Big file_utils cleanup (#16396)

* Big file_utils cleanup

* This one still needs to be treated separately

src/transformers/modeling_flax_utils.py

@@ -681,7 +681,7 @@ class FlaxPreTrainedModel(PushToHubMixin, FlaxGenerationMixin):
                 </Tip>
 
             kwargs:
-                Additional key word arguments passed along to the [`~file_utils.PushToHubMixin.push_to_hub`] method.
+                Additional key word arguments passed along to the [`~utils.PushToHubMixin.push_to_hub`] method.
         """
         if os.path.isfile(save_directory):
             logger.error(f"Provided path ({save_directory}) should be a directory, not a file")

src/transformers/modeling_tf_utils.py

@@ -1401,7 +1401,7 @@ class TFPreTrainedModel(tf.keras.Model, TFModelUtilsMixin, TFGenerationMixin, Pu
                 </Tip>
 
             kwargs:
-                Additional key word arguments passed along to the [`~file_utils.PushToHubMixin.push_to_hub`] method.
+                Additional key word arguments passed along to the [`~utils.PushToHubMixin.push_to_hub`] method.
         """
         if os.path.isfile(save_directory):
             logger.error(f"Provided path ({save_directory}) should be a directory, not a file")

src/transformers/modeling_utils.py

@@ -1036,7 +1036,7 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
                 </Tip>
 
             kwargs:
-                Additional key word arguments passed along to the [`~file_utils.PushToHubMixin.push_to_hub`] method.
+                Additional key word arguments passed along to the [`~utils.PushToHubMixin.push_to_hub`] method.
         """
         if os.path.isfile(save_directory):
             logger.error(f"Provided path ({save_directory}) should be a directory, not a file")
@@ -2129,7 +2129,7 @@ class SQuADHead(nn.Module):
                 Mask for tokens at invalid position, such as query and special symbols (PAD, SEP, CLS). 1.0 means token
                 should be masked.
             return_dict (`bool`, *optional*, defaults to `False`):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
         Returns:
         """

src/transformers/models/albert/modeling_albert.py

@@ -610,7 +610,7 @@ ALBERT_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/albert/modeling_flax_albert.py

@@ -144,7 +144,7 @@ ALBERT_INPUTS_DOCSTRING = r"""
             Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
             config.max_position_embeddings - 1]`.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
 """
 

src/transformers/models/albert/modeling_tf_albert.py

@@ -747,8 +747,8 @@ ALBERT_INPUTS_DOCSTRING = r"""
             more detail. This argument can be used only in eager mode, in graph mode the value in the config will be
             used instead.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple. This argument can be used
-            in eager mode, in graph mode the value will always be set to True.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. This argument can be used in
+            eager mode, in graph mode the value will always be set to True.
         training (`bool`, *optional*, defaults to `False`):
             Whether or not to use the model in training mode (some modules like dropout modules have different
             behaviors between training and evaluation).

src/transformers/models/bart/modeling_bart.py

@@ -679,7 +679,7 @@ BART_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 
@@ -770,7 +770,7 @@ class BartEncoder(BartPretrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -993,7 +993,7 @@ class BartDecoder(BartPretrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -1799,7 +1799,7 @@ class BartForCausalLM(BartPretrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
         Returns:
 

src/transformers/models/bart/modeling_flax_bart.py

@@ -138,7 +138,7 @@ BART_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 
@@ -169,7 +169,7 @@ BART_ENCODE_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 BART_DECODE_INPUTS_DOCSTRING = r"""
@@ -215,7 +215,7 @@ BART_DECODE_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/bart/modeling_tf_bart.py

@@ -625,8 +625,8 @@ BART_INPUTS_DOCSTRING = r"""
             more detail. This argument can be used only in eager mode, in graph mode the value in the config will be
             used instead.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple. This argument can be used
-            in eager mode, in graph mode the value will always be set to True.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. This argument can be used in
+            eager mode, in graph mode the value will always be set to True.
         training (`bool`, *optional*, defaults to `False`):
             Whether or not to use the model in training mode (some modules like dropout modules have different
             behaviors between training and evaluation).
@@ -715,7 +715,7 @@ class TFBartEncoder(tf.keras.layers.Layer):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
 
         if input_ids is not None and inputs_embeds is not None:
@@ -894,7 +894,7 @@ class TFBartDecoder(tf.keras.layers.Layer):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
 
         if input_ids is not None and inputs_embeds is not None:

src/transformers/models/beit/feature_extraction_beit.py

@@ -120,7 +120,7 @@ class BeitFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
             segmentation_maps (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `List[PIL.Image.Image]`, `List[np.ndarray]`, `List[torch.Tensor]`, *optional*):
                 Optionally, the corresponding semantic segmentation maps with the pixel-wise annotations.
 
-            return_tensors (`str` or [`~file_utils.TensorType`], *optional*, defaults to `'np'`):
+            return_tensors (`str` or [`~utils.TensorType`], *optional*, defaults to `'np'`):
                 If set, will return tensors of a particular framework. Acceptable values are:
 
                 - `'tf'`: Return TensorFlow `tf.constant` objects.

src/transformers/models/beit/modeling_beit.py

@@ -618,7 +618,7 @@ BEIT_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/beit/modeling_flax_beit.py

@@ -111,7 +111,7 @@ BEIT_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/bert/modeling_bert.py

@@ -837,7 +837,7 @@ BERT_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/bert/modeling_flax_bert.py

@@ -164,7 +164,7 @@ BERT_INPUTS_DOCSTRING = r"""
             - 0 indicates the head is **masked**.
 
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
 """
 

src/transformers/models/bert/modeling_tf_bert.py

@@ -1025,8 +1025,8 @@ BERT_INPUTS_DOCSTRING = r"""
             more detail. This argument can be used only in eager mode, in graph mode the value in the config will be
             used instead.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple. This argument can be used
-            in eager mode, in graph mode the value will always be set to True.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. This argument can be used in
+            eager mode, in graph mode the value will always be set to True.
         training (`bool`, *optional*, defaults to `False``):
             Whether or not to use the model in training mode (some modules like dropout modules have different
             behaviors between training and evaluation).

src/transformers/models/bert_generation/modeling_bert_generation.py

@@ -242,7 +242,7 @@ BERT_GENERATION_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/big_bird/modeling_big_bird.py

@@ -1844,7 +1844,7 @@ BIG_BIRD_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 

src/transformers/models/big_bird/modeling_flax_big_bird.py

@@ -181,7 +181,7 @@ BIG_BIRD_INPUTS_DOCSTRING = r"""
             - 0 indicates the head is **masked**.
 
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
 """
 

src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py

@@ -1724,7 +1724,7 @@ BIGBIRD_PEGASUS_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 BIGBIRD_PEGASUS_STANDALONE_INPUTS_DOCSTRING = r"""
@@ -1751,7 +1751,7 @@ BIGBIRD_PEGASUS_STANDALONE_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 
@@ -1834,7 +1834,7 @@ class BigBirdPegasusEncoder(BigBirdPegasusPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -2188,7 +2188,7 @@ class BigBirdPegasusDecoder(BigBirdPegasusPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -2999,7 +2999,7 @@ class BigBirdPegasusForCausalLM(BigBirdPegasusPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
         Returns:
 

src/transformers/models/blenderbot/modeling_blenderbot.py

@@ -625,7 +625,7 @@ BLENDERBOT_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
         return_dict (`bool`, *optional*):
-            Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 """
 
 
@@ -710,7 +710,7 @@ class BlenderbotEncoder(BlenderbotPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -935,7 +935,7 @@ class BlenderbotDecoder(BlenderbotPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
         """
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (
@@ -1518,7 +1518,7 @@ class BlenderbotForCausalLM(BlenderbotPreTrainedModel):
                 Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                 for more detail.
             return_dict (`bool`, *optional*):
-                Whether or not to return a [`~file_utils.ModelOutput`] instead of a plain tuple.
+                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
 
         Returns: