r""" Instantiate a [`ModelCard`] from a pre-trained model model card. Parameters: pretrained_model_name_or_path: either:

Analogous to the to() method of torch Tensors Args: device: A torch device dtype:

Number of byte required for this data type Returns: Integer > 0

Wait for the background thread to finish. Args: timeout: Maximum time to wait for the thread to stop

Analogous to torch.unsqueeze. The dimension is relative to the shared dimensions of the rotation/translation. Args: dim:

Concatenates transformations along a new dimension. Args: ts: A list of T objects dim:

Will look for the words added to the trie within `text`. Output is the original string split along the boundaries of the words found.

r""" input_values (`torch.Tensor of shape `(batch_size, 1, time_steps)`): Input audio data to encode, n_quantizers (int, *

This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of thi

Save a model card object to the directory or file `save_directory_or_file`.

Called when a proxy object is has the keys() method called. This is what happens when ** is called on a proxy. This should return an iterator

Loads the dictionary from a text file with the format: ``` <symbol0> <count0> <symbol1> <count1> ... ```

This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring

Returns a copy of the Rotation whose underlying Tensor has been detached from its torch graph. Returns: A copy of the Ro

Similar to the `apply_chat_template` method on tokenizers, this method applies a Jinja template to input conversations to turn them i

This method takes a module_path as input and creates an import structure digestible by a _LazyModule. Here's an example of an output import

Converts a token string (or a sequence of tokens) in a single integer id (or a sequence of ids), using the vocabulary. Args:

Main training entry point. Args: resume_from_checkpoint (`str` or `bool`, *optional*): If a `str`, local

Converts a string to a list of tokens. If `self.legacy` is set to `False`, a prefix token is added unless the first token is special.

Pad a single encoded input or a batch of encoded inputs up to predefined length or to the max sequence length in the batch. Padding s

Updates the cache with the new `key_states` and `value_states` for the layer `layer_idx`. Parameters: key_states (`torch

Updates the cache with the new `key_states` and `value_states` for the layer `layer_idx`. Parameters: key_states (`torch

Creates a random float32 tensor

Log `logs` on the various objects watching training. Subclass and override this method to inject custom behavior. Args:

Converts the old size parameter in the config into the new dict expected in the config. This is to ensure backwards compatibility with the ol

Serializes this instance to a Python dictionary.

Prepare a table and a string for the model. Args: table (`pd.DataFrame`): Table containing tabular data.

Override original method to fix state_dict keys on load for cases when weights are loaded without using the `from_pretrained` method

Upload the {object_files} to the 🤗 Model Hub. Parameters: repo_id (`str`): The name of the repository yo

Find best matching expectation based on environment device properties. We look at device_type, major and minor versions of the driver

Returns the mean of the distribution.

Prepare a table and a list of strings for the model. <Tip warning={true}> This method is deprecated, `__call__` should be u

Simplifies an object by rounding float numbers, and downcasting tensors/numpy arrays to get simple equality test within tests.

Returns the (height, width) dimensions of the image. Args: image (`np.ndarray`): The image to get the dimensions of.

Serializes this instance to a JSON string.

Embeds the upper level music tokens and upsamples them to provide as audio conditioning.

Save the provided data object with the representation for the current [`~pipelines.PipelineDataFormat`]. Args: data (`di

All context managers that the model should be initialized under go here. Args: dtype (`torch.dtype`, *optional*):

Constructs a `ModelCard` from a Python dictionary of parameters.

Infers the channel dimension format of `image`. Args: image (`np.ndarray`): The image to infer the channel dimension of.

Passes over every char (utf-8 char) on word and recursively adds it to the internal `data` trie representation. The special key `""`

Create a standard causal mask based on the attention implementation used (stored in the config). If `past_key_values` has an hybrid cache str

Overwrites the `parse` method to incorporate a skip for CUDA tests, and remove logs and dataset prints before calling `super().parse`

Returns the sequence length of the cached states. A layer index can be optionally passed.

Reconstructs images in the original order, preserving the original structure (nested or not). The input structure is either all flat or all n

Args: tmp_dir (`string`, *optional*): if `None`: - a unique temporary path will be created

Converts `image` to the channel dimension format specified by `channel_dim`. Args: image (`numpy.ndarray`): The image to

Resizes input token embeddings matrix of the model if `new_num_tokens != config.vocab_size`. Takes care of tying weights embeddings

Groups images by shape. Returns a dictionary with the shape as key and a list of images with that shape as value, and a dictionary with t

Cover moses empty string edge case. They return empty list for '' input!

Check if `pkg_name` exist, and optionally try to get its version

Converts an individual number passed in string form to spelt-out form