Returns the base model.

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

Get the index (e.g. 2) from the prediction (e.g. 'C')

Get the index (e.g. 2) from the prediction (e.g. 'C')

Given a task returns its corresponding adapter layer. Args: task: Input task name. Returns: Adapter layer corr

Check if Llama is available in the transformers library (it's not in earlier versions).

Usually, layers that produce the same feature map spatial size are defined as one "stage". Under such definition, stride_per_

Predict the answers to questions in a data split. :param eval_tuple: The data tuple to be evaluated. :param dump: The path o

Predict the answers to questions in a data split. :param eval_tuple: The data tuple to be evaluated. :param dump: The path o

Predict the answers to questions in a data split. :param eval_tuple: The data tuple to be evaluated. :param dump: The path of

Predict the answers to questions in a data split. :param eval_tuple: The data tuple to be evaluated. :param dump: The path of

r""" This method wrapps the entire protocol for preparing a model before running a training. This includes: 1- Cast the layernorm in fp32

Set the state dict of the Peft model. Args: model ([`PeftModel`]): The Peft model. peft_model_state_dict (`dict`): The state

Split a full s3 path into the bucket name and path.

Temporarily set the random seed. This works for python numpy, pytorch.

Gets back the base model by removing all the lora modules without merging. This gives back the original base model.

Initialize object. Args: model_type: The transformer model type. This is used to retrieve the right method to