Big file_utils cleanup (#16396)

* Big file_utils cleanup

* This one still needs to be treated separately

src/transformers/models/unispeech_sat/modeling_unispeech_sat.py

@@ -1116,7 +1116,7 @@ UNISPEECH_SAT_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/van/modeling_van.py

@@ -444,7 +444,7 @@ VAN_INPUTS_DOCSTRING = r"""
             Whether or not to return the hidden states of all stages. 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/vilt/feature_extraction_vilt.py

@@ -145,7 +145,7 @@ class ViltFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
         Args:
             pixel_values_list (`List[torch.Tensor]`):
                 List of images (pixel values) to be padded. Each image should be a tensor of shape (C, H, W).
-            return_tensors (`str` or [`~file_utils.TensorType`], *optional*):
+            return_tensors (`str` or [`~utils.TensorType`], *optional*):
                 If set, will return tensors instead of NumPy arrays. If set to `'pt'`, return PyTorch `torch.Tensor`
                 objects.
 
@@ -208,7 +208,7 @@ class ViltFeatureExtractor(FeatureExtractionMixin, ImageFeatureExtractionMixin):
                 - 1 for pixels that are real (i.e. **not masked**),
                 - 0 for pixels that are padding (i.e. **masked**).
 
-            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/vilt/modeling_vilt.py

@@ -660,7 +660,7 @@ VILT_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.
 """
 
 VILT_IMAGES_AND_TEXT_CLASSIFICATION_INPUTS_DOCSTRING = r"""
@@ -715,7 +715,7 @@ VILT_IMAGES_AND_TEXT_CLASSIFICATION_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/vision_encoder_decoder/modeling_flax_vision_encoder_decoder.py

@@ -107,7 +107,7 @@ VISION_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.
 """
 
 VISION_ENCODER_DECODER_ENCODE_INPUTS_DOCSTRING = r"""
@@ -122,7 +122,7 @@ VISION_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.
 """
 
 VISION_ENCODER_DECODER_DECODE_INPUTS_DOCSTRING = r"""
@@ -161,8 +161,8 @@ VISION_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/vision_encoder_decoder/modeling_tf_vision_encoder_decoder.py

@@ -132,7 +132,7 @@ VISION_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/vision_encoder_decoder/modeling_vision_encoder_decoder.py

@@ -136,7 +136,7 @@ VISION_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/vision_text_dual_encoder/modeling_flax_vision_text_dual_encoder.py

@@ -114,7 +114,7 @@ VISION_TEXT_DUAL_ENCODER_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/vision_text_dual_encoder/modeling_vision_text_dual_encoder.py

@@ -89,7 +89,7 @@ VISION_TEXT_DUAL_ENCODER_TEXT_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.
 """
 
 VISION_TEXT_DUAL_ENCODER_VISION_INPUTS_DOCSTRING = r"""
@@ -104,7 +104,7 @@ VISION_TEXT_DUAL_ENCODER_VISION_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.
 """
 
 VISION_TEXT_DUAL_ENCODER_INPUTS_DOCSTRING = r"""
@@ -142,7 +142,7 @@ VISION_TEXT_DUAL_ENCODER_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/vision_text_dual_encoder/processing_vision_text_dual_encoder.py

@@ -60,7 +60,7 @@ class VisionTextDualEncoderProcessor(ProcessorMixin):
                 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*):
+            return_tensors (`str` or [`~utils.TensorType`], *optional*):
                 If set, will return tensors of a particular framework. Acceptable values are:
 
                 - `'tf'`: Return TensorFlow `tf.constant` objects.

src/transformers/models/visual_bert/modeling_visual_bert.py

@@ -669,7 +669,7 @@ VISUAL_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/vit/feature_extraction_vit.py

@@ -98,7 +98,7 @@ class ViTFeatureExtractor(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/vit/modeling_flax_vit.py

@@ -79,7 +79,7 @@ VIT_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/vit/modeling_tf_vit.py

@@ -626,8 +626,8 @@ VIT_INPUTS_DOCSTRING = r"""
         interpolate_pos_encoding (`bool`, *optional*):
             Whether to interpolate the pre-trained position encodings.
         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/vit/modeling_vit.py

@@ -500,7 +500,7 @@ VIT_INPUTS_DOCSTRING = r"""
         interpolate_pos_encoding (`bool`, *optional*):
             Whether to interpolate the pre-trained position encodings.
         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/vit_mae/modeling_vit_mae.py

@@ -631,7 +631,7 @@ VIT_MAE_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/wav2vec2/feature_extraction_wav2vec2.py

@@ -118,7 +118,7 @@ class Wav2Vec2FeatureExtractor(SequenceFeatureExtractor):
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
                 values, a list of numpy arrays or a list of list of float values.
-            padding (`bool`, `str` or [`~file_utils.PaddingStrategy`], *optional*, defaults to `False`):
+            padding (`bool`, `str` or [`~utils.PaddingStrategy`], *optional*, defaults to `False`):
                 Select a strategy to pad the returned sequences (according to the model's padding side and padding
                 index) among:
 
@@ -156,7 +156,7 @@ class Wav2Vec2FeatureExtractor(SequenceFeatureExtractor):
 
                 </Tip>
 
-            return_tensors (`str` or [`~file_utils.TensorType`], *optional*):
+            return_tensors (`str` or [`~utils.TensorType`], *optional*):
                 If set, will return tensors instead of list of python integers. Acceptable values are:
 
                 - `'tf'`: Return TensorFlow `tf.constant` objects.

src/transformers/models/wav2vec2/modeling_flax_wav2vec2.py

@@ -281,7 +281,7 @@ WAV_2_VEC_2_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/wav2vec2/modeling_tf_wav2vec2.py

@@ -1413,8 +1413,8 @@ WAV_2_VEC_2_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/wav2vec2/modeling_wav2vec2.py

@@ -1226,7 +1226,7 @@ WAV_2_VEC_2_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.
 """