[super-hot-fix] cahnged h_n to c_n by mistake;; to be consistenet wit…

…h prior trainings changing back to h_n
AI4Bharat · Aug 20, 2020 · ab742d5 · ab742d5
1 parent 7c72584
commit ab742d5
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 6 deletions.
diff --git a/algorithms/lm_fused_rnn.py b/algorithms/lm_fused_rnn.py
@@ -226,7 +226,7 @@ def forward(self, x, hidden, enc_output):
             hid_for_att = torch.zeros((self.dec_layers, batch_sz,
                                     self.dec_hidden_dim )).to(self.device)
         elif self.dec_rnn_type == 'lstm':
-            hid_for_att = hidden[1] # c_n
+            hid_for_att = hidden[0] # h_n <<<check
         else:
             hid_for_att = hidden
 
@@ -267,7 +267,9 @@ def get_hidden(self, x, hidden, enc_output):
             hid_for_att = torch.zeros((self.dec_layers, batch_sz,
                                     self.dec_hidden_dim )).to(self.device)
         elif self.dec_rnn_type == 'lstm':
-            hid_for_att = hidden[1] # c_n
+            hid_for_att = hidden[0] # h_n <<<check
+        else:
+            hid_for_att = hidden
 
         # x (batch_size, 1, dec_embed_dim) -> after embedding
         x = self.embedding(x)

diff --git a/algorithms/recurrent_nets.py b/algorithms/recurrent_nets.py
@@ -99,9 +99,6 @@ def get_word_embedding(self, x):
         return out_embed
 
 
-
-
-
 class Decoder(nn.Module):
     def __init__(self, output_dim, embed_dim, hidden_dim,
                        rnn_type = 'gru', layers = 1,
@@ -199,7 +196,7 @@ def forward(self, x, hidden, enc_output):
             hid_for_att = torch.zeros((self.dec_layers, batch_sz,
                                     self.dec_hidden_dim )).to(self.device)
         elif self.dec_rnn_type == 'lstm':
-            hid_for_att = hidden[1] # c_n
+            hid_for_att = hidden[0] # h_n
         else:
             hid_for_att = hidden