qidZddlZddlmZmZmZmZmZddlm Z ddl m Z ddl m Z e jeZdd d Zd ZGd d e Zd gZeZy)zTokenization classes for Bert.N) Tokenizerdecoders normalizerspre_tokenizers processors) WordPiece)TokenizersBackend)loggingz vocab.txtztokenizer.json) vocab_filetokenizer_filectj}t|dd5}|j}dddt D]\}}|j d}|||<|S#1swY4xYw)z*Loads a vocabulary file into a dictionary.rzutf-8)encodingN ) collections OrderedDictopen readlines enumeraterstrip)r vocabreadertokensindextokens \/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/bert/tokenization_bert.py load_vocabrsw  # # %E j# 0$F!!#$!&) u T"e  L $$s A''A0ceZdZdZeZgdZeZ dde e e e fzdzde de de de d e d e d e d e dzffd Z xZS) BertTokenizera Construct a BERT tokenizer (backed by HuggingFace's tokenizers library). Based on WordPiece. This tokenizer inherits from [`TokenizersBackend`] which contains most of the main methods. Users should refer to this superclass for more information regarding those methods. Args: vocab (`str` or `dict[str, int]`, *optional*): Custom vocabulary dictionary. If not provided, vocabulary is loaded from `vocab_file`. do_lower_case (`bool`, *optional*, defaults to `True`): Whether or not to lowercase the input when tokenizing. unk_token (`str`, *optional*, defaults to `"[UNK]"`): The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead. sep_token (`str`, *optional*, defaults to `"[SEP]"`): The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for sequence classification or for a text and a question for question answering. It is also used as the last token of a sequence built with special tokens. pad_token (`str`, *optional*, defaults to `"[PAD]"`): The token used for padding, for example when batching sequences of different lengths. cls_token (`str`, *optional*, defaults to `"[CLS]"`): The classifier token which is used when doing sequence classification (classification of the whole sequence instead of per-token classification). It is the first token of the sequence when built with special tokens. mask_token (`str`, *optional*, defaults to `"[MASK]"`): The token used for masking values. This is the token used when training this model with masked language modeling. This is the token which the model will try to predict. tokenize_chinese_chars (`bool`, *optional*, defaults to `True`): Whether or not to tokenize Chinese characters. 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). ) input_idstoken_type_idsattention_maskNr do_lower_case unk_token sep_token pad_token cls_token mask_tokentokenize_chinese_chars strip_accentsc ||_||_| |_|9t|dt|dt|dt|dt|di}||_t t |jt||_tjd|| ||j_ tj|j_ tj d |j_t! |Dd|||||||| d | |j$ |j$nd} |j& |j&nd} t)j*t|j,d t|j.d t|j,d t|j.dt|j.dt|j,| ft|j.| fg|j_y)Nrr )r%T) clean_texthandle_chinese_charsr+ lowercasez##)prefix)r$r%r&r'r(r)r*r+z:0 $A:0 z:0z:0 $B:1 z:1)singlepairspecial_tokens)r$r*r+str_vocabrr _tokenizerrBertNormalizer normalizerrBertPreTokenizer pre_tokenizerrdecodersuper__init__ cls_token_id sep_token_idrTemplateProcessingr(r&post_processor)selfrr$r%r&r'r(r)r*r+kwargsrBrC __class__s rrAzBertTokenizer.__init__Os+&<#* =IIIIJ E #IdkkS^$TU%0%?%?!7'# & " )7(G(G(I%"*"4"4D"A  '!#9'  -1,=,=,It((q ,0,=,=,It((q )3)F)F$..)*(3t~~3F2GrJ'(T^^1D0EXcRVR`R`NaMbbdeT^^$l3T^^$l3* &) NTz[UNK]z[SEP]z[PAD]z[CLS]z[MASK]TN)__name__ __module__ __qualname____doc__VOCAB_FILES_NAMESvocab_files_namesmodel_input_namesrmodelr8dictintboolrA __classcell__)rHs@rr r )sB*I E.2"    "'+%)8 T#s(^#d*8 8  8  8  8 8 8 !%8 d{8 8 rIr )rMr tokenizersrrrrrtokenizers.modelsrtokenization_utils_tokenizersr utilsr get_loggerrJloggerrNrr __all__BertTokenizerFastr7rIrr^sa%SS'>   H %#.BRS^ %^ B  !rI