revert changes in models that don't currently support cross attention

huggingface · Oct 27, 2020 · 741dc0d · 741dc0d
1 parent 56622d4
commit 741dc0d
Show file tree

Hide file tree

Showing 7 changed files with 7 additions and 7 deletions.
diff --git a/examples/deebert/src/modeling_highway_bert.py b/examples/deebert/src/modeling_highway_bert.py
@@ -65,7 +65,7 @@ def forward(
             hidden_states = layer_outputs[0]
 
             if self.output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
             current_outputs = (hidden_states,)
             if self.output_hidden_states:

diff --git a/src/transformers/modeling_longformer.py b/src/transformers/modeling_longformer.py
@@ -917,7 +917,7 @@ def custom_forward(*inputs):
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states:

diff --git a/src/transformers/modeling_mobilebert.py b/src/transformers/modeling_mobilebert.py
@@ -577,7 +577,7 @@ def forward(
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states:

diff --git a/src/transformers/modeling_tf_bert.py b/src/transformers/modeling_tf_bert.py
@@ -404,7 +404,7 @@ def call(
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states:

diff --git a/src/transformers/modeling_tf_electra.py b/src/transformers/modeling_tf_electra.py
@@ -257,7 +257,7 @@ def call(
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states:

diff --git a/src/transformers/modeling_tf_mobilebert.py b/src/transformers/modeling_tf_mobilebert.py
@@ -587,7 +587,7 @@ def call(
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states:

diff --git a/src/transformers/modeling_tf_roberta.py b/src/transformers/modeling_tf_roberta.py
@@ -447,7 +447,7 @@ def call(
             hidden_states = layer_outputs[0]
 
             if output_attentions:
-                all_attentions = all_attentions + layer_outputs[1:]
+                all_attentions = all_attentions + (layer_outputs[1],)
 
         # Add last layer
         if output_hidden_states: