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>

examples/research_projects/bert-loses-patience/pabee/modeling_pabee_albert.py

@@ -253,7 +253,7 @@ class AlbertForSequenceClassificationWithPabee(AlbertPreTrainedModel):
 
         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).

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

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.
     """
 

src/transformers/modeling_utils.py

@@ -912,7 +912,7 @@ class ModuleUtilsMixin:
                 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:

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.

src/transformers/models/blip/modeling_blip_text.py

@@ -613,7 +613,7 @@ class BlipTextModel(BlipTextPreTrainedModel):
                 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:

src/transformers/models/blip/modeling_tf_blip_text.py

@@ -633,7 +633,7 @@ class TFBlipTextModel(TFBlipTextPreTrainedModel):
                 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:

src/transformers/models/blip_2/modeling_blip_2.py

@@ -1059,7 +1059,7 @@ class Blip2QFormerModel(Blip2PreTrainedModel):
                 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:

src/transformers/models/bloom/modeling_bloom.py

@@ -256,7 +256,7 @@ class BloomAttention(nn.Module):
         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]

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):

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`):

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`):

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.

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):

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 @@ LXMERT_INPUTS_DOCSTRING = r"""
             [`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 @@ class LxmertForPreTraining(LxmertPreTrainedModel):
             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 @@ class LxmertForQuestionAnswering(LxmertPreTrainedModel):
         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

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 @@ LXMERT_INPUTS_DOCSTRING = r"""
             [`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 @@ class TFLxmertForPreTraining(TFLxmertPreTrainedModel):
             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

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 @@ class Mask2FormerMaskPredictor(nn.Module):
                 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__()

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).
     """

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).
     """

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):