Rename add_start_docstrings_to_callable (#8120)

examples/bert-loses-patience/pabee/modeling_pabee_albert.py

@@ -20,7 +20,7 @@ import torch
 import torch.nn as nn
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
 from transformers.modeling_albert import (
     ALBERT_INPUTS_DOCSTRING,
     ALBERT_START_DOCSTRING,
@@ -87,7 +87,7 @@ class AlbertModelWithPabee(AlbertModel):
         message = f"*** Patience = {self.patience} Avg. Inference Layers = {avg_inf_layers:.2f} Speed Up = {1 - avg_inf_layers / self.config.num_hidden_layers:.2f} ***"
         print(message)
 
-    @add_start_docstrings_to_callable(ALBERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(ALBERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -230,7 +230,7 @@ class AlbertForSequenceClassificationWithPabee(AlbertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(ALBERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(ALBERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,

examples/bert-loses-patience/pabee/modeling_pabee_bert.py

@@ -22,7 +22,7 @@ import torch
 from torch import nn
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
 from transformers.modeling_bert import (
     BERT_INPUTS_DOCSTRING,
     BERT_START_DOCSTRING,
@@ -92,7 +92,7 @@ class BertModelWithPabee(BertModel):
         message = f"*** Patience = {self.patience} Avg. Inference Layers = {avg_inf_layers:.2f} Speed Up = {1 - avg_inf_layers / self.config.num_hidden_layers:.2f} ***"
         print(message)
 
-    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -254,7 +254,7 @@ class BertForSequenceClassificationWithPabee(BertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,

examples/deebert/src/modeling_highway_bert.py

@@ -2,7 +2,7 @@ import torch
 from torch import nn
 from torch.nn import CrossEntropyLoss, MSELoss
 
-from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
 from transformers.modeling_bert import (
     BERT_INPUTS_DOCSTRING,
     BERT_START_DOCSTRING,
@@ -134,7 +134,7 @@ class DeeBertModel(BertPreTrainedModel):
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -288,7 +288,7 @@ class DeeBertForSequenceClassification(BertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,

examples/deebert/src/modeling_highway_roberta.py

@@ -4,7 +4,7 @@ import torch.nn as nn
 from torch.nn import CrossEntropyLoss, MSELoss
 
 from transformers.configuration_roberta import RobertaConfig
-from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
 from transformers.modeling_roberta import ROBERTA_INPUTS_DOCSTRING, ROBERTA_START_DOCSTRING, RobertaEmbeddings
 
 from .modeling_highway_bert import BertPreTrainedModel, DeeBertModel, HighwayException, entropy
@@ -45,7 +45,7 @@ class DeeRobertaForSequenceClassification(BertPreTrainedModel):
         self.dropout = nn.Dropout(config.hidden_dropout_prob)
         self.classifier = nn.Linear(config.hidden_size, self.config.num_labels)
 
-    @add_start_docstrings_to_callable(ROBERTA_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,

examples/movement-pruning/emmental/modeling_bert_masked.py

@@ -28,7 +28,7 @@ from torch.nn import CrossEntropyLoss, MSELoss
 
 from emmental import MaskedBertConfig
 from emmental.modules import MaskedLinear
-from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_callable
+from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
 from transformers.modeling_bert import ACT2FN, BertLayerNorm, load_tf_weights_in_bert
 from transformers.modeling_utils import PreTrainedModel, prune_linear_layer
 
@@ -498,7 +498,7 @@ class MaskedBertModel(MaskedBertPreTrainedModel):
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    @add_start_docstrings_to_callable(MASKED_BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(MASKED_BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -671,7 +671,7 @@ class MaskedBertForSequenceClassification(MaskedBertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(MASKED_BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(MASKED_BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -756,7 +756,7 @@ class MaskedBertForMultipleChoice(MaskedBertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(MASKED_BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(MASKED_BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -846,7 +846,7 @@ class MaskedBertForTokenClassification(MaskedBertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(MASKED_BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(MASKED_BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,
@@ -932,7 +932,7 @@ class MaskedBertForQuestionAnswering(MaskedBertPreTrainedModel):
 
         self.init_weights()
 
-    @add_start_docstrings_to_callable(MASKED_BERT_INPUTS_DOCSTRING)
+    @add_start_docstrings_to_model_forward(MASKED_BERT_INPUTS_DOCSTRING)
     def forward(
         self,
         input_ids=None,