Big file_utils cleanup (#16396)

* Big file_utils cleanup

* This one still needs to be treated separately

src/transformers/models/encoder_decoder/modeling_encoder_decoder.py

@@ -135,7 +135,7 @@ ENCODER_DECODER_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*):
-            If set to `True`, the model will return a [`~file_utils.Seq2SeqLMOutput`] instead of a plain tuple.
+            If set to `True`, the model will return a [`~utils.Seq2SeqLMOutput`] instead of a plain tuple.
         kwargs: (*optional*) Remaining dictionary of keyword arguments. Keyword arguments come in two flavors:
 
             - Without a prefix which will be input as `**encoder_kwargs` for the encoder forward function.

src/transformers/models/encoder_decoder/modeling_flax_encoder_decoder.py

@@ -122,7 +122,7 @@ ENCODER_DECODER_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*):
-            If set to `True`, the model will return a [`~file_utils.FlaxSeq2SeqLMOutput`] instead of a plain tuple.
+            If set to `True`, the model will return a [`~utils.FlaxSeq2SeqLMOutput`] instead of a plain tuple.
 """
 
 ENCODER_DECODER_ENCODE_INPUTS_DOCSTRING = r"""
@@ -152,7 +152,7 @@ ENCODER_DECODER_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*):
-            If set to `True`, the model will return a [`~file_utils.FlaxBaseModelOutput`] instead of a plain tuple.
+            If set to `True`, the model will return a [`~utils.FlaxBaseModelOutput`] instead of a plain tuple.
 """
 
 ENCODER_DECODER_DECODE_INPUTS_DOCSTRING = r"""
@@ -198,8 +198,8 @@ ENCODER_DECODER_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*):
-            If set to `True`, the model will return a [`~file_utils.FlaxCausalLMOutputWithCrossAttentions`] instead of
-            a plain tuple.
+            If set to `True`, the model will return a [`~utils.FlaxCausalLMOutputWithCrossAttentions`] instead of a
+            plain tuple.
 """
 
 

src/transformers/models/encoder_decoder/modeling_tf_encoder_decoder.py

@@ -143,7 +143,7 @@ ENCODER_DECODER_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*):
-            If set to `True`, the model will return a [`~file_utils.Seq2SeqLMOutput`] instead of a plain tuple.
+            If set to `True`, the model will return a [`~utils.Seq2SeqLMOutput`] instead of a plain tuple.
         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/flaubert/modeling_flaubert.py

@@ -123,7 +123,7 @@ FLAUBERT_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/flaubert/modeling_tf_flaubert.py

@@ -165,8 +165,8 @@ FLAUBERT_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/fnet/modeling_fnet.py

@@ -507,7 +507,7 @@ FNET_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/fsmt/modeling_fsmt.py

@@ -282,7 +282,7 @@ FSMT_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/funnel/modeling_funnel.py

@@ -913,7 +913,7 @@ FUNNEL_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/funnel/modeling_tf_funnel.py

@@ -1079,8 +1079,8 @@ FUNNEL_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/glpn/feature_extraction_glpn.py

@@ -86,7 +86,7 @@ class GLPNFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
                 tensor. In case of a NumPy array/PyTorch tensor, each image should be of shape (C, H, W), where C is a
                 number of channels, H and W are image height and width.
 
-            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/glpn/modeling_glpn.py

@@ -467,7 +467,7 @@ GLPN_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/gpt2/modeling_flax_gpt2.py

@@ -103,7 +103,7 @@ GPT2_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/gpt2/modeling_gpt2.py

@@ -607,7 +607,7 @@ GPT2_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.
 """
 PARALLELIZE_DOCSTRING = r"""
     This is an experimental feature and is a subject to change at a moment's notice.

src/transformers/models/gpt2/modeling_tf_gpt2.py

@@ -715,8 +715,8 @@ GPT2_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/gpt_neo/modeling_flax_gpt_neo.py

@@ -101,7 +101,7 @@ GPT_NEO_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/gpt_neo/modeling_gpt_neo.py

@@ -463,7 +463,7 @@ GPT_NEO_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/gptj/modeling_flax_gptj.py

@@ -103,7 +103,7 @@ GPTJ_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/gptj/modeling_gptj.py

@@ -390,7 +390,7 @@ GPTJ_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.
 """
 
 PARALLELIZE_DOCSTRING = r"""

src/transformers/models/hubert/modeling_hubert.py

@@ -931,7 +931,7 @@ HUBERT_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/hubert/modeling_tf_hubert.py

@@ -1387,8 +1387,8 @@ HUBERT_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).