rs^2a-{^?0@kFw(`OQNn%RjMw!eGNe?8WWU>Ey zt73RH1~~M{xD5iginx|FVEztC4eD7OP zd0dc4)Q{lnaCZoHBuY=3qp)r&Yi5}Fq?x#?D2N(xnxhx+qhxxG7}Q(+0zPq>U~)}P zJspmcjjfL@!#x2!9VVs+^uEN9fjAI)(i|n@!e0B^!+g?A+$=D2&9u}D_))TU*lSDG zOC3nhi@052_Ld~vIt3Zg0pnR1}9>O(!vdYZ1LS6!6URJ|3=PQuWq=-*B7&(Ej4xLj&P1d zY8c0`!^d_22Th0{ z##Aj2nR=x@V%N7;|20yCA~=pDIu5nKP_5XbXJgSa-~Y~u%Dn2KIiS`$4aXWK#dPqm zGuDlnUeJfTFuVhsu#Gs_0EsT_Gr~4@hBxF=YqxutqF|PH2a-VXaIrXSej<`g85{InA0en7 zRD)jo5ylgH@bkz3TvN(Fq9~z3vn(U-urd&0WyWgC9XU z;*@Dat!TVnu^OPS|fiE&54?bVd z!#GGU+Iy$p-F=>z64Hk7HD$OD?S@iR6YCDBPT)_Jdiq~e*7D=$_%&s1KYl7*Q^u(E z05xTu(0VVUvGv-VD7w{@@j;W2##XoR`5D;mQMUFrHX@s22JH<}Q?~Ja@KX{W`Z9!C zYI&dz*2NF*7mYx`Rk?afBCwWTkN=46tGu=$Y^z%VUVBAS#oor<3Q4v1_T8jD&+!W- z)j%hu)B6CS^)FY)4lzT>w3@P>F@9|W5n6`(e{hzK?$nNZvXp4_^~o1@1ZQYTjR-tT zsEMsPc^*;xXWKv|)QI;ajVg|6*ciUX?29hmA5B=*%TK3JFuUy3~@!Ybm1eDrx9 zd}-F)cYVcT>_%|8Q~LbFEJ!)lev5Dgh~TqYlKf^QKa(0V zF{Uk`7S{aVtBTtI{&9iuI>S#v&~v?qeW2c%48_^01;@wF3E?%P!scM|yr*Fg97;7E zdTSTPF$GJwn*uQ3`s4fF+}?Tt8bV;GCr%_^-jy7hMpfyFD!H7GX^~b~Z1US-T?;@>;~?4mBMi9lC`gWKBfB$C4c(xACT6O}RE=#WfX6{2&+L zohEB7=^q$41M_X#-l&!gt}?A*D(`M2|0HXmSpFu9cz3Si?mu_|s;VxeM8Ax-jf5A) z?ss@ZTj3Rsf7{$z^y-b7ECte_;nfiTiv0Bg8GpS%#vco0sB4}j7jKEfCTjH^&FWsU zCw4oj;yLNAyYT9PxU=NoutESGJf~5b!%s9rP+tz}=(&i3{(4v)LCZI%f-3ObOhJ3Y zEeKk~L4}_CDQMro1_-LTye_ELo?lRqQC1s4)j6n*=T8(=v@8)pn<`_WlPvQ5gM!lC zQW3OjtmoN#3VHxPgdeqpSF3@!*d(6?mT^0?!g}_-3lO3Otp{#iZWQtgX+f3~Q zX+tPpWx3Jlmm4VZpmj2Je#4l?8Apj!mb?8=zE;^Rkso2*D9{rq#wkkdm}TIwXOY@I zUhic!-j_*8efNi`n*^`cN$W$VK=FZbyS5v?m2k;7kY4n<3Mh`>_>$krTTtYNCPLC? zwelRr-;B#gNBsGGE^A9KAJ191r|5n%4f|`yNObnU$Z{*bzgDdVtpn{x%g7RoSH+4@ zo1gVBpt0n!`5R4cwGpZCXTe#9zJu#3r34!F2h|ZB5ZY5K@{YvTwoS3?fk;?UZV|Q+ zYbx%fQ1OfdGljTulmdTD))r=6bV_Zjz68oTt;#8k&^sBn2qAs=={Qs~7K zTdiu8t4N8Ml7s7aU&=jTscZ-1uRdNy+@L6AwJy$T2q=w6X)A`r${|4wUPRqUE5-Kr zrz8c&%Av+Z-5`Nd&BV%a*6UwIihKmOaZ2w++9A#JHomVex+Az$ zOKNv$+LBxhq06XPxzaoSl3hFxV37%t?%3^)UVezh_XlpCjOH&0YzOo0r?~1theJemd@{7$rl$Cp-3qKEBMt-L*vA(zF)wqED>s6jb~YI(YKM%FagdmkmW{ zlwo%`OSSj4nL|=pf82b=Mo&EfWyZ#eez2 zdSWYZb-;II-&Z^l{$%`L3v@>W#{w%cOq*l?hgG{(4$M9pFbN zrJEr_?*0%g;_T5fs1D{mNt>~0zCA|z(j-VE-S1RM*yClr?d$RVh_JDf>j`_Jv^IW1 zUBN6szDp_Z>4#ue*hG?TvR+X_nxF`~A(cDjVOOQFXGq^^_51{#h~sdVv@B;K3H3u! zMRl^u-dq~_i))F8pen&0%UpuUGRe?V`6sBq@Cku%t?H&i8Puwd560P}JZWcUFN$^9 zs1nxF@!1JGUTGSFu>3(^`e47$kd!#=_gTZJf$vT<+0q=}omlD_FWkg-_mOy7g)1NS+;<$!BTI=bXJcnKKJVx^43<}E-%2^Q;yxoT6+PLQ5-%3XRz z3VQ>Kb=~;_ZhXTbjdjYmaEHupw^S}ji4@ClCSHNG#3@($+F#gHEWTT2W7Q4n)-Fi< zOu4$@jnm)EVwIBu#AgWpjwCKI?sUU)ZdMoeY)io2wH|_w>n8NO&gffj)))4Ci`9K| zu&9Lc4~0|F2^L{*j%y<9tu5B~>NaNEZP1g6QAge?qmHl_St?(^O(n9u2vR3gu1=@Y z-pOLr%55e3BX}lBY#e{(l}?o89~uuN=lH%Hfk>yn!^Jd}^l}|T*FXwfYFRfbQ*4-r ze{O5>&t63u4f60B9@U`ZaAzo10;-LQRh@7qqg7EzTPu*)KXhFK< zO*ijID#S2(p!$YkwfGhTtx#8&zIWuJg-xLy3b^q=6~+Y8_-dfEO?TWLjh8;w*AO;K z(x?|NeZ8LT=pnF%g;&|B_jFFRx?&mIPTR;4GPa zTYfMx7v4Db;vLJ!-)RMJUG`dT%d^jUz#GqAYh$l~w~)Mz*o%DY?X$ycXD{-ddXc=1*^7L4U(SNJ344+6 zz&W+xO=2(d^{(6o-emS7-!D%V!kfZg&-oVmj#xc>@GE#BRfj_{jYOgh@F|UZD*zwk zmeJM$3+T?&1$-(%n zp-eDN#pH_th2f$u^ah#{w%4v3kn0=Oeo5o`LII?tt{BR)ZDq<1~_%N*R3OtWlg*3NNy6w^Z%ZC_J z9Fyt>GTguS!$fmZJrvL}xZ?zu&{$EtcYi(60QV*+mbLheDv|Kj;u&_p`*I+1zx-rv zAt|B-xpNM>lc7LS)KFITp(8kIF}4JtNPK*X8ljjeXujp}RkSJjv|RZbW5z?MQPupQW0Jm_+Q#$dJeJOP3{kUG|oXnI$jzj^0cY^N6jk*ivGzw@4=TS$U9I@{+IZ+hO8e zVrwdP1+ll(vGtL^GfQ5(>?6~9;$zIj#wqp)vD@e6QM!F)jaR80$V{FjL$lwzXovA~rIb%VDsLSxn_XUb`GD{~nYlA`oIJR_x!z-ot0uknRxK zj#=`O@3C|BMN48IS8P1~mqw35ziyX9<#=YvOTN|zJ;exOpHgfmVh8NRhBtOuDmO7p zUh)|m>x$Qi?Wx!)#2QcIfr?!YlV3ASUh;V^PZOUJ+fT8t6I)y&iG5DWC3I}$CEtPf ze8hcXhbZuWg@fWC10NnULqDDmJy2e!2i-Yd$NcfAxoGgFZqgIaTmqJ zj#F$SVox++8%E0M%#znGN6Ib1Az}itQxy9Yu~&Y=qbIu@C3i4OUh*A0)vM(ldU_9k|4XIwhi`ukLOKj#m+#9jWNiuL5l>>Rnx9+hV;f)Z>TZ+9*?9dqvh@C8(F-ur-T3X31-pQ{?UP+M+wLA1OAQ*bPw@N_VPUz$|&m_Y|JM zyhQ8~#SS8N%+3U2r^yePB`^6pU&#^W#D1aJCB$~E5k)LM2*E6Q$v0t5xVT8{NyQ!_ zwrD>_RJ)vM9fJ|fMmOH=a+a}Uv=={ythOB2zi^h*h$4TXcn)T^QQqH=nP$)nOy+6@ z=0)SevG-F!PPlC_DXGA$G&bTDda4`0LceV?Pgh`88C4EpM-4EQR^j>=P91x%3e0Nbv4yReu|ZEa znf?`+*Nh1V)w$>hy_?Alt-!1?_Td#H%4;f_IlNn!uS!eX} zX~XG!0R2;wnOcEaZ*-4t$jk-kH%#V*3d{y06yHK1d%RX-LsmF->@QbfHX2`D%3>xF zdWy*`sK9J8LRb4R(+>KRCbP5xv)Kq=rt0Bv=#xxlbp>XN@n#!;PG>RnwI;Kn0<+aH zCU`QlANpaF*;awsW<0SEQ$rlrIq1Kc%&rQ|>&Ei0Ycu1H4ln>t9s9lt%p1m+dBM!o zh2Gd?-mAcDH)?t*rVaGRP39mm>WQXZ?l2P8+Bu;?CUBGh;dK+8@f&VSI?+^lPo={` zeZR`BlkjVS3v~Jscn+p7DYonn5Ok4IF7<_jZ@(&nE+xvPzC1XB=_Xw;luLaPF!0+p zf-Vcn&8hyK^YsNy{>!DF&==3J37X=UOFeC$Roqz6M7~_=$$34ipP=b_xztneC;R#c znv|DIJ<*=-V-+;@E;pysA!$B>CfMatPnz!})E6|3E|+?0d}TyyL6hZjsVBhp<0*os z#N|>?bE_7$5;XBGmwGbW>9$4C^tRlbvW}b9NYEsZzpLhju}e%W`wVcphJSqA4WOaZC%JpBF7?qF65V>7fmcI8q}*G7gU37Wu_o0GCmU#AJ0rj<)Q^*VUCk)X*~ zxzrP^HxAeYO}WaYo<_9`Y%FMERW9{psY9@vpy^Y&)Kj8MJsJy|B$b;JpLUrAf~G>{ zQcrL0o~a~g!c#8wBxcQo1VPgp*yEV0+@0htX!24n^@L>Zlt@8Slya%39g`0D2%3m& zcN;WFG{)_UmA{7J*=$UFEn#gU!#uZarTYLuwc3b)Ch|$|(d+PhWj(%`LLryZ+lb`| z*a&C&1uaEJ(MA1Xi!VferFas@Y$Ch*jjbIaJ^^>aWJe|AKF|)gqIQ)<6Zv$dC)=b6 zNy*C__<5i)b+x}k(nk}T$Zqj@p>AUCCM(~}Bf18RsRd_=<@*T%8+?HE2&csHRFsI< zEp#wVWQ1|KCGG&o$3Qoq@e=C|r&JGT9ZI&BfA0)-+C* zVeQnbP}i757$r24zEjmi)~a;q6!syeboK){{0O9sEUGpn-A|CA@;rb`1VvFI#q?aK zi5wY4$BDxK+`Ea0Y9jk_;s3b-H(}_oM8}Vp)euj8>La2MlmKUG^$ccp;F^Qb zSAeb5>8f)y$@0(kL_`?mH}AehyaVxLIJb^i4}yEVY#aT!7$!sXMgL@a`AaF!+AQ3}6~*QHQxZ=T3TcpFo@ zPMWh6hOek?mcn1KWH7+t1m7XfECsrwje$W*5mvcd@&dH~{e& z97=)q%Tl)S&RLWAfT$EM)y9Wkz}!(B<)gSILscnIi14EP7Fce*FGNFK3g2&XDTQs* z;zd0KCBs=h?txPH$}ELOcOpdr!rPnLx2~lysvyrS1q-0zaAOHR<5CJ&YT?aURTr+* zjYTU2vI=esp_HdCg)McmxfI?=4iSeT9*09I{IC_Jz_!spI>(fv@S=1USni%q6mk0K z&TV!%y2;DZg*Spi;4HU$;^^KskIu(EO2i`EW@;a}9$o%dh33)yh^*QJEFt(~m!o@j zfO(F8`?W3}>;s(u>;;{!KDvv$RZai#=UA~G;tn_*-IP46KV{qKA8I-s5S`;iH)HWh zDm?g69Oa|98N{I!C`5QsehVxO<~DWK^k25Pl!D*1W+DJV;c%8{Y`P}bnx)YAEg|Y5 zJlWJXyOx6Z$ZnRxMOab-@M(g#yOaVwTyUJ@zi~Ao37rP)Wu2}rg-_papw96;A+|!? z3x`s8+aKqcZKHoE1v(%q1#4L&aRSU)#Zf+r+YzYh0zc&NqWsRtGg*{)g72Rzar?v@ zqfiJON{VA~*Uhoc$j8r887KHYGaUO&*#TuJkPDeXF61-vaGEcJ1-ZCE_avZa704}) zL2hwpJ^ERh%O5-TpTYZ(hZ5cDmE| zhLdDRpK!eWAt_lY$0NoJI5t=v-5;{v(-v?cmgvd2JFs>qkH-FN##W$vH7@_4;p`82 zIl{22cp=sAsxTgK7n3qg8W0%~_rYal>YVp5_Q*#8r-cvU2dd;nfA52c%(Cf?fdEo|0KZWzV+lavy@tsgq@u zT5(q$MT<|P5j-4jtcEdK(1OlnS5En-Cj~z1cjwiV4wkdVJ zH0MFwLAWn8Cr`D&!1lhZhC`VACd!(KpGYOsbRD9=qn0D5o8#k7nLcVkP*dsa^X+MR z{`OjAmS~EFYH7Kl6?{GND*9&SV*|gmh+3jsQ(40r@+MY*Joj}VqD2~`HUx>1H?Z{` zoFya)1>!*~TKK2~mvXD`684IEVi-loQhjMkiC9>{i@b{M6Z22GQg z=?8td$ta^Vv$v=o8}$x3N>uA-m3&}rjdQM z4rlm1$cvhmHmVS_yNKhR!{*qG;CLo7Qm>-jGeFNa)#+3^EEAbJT{r2Dyn^SgWlg{cvLl+K|$#}o=<~5 zlNe=`&gRsGcu|N!ehs8eItg4|;cc6>ij>D9yq%^bd3Ko^%tW<*@NsKZ|%ZtleaY>~>%Fs!Ol8bzbI~r6{S%&S& z6W%bY#E6Sgb*eYWp_-PeC(%sp#`$dcOy~@w`#xL0x5+mK<1XaEyHzF^X>!`>^E-I;j}`Zkks9rm6W)0FGQA?TkHQwNDUhrRRP zHRpP90n!zn#D5NZ+4ZY%hL!iC>xWZ^;5zJmFv-eHGW2YdaUJ%$Jllwk>H_^4Vw6#O z*c*^zXQReInxvEX&tdQVb9T;fBgkEvmdO!xn0m~L?BAG~PoaNfGR<5VIR*1JgxrAs z2QkVhokVh1L*zjAe+$(4_!Z;I_oK1kMy_}wJ$_aLW$7r#3(zpye+Ar$bj4$Qn60LIqXq( z7(Eo79W?58RUY=nfSj)BM~1zpY>n8c70@@Cj5_Oj%-WSxhnWM=kC}`rV9lhY)ZhZX z2>mx=l;K+bLEySpj(!<{1g-P6#d<5x;PQrEA8{!=H zZu+WT;wH$4ns%H%J!Yk}%VDFcyn`2i;Z$RA9rg~SyYuNwg`P!>GD;^AYg4DM1k%$w z3D;q7lylgti-`!;m*+vw)AS?5Ud~@Nnb`vUO_OmQ_Imbiz{B3>(7z@|8Ks9k%OOu5 z_O3y?t&?ya_QvU9&mz1MF;loK)S^_iS&|^);b&RWjDTR&7gq}-`GD?qeckZi8yQd)a z)Jgp37YU-5AV1KwjHU|V9OL3>j3d&YXjc(<*Nj$&qM6Kz;X#5D3Va`Lzr#7~)MMOx&0C8^$PFPSkxX^pImS_TnD^^3 zZo`2b9^;+@*;mt#jB!U_OX6dh0{umkQ3b3AD32(`tcSkCWK;obW;VV?NU@GU|AH7* zz*;`lv=JBZuaK_mBt(fG|!gY*8C#eUYQ!7HrR9=aO}L~gAHdt+aE?=~FYTzLCOOBrH#~4V9~m|W z*;><&jBztQS8b>l^dTnWI>uFfM{rN_0`!-OQAX)8?w7?`+`Vmt^tw*Mb&Pw}ImZ36 zA5YU@)Jc%%H2ugJH$6(7zWdPK(Y32n>^jEf^|$lsi-2C67-f_m<1VJE)7K1Aj!wdL zj9cX#CgVEBJ#)*-W85z2`-o9S=`rq4O9qc| zUqSj-C*eBAt<_^3Ufm7E2!Q>J8jGw{=@&HN6PzLPrOFVa;mSl2`VqFYg0mdU#u!Kc zocA!%*Npul18C*2v(Ms^kW^bHHV;dcRgG0E^Tct)DM4UQGlB0A;$%9Cd}hW&pJ6g* zU6{?2{RG8Y4SfSK7hL4~ZwHGj2zejUA)SOM*@_EUvVu$!hV8Xc_28{$M4Yv8hSv~$ zPt%Uds+n(Q6*3ca5cj{})RC*os+so}sj)Z-dO9)6C@qh9O$~b;A$8G7h?4W$bJz;U z9nHKQf$)YKFZhLUhEqYlq-lpyn#oMgWo85PohGA<(#%5}Mi$)rJqrCxVw6!@9(P1t zZ(fFULnk3heoSh{R*+E`_O=h8f~#TddbKWR72&=oa$AJGyDeRc}>fCct=Q>%mZ)LLLWf)IP5q~*?3q* zCIVM56e|*X3^B?molOV4M%EP(G9b0kNr;j+Pd8;N8sY8{W-U9g(@m<3GluN+VxtCt z9I0tXNol6bZZ~G;KwoMys-!ftx|2G6yP=m8qm0t>bFB-xq>e-SPA4HsMh~X58^@;) zm*KDbWy2e9gwIs1Pon4Zg{xT6H8b2=gN=%T-q2)JMc2%y5ly(F7earG7-f`}|FPHR zirxp(Ae{tCYDxrKK^2|8jC%(i_G;O5n1`4l)8w_lEkA^bWk>NsJN*%|ZE!>t5uSqO zaNCi_K}zSJ1&;LqmOm!p(Wdu8JRxK3%!_!L{Tm>^{Rk zuxCe_ylZ`RBX5f2&MAxrFpZ$<$VPoyx8&i(Ab|I)tj3XBWl8E7%K0yEdWPIkmK?`k z5^3@wdsAeI=bozY{>|PjS<>L=LU{jSZvmEk-#VWGzeQSq#Jhi%Uy##HGdEr&GF?71 zBMgWkt*d@!amv&JW9&{v@97HPA5yc!? z8HF2zr?Qfd^BE{ZNp(G8>dExB*-5ye6eZ;IAm(XGolKu+?s|*Ti8%dUl+Xgp_qfv^ zmnvTkqT*Te8P;HNLDcKFU)i^&H5Fnsyn+Rv6{?bJ?g@p|3HSio@6n;r)WacKkEBF_Xz|a2|dxkJ)CmCKfQ^+`Q&Hjabq>~)c-J3hfjYwKP3BJ3I zQ=yYww;MZxjD|4L6rG)9)kJ?v!JSiB0bmnB&;EbzB;R@&oun_yvMQX)?a@y1w@VS& zJ_>RR7v$eN$&m}=d9gQjl5`}NK8Q3(M|5J5El=+evxgy8H|KNbN|d+>~p_Ay^>CfS;t}RHz->pT;=5 z2*Pqxbk>e*J{Vt7J2-`R0DMf)`v1?hW2RSIVTngsHi1*QJz6_<%!?D2Hjul!Apc%F zrWIEPr3{f~>xhopu{{y5L-KL1g|eGe*Au2{N5t|vV$W{Wjw2wx)|BLv$+ctU zII10#&}#Y4#~I=<`i`)`%IEP&|0HS${Xy+GH>A4wBLaO#;BnmbgtPqUT05?A?I7&c z#aIiM0;J8qV85%{L0lR>O^<1jkUhbT{ulO7V)XQ^f|&G;;7Yp zZ?@cfF9DV}v99#S;_+Ef=jsH-$YFnaLrboDwlG_CUxU{Lw?NyWQxJ~HVp_O&*}oau>?O>t$OCUfh)jWvp0 z`uYh=1gQG|Le7?n3ut*E??54a=$!amZVreK>-)IZS#sQeW_*L9o<4eHc2jP4%a z;35!?ZPcwJ#wte8D(nDvP9Pq7re^-LL8Zi^4?%l1Eeyd3C3!V6GBk2a_-^bW;osONQp&IWb#rG~0O{YBchjuig8L3RD1 zH8u?ZSD#oqizq(bpn_YWLFGc|s1s6MK37t+LG{?0sv6W_P!s=!bT+69LxZp#1&}SS z*uOO>OE1h?4uJc{WdEx{eR-vcYEZuem8hqx*cku6HmICRVM6k`$U}miC{fost{POp zD7&z4M^zjEVx*=d*MDwMAC9C3MG38ztu6&P8kBJ%Onh}34T}Ca8r0i+F(&Rp5{D_7 zaX1wf8r1W2X1GC#5AnABDInMX1$&V&XM-B|7uI2WoWiPixJQ$Gq(Su@rEUR~Az7IU zK1;``(4fxsg^isdbTvhDg!(>BO>Q}XvC!ve=06+Mb6C4i4T__@4*Xp+%D*(IG3_v= zM1x}bG|1mf`rjJVrKOy>m2sc%@#r7UG9M-MXoH&B3zjEBwG%-l?`%+6g+eu`HqhGZ z6#lzGU3@Z2SVn`JXr|?CP+yPDNBxJeUMKW#4eIR5Ox2*?2ld6jkj@5m_%qDge*yBB zEB0>2dB#6zZz7%^^H`6$_85OivM35RA+p+NAkH?gakKGqONsZHK@E3 z?&6PaXi)n>9M+WN`p*q&!EkC&l#nR>3>)_qq|3k{@AG)XLB4buWW3ruSUf;XpC6FN zSv(V~@N~nr8ecj|$8!+(Il6HG0G?{PmO88H>RG0G)5;OO1^s2&Q{RXLyBRW-h%2>jIk_cj@R{mEp$8D5EBTAT=dIx)&9 zoz1l#dA#Y^3P@{p5~AeucA0Di^~qUAN9xROjc^wqf%sfga$PgL_oO}t@&3k4w&ViD z8zx}g2ws#sQbFYi(Y{#ZMq(>@~6vQLH#eG+azDQ>}_=4?j;(bAI*ld zq(=W+x7pXYJIYJsLhAykw+UA4Hv1a8@dgz!6QR#EnTp+JUnA*^5BE@Opl>3k6(Zxu{f!w_Q`xA_kh*p@S#4swpBU3#d2 zMsleqGn=6AHkpb&)Ih_dD2;olW6+Niqm0toygLCmvC%{Q2I;0wLibPu4eFuLZK|$g zu;CR`k24IzebyK_RSN16qGqn0Dr6=DdZEcw?4br5<638P57ixdA7YeII*GY$bGe6_ z4C#5DgzljR8`MLg+f-dgsUZiZa)xh!EZ4N7GtkU>n-4R`p`S6Ciak`R@%@Gb?hO8f z{*V}Dlulw{-{#yIRJnjVA#i-`x`!&|TjzMLNlP8l<#1zlt8~t=1<1!VE!%M|)y&X+ zis=J=n8|c@VJ=T85EN@R^m)W6qjVCJ@kz_BkhehEp_5Qc9mt9yXrT3Uw>io<*CC6I z`VQpJns(_nM;ZGntCD&M-Fne+n6hHGIm-Aa%E~1b4ZRLA$|%)sjxu~-%;u8Hf|RF| z5GAVH9K|KoSa+Laj4r)B*r-7uM{C-p+Z?1eY5LxxA`@s?{yNo+k}j669dc{ylbB>#~WY2>&F=yzu@@~oT}(*)YD9- z+KQzHaIPem_V6z3thCN<57eA~v%JJ3HinTj*1X+}^z+-Aa5=||`niBU$Wnbb7n@>nmPN&N%K z9Tmt?(bY_98qcH(jE>apA6yM>Un#CphYQZj>) znWWX${3C1&8l?gXH;#^@QlT7xAlJNC)*$OfYV-77iZH6r|evb*}410nctmz8xZV=OG z5w?9nA~T@RHyLG=PG?qpIvceW`c7h$QCco;7{o>$h4iIPLal419LOki@3bDW8FttB zHSk z3AOB!Vv|wm-f6vPGb~$aw7Di{_!7v)ns$_wW!=dCR4ETT&noPxdSY)ikrpXlRDd;_kQAX(`{8rl8 zsPT}d>LheMEMlY3I8+5`Z+tum??<2%wt#$7(~gqT%>Lym%zO_0l*v@AhwY8u+g4_y zZb83Cj510m5pl+sjS5253Weii*Y&VH8`Ti?a2+a8>DN8&!kQsJ3c5q}kbD{PpwVY2 zws%KB7m7U}7OMp%C;20=8S=2vB_~=eM)+_5lT3)O44lwhv~z-%09a>2JPUV1_b2D@ zwZVG;4idByG4Snc`rIt#^!dqpJP$t$?Sf83l+XuXNmln%uyPerX*s@ogM;}k=)iRx z4EZwTRpZ2TZ&3pQad4J(|8p=mj3+;{h%E@u2GGWYTn^@j@dG|GO3tw`osZJ5B(dX!6*!!8~N?@S~&A>hvK^tz;VUa&W z`dN$_$L-=E;(twPd_spsmk#_RI?uV`1}ey6oT=Vcd=BwCsQV_#7d1|(a)UsIf`7;H z!>Q!>s>TVq|J9r?YnlScB=jOu!3t1a(Kq2|eE`hoLtBl$($;%}l9(USl@x1b|ZnDbq~-DZd7M$#omreVqtQ z+chX(nL?&njn30EsMJC&MqRf8&NT)_-Vja|2+xq6P)4&<7u*^vQ)rqTfbM4nPTpsh;I!JJVFtd267kC zLCQ2!-*q#LE9!D+Yjh$ojn)^lDO6DdjLx*;Ekky;7|Rwm<6J)j{GE=VE_$8NC_JvC zvwjP}-zKCgs+rQ)TXOkgF7ysw0*7;$X6k>APveT30xg4ds$?*YUh8L5sG^1%ox50~ z;C}*N{=lF}x%LJ8oQ~iy&4gb5!k3|!04z2kWts`C`L-sTwhO>Mf>f!Q`gbD&*|g)( zzSD`owBL5tP^OS+)r`(B&-H`9tHtR5ZYJmI_6Kf#!l?$Mnzsps#;BsL3!t$HDbq|S zs#+wQ))qhqf|O~de!La#6JaZhLC~JliNLhkFZ(J}$h0t{^W8D^;O}NJS}smx(^dlB ztRpy@feBTAOEUBkfX_`xH3JhGo~2fB{tDnaLCQ2!A6laB*mh=gQo%XGa+S~3Ela!0k_a;0A>=TOf&W7 zof>irT?=iqP6Vaare_0X3Yiv%PH`*xvC><6v9nnwuAX<_!!(9`nYeo1_5inm5cf94 zzY4P}bcnX^Zv_AfHm*0qjbvz*?qcc(r$FuxO=iaT{><2*r<)8Pk;yDeEaU^~2)zq2 z_mHSCxO%~q&RkeBdg$}r>iJ_Vd% zGRSOAJMOMgA3==l7s^Z*=)FybTa1~``qb*&U?xJJPK+{2Cz0M%wU-r;){;zTPVn9# zl!G1&Kfp;X-SSFHk(DWvgYN!?^G?1@d~bVz6<5HB`5VO!uBFfWgM3sO2b5;G<8#LH zJ{G~kDTux~Danf)G^0N;kW@;_-#K1*T?}E4c_ha% zyNyL8a_#|YkV#gUAQQB74n8OjXa<1!Cd6mZnbNtB;`tPA1F(yr7{qYQ_9>Cg<1@Ho z9fNjUC!!|EBs(hBD}B6hJk5=uT~#^D2OvGrJRHT7gdJ0DnhC=(lUb3_<4xw_S$d$U znWxs-`J2%>&JB)c%y(NzV-8VG$D zG0G^N&GUX?q8a4bkml(mM9H%6^4SWiR&hp0s@Z?sZ7p_!*sm#Btsp`_?egf~nk_jF z@r((qI2k4Iv}-eBM?P#Ie$#@!8MzM)q%`dl+i53%YM z&hi)xboQNUMrjN7N(I^82wU$jIs#}5vc#lS2dbH%7TL^{LLYB3s;|;a&E@L-!iCV6 z5z_^U3NxF>=ct#7_CR`DCm~9@_N&ZRIO_S0Pu=0|XbetD;tYQRc}>%f4pcL(GV3wp z`8S?_!>PkmMro!`d$m#^0lJMCWt5gD+*5nYwS-iplfWI8=peSj(SfS&wzF~jN(5&( z3glExyY%XvjcJ43m{|sWgUM9v)jJ!#E<|yc@ILfI#3-Y5Hq(A+z&*lQNEdVxx>xVa zJp$^v>TXMngp;*7!%8^&p>V1c)RO?sgtyOP#s)pzWGeRRB}Trt+7YTF^e)6GqjVCp ztGe+l=s8GZbP~E(FLAt>A-o=Qj_O4;b8%=P_aYI{YZEhy z>!Fr&6B~0c(hO3LPC}Fv{T0DhuuaBE*=xu!_^TMres97#4gfk*wEDWPh z#Mb64-B85-aH>SrY1fR~sp`zcLARTXI_;Vn7aGljd28s!#HiBJ@}7?BlA%AOp*jgs zQusP<>^cVXD5E2t;?o}_iMb$_X-eKqGP}2LhxxN5J0b2jfxQR0drL*^$$RRGqgv42 z+sio1rCs}CbFnO$8PvrMm5F>=GApS0+6eIvcy9|#o4Gkgo=hnjY*pV7=$4eK*=8u~9L!&S+qnIi}A0IV`bS0hY#z&VW4 z@_=fUxgteEiXoZG2VHw=AzPttl-Mo?MZ;TQv`e#dhHXGTu4#u+n%Oj2Ff$bTIFnIE zX{Po|b=jx|(3cXUN=nPQYm(Wh-H^(45~5_{GgP|^_>p-4nq7bRS{a{zUzhVc2kqe)L#QsT)xS>|~!>LN(XmSd9tVM0Ms1D>rLf9gmM^2ppHYgWLE1dv7=G^~jHbFHv z;dL67tTgyN?9!Medj_5LKpiAsmh5G?_jD6e5%VRAJ?GIn_^k2&I6KeysEV(T&+H|n zLlQ!7p(h~_dPztiBq0eTp_hnMsnTmu1nEc-BE3UEDbkzp2LWkP1Ox>HX;MT~KoLQt zKEHEjXLdJfXB46~4_~i=zRfVsb+D(ks8$#=-$1`;80R|J z)B4edMOt`9{txI+h*{!P2YXsi8mh}QdGPd5AP#m#*TJ5yb+ES;5m}BitOqhd(=K(e zx3y_&C1yH9?_(I}I@sIla|%1Xp$<-fK9v~dl&XWhtq-vKG<9aHAg$9TbRF!?#f=H1 zsvLc+sP`IhhUY+D)wD|;>|^~A+K8Da&;^P?Mcla#_OUjdO=71CLoZ2;a!O}&p<)<2 zRR>bEHlgcaAD24V&nmg780Xg=;6M$l!CqCNe%5zGsxa~qW$nzm1-G}BY$ zXXaPve;YdJoQR>`fF))X{LWAYIlrgfmb+m;|_~7d=p&v6$0KRcB?FU%Q`~>}X!xVC1dc@Wh zlqwf8_rhT-;UXU>o>!E`H)SA&kW8+qyOiVw`-GM?4A#%s3iK%;5*x~m(jyv|S(izV z3*Mrk+$61xx{0DYQV%2xt>}US-w19?H=Y~HZPFTACq{I__frAPH4xu(bwClt@-ws< zz#aqf{Y?k-=_i%=2IqMIKN8d(W`#pvSRxPK;rtugb8Q6I)9B^S?25WY-If-EG?Y7~ z74aeu=UUDi3j=T{|JPC&98j%q6Bud?ptXTExk9V3T%FPl05FUo<(kRqu`R#Y0evR4 zx!MR^qm?D>O7CR6Zpmufc55s?W0#)2Q;Ku0UjhC`Cy)*Kgbnoa7s1fa0RA*kqAT?G zaqQoTue^M)P6UVDHDhFEXkL*DJrr6M(y0*P+DB^(DOcJwFNELcZJ+0_g3n({D{OW& z=h_}{51l|Q_Hsadf|UPb08BFwul904V}dL3a<9(-tS3lC&E)jI>os`2*HLKSXd~nr z_C8@(> znwnnQUzCeC21cVgov}@krA23M1hIZDjUaWCO}vY=!zg`Zc{Bogo0e16oRqtOXoT-) z0$5}qei@ph{yyIx!_W=@2MomT_;NtMhfE@{)P5N8*y#~ z?3v}L-=&I0bFQJdN>dYu@_!~*B2#~x?oVc@DS)>OG}jesurXOsx}gBxCrG(wjO6bZ zBNjuS18tEu;@k*g+Z3XEE2e6;eOR1xJqq}&P9Q`0gbnn?ra*>%1MtK^;jU1Nd}Rft z%bT}|$dAMBnlW;GM2M&fJset1(y0)g8$p%!w6@Sub$1mA;#@le?xPcAZ3MR-$1*e# zK!$;`G=fps>6_L(ehy#@K`Lsd5&TKVIR(_Oc#jo|0ofiOd!$2UV=xRrswnCDk5OJ)re< z==b`Sg1*ON5MB>O*Yf2gXw!|(dsM~B??PxV3^8Bsoz_73N8@?ob%apl^sjqzG3nl? zmGR(U-{*KTUB<~c_cv?tdn!NP8%h7*!9h|M-N9Q5?jyMjfQf%9*-fa|V*G`@Q3PX$ zHrEj{ACTZ2u+++2WMnQ*CRxepxltAk05;1G;{ij(CHs`{NEWgukPotBa#R>>YPN;X zx#6h-Gx8#0Z4T`CLR5?;MJ{CDu0gnW0nwu<9veRd^f=*BO$`te0))uxS9v)ZTX4_4;(7}19@vsUI2|ezxx}}O5f^R4xf7Z{QR;c zQ?e@1+JsXO#REAsE)_t6r|pRY@9TdtwLDX@GteH!`U%;+e>dQReEK}BpzMu>IN8|K zz;OS27ebYVl@Qk&3+Z&(BuDsT(Wc#THgIRx#vLyyT@9QqiKor4Ogx^)ZMJ%@hi zW-RoH9%I&4Q@Sm%i15Y1=Pzl;8a&K->kUy;J=T;cC}dfvp+xX0uawkC;^ZeqLnc;M z)!*hP4^Z7o0o2ZzN>0S(74>?9IJv}Q#R@#?j;FDiOb0c}m{hL}h?C1bdVO4yAUB4V z09a$pJsAyi>F+fxjl5SEZCeAcgrlZ+5Wpz|vHNJC0|tURTy)N4e?|QQ;xD4;6!b~1 z{mlc^Hy-owLNw`#788g=5nL}@l*X&x;^ftO@BN7?GO$JKJoxmJ$GbJ^irOF(4DHzF zCqwo@^SFE+QR7qS1ZIHdxYcr#6PY%8(@pQSULp~-R*ut%U}f{=fOlou%Va@cWXAWw3eCM^6)5cV$y@-hdkdN?ZaCL3|_ zGRms{2(QJFfdz2Y7e^MGLX~{HOf0&md7LTv*)S^;fVZF|rD0Gym3+Kx=}{pU;){}J zauBF=Z5-8uUJe#7+jv~Rg7E<*&(J&onZ_KK{B-Fb(HP`g9%p7m3b`BVVQofj%cPoa zyo~VNAAokvO*dY?=QF3g#i+aw{x6-z-gM(-r00Dne6;^3e_@OSC5sj;B1+-la}ug$ z#>??OAOB@j_Ub`wNC3s&03Z2>4XC9obb$DdlLZ<#;^ms6y@ukofK+)$L!6=kNzGZ+ z@Ob&ZqNjeZub@=`Hf4dR7l@Y!yvB`=QoX=&0B1BL(rJW=mq$DzYxt^Geh=D1qfeJq znd9YgkB{4+s=A?O<;TG%!fvRD4otk`59%+Eo)_XoD1a&&65KZ8<5O79%Fr9bv1r(k^! zZHq~-CLxVv_i}p@8j79dz&8-jIRQPp4@C}r@{IS6QXbxi_!j{jrV~;Dg97K@%B5WN zMZELju%9ATqDB}5rq2vfA*ckcy3tdq3?{9{>oelQz%b6zsWs{=Cs~?g7U5JkiU{U__TMweeK~C96wk%&Q-0E`}GKY^ov>WRn7QdX?eD z0kR9Io7Xc~%9BT;s6&$}Aea7!X!^bZ0?O7|D#gnSCnyBM7M zuBywZv64p(!h$Ns2w5CRwH&b4)p*n(E>U*1PUH1!QZ@$oR(2S(JH_?&d2d7mAqN4O zlpT|m;E7>}`}i-1n@hoM%Yj|76PjS}UUqFriLH>1%BoSD3UG(SZ)NdssLzc>O$18+EsgUk8wE-f z5yfz58#Hhw%IdP{@V?zzrt z)z~!w#8gd*cc|M*l+C2a*H4P6e&;iY>kY7sX99_`m-MjBsU{9VI8GwfW$}}wNn4`q zCyR8M!z0s3)Y6X0b%o@z{1c==lgLu@?IY8}x9W(dRA=&OL~@f5o|HIK27a9nm%Njt zkNf;j@l&-lnil;d0m;oE_T)^B@*dHuiYnL( zAbui>@=~_=hf|WA;GJB-QqXe%Zm7Z8p=tg<9IT@9D+wT!Ag)JT?s&m1s<7_?oD)~I zhLABERCb)IdMefQBmTIc&<*N)*$L@h9p$Qi2}PI38-?&uPS>0~`Wc=g6`qtwQm%{D zu;$@nyd?^kDI!woACJFrSz$lu?soOaU&- z3ox{=I2!CqDAEJAvsi)yBwdq7=QUPy(y(&b=A@_2y3R@OK7p|`80$m^50Nj9IqB{9 z;P-F{W3`xdPP+L3#-0UGR~dqHaLh@!H*TP6+o?y;u>Rtf7sBR^e2&9h2S8xV@BsWsb>ZB zgqQK%9r~UO%L~bRPTKNK0WskU=7i57x}h)?Fcn{^IqB2M=xhr@D6K^`E#x^VRR-Z% z0%5x*k6{4zYb;yK6##?jJ4~_Y%g;x{LK?p8MiwR`YdOp0KgGYEXiNFO4{@RaI#B-e z<;*kQc*+yMN#H~q7?EAnoL zPi-m1KdyOKt*P^Vg(C~~@%9jzI8lQcNayOOBW zMEV633B@ld`qTvL6q%-AJ?v%~qRf&AJ{4GeGwF$M#on-qh z{8hb*!>NK+zh1c4i1Z(UoULj5Equ*fK8))OU^YPCWfM05NV@pC9xOk7Jk*YY8S-EBGz{IT%2L?U{q`zDhg^Dp=N5eD$Ps`^kl=R zWrUjX3Jd0Cgx#U{Ax1f+jT*f`(otch28M6xb<&wKhAI) z$bFi&FL2RJzE?J8E<(R)7`4DfGvf!URX5L}ixA_KmXDuT56KmVRFY(>QzETlULW>? z)->T2AhXwb;A1Jvy+;sdm;kbcrtNDAG}Gs7erEbWA8r`?5<xamG(@aKhsM6PiFA9>61zFAc3)zuJvE z-Ria@oSFO;@pb|n%A#s$nyGx)hZ|a5==F)|0;7&>{5z=KR@y=8s7>hY*wS_DSNo)m zv?7*ybB2>Zexhl6>(`9upoYwR4t=L#lvA3K)6~ zv5jP>u#${cWYm>ot>5v?B1NibSP@Ykhl+ypPTOOx&F_0KQy+R`!#MA>J=XI5HHdc- zeg}GQVw6)ln`R?Q@lL|yAbq4wsMiIM7gTy!qD@by*TrqUqpe~%!!029YT9L|?a9_J zy;KpOhknB_&O2>Swk{W|%1%9l{+bx&l+I>H&2sEi!AeC$aU6W^dONns>=YId({ucF z<;Pa;1}X~iAe(7g_M{4-w`1$0cG@P@B&+vLGM~!__;RvI){pCvHa)y=ML~vEKwnLa z@>9$6fAtnaARmBqRGZM-v3+b&c0tcM^fN=R$931QzqdIk6S4fG~stE_x0@ycox-&v4X zG%er5H>irXuW1A4cX{t-Pg3n$O0Tkdr|0Kn?&wqlaHvxMkdhh3XK8UkhAKj@N{sST z?K`&0`X{Kem|KO$$Dkmm%%;25gtc7}hWmd&F6Myw(FU#4IiqB`!+LTE zcdH+O^Fdx&O*&?bJZ4I+ltGF*Ii&gwo{RMRsWnBi-0QW=R#aUoX#uo@ z#?^|C3^@hvzI^46DKK1+!yr%8h~w1kzlB?`E+D+%y%J01wkrLGSw-^KuAuR z0p??Cd4?*QRFG+!mTf4BV<6*EwbhSovfOpFSi z9thX}5+k}oJ`U-OHsKfuDLafOip~fq(88Ms!l(FJ*0i5hmVvNZcrA9S2=sD>QKha& zvG5u;W}=}d8Ab)zFq_MAfV)EPMT~M>%Y8oa=Kzm|G+CQ)41^TmM8`mQ<4`p*z-JN1Uv>Ck5pqny$M;j4CKcpzL4X{$EjIuLet z41`gEDV*VDkasjKd#S4EG7yq#2Dx*CYH%_?bgD&hsJ7mptD+w1uH#o28L9z2f*9qe zmZOK15kn!jgp{UDI0iz>uA3eRv1D`|`rk~scxkfu69eJ*{uu|0h#%@;ApH3N4}_#F zJc_$ut6*z4**gm3j?qHa=Yf#0!x^|Gc@oHt9I!WZ4%0cKWO>1=GuMXuf8g>XudF7s zj2YMMsjm2JwD6YrFrmm@L*|t;RhelH{cXdz z4w>$|)Rm)Q&_@zeA4VP7+>29B($0ajNSpZ2A=9gGWzO&r$kUq6GGw059l^|9=#LEJ zI%FOX4dgDtD+1#T4!cu&$joO`SB^p;g_BH?`p+Tr>jPg^< zD<%hv=8#`NlJ)GT>KHO9JN&TFL*|*=2|Q$mfUKeEEJJ3Ik81O|q(E1K{% z-;N_3L*|@5s?^g#j?;9OA+uGxlI+xC=&KFmI%JmnUA;f#0Q958D5vz0xfR%Wokce1ww#vP-Dd@_rEJCcDt^Dwf$0uq4qS+R$74j3}Q zCCh)Uzp7RgG6Y;=4yT{Ag6HBL_1h7fxAt4)tH#Wk zU>9lLjD)bbA|LLzu7kgF8{|V8ah#idBy7>Pnz#q|Y>w&}eH(79;8UHG7tA0TGNvoA8C!T|-2QH?szfw*Zopto1~a@(2&U^+pfE+X&Ht=H<=@YT>ZXd`&G;lMU@CK9fYk9e7$ z8mt1P^;_w1&h-btw{!y4=pE3u^|=^&3BWx@pRRJvSc+e(X3C`iR3J#XX7m#8RN|kK zG0+mU5x7?9x`1nNg#;Wsy;HW=_Ioj_f|bU?T8PsrvY3tkaEpHG1EQSwZ8y7BS^Vs^w?5m z*|mUJcz}b00M{OrY6#ch3b}?Cfz`oox^-n3woMMK#ks};PSpwQt{Le1Cov550x-lt z$~6Oh{8M#y?PCD52~w^Z{p3HM?Ak_X+qDt6c9&g)E94p;!Bb;W6=|*QT#$3U4)|A{ zz}^fDWb0FkAuFzk$b&buBo4laD|#d z74@pVy0Qj39=tx6YWRDAC+Gxf1msusF?4E?7emVctTPY~fey&6k=i)@Fn|*TDc95( zh@7tf>BfWLZD_x0BXEsg)yKKeZ8JQxrKT4V(#oGJgmVpy$E`mc&dtC;{dPw)R2M*; zfmDGSD7j{FF3^qux)G#YGx~>w`^*N&L!3LU6LIMd%TEZckkib$0nfa1 zQ^cp+Ft@-)fJ`0+bw-o&X+A_GL;j119Ik^4&VEA`O&f&QRcM8+Vqi0kkJbxn_(!^jB}C9tLfsHUiiFbIvfb zcmvLLHQ=o}fx6i1m=2Vv6VA{#04^HH@oYQ~`6rep@YTpa0X!v0xn|}JLr%u=K|4B%v)!0wtE?mrrk$WRXegAAlxGf=@tjoGzn06rl| zxn}f#Pph}~uZOl(8-Z(P&Y&(vJLU|%hUem3uLAx>C$PI_hWnvu?hMH$xF>`|mAZ1x zK$#Dd*tPNi!U$5X8U5y(8XUDo(2}(gxMt=Ij64;^xh4W`sT0@>)C~8J%cU~Z55NcmsRA|7tXYM)KxYA1K#+3H=%H@) zxvp)4wp$xP)XbcL3beD+oT1yds$aVe_@Pc93sP^Hb=q0&kojGxvXO>wR`vf6aa87RDB2|Ji9-;aATby>ve`= zqB=v}01PltmN~=a7r{Jdm;zuXLCQ5VXQ({NpIuuAZL>B4*Zy?Aj9mF9=eu8NEY%7`v9g6@q|6Hv=rL{(7lLeBs&k09lm$!S^%b!U1Dv@f&~M9pQ+Fwkkv@Z;N6IMXz16iMfT{#3*UX&3?@9#E8Ja+Au8qJompQ`-r#VCU z6*bwl4**Zr3G4-GhWk@HOEHuQV1t2Ffg0%j^lDt7UjsNrkaErFZVyAbK<`4kuZ_Sp zGiRUz9pf}-7*sP4=UNb*c4-_cAYW5&nRU)UvSZGW(n&RqX85+fNqCC8Su-c-{(YDr zTf+d1BuKey^dG(q7T-ai3vIDB0(Z@vf%06S=M1zMScf3VM>!;W8@}u749lh0o(GN9oM9E%?V8VF z&ahHi318t!Xk5+0HG$lF=q!H#e@&+`a|VzZi>sJ91CtWba>v1E3{uS*K&D04Gjj$e zD}t(}Nj+yMiHQ8Q&Rs}CWGdv28p$?isP)=Yv_&B0AdN^4zwo5QNiQeW6`tcDU6TcN z#jB^S7QL12X{-N5c)7d*AQ0C#jyh(x*e4li%$~pT?_i249k4= z&C^z11w?@}L3sWd;t&HA!&mBQtCn4=V(mYK3@viCQEvdC${;+aMZ`KClFQ@h!?+}M*jC8J~qG~_zF*3wLtQ< zvLu=fY;sbSx$s;LG|8)d2p)WBFbm5Pp{LS6BM&wZvIagSQHt~^*S-WFRkm?a;bJ{L z$xguD)u_C=29Gpemi1xi-Uz&RqYxB1TFXxDMR=05PsR<57SrCQE9>~D%PRa%*@Ewn zQVLtNOWyVHQRN+StB$CJPx1<|yBd{SX;aP&84E+ZqnZdg2a0@+KY3t3i#0r3Pu)PO ze9yANV@xkVNo-Yt*V9Q{0k7_lJCbXW{dFp%)1 zFZRi=%cKbR4mfxEXUhh(;|*XA68Ev|xgRsgT0s{E*} z8BpX?ExW{>q6@7abWHKYT`$7pf#}08kI@w(AX}^Xc&rojl1r7pva+Gc347FCuyHt z`v^_46Y{2iipcUpuEzI!DMi&7m%RJnql*5TZ_A7B_$1E(yQ)z+lipr;S+0Sh1}pK) zb}AJ4rf*2!k3%kIh!<5?jtB!3 zX-RBER|Z-OKphOozg7(sJIs54YfX*)l7 z7iXu!q1PlvIi=-eKQv;eQX#d{CPdm#wW8PyataR^ZRuME9~)Y2ViGvR4?s@VwB0Gq z*!IOUvjX~h!ziaT(-CVIC{jnDpCm>(rRDtCbB-c)8`7`Zgh=ai*T!CuQ&=SS&d}oc z7-uD2_TvowyI|=n4i$xPiv2X%B|Q~Ttq7mq*sGXSldRFOvYadtU$!*KYI2cj)0_9# z5@e_^^ufd^KegOih@D}hLO8<_ zAjfOk9x2U)zp2H{66k9TqavjlTT8W({a4VxCPq1>W$%7*9H|?Se%2;L+IzSkLtap% zXfxN`=n@8p6|YKtURuD#Ug-1aQ-plWnp(t96z_(&{^PL4plq}Ec^#}DPgD{K&>8`1 zZb0WguY+}=Zy{#hg+9zM&V617Oa4%WTk;I(bBIZy+Gg((J6IzM7vz?_5z=;TLic$c z?2CZx{aGig`n|fG;T4c~HSN;pb+VS>T{z_U8|Yr$^%*+%d7Uh=Qmxi413iQo<&^64 zI$49qtJS&XFnGonT4CCCF^t1-Nr8Z4Wf!>lB<&@5*ZhRH4jC~;u)+Tgc(vvGAIylwH z-?dibb?sC%i$H#+X_vm_U28(e{LJi!e!?)$eaXAl+?grdm)wGWj~L~Y&SrO>hU}E= ziT)1>%OGFHF3Jd8HR(5(6mcm(%%}iSru_h=p7B?+?VvXtm^f+FG+_! ziWudT>Pz}t$Gz0U=kp*f(I#|X(x3Yh^t!5%547qAs-igp@>@;2^d$qWV4oW7_&w;4 z4ddLG47AQ{P3B1Xp#2Bnusfx*`MG2=N2(&Esw7iyp!<@694V}8-h_BY$iY^{@==^& z8pv*%wpVn`L_ZE;W)$>~45KQ#X6Emz%oRNo`Wj+Xq_jMuL1XR`zJm0%Hi1ZC*BJ7G z26yV0nmpL(}`ke@8CYPhP}zdXEOuR9Bo2( za38vMaAU2;JpwtyFF_vDv`Ys!*4kJhmYJWR|85xP4sNXF?_HCf%JnWfK^%6cR0lWK z3i?njzAOVNgk%bm?%>9{c5sudL5ZQ9VRMjgYucrQn`AZIlFH05=wl4y+`&zgPUbV&#>636VT5a#<_!=Wu4ld z#F2Uc{V_4hDb>NvvfL+ya-@9FJp|yeS9CR5nq}|cgy$7>7K2m%!o)X1ejadW0QwU8 z6d^ZSbGycg#z^@VrS6UpW$#P2SX(-Ji$2hX02*yT=e}f%Rr@1fX68d*X&C3eWQ(<7 zmD<>OFZ6@N45D^v?@P8=C#$PPMn6LONt@7p$rjhXWSb>mjGXf5kEMS&l=sRh)e&s7 z=7kDo!l2hSjB{VI&GN3L?%yyENOK)=x#tGIJ36DZ@B7+nv__#`v9pCh#-#--%I9>1<{{_Th5Rg`)Jr!RM}< z?M^Ot^8<006;!b@XIKqnl%{1m#a{nFT(5p0l4}2fxXb$R>++neH@+Ndl8xdbQ@;Rr zS;x;u3NrLD^x4EHKee3sTNyDP@&-uTvL2d(68G2%C1&nfj3n95#1k68WwsV8Pa%Qq13|Hq*~=lXfX^6-mgrZ)73hHZD z|7Vh!Q_z1fjC1|`+PX8YFgx`p^ryrqr&Rs?+S)Wtt>({*sO87O=dSDL*RJ*Rgw^FO zFU~LuWTK{B>gNgTv$d6(=>om4VVvvd32XIu)q_ujK8+aVl&YU6tihA;N=a1TPa%D- zP3Zc0!d^dx=P?xO;7SE^Q@MW^ux$v+ojyg#@2y1}6T~f~{EJdA&avDtS{;iQ5Sh@t z2jlt=4h1@w`$el;vM)2?&?5}vT<#aG)FCA~#4Vwx5wpgr+%H;f$5iEV9}H=PHlfS? zqHDQdvO+&lPOSjBUehk+e#wgaAex;z0{vUVIG6h+tNMW^?9@Hz4~bDusdB$$Jt`N< zPPrp$zBu^Yb-7=1E%z&ycdiV z3(AY2rm~j%3<%fxV^dkjXHq4sT6zMbd50q3jZ`J*ou66$i1bI;@usP4=I=eZzIqd6 z2>5VHvzpH1w91tw}$YZ({~0A128e5TQN%kc@eDqIP7OR>&vpxr+N$wvA=CF99m6>em=1R z^qC&l{OxPDnnG)B^cLl3r9hwK#jjOu(P2A6Be#V419k;$FT`nwe>x-0=Ubs+x2-*EWS^g96a&3Ize7dtu z*hgne3Hd9K*EwKS8=*L-b4HQa)$3IoypcQqaFjKUtS0GK6^|UKP-Hk#_VL-9QU~up z2HZY}1jES(enm>8>{sH&PA-jp*3Xy z4~~la+=icW*m0#KfN%r-IIp;BDcHT$Jp|;?Jah#uL#B5=P41+Zh+dD;9;aw)heYWn zA-536AC4oRafbX6aZI0s7D(R9nMkr+Cv&Y?VG7Mf;8%4LQ(Kl=Uo1tNL&L&42`>iU0CIcJ9uE6$PqO$$ z=SHTsFc=J|x&~cp%Zlb#~*_ViR z?V3;I<5SXzs%65H5@*VU#PZ^qjFcz*^WDNzuOFbc-uaGeD=Cjh;g0Vzq})!a--R7V z2J#FVqDUp|&`30$qdj)p2%PuG>?WOSt4eU%XGR;mG>YE6 zSUZOV=g9|q-}Y^Z{n3S0-_{&x7mer8x1F;ZZT2^{br5+y0r*s%#MIVvmaTAkQ(GC_ z2xz+#sQR{Z*0@-2Q(GCl2`7^CQQ#AG8Wm>sNEhSvaZB3X*?2JC z7wBM(i^seeg2{x_-nvG(GdaspSc-z2?%O83Nfs;6SfZqVnHzYP{zJ5uwP(1skW~5t zmSCMis+*K@49?jRU>V1)g|PW}hF3lY;xRfqwuj<2Q4$v^>r3l&!%!iE07m2hJDST9 z_AOGzNS}5$L-4>E=$<*GXiQ%4Bkz$iR+g9_)L1o`abRa@-l=<6kI_Y9KLKmbs{AHC z%jg8Z0RN>aCgx231nRC6shSLIcEzKX_Q&Wvne;+P^W)&?slr?;)i9qZ z^U;s2W->#LM|Ap4#Tp|@Bcs8m>NJjs+pF{?yoF&0RlhD8QAHs*!~3?`g;be#mj3In z;ng~0QKsWhaG8=cyb9ipx&ldeQgZJkmoj}kmwG_LS{}nY>W%@qngbS(R47y8B4t-; zB^9hK`5v2OXJ1ihs!qdzto-UZp~k)RM(EV0&oZsT0&? z+Dlr)FWXnu;ep4&ED$`kr&YM`ACZ&-9D2CEFvP9forUe#DQ6E8-H8mo+D1;{^PI_IbF#4*iQ%j0N z&{OH3@6xfaM^AkEfKq(tT8uu8bzfDya&D0upX4lHOEoG7?nPg9S^7^27DrEF>f8j1 z+^S`#DknTi+9%IqkyG?I4Z;1DaAe4yh}IH=baH4 zD2m~eoC<8NM&;)RIPXg^^v)~1dny)+T&HE1yh%!GRjW1}NSuYG6yxCl{WEC*=9)i2 zd_u`KBM(P1`KxiD--LqVHaXypgavRgDvuvz2NGcDMJU!^9f1)Us%3{IoIFYUHCe{d_psVfz5Z zWbv=X9@4ttMF=VC}g~PFA*B5hQ zIUmAGEoNP^yJK`k{0xJ7#1Nc=W6AFNsd<&ZH$ePM)DX(VEZOaqQcX~y$T!g3CTb__ zdYizFJ1dL#NG}Pk9O){(Ub4HQQ7%<(^`XT(((5I=MLLxhRLZgwwC*OoW65rMuezcc z0b?Leasu>{-N8Rv%EL^EYY5;lIhO1`4XCYLJOuH$v0yexX#7@)3c)RC_l(X|0vOLn_^V?|ya2(en^+NhT7Qe_aH^%2$$ z)@Z+c!mk9Vv8UF6HpNJ&~mOxx%fFyEY zz8r<*6W+rp_Jvkd?s!-oUwmULR<)F9F1@_>$tqU7NSurYN&k#_hMCU;N`h}}_fuRN zWbIVlAZuV-$*TAy3jr&mQJHa+JA-f-DuuWI%5yLzYirr5F$zzTuE}(~&z-lhYdH87dKMv)DX${wBB#4R>luFR5ldjV1*5H+xsDhRPt)(Nq zZVmf}*B2CG*$>(flitx99zT=9M!*b+bDRL(8m>PLRUU4JxRU@5lcP2Kl{ZGYcoyPC zW5KkBY4P@LcK(F+)acwA3TGtZ`A?KnJ{)XV`k4jfKJ|jcm$ZO91Y!*X9HIr}yoKG_ zPwV6Q7lh_obZQM$8H8sFLVgWT6@&@D7BWG83ow|zLt~&XSKrTz^2e96$-<#1G=|go zqeh55h4P4*l>a)2I}9*aEnwe`f%c(jH1BGYgBeV#!cWW7_vS@4}BhnWQ3L-mT~e-~^v{ zaTpppAJ-aA!HBG;WtY538eC>sut;o)O_cY}z>OIC)KcaTco$FW_d%kOl(1uSlY zSH1JBE}%lz|J;K_Z)kG>EjOUL5u+Ik8+sD68~S0xsCyoox$_8@+OaP7D)d{#3_t>5 zvMGO1?FR4ylALb$N2G15UW>gT$FPF<>)TQH zGmLUdGku>Zrv^b!Cq_A?IQ&>Sft7Z*+EMYamLUxM6 zevl_LZFfpDn->;i<`(n^hEYywrfhZf@NMpo@e&9ecBi!5G_o=~RSr@`lBwjdYs#`P z_JW+k4Psi`&=R|FtXtWTGi(X6qo$qKh!OMr3r}X!p^r0+V+jS@>6glD#1=!(Bt}I_ zXEX3{Jzhew2hvyCgjzyDscF3_R@0SzQ2~x zqNL)U_ri-p zDn&8{NjJqhu1zu0`by;C43j{%(zHucjI^e8DauSg=pzi{+!Q0Nz)PjssZXHKCq_j| zHN{BFzlB;gum#c=+JtV3k?fRdiqY0n+`XjYyAJYKO}jM3Xsgn8<+$w=-1^0#>Vk7q zjJA%?Ld5eG73HCa5u=<^O)=Wq*}E|}#dt_fvrP(Si^cmaZ~&e(ofoiZi+FMy(v;_I5!x5 z1)7F84t4f^ElpG0{Ujf%Ypvay0 zb5tV^Yxes$tz>n#eV9I*IxvtZpI_y$%xx{)@4pfb-_nUiU2@}+ z%W?>xYccE0ztULTb2MJj)P0r7yS(v*pr`TuaEjGz*c?sL!xa409t z%zu6!n+T>tsQ|4K=_`BB7QWagQ}d#U%tI5Tf1cOCjj!ME{TqTO!h)mG$g?KLmTnQ^ zB;}YN34?GjDtpmyjSQIrLy@y^o8b(M$U2Y{HR7;ln&5h?;Ta#(1SyHF5AfkSiE9&F zZw>E<^&;qw=m~bF7lK@_Y10JPTbr=-pUV?l-7>h}K*E#0xF(0xD509*r3KlV;EuAc zO>kyqT%~lIk4_MWZ9B!n(FCX8$cOuX5GrXgYZHu~Tu;b2s7(#QIXIf&_(wIBzwd$= zNYsAH#5BPhz5jhoFhPI;hHi3nk6&FWH{~X#DM|#}^zdDz!3hgAcvySw-33hzD zu%IHAzd`%cq<1vIT$?J3#RPaGTK+hcgSv(F$unoCyYes$Vl@IdOd1#z*s*Pxaxn#B zOJl(_!QmcJDg^zY4KX@5!Pc#DePbfN%b@Scu>A06-2}I|d5MSj&?;6#+-iWo@s(b@EW`sTogMS2DO&?0}m`l!PBSK zvY^|mg`r|7_z9HeBAqU^SGB0d4i$E*8HszH0M;3Z=Ew@FI%&B(EERJ*mQY6gg0Bw& zI&DDiKMZp!UobOwp+7PVcOiy3h*de1%4-pVfW!7LJ_sYP?^uAlp%6&nB$LyaPV}hF zUXb_rwMfhJTg#GegRaDIhHXK1)wF#!sF}n5QOt~lKFKg@HmI3?pH$-6;Bx4n5u=>a za@#Fx-n<{u5p6;}1wdZdf0&iq?hB_%yZJ5*;|%YE{72JrR7E;X&8#01!HgdYt0)dt zHiKN4nOD68rK$lvf*9qLHWA#_Tcks72`NpR5NXfX1hN;jD|{iVHB|)9F7?EFASP%^ z{?HH?eAI8CB~w+IErR%&0S27qiB}P%e!jDi*scX#<(i{(2Ny0`ld4>$fNIN7<>*su zTm>E7J5Kxp{vS%?M%9_>r@hK4sD@QwcB=4Mf>{p^1<_p4UgegbtO!eWtsftIiURn$ zCZGleOy^MJvPuHr#gXrSJN(4uD%sN#G9G%&}SINxyseI zaxb@Wm0JUSBQYvcI-5?{+_}nq4e69Np{rbd%UUf6i0z>`&^~MEuC^8d&wra#-%|6j+BdfW2 z#5?Btl*D!(e{Sd`&Jzuy)A8beJ<(v=2IoC+u)Fq&27whb%9x1;gB1)UJn4&l@+B7L zhpxc6(?73%iV?CYm~O;*wQ@Q4IDAy+K4!DKh{h-RKCp=zmD4EL8S(`TJ==~UsX7$7 z6n{2qB!_b!4QXFUZ?p*mbTLuYl)_=F2y-$y5exBl z%cxZiK)3Li?1rCRHa{DY^#L^{kWNk3GBj$jXe5U3$6{+X*b_{;<{TkVRpMOddSWdu zam$ZEEYOq+k9U`6UV<)a3r{kXDa))6vd4W5$Rew9j(dmVFn1NeBQjSHemOFg2^1ra zd&lFrd10+YtH+TQOEK7mWw?133M3*s_Fr+|SC8XPPRi~`)<-9Giu>r7#Z258oB?Pd zfh|oW9dUnJ1+Py*+=<);=$N*cE$(SM@%m3TcmvEInsbCe#r;B+a^ekMizD3-ML!%I z0TmvOJH3cTc#@$^xx8?^J?i!s!!|63xQ||njty}q zG8|BS0_oJU#r-_?6qIbREtsB~%Nq9-jK<+>5%*CbKGKxr@NnGenHvM=9GS{#CZG-XrC|LV5RcEYW1kqT^>SQkDgq=lJN6%O zKRbvkrywU~Yb0y0lRCw{V0laW7kv;!cJ#<;PvC+vA=W!7n#&sZ!u=CP zTZ(%>5F<1tIXoQq6GbWRWGGWkDjH#r`(%*M*XA7eoh#6RdJ<0d`>UI{GuR){dj$SM z;mlIs(}OW9LO(;~d_Zfo#cXj;uNa9XQq}+#Ohjmh?xKAG%YU0k|T|kcr%!|CU#(il7Zfhg% zMCQeLm&C!RmM!khHe#B_25W;!&|KEIKgKmR|CNY)2N1n9B{@7C_XGYEcQTYI_u@$o zHIGaO`P=%Oc$2D2gzI$ytLDiscggV@T7U9UvLms5&^lF1NOGT{)a~q<#Qlj z$S#Yu|A>3_3lSW5a#GervPhlODej%~2b;Jv*bz`~0=pvbtZ|>}hZh$j?nF)oG+$fH z7Wc3IzzP*MxEag=%{fA#>ifRdeqs>b_9icY_(@YLJREm=znkzRLz%K^ToZfT{{Y!_ zW6p6uvIlEM%5K1`&v4lKxyF5j3Fs$18sXLqkj^<^qYd`FJ?6uMfsD(J{YTtCIfMp> z`c6*DjYzg#Cv}SZB)ped&)*rm2c6Ue0;uoVUya2(gMe*^M7JN6%OA25x7LTVf+FSq4a-@dxp)<-PHBuIq z9xd={In+uwdU3IwhXk3F=nTd2$!Prke1~2jo-Aw0V!zdi;+H<#z65bdQ{tW4*qAIC zn@1HIR;i|Vk${H~uR(lC0G*FohEF>b@vEjCe{k8M7F-ToiC=7-dmv&{7Dtvas7~g~ zR@KCoVhDN!hjcu3ZF(6lSz}u@2q|PtH&7e(f1HJ63{elFbqsEs zv%ythwrkE20#zKt>Qohie(awBab8m@JWjjDGA&LigFVexgAY_1T=xk5xBpO z7n(`GZCLP(!*>AvALA0q+%K6}z>NL0`H?&bT80_M4a6Fpn4hVaUxJart)ZfF+3qL>6gxJDm z0Cx!b6E3MH#E)Sn%eloUXAha?d9?(LF63vfcgr+-{61MO<&R& z#=ZlkGvN!=05I!yM;pl&&;f~?HdoI0Li?)fqP1NkintOj9D9n|i^bdd%b z6~vL%W~DD&JE#%A`*L(?Hx^kN2^vzO2&&{+I;ggJ8}LpxwhkcP(UemMm6Tpdbx@-r zE+v4@$Eky=T>{fR^T*ynS+@}Z&V4769M#DjVNk*UCnHY0$cx}#0dZGTDn$0++qh*+ zN~I2p4BoN@`_eY3snVTxp~;l{dL`RCC@P8%Q3^SAP>Vjn&(#9WGhU3O=jX<A5Ip|yliUIv9n|w`xK)j4Q=^o1z(?yeP94-AKVUR!4WYde z9UWBBEvPW)pxDCu045UjHWdRmA=5!o&RIIB4A%~7folh~!rnps^DBbI<}btauiA&K zh4R7MZlWv|${P?~2s59$Y(;?Y$D`vAk%WdRuJaYK}j$C{8$BWcMfo^&8{8Psw+4lM3=Uhl$Y@5 zHYHkZqU)GJ^*Gp&x97CI29XQr!IV=6wbcq#9aKq((FD->ICW68$1QL5s`h(TGpHsN zaCMP$Z-XQ~bh50$-%v9|G_v692OuVEN^*$o!I#`JX!k|oNd|A(niRzT^s!+J zqMG#6Mrw*!nf3$>K0REv|zQ-vze392F*4#neAApMOxd*G>ab(5r80@2w zSglkANQ3Oye{@h|@d`m|GUTM}gJgqrQl}2;?5h%{gJN(FpydSqMnTWgLG6gZ0EZ5W z$bEoLYm3=BsA;9k;U^>I{4rKxdk{ns6h)0bF`^Z%FFn3L#J`-ppL9V{GLL1X+%c{)qf58bTlEhP!wlZ9*6BX z6@zOBMLB2bpnAD>Py<~%s1f!KDzpr2vHAC5dMcTBjgu;rUvlGp<5VasK&;o4;`a{pa8g(rEvb=)7$x8F?gR4Iam;JtAHa*oK)e&1eF)Lo}ms$fk@Pp$NTno&WY?f2@ot0f&~Z!m*5s065NVQ2v*!$D6~Lvin|pFS{#ZM zD^`j_(c&&)ovyR)-)#;9#|D&PRN!~1e^cGcc{ z$zqRR?4j$s5XQf?K-7xD!t9%P$FJ6D<%Wfg%AX(_ zXL?L<{OVF34_k`ECD3d#b*>N?y|JiQn3kKpaRiEUrpgG9TQAU2v}zNUJoW@@;vUqG zC4`CCUF7`Ca(MrJH@Y5+t#aIW*O5SS7wO4QK9zu^RwA-*D0A&1{^7V!&@LC{vhPcL7x|$suIKEB;36yO5Vy;ENtbaK*&B`%smGA0y$NOgEZMKK z5vi1rWKAgh$}Tcc^O8IwJy$Z9?35&z3-P7x^y?Pn1|$1XAfonJMmp)U>_W=?nCg2m8_53WjmzHxQlq4MYEIb$E8m!@nw50 z+5X`;jVKC9m4vdd>>^t`%cWh7jZ{5Rtgl(rZ5Jsp&Cl9JL^vCwB@`ZH-^ANReC$PJ z7ol<=M5j%U33ib>6R;(U!$;73Fm>^Fk<>4eX~}P*H9I%zl!Qli5u+FAC|Xsx zmp}HtGgji~E=T3D+I>TrxAV!@y2PID{(*lGbT*f-q}cX0k~!s zb&LD){I27S2yIxMBv@?u5{dhdNpO0BxKo)MqEZxc)e^*gZEhUo#9@7CqD-AD1V-Eo zw+qzvpF`Y#fMS@bGQuNqUlGi>(^0hARNZ_QW(t&EM-z|xi`TgDbP$3obXNk6b5*J; zSpn^lwH)Suka!$PD2uRU`Eco1`9qRDq3kPhkHI(88Fw~P)kU#JW>L4e|BhFi&Fx); z{UI7oVI77u-tm26MI00ncPbY_w88Y4Anu;)aCKQ69)aePsdI(Ei2LluMYNbRi2DmD zKAI{cJQDYXSr~UZidMrHRgB%tQfC(NR5arbTMwBjzp{~-EAVXs8uo^w7< z?+|w?1F_z@vB;_=h`VhbE|iPIiqOeB;!fohh!&Y16U6;Q z_As0>vgYm395HqA<6Z;rkW9z>1nLG9k4+U`GsW;Hapx2KnuCs_)p@&5Y}`LU+3lai z<6ip+Zh#dyfxF&VZ1dsCRjFEG1@z8g+(vB#NtZ-q>nz#ELKx8oLNY#~>??6!(ioQ) zahzcz)m9YSWfpad`@NhMtnFQdHz9gL;bznw|2VVu1+MrY?o|3Z^M@P~4u9<&q+>i*b>*tkD}vgPT-<6h$yE`DV`g%6%%vE7Or_or4s zr#3*XRUv7fi0qXm`=bQ5zHX2VOep(G+z*f7?P))aja17~Y^_<;E$(Te1FX1<@DxN> zD14tt+!u%79tPr07$4 z5_f(SNORCpw7OEzoUcYfdG1W&abNo*2F#fdtfo6pd}7m8sq&NWe)9Z%YD6wP{|d>O zL}dP!tQ#MXxeLkbgtD*1{cQ&0t_>Thg3!twSfn9talbmJq7`=$MnKe%!t|&+{&8ke z&b$(LDtkjT)byAj?)yfU#3LiDc@8uyO`R(Q#&IU?==|D9e3DA-f#RsCGQuNq=SQeC z2OUMLseM%J_I?Sr9XbhOeAgPgvthgl`p9Hd& zkaSBZ`%2t5&Bpp*duJonEEJn>7Ilkzo7%4VstET&bd18XiNt+b9(+v<+dGwyAo^f> zOc3{=&)D9@VM@d&E0%a6FybDXBUHPJui>c@P*gHiO2Q*?H$HhrN71TmlzIMMAIiV~ zO+4;j`eJoQK`@u@!sEuhffZ2i?Km-C3&~%J$eLQRP4#gec@C1h31wf2d&-r@@tuuS z$DthK$1up~s4DQulc+n;sLyz1CrzM~cHS(9AJ) z@wfNlkMd|q|3ci?L$SkDDG86n-T3e#9Yw3R_&GV_nd&1@mc5X8+`TGR)RcA}Up>ZR z>l`=k-&+CwI~PuSazIi#5!sKHEa}K7$468FD`@Hj(9(P}{^uh_U(gmUx6#N*zsBi>@{2f;YH8y7e3GhG3#fVj_x zWMd+-d6sNNc3j{48s0f9N6N&p` zeqIyXJC*eziZVSWi2H)gIGYoP1E3jW>f&$jp8Eo|)4LG&Xed^hDkb5Oxbv&OnuCs_ zmH&o(v2ouHWxmUa$Nj&v*g~H};CTsau{Lhpw^;#weZ4sDe?pQc5!r4_*7zUn;bkGI zl~DGTxEH!38>hxbsvl5nkXh6%?)RVPw2m_(TnN!>3ja6OIe zKJs^rEFB^EiSACujr(OQps(xWQyLQ>S)7RMrX@S@J;v3|ko=uc_LaB~x`B?zafXdl zFMu~@QMb4+xahFrF2ZzJonS1s`-#Lo=T_`>*xsqE2vL0sxoQdG{(C>nSH)pxXa<W?M9EYy3nh`0K8=l3-=bJ| zv#49#OaAIQj}+l#h~`pQ0(Hkfk8FbHP8fG8w?lNq^q3&--B;isCk}5x^U~D0LSSs~ zoO}(~inu!v#Z*`%0!DZw?)3p0$W==a_lPTaw@Mr~gQl~oiy!wkdF)!?7R3E$D8`y9 zCE<~{^Y;}r2OUMLX?PIdIG>pVWzw68$9DBQ*Jwa#_yk~cTjjBk`gE*JQ8>Q z+=%9&qiA*WfEjoGLeGd>iN`&UKjs+KAZS5%cKnEwt5TIzepiaenGFA8^}0hcED>2M zOE&d+D!l&!$pW6~y&(FD!gdU2yzTu}4Q$1TJC$=F zT4{Pr5cfPiaaJJ?_d#>o)VV@n#J$GnK&|OI#QgyjuS}H@9*H}Di%N6QQ8XT|#lx-g zO%wjw-uL(`v8QT>%r1xLzaglJ#WpK$+!tE`{ZJSKP%B9KBqCd3$+B_&Hv*FB31wf2 z`w&lD!o=~NjZ}Mp17=aTxL1vG9p6Rx5TY0g*Ci77?(5{$85@-;vEEs+$f_lXd(?BB zzl+0C&{Q{d@#DTfeOfKU8pJ&kiY}%~Nq8jg{JA^LK}XRl@Ll=XxDSBx>Al3`p3#YO zu5}RXr@K>e<9@{o=#3DZ68{Uy<3wb)EZNIRrpE+vpI-*YMRB+Rnti4&e%wQw zmdC5Vi2DU7ZksA4;gPui(voqfqi8j_Pnp=bzl5^=gT&*W<0Q8Cko)-6V=T5;apV5Q z3h2ZeW$^qzByAIsdCoK+JIKBq*Jr+mWJE&Q*W!L5yTqN1RO?Y}i&@ky?t{DfTX7fR zMTqWD=!3fBA7`R=8uL{u?O5-WSY*`_#Jx*qOdiBxZfHuGy7+MqpX{N{UXHleg`&Bs zQW74Cds2xz9c}Qqxef0LD<7W?9{lyz#>&@ctw*&sUPz_PkTE)oyn3Jx?!pAq3!fP0 zzbmT4;|yBkCMt<{pXb<*?M^(~;@3p&%B;d4;RH(Y?cSZyzHyXGAwxzUlRkT?96Th3iavBZX4cp(Y^6A^>K>7bNcl64s z6~HI()pgkJTK0Ojw;`~<3%s1mfyZDNv>xwtgc&hRj>QPr?LORCYxl0MUE73vd0An| zXBws7@$p9%jogJJd>PZCTC#9M-q(7q|8y>5$ zEA`VLJj*o-<|&q>3Upf3N+0%C5|wTfCMh*obPW_+=|9elvk$MY>{ju33k*%?1kw)jTPkM_jMg_v0-IN&XZMHfEId?GIfsl%VR$$I<1v>9Lr_cQFL)}#@Ngf-r!Ega zRLBAk4OPx>nuZ%598WO+!(#N&!-0NS6+_iR`FO2OX}HM`MMss z)kLeZJjD~RSR^po6V<0uEhzjQt=;swRKbBs3yVzEgSIjC(%Vl}yKZfvFQD>-v?^zK z>;r3m%cH3%q6$uJpU3bq9oAnhA5rtS<%NsgO5fN%gT|n%EimqM3GR0df?&V$IvHNg zSk?0|UbO_nRqLjepm?Rc-p$K~`@gWLXK_2bUc;nN3{-liV#?^o^ibrnRB{-djV{5y z``beu+M&uTF9sM@37XoLPWB~JGFEwoO{k3LUtsQKNo4mPt|s-W3NOo)SGUPH3-|@* zX_iDbhvDkx7xv*c<@NI*JlnMz<_)Gqi^`;Vz<5HH{+K*mOTi8DIE-f~F!r?U0X!O> zQ>nUbNUt^6hn@Nvj4>|3t#TzGxUT$*7xB^BEJ7>%FuM3-iS=;$1CN&vRrY~e5p5gx zZ&eV+VwRv@)>}RhYzVHk+*{M?bg#_mQoFCnhLsQS&a=-iNCzUBZW|%#rJa z^N%`3v=OOby8zWSQ)$}I_^s}kLF-TZJJ`O^K9L6{`5ZGxwg$JVXx(WKc!4kBV3C?k z`&GZHVL{r0w6K?eEsS<4?;6>j;3Xs(*_y)G+7ehJ+tnl4jRy9IZ7}Uz71ziXoE#sD zWCLcyw!l^19N8+hOQv081Gd4o+p=3D+mH}*WIH$!&yZb*;t^F6usN(shP}wQm}q^1 zK}V1cWAFYL_GT!Kd+G#^JZVgeaqQO|qYy6F$QG4JVf=?Cg1UZeQv8u^^L(6)w1K&+ zB^e5x+sKw?O-gMvi;jX~N?es|Wc%eGDAY=rH(QbkbmSV@ru8UjI8vt|xe{0A8rfbC z!zCIO1GDxrw$%wnw$~H!AW2%7vziihF$sd|8rfPEDrSsqWud4<)ihZxYh+s+o=N+K zTWBj7+nH|MMz(8OWsRMo2E#bQ<-r=+V)j=sJVe8|%;muv+1A|B4G()^Jm~UZjck9W zC~J7Q1>;lGL!2=SBU{X*qJ|sCD~x|wWL4s*FtSz3>NGam{7{snO8PtA$hNq#hq384 zgrc?OE8fU<_i$xn4;ui*hWLO z8kR(MZ?}={NIT40+rr%4lEfd`(v0Bq|1gg;C2k{|V?buDHaCOcU|d6iv8P!h+tK3H zjFIgSj3-%YIF)ueDLV8ajL%&HYh+7}rv%#vLg0hpCpi|Qim!}p-pUbwWQ+1B zz>~>7NP)qF-D?|&nzPm z<+IU9wDmb+B-;60F%s>49vO)aKA((4M<1Vp;`v*jEJmV}Phlg`+2^Q{=;CwIXlhrV zG-w8X1aYUQrZ-b*ydS;eK;9u$F|CavXTi->mDEW> z{j^W%U~dOoXIFW{UOlk?!3Zrg%MXWbjH~>SYk5(=j=j#Ef|^_fSOVKhmz}4>%~V6@ zoaL2`W5#~ij#&04Q-@WAy?Lt2c`<6)%y+5As|K1xz)=qoxuUM;DysRhcSe73@10yj zG5q_##|v{e<6!CgzS1p|KHifmFrHI!%9-GlSKf*fFQpD?> zu$egFod@f3OQC0Q@12GhJ+zB#-aZHpnIc#3d`OvJV+nN&wgm8e`cb=mouc(JF(`aegJ2^hN_s*;WIDgp)XS?Wd3O|z*uXoP9$Ccc3u->o~ z@p>o8A&e7mVYPoWeZ=dXpPJzeC>^Z9mLgv7{4^Z%p3<;ZwG{Kv7&H2%QxAzqqV-_& zA|dEticEX?K<{lSHI`66!#0xk#Zunvo%H3BYprOHhHaUvyt(f*-<(DpLi--r4!Fvj zy>nC*HG1a;Z1-Gtvv)Gy^)q@$A->*NjQXwK$z0s6ckHg-8IQPKQK9XsX_?tO`#-z) z&W%u9X{-llE$Hx7Rx~eO@BHkK-suDDAWIRiccRbX{^@jBqb)_e-ucfH*Ngsub)Th( z*E_TELz(I#tamKMdo;$3e(BU@&L`2%v3Xx0@WeVuV@-Scz{zzzHI`7BU<;;Q%f|9% z@A&V@t{tX547SRy@@DT;80ewhr@a+y?Of%}-r2j!&*+^&unl+F&EDzKI@IW$`LHdq z>{jo*uItu29t{<5uXDbufP-UdG%Yh0qaF9+bk2sunzoP4D+xh4 zQ{?KLCDZ*imQYP$YfbwTDQ|YptR}UzEwuNCZLq7n**OckWYo^mJ{z_LuJUH**n4L* zI%gYfyIpp(b9U@6ZFJ5>*sfc4t8+FrbL$+t)j1sykSnTEws7N8%y2z0&P|4-y1O^!J7>CMuo|blc=^(buyKv{fyt^6=mhOP1G>h zexbb?JeikTE;xGy2yx@_`zC1zIYCd^MPNnSc~T{cNIy0GC$IXnWjR9K8AiOhA`#OA~afyx7OF;kM@ zvdNAm7}{#X+|-n))*BE2*JYE4VS&bFlkcDyKvk%$mUY=AUni&5kXzGq7=NXJ@o~Fs zQreM1i=cM4#Ca%g#q(udHhCV9(YTiI848D&*?Otbb=f5SO4nr& z+z|NPnOU?d6a`AKkaUjghCqay9+yq3Ep}^SF ztji|H$NL(WO*9 zTbAImN!+U+GITgRaHy=)T&4Bl?dmDE;1s2`<|@c%8Qg~l*UhDAy;}T(Yfib8kA4Js zV-)Wf=Tg27dzB;_#n@tRE8u6EZSt5P&98-eqqoeLREuy1YWbMZ z8Zv)msXW?nl%0shmS!N%aWVjg?fj%Ye@akuusI*p6Cu-fz{` zs~>R&*+RWf{cu@n?NN2?gEl`q8~7BJ{mvPo`;e@|52dw1=@B(Nib9<~+L#`T zEhp>SXY1O#19YF0air3Bt4XLo|47_NqqZz{Y`A%wz=%YvQcBCMvUxWilTwROL(}`= z)UtK5h#i{7&&-TNA@jln`i=7!`TUIWImH;&p;I~0x5FdE)bDiZc|z&Z^CPkFAJ{2O z^jRBJ75x*Jp656&{jp5Yk5TUiR??cPoZi_B1)y~EqNQU&XygfZ&y zo0?MMey2QIWt509N^mQb638goHE$RrHOQ1RRJ)T^}&AYv{ma2jLUmN>T8+v0x?c$a#Cl98omZ+9s3O9->CfrnK zRY*t)_?qa;l1r$N9h1-y_?pKiLf9Ix!mA9k;NW+gnMdosxFyt+p}e(hibor=*?w59L#8n>-Iyh|stV zI`u!kOQ|`OA6Ma*!6`vzI3?{!i)^Ppe;!sit0-11U`t3xVX<9@e@B?uw(n$B%+U1@ zlGguZxiE=ye?I$msppZsg=#(7C7YGCNN+KyJ<>w+2 z4u2f_oce|#NCqLPOH$VlXe>S)8<|r-b{h4R72Qu~kPMgPWF{L(^8Q&QTS>Al$~iho zazjQN3NOuMgCk};^_7y2$jlrgy#|^9eP0GDNnIEq+gX5)bwz%;Ae-jc375LwFO-ts z3zE!sO4@Nz>iY>s#~n#NE{f!HDN-gkyE|oZ*28RFFD5xgIxX_ix)OB7v?DkLqr{|B z{}$Q0{u?$RY;+~*z{ueyrN^ebB45|9lp?oF`T&SsMD&LBbZM6T42@H$$EMd8L50&% zp{{2^%bfZ>X&?J4sY`}a&lkqDy)5Y`WR6omiw)j!6LP0MyJAXI$F)Q@ z`}Lh<-yDXdM@lO0F_}z~2l2_bF=}_^YFaBbEg8;2@z-9Q8@5u@{doE4qiycO()i2M zE(dLF3@G5#f5b}iYXYl4+-2b`)6=bc{s3 zx*j7&W;k{27i76T(8b8rDhd-xYyMpvC za)S2281ygkgs}e!1Q7X&%0x4w6>bm)btMK6qXDO0bt#TuuhAM^f4qzbrrT(s!y7-d z?34^g3M6q9&2`<-wW?y?4m5yxdnoH}zHuOJN2ABp6=l;;fxhmSv3d=fgwsc+hcjP%HG>b#-h zlx#;{$p~r!AH_x7C6bjf!>RW~+Nm?g(NJ`K#aLd9KQgu@&`zIX+J8N#sL{=r@mftJ`QF+(@tIeOftu*dn2vuCneuk(#3`_hiRu?8fm9q z0d%xZ!=R4)g+^vL^~p#(CC91Hl>9c5KNo4o1d$&cLpd{WXhGVkp9USP#F%j^jm&WB zS&???%yAro(W&PY<4ZB-WA=3^(_|XUFarZA(!1@?=BCx6tN9fTNvUmB{ccs)qH87N zp)Gq_oErrrR{)!L>9SgMy=2TPLmL4v!?LpAlHej*bagQ{rICP<;;m{Sjbr4*wJ(hW zQ3TnGnLS>z$1|HinSxTL?ihb9x^GHWveGhcRfpbX)1n7R_FgL+t1_J{bVb;6CeyuU zkp*Ux#paMlLC0Mb()IekG78MVi51e0Pm!+I7X>DU82${479kNL#Cv%5xa>&s1%{()cr2rC{-!SX8JT zXXpg1skY^TwUUy*J=4L;6JAxOK42SuviyleOH!ex?O z6I0=@vG$+a`f2xJe~QJ{^D4HlSkpNBPe(m8JG>;tA|`bORcJl_l%f;t*^;K$vci_n zv>O*W3hw!$;ChmNAgf(2FII)UCN)Fpz926@Myjd$jijMsZVyuzmw8yOyfDwuhm|re zP7a4@jAiaI?{8dtQw#L;5uVxtWG|&j+FkEaJujG+IHi?;cxGx<6@M-bWL&P>wqpHs zT<*sSq+jpzs=P~SWEI(V>mRLWHd`qXOCVu>|4uMz~W?^9id&K%-N+xS^V|d^$l+3M5?y>`i!WfGzy@O1+{V@Qtj;(XJ*F7Optey zj!sG)hOLAB_nMk^7uhdNlTk;tak{kB*IXLD#jJM z$FiILV~0Ezkb^R<#j!~^~ zy)jbV2;u<)`+#hkSJaGetME|q!n=P6q88p0J(G`JrsZQ-;d#fpsgt9ahq50XSOM?9 zX2MM{EVf%{EPjZ_uzD)rbrtZ;MIKnHm{!Gbt#p-Je6nXN;dIX{GEvpQ8oyU|thPV#L7IbyfBfQ$z9n8)!~ix)^s?y_LPd zSSMl!^Sd|_!c6{&tu{*@=C&^|ECse%HvOju+qP|~ni)dc0AxqE3tRT7HImMXT+ zI!Z)fPTxe;GPYdFaz0h zS<)MSX;Hy`It1%g+I1(Y;hPr_`~l;iEXS;bt5r@z&>qGvmcY2d7pd^6=jq=#jEYCWI-UyI^Gt<@ zEk!YWcZ4lcD}Z%W7?mWy$^k~VrARV4R&WHylP&@8iAE}4oA2?m>5V!b!}x{*)?plq zq6}3UTZ)HK!A3nPvIc4yut+_zs>Ve#!B;w)Z@0!7WSCIJp{N*7<$aW$kiizT?ocMJ z3iq;>P<%_3TaAV)n=Nnqjec5ns)j)^A)YVZe2i3KwhWi*7SrlbwE~Jwmain-E}kfk zNRjHDgQJpi&>UrytYFJfYqqCR@;0)cStV1-7GU_-K?*_EeGofn|^yJ&KU;6GTO8%9(**pw;ic2*nFFvC}cEZF%&DP zGPe9yJ-Ovyw)xsJ80S7}Ka59Qg2u-KAh>4pExR;Ua2LkMF2N6XLm;?mOA@gtw~T1Y zJ3H==V2Q0_<*KX@+_Cxg@0i1ALkNtarXY5&m~o1G*nOL0TF1f~FN>*KP&6=Ah{Y69 zJhSQL=i*h_F0k~aRrUrJ(4q{*e>TUc$2P;qWGH4@KB9OFCQ`k&=|`U!ryptqj9V?i z7*#W!6N*=kHaPFPt}5g6jFIYt&5siN{f#XYe}A<3a@*3cD?i*s!okTFx+`O#=67B7 zZJ!2zf40dfb@Jhk zSt;#_t-JV3Vo#oWz89XT(zLcPD%r3b_1>(ba0i&BE+iLo`p#euRfy zl)qiy{Gouh0M?aMNX%p`nx|53<|Yu+B{WZzXQf^?E1X2co!chM*dig3b;nn4r3lPPbELey|txXU-z-HmKVBEQ_Hg~ zIT>!}ozgfxo|?x)&bxeqrdOD>17YiF$lhUg)hLNj3?7i@K?Uow{U~abuyhV00ZPk7`Cw)FYcg=dzdHFExVM| zP9XcDX;N=!k2_@%f4#QP#nXnIvOLU*I!2>_+mz+wZ+ONtEesj4*gi=e=9K03oKP(k z*=0?WQHM2UDTq(jv9BA!-p11WdK48{Q?5eL-$l^fH4S_-H}Mh8eAri5 zejE^*ej5392+S&)eXt+5{D?-TEWbylF{Ug#x5Q3aWaS-^Tt62gF!7H# zQ#n7&2qArDm;GlryEQs1-qWIn$&9X^%T)5r4hbCBYcLDNEbjsN)%$=r(0p zcs?nPe{eF4j)J5PbIOuxVF7IxvKN~sqYi7zlG|BT<_$YxKW1qd!8l>;tmmCmU4(Zb zdL38j9?Nd}wZ*kWm<%n=ghd*psEOmTr|#q96k&0Qs>Kz4E%KbQ7y&6ix*P*?T_1&# z38pO4J6BZMrkRW>OWq=g+DY`CmXBSgq!+6tG6(IW+|h!)|6$=1YbiCkRL|^EQY|GkYLIZ z^0A^(aVc2KQy~?b3QSqNa-=kTw173r@)2cCS@Qc=G6Vx*9Oe>OQo5*Qt|?2?9pw!-`=R*9bQ7zxrYt2k$DT$$g5q^Nl{IB~Q!J%+oO@aF0(kTbi+hdM zlqI}U3GFmh#h|D_6~h->qcvqY+rGSZo~jm5bhLc6;CA7fvT#(w{1;}|lx47Ah*5F| zvKLw%Zcj|-@qIhR}yc^vi4jSQK+9G8*53q4sKJH zt1ojK&0hw^mN+WqF2$7PQ(iwg8&IbpyJAV@w8Cx5;#(6_mKc~nm=dg?HDwuj9~bdc z7Q+2NEXH;{Tv=0=F1hh+Ye85_n^whxkK2@GP`CU>BN{=`f+}P8vZgFqmd9?4{a_sE z5?E7~4$G1l9%jKf-zBi7EX&$umk~{EgK@V@U`<&X-He^PU4rqhDTv)httrcIbut*! z4Hb&(-&l;6x=mTU>v-b*Cs=aPD*Io&Da(d5nG7G5p{Q>8a7|e{_suCM3#vVgT`YlX z%JLBBU2anrO8&1YOD~My{~uG9{pGSsD_)_R_iT+ExZI{JJGUp5DNCBd*puQ-Su#)8 zq+<%gRvL@3=ZnZ~$}*})Dy=@OEvb;0$ynqzWqE%)hxQ$;gDr(@LvB-+3YS8)8L-YZ z74B1(Tgj?u8(}zP8Wn?<_>`r)XI1SwbazcHrz~WIFZ!%-RrSk26gw?+MuBPFciXuPXmUu#O0QZc)@x8U@^@EM;O~Jc7KF zbo5T@FsCdp3l`VzBKw7DGU~9VEDzRXla6)5Z)z-3T*>3+)|4gvjT)lK2YU%>zMzUY zng+fF@Bj)ohkCHLu>5%5LqlWzH1h4?YZuM;u>WlN5sge)_(;9nWx233cFH0v?^uAg z_Zs3QXV3X`uVRQmdyhC%7RT`HnmPuTr|H&*wOUc$+c<}IAK5QWlj>$g*}BUj{(3De zlgpU0q$!R%7NUULl;zGiM7ba;FNVdIT<)Ry*j3gkGVZcm$X*5S zzrt02EVir+p*cCglx1H4-YT99%PP~Vn34@v;;C}7#Mh@RzmCDw;WUceit9j%nP=db zvb-&Yn+qSH@hcJA47Z9zY_>5%aKN1jlrbDNB{9Rq_32s2^J;@-u#> z1XGsg`|-koM@iiE!Xo{lD2eL|QlqJQZvKkK&DkF^9Tmoy#QaF9ECLs@lv9cv_o3b=Srs}lO zFRwZIz?7x<;k+VN-@!P5rDO+io3f0NQXh&t%~F`MRDP*MtX9Ifo~4Y|T2q!}o6;D9 zqcEPb1m=VUQx?yKPNU+du)d-~DmE3EvRt^6%JAV=8gpVSv6Wa;mOo5*Qt|?37$|Vgq{h%0Xx`|a;Q;3 zm|atrokyw~B_ojC&?;G9wkYc^%fD+g7}4ng#b8UtjX+5yt|`mU_ks=S9LQG2m5Nul zDa*%wxU9Ypn&Xy^#~t@6%cDHC@cU0Nzpy0nrz|_C;W*(ai-Q0bW5q=hZ^}~j`#hph zg&_;Gq+AELDNByirHtk`g`!g&m2#J2%CcPdG`e{>WD_l^+na83m!-;EMEy6I*O(Ib zDN7Pw_W2u@qi$Af%2NIxj3~EZd1hJ_4?b>FmZIZ~tEkEuhT#B9?3Qm$S+v|sz)@DwapXqCMo-jt>M++v20w@`exe7L48?^kA$lLZw}4%a`h z7~ybDStk7%XUam!|21U^<{R~ zO{Oe7_@@mVjpaH2-HZv9x#NO=W3*wP#k#~~vxe6PqQczT}RGH}Osfz3a)lzS-dDUte zsz@lhSSq=L9IIO8?K|vfRc#Jy90|o_s)FGrw#NIlv%}X$@4Q<^;hSK51Yk838%$sP zxEP;G!^g#Hw(`=(t7>xdZM>>1x7^07@^Yh1bL_{euCfO=4~vb=t4fFrXRvW1qj$sI z_@pqU!eU#?yI!uy%sEn1`<7F_{4ka@1+mvutjMg~R>)8_grc>jx)3)qn|k^fssT`p zuvBuP$`zSz{hat{H)@QAVii^0S-+^FtjL@l?`K>uJOITp(-$Md**zlj@4VC!8P4$; z8P4n(8P4Z5$2G)qmA(49G_jFMimfk1b`i&iGdz?qGCY7VGCXYHbr{4_3;H!3A7JMG z@07y@USGprhG*90vE3xc_z3S*TsX#C6T+IsQ);j6#@l2~kse}e&;6Mh#(RcNy}e`) zVYa*<>2B)UYmF9~g%m@%VCC9NJ zMyKAlB)u&ao)OwgF{c2CH-*GkpTeo}{)5Pegd+I`$pD?NEIak?uwk`}lQ|dg>Fwl2 zAjwjd&iTd*9ugN)5+Fuhh^NWL!Ilz`*A@AGNF2>2`8#tvNYY*w$zGD|Efq8sT%7J9t&s@rUS`bm|kEAmfnO<~2t0jM(-`pk_#*#!&G{6tk#!Ey=pk>}0+lX| z3|IsK5>gi+M!Z=f)foBZ!WjPzW${8$r>3d3pnk*%Vk)taSO@6(`D6j$%t)sBw(~~$ zCjSuBWV|Ad&cV@K*X#DCqV|tmm3F_NF;k^6;bW1UkEE_MiPdDrI>}(tsc#>Lj6EX0 zHwnq3NCxOk>bmzdl4-oGg0$m-6l4*+q%#y5@1>wGYmPS^QLTtIA^`A*%R z<#R^RIg`4+WCqE!Q__wqFrv$m?Wils!ZVRvX5sCOgnY{-a#K? z4#cwaJCK_j3I}s^ec%)pVMc%u#|z{*{t(|OP*4{iPTgk|UhFy{W%lFvt;JJwyntz|XForsT~F>yLt}q@1aSab9yFSmK`a5JWGS;G9$U$c^nWV( zW}9{+le`W6pZfQGkai=p;)OiCIJ%mzfCa8)ddYfjWK1J&lAicG-wz?3`ZP&r-@qKE z$EfQQQfY0JhxcOCk+)`eIn6I^N57-^qxh|wNwl<`{O&ePuW7}9d|FjY+c{uj`}CTw zIVk>5bu3d-vuBbv{HOj&RY|i4%kN1z_35xX9sy4M)I9XeYb163^#XR_^*PA!#>k5e z3M=bKfn03G+mqan$3kPcEDppkTV)}$9BqYzicUow^N*9w{_Ggnm)Yn#X{3>Ug?zj+ ze1eT++L5Fd*NI6)4cU&2$iTa1;;YL)^ulaiXGMB!x+n5=J;Nz-m!yw?cpngSv=Wbo zWWZ@UC<`H8eFYsqigzQUC0AvfQ5}*_{jfBb*?46OD~B&HpuA(9IAl_n45!}u9Mfmc zkT-20A1$LT@IS#(*wq zL-P!8-QeZdG;}~>yh09!B+0K1C0@$kioQRD`hpH46?B<+N@&}F6htGuKoh;l?dqAl?~F^ZT?tRns-{vmD>uL!Td zfV4y|qBK!}o+}Uyi1tKZVl**_SWVQTxHYkhf1V_65ix}KZXkfjO_U*O6Olx3VmL8_ zSWfh!tL;o3Bd!rI2>Tu&HIaiDM%{GcH)0!cl(mY-EFgX-{vys1_lfs}?*SkqQGlpGG$7g&{fJ+PUx_uu zZsHVin|Mv=2Z3}%9-=H!hiF6eAx020i50{S;y7`g_>b^71o#snL`k9s(Sqns{7g(C zmJnNrL&QbmG4Yx3I}8L7p+se(5z&$8PmCqz5$lLO#A)IV@rJ;Q>snACk(US~>Jn{; zKEw!OCb5FpK^!Np6aNt&M*)8#geXbWAX*YVh#|yOVkz+lahSMFJSDVaKnfxoQIx1g zG$pza1Br>mLSiHFH*ubLNPHxc9tScLg@{VTH$+FGKQWe=N30|E5@(3J#9P970!UBf zBmU*4T#l&+Ono5g@z1uz_e3P8CcTN_#0+8-4YP?=#Gk}J#7*KA;q?!YmdHhvCTbC_ ziC)AoVmk2~v5h!NTqT|pwv#|AB0Eu>h#;C1-H5@&WMVO~nK(#XBpwr=3BOZ75D`jL zCK?gn5(9{F#C&2sv5z=Q+#}u*KBs{UM1G^d&|UbBNW%F5)C{i-;k-&j0~L zZlVlPn}{TO6T^ua#ByRgag4Y|yddmnfz(6}q6AT$XhC!*ekP_6ONcGRA>tD8g!n=v zKL=zbiV#(aCPZiACt?B-O>7|c6aNwqhz~@Pe}POyL82njkmx}CK#U>g5^IS)#A)IV z@rH1m2Lg$_L>N(*h$6ltMiR4#mBdcs1aX6SNqAlW(hxa`QbbLn715IzN=ze`5nG8P z#1-Nhp)LX`iD04_5l%ECx)OtkNyH*z6LEmJKs+Ko5y>tAK}0A~nP^0OOAH{!5%Y=l z#6IFIagTUM_*@1u5c!GnM17(i(U%xa%pq12yNHv-Eh2{Sz5)agxrs7FZ6cEBO$;Yy z5X*_}#4+L;@q(~l1yU0^h!R9~q6N{N_?ehOEFrcKhloqW6XFYz{2GvzC_+>vnh>3d zpNI)WG_ir$Py9Offz%~CDsyqh||O!;tk=r0R$3xi7=ur z5k-7Qj3j0eD~X-N3E~FvlJL9cBjyw9iG9Rb;vVsi@VNtI zAo3ICiTXr4qAxL;m_w{4b`d9uTSN@ueHREIaua2U+C(JLn;1^aAeIx`iDSey;ss&9 z2c#x)5G9D}L<^!j@iQ@nSVC+e4iT4#C&U*b`F$WOQG}>UG$A?@KM@m%Xkr7gpZJ$} zKztyQJODBg1&NA8L!txm12Km9l~_aUCQcEziPwbw5J*SlA<7bUh&DtYVgxahSV8O{ zjuY33{|JvqfIksJlq6~pEr}k)5MnB^l=y=n475?zRa#6)5t zv61+jI8QtzJ`zcv0GWwGL?z-Iq9f6t7)#6}))9M&GsIouE#Z6$q$lzb<%oJjTjG0S z6fv7vMf^$pL);`@5nj)Lv_vkVG*OFaP4ps$5z~p^h;77C;wtf+ussJ-5!s32L)NqPW(ljBkmLL3E%&Kj6?yV z0?~kIPxK>xA$}#+5W9&}#BJg=p}z#u5qXHRL>;0H(T5m8%p_J2JBZ`Nb>ct5;}zgf zgb*c(8bnK?2Qh@0N-QP*APy6kiKm1X1Ee6b5k-k=L{p**F_4%@EF?A(e-r14hr~xB z>1!Y}QIM!eG$c9@KM-Sxxx`vx4{@5fL%bmzZ-8_}9-=H!hiF6eAx020i50{S;y7`g z_>b^-3-}WuL`k9s(URyv3?ZfxONl>-!^CCcDWSasQV`jQqC_>KDba-(NK7Oa5*vxX ziSxum;v6wL{uWaAvzNMiLt~yVjZ!UI78ee-V)9aKzbrC5k}M{qKNN^k;E)w zC9#t@LEIo-5}qG{G(=9K6j76CMf4 z60wNbL>wS45RZsYM6%C75D`jLCK?gn5(9{F#C&2sv5z=Q+#}u*K3{+gM1G z^d&|UbBNW%F5)C{i-;k-HGhnpL~f!CQJaV)dK1Hm8N_m8J8_J-M!X>G3P?@lAW9I` zi55h6;%8zCv4q${93n0ePlzu>avPA9C_+>vnh>3dpNI)WG_ir$Py9>VBi<1{b|3?h zpD0h%C)yExiP6LyVl}agI7!?hVhC>!Ab`kClp$&pkwkA|I5C4*PHZQR5!Z+pgxwQJ zP2?a-5Y>t1L^onEF_~CQY$gs87m3HjXTr}5$U+n*st}EdPQ;JIcwzzZJMkBBj<`>} zCw%doVo*k+08xQxK(r_N5x)?>5^IRv#3|x7@tVL5HXON#JVaTd4$+3_LyRD15-W%u z#Bt&}p&UR;BA6&fgcHq(uEZc>60wNbL>wS45RZsYL^3B3M1&HRiAKb?!~kL(F^^bB z>?O_+cZs)z(+5aTM0uh< z(T?a#j3(v~tBGC2N#Yg}LwF|#0*Ks18KO23N%SU$6EleA#CGBsagBIE*i!(hi5x@; zqB_xn=uZ4hOd*yKTZjY1dEz1Qkw}^n$V?O>DiPlh9f|(LSYjTrj@V0_A?^}y31=!G zJ&}(nN7N(Q65kUeiJ8Q5VjFRkxJo=HY^i}%M0TP$5kWL3x)Fnk$;4t}GjWi(NIWJ! z5=s4m%tRrg67dbuk?2p1CFT+9h`q!a;x6%q@J<5+5V?smL~SCH=uHeKW)RDX?Zh$S z8u5a#rv*|IIfxQOb)p5)o%orULM$P+5C@10#3SMpkt_fRB0`DEL?hx`VgNCYm`|)H z_7P`^d&E1!CmoQ1$WN3f>J#mVzQkx^7O|4pNt_^V5HAVOKp+i~lPE>hBw7(YiJ`CcZ5$yAOn$~C{NTU+7W$;(Zn2LHL;60N!%i02=7cl0Fj$0L)0cB ziQdF;Vg|9C*iIZHt`RQ?duAXtk%K5fR41Af-H5@&WMVO~nK(#XBpwr=3BMp93sIP; zLNq2i68(v>#5`ghv6nbQ+$G);&MZKBA|FwXs7JIVz9&WzvxybNcH$^;m3U6rvI41y z>_l-Qf@n^3BL)+biN(Zb;vjL6cuagI{IUUAh{8k_qA}5l_>mY({7S4Lb`z(F+r(=^ z4+hc^d5E$^9ik23)KA{R?eP;`c>}sV?le0tN-_oyAvx|yGBch*aXd$oH?(LmDlremb#!*AAAS$s*ndr5;wu`QqF9rUJ_|X zTk)0Z9&i6MBLEL3Q9|MqsVqMd5~m*afZ1(D$ZR|l@{p2B$aky|FGrApFX=ua-^b7w zjk*wz%pu2dTG~=rBtcR>XFJ>=e<pP~OfR41?*`PkS z7mHi^_O4BlajsyI(6ndHi<_=5}@pGkP1KR=M*|li5DXy&3mYCNybVN+~_c?~WZ)6^BGm$Y#rbMO)l}bXT(Je}4$UK&W zQie(qA)%r&6lJJ151K_tlBv>!QZm%{d0p!q)#vg4J$`?j%jJ5#)@!e|_PWq7 zCH`y;e7#m{NMW_4dB9qAmG&V1Y-jFNQdrjKTrV|FbbKi+9|HLu`Y2i=7*FZ1b0C`Z z`oVY}LNa>gN*%9zZXG-SHR8gQb$v-~E2_u|_aE|a5^MOv)sG)sO*oxd-$aem@ z6jPDod;@f--AyH@2oOa(;H)>cSTe*sJ`Q7=Xo_1Z;xJO~%cMif56!1qSk?f-mn zA?0-RqE-UW4QI#vcEaoK)sZeNs#IIYE3lMQHkU&Cc6mbNs{AzLea_SUFY5OxjQ?;B zCK4si9H{)Jis^SCfhsV+(b(Z7{j$4&$*5j&+|R_FdVo?qqc~gT%h=Je-Y%goNkS@G z1CNYs3oe)bxUlzqDd~anpxc>gc?(`Wcf^T*D)}<9(0<~*whZ02-$|%p`apoYU7fnX@A;(1#qUhtYN+}`A#?Mj6 zOL<}Ca>6>76m1k|O zRz*I?%=5CBHan-oQ^qKIc$sQ$9Va?zjIyy4*^i=9#;Pug6nxy@<~dW=uj>h=5W&5! zl=z#$D!+EHj(>zs%{*ijr$_bAd3hUtw($`8t!oP{24&NKRN^7fDRQ*9mOTbt1Z@bd z)Q*cwrEY@Wg5iSu02b;ImJc*IxIXwXHS@1s)-HQW&t*rO~}?dIHJ<>sqQ<$&q<7`op{rIz5G}{UZB_ zC!;g*R2F5d^t&KeM^$s?nRwO2D~jquscWO;C>rEq7E#tNvdhCW zRB;kmWDBzV7$q5PznGfq=VKKiY&^o=U!dZLKpG!SN<71Uhd2T+aCAm^+Cqm*pf5O~ z)KnyaH&#G~E3?VNrxB{XHU3 zs8`id-$v4S8v?~fi1*qiH5n;M{JU-CF66W(hAZ83P!{nTe>NU79wIqI8ArXQqjE0f z^@59q;(Q_q9=CItLr^{xWZY*LAs0^^<36|~qjH^~gYDhw5_!wUc#GFM`IZa|;o}Ew z+=&oHT~`X1veGgvv@7F1q|OW@Zty7Lm2p2 zcsvU$fqE*Ogp@HE%}e{Z=hm8JSB(GBPBxqm=k698*zwyX3>2Z@l;fs3zi}K3XRo?Jm<IT67J&wBNJ$cXM6&yyp zw~PPG5u)fG^HSF-N{goabc>ErMf^d)lrf4*&XBv5WHy&0V>3d=t?XVFmsBK-qNcNi z;@sqz%P|? zgOzr>6fAHdV|WziD@LfvJYx)eGd$iwNG@xmb)F@pjJ)kKQtm-NK1RIxIW(~Xce*C- zAY~~iSEW0ZCV@b87#oe1_eC5uKF2N{ENn0aei$Biv$5&jN+}^_Oh!e=DSgn(Ob~yO zSX_nR@%>KJZlc^J&=ncazDHa?2wpCM$E)lp%FAZKc&}Un{~FGVthS2Bd+lC*l3ddF zD|&E|lC<(*=K8bF%Sb0`=DTis&ewR-iMshNcrfQ{Jn6*!*?8ThoJ;r|)3mHD-8yDWL!O;u z`q$TZ(uoJNPTPN0#&T@ya_nKZAAgM}otPfo(W$jJuh@kBOZlu)cz=y2otP2Ld$)tl zzhOH1k6B;iNhfASyZ4r}xhUmG@fTA2Ydq=1f@mAVsyfK^Wpchdt-GcODW6V-{~f-r z5dT~3|3&^%Pdc&0`x6ve>o4`B6VLcdJ$$O?d=6iVtsPu7@!9-uz=pQxW+c*ydeL4E zQ0IyEyc!-hkxTQvrhK#6VCoqRD|GZ7t~Kp@olexx>bi&>fRR7TF(vlM{@+dH9B)9z zarOjGCpz+0C&##x1lgO&r4yHAHM@jQj}*dCg1^lDK87}tOD8VPT6xpy8MTlb=9t1U zyuDdE(JB9eF=40qi?LsM(&r7vm0|O@bmH>-ou0nPTTI-JeYAWIaEQdeeY)lU^1%@2 z5$y9%#`#@&$v)Oq`3DXT-(Xma{q2)+?Bh-3G%oTca;jW^s>g+oRj&gEGuP)@zTj?Z zVDo=#c}&aN8HpkoO7WL@+%=F>%fIm!+3CoQb4-cMhjVKAO*;r<#K(~xK9n8Go#OMXjq(IcUS zuEhSbnB0&x+(9#>F57C{<|&*M9C9jhswO*CbnzmET3HLp$NLWNx0mCeS}iG9%>q@3hW^`sNuR#I=mZ@_2gtHej_t5G+d_#PaWq`KSZ04QGF=C(XCivTpy9O@mSs%4F2w&+$V9r_0F) z6YW*M_)lKtlx>k?U0caiaG zc-K3bqKYaoCkH(fSsmNct#rVA<$qj49pw12McgHoIrsp4MxFGl`R`D}5?OV7unoy{ z%nSeH5<+=@xO8GnqQtEYPtB-H(!_e)Hk@>I#U@B9~4q zO=Q(e7NHf~!`3JK`Fw}#%Ey`dR!=&SD-o+XCSuvWja-^f?WDSoYoF0jc{mTbgOWB? zVZ4o8I`Lwn)Z7YZXQ-1%^u*Zb9|vtCmrkrnlpL3@Xoefhc#QX*bO_tXr4xq|@odDz z%3P}#l+1|Ie55*=xbNOF8LM)=dV{V!>l3+$4`P4DxA5Aae)(M)4aN ziEZpjd%goh8fIS)bp(H$D4jUWPTwt3uAIa(_^yz@fAV45ymaE%MEp7;wbb?b3v+5| z#iak#Qin6Sw*G+EK_zcBu-t!Y>Cyp(Ga?cc<ZIMZ)NeAKqf4D# zOGgq#%AQrzm6P}k-!1YFtx?+qrW3~!MQ$vX2oAqu{PUzk*aR@0_$!fr+UmBUqKn_h zCV2dXiq7#!WEQ#O;esJXBdpEj5e}kl0{NX}7MXOkX>jO<@tS`ev`t`|uQlbr>iS~g z7{f4*lY>rXxY&4`KziWJA~inB2w~>pu{bA8wwJaEOeacY=6|Yw`LJ!;TX=l(f4sB- zU^KW(E;^+GV9>2@$Uq=jH6*Euo`%=A(7V;{5Kl`8a=aOX~QQP>X6HPPA z)w`v3MjLrG!lRYE9IKUn_hFmBbmH$=7N>oZO_UEL_8?2kP1WXg8#F+%*d zY}EKs_8E8tk0(!hsnI8VknDhLg8qH}a-kw$$L+KKx;kpuCNQ12EHi3c@fhwP56}EMGMe5l;SUxN|!}k^f9J1AG?kE?cy?77xOv)*X4_+is3RL zGb&$)y^b%#tZpSl5evVku!mflkK-++H~nXGK}PnOWezMXo6%pc(}3A0Q>)}*bHQ|C zQD*+B%ZrCXzlg^>|M5!v+w~({D`junw+pwQPP%%y8TLa@C$?lpUGFWHk;pxTJ=6Ge zjw!PCLr?Shqv-S{3|vmdQsZA%Z7!Hje4iQBJI3|8C6)_wtO;F{!sdc>Oquy7t5aV(@t~wc`hd$wy<^$-_sZ&n( zATlu+5n$06D70Gx|xw&_I85lM5(NQf3%4o zd)B`{+LTU|&i-iA&ctPppF_MdSv8J7l9f?$XJS?1hWx*LI9~Ox&)3x9DrObhKC_LR zZi#7h0{n#r;g3G06E(65-L<-e?G3RtJ!ucy5$aZ~aHV3kyye{$82_~$VLEYkR{jFZ zTlz%Yj%~COt1bL{JHm9LSyr_8toj+#vCfl+YuSUEupMDK(JCw2T)j}ni&$Sj>ETv4 zhmZg1m)bEaDmfr80iN z`nx>(=&1kNWRa0(-ssjP%`ys*+oJs0BYTrY-I1N#WKnl)&t(1mS7Y6&Wkh+eBvN%3 zn)|0^Y_iCYEXq3VCGNnkAkMXl)zMv|w!%v%%0&fseO@=?fbLj_W4t@ZA-g*YUKOJP zTN>2}Ui0u+a?)!#k1N84dFjNdQJyz))e2s3gnQ&(82Y@{f4@ zA+Ia+!?Tw+%u6TEj83mwvT3+L7okk0_zM~DxRg@8vXQR^)kJHMqlX&=_8C-b$<-zN z)6PvMR~z5;YA-#t4xW^9NLHsa$HSTjIFAjfM%L-jnvzl6L8#UjYfQ|S=}E+cj}6TJ zPezEM<{xTOUvosUCYqR>M2IY~g%E8sUS&T~o_la@jo+~_8_HvQkmrmTSvuxJsy!z? z1r%i;O>-1cc2?B3qiDO+e($uCrz*nNU|f;gnPgP$0R@&2MHL>@1Y%I+$Zl=6RzZ)V z^B+=D@syFJDkyAy_beS=TZ$ZkL8 zdz8(0$R^JQr^#5^*^!)#)cwbMFs4i+BMW#K3`J4FnZmM0DXeAW=NzjKW=W9iLsJt> zUL!MYZLGqr3cDkrvjCFORVvcBkB6|F=vYmk8Fos>vLQ!o$l#F$j6Q+Se5CU#$D}VaiYFu8da;1frP8e?4=U}iQ7`2L+33l9Ejfwna%V?l`A!EB#w`(&(R+{6 zdM~u?2KKn*A>smrt1ZaJ%CV324DHiTgprKHZR~3AR*MiOS&)tODsy%8=jKu+^DP&c zuLt@PQbra-pv13Yq&X-f@7wR!{i+ZNv=!%4O8hNGmfNvXiru=rf@%Y&A^yv0W`3eW z3gdz^6|XZ#k4r;&E(m4qma*zQ@p{HC44+ws1hTl~aopSEo-&Q1-1Frqna%Uy@r{mk z0Lm76#yxfs@=R6f@9~k#EQCH}?-ozUTec`VfXp)j8JFAm$!T#hpGyugl9 z;%zwc?Ab0SEFV!#9CZ5kS191EKOXMQ9{86zXl zb**IBddNxN*+MOJchc$GlwDzbD*}BA-tkb|Hbd!Nj3qu6Kh@ajOBM7QkY{b5;Ql1; zDO0vBM2O7U+VWG3Z#vdDP*zG9H-S_%$TMA~|JFy&|2g!Kz0dqi-m+=eIQbQpvEXAW z?Tk&53vL!Zyi<-cve~O0rNoVKs}Cv2Sit>nVr3F{w{ql_%Kv!j&wv^gqmmR$~8KCy?=HbZ84$hXS4 zFz$%Jp1i-3GkDP#WcROC`%GJ5qe?h}0kQGQOp3&oSsFnGMsu`c^V zd6Pg<;0!iE*3l~cyFT*hKcO4#{e-<`<7x6y`Q0Vs0UMLYB;RQ)nl~2#Wn_6cNX4PV zsktih?RL4&>61t32kIizrhbQ>1EsG=r6D1(hzI1_CBJU=s~%}R*WbqwZn7{`wt-G4 zl|60|_>qH*_^4`DFHl*WAM(KO*y(Shn&PLOsBB>!e${@2H5O!}Xmknj1`D_PC;hU1 zf7>m{#(u8ubpAvco0*))XxIyI7K~w}%1)mA&|;}k^ocL1KdUX5u^=1oEzmgQ+7`~} zpe(H3h0x4`Y`l8$BSJ?DH~cIi;EE)5+0%0Ce7QJ)1S;BmD#!rw7gA2N%L@mTQW(!f zU^QNTk}t6BIFvU(j7S1bsHQXuYPepJnfTwSQR?gf^0nKSH_lE5T&nJkH>2kb`De4 zC%=yHnFZOB(SryK5^enOO>rs7==bAlZ@G*I!8q@HrT;ObB>TUb=BlJj^*n*?RD{Sj zMo=P z0jGSTq|!*BuQWc3LeuwRmjKv2~xg%QllRsKdz_5g{f@iI~KJ@(bFCe2QExn zVKos0`DC=wzl?K!)fo?Xh&trQS4tT88|I%FkKh?axqg%X?3C(X$o09jP@IM6=Ecg$ zD$cuHa8{gw$!LmwnwOK3wvQAR(&p8Vq8!NBKN6ORg=ZJl4=3nC`amm@tw*hFIig&( zII3I)?CAJk8w0OUUPq~IJb@=S!N=tPCrZi~Me{4jM+!eB^ZpfO6{lgtY1kk?1N+Rb zEaf%(NTD5{s~le5OgY~UkD^PO3vX|)QUqPd&6$-Jw${Tl6)kF`j&|~?iZi&1P@KUf zoLkncl3?A;l6VHaoRxK>)e`QpPlFL^Z&GM8<^ea_+PSLY1U?!bZ(XQg?OWA!CKZ^GktPSlH8v?!90 zGBW8|UkCii=i^C1^;RYkbW1aHd)0IX7^<(@D#ogpe9NbMN!vEjO`G3rtaE^0! z9zwovdVX%`CMfq}UddfwwtcqUorE4Vy}F{hx6QI;KB?l$#_BKxR*Hei=w1~1t9(*~ zEYQp^YzI$eTN_xZTZl023B#gMGR=<=Ilj_y6@^=BJW`hV%{om3r5WWk{R%@T+FO)~ zm%r)8XIeNvQ5ZqU0G%Yz7qB(R7R&LU2b8)6*eichInJ8`y zoWb0SCEwOr$CkiEi}_mP=eRQ#a-s>BD<47{br+#{GHMBrFLanPUF9Mn6%~YYBLya- z;W);9abbh&DZw>V^11~`e`2Mft<|0MmC~H| z4;}qC&UL)SQC6d}p}CuYkZ~&tlc+Z;%Jb#LYnQ=2?-JF3gp>s?Cdj-yf)ZD;ap?f1 znOsrLDd6i?THj8l&&MSh1>sf+-JE9kAmv#?%Ba^06h(SG!Y;25kuanh$GOiz-a&{T zv90DXC6vHU@y0U4<$VjqXMp2Eyii1Er))8u(ts85VTG_U@Pp=m8More4Zi)~AEEd{ zE@C5v;>>=Sw;Cm@cq(cLkBrIaW&7MZM#=;Bk-~4km(o5+uJ}YtZ{dAlDqE3GE2PS} z64}}>_`5=f5RG}fQ&s0XYETz5o>7Gss}6kv%B%${?>QszRyfEzO(-;DkGBU2U+Q z93Cg(!kQejt%Dpzh%dEm!gwW=z_86Ybb|OKkXLvo3IlHlkHNB z^xatb>D1zID}A`{5aRVt+2lT@k-+`E@jMh>p8lZyh&h|&@O%F%Q5KlvGREe9WL)OmF-D{ zxCoTjUl%Bi1STSlTT!@AVGUi}z%I|0P@l5{Id1FZ5_oCcetQ=xrI7m~FuuuYMl6yn zfzOJVziYg?q>>H>`C7u0Ia0;3>G3+d-1D-8THHItA3Dfngm{Z>2VYS_30$*`yIvJP407+WUKsd1cwAtG zj#>B(xkyMEDcfsulpfc$OBpCXRo3azUr0XR@_;v$MgmVFj3ZH^XbQ+Q^IO8eo4}cX zdq*hFiz>Euds{w1pM;gII+{?hMTF-ANOqs~uG|9;xDX|ZUIH2ay+>{~**yq7B`+Ixq5f<6l?J!x~rpJ&e{JLMU8z?CRb z^e&i;+J7Sq{B}5xX?6?6nas6y+AjG7{Y|WNdo2|IJ$rt>N1lNP+=0UF4anoLy~4m} z!@VBhz|4~-)l#$5$ z-!!o~QV?+H>e(c?KSEED8_$ zK-%{2!oXX=x&1gI6xSrYBmW@``ZTP(er}(~|CCSQ0hge#Zwg4OJ}M0SCO8ktj|;_l zcy8;2WAX|58myeJHp=;{_I&?@JOdB-ISQK`f&5kU9=Ow8LPR#|uh%+dfk0HVetR$olDprOY+HUPcjuF67?O z$~0T=A{DPamzCbF9G{{azgh(k!dMB9AkeK>ke>wJeFo_S%L}E)YwYqxB?;459ujYG zknITZX4{&ds)Q1luQb-Jk~_?y#cQ(1$dyPvyR2&VHCN>3^5Q`mjWY};2;2_-NoZTzB^xQ3+M zlj+Jo&yJ!D8!tO%Q5I{>x%DFE)F>t>k;A;PWj!L zN+W?KNXD)8^_LpW8hDzTHn$*znr-N~#@q(ICE-|jJt(`!E zvWtCN_^_gPZxQme)LR188>)?0s)8&FVvC)4GDo8j{dLTJYY^sIkd2YHbM@(2%G(}q zb`p+}&ZQw{y@tB!=$|DXlzdMUV5b8VD6WWY^r5nY7G&e+=_fdSTPWfIW8kG*QaPp9 zDKFQzR<%gri>n}8ZM70gkDJ-0MH>mlCaLU)6B_>_^S&#Tk*2X@Te+lgiTh(@%nW7f z(m2V+{O8Lh@P_dCQ3oH@UdNZfJ+WOnx07SgCt&5q5scrn`wJc9D1q4r<0~jUlf3|! zAMG;oLJ42Gowu&2E}^E3@yp7WgvbJuM0Q*Z@&x-Lp*Wkmb;P}~-K%WYBe17F6+UhL z6APRe62fUUR&;1yW&T3?8)SybHCr278TGmJm(9zYOQ)4@Vf2=$XEw$W7aIdV z{c7pYf$?;bCZiy1k+8_I+Fz$aN?>0Ud;flo(nycrvrF|GBqZ*mp>0E86CT%ost5gE z6Iy%X)$D$+AZ|Ozax8^ftPIU>x|BS{B~35i3#|<0u{9A`lK`^5Px9H8kL**0N#Nex zxDDhyp|fvgMzA_O+nyHqG|8qj6Jpr6@A z{JTv$#^?52d6mNKzL7AWId0!xa=Fv^pz-^@3M>~!tza^e!g|g=(u;uHirpf;{{2*6 zYQ5?u-?uu(3GyD-v=FH&H8AZ8rD^O?*(GyQ;oLY|k0+Pb{nv}14o?|b7iC?injz}|YGOOtCw(U0QeI3n%OshXnj-2P|b zUR$N`g#3V%TS3ONAkUx8Ct82#p!6p}E~@6m4~aMYO?ca{!aZ;v=3A-j+L%#Q^^^8- z#hLFie-?}zy{@ViWx0l{r;D>GPQ!Ar+YT$FIGYiSQz5RsTNwDJJj&)mW2pGYI7mE| z!1V5?>DaduAb#9J>UTvc5Aq1~52Xxznw2@WO6Rw4>~6f+*nEic^u}GZh@YM9;}&8t z{@c+f4OJ_WzzxMPp}cwSIb0a@&qgWAAp3N|%BYXh#4X-XMbsNF$McL`j8{%kNV)LD z4W!pV;g@%h^vxj6|6XB_`=lH+?*~stXYl*e!z6|ua|zFOlb1mko6Z+qY1+FpbiI$? zq?^V^66iWWF6}6PWL-pz=rAsx4^{Q_MUSPhEyk5{m%ut1*VJoZT;xq1GNz5ixAG^LRq53|b(D9aoABHU}C zGF<`_=Lk%k8`s(8U03;9nh`rDIqr22$|vwI1}oK%{%s0bu3~-~ndYQ{%wd>cdGMuw)<)Lts}!#yu)cG)u=66tl9BEGKn4m{ zPPMhTlP*JvTRG{SPsv>Zcg`SlYbf!x7#Sge;p^vj!@|)Mymxv4fq1Hr~1Q=5~y z?o4wy$?QFZz;cYMXtGZ^P-U8I{JcHBMP(iqfh#1)&9(eKvEM@+o12bn5x%i-Wq0V0 zPN<-)`NYPB8=2j4Kr+c81*t6$3m0U89 zcdP{77M=Nnr3xh;_aPB;_fR@rpx1I#>lS_9Gf{F3=Lq_a2fy)+Uo1TMQfE_azK0Zdbwbzb^ zEz~+Hxg5yNiIu-`;oT7;an&<68>O*eeMM7{|LZ7QkL(al<{Z_g=?$WPxx@+#5F$(#*f=Yl)ad!fp~?5 zJ+2H1tW#6$_blhlqpV2qaQ6VByMEmjmLPO2h4p_vVt$Bgg%GxE~RiU_oFQ zyGAH)mbk1F2K}s8m1whlUVKx0{yPL%?ud2Xm17HIC*xb6D5P9i>IKs6qQpDxzGH)o zpMtCa`#{EHs*Dph_Nb%k->1&vlnts32@JjwSZs1Va#dtTJ5VyDs0}Uj`c&~GFuF2- z!C3JOrLPV0M$uNGfY0?ztaQS=h2l&o`PAM8 z#+5%%owVl%z;i?6^?$4FOJT(v=ZCGl`N8m<0b~qmej`5d)lRz7?n=bZF@uN$k!hqB zR;YgOw%_y{+>1LGflp6koNwa>jPbKhI2AP&NnoJtG*aUC?N_&xn)Rxd5W3EbssenG2x&&a_`A*J{Vx5L#JyN4ATc zP%5i75LmSVvUXU0H`}j&KjsaNSzb(XW@-0u zu9w<4A4-Q~+-3a5Shb@wYmG5>ioXII{>K zH{wcP{v%b&@b95DpiKDLTQ-)H?U&ye8Qa>J=Lg9{jk69bpp5Kj4>DJd68FWCAHH3> zJALo|D%@adftgj#W-5Br=d_3QYX-@>>~{tD!is#@HQp#|$~v>Dwg$a^W*+==pQGBi zwW(_3Y%AjYbj+({<$(*6Q4RAoMsa@jx$ub*~gz~jVXYgwd6iXrOxjOzIB#%c~ zc&wx%%?IhxQz~TO#Y;)43^FE1rx!EdY20XRm|OW;Q<$^$zO(+Tg$(1^Jj#$b*RZ@o zInrrF!l1vGPXTVR&qY}2v5%*pYy5e>gD4yE(4jGIw5TK zx0DU$9pKE`)AV_0;a2>1V`^fslaAmA$d8_88c(nwo0q&k#ecJ}+wlyz#g}u$7UBON z=KMd&NDxjx<63G1##<~jCSZq&4N)byqaVF4q0Yc zMw#i*%!LfmTnp!areVD{!|e_Wqq-&|%BY{ezND(Sa7rg2a3P%-iY|&83QjwjNDRmO ziswOoLb521`iIemgNbv$=`gQ3z=t^j`2J&SjUwwh`klt==P_j8fd@}^ODl?G?ib4_ zOF@x)l+E8pXR@NKjv<~hCZnHGc*;c_R>2oY=i!*M{HF666lI^wrBexcFB;^Yrn2~Q_Bvsv#~75~(CODtrCn~^ zs6q!}u7o`ZjAyDUjf9kuwGCC2MtYq0bGaOVvZAgHHCGCO>3rEvwXIw=C6vIp*;oyQ zjt8uC(+(OEpJr8?Grg;j~oT8jte z4*Mt0RI0#h!Q=6pRQ&_aQbr{385X;=uP4W#kHpGdH5jk7`_uL1D1pXpd>(~CA^C{6 z;KDtHGxHKM-E#xE2Oe-83UBo{R3dTq%?FtaL}8|?k(9s#euu&g!r8)xjfJg1o=@9T z$PJ&Z-J2+F-~q2k;cb?4giFoWf^-!24A#v}mFhS?oFID!-e$BeAtOfW!rubUdqL(x zE)BNkZl+XoQuuv3`1=otx1NIg zjdoWkCcKatY9ce#1S?%}C7nla!4QD4@Iy-XF>b865Go3SA837DS-j?J7!goMH*5*GqjuP(}dk^C3u z-`gnp`EB{FKknm9IbX+-zzlUe9W5ww}v zwQ|sTe(gLd9~q^vGMB1P8QJd~&KY!dUkZ8v^vbsEQ4!Vh9GrU!S8A!*I`x4I8PA~b z8|QgFGec)-#Im8Il=9YZc}%==eKocl7V15^&AesDr)A8A^dx@o?rJM1*oS~i07&QW zJqe`q-1upwEx1^E;6fJlq3|f4qwrvMj!-;hJ18=mz99(w)D`J`wr0t z5Z{b6Bju&HD3>#g;&e4eDO`*D$r_Z4;XDX;ZC=-2^}D3BnzcBuIGKN1MlDB3m(pKp zw;9Fx4cgk(AJ}SARz8i&$-jxM7Y~q6Q!9bbGoOM^J=(gdy!_8|%;DA5go6@rNF~LT z&O3{z%DNI{;;@zY>8n&<+geL0vQV{F(REkAJbGG6kK2`pJ5iV!$DgS?bOvQsgv=jU zKW&N5*+HL!FBVHLfG10x;Hhjbh4uKhPIP3YN=6?%C%w%Q`71(XC6%p6XO6&;cY(Y( z#+({YvzLiW;rW5BLDAwWz8vQ{j04ZhGxUc$I3k1X)42}E0ILZs2$RuigJldD!k@R3 z=AnJCJ44(&)Xh$_-)5sY54?<0m@po$C_$;@W86#P7ZN(gqp*YkkFvHc;y+_EU(C@H zVgwm-1{ZX39_iu7RQz3vJph?$4L7b2JE*z6%?d2urhqfEWUj*T3K#w?dl)AaXSKlH z!b)Y-F^6hcWWK;PAf3s-^1>zN-;@>K2-1+Pti;M91LDv;D~SK*_|jQiio#RQio)+~ zmBMZ1sZutBd|suBc*&}|APfPygF8)F+c>|Pd>Vr^5SOUQg(~9}jxWv=4D)5qiI6FH z=izOmI7^)!r*?Hk6ZcF_iCx)eP}SSqaVuQp4iP`*-Hh9vW_V5IFYt?NC`xyb9}zlJ zJLaX$3m6}9&6asU_0oB~a&Wzs-Ed|GTwh;LkTv+!zDGRhHSd%EhxU1Hn)qFAKLtm~ zr;_n3klSkOLY~+7?7li$`U;RPWsLOuou4trvXzzFJ|M$E^Ugj8LI2z5sbeMi$jUUF zm93P5AEV0>O;`NDh3p`K!pq^02p1VMQR3@%tN-|UhB6|V`|z1c8Wg#618VCJ=)4m& zO98~WbpV+fL}AL1I1I{WOAm@7^Z7tqOQLhdd`#HWygx|x?H{b8<|yq``0xbYo`ElM z9x9Rt7H&8vK~eVUrCd;9WX|%&;Qz3c(~Uv7d=H!LW#zv!Gb2|osJjGJVXe9H9BdS4 zPSPla5!n;62IU*~b5&@1Tthml0Sb9mYyF(1>Z^qGsZRBZG4Q|4^DR)c_7^MFn5NaOL72 z5YEk-`8Z$8glx6;qK0b9q)Bpzrd5%rL6>hSxs3>u2zI9 zAYUr2CFKK0KD~jil0gYrvyP4{olkbwm9;m>)chPh#Qp6&S-*0$%4-x;oX3)IM&ri! zUzQ$}J73iKUi^};r{+?)D{^_4c2*74<5;{@3TF+Dc_SZPNEsb%^ec)f&S=5q$vHu2;3qxZm-#O0CmMT$D0)Ev*siZS{YbI+MkX@AEC^ zQUlZxkdvdR{jVxdXRa5DGwOsh0yS=ZLwZoYgY)>}P2sjn*&1y*aTs#EB^Sw2H0*6P zqoBy0*&tg-qes!>??A*RyE`bZR5u>ox>%yUM9ss~u%4jZR3eeWQ zK2)Q(r7S~zm+ANNf_`{Iwzk zUC5(;D`(qU%c&|m|M}Ocynzc-*;6EAu24z8y;7wPsQYG`EpfQR-YE?HV)O3C1y?E6 zGLS`~-zX14nv7AL{YT8_?UGfTI~_O^OCaaeJ{$H(Iba_ty!^OV)=#W=-6!Qe$6T(+$AmH)*{YUc_6##UC+3+?+Qb-F(h#BeOSi` ziX6HA*}4OruTcG=q~ctYKzc0{Z9^s{@PJ7ap2q(v>}H+@dAZS^!P@VrQq97LxfFW_ zK1W$A^P~EMq8yyHP|kv6mcaS7&y0h(oQn%j*q}U51DWnq%F9E_Ly{{RC373!lPVa1 zaZ=(}`p9C5@_8rmzpoPCtZk4H^-Sc++x=(0i;?*ANSx6#5oL|MH8Ufz9FG_I%glRe zxr~e**~U2kpps=X5+7mMnqy3S-H|(loaD)@SBE#uNF2a=?4(Bl#kenV#0SWq_17Xk z4pihPK7GNT4-oeGeJ|0g=RmShAn*2nwG76A(89dhm&A;2(fV@?Jba&3dE*b8OVTTvs}XJVWCk9|o#r7v9Y(;EtB zq==YUjqS}GdnoFV;8dhw)+io-$Y}=--=B1HX@}@D96b|7^L~(VS~yY`l^pXIYV?Q9 z9+S%AP%Qu56V3~}wO7GdQw}alze(>_!(K9~)}>KKbL2KB)4q|}Z{PrIrSiDy*;RYN z1)PFdXLVAdY|C>v#})FfqLtC{_g4ESDP!m~bU2jX6)LX=ch@g@=+*iq3%N8{moy{+3mofi#>J4zi5NJ{=Ny zQTp0P3ipYbd9hw>eKS0Y+GPp*6v#`2pbNR5u+qfV+V;F3g%9`M>?@+sa!r5=@5slybvIJX#{?*7q}?n0hT;m8Vt0L2_F{#}YC|FpBe1t@-$pvWm0l2TlhAvg+DC%TrP|+eZqMjiqH> zWxZJ$DNBuiS5<_d3wg~9C7ygh{mJ8YdC*B)-b4#nh|C1hZE7k%eZ&=c6#@(CPE*1g zK0|oN9y^SGT}P5FroFH+#mr2EAh) z@g#A0wh!lZ+B4)Cc)*M6$|skrB2>sI`` z^PwO=MDx}7^J?~)LU9%v))OA#9-0fNt)ps*cQ%*K?Oq+>SahBq)D|uV8HHIHjg^~O z;_yWD4Dkx)(z%C5VS>D_aJj8g_@4EdQZ5F$`7)ncPxBB5Kvpx<7cM+YSf+t|mRl+6 za*gP%lGb&6an?YX4|7h0imdz`G>S9T=Qs5^W86@9 zImlOQ8!3N*ziVZatg#-qM^UnYF!*$BC?537 z&-Pid&(?FrpLd+*E#z~lv4?SWJ4FjtA=@FLusc?3>AOIV(nfl+y|5>xt2E}iy)Li=bYTyCor)>gHUG8beXsipW%kQc&RiND-hc-XwpdGbks%%iz1O>S}f zaC~vvhxst)M974&%hJm^5ofWu;~X}MGjHZN6>e3uqgLj%k-iY*LFf6(PvE86N~sDm zTterz+I+Whv+>jmb(AN&sUCKk|8#+rz8C3CiE{<&C=5O?T`V5-Rh{H{1*w>eBm(`f z#TCfV~=}_1jvb*%XAd^1VNH1`MunMK4)!n8ts;{J#!tpV$hYt_7 zjN+{N>LF}L8S%0Hq4|eCigX_9cMyJN9<>)g+D`sK`3NgZ6o|tz$P2`0m`mr~X%r^N zE)pg#lvN7t!OC2a_Rvw*ognSuV)2(R5gs=0bE$k1AamC)%b$H!x;q?SoCUk)?>Q$z zu7xhk&CZE9?ZI&l8^w9g+;MjFb9=BduaoqJAnl>E+k<(j%cN8VX%FbM2lL&=&Bk-m zI?Aj4R1ZhYa~mi0)R_|J4eA?(!RM1+;z9rDW~J?KpH4(zY~eUp_Lt8s#?i*lhbdaH z3VAAr!f$zy^b-Tcs}7dFn{u&6pYNzK)rAl181+rHZ2rk`c?wxMV1DBWDdIe;=%O;V zb4f>EF5VAhTxF$GS1F~h5FUdwGj5!Am6W?cmPdmuNHd>feQ|f`FM~`6nb*2n{K9L6 z@9A7k&h^3ROg%*7^(b=Gg5J!an~uDIKPW~WPQUZ z&Tq$c^0MPUHc&^o2Y=QzT`#N;(!y_0DFT1e$_utiXX@5Cz}VCHFnOS6e3Z*ZiC<@B z7o6L8=WxepJYj%13Mj0hX(K9#8I0xj)26AOHUv2%ZKGL^< zJex9ady9Cl+k|s(6&CL+pCwiblKJ?wV;UA!8LEQatvohD;JhQmt!$H;NF0&xY zL_5Wb(-!(XtddLMByxawgylUxmIQv6i{x_alR|M;zPQqA=s1gU=e@jP(3Pg&E694& zMg5=~9B;2v9|Cz=C3%ZY(iFQ9fNr8LH#L zwwM-nK7w$z)7&sqxs0cs@Q1 z84+mC2#kX4SbwDs@s=_0+^eJ%F&?&mVx+?CKw-AX`h*wdBhK3#=3QQrRh%7|;0*ae zPP={XTq9+XeWY+_zE;-p)(fnYGR8RnZAA#WknX@re_MMvRZHi5?{JkuSm^-3Vm};v zKU1%I2A|DM=$UfwatqJoKLz|BCIH`2xT(g#*TA`Mzb_O|85yO%E6vpUKA7|4nu^N(Eie$Jsd^Sq98%NY5H(+sw$EVI58ilx%h^ulkDk9YZ49koB{xmeyIEDO>c?o_D)pJHXUtVl$%37 zcCC-a8(UceXL<(CyUw2ogHP6H;z3VrlK%qx+)V@?glw1f9phHx55`~iDq64#`EjFA zk2|IJ0h!(ZMtEeGl=0@dcS|{CpZup7178h~*Cxth%=lg|5>iHG2M`B%ur~ z-DV3)K5V(sex;GXFU;5yg{F*STq;w+Lpwh<2B8K5v9~eVQs1-yU1i+dZl9sDhy{U5 zyL(V6ad-P&TTRnz)xJi!)k2#up+lWeD(hGfm{vAs+@W9AL^2Z3vdiT^Dwh%>BX`&+ zERn@JHggjgUObo%L%tGYEER7kk63;ZN~VXt({nT+KdYVU>1&e#etK~B{WTXdMzyss zX<4yh{~z{9Pk$r--6*W)MM-5V(is(_$DOiNQ9n7+pX20M`flOAJP~@Yyu!ez!5NCg zLUGn~*}9aoN>^rm7sw?iw|q7dfyp=XfIr%LCBFoo;@+j-L%@ShJC;^J;mlt*|JXQY zoct$&^kMlFQ^-20{6cXm*}P~0S;bje3unb3$WXyP=N6RG!#+~DHvFs#8}BMF@QbQ^ zjB);-ZqwFf<=SASzpXu-sx{HrlcJEY>ICKR{)x)rF;(i5xm2kV=ulj>^~b7Gg)EIC z5%Z45z)KXCa;mYzJxbLDeZ zk#Q>QKX2zh<(9k788~7*FiDl!?S56}4wA<$NyD8u0>cW&d!UAnCgdVwUcIK2z5U-`&3CX)roYQZ2Cq{OGoDuw>=3|YUo-mk8d zUB=w!$S3GRy(K8+b6ekbsyimAzCZNE;ZhAGC^qyX9Zx|U~XymsL;xvh+R_Hy=8QQ*Q#FF(j==>x%+g8hO& z1<^x5Nx^A?27>bhmkDkV^cUPAm?D@1aIfJ~uAR#5M6O8P<*5(oGz8pBrsFb}2RC(o z5V#DQzhT^oI|Fwon$k?U6>?!bS18VmJv_c4L~kw^3G4$2XI%-%L?Djw-MDa9&{Fd4 zc9cw(TFK=y^!O=9*@nQ|Rjq}6FH$TSnHmB$IVR;?TkAUM`bU(>>N8ccyRgQ;_(&3% zFG1iJ*iiw5Tw={DUMxkNTaQbGjmPS&>YIejF_>?dBQBj8g1d#Un;*GLd_PEsY~>ZK z%wQ9T@#J{%66Vsm)}k;vohUqLs}z=U-794n$m}V|zO&{JS)Vsa`tu<3N#>RA6K^$H zIQ)L$0XWZJtej4HY3X#CH83E%4 zmLHy?BS@g#U#;>Ly+$a`nmlUPL$`rAd*MqX{NCV@u@v4-nx&k~B1n9MV zn-b`}FB8g}w;ekRgFdXQ5|t$>H*?*^KfRU!4f3g;uD(ukU*ny|`8O$~T(}Pgxy+$3 zfJ;kHfNTQLQ#koXDP_%P!I>$bI^xL=T<9^?Oh9c;7%xO-R-ah$GJBkNvw}!q%L?;p z#t%tJNAL7k_t7DQywY)tP@J(cJpRLB4np}k8u!{o$de2l~IMCJxaGuWiq!%&PyGN(~PLSn@<~RG767=H!Wu=X>@&ylb4Kw58F5#WiRcBZE zk=I)H7gCbZJ7u&^ZcSOe0ejS$8ZlB?3w$A*?r*eEoXLG#-xwvIpl`TCylZ<64L{}? zI7?A^h|Q#qdBAmd%ICIu%EJ<$=BMV1OJTZjj(FBu%>dkFA1UbdRSivCh5cPUfnPsX z`6+vkDsS!E8jw`DPs$8lk7fP}&LwlYDr}M~@3jZTp9dKaTA6O^)sIL&_hIQ$Xk}Ig zgWTPK+;E#uv;No&>B~Txym|SV;>~9Xhde6$2~MYLOE4Ho7aVFpDDG9Tc&f3f4g<5j7n#I zSTv*3`2bJm_`go59Bwu~Y<%7Lv$6CevbHr2FwQZ)Z~Wa@d4_yC8iyMf89y`rWvn?< zK3$Asjn5gk8S^|UYdzyN#(RycjJu4d%#u$NTy5hVWDQAIiBAa_8|An&vHg2j@T8zu8oP~Bt| zRoO%t0vMM(qksYbru&Fb!-WUoi_~8U<3f(o|oSO z8MoS4>jla4jWu3XKpB~!1L<5*;>=*`B={GEVFQG01H_h+K zPXaenHs*a@X{5)O+2x}R5|&YuaoRyPBE)@ddwZi2O5kqQ`1$AZ-fFD53C9^uQ{Xw0 zKVkaTX4L1LTn^kKeW4(d8S#*i9^HIDI`5}%RY380H;GcFOwS?5zd6d&#v8UNiez5h zw-R_SbOyBBg)7Zp1DU@8;|9eQF}U>qN&)KQ$fXsGJ7HvE!a`>irH6(0pggitDF#}Q zjp;H<$Tlhji6=R#D6iSiu5#@~|ko!7S^JZiw%%Qyb zN@d62+E0`fG$2oQUDwrhgj^+*xE0wa{~g*A*v2*P`3¬P`*D8|ch=@fquigRUx z$IBe6;*Y9y349U?&O4wW3+R-7BQ6~IXXp-l_p`Tb3@Lw+-z730v+)ZA9yA>gE;_1! zGO}_IWYsuIT!odUEFQ4S=}tdab#M+cKe+>?K?!WD;YyGacf&{rv78F(VZWL$sDKk? z8)%z^c#LgQIopS@O|>A(FHi@6+`=WtlqU&1!8U))_$@(LZRFUkPvHIu?z~+OWprv3 z=Yf=rKRMPHf2oub*rf(eivXE=lS`pe%6DdhZ!4C8Mp3oD51AKeWNd8X_nI*KNl=ylaQ_&xG++%r|>R_nTPxwQXCxH(;yAq_t(=qb?o*M2v)AlQ= zt0hh{oHB4T|02RV3$k(hT}{WmzPd2*^)-~Au_(+kfbkD@55ko+xA?e)W^OJIEk!7h zQi)~b#;2x2tu+R|yOsiOM`1h##!c-Wgo(<~MHU`*4qso5aE%4ocrM3|D6G&1S%p|z zL7sV}82aWJ#n1y6vNt&jkI^5M9=MSA%2Aj^m?b@MA(IFwy!0_!df-ClVNrOB^jI-| z|IUa69xZ^hSY%dFfehULA6?%aSVi%Dzq_fMEXhlKFAyM*5CQ~9r~v|m9y%<&h;%{; zy?5!ogY+)V1_%mR=wd-Y6fB5{BBG!a5fKG`=iJ#BLO#FmAGx{r+;i{F&hF04&d%;z zPh9Cr?gxTm3QmcYL>q|Igv1)OjpQqx5?31#JbB$jTo1%g`_!6|VQvYAK;NUQ_g zLcY=|F_!`ak8rjUSGtmif#8CG#0sWYs4E##*8`+dJrDRwr^LAo#QZKl59jZY!|_dT z?J*RYh~N>ZR~iS92e*Sgxl}?FgIY98*!yCkvuH;`pFwrA8CC-mGiR_2O zE!S@Hl}?FIc>%#Wv6r~gm23l^Jd(}ayU>#FereES$If{CPWg!lY>g2+!Bg{@9@w) z7Jxpi&-oWXNI@RUm)>HL_wVBlDV-8?0YK2f?#HpI3?j}`p-^JI24D=)AhELNZH6jc z$*+K*#s#OuaQhu1-5@c%J3zkDDY2#%2xjjN5?8vCbAjOIU2sa&`ynEYA@RiPUGkMq ziTwtG^_1@sSGtl{f#73OkoY*o`_z@}fn$K}HsG;QIwhJCAZBtQ?ldb3OK(wsIC(@i z2QEkDMJRGq_yMN4G*mee+=4+x*9(bti6V($@-7{eDMC^30v?2=0?>zB^~V6V0>Hf0 z2P|yk!`v06Q=%^if(LU)a8y2%NCN1^A<-fMM*9PaT0Tl#=}Kk;L0t<@iL+afzeRs? zjC`e2;;T-_$%#EdTspkBFT_=m+%Yn zl}?EXHXxY0|B|@Ul{^Lng9k`7G#98V`4(yxU2{X$xY8-{u09ZS#HdNLY7y2uRdic7 zc=I?B5W-M2vy8>`$zI{fbvv!qU^gI@l30ZtI<`I z*BSw{08~u;0Y!^b1HNPdsN&v&Ky?9%)a2Do00S`52a3FfBv{TMxvOS?J^*wqA%u!Y z=pw1(PKM6elIu(@jD1RjHq!5kzihLqPp!P~WUpuQvQLbb`Gk6#iZ?+Kk>igS8B z!heg*--1Ls@jVzw%&Gw~AA^V+SfN~!_=Bh#Gcuk#P8BN+|6{$WuVoZ|#N!(ViBB6s zVsR0&U>##i%E!W_#7y5)%9?WA7eY7tNJ^^c1$17X-$A0C5WSNxk*<)K1sDB_Kap#Y zcp@VDV3YbG-9A78@M^qGNgW3eI_`Ew{?vxJLXy!;pz?uWy6_3N zF+%bxNd(soIGC0VB2(!b+Zy=(cLHlAwgW%6M*u@)=w>azSR=t#lbeY8rwz58kcr~% z1OWOMRE9ZF7(R`mjuG_{RIKY4`IpE$&53N3KDlobUWG)^Zi}wu5yTEJhn+tMMQVXl zw&mM#KSztCj`l$$5p+kL=vBxk*iZCFK=AsI#>w?xT5zj>VM}NOf#6q>rFMQ4{Bmo0 z(}dr)B{?bs@t8n`P-3P8wPv<1#+srEARpzIZRq?_(wr6jlMY1Uq?4=DR#F1Notx-N zZbu&3&3B;4ZnD<*%Z@w}NgZF85J?2H?=NyU719Sgg2^8sSofrJH;;DYR?lKfSh@@Z zO}gyBsdXavun$2BG7c7o8G)PFEH=pc|zHiB+E6c>yb3Nf!_-uj@e$b*$@y#8Co* zRX;t6C|${PAZQYMQO*&41teBo3X>9dL*m`to3Tn)az7AUL0ObOeJDdAQB}gE#QKc; zF5u&av6`1XpiTr!nq^P-ikyH9c){@^dz;N)u`g0OCDxGuF&7nQr3QUNCjq?R0f{aL zi1|94m|6dpFag{KLgM)ukf6X$kmfmwUv`O30(i0|aupD~atvwy2?|!sLd8Ok?+CvD zX$B>6%j+*Qk^r6;igODPv$!}Pd`~9jcV5Obn9AN@!t0Srb(J_Ukv7Sh}#loLM@Q1Tk;(mWt={oXuv2%wpiDkg%P z!@jKOvmzVyqkd9kMt|!2uOsh2k|5#+H%$Noen@;e4u~0Df^{+VS2_vcd$^Fe%L8JT z7w5y@=p=w~HzYa$AZ8134!uPu0X#Q`#Gnp{nJvz{ztc$oD~KS`egiS*h;zhkItidD zhs5K2Am(;)KDk3D0laq!i4o~tI8TUk-aR@A;3S2_j5rYUsyGYZr;`BY#UXK{3B-IN z&UgNzlK^hdA#n)fjyo{9v}oykc^1d}sp8O`3HU~UE~RtJmRAI1ig(wr(zHHE!SPS8RF zc$z9b6FOF-45XxvG~(@@O^(7>!HY$OM>yVUc8p^=B{>CV{)qG#BT9m@gP2C?gG3)M zGWa+{-H<5oVA%*Hp7o0)f;qtm*Pe46NC~&4dL7zMviMJIYclu6H9OJ0@Qzp zjvs+4NI3=bb$HGxI3?!j1ZgSWIudmU+3@^nIOWY#7?Hl7%wu%$V-}Or`~U!+2!mmM zEWW8Dn4AFK#}WOU$dR9L2d6;d0sctPAA*kG8G?$YY!oGR+_*qDqe^o-3r4eu1Tc*V z9r2KOkj?lsP*8}mQ0oZ$dtuXuUM7^h??T}0D$X7*!rzMQhP}l!BAgihL1GdBh&dZX zd{Ig$2POUkBuAe#j>M4h)E7dUdu7WCI7r~QY~t)B@(5rZu&}?Hz!>7@b)mfR5`n`! z=l+%?ZvilkGThlIQyEAA>qm@|?pHH^?Gn@{esc=l|;zNaJ0emk2eYhXS!8dOUFai&HFl=XtR+f8DA4csSY%vS~ zR$k`kdunGuZH79hulLv`gGRB{iEq7i9 z^^&k(5;lGKl+1GSy3zSioHqdAid;c?eIpaniQ70xG$=sKA}K6TOQF<|_@ftD<`Q3X zZzBwRN}?_(n9>D+CuRUJ81muBlMk@TOlD)izC7{H@pskeD zaVSL(eT81?=;ffJmxIJn6He9bL|zmQ5e$%a&|6dR520iFbSGt}x0pfUlo&V(Qc}FJ z66K4%z|jt*q>iZ)*)Jk!A;00CHj=%ba)~+-Ookv28sL4DO7FOz+ZqG}s}luR`bOY5 z!`~)z8zd&W-a%c@Ph~|ape2Fz0MLaUB98#}9}>9^P!czf3Z?#g1U|+oX`YuPjRBaK z#h3pB1`YidC0X*#%S@9#XR~7~PflwHTwNU_Y1jYHW zNTvS?9gjUerKBDs5?B8xOr!9+;%x_s_65>xUY`4+0OK@(t^m+}o?#XO=sO{U0^>fT zByLU>%Bgb%HnVigB*|$2<_7T%IL|-=`1qR07rr37P(0Yjgq>WB^&SHNw-&NTN*^r5 zyCRid;7g|Ievy(ozGel98~zKF3LhZeNsvK-+E*w?@e1581(YO^1^_+%x6DESUqFV$ zwedA2akG<9ZeJp>3&k@BNRs;i%u(W7cA0?$@bp^b^zX?olE~d)3|0|}zEbd*D+t{y zSrkx*z;Ve{;zwp7fSYJYv~xdD5;w04CGZ-7L?kr-mL&NAm_|kJ$?HEekN}>^i(G%5 z>~*qv#-JsJ#35pQWeK8+sicy0?h$&qUy!jioand(=uDuOuxtOuZ%`69 zX9=aqEdqs+&|D=+5&@W7#rNUw3?zULv5I`>53*Ot<{5)^CL%u(yzy;%l|BMGhVOSN zspAKK#JltkIST&}UOYH^6SGxOl~^TI@hQb7MN-ZGqP+jD&WrG!+hvy&A4}dY200^ST}y0`!B%R!`$7YkeE4!#E@3vDN%Jb zm}PY}?%X^`Y)ND>;CMnKUL{jx+j}6*K}d`>B?4R`Fj=;A)Gr9}1aMx6^B*6Um$sL^>_l2~pCow&fO%Ma^-Km5z;wFENxRL5Fb%2(GTWEDe>?a(rhWtP>V1De0Wdv z&La0*X6N@lBtBV}5A?~P7o zbP={fNg<4DJ&crFz|Ef|%dl|5zlp3>kVHDMNCXm3je(d}RhH#Dp?oay^#P-2CZJ)N=-H)2O@9FJf&ld4$4ar*3@!T+t|Q4V%(Q<%nv*51 z0z7d57E8>NqTCJvOu#^*VJbpN+}tOW%LxQ*97OY^B>545c}{#&5*bJUFaC%eR-Ejp zOde*&U?DhkG~R+xVuG?6@yAVhLi_=Vc~p@zBsL%k>C0AQEhr$BKn4KZIhW#=2*ilQ zCxuFId&JEiLV1u(U2`e zpe^xKNPJRJbR{=PY+MS`e=l27KqCUTWqT>*xg`R4}Y@RWgFcvvP@OvVa9@~{sJ-Sm;$D2Qp zxQXaSsqk#^PKLw^>mJzbYAJyNN)dQf$`RI^SqR`qq9D;q^rR$io)OBOJ_P1-H?ByM zzW|uOiEnvd1`@zTyvP~-$oB2R9bgRZctu|&c>ezMD!ndr9QtfZ>M&k+Wa}@o5Obsu)N$+S!_N^P^W_w?(kU^L1%lar zKjKQK#A6j8=uZQPE1eR{jDesN4J58~O3Zx&K^J4vOFc#;IuIZ@K7!8)BCd2L7Xm>$ zBRD16^k5<}kb!}57Wqo2#EXGI@N_dDaiuF+2MC`G5~swIc0qhXsAq%{S2`tD-~mB% z;v%keC07B#83&1FDq+->yn^X;T(boY5kyo#42&G$h7k9}`#l zLuA8;qa(OQrBh<=1qdFZMG{v!CFX^I;3IfZ^iszYHApOH0)lr31n(P7TUDlK#+%`KXuF2MJIwQ&qGeZ7~)D-vM3O|u_QPpZaD<`QS`zE$X7Zgra^$<<|dZ7 z(v=(y1P>h{(fSpnuB3%}!@YYy)~C`b@vBBaFmEeoFD2@>2hZe^-8j&Qg+wC;iHUT< zDX}mni}DgA#_^&T%_dSD5{q+0KQ@SZV@NDq6rB?1zvz@0aSWk642k8=qVF9>#khp!C1iEiM zj(e$M{%SHMbqt4~n?>Nn1KF2&krTi&5}{CHWf|k^f)W&1aVpg2!tN++`VcvdysC5# z5a&?;frcs|eIQeA;62J>7B8RLb`-4$D5{$tr80c7k z(;v^gDkBJUb~shc+7f>sD1M|709q_LgNR^Jkz{#c3s)}uF^EW?0SFo;K?JY_0RZls zMeY=O$yF%VYT1MWS`&CxVydoX5edYIM6@EW^NJ)X1i<`Fd>?FJ zAOXCaC-TjWWDgk2J!A~J0@05NUS|_Ri#6chDj=IcB>?EXwlWI=EYgHT8@rj3xYtXB~E(GId#k7pqmgWGG-FA3gxCqj2i76r^8a8z;) zc#T;IV0||vZqIj75;w02rSTgCHXxz-mn3Ncz|=CiCwJatAOXzLh`hR+>{Sz32*%*+ zZ5;DrcCd~k6uC2@0*PRv%e$z zkZhhYxRn(-S#bR_y-F_y9VfsQO6nMii?`wTL06GFld`smjC2@1TP@Y~RaDuz>rX=zG8NNf}+j^aW1TbzFx%?Nht4-w&Fa|et zqVEvA#0`Z0C0P{EihzF;7NyEfW+8wvIV2jcUnz;3C4}Pnoj?W>npFYd=YIj1^~Cr7 z9}Fab;f%;PZj)Ve8rWIFZaRZKNL2uVk_0;r9dDq6XwC%%kD#Gqt@9m9>iz!NT?&lD zVf>2J-f+sBhg|6YWt-cE@PS;h_W^t&0DTxW{md?Y&UJS2N|wLNgeOEQ-S-}mFp&*@ zA-@$Q-X;fvNnXKc{7DXVe9A`jHuveJj<1R_bTu!0vGWc zlvqzyk@Bp_do}r++6t${Dh5H0iuWxSlXWR<;5%2^;``@$n;Wg8-I= zKw_oELrUW2C86Z|kHF`sc=MJd2?t>QExx^v8At#ZlE^Jj$Zk3l?EKBR$xgF*d@=!` z9YYdqP4u^gSotY()e)isx)Ep%03QrBf*}dusWK!UA8Emm#LZDcN%SS~B{pHsk|d=7 zm`la?sb4Sx3E&zN`JO-7jpwisjKNS+^s|Du2teo=$)bR<1g=P~u0hN~0IR1UaVr)` zN!&D=v$QoW0xgixi~xWS!UHf1iSJrI1`^;mvM4WvkbO!v<2_B5w%ZHx!{AB#Q!; z66hzn#)LDTK#WK{Cvq_tadW;<+D8&N&7!QABrgCkcZ%#|>KmwR%6xpRL*=06x^Nc~OFZvL{??IX;C5r-L zQvh6)Tt<0jA%OdKNc6m^l*G+{gmSe4f$~Ub2DRq?TnAuALB-ufB?c0}<6e=YDwADu zBMZS8ObS2`3LG!^XCgOYiA;;$dU307nFuPSLS`$aOvuf$f|R%3FD*(1>rbUrvd$wG zUL`QiLE^ni4Jb=|yLBI?ZO`8e2t_(!|PCN>i(^Qkq$Fl+wc5tdy44aVY;8pEvN)`sjs?(CT7-A!ZR}X7xT$ zL(9?=LJsSimQ{0ZDH!FX9qET5H`Kva8QB91w`mc8jX?^v+F%_}nbM=-!wP6j7 zyxs;?)lW}U6;DMfo{CgF6{&bCtau?y1^2_T9ix!q)>0@J^nwk2wf=fVov&ncZ;N9> zg(1bQMlhiyS+A>*BJa`xu&%0*64t*GGG%fqLaIo}N4Y~v zSi4n7Nh_*~6t$#P0m=ov{n2^|siw-)&M8<<=}KBZskl;BD-~DD8VKdF(PPjs#MKaH zd2bx|qm)$uCX{4rwhBqMUXhSteFq{WLqdAz4oS9}sgTmvQ590!x-22XyD_A;gsjRP zQrdb+g_N=WQ6Xilh^pM7QIVAq(m+B^=ME`j?NcFTtPFFfqr5k5!|vm8|zwU?uCa1P&kH41ohBaEv!lAAz?+6umPr z#8tK)tGLQmp>&pU@SstMnmp=_gzWc*h{Y|W zTPsyuO{wPtB_LaSyjAB2d^?lDfO*YUge-y`9>)Xtbe>pOok+EU{z5{ zrq#i#jPWX~l+w^T=vBV)D*q^@krh)*@-(ulc$E%HX=aV_D$Af;(2GnDLEX2J+D`7A z>!fIAMZ!e=H@DtYAK)8AuX(*RY(g1WbTlb*6S+7wpywX+Zv*_l3X1jCnV&tH$*HZ zwyhsjTq|p_ifd)P4h2ow(B6oc4dg|xBq*I^wjtyK~sQzT?X?vOUtL>1E3st6`F*VbwwA-6+ygp87qk8_8# zwf3lxc2<8C($1O-<+eV_CmA92b%`;$I6EU2Gq4@{_ zGK?(rVy>gz+46&lvb|vau3}!W0_$;G_oIUlb6vL8GIz)eRuvV}#fk?Lv0bbhP;Tp0 z62cLZuItI@FsA2@>0%8~F)vyjRLqOkaEUn<$E&2aiaD4&=0$6xis@=CR54wxT~IFQ zak~N#GggXrD|bj&>q`~V%{r+-L+NL&2saesZfgcR-K~FBNOvo` zKDTmHDs6&<*ttWxTUAv^4=WW+#P+b7K)J0~l?te;>--YSjNCCjtZWt2)9RyQdRkK? z=D1WqhKf0qJEo_#NyYTC)~T3Y)&YrG-IJA;s7|0ia>w+tE~uE^))y+KxAmvQ#Fwjy z7#|f=yj$Ljq_<@P!@l&fA{xj}^|8uBxvd{4mWe<=71%W=kb?rnudM>JtX3*8%Nizu zCm(W)E)}>cC(zr^XIbM^U|(yg3hZkgkifL-90QkEfnPlrm@{(fYwc6f{jAF>x}Wt> zqGwd#pj%VX!QFEm>weZf71Q5}$drBTZ^TOYrXWBZaSre^M#{#GJP)XM;?p^6z` z^_7^upEgEJQ57@V8zYBa@jI%(Y-_d(%(iw);EFd&BQR10?#T)Cj_tCor7Ccs^|1;Z zXkCMHLGOH=?Z!YmxUGTKX%#Zada6POS@|1s$2#CA0s8A@)V4f5a#aSXh#6Y? zoAR1#fSw&V7tZ(n0(_g})E%G?4zgi4@J}zUX$_98YOLuEPM33QhGuI)72qEpP!@j8 z9$T>itkEV5?`zjS5Cvqn@qraV_c{ ztJyWqRoAlhy?!TPX!g>E^f#Vug=i(&oxY-Vz_lPN zR!gz1sr6tT6lR+J>(BXM9SVC5z6^Wv%`&jw4Z8v>(>5YYzB*Yg{6%z_!@S@tVs3d8y|{0nS-k^OS5rm#+j2P1B!J?Lz4SZ5!BBaw&b6^3}ukj@aG0SA=yXtR8ZH zWdHeY2CN^YCeGQ@rqlOx*bw->wrBUH?|S$ISU=cpt3|>3HD*4xe#0JcqzSCwVphVs zV_U@=z`7ap0M^5ojG%8lFmJ{1h-vyu#!nr&i@%57#LwXd!v1JQDcFC6&qiVwY%7Y% zZig?2?Sb9><4D+d)PaP(z6%R-H)Iq1Nw9C+jDvkI^c&bIu-$3RVE>u#b=YaJTQs75 zKV%9IDD0$N;jsUb3dn>V6V1c;K+4ks_P$7N|8KF|!=6030PKficZL0aKpO0S#Lj{p z{u57;M`90#o#9~x|0^fgXxNoLD+2pTI4_aO`avWA{YJ1gT?n%P^mxY5Vs8RF8eS{l#!G+qFnp&={y%fe~jp77H=!=#>E zTD$>{XSmNA(Bh|s6wy5MqQ8TV<%EVctHqDql%aVRMGq6k7-8`G zoc)Z^+*L#KERI?(j5Wew!FPKZI{8|B_RlE&(vbIrc1&p8)OXL(oMJ8YudI2NhujeE zA35AWoE`tO3q_LhX`YuuBjLkxXvlps7ApH3SBoG1XQt*^8Cp$fwQ^{@-c-t0Fe|0S zuYQuIc~*sX7g|;hjpwDJNnXW$bw~-#v({C)RD>2^GjzXa}NAWa@-@Cew=GhWe7BN`g1G!JXk7L1h#EC38qp{}M8r4!5?S#Q& z$JRt`g_fG4j33{wn&x>$WnTM}rg^q0z3P&9&9gm<=UaSsCvbMCl2krXM)T}c`pVYj zHP5c7gR-5&Ion}#p!1S>wlOXKy8@*&&#O^C3-@LYSL^5H%K3Cpuk@`;^PGv}Wp_^x zJ5gBkoOO(9UslyT=ft??6XFj~iV@B0@0!oAC2(d{Ro2jo&6{W~vh^g_ZR9>~bRW?h zmvo6>Il9MuYoX2sY}{T_BVU-sv!#=ozJxK*AO(u>X_j<10 zYjahg6@BU;ey+a#R!vxyL&MN0VJQQoI{8zfsVWtXCl367NzAF2-H1tZ)svXUkW&mc zYR~vcT(~vc|Ef{_gftjZov{mZh@$;A(WT;${VwYF1HU+Q{4cteSsra$Hah?MZ^R-` zPZwJ;%|3b82Uah!0_`^bY-4+i<+6`qNSUqo5zAvgSiz1uORPBikJoC$>MPxElD)G$ z`=WlK=a4hS{=8futo~x9*Pg6BKCVt*@O=iUkiKHX@6LQWDD)> z&=HIqELK-Ls3SYoA!23O6@99#l33^LcA-4ACX4m8ePL88te3?4!G7fj1J+crZrIcQ z;V@{LSa)pyaQ58O#lpbgtuy&y&JdFW1H9w|dzS1J2L^vu;22|$_&G3$4`4?;S8R`t zGie1*g=~GE&vKkOCk$L|3tU6<)z?}K#Mu!&faHbw*xhLM_%RJ(Et1ecyV_|U*Tv!N zXk2#pO|0i7VtMR`HtTt*)N`EuZ&ME3mPN6{NwUWjPKUKzd@1(fs~lFXaIxn|v-cGU zgSAq&kYP`Itif6(IWtkx;V4$NzS?IWPTUiQCtF|Z8nP)wTgj3ZC|eS~bxukDjOOTL zy;D-(_gO_7BG{9;kY^Zzvh|HVT@WaFHoG2e_0xV}o+_)!-6GY8tO=`^B;n;^=&@eB#$dw-{Jb#)mY1q5u5$id>+A_kv^zZ%|1Mxd$P+@8CD=7AA^>y zzv}Y`BJtU7k-QwSKtOcv68cIc8k#?*wiY)~FCWN3qJ8#KC2gQyA)KRg&F*ODgHU&>ZV}eJnl$%3!~s9>I#u z@HuE~O65ih>C%I=0PtBco{fd{81|~|0eo&(D4M&zO?HOJu4thw~_v91^SjV!-v^ZQi2H z(!&Y?Iqc^dtXB&CZ+CvJ%+OfvtdDE3Uey{5kG*YPC2g=?ErNpt&9*)Ygq1E98Ly$k zJCd#;s*%Gx;5(uR%(RHk=L%>Ce0vWZ39EWQCgNWAd1ZAJtl9xA{p&shPM3q#Ab@3j z-Dkli+{fvS16UGV6HlohI ztHQ99aq)Z=82OYjeM@Z^AMwBud;(N2+$PXv6d(p*#^hc0{r- ztfE2RrABCRsLBEe8m5=Ntjz4rYEaSFD_LEMrmb)Rj;lM*9SSC>a)PKepf44S5Pw_4CE|8%ETSjYfBlR};#agFSBlWgUsfJ zR32Urb=ua`#^_OsmK9Y=8><&r z#CAQCs(FPdVUvO9{U9?N}Hq?SGW0Smza32z6zm}^dz5#;AHj3={rd; zrAi!sqY4b?XPjRe{k|8PSJAU^G+4(yTHtS}3FH#zYt`9ooCeFl=S~_Hu&ulD7m!?}ou} zDbLIPNif2cQ6akw4Ci>Rzv&CZIiCAoh|s2Z58;H!7#P7SX4n_Cw5fVMb@KI#tq#K} z=-iD445y%zqH4luu41yMmPhXzBF0=j;kz=h!sWhjuFf8Npk6hWJ?;r3ZJ^%GwJNiK z)_0V177Wz8`^j8`J!DNhm_5Sja@p164Ol(HdCWZa^ann$dWjWh-+qf*=q*-~orSrM zfqEaA>PW#+{u}-OTzw5TGEmR*@k6$(-RDBIY2Mx4oL&Y-2`TN+t3495q-lDx>w9m3 zrs<`9RDnEKBS0(@dQA$n5_h=M^s*tmvGLe--u8!;l8?oU(;oQnKYpSS%(*&qPSaC; zUWPUFOIJh9HBGPXiXV+Vp0lAATs7ogOSAvF&6G7|un?#XT~{Pc%bKP)cI_th;hP>f zn?(MGESlZ<{erNXhH|KrWIvjp0;`!=DfWsoWnnd!lBL;mrqzMf!o_OLuzR*G469`< zQ)b!+%Qt~#yEu)|!e0JC30SRyIn~nM{-lZ@tkxFi4!UAr9?HJ3Ow;r>J}iA!!$H2< zbno@?@LVvAF6s=JSg#n27nSk$)iT-)r?J3@a)zGhYW~DmOBmxE@)>$@=a6sRp8#f( zj5s)kYzf*7G}RBewS6~>!Ei3kBQ;aBnR;}7?tfNf^=2?U%1FMR0VBWaS$d@S!6>MV z6$4VWSiXUC1V*lguS?0?(a-^MkCRY