Fix docstrings with last version of hf-doc-builder styler (#18581)

* Fix docstrings with last version of hf-doc-builder styler

* Remove empty Parameter block

src/transformers/benchmark/benchmark_utils.py

@@ -79,7 +79,6 @@ def separate_process_wrapper_fn(func: Callable[[], None], do_multi_processing: b
     measurements it is important that the function is executed in a separate process
 
     Args:
-
         - `func`: (`callable`): function() -> ... generic function which will be executed in its own separate process
         - `do_multi_processing`: (`bool`) Whether to run function on separate process or not
     """
@@ -210,7 +209,6 @@ def measure_peak_memory_cpu(function: Callable[[], None], interval=0.5, device_i
     https://github.com/pythonprofilers/memory_profiler/blob/895c4ac7a08020d66ae001e24067da6dcea42451/memory_profiler.py#L239
 
     Args:
-
         - `function`: (`callable`): function() -> ... function without any arguments to measure for which to measure
           the peak memory
 
@@ -228,7 +226,6 @@ def measure_peak_memory_cpu(function: Callable[[], None], interval=0.5, device_i
         measures current cpu memory usage of a given `process_id`
 
         Args:
-
             - `process_id`: (`int`) process_id for which to measure memory
 
         Returns
@@ -336,7 +333,6 @@ def start_memory_tracing(
     https://psutil.readthedocs.io/en/latest/#psutil.Process.memory_info
 
     Args:
-
         - `modules_to_trace`: (None, string, list/tuple of string) if None, all events are recorded if string or list
           of strings: only events from the listed module/sub-module will be recorded (e.g. 'fairseq' or
           'transformers.models.gpt2.modeling_gpt2')
@@ -483,7 +479,6 @@ def stop_memory_tracing(
     Stop memory tracing cleanly and return a summary of the memory trace if a trace is given.
 
     Args:
-
         `memory_trace` (optional output of start_memory_tracing, default: None):
             memory trace to convert in summary
         `ignore_released_memory` (boolean, default: None):

src/transformers/generation_flax_utils.py

@@ -208,7 +208,6 @@ class FlaxGenerationMixin:
         post](https://huggingface.co/blog/how-to-generate).
 
         Parameters:
-
             input_ids (`jnp.ndarray` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             max_length (`int`, *optional*, defaults to `model.config.max_length`):

src/transformers/generation_tf_utils.py

@@ -418,7 +418,6 @@ class TFGenerationMixin:
         post](https://huggingface.co/blog/how-to-generate).
 
         Parameters:
-
             input_ids (`tf.Tensor` of shape `(batch_size, sequence_length)`, `(batch_size, sequence_length,
             feature_dim)` or `(batch_size, num_channels, height, width)`, *optional*):
                 The sequence used as a prompt for the generation or as model inputs to the encoder. If `None` the
@@ -1336,7 +1335,6 @@ class TFGenerationMixin:
         post](https://huggingface.co/blog/how-to-generate).
 
         Parameters:
-
             input_ids (`tf.Tensor` of `dtype=tf.int32` and shape `(batch_size, sequence_length)`, *optional*):
                 The sequence used as a prompt for the generation. If `None` the method initializes it with
                 `bos_token_id` and a batch size of 1.
@@ -2070,7 +2068,6 @@ class TFGenerationMixin:
         Generates sequences for models with a language modeling head using greedy decoding.
 
         Parameters:
-
             input_ids (`tf.Tensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             logits_processor (`TFLogitsProcessorList`, *optional*):
@@ -2323,7 +2320,6 @@ class TFGenerationMixin:
         Generates sequences for models with a language modeling head using multinomial sampling.
 
         Parameters:
-
             input_ids (`tf.Tensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             logits_processor (`TFLogitsProcessorList`, *optional*):
@@ -2600,7 +2596,6 @@ class TFGenerationMixin:
         Generates sequences for models with a language modeling head using beam search with multinomial sampling.
 
         Parameters:
-
             input_ids (`tf.Tensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             max_length (`int`, *optional*, defaults to 20):

src/transformers/generation_utils.py

@@ -1555,7 +1555,6 @@ class GenerationMixin:
         used for text-decoder, text-to-text, speech-to-text, and vision-to-text models.
 
         Parameters:
-
             input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             logits_processor (`LogitsProcessorList`, *optional*):
@@ -1789,7 +1788,6 @@ class GenerationMixin:
         can be used for text-decoder, text-to-text, speech-to-text, and vision-to-text models.
 
         Parameters:
-
             input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             logits_processor (`LogitsProcessorList`, *optional*):
@@ -2046,7 +2044,6 @@ class GenerationMixin:
         can be used for text-decoder, text-to-text, speech-to-text, and vision-to-text models.
 
         Parameters:
-
             input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             beam_scorer (`BeamScorer`):
@@ -2355,7 +2352,6 @@ class GenerationMixin:
         sampling** and can be used for text-decoder, text-to-text, speech-to-text, and vision-to-text models.
 
         Parameters:
-
             input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             beam_scorer (`BeamScorer`):
@@ -2672,7 +2668,6 @@ class GenerationMixin:
         decoding** and can be used for text-decoder, text-to-text, speech-to-text, and vision-to-text models.
 
         Parameters:
-
             input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
                 The sequence used as a prompt for the generation.
             beam_scorer (`BeamScorer`):

src/transformers/modelcard.py

@@ -80,8 +80,6 @@ class ModelCard:
     Inioluwa Deborah Raji and Timnit Gebru for the proposal behind model cards. Link: https://arxiv.org/abs/1810.03993
 
     Note: A model card can be loaded and saved to disk.
-
-    Parameters:
     """
 
     def __init__(self, **kwargs):

src/transformers/models/auto/auto_factory.py

@@ -563,7 +563,6 @@ class _LazyAutoMapping(OrderedDict):
     " A mapping config to object (model or tokenizer for instance) that will load keys and values when it is accessed.
 
     Args:
-
         - config_mapping: The map model type to config class
         - model_mapping: The map model type to model (or tokenizer) class
     """

src/transformers/models/flaubert/tokenization_flaubert.py

@@ -130,7 +130,6 @@ class FlaubertTokenizer(XLMTokenizer):
             - Install with `pip install sacremoses`
 
         Args:
-
             - bypass_tokenizer: Allow users to preprocess and tokenize the sentences externally (default = False)
               (bool). If True, we only apply BPE.
 

src/transformers/models/fsmt/tokenization_fsmt.py

@@ -354,7 +354,6 @@ class FSMTTokenizer(PreTrainedTokenizer):
             - Install with `pip install sacremoses`
 
         Args:
-
             - lang: ISO language code (default = 'en') (string). Languages should belong of the model supported
               languages. However, we don't enforce it.
             - bypass_tokenizer: Allow users to preprocess and tokenize the sentences externally (default = False)

src/transformers/models/perceiver/modeling_perceiver.py

@@ -1960,7 +1960,6 @@ def build_position_encoding(
     Builds the position encoding.
 
     Args:
-
     - out_channels: refers to the number of channels of the position encodings.
     - project_pos_dim: if specified, will project the position encodings to this dimension.
 

src/transformers/models/tapex/tokenization_tapex.py

@@ -1398,7 +1398,6 @@ class TapexTokenizer(PreTrainedTokenizer):
     ):
         """
         Args:
-
         table_content:
             {"header": xxx, "rows": xxx, "id" (Optionally): xxx}
 

src/transformers/models/transfo_xl/modeling_transfo_xl.py

@@ -523,7 +523,6 @@ class TransfoXLPreTrainedModel(PreTrainedModel):
         weights embeddings afterwards if the model class has a *tie_weights()* method.
 
         Arguments:
-
             new_num_tokens: (*optional*) int:
                 New number of tokens in the embedding matrix. Increasing the size will add newly initialized vectors at
                 the end. Reducing the size will remove vectors from the end. If not provided or None: does nothing and

src/transformers/models/xlm/tokenization_xlm.py

@@ -791,7 +791,6 @@ class XLMTokenizer(PreTrainedTokenizer):
         externally, and set `bypass_tokenizer=True` to bypass the tokenizer.
 
         Args:
-
             - lang: ISO language code (default = 'en') (string). Languages should belong of the model supported
               languages. However, we don't enforce it.
             - bypass_tokenizer: Allow users to preprocess and tokenize the sentences externally (default = False)

src/transformers/testing_utils.py

@@ -1286,7 +1286,6 @@ def pytest_terminal_summary_main(tr, id):
     there.
 
     Args:
-
     - tr: `terminalreporter` passed from `conftest.py`
     - id: unique id like `tests` or `examples` that will be incorporated into the final reports filenames - this is
       needed as some jobs have multiple runs of pytest, so we can't have them overwrite each other.

src/transformers/trainer_pt_utils.py

@@ -377,7 +377,6 @@ class DistributedTensorGatherer:
     For some reason, that's not going to roll their boat. This class is there to solve that problem.
 
     Args:
-
         world_size (`int`):
             The number of processes used in the distributed training.
         num_samples (`int`):

src/transformers/trainer_utils.py

@@ -337,7 +337,6 @@ def speed_metrics(split, start_time, num_samples=None, num_steps=None):
     should be run immediately after the operation to be measured has completed.
 
     Args:
-
     - split: name to prefix metric (like train, eval, test...)
     - start_time: operation start time
     - num_samples: number of samples processed

src/transformers/utils/notebook.py

@@ -120,7 +120,6 @@ class NotebookProgressBar:
         The main method to update the progress bar to `value`.
 
         Args:
-
             value (`int`):
                 The value to use. Must be between 0 and `total`.
             force_update (`bool`, *optional*, defaults to `False`):
@@ -204,7 +203,6 @@ class NotebookTrainingTracker(NotebookProgressBar):
     An object tracking the updates of an ongoing training with progress bars and a nice table reporting metrics.
 
     Args:
-
         num_steps (`int`): The number of steps during training. column_names (`List[str]`, *optional*):
             The list of column names for the metrics table (will be inferred from the first call to
             [`~utils.notebook.NotebookTrainingTracker.write_line`] if not set).