Moved functions to pytorch_utils.py (#16625)

* Moved functions to pytorch_utils.py

* isort formatting

* Reverted tf changes

* isort, make fix-copies

* documentation fix

* Fixed Conv1D import

* Reverted research examples file

* backward compatibility for pytorch_utils

* missing import

* isort fix

src/transformers/models/imagegpt/modeling_imagegpt.py

@@ -38,7 +38,8 @@ from ...modeling_outputs import (
     CausalLMOutputWithCrossAttentions,
     SequenceClassifierOutputWithPast,
 )
-from ...modeling_utils import Conv1D, PreTrainedModel, find_pruneable_heads_and_indices, prune_conv1d_layer
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import Conv1D, find_pruneable_heads_and_indices, prune_conv1d_layer
 from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
 from .configuration_imagegpt import ImageGPTConfig
 

src/transformers/models/layoutlm/modeling_layoutlm.py

@@ -31,12 +31,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
 from .configuration_layoutlm import LayoutLMConfig
 

src/transformers/models/layoutlmv2/modeling_layoutlmv2.py

@@ -31,7 +31,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import PreTrainedModel, apply_chunking_to_forward
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward
 from ...utils import (
     add_start_docstrings,
     add_start_docstrings_to_model_forward,

src/transformers/models/longformer/modeling_longformer.py

@@ -24,12 +24,8 @@ from torch import nn
 from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import ACT2FN, gelu
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     ModelOutput,
     add_code_sample_docstrings,

src/transformers/models/luke/modeling_luke.py

@@ -24,7 +24,8 @@ from torch import nn
 
 from ...activations import ACT2FN, gelu
 from ...modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling
-from ...modeling_utils import PreTrainedModel, apply_chunking_to_forward
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward
 from ...utils import (
     ModelOutput,
     add_start_docstrings,

src/transformers/models/maskformer/modeling_maskformer.py

@@ -29,7 +29,8 @@ from transformers.utils import logging
 
 from ...activations import ACT2FN
 from ...modeling_outputs import BaseModelOutputWithCrossAttentions
-from ...modeling_utils import ModuleUtilsMixin, PreTrainedModel, find_pruneable_heads_and_indices, prune_linear_layer
+from ...modeling_utils import ModuleUtilsMixin, PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     ModelOutput,
     add_code_sample_docstrings,

src/transformers/models/megatron_bert/modeling_megatron_bert.py

@@ -39,12 +39,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     ModelOutput,
     add_code_sample_docstrings,

src/transformers/models/mobilebert/modeling_mobilebert.py

@@ -41,7 +41,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import PreTrainedModel, find_pruneable_heads_and_indices, prune_linear_layer
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     ModelOutput,
     add_code_sample_docstrings,

src/transformers/models/mpnet/modeling_mpnet.py

@@ -33,7 +33,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import PreTrainedModel, find_pruneable_heads_and_indices, prune_linear_layer
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
 from .configuration_mpnet import MPNetConfig
 

src/transformers/models/nystromformer/modeling_nystromformer.py

@@ -33,12 +33,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
 from .configuration_nystromformer import NystromformerConfig
 

src/transformers/models/openai/modeling_openai.py

@@ -28,13 +28,8 @@ from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import gelu_new, silu
 from ...modeling_outputs import BaseModelOutput, CausalLMOutput, SequenceClassifierOutput
-from ...modeling_utils import (
-    Conv1D,
-    PreTrainedModel,
-    SequenceSummary,
-    find_pruneable_heads_and_indices,
-    prune_conv1d_layer,
-)
+from ...modeling_utils import PreTrainedModel, SequenceSummary
+from ...pytorch_utils import Conv1D, find_pruneable_heads_and_indices, prune_conv1d_layer
 from ...utils import (
     ModelOutput,
     add_code_sample_docstrings,

src/transformers/models/perceiver/modeling_perceiver.py

@@ -29,12 +29,8 @@ from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import ACT2FN
 from ...modeling_outputs import BaseModelOutputWithCrossAttentions
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     ModelOutput,
     add_start_docstrings,

src/transformers/models/qdqbert/modeling_qdqbert.py

@@ -38,7 +38,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import PreTrainedModel, find_pruneable_heads_and_indices, prune_linear_layer
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     add_code_sample_docstrings,
     add_start_docstrings,

src/transformers/models/realm/modeling_realm.py

@@ -31,12 +31,8 @@ from ...modeling_outputs import (
     MaskedLMOutput,
     ModelOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
 from .configuration_realm import RealmConfig
 

src/transformers/models/reformer/modeling_reformer.py

@@ -30,7 +30,8 @@ from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import ACT2FN
 from ...modeling_outputs import CausalLMOutput, MaskedLMOutput, QuestionAnsweringModelOutput, SequenceClassifierOutput
-from ...modeling_utils import PreTrainedModel, apply_chunking_to_forward
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward
 from ...utils import (
     DUMMY_INPUTS,
     DUMMY_MASK,

src/transformers/models/rembert/modeling_rembert.py

@@ -35,12 +35,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     add_code_sample_docstrings,
     add_start_docstrings,

src/transformers/models/roberta/modeling_roberta.py

@@ -35,12 +35,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     add_code_sample_docstrings,
     add_start_docstrings,

src/transformers/models/roformer/modeling_roformer.py

@@ -35,13 +35,8 @@ from ...modeling_outputs import (
     SequenceClassifierOutput,
     TokenClassifierOutput,
 )
-from ...modeling_utils import (
-    PreTrainedModel,
-    SequenceSummary,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel, SequenceSummary
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     add_code_sample_docstrings,
     add_start_docstrings,

src/transformers/models/segformer/modeling_segformer.py

@@ -25,7 +25,8 @@ from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
 from ...activations import ACT2FN
 from ...modeling_outputs import BaseModelOutput, ImageClassifierOutput, SemanticSegmenterOutput
-from ...modeling_utils import PreTrainedModel, find_pruneable_heads_and_indices, prune_linear_layer
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import (
     add_code_sample_docstrings,
     add_start_docstrings,

src/transformers/models/splinter/modeling_splinter.py

@@ -25,12 +25,8 @@ from torch.nn import CrossEntropyLoss
 
 from ...activations import ACT2FN
 from ...modeling_outputs import BaseModelOutputWithPastAndCrossAttentions, QuestionAnsweringModelOutput
-from ...modeling_utils import (
-    PreTrainedModel,
-    apply_chunking_to_forward,
-    find_pruneable_heads_and_indices,
-    prune_linear_layer,
-)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 from ...utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward, logging
 from .configuration_splinter import SplinterConfig