fixed docstring typos (#18739)

* fixed docstring typos

* Added missing colon

Co-authored-by: 김주영 <juyoung@zezedu.com>

src/transformers/models/bart/configuration_bart.py

@@ -77,17 +77,17 @@ class BartConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):
             Scale embeddings by diving by sqrt(d_model).
         use_cache (`bool`, *optional*, defaults to `True`):
             Whether or not the model should return the last key/values attentions (not used by all models).
-        num_labels: (`int`, *optional*, defaults to 3):
+        num_labels (`int`, *optional*, defaults to 3):
             The number of labels to use in [`BartForSequenceClassification`].
         forced_eos_token_id (`int`, *optional*, defaults to 2):
             The id of the token to force as the last generated token when `max_length` is reached. Usually set to

src/transformers/models/bert/tokenization_bert.py

@@ -383,7 +383,7 @@ class BasicTokenizer(object):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/bigbird_pegasus/configuration_bigbird_pegasus.py

@@ -85,10 +85,10 @@ class BigBirdPegasusConfig(PretrainedConfig):
             just in case (e.g., 1024 or 2048 or 4096).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         use_cache (`bool`, *optional*, defaults to `True`):

src/transformers/models/blenderbot/configuration_blenderbot.py

@@ -78,10 +78,10 @@ class BlenderbotConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):

src/transformers/models/blenderbot_small/configuration_blenderbot_small.py

@@ -78,10 +78,10 @@ class BlenderbotSmallConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):

src/transformers/models/detr/configuration_detr.py

@@ -74,10 +74,10 @@ class DetrConfig(PretrainedConfig):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         init_xavier_std (`float`, *optional*, defaults to 1):
             The scaling factor used for the Xavier initialization gain in the HM Attention map module.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         auxiliary_loss (`bool`, *optional*, defaults to `False`):

src/transformers/models/dpr/modeling_dpr.py

@@ -404,7 +404,7 @@ DPR_ENCODERS_INPUTS_DOCSTRING = r"""
 
 DPR_READER_INPUTS_DOCSTRING = r"""
     Args:
-        input_ids: (`Tuple[torch.LongTensor]` of shapes `(n_passages, sequence_length)`):
+        input_ids (`Tuple[torch.LongTensor]` of shapes `(n_passages, sequence_length)`):
             Indices of input sequence tokens in the vocabulary. It has to be a sequence triplet with 1) the question
             and 2) the passages titles and 3) the passages texts To match pretraining, DPR `input_ids` sequence should
             be formatted with [CLS] and [SEP] with the format:

src/transformers/models/dpr/modeling_tf_dpr.py

@@ -493,7 +493,7 @@ TF_DPR_ENCODERS_INPUTS_DOCSTRING = r"""
 
 TF_DPR_READER_INPUTS_DOCSTRING = r"""
     Args:
-        input_ids: (`Numpy array` or `tf.Tensor` of shapes `(n_passages, sequence_length)`):
+        input_ids (`Numpy array` or `tf.Tensor` of shapes `(n_passages, sequence_length)`):
             Indices of input sequence tokens in the vocabulary. It has to be a sequence triplet with 1) the question
             and 2) the passages titles and 3) the passages texts To match pretraining, DPR `input_ids` sequence should
             be formatted with [CLS] and [SEP] with the format:

src/transformers/models/encoder_decoder/modeling_encoder_decoder.py

@@ -136,7 +136,7 @@ ENCODER_DECODER_INPUTS_DOCSTRING = r"""
             more detail.
         return_dict (`bool`, *optional*):
             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:
+        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.
             - With a *decoder_* prefix which will be input as `**decoder_kwargs` for the decoder forward function.

src/transformers/models/encoder_decoder/modeling_tf_encoder_decoder.py

@@ -147,7 +147,7 @@ ENCODER_DECODER_INPUTS_DOCSTRING = r"""
         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).
-        kwargs: (*optional*) Remaining dictionary of keyword arguments. Keyword arguments come in two flavors:
+        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.
             - With a *decoder_* prefix which will be input as `**decoder_kwargs`` for the decoder forward function.

src/transformers/models/fsmt/configuration_fsmt.py

@@ -95,9 +95,9 @@ class FSMTConfig(PretrainedConfig):
             End of stream token id.
         decoder_start_token_id (`int`, *optional*):
             This model starts decoding with `eos_token_id`
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             Google "layerdrop arxiv", as its not explainable in one line.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             Google "layerdrop arxiv", as its not explainable in one line.
         is_encoder_decoder (`bool`, *optional*, defaults to `True`):
             Whether this is an encoder/decoder model.

src/transformers/models/layoutlmv2/tokenization_layoutlmv2.py

@@ -1362,7 +1362,7 @@ class BasicTokenizer(object):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/layoutlmv2/tokenization_layoutlmv2_fast.py

@@ -108,7 +108,7 @@ class LayoutLMv2TokenizerFast(PreTrainedTokenizerFast):
         tokenize_chinese_chars (`bool`, *optional*, defaults to `True`):
             Whether or not to tokenize Chinese characters. This should likely be deactivated for Japanese (see [this
             issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original LayoutLMv2).
     """

src/transformers/models/led/configuration_led.py

@@ -74,10 +74,10 @@ class LEDConfig(PretrainedConfig):
             The maximum sequence length that the decoder might ever be used with.
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         use_cache (`bool`, *optional*, defaults to `True`):

src/transformers/models/m2m_100/configuration_m2m_100.py

@@ -76,10 +76,10 @@ class M2M100Config(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         use_cache (`bool`, *optional*, defaults to `True`):

src/transformers/models/marian/configuration_marian.py

@@ -76,10 +76,10 @@ class MarianConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):

src/transformers/models/mbart/configuration_mbart.py

@@ -76,10 +76,10 @@ class MBartConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):

src/transformers/models/mpnet/tokenization_mpnet.py

@@ -340,7 +340,7 @@ class BasicTokenizer(object):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/pegasus/configuration_pegasus.py

@@ -71,10 +71,10 @@ class PegasusConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `False`):

src/transformers/models/plbart/configuration_plbart.py

@@ -74,10 +74,10 @@ class PLBartConfig(PretrainedConfig):
             just in case (e.g., 512 or 1024 or 2048).
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         scale_embedding (`bool`, *optional*, defaults to `True`):

src/transformers/models/prophetnet/tokenization_prophetnet.py

@@ -94,7 +94,7 @@ class ProphetNetTokenizer(PreTrainedTokenizer):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/rag/configuration_rag.py

@@ -49,7 +49,7 @@ RAG_CONFIG_DOC = r"""
             `"compressed"`.
         index_path (`str`, *optional*)
             The path to the serialized faiss index on disk.
-        passages_path: (`str`, *optional*):
+        passages_path (`str`, *optional*):
             A path to text passages compatible with the faiss index. Required if using
             [`~models.rag.retrieval_rag.LegacyIndex`]
         use_dummy_dataset (`bool`, *optional*, defaults to `False`)

src/transformers/models/realm/tokenization_realm.py

@@ -132,7 +132,7 @@ class RealmTokenizer(PreTrainedTokenizer):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/roformer/tokenization_roformer.py

@@ -103,7 +103,7 @@ class RoFormerTokenizer(PreTrainedTokenizer):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
 

src/transformers/models/speech_encoder_decoder/modeling_speech_encoder_decoder.py

@@ -143,7 +143,7 @@ SPEECH_ENCODER_DECODER_INPUTS_DOCSTRING = r"""
             into a tensor of type `torch.FloatTensor`. See [`~Speech2TextFeatureExtractor.__call__`]
         return_dict (`bool`, *optional*):
             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:
+        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.
             - With a *decoder_* prefix which will be input as `**decoder_kwargs` for the decoder forward function.

src/transformers/models/speech_to_text/configuration_speech_to_text.py

@@ -70,10 +70,10 @@ class Speech2TextConfig(PretrainedConfig):
             The dropout ratio for classifier.
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        encoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        encoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         use_cache (`bool`, *optional*, defaults to `True`):

src/transformers/models/speech_to_text_2/configuration_speech_to_text_2.py

@@ -64,14 +64,15 @@ class Speech2Text2Config(PretrainedConfig):
             The dropout ratio for classifier.
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-            https://arxiv.org/abs/1909.11556>`__ for more details. decoder_layerdrop: (`float`, *optional*, defaults to
-            0.0): The LayerDrop probability for the decoder. See the [LayerDrop paper](see
-            https://arxiv.org/abs/1909.11556) for more details.
+            https://arxiv.org/abs/1909.11556>`__ for more details.
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
+            The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
+            for more details.
         use_cache (`bool`, *optional*, defaults to `True`):
             Whether or not the model should return the last key/values attentions (not used by all models).
         max_source_positions (`int`, *optional*, defaults to 6000):
             The maximum sequence length of log-mel filter-bank features that this model might ever be used with.
-        max_target_positions: (`int`, *optional*, defaults to 1024):
+        max_target_positions (`int`, *optional*, defaults to 1024):
             The maximum sequence length that this model might ever be used with. Typically set this to something large
             just in case (e.g., 512 or 1024 or 2048).
 

src/transformers/models/splinter/tokenization_splinter.py

@@ -111,7 +111,7 @@ class SplinterTokenizer(PreTrainedTokenizer):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """
@@ -340,7 +340,7 @@ class BasicTokenizer(object):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/splinter/tokenization_splinter_fast.py

@@ -87,10 +87,10 @@ class SplinterTokenizerFast(PreTrainedTokenizerFast):
         tokenize_chinese_chars (`bool`, *optional*, defaults to `True`):
             Whether or not to tokenize Chinese characters. This should likely be deactivated for Japanese (see [this
             issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
-        wordpieces_prefix: (`str`, *optional*, defaults to `"##"`):
+        wordpieces_prefix (`str`, *optional*, defaults to `"##"`):
             The prefix for subwords.
     """
 

src/transformers/models/t5/modeling_tf_t5.py

@@ -1008,14 +1008,14 @@ T5_INPUTS_DOCSTRING = r"""
         decoder_attention_mask (`tf.Tensor` of shape `(batch_size, target_sequence_length)`, *optional*):
             Default behavior: generate a tensor that ignores pad tokens in `decoder_input_ids`. Causal mask will also
             be used by default.
-        head_mask: (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+        head_mask (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
             Mask to nullify selected heads of the self-attention modules in the encoder. Mask values selected in `[0,
             1]`:
 
             - 1 indicates the head is **not masked**,
             - 0 indicates the head is **masked**.
 
-        decoder_head_mask: (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+        decoder_head_mask (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
             Mask to nullify selected heads of the self-attention modules in the decoder. Mask values selected in `[0,
             1]`:
 
@@ -1084,7 +1084,7 @@ T5_ENCODER_INPUTS_DOCSTRING = r"""
             Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
             is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
             model's internal embedding lookup matrix.
-        head_mask: (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+        head_mask (`tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
             Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
 
             - 1 indicates the head is **not masked**,

src/transformers/models/tapas/tokenization_tapas.py

@@ -293,7 +293,7 @@ class TapasTokenizer(PreTrainedTokenizer):
         tokenize_chinese_chars (`bool`, *optional*, defaults to `True`):
             Whether or not to tokenize Chinese characters. This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
         cell_trim_length (`int`, *optional*, defaults to -1):
@@ -2053,7 +2053,7 @@ class BasicTokenizer(object):
 
             This should likely be deactivated for Japanese (see this
             [issue](https://github.com/huggingface/transformers/issues/328)).
-        strip_accents: (`bool`, *optional*):
+        strip_accents (`bool`, *optional*):
             Whether or not to strip all accents. If this option is not specified, then it will be determined by the
             value for `lowercase` (as in the original BERT).
     """

src/transformers/models/trocr/configuration_trocr.py

@@ -67,7 +67,7 @@ class TrOCRConfig(PretrainedConfig):
             The dropout ratio for classifier.
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        decoder_layerdrop: (`float`, *optional*, defaults to 0.0):
+        decoder_layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         use_cache (`bool`, *optional*, defaults to `True`):

src/transformers/models/vision_encoder_decoder/modeling_tf_vision_encoder_decoder.py

@@ -136,7 +136,7 @@ VISION_ENCODER_DECODER_INPUTS_DOCSTRING = r"""
         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).
-        kwargs: (*optional*) Remaining dictionary of keyword arguments. Keyword arguments come in two flavors:
+        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.
             - With a *decoder_* prefix which will be input as `**decoder_kwargs` for the decoder forward function.

src/transformers/models/vision_encoder_decoder/modeling_vision_encoder_decoder.py

@@ -137,7 +137,7 @@ VISION_ENCODER_DECODER_INPUTS_DOCSTRING = r"""
             more detail.
         return_dict (`bool`, *optional*):
             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:
+        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.
             - With a *decoder_* prefix which will be input as `**decoder_kwargs` for the decoder forward function.

src/transformers/models/xglm/configuration_xglm.py

@@ -61,7 +61,7 @@ class XGLMConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         activation_dropout (`float`, *optional*, defaults to 0.0):
             The dropout ratio for activations inside the fully connected layer.
-        layerdrop: (`float`, *optional*, defaults to 0.0):
+        layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
             for more details.
         init_std (`float`, *optional*, defaults to 0.02):