Use more self-explanatory variables

src/transformers/models/decision_transformer/modeling_decision_transformer.py

@@ -530,21 +530,20 @@ class DecisionTransformerGPT2Model(DecisionTransformerGPT2PreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.size()
-            input_ids = input_ids.view(-1, input_shape[-1])
+            seq_length = input_ids.shape[-1]
+            input_ids = input_ids.view(-1, seq_length)
             batch_size = input_ids.shape[0]
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.size()[:-1]
-            batch_size = inputs_embeds.shape[0]
+            batch_size, seq_length = inputs_embeds.size()[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
         device = input_ids.device if input_ids is not None else inputs_embeds.device
 
         if token_type_ids is not None:
-            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+            token_type_ids = token_type_ids.view(-1, seq_length)
         if position_ids is not None:
-            position_ids = position_ids.view(-1, input_shape[-1])
+            position_ids = position_ids.view(-1, seq_length)
 
         if past_key_values is None:
             past_length = 0
@@ -552,8 +551,8 @@ class DecisionTransformerGPT2Model(DecisionTransformerGPT2PreTrainedModel):
         else:
             past_length = past_key_values[0][0].size(-2)
         if position_ids is None:
-            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
-            position_ids = position_ids.unsqueeze(0).view(-1, input_shape[-1])
+            position_ids = torch.arange(past_length, seq_length + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
 
         # GPT2Attention mask.
         if attention_mask is not None:
@@ -603,7 +602,7 @@ class DecisionTransformerGPT2Model(DecisionTransformerGPT2PreTrainedModel):
 
         hidden_states = self.drop(hidden_states)
 
-        output_shape = (-1,) + input_shape[1:] + (hidden_states.size(-1),)
+        output_shape = (-1, seq_length, hidden_states.size(-1))
 
         if self.gradient_checkpointing and self.training:
             if use_cache:

src/transformers/models/gpt2/modeling_gpt2.py

@@ -776,21 +776,20 @@ class GPT2Model(GPT2PreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.size()
-            input_ids = input_ids.view(-1, input_shape[-1])
+            seq_length = input_ids.shape[-1]
+            input_ids = input_ids.view(-1, seq_length)
             batch_size = input_ids.shape[0]
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.size()[:-1]
-            batch_size = inputs_embeds.shape[0]
+            batch_size, seq_length = inputs_embeds.size()[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
         device = input_ids.device if input_ids is not None else inputs_embeds.device
 
         if token_type_ids is not None:
-            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+            token_type_ids = token_type_ids.view(-1, seq_length)
         if position_ids is not None:
-            position_ids = position_ids.view(-1, input_shape[-1])
+            position_ids = position_ids.view(-1, seq_length)
 
         if past_key_values is None:
             past_length = 0
@@ -798,8 +797,8 @@ class GPT2Model(GPT2PreTrainedModel):
         else:
             past_length = past_key_values[0][0].size(-2)
         if position_ids is None:
-            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
-            position_ids = position_ids.unsqueeze(0).view(-1, input_shape[-1])
+            position_ids = torch.arange(past_length, seq_length + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
 
         # GPT2Attention mask.
         if attention_mask is not None:
@@ -849,7 +848,7 @@ class GPT2Model(GPT2PreTrainedModel):
 
         hidden_states = self.drop(hidden_states)
 
-        output_shape = (-1,) + input_shape[1:] + (hidden_states.size(-1),)
+        output_shape = (-1, seq_length, hidden_states.size(-1))
 
         if self.gradient_checkpointing and self.training:
             if use_cache:

src/transformers/models/gpt_neo/modeling_gpt_neo.py

@@ -525,21 +525,20 @@ class GPTNeoModel(GPTNeoPreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.size()
-            input_ids = input_ids.view(-1, input_shape[-1])
+            seq_length = input_ids.shape[-1]
+            input_ids = input_ids.view(-1, seq_length)
             batch_size = input_ids.shape[0]
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.size()[:-1]
-            batch_size = inputs_embeds.shape[0]
+            batch_size, seq_length = inputs_embeds.size()[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
         device = input_ids.device if input_ids is not None else inputs_embeds.device
 
         if token_type_ids is not None:
-            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+            token_type_ids = token_type_ids.view(-1, seq_length)
         if position_ids is not None:
-            position_ids = position_ids.view(-1, input_shape[-1])
+            position_ids = position_ids.view(-1, seq_length)
 
         if past_key_values is None:
             past_length = 0
@@ -548,8 +547,8 @@ class GPTNeoModel(GPTNeoPreTrainedModel):
             past_length = past_key_values[0][0].size(-2)
 
         if position_ids is None:
-            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
-            position_ids = position_ids.unsqueeze(0).view(-1, input_shape[-1])
+            position_ids = torch.arange(past_length, seq_length + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
 
         # Attention mask.
         if attention_mask is not None:
@@ -588,7 +587,7 @@ class GPTNeoModel(GPTNeoPreTrainedModel):
 
         hidden_states = self.drop(hidden_states)
 
-        output_shape = (-1,) + input_shape[1:] + (hidden_states.size(-1),)
+        output_shape = (-1, seq_length, hidden_states.size(-1))
 
         if self.gradient_checkpointing and self.training:
             if use_cache:

src/transformers/models/gptj/modeling_gptj.py

@@ -577,22 +577,21 @@ class GPTJModel(GPTJPreTrainedModel):
         if input_ids is not None and inputs_embeds is not None:
             raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
         elif input_ids is not None:
-            input_shape = input_ids.size()
-            input_ids = input_ids.view(-1, input_shape[-1])
+            seq_length = input_ids.shape[-1]
+            input_ids = input_ids.view(-1, seq_length)
             batch_size = input_ids.shape[0]
         elif inputs_embeds is not None:
-            input_shape = inputs_embeds.size()[:-1]
-            batch_size = inputs_embeds.shape[0]
+            batch_size, seq_length = inputs_embeds.size()[:-1]
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
         device = input_ids.device if input_ids is not None else inputs_embeds.device
 
         if token_type_ids is not None:
-            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+            token_type_ids = token_type_ids.view(-1, seq_length)
 
         if position_ids is not None:
-            position_ids = position_ids.view(-1, input_shape[-1]).long()
+            position_ids = position_ids.view(-1, seq_length).long()
 
         if past_key_values is None:
             past_length = 0
@@ -601,8 +600,8 @@ class GPTJModel(GPTJPreTrainedModel):
             past_length = past_key_values[0][0].size(-2)
 
         if position_ids is None:
-            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
-            position_ids = position_ids.unsqueeze(0).view(-1, input_shape[-1])
+            position_ids = torch.arange(past_length, seq_length + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
 
         # Attention mask.
         if attention_mask is not None:
@@ -641,7 +640,7 @@ class GPTJModel(GPTJPreTrainedModel):
 
         hidden_states = self.drop(hidden_states)
 
-        output_shape = (-1,) + input_shape[1:] + (hidden_states.size(-1),)
+        output_shape = (-1, seq_length, hidden_states.size(-1))
 
         if self.gradient_checkpointing and self.training:
             if use_cache: