Cleanup: continue the init refactor (#35170)

* Round 2 * Round 3
huggingface · Dec 11, 2024 · 7d303ef · 7d303ef
1 parent 5fcf628
commit 7d303ef
Show file tree

Hide file tree

Showing 97 changed files with 573 additions and 1,407 deletions.
diff --git a/src/transformers/models/big_bird/__init__.py b/src/transformers/models/big_bird/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2021 The HuggingFace Team. All rights reserved.
+# Copyright 2024 The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,133 +13,19 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import (
-    OptionalDependencyNotAvailable,
-    _LazyModule,
-    is_flax_available,
-    is_sentencepiece_available,
-    is_tf_available,
-    is_tokenizers_available,
-    is_torch_available,
-)
+from ...utils import _LazyModule
+from ...utils.import_utils import define_import_structure
 
 
-_import_structure = {
-    "configuration_big_bird": ["BigBirdConfig", "BigBirdOnnxConfig"],
-}
-
-try:
-    if not is_sentencepiece_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["tokenization_big_bird"] = ["BigBirdTokenizer"]
-
-try:
-    if not is_tokenizers_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["tokenization_big_bird_fast"] = ["BigBirdTokenizerFast"]
-
-try:
-    if not is_torch_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["modeling_big_bird"] = [
-        "BigBirdForCausalLM",
-        "BigBirdForMaskedLM",
-        "BigBirdForMultipleChoice",
-        "BigBirdForPreTraining",
-        "BigBirdForQuestionAnswering",
-        "BigBirdForSequenceClassification",
-        "BigBirdForTokenClassification",
-        "BigBirdLayer",
-        "BigBirdModel",
-        "BigBirdPreTrainedModel",
-        "load_tf_weights_in_big_bird",
-    ]
-
-try:
-    if not is_flax_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["modeling_flax_big_bird"] = [
-        "FlaxBigBirdForCausalLM",
-        "FlaxBigBirdForMaskedLM",
-        "FlaxBigBirdForMultipleChoice",
-        "FlaxBigBirdForPreTraining",
-        "FlaxBigBirdForQuestionAnswering",
-        "FlaxBigBirdForSequenceClassification",
-        "FlaxBigBirdForTokenClassification",
-        "FlaxBigBirdModel",
-        "FlaxBigBirdPreTrainedModel",
-    ]
-
 if TYPE_CHECKING:
-    from .configuration_big_bird import BigBirdConfig, BigBirdOnnxConfig
-
-    try:
-        if not is_sentencepiece_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .tokenization_big_bird import BigBirdTokenizer
-
-    try:
-        if not is_tokenizers_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .tokenization_big_bird_fast import BigBirdTokenizerFast
-
-    try:
-        if not is_torch_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .modeling_big_bird import (
-            BigBirdForCausalLM,
-            BigBirdForMaskedLM,
-            BigBirdForMultipleChoice,
-            BigBirdForPreTraining,
-            BigBirdForQuestionAnswering,
-            BigBirdForSequenceClassification,
-            BigBirdForTokenClassification,
-            BigBirdLayer,
-            BigBirdModel,
-            BigBirdPreTrainedModel,
-            load_tf_weights_in_big_bird,
-        )
-
-    try:
-        if not is_flax_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .modeling_flax_big_bird import (
-            FlaxBigBirdForCausalLM,
-            FlaxBigBirdForMaskedLM,
-            FlaxBigBirdForMultipleChoice,
-            FlaxBigBirdForPreTraining,
-            FlaxBigBirdForQuestionAnswering,
-            FlaxBigBirdForSequenceClassification,
-            FlaxBigBirdForTokenClassification,
-            FlaxBigBirdModel,
-            FlaxBigBirdPreTrainedModel,
-        )
-
+    from .configuration_big_bird import *
+    from .convert_bigbird_original_tf_checkpoint_to_pytorch import *
+    from .modeling_big_bird import *
+    from .modeling_flax_big_bird import *
+    from .tokenization_big_bird import *
+    from .tokenization_big_bird_fast import *
 else:
     import sys
 
-    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
+    _file = globals()["__file__"]
+    sys.modules[__name__] = _LazyModule(__name__, _file, define_import_structure(_file), module_spec=__spec__)
diff --git a/src/transformers/models/big_bird/configuration_big_bird.py b/src/transformers/models/big_bird/configuration_big_bird.py
@@ -171,3 +171,6 @@ def inputs(self) -> Mapping[str, Mapping[int, str]]:
                 ("attention_mask", dynamic_axis),
             ]
         )
+
+
+__all__ = ["BigBirdConfig", "BigBirdOnnxConfig"]
diff --git a/src/transformers/models/big_bird/modeling_big_bird.py b/src/transformers/models/big_bird/modeling_big_bird.py
@@ -3126,3 +3126,18 @@ def prepare_question_mask(q_lengths: torch.Tensor, maxlen: int):
         mask.unsqueeze_(0)  # -> (1, maxlen)
         mask = torch.where(mask < q_lengths, 1, 0)
         return mask
+
+
+__all__ = [
+    "BigBirdForCausalLM",
+    "BigBirdForMaskedLM",
+    "BigBirdForMultipleChoice",
+    "BigBirdForPreTraining",
+    "BigBirdForQuestionAnswering",
+    "BigBirdForSequenceClassification",
+    "BigBirdForTokenClassification",
+    "BigBirdLayer",
+    "BigBirdModel",
+    "BigBirdPreTrainedModel",
+    "load_tf_weights_in_big_bird",
+]
diff --git a/src/transformers/models/big_bird/modeling_flax_big_bird.py b/src/transformers/models/big_bird/modeling_flax_big_bird.py
@@ -2633,3 +2633,16 @@ def update_inputs_for_generation(self, model_outputs, model_kwargs):
     FlaxCausalLMOutputWithCrossAttentions,
     _CONFIG_FOR_DOC,
 )
+
+
+__all__ = [
+    "FlaxBigBirdForCausalLM",
+    "FlaxBigBirdForMaskedLM",
+    "FlaxBigBirdForMultipleChoice",
+    "FlaxBigBirdForPreTraining",
+    "FlaxBigBirdForQuestionAnswering",
+    "FlaxBigBirdForSequenceClassification",
+    "FlaxBigBirdForTokenClassification",
+    "FlaxBigBirdModel",
+    "FlaxBigBirdPreTrainedModel",
+]
diff --git a/src/transformers/models/big_bird/tokenization_big_bird.py b/src/transformers/models/big_bird/tokenization_big_bird.py
@@ -319,3 +319,6 @@ def create_token_type_ids_from_sequences(
         if token_ids_1 is None:
             return len(cls + token_ids_0 + sep) * [0]
         return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+
+__all__ = ["BigBirdTokenizer"]
diff --git a/src/transformers/models/big_bird/tokenization_big_bird_fast.py b/src/transformers/models/big_bird/tokenization_big_bird_fast.py
@@ -227,3 +227,6 @@ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] =
             copyfile(self.vocab_file, out_vocab_file)
 
         return (out_vocab_file,)
+
+
+__all__ = ["BigBirdTokenizerFast"]
diff --git a/src/transformers/models/bigbird_pegasus/__init__.py b/src/transformers/models/bigbird_pegasus/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2021 The HuggingFace Team. All rights reserved.
+# Copyright 2024 The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,55 +13,16 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
-
-
-_import_structure = {
-    "configuration_bigbird_pegasus": [
-        "BigBirdPegasusConfig",
-        "BigBirdPegasusOnnxConfig",
-    ],
-}
-
-try:
-    if not is_torch_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["modeling_bigbird_pegasus"] = [
-        "BigBirdPegasusForCausalLM",
-        "BigBirdPegasusForConditionalGeneration",
-        "BigBirdPegasusForQuestionAnswering",
-        "BigBirdPegasusForSequenceClassification",
-        "BigBirdPegasusModel",
-        "BigBirdPegasusPreTrainedModel",
-    ]
+from ...utils import _LazyModule
+from ...utils.import_utils import define_import_structure
 
 
 if TYPE_CHECKING:
-    from .configuration_bigbird_pegasus import (
-        BigBirdPegasusConfig,
-        BigBirdPegasusOnnxConfig,
-    )
-
-    try:
-        if not is_torch_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .modeling_bigbird_pegasus import (
-            BigBirdPegasusForCausalLM,
-            BigBirdPegasusForConditionalGeneration,
-            BigBirdPegasusForQuestionAnswering,
-            BigBirdPegasusForSequenceClassification,
-            BigBirdPegasusModel,
-            BigBirdPegasusPreTrainedModel,
-        )
-
-
+    from .configuration_bigbird_pegasus import *
+    from .convert_bigbird_pegasus_tf_to_pytorch import *
+    from .modeling_bigbird_pegasus import *
 else:
     import sys
 
-    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
+    _file = globals()["__file__"]
+    sys.modules[__name__] = _LazyModule(__name__, _file, define_import_structure(_file), module_spec=__spec__)
diff --git a/src/transformers/models/bigbird_pegasus/configuration_bigbird_pegasus.py b/src/transformers/models/bigbird_pegasus/configuration_bigbird_pegasus.py
@@ -407,3 +407,6 @@ def _flatten_past_key_values_(self, flattened_output, name, idx, t):
             flattened_output = super(OnnxSeq2SeqConfigWithPast, self)._flatten_past_key_values_(
                 flattened_output, name, idx, t
             )
+
+
+__all__ = ["BigBirdPegasusConfig", "BigBirdPegasusOnnxConfig"]
diff --git a/src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py b/src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py
@@ -3028,3 +3028,13 @@ def _reorder_cache(past_key_values, beam_idx):
                 tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
             )
         return reordered_past
+
+
+__all__ = [
+    "BigBirdPegasusForCausalLM",
+    "BigBirdPegasusForConditionalGeneration",
+    "BigBirdPegasusForQuestionAnswering",
+    "BigBirdPegasusForSequenceClassification",
+    "BigBirdPegasusModel",
+    "BigBirdPegasusPreTrainedModel",
+]
diff --git a/src/transformers/models/biogpt/__init__.py b/src/transformers/models/biogpt/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 The HuggingFace Team. All rights reserved.
+# Copyright 2024 The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -13,49 +13,17 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available
-
-
-_import_structure = {
-    "configuration_biogpt": ["BioGptConfig"],
-    "tokenization_biogpt": ["BioGptTokenizer"],
-}
-
-try:
-    if not is_torch_available():
-        raise OptionalDependencyNotAvailable()
-except OptionalDependencyNotAvailable:
-    pass
-else:
-    _import_structure["modeling_biogpt"] = [
-        "BioGptForCausalLM",
-        "BioGptForTokenClassification",
-        "BioGptForSequenceClassification",
-        "BioGptModel",
-        "BioGptPreTrainedModel",
-    ]
+from ...utils import _LazyModule
+from ...utils.import_utils import define_import_structure
 
 
 if TYPE_CHECKING:
-    from .configuration_biogpt import BioGptConfig
-    from .tokenization_biogpt import BioGptTokenizer
-
-    try:
-        if not is_torch_available():
-            raise OptionalDependencyNotAvailable()
-    except OptionalDependencyNotAvailable:
-        pass
-    else:
-        from .modeling_biogpt import (
-            BioGptForCausalLM,
-            BioGptForSequenceClassification,
-            BioGptForTokenClassification,
-            BioGptModel,
-            BioGptPreTrainedModel,
-        )
-
-
+    from .configuration_biogpt import *
+    from .convert_biogpt_original_pytorch_checkpoint_to_pytorch import *
+    from .modeling_biogpt import *
+    from .tokenization_biogpt import *
 else:
     import sys
 
-    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
+    _file = globals()["__file__"]
+    sys.modules[__name__] = _LazyModule(__name__, _file, define_import_structure(_file), module_spec=__spec__)
diff --git a/src/transformers/models/biogpt/configuration_biogpt.py b/src/transformers/models/biogpt/configuration_biogpt.py
@@ -129,3 +129,6 @@ def __init__(
         self.layerdrop = layerdrop
         self.activation_dropout = activation_dropout
         super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
+
+
+__all__ = ["BioGptConfig"]
diff --git a/src/transformers/models/biogpt/modeling_biogpt.py b/src/transformers/models/biogpt/modeling_biogpt.py
@@ -1028,3 +1028,12 @@ def get_input_embeddings(self):
 
     def set_input_embeddings(self, value):
         self.biogpt.embed_tokens = value
+
+
+__all__ = [
+    "BioGptForCausalLM",
+    "BioGptForTokenClassification",
+    "BioGptForSequenceClassification",
+    "BioGptModel",
+    "BioGptPreTrainedModel",
+]
diff --git a/src/transformers/models/biogpt/tokenization_biogpt.py b/src/transformers/models/biogpt/tokenization_biogpt.py
@@ -356,3 +356,6 @@ def __setstate__(self, d):
             )
 
         self.sm = sacremoses
+
+
+__all__ = ["BioGptTokenizer"]
-Original file line number
+Diff line change
@@ Expand Up / @@ -171,3 +171,6 @@ def inputs(self) -> Mapping[str, Mapping[int, str]]: @@
                     ("attention_mask", dynamic_axis),
                 ]
             )
+    __all__ = ["BigBirdConfig", "BigBirdOnnxConfig"]
Original file line number	Diff line number	Diff line change
Expand Up		@@ -227,3 +227,6 @@ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] =
		copyfile(self.vocab_file, out_vocab_file)

		return (out_vocab_file,)


		__all__ = ["BigBirdTokenizerFast"]
Original file line number	Diff line number	Diff line change
Expand Up		@@ -356,3 +356,6 @@ def __setstate__(self, d):
		)

		self.sm = sacremoses


		__all__ = ["BioGptTokenizer"]