hezar.preprocessors.tokenizers.wordpiece module

class hezar.preprocessors.tokenizers.wordpiece.WordPieceConfig(max_length: 'int' = 'deprecated', truncation: 'str' = 'deprecated', truncation_side: str = 'right', padding: 'str' = 'deprecated', padding_side: str = 'right', stride: int = 0, pad_to_multiple_of: int = 0, pad_token_type_id: int = 0, bos_token: 'str' = None, eos_token: 'str' = None, unk_token: str = '[UNK]', sep_token: str = '[SEP]', pad_token: str = '[PAD]', cls_token: str = '[CLS]', mask_token: str = '[MASK]', additional_special_tokens: List[str] = None, wordpieces_prefix: str = '##', vocab_size: int = 30000, min_frequency: int = 2, limit_alphabet: int = 1000, initial_alphabet: list = <factory>, show_progress: bool = True)

Bases: TokenizerConfig

additional_special_tokens: List[str] = None

cls_token: str = '[CLS]'

initial_alphabet: list

limit_alphabet: int = 1000

mask_token: str = '[MASK]'

min_frequency: int = 2

name: str = 'wordpiece_tokenizer'

pad_to_multiple_of: int = 0

pad_token: str = '[PAD]'

pad_token_type_id: int = 0

padding_side: str = 'right'

sep_token: str = '[SEP]'

show_progress: bool = True

stride: int = 0

truncation_side: str = 'right'

unk_token: str = '[UNK]'

vocab_size: int = 30000

wordpieces_prefix: str = '##'

class hezar.preprocessors.tokenizers.wordpiece.WordPieceTokenizer(config, tokenizer_file=None, **kwargs)

Bases: Tokenizer

A standard WordPiece tokenizer using 🤗HuggingFace Tokenizers

Parameters:

• config – Preprocessor config for the tokenizer

• **kwargs – Extra/manual config parameters

build()

Build the tokenizer.

Returns:

The built tokenizer.

Return type:

HFTokenizer

required_backends: List[str | Backends] = [Backends.TOKENIZERS]

token_ids_name = 'token_ids'

tokenizer_config_filename = 'tokenizer_config.yaml'

tokenizer_filename = 'tokenizer.json'

train(files: List[str], **train_kwargs)

Train the model using the given files

train_from_iterator(dataset: List[str], **train_kwargs)

Train the model using the given files