From 3000b3551c3bf8081925233696caae0b3b2cd64d Mon Sep 17 00:00:00 2001
From: chmok <52270280+chmok@users.noreply.github.com>
Date: Thu, 23 Nov 2023 16:44:52 +0900
Subject: [PATCH] Delete models/rnn.py

delete rnn model
---
 models/rnn.py | 46 ----------------------------------------------
 1 file changed, 46 deletions(-)
 delete mode 100644 models/rnn.py

diff --git a/models/rnn.py b/models/rnn.py
deleted file mode 100644
index 221960e..0000000
--- a/models/rnn.py
+++ /dev/null
@@ -1,46 +0,0 @@
-import torch.nn as nn
-import torch
-import torch.optim as optim
-from models.train_model_multi import Train_Test
-
-import numpy as np
-import time
-import copy
-
-class RNN_model(nn.Module):
-    def __init__(self, input_size, hidden_size, num_layers, num_classes, bidirectional, rnn_type, device='cuda'):
-        super(RNN_model, self).__init__()
-        self.hidden_size = hidden_size
-        self.num_layers = num_layers
-        self.num_classes = num_classes
-        self.rnn_type = rnn_type
-        self.num_directions = 2 if bidirectional == True else 1
-        self.device = device
-        
-        # rnn_type에 따른 recurrent layer 설정
-        if self.rnn_type == 'lstm':
-            self.rnn = nn.LSTM(input_size, hidden_size, num_layers, batch_first=True, bidirectional=bidirectional)
-        elif self.rnn_type == 'gru':
-            self.rnn = nn.GRU(input_size, hidden_size, num_layers, batch_first=True, bidirectional=bidirectional)
-        
-        # bidirectional에 따른 fc layer 구축
-        # bidirectional 여부에 따라 hidden state의 shape가 달라짐 (True: 2 * hidden_size, False: hidden_size)
-        self.fc = nn.Linear(self.num_directions * hidden_size, self.num_classes)
-
-    def forward(self, x):
-        # data dimension: (batch_size x input_size x seq_len) -> (batch_size x seq_len x input_size)로 변환
-        x = torch.transpose(x, 1, 2)
-        
-        # initial hidden states 설정
-        h0 = torch.zeros(self.num_directions * self.num_layers, x.size(0), self.hidden_size).to(self.device)
-        
-        # 선택한 rnn_type의 RNN으로부터 output 도출
-        if self.rnn_type == 'gru':
-            out, _ = self.rnn(x, h0)  # out: tensor of shape (batch_size, seq_length, hidden_size)
-        else:
-            # initial cell states 설정
-            c0 = torch.zeros(self.num_directions * self.num_layers, x.size(0), self.hidden_size).to(self.device)
-            out, _ = self.rnn(x, (h0, c0))  # out: tensor of shape (batch_size, seq_length, hidden_size)
-        
-        out = self.fc(out[:, -1, :])
-        return out