[`Styling`] stylify using ruff (#27144)

* try to stylify using ruff

* might need to remove these changes?

* use ruf format andruff check

* use isinstance instead of type comparision

* use # fmt: skip

* use # fmt: skip

* nits

* soem styling changes

* update ci job

* nits isinstance

* more files update

* nits

* more nits

* small nits

* check and format

* revert wrong changes

* actually use formatter instead of checker

* nits

* well docbuilder is overwriting this commit

* revert notebook changes

* try to nuke docbuilder

* style

* fix feature exrtaction test

* remve `indent-width = 4`

* fixup

* more nits

* update the ruff version that we use

* style

* nuke docbuilder styling

* leve the print for detected changes

* nits

* Remove file I/O
Co-authored-by: charliermarsh <charlie.r.marsh@gmail.com>

* style

* nits

* revert notebook changes

* Add # fmt skip when possible

* Add # fmt skip when possible

* Fix

* More `  # fmt: skip` usage

* More `  # fmt: skip` usage

* More `  # fmt: skip` usage

* NIts

* more fixes

* fix tapas

* Another way to skip

* Recommended way

* Fix two more fiels

* Remove asynch
Remove asynch

---------
Co-authored-by: charliermarsh <charlie.r.marsh@gmail.com>

src/transformers/models/gptsan_japanese/configuration_gptsan_japanese.py

@@ -87,6 +87,7 @@ class GPTSanJapaneseConfig(PretrainedConfig):
         use_cache (`bool`, *optional*, defaults to `True`):
             Whether or not the model should return the last key/values attentions (not used by all models)
     """
+
     model_type = "gptsan-japanese"
     keys_to_ignore_at_inference = [
         "past_key_values",

src/transformers/models/gptsan_japanese/modeling_gptsan_japanese.py

@@ -1288,7 +1288,7 @@ class GPTSanJapaneseForConditionalGeneration(GPTSanJapanesePreTrainedModel):
         past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
         **kwargs,
     ):
-        if type(spout) is list:
+        if isinstance(spout, list):
             spout = torch.tensor(spout).float()
             if input_ids is not None:
                 spout = spout.to(input_ids.device)

src/transformers/models/gptsan_japanese/tokenization_gptsan_japanese.py

@@ -375,7 +375,7 @@ class GPTSanJapaneseTokenizer(PreTrainedTokenizer):
         verbose: bool = True,
     ) -> BatchEncoding:
         # This tokenizer converts input text pairs into Prefix input and subsequent input
-        if type(batch_text_or_text_pairs[0]) is tuple or type(batch_text_or_text_pairs[0]) is list:
+        if isinstance(batch_text_or_text_pairs[0], tuple) or isinstance(tuple(batch_text_or_text_pairs[0]), list):
             # As a single text with an explicit un-prefix position
             batch_prefix_texts = []
             for pref, txt in batch_text_or_text_pairs:

src/transformers/models/graphormer/configuration_graphormer.py

@@ -129,6 +129,7 @@ class GraphormerConfig(PretrainedConfig):
             >>> configuration = model.config
             ```
     """
+
     model_type = "graphormer"
     keys_to_ignore_at_inference = ["past_key_values"]
 

src/transformers/models/groupvit/configuration_groupvit.py

@@ -88,6 +88,7 @@ class GroupViTTextConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "groupvit_text_model"
 
     def __init__(

src/transformers/models/groupvit/modeling_groupvit.py

@@ -783,9 +783,7 @@ class GroupViTPreTrainedModel(PreTrainedModel):
             nn.init.normal_(module.out_proj.weight, std=out_proj_std)
         elif isinstance(module, GroupViTMLP):
             factor = self.config.initializer_factor
-            in_proj_std = (
-                (module.config.hidden_size**-0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
-            )
+            in_proj_std = (module.config.hidden_size**-0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
             fc_std = (2 * module.config.hidden_size) ** -0.5 * factor
             nn.init.normal_(module.fc1.weight, std=fc_std)
             nn.init.normal_(module.fc2.weight, std=in_proj_std)

src/transformers/models/hubert/configuration_hubert.py

@@ -159,6 +159,7 @@ class HubertConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "hubert"
 
     def __init__(

src/transformers/models/idefics/configuration_idefics.py

@@ -72,6 +72,7 @@ class IdeficsVisionConfig(PretrainedConfig):
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
     """
+
     model_type = "idefics"
     attribute_map = {
         "hidden_size": "embed_dim",
@@ -134,6 +135,7 @@ class IdeficsPerceiverConfig(PretrainedConfig):
         qk_layer_norms_perceiver (`bool`, *optional*, defaults to `False`):
             Whether or not to use qk layer norms in perceiver
     """
+
     model_type = "idefics"
 
     def __init__(
@@ -236,6 +238,7 @@ class IdeficsConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "idefics"
     is_composition = False
 

src/transformers/models/idefics/processing_idefics.py

@@ -119,6 +119,7 @@ class IdeficsProcessor(ProcessorMixin):
             An instance of [`LlamaTokenizerFast`]. The tokenizer is a required input.
         image_size (`int`, *optional*, defaults to 224): Image size (assuming a square image)
     """
+
     attributes = ["image_processor", "tokenizer"]
     image_processor_class = "IdeficsImageProcessor"
     tokenizer_class = "LlamaTokenizerFast"

src/transformers/models/informer/configuration_informer.py

@@ -133,6 +133,7 @@ class InformerConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "informer"
     attribute_map = {
         "hidden_size": "d_model",

src/transformers/models/informer/modeling_informer.py

@@ -1222,7 +1222,8 @@ class InformerEncoder(InformerPreTrainedModel):
 # Copied from transformers.models.time_series_transformer.modeling_time_series_transformer.TimeSeriesTransformerDecoder with TimeSeriesTransformer->Informer,TimeSeriesTransformerConfig->InformerConfig,time-series-transformer->informer,Transformer->Informer,TimeSeries->Informer
 class InformerDecoder(InformerPreTrainedModel):
     """
-    Informer decoder consisting of *config.decoder_layers* layers. Each layer is a [`InformerDecoderLayer`]
+    Informer decoder consisting of *config.decoder_layers* layers. Each layer is a
+    [`InformerDecoderLayer`]
 
     Args:
         config: InformerConfig
@@ -1781,7 +1782,9 @@ class InformerForPrediction(InformerPreTrainedModel):
         ... )
         >>> batch = torch.load(file)
 
-        >>> model = InformerForPrediction.from_pretrained("huggingface/informer-tourism-monthly")
+        >>> model = InformerForPrediction.from_pretrained(
+        ...     "huggingface/informer-tourism-monthly"
+        ... )
 
         >>> # during training, one provides both past and future values
         >>> # as well as possible additional features

src/transformers/models/instructblip/configuration_instructblip.py

@@ -192,6 +192,7 @@ class InstructBlipQFormerConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "instructblip_qformer"
 
     def __init__(

src/transformers/models/instructblip/modeling_instructblip.py

@@ -1176,13 +1176,13 @@ class InstructBlipQFormerModel(InstructBlipPreTrainedModel):
         # If a 2D or 3D attention mask is provided for the cross-attention
         # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
         if encoder_hidden_states is not None:
-            if type(encoder_hidden_states) == list:
+            if isinstance(encoder_hidden_states, list):
                 encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states[0].size()
             else:
                 encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
             encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
 
-            if type(encoder_attention_mask) == list:
+            if isinstance(encoder_attention_mask, list):
                 encoder_extended_attention_mask = [self.invert_attention_mask(mask) for mask in encoder_attention_mask]
             elif encoder_attention_mask is None:
                 encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)

src/transformers/models/instructblip/processing_instructblip.py

@@ -43,6 +43,7 @@ class InstructBlipProcessor(ProcessorMixin):
         qformer_tokenizer (`AutoTokenizer`):
             An instance of ['PreTrainedTokenizer`]. The Q-Former tokenizer is a required input.
     """
+
     attributes = ["image_processor", "tokenizer"]
     image_processor_class = "BlipImageProcessor"
     tokenizer_class = "AutoTokenizer"
@@ -142,8 +143,8 @@ class InstructBlipProcessor(ProcessorMixin):
     # Copied from transformers.models.blip.processing_blip.BlipProcessor.decode with BertTokenizerFast->PreTrainedTokenizer
     def decode(self, *args, **kwargs):
         """
-        This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer
-        to the docstring of this method for more information.
+        This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to
+        the docstring of this method for more information.
         """
         return self.tokenizer.decode(*args, **kwargs)
 

src/transformers/models/kosmos2/configuration_kosmos2.py

@@ -77,6 +77,7 @@ class Kosmos2TextConfig(PretrainedConfig):
         use_cache (`bool`, *optional*, defaults to `True`):
             Whether or not the model should return the last key/values attentions (not used by all models).
     ```"""
+
     model_type = "kosmos_2_text_model"
     keys_to_ignore_at_inference = ["past_key_values"]
     attribute_map = {
@@ -271,6 +272,7 @@ class Kosmos2Config(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "kosmos-2"
     is_composition = True
 

src/transformers/models/kosmos2/modeling_kosmos2.py

@@ -1374,9 +1374,7 @@ class Kosmos2PreTrainedModel(PreTrainedModel):
             if module.out_proj.bias is not None:
                 module.out_proj.bias.data.zero_()
         elif isinstance(module, Kosmos2VisionMLP):
-            in_proj_std = (
-                (module.config.hidden_size**-0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
-            )
+            in_proj_std = (module.config.hidden_size**-0.5) * ((2 * module.config.num_hidden_layers) ** -0.5) * factor
             fc_std = (2 * module.config.hidden_size) ** -0.5 * factor
             nn.init.normal_(module.fc1.weight, std=fc_std)
             nn.init.normal_(module.fc2.weight, std=in_proj_std)

src/transformers/models/kosmos2/processing_kosmos2.py

@@ -52,6 +52,7 @@ class Kosmos2Processor(ProcessorMixin):
         num_patch_index_tokens (`int`, *optional*, defaults to 1024):
             The number of tokens that represent patch indices.
     """
+
     attributes = ["image_processor", "tokenizer"]
     image_processor_class = "CLIPImageProcessor"
     tokenizer_class = ("XLMRobertaTokenizer", "XLMRobertaTokenizerFast")
@@ -390,8 +391,8 @@ class Kosmos2Processor(ProcessorMixin):
     # Copied from transformers.models.blip.processing_blip.BlipProcessor.decode with BertTokenizerFast->PreTrainedTokenizer
     def decode(self, *args, **kwargs):
         """
-        This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer
-        to the docstring of this method for more information.
+        This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to
+        the docstring of this method for more information.
         """
         return self.tokenizer.decode(*args, **kwargs)
 

src/transformers/models/layoutlm/configuration_layoutlm.py

@@ -101,6 +101,7 @@ class LayoutLMConfig(PretrainedConfig):
     >>> # Accessing the model configuration
     >>> configuration = model.config
     ```"""
+
     model_type = "layoutlm"
 
     def __init__(

src/transformers/models/layoutlm/tokenization_layoutlm.py

@@ -262,8 +262,8 @@ class LayoutLMTokenizer(PreTrainedTokenizer):
         self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
     ) -> List[int]:
         """
-        Create a mask from the two sequences passed to be used in a sequence-pair classification task. A LayoutLM
-        sequence pair mask has the following format:
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. A LayoutLM sequence
+        pair mask has the following format:
 
         ```
         0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1

src/transformers/models/layoutlm/tokenization_layoutlm_fast.py

@@ -175,8 +175,8 @@ class LayoutLMTokenizerFast(PreTrainedTokenizerFast):
         self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
     ) -> List[int]:
         """
-        Create a mask from the two sequences passed to be used in a sequence-pair classification task. A LayoutLM
-        sequence pair mask has the following format:
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. A LayoutLM sequence
+        pair mask has the following format:
 
         ```
         0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1