From 0148b0a10af72885fcdd9ab92aa1c02c50ffff42 Mon Sep 17 00:00:00 2001
From: zspo <songpo.zhang@foxmail.com>
Date: Wed, 24 May 2023 02:50:40 +0800
Subject: [PATCH] Fix some docs what layerdrop does (#23691)

* Fix some docs what layerdrop does

* Update src/transformers/models/data2vec/configuration_data2vec_audio.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Fix more docs

---------

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
---
 .../pabee/modeling_pabee_albert.py                 |  2 +-
 src/transformers/generation/logits_process.py      |  2 +-
 src/transformers/modeling_outputs.py               |  4 ++--
 src/transformers/modeling_utils.py                 |  2 +-
 .../models/align/configuration_align.py            |  2 +-
 src/transformers/models/blip/modeling_blip_text.py |  2 +-
 .../models/blip/modeling_tf_blip_text.py           |  2 +-
 src/transformers/models/blip_2/modeling_blip_2.py  |  2 +-
 src/transformers/models/bloom/modeling_bloom.py    |  2 +-
 .../data2vec/configuration_data2vec_audio.py       |  3 +++
 .../configuration_deformable_detr.py               |  2 +-
 src/transformers/models/deta/configuration_deta.py |  2 +-
 .../efficientnet/configuration_efficientnet.py     |  2 +-
 .../models/hubert/configuration_hubert.py          |  3 +++
 src/transformers/models/lxmert/modeling_lxmert.py  | 14 +++++++-------
 .../models/lxmert/modeling_tf_lxmert.py            | 10 +++++-----
 .../models/mask2former/modeling_mask2former.py     |  4 ++--
 .../models/mpnet/tokenization_mpnet.py             |  2 +-
 .../models/mpnet/tokenization_mpnet_fast.py        |  2 +-
 src/transformers/models/opt/configuration_opt.py   |  2 +-
 .../models/pegasus_x/configuration_pegasus_x.py    |  4 ++--
 src/transformers/models/rag/modeling_rag.py        |  2 +-
 src/transformers/models/rag/retrieval_rag.py       |  4 ++--
 src/transformers/models/realm/modeling_realm.py    |  2 +-
 src/transformers/models/sew/configuration_sew.py   |  3 +++
 .../models/unispeech/configuration_unispeech.py    |  3 +++
 .../unispeech_sat/configuration_unispeech_sat.py   |  3 +++
 .../models/wav2vec2/configuration_wav2vec2.py      |  3 +++
 .../configuration_wav2vec2_conformer.py            |  3 +++
 .../models/wavlm/configuration_wavlm.py            |  3 +++
 src/transformers/optimization_tf.py                |  4 ++--
 src/transformers/pipelines/audio_utils.py          |  2 +-
 src/transformers/pipelines/base.py                 |  2 +-
 src/transformers/tokenization_utils_base.py        |  2 +-
 src/transformers/training_args_tf.py               |  2 +-
 ...uration_{{cookiecutter.lowercase_modelname}}.py |  4 ++--
 36 files changed, 68 insertions(+), 44 deletions(-)

diff --git a/examples/research_projects/bert-loses-patience/pabee/modeling_pabee_albert.py b/examples/research_projects/bert-loses-patience/pabee/modeling_pabee_albert.py
index 5e17352dc19b..57b649ec067b 100644
--- a/examples/research_projects/bert-loses-patience/pabee/modeling_pabee_albert.py
+++ b/examples/research_projects/bert-loses-patience/pabee/modeling_pabee_albert.py
@@ -253,7 +253,7 @@ def forward(
 
         Returns:
             :obj:`tuple(torch.FloatTensor)` comprising various elements depending on the configuration (:class:`~transformers.AlbertConfig`) and inputs:
-            loss: (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
+            loss (`optional`, returned when ``labels`` is provided) ``torch.FloatTensor`` of shape ``(1,)``:
                 Classification (or regression if config.num_labels==1) loss.
             logits ``torch.FloatTensor`` of shape ``(batch_size, config.num_labels)``
                 Classification (or regression if config.num_labels==1) scores (before SoftMax).
diff --git a/src/transformers/generation/logits_process.py b/src/transformers/generation/logits_process.py
index 95c8064ee404..dd51610afd43 100644
--- a/src/transformers/generation/logits_process.py
+++ b/src/transformers/generation/logits_process.py
@@ -678,7 +678,7 @@ class PrefixConstrainedLogitsProcessor(LogitsProcessor):
     generation. See [Autoregressive Entity Retrieval](https://arxiv.org/abs/2010.00904) for more information.
 
     Args:
-        prefix_allowed_tokens_fn: (`Callable[[int, torch.Tensor], List[int]]`):
+        prefix_allowed_tokens_fn (`Callable[[int, torch.Tensor], List[int]]`):
             This function constraints the beam search to allowed tokens only at each step. This function takes 2
             arguments `inputs_ids` and the batch ID `batch_id`. It has to return a list with the allowed tokens for the
             next generation step conditioned on the previously generated tokens `inputs_ids` and the batch ID
diff --git a/src/transformers/modeling_outputs.py b/src/transformers/modeling_outputs.py
index c69e426ab531..aceec7abd406 100755
--- a/src/transformers/modeling_outputs.py
+++ b/src/transformers/modeling_outputs.py
@@ -1522,7 +1522,7 @@ class Seq2SeqTSModelOutput(ModelOutput):
         scale (`torch.FloatTensor` of shape `(batch_size,)` or `(batch_size, input_size)`, *optional*):
             Scaling values of each time series' context window which is used to give the model inputs of the same
             magnitude and then used to rescale back to the original magnitude.
-        static_features: (`torch.FloatTensor` of shape `(batch_size, feature size)`, *optional*):
+        static_features (`torch.FloatTensor` of shape `(batch_size, feature size)`, *optional*):
             Static features of each time series' in a batch which are copied to the covariates at inference time.
     """
 
@@ -1593,7 +1593,7 @@ class Seq2SeqTSPredictionOutput(ModelOutput):
         scale (`torch.FloatTensor` of shape `(batch_size,)` or `(batch_size, input_size)`, *optional*):
             Scaling values of each time series' context window which is used to give the model inputs of the same
             magnitude and then used to rescale back to the original magnitude.
-        static_features: (`torch.FloatTensor` of shape `(batch_size, feature size)`, *optional*):
+        static_features (`torch.FloatTensor` of shape `(batch_size, feature size)`, *optional*):
             Static features of each time series' in a batch which are copied to the covariates at inference time.
     """
 
diff --git a/src/transformers/modeling_utils.py b/src/transformers/modeling_utils.py
index a8e8e4b2e241..b7b832b4640e 100644
--- a/src/transformers/modeling_utils.py
+++ b/src/transformers/modeling_utils.py
@@ -912,7 +912,7 @@ def get_head_mask(
                 The mask indicating if we should keep the heads or not (1.0 for keep, 0.0 for discard).
             num_hidden_layers (`int`):
                 The number of hidden layers in the model.
-            is_attention_chunked: (`bool`, *optional*, defaults to `False`):
+            is_attention_chunked (`bool`, *optional*, defaults to `False`):
                 Whether or not the attentions scores are computed by chunks or not.
 
         Returns:
diff --git a/src/transformers/models/align/configuration_align.py b/src/transformers/models/align/configuration_align.py
index 488b7f6fe458..0436b278f0a7 100644
--- a/src/transformers/models/align/configuration_align.py
+++ b/src/transformers/models/align/configuration_align.py
@@ -184,7 +184,7 @@ class AlignVisionConfig(PretrainedConfig):
             List of output channel sizes to be used in each block for convolutional layers.
         depthwise_padding (`List[int]`, *optional*, defaults to `[]`):
             List of block indices with square padding.
-        strides: (`List[int]`, *optional*, defaults to `[1, 2, 2, 2, 1, 2, 1]`):
+        strides (`List[int]`, *optional*, defaults to `[1, 2, 2, 2, 1, 2, 1]`):
             List of stride sizes to be used in each block for convolutional layers.
         num_block_repeats (`List[int]`, *optional*, defaults to `[1, 2, 2, 3, 3, 4, 1]`):
             List of the number of times each block is to repeated.
diff --git a/src/transformers/models/blip/modeling_blip_text.py b/src/transformers/models/blip/modeling_blip_text.py
index 85585f3d5dd3..1f269cf852ee 100644
--- a/src/transformers/models/blip/modeling_blip_text.py
+++ b/src/transformers/models/blip/modeling_blip_text.py
@@ -613,7 +613,7 @@ def get_extended_attention_mask(
                 Mask with ones indicating tokens to attend to, zeros for tokens to ignore.
             input_shape (`Tuple[int]`):
                 The shape of the input to the model.
-            device: (`torch.device`):
+            device (`torch.device`):
                 The device of the input to the model.
 
         Returns:
diff --git a/src/transformers/models/blip/modeling_tf_blip_text.py b/src/transformers/models/blip/modeling_tf_blip_text.py
index 6e8ed8a891c0..02f592c25902 100644
--- a/src/transformers/models/blip/modeling_tf_blip_text.py
+++ b/src/transformers/models/blip/modeling_tf_blip_text.py
@@ -633,7 +633,7 @@ def get_extended_attention_mask(
                 Mask with ones indicating tokens to attend to, zeros for tokens to ignore.
             input_shape (`Tuple[int]`):
                 The shape of the input to the model.
-            is_decoder: (`bool`):
+            is_decoder (`bool`):
                 Whether the model is used as a decoder.
 
         Returns:
diff --git a/src/transformers/models/blip_2/modeling_blip_2.py b/src/transformers/models/blip_2/modeling_blip_2.py
index be499dc1d516..381a3fe0ed7a 100644
--- a/src/transformers/models/blip_2/modeling_blip_2.py
+++ b/src/transformers/models/blip_2/modeling_blip_2.py
@@ -1059,7 +1059,7 @@ def get_extended_attention_mask(
                 Mask with ones indicating tokens to attend to, zeros for tokens to ignore.
             input_shape (`Tuple[int]`):
                 The shape of the input to the model.
-            device: (`torch.device`):
+            device (`torch.device`):
                 The device of the input to the model.
 
         Returns:
diff --git a/src/transformers/models/bloom/modeling_bloom.py b/src/transformers/models/bloom/modeling_bloom.py
index e954cfb44730..5c0d570cbe9c 100644
--- a/src/transformers/models/bloom/modeling_bloom.py
+++ b/src/transformers/models/bloom/modeling_bloom.py
@@ -256,7 +256,7 @@ def _merge_heads(self, x: torch.Tensor) -> torch.Tensor:
         Merge heads together over the last dimenstion
 
         Args:
-            x: (`torch.tensor`, *required*): [batch_size * num_heads, seq_length, head_dim]
+            x (`torch.tensor`, *required*): [batch_size * num_heads, seq_length, head_dim]
 
         Returns:
             torch.tensor: [batch_size, seq_length, num_heads * head_dim]
diff --git a/src/transformers/models/data2vec/configuration_data2vec_audio.py b/src/transformers/models/data2vec/configuration_data2vec_audio.py
index 2ec526924f36..066d81a5daed 100644
--- a/src/transformers/models/data2vec/configuration_data2vec_audio.py
+++ b/src/transformers/models/data2vec/configuration_data2vec_audio.py
@@ -62,6 +62,9 @@ class Data2VecAudioConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`Data2VecAudioForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/deformable_detr/configuration_deformable_detr.py b/src/transformers/models/deformable_detr/configuration_deformable_detr.py
index 54614eef96f9..dbe5fd7f0a78 100644
--- a/src/transformers/models/deformable_detr/configuration_deformable_detr.py
+++ b/src/transformers/models/deformable_detr/configuration_deformable_detr.py
@@ -77,7 +77,7 @@ class DeformableDetrConfig(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.
         auxiliary_loss (`bool`, *optional*, defaults to `False`):
diff --git a/src/transformers/models/deta/configuration_deta.py b/src/transformers/models/deta/configuration_deta.py
index 836e9732e68e..dc85ea91df36 100644
--- a/src/transformers/models/deta/configuration_deta.py
+++ b/src/transformers/models/deta/configuration_deta.py
@@ -71,7 +71,7 @@ class DetaConfig(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.
         auxiliary_loss (`bool`, *optional*, defaults to `False`):
diff --git a/src/transformers/models/efficientnet/configuration_efficientnet.py b/src/transformers/models/efficientnet/configuration_efficientnet.py
index a6e4d172f442..e6b6a1c261ca 100644
--- a/src/transformers/models/efficientnet/configuration_efficientnet.py
+++ b/src/transformers/models/efficientnet/configuration_efficientnet.py
@@ -60,7 +60,7 @@ class EfficientNetConfig(PretrainedConfig):
             List of output channel sizes to be used in each block for convolutional layers.
         depthwise_padding (`List[int]`, *optional*, defaults to `[]`):
             List of block indices with square padding.
-        strides: (`List[int]`, *optional*, defaults to `[1, 2, 2, 2, 1, 2, 1]`):
+        strides (`List[int]`, *optional*, defaults to `[1, 2, 2, 2, 1, 2, 1]`):
             List of stride sizes to be used in each block for convolutional layers.
         num_block_repeats (`List[int]`, *optional*, defaults to `[1, 2, 2, 3, 3, 4, 1]`):
             List of the number of times each block is to repeated.
diff --git a/src/transformers/models/hubert/configuration_hubert.py b/src/transformers/models/hubert/configuration_hubert.py
index 139df45bbb79..1f326871c3c9 100644
--- a/src/transformers/models/hubert/configuration_hubert.py
+++ b/src/transformers/models/hubert/configuration_hubert.py
@@ -62,6 +62,9 @@ class HubertConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probabilitiy for the final projection layer of [`Wav2Vec2ForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/lxmert/modeling_lxmert.py b/src/transformers/models/lxmert/modeling_lxmert.py
index 9fe7ecb730fd..73586872e8a1 100644
--- a/src/transformers/models/lxmert/modeling_lxmert.py
+++ b/src/transformers/models/lxmert/modeling_lxmert.py
@@ -111,7 +111,7 @@ class LxmertForQuestionAnsweringOutput(ModelOutput):
         loss (*optional*, returned when `labels` is provided, `torch.FloatTensor` of shape `(1,)`):
             Total loss as the sum of the masked language modeling loss and the next sequence prediction
             (classification) loss.k.
-        question_answering_score: (`torch.FloatTensor` of shape `(batch_size, n_qa_answers)`, *optional*):
+        question_answering_score (`torch.FloatTensor` of shape `(batch_size, n_qa_answers)`, *optional*):
             Prediction scores of question answering objective (classification).
         language_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
             Tuple of `torch.FloatTensor` (one for input features + one for the output of each cross-modality layer) of
@@ -153,10 +153,10 @@ class LxmertForPreTrainingOutput(ModelOutput):
             (classification) loss.
         prediction_logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`):
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
-        cross_relationship_score: (`torch.FloatTensor` of shape `(batch_size, 2)`):
+        cross_relationship_score (`torch.FloatTensor` of shape `(batch_size, 2)`):
             Prediction scores of the textual matching objective (classification) head (scores of True/False
             continuation before SoftMax).
-        question_answering_score: (`torch.FloatTensor` of shape `(batch_size, n_qa_answers)`):
+        question_answering_score (`torch.FloatTensor` of shape `(batch_size, n_qa_answers)`):
             Prediction scores of question answering objective (classification).
         language_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
             Tuple of `torch.FloatTensor` (one for input features + one for the output of each cross-modality layer) of
@@ -828,12 +828,12 @@ def _init_weights(self, module):
             [`PreTrainedTokenizer.__call__`] for details.
 
             [What are input IDs?](../glossary#input-ids)
-        visual_feats: (`torch.FloatTensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
+        visual_feats (`torch.FloatTensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
             This input represents visual features. They ROI pooled object features from bounding boxes using a
             faster-RCNN model)
 
             These are currently not provided by the transformers library.
-        visual_pos: (`torch.FloatTensor` of shape `(batch_size, num_visual_features, visual_pos_dim)`):
+        visual_pos (`torch.FloatTensor` of shape `(batch_size, num_visual_features, visual_pos_dim)`):
             This input represents spacial features corresponding to their relative (via index) visual features. The
             pre-trained LXMERT model expects these spacial features to be normalized bounding boxes on a scale of 0 to
             1.
@@ -1171,7 +1171,7 @@ def forward(
             Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
             config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
             loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
-        obj_labels: (`Dict[Str: Tuple[Torch.FloatTensor, Torch.FloatTensor]]`, *optional*):
+        obj_labels (`Dict[Str: Tuple[Torch.FloatTensor, Torch.FloatTensor]]`, *optional*):
             each key is named after each one of the visual losses and each element of the tuple is of the shape
             `(batch_size, num_features)` and `(batch_size, num_features, visual_feature_dim)` for each the label id and
             the label score respectively
@@ -1398,7 +1398,7 @@ def forward(
         return_dict: Optional[bool] = None,
     ) -> Union[LxmertForQuestionAnsweringOutput, Tuple[torch.FloatTensor]]:
         r"""
-        labels: (`Torch.Tensor` of shape `(batch_size)`, *optional*):
+        labels (`Torch.Tensor` of shape `(batch_size)`, *optional*):
             A one-hot representation of the correct answer
         """
         return_dict = return_dict if return_dict is not None else self.config.use_return_dict
diff --git a/src/transformers/models/lxmert/modeling_tf_lxmert.py b/src/transformers/models/lxmert/modeling_tf_lxmert.py
index 948053c93e28..f02d0ff6446a 100644
--- a/src/transformers/models/lxmert/modeling_tf_lxmert.py
+++ b/src/transformers/models/lxmert/modeling_tf_lxmert.py
@@ -111,10 +111,10 @@ class TFLxmertForPreTrainingOutput(ModelOutput):
             (classification) loss.
         prediction_logits (`tf.Tensor` of shape `(batch_size, sequence_length, config.vocab_size)`):
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
-        cross_relationship_score: (`tf.Tensor` of shape `(batch_size, 2)`):
+        cross_relationship_score (`tf.Tensor` of shape `(batch_size, 2)`):
             Prediction scores of the textual matching objective (classification) head (scores of True/False
             continuation before SoftMax).
-        question_answering_score: (`tf.Tensor` of shape `(batch_size, n_qa_answers)`):
+        question_answering_score (`tf.Tensor` of shape `(batch_size, n_qa_answers)`):
             Prediction scores of question answering objective (classification).
         language_hidden_states (`tuple(tf.Tensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
             Tuple of `tf.Tensor` (one for input features + one for the output of each cross-modality layer) of shape
@@ -873,12 +873,12 @@ def serving(self, inputs):
             [`PreTrainedTokenizer.encode`] for details.
 
             [What are input IDs?](../glossary#input-ids)
-        visual_feats: (`tf.Tensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
+        visual_feats (`tf.Tensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
             This input represents visual features. They ROI pooled object features from bounding boxes using a
             faster-RCNN model)
 
             These are currently not provided by the transformers library.
-        visual_pos: (`tf.Tensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
+        visual_pos (`tf.Tensor` of shape `(batch_size, num_visual_features, visual_feat_dim)`):
             This input represents spacial features corresponding to their relative (via index) visual features. The
             pre-trained LXMERT model expects these spacial features to be normalized bounding boxes on a scale of 0 to
             1.
@@ -1297,7 +1297,7 @@ def call(
             Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
             config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
             loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
-        obj_labels: (`Dict[Str: Tuple[tf.Tensor, tf.Tensor]]`, *optional*, defaults to `None`):
+        obj_labels (`Dict[Str: Tuple[tf.Tensor, tf.Tensor]]`, *optional*, defaults to `None`):
             each key is named after each one of the visual losses and each element of the tuple is of the shape
             `(batch_size, num_features)` and `(batch_size, num_features, visual_feature_dim)` for each the label id and
             the label score respectively
diff --git a/src/transformers/models/mask2former/modeling_mask2former.py b/src/transformers/models/mask2former/modeling_mask2former.py
index 289b4919e369..4cb2493e58c8 100644
--- a/src/transformers/models/mask2former/modeling_mask2former.py
+++ b/src/transformers/models/mask2former/modeling_mask2former.py
@@ -1767,7 +1767,7 @@ class Mask2FormerMaskedAttentionDecoder(nn.Module):
     of the predicted mask for each query, instead of attending to the full feature map.
 
     Args:
-        config: (`Mask2FormerConfig`):
+        config (`Mask2FormerConfig`):
             Configuration used to instantiate Mask2FormerMaskedAttentionDecoder.
     """
 
@@ -2003,7 +2003,7 @@ def __init__(self, hidden_size: int, num_heads: int, mask_feature_size: torch.Te
                 The feature dimension of the Mask2FormerMaskedAttentionDecoder
             num_heads (`int`):
                 The number of heads used in the Mask2FormerMaskedAttentionDecoder
-            mask_feature_size: (`torch.Tensor`):
+            mask_feature_size (`torch.Tensor`):
                 one of the output dimensions of the predicted masks for each query
         """
         super().__init__()
diff --git a/src/transformers/models/mpnet/tokenization_mpnet.py b/src/transformers/models/mpnet/tokenization_mpnet.py
index 1f5ad2f41aae..57a06beeead0 100644
--- a/src/transformers/models/mpnet/tokenization_mpnet.py
+++ b/src/transformers/models/mpnet/tokenization_mpnet.py
@@ -119,7 +119,7 @@ class MPNetTokenizer(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).
     """
diff --git a/src/transformers/models/mpnet/tokenization_mpnet_fast.py b/src/transformers/models/mpnet/tokenization_mpnet_fast.py
index 288c69c62b3c..82d8ffec08d9 100644
--- a/src/transformers/models/mpnet/tokenization_mpnet_fast.py
+++ b/src/transformers/models/mpnet/tokenization_mpnet_fast.py
@@ -98,7 +98,7 @@ class MPNetTokenizerFast(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).
     """
diff --git a/src/transformers/models/opt/configuration_opt.py b/src/transformers/models/opt/configuration_opt.py
index df13b3201998..d2b7a4347ea4 100644
--- a/src/transformers/models/opt/configuration_opt.py
+++ b/src/transformers/models/opt/configuration_opt.py
@@ -67,7 +67,7 @@ class OPTConfig(PretrainedConfig):
             The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
         attention_dropout (`float`, *optional*, defaults to 0.0):
             The dropout ratio for the attention probabilities.
-        layerdrop: (`float`, *optional*, defaults to 0.0):
+        layerdrop (`float`, *optional*, defaults to 0.0):
             The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
             details.
         init_std (`float`, *optional*, defaults to 0.02):
diff --git a/src/transformers/models/pegasus_x/configuration_pegasus_x.py b/src/transformers/models/pegasus_x/configuration_pegasus_x.py
index c393a6b8a910..f48e19bdcbca 100644
--- a/src/transformers/models/pegasus_x/configuration_pegasus_x.py
+++ b/src/transformers/models/pegasus_x/configuration_pegasus_x.py
@@ -70,10 +70,10 @@ class PegasusXConfig(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`):
diff --git a/src/transformers/models/rag/modeling_rag.py b/src/transformers/models/rag/modeling_rag.py
index 6941bca09c74..019b26ef08e9 100644
--- a/src/transformers/models/rag/modeling_rag.py
+++ b/src/transformers/models/rag/modeling_rag.py
@@ -1430,7 +1430,7 @@ def generate(
                 priority: 1) from the `generation_config.json` model file, if it exists; 2) from the model
                 configuration. Please note that unspecified parameters will inherit [`~generation.GenerationConfig`]'s
                 default values, whose documentation should be checked to parameterize generation.
-            prefix_allowed_tokens_fn: (`Callable[[int, torch.Tensor], List[int]]`, *optional*):
+            prefix_allowed_tokens_fn (`Callable[[int, torch.Tensor], List[int]]`, *optional*):
                 If provided, this function constraints the beam search to allowed tokens only at each step. If not
                 provided no constraint is applied. This function takes 2 arguments `inputs_ids` and the batch ID
                 `batch_id`. It has to return a list with the allowed tokens for the next generation step conditioned on
diff --git a/src/transformers/models/rag/retrieval_rag.py b/src/transformers/models/rag/retrieval_rag.py
index 261255b9f62f..88cb54115bf5 100644
--- a/src/transformers/models/rag/retrieval_rag.py
+++ b/src/transformers/models/rag/retrieval_rag.py
@@ -573,10 +573,10 @@ def __call__(
         Retrieves documents for specified `question_hidden_states`.
 
         Args:
-            question_input_ids: (`List[List[int]]`) batch of input ids
+            question_input_ids (`List[List[int]]`) batch of input ids
             question_hidden_states (`np.ndarray` of shape `(batch_size, vector_size)`:
                 A batch of query vectors to retrieve with.
-            prefix: (`str`, *optional*):
+            prefix (`str`, *optional*):
                 The prefix used by the generator's tokenizer.
             n_docs (`int`, *optional*):
                 The number of docs retrieved per query.
diff --git a/src/transformers/models/realm/modeling_realm.py b/src/transformers/models/realm/modeling_realm.py
index 261b2b4a0b8c..339cc27e9289 100644
--- a/src/transformers/models/realm/modeling_realm.py
+++ b/src/transformers/models/realm/modeling_realm.py
@@ -726,7 +726,7 @@ class RealmReaderOutput(ModelOutput):
             The index of the retrieved span candidates in which the predicted answer is most likely.
         start_pos (`torch.IntTensor` of shape `()`):
             Predicted answer starting position in *RealmReader*'s inputs.
-        end_pos: (`torch.IntTensor` of shape `()`):
+        end_pos (`torch.IntTensor` of shape `()`):
             Predicted answer ending position in *RealmReader*'s inputs.
         hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
             Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer) of
diff --git a/src/transformers/models/sew/configuration_sew.py b/src/transformers/models/sew/configuration_sew.py
index af7041843de3..07e3a7df26d8 100644
--- a/src/transformers/models/sew/configuration_sew.py
+++ b/src/transformers/models/sew/configuration_sew.py
@@ -63,6 +63,9 @@ class SEWConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`SEWForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/unispeech/configuration_unispeech.py b/src/transformers/models/unispeech/configuration_unispeech.py
index f4e8df659e9a..4054c49f3e0c 100644
--- a/src/transformers/models/unispeech/configuration_unispeech.py
+++ b/src/transformers/models/unispeech/configuration_unispeech.py
@@ -65,6 +65,9 @@ class UniSpeechConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`UniSpeechForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/unispeech_sat/configuration_unispeech_sat.py b/src/transformers/models/unispeech_sat/configuration_unispeech_sat.py
index 222f982fe769..8bd482f394d0 100644
--- a/src/transformers/models/unispeech_sat/configuration_unispeech_sat.py
+++ b/src/transformers/models/unispeech_sat/configuration_unispeech_sat.py
@@ -66,6 +66,9 @@ class UniSpeechSatConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`UniSpeechSatForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/wav2vec2/configuration_wav2vec2.py b/src/transformers/models/wav2vec2/configuration_wav2vec2.py
index 7afcd3f0ee28..6f7709e535d1 100644
--- a/src/transformers/models/wav2vec2/configuration_wav2vec2.py
+++ b/src/transformers/models/wav2vec2/configuration_wav2vec2.py
@@ -63,6 +63,9 @@ class Wav2Vec2Config(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`Wav2Vec2ForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/wav2vec2_conformer/configuration_wav2vec2_conformer.py b/src/transformers/models/wav2vec2_conformer/configuration_wav2vec2_conformer.py
index b9f24c7e7080..24b7dca73944 100644
--- a/src/transformers/models/wav2vec2_conformer/configuration_wav2vec2_conformer.py
+++ b/src/transformers/models/wav2vec2_conformer/configuration_wav2vec2_conformer.py
@@ -65,6 +65,9 @@ class Wav2Vec2ConformerConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`Wav2Vec2ConformerForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/models/wavlm/configuration_wavlm.py b/src/transformers/models/wavlm/configuration_wavlm.py
index becbe100240d..c3ac6ba196a8 100644
--- a/src/transformers/models/wavlm/configuration_wavlm.py
+++ b/src/transformers/models/wavlm/configuration_wavlm.py
@@ -62,6 +62,9 @@ class WavLMConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         final_dropout (`float`, *optional*, defaults to 0.1):
             The dropout probability for the final projection layer of [`WavLMForCTC`].
+        layerdrop (`float`, *optional*, defaults to 0.1):
+            The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
+            details.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
diff --git a/src/transformers/optimization_tf.py b/src/transformers/optimization_tf.py
index b42e04041b8f..382eae2a30e2 100644
--- a/src/transformers/optimization_tf.py
+++ b/src/transformers/optimization_tf.py
@@ -117,9 +117,9 @@ def create_optimizer(
             The beta2 to use in Adam.
         adam_epsilon (`float`, *optional*, defaults to 1e-8):
             The epsilon to use in Adam.
-        adam_clipnorm: (`float`, *optional*, defaults to `None`):
+        adam_clipnorm (`float`, *optional*, defaults to `None`):
             If not `None`, clip the gradient norm for each weight tensor to this value.
-        adam_global_clipnorm: (`float`, *optional*, defaults to `None`)
+        adam_global_clipnorm (`float`, *optional*, defaults to `None`)
             If not `None`, clip gradient norm to this value. When using this argument, the norm is computed over all
             weight tensors, as if they were concatenated into a single vector.
         weight_decay_rate (`float`, *optional*, defaults to 0):
diff --git a/src/transformers/pipelines/audio_utils.py b/src/transformers/pipelines/audio_utils.py
index 62c2b00c467a..f17dd68d6439 100644
--- a/src/transformers/pipelines/audio_utils.py
+++ b/src/transformers/pipelines/audio_utils.py
@@ -119,7 +119,7 @@ def ffmpeg_microphone_live(
             The length of the striding to be used. Stride is used to provide context to a model on the (left, right) of
             an audio sample but without using that part to actually make the prediction. Setting this does not change
             the length of the chunk.
-        format_for_conversion: (`str`, defalts to `f32le`)
+        format_for_conversion (`str`, defalts to `f32le`)
             The name of the format of the audio samples to be returned by ffmpeg. The standard is `f32le`, `s16le`
             could also be used.
     Return:
diff --git a/src/transformers/pipelines/base.py b/src/transformers/pipelines/base.py
index f57d08efffd6..4fc256266449 100644
--- a/src/transformers/pipelines/base.py
+++ b/src/transformers/pipelines/base.py
@@ -514,7 +514,7 @@ def from_str(
         Creates an instance of the right subclass of [`~pipelines.PipelineDataFormat`] depending on `format`.
 
         Args:
-            format: (`str`):
+            format (`str`):
                 The format of the desired pipeline. Acceptable values are `"json"`, `"csv"` or `"pipe"`.
             output_path (`str`, *optional*):
                 Where to save the outgoing data.
diff --git a/src/transformers/tokenization_utils_base.py b/src/transformers/tokenization_utils_base.py
index a186c8338bad..ecc9d5011d78 100644
--- a/src/transformers/tokenization_utils_base.py
+++ b/src/transformers/tokenization_utils_base.py
@@ -2093,7 +2093,7 @@ def save_pretrained(
 
                 If `True`, will save the tokenizer in legacy format. If the "slow" tokenizer doesn't exits, a value
                 error is raised.
-            filename_prefix: (`str`, *optional*):
+            filename_prefix (`str`, *optional*):
                 A prefix to add to the names of the files saved by the tokenizer.
             push_to_hub (`bool`, *optional*, defaults to `False`):
                 Whether or not to push your model to the Hugging Face model hub after saving it. You can specify the
diff --git a/src/transformers/training_args_tf.py b/src/transformers/training_args_tf.py
index 847bbdb78a15..461c4086acc3 100644
--- a/src/transformers/training_args_tf.py
+++ b/src/transformers/training_args_tf.py
@@ -66,7 +66,7 @@ class TFTrainingArguments(TrainingArguments):
             The batch size per GPU/TPU core/CPU for training.
         per_device_eval_batch_size (`int`, *optional*, defaults to 8):
             The batch size per GPU/TPU core/CPU for evaluation.
-        gradient_accumulation_steps: (`int`, *optional*, defaults to 1):
+        gradient_accumulation_steps (`int`, *optional*, defaults to 1):
             Number of updates steps to accumulate the gradients for, before performing a backward/update pass.
 
             <Tip warning={true}>
diff --git a/templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/configuration_{{cookiecutter.lowercase_modelname}}.py b/templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/configuration_{{cookiecutter.lowercase_modelname}}.py
index 3221696317bd..2898b5cf6f8f 100644
--- a/templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/configuration_{{cookiecutter.lowercase_modelname}}.py
+++ b/templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/configuration_{{cookiecutter.lowercase_modelname}}.py
@@ -107,10 +107,10 @@ class {{cookiecutter.camelcase_modelname}}Config(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`):