o ZhS;@sdZddlZddlZddlZddlZddlmZmZddlm Z m Z m Z m Z ddl mZeeZddd Zd d ZGd d d ZddZddZGddde ZdgZdS)z$Tokenization classes for OpenAI GPT.N)OptionalTuple)PreTrainedTokenizer _is_control_is_punctuation_is_whitespace)loggingz vocab.jsonz merges.txt) vocab_file merges_filecCs|}|sgS|}|S)z@Runs basic whitespace cleaning and splitting on a piece of text.)stripsplit)texttokensr]/var/www/auris/lib/python3.10/site-packages/transformers/models/openai/tokenization_openai.pywhitespace_tokenize$s rc@sXeZdZdZ     dddZdddZdd Zdd d Zd d ZddZ ddZ dS)BasicTokenizera Constructs a BasicTokenizer that will run basic tokenization (punctuation splitting, lower casing, etc.). Args: do_lower_case (`bool`, *optional*, defaults to `True`): Whether or not to lowercase the input when tokenizing. never_split (`Iterable`, *optional*): Collection of tokens which will never be split during tokenization. Only has an effect when `do_basic_tokenize=True` tokenize_chinese_chars (`bool`, *optional*, defaults to `True`): Whether or not to tokenize Chinese characters. This should likely be deactivated for Japanese (see this [issue](https://github.com/huggingface/transformers/issues/328)). 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). do_split_on_punc (`bool`, *optional*, defaults to `True`): In some instances we want to skip the basic punctuation splitting so that later tokenization can capture the full context of the words, such as contractions. TNcCs2|durg}||_t||_||_||_||_dSN) do_lower_caseset never_splittokenize_chinese_chars strip_accentsdo_split_on_punc)selfrrrrrrrr__init__Es  zBasicTokenizer.__init__cCs|r |jt|n|j}||}|jr||}td|}t|}g}|D])}||vrH|j r@| }|j dur?| |}n|j rH| |}| |||q(td|}|S)aj Basic Tokenization of a piece of text. For sub-word tokenization, see WordPieceTokenizer. Args: never_split (`List[str]`, *optional*) Kept for backward compatibility purposes. Now implemented directly at the base class level (see [`PreTrainedTokenizer.tokenize`]) List of token not to split. NFCF )runionr _clean_textr_tokenize_chinese_chars unicodedata normalizerrlowerr_run_strip_accentsextend_run_split_on_puncjoin)rrrZunicode_normalized_textZ orig_tokens split_tokenstokenZ output_tokensrrrtokenizeUs&      zBasicTokenizer.tokenizecCsBtd|}g}|D]}t|}|dkrq ||q d|S)z$Strips accents from a piece of text.ZNFDZMn)r"r#categoryappendr()rroutputcharcatrrrr%{s    z!BasicTokenizer._run_strip_accentscCs|jr |dur||vr|gSt|}d}d}g}|t|krI||}t|r/||gd}n|r6|gd}|d||d7}|t|ksdd|DS) z&Splits punctuation on a piece of text.NrTFcSsg|]}d|qS)r,)r().0xrrr z5BasicTokenizer._run_split_on_punc..)rlistlenrr.)rrrcharsiZstart_new_wordr/r0rrrr's$     z!BasicTokenizer._run_split_on_punccCsTg}|D] }t|}||r|d|||dq||qd|S)z)Adds whitespace around any CJK character.rr,)ord_is_chinese_charr.r(rrr/r0cprrrr!s      z&BasicTokenizer._tokenize_chinese_charscCs|dkr|dks@|dkr|dks@|dkr|dks@|dkr |dks@|d kr(|d ks@|d kr0|d ks@|d kr8|dks@|dkrB|dkrBdSdS)z6Checks whether CP is the codepoint of a CJK character.iNii4iMiiߦii?i@ii iiiiiTFr)rr?rrrr=s zBasicTokenizer._is_chinese_charcCsXg}|D]"}t|}|dks|dkst|rqt|r!|dq||qd|S)zBPerforms invalid character removal and whitespace cleanup on text.rirr,)r<rrr.r(r>rrrr s   zBasicTokenizer._clean_text)TNTNTr) __name__ __module__ __qualname____doc__rr+r%r'r!r=r rrrrr.s &  rcCs6t}|d}|ddD] }|||f|}q |S)z Return set of symbol pairs in a word. word is represented as tuple of symbols (symbols being variable-length strings) rr3N)radd)wordpairsZ prev_charr0rrr get_pairss rGcCsn|dd}|dd}|dd}|dd}|dd}td d |}td d |}td d|}|S)zm fixes some issues the spacy tokenizer had on books corpus also does some whitespace standardization u—-u–u―u…z...´'zD(-+|~+|!+|"+|;+|\?+|\++|,+|\)+|\(+|\\+|\/+|\*+|\[+|\]+|}+|{+|\|+|_+)z \1 z\s*\n\s*z z[^\S\n]+r)replaceresubr )rrrrtext_standardizes     rNcseZdZdZeZddgZdfdd ZeddZ ed d Z d d Z d dZ ddZ ddZddZddZddedeedeefddZZS)OpenAIGPTTokenizera( Construct a GPT Tokenizer. Based on Byte-Pair-Encoding with the following peculiarities: - lowercases all inputs, - uses `SpaCy` tokenizer and `ftfy` for pre-BPE tokenization if they are installed, fallback to BERT's `BasicTokenizer` if not. This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to this superclass for more information regarding those methods. Args: vocab_file (`str`): Path to the vocabulary file. merges_file (`str`): Path to the merges file. unk_token (`str`, *optional*, defaults to `""`): 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. Z input_idsZattention_maskc  s4zddl}ddlm}|}|j|_|j|_Wnty.tdt dd|_d|_Ynwt |dd}t ||_ Wdn1sFwYdd |j D|_t |dd} | d d d } Wdn1srwYd d| D} tt| tt| |_i|_tjdd|i|dS)Nr)EnglishzQftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.T)rutf-8encodingcSsi|]\}}||qSrr)r4kvrrr r7z/OpenAIGPTTokenizer.__init__.. r3r2cSsg|]}t|qSr)tupler )r4mergerrrr6sz/OpenAIGPTTokenizer.__init__.. unk_tokenr)ftfyZ spacy.lang.enrQZ tokenizernlpfix_text ImportErrorloggerwarningropenjsonloadencoderitemsdecoderreadr dictzipranger9 bpe_rankscachesuperr) rr r r[kwargsr\rQZ_nlpZ vocab_handleZ merges_handleZmerges __class__rrrs,      zOpenAIGPTTokenizer.__init__cCsdS)NTrrrrrrsz OpenAIGPTTokenizer.do_lower_casecCs t|jSr)r9rerrrrr vocab_sizes zOpenAIGPTTokenizer.vocab_sizecCst|jfi|jSr)rireZadded_tokens_encoderrrrrr get_vocab#szOpenAIGPTTokenizer.get_vocabc s~t|dd|ddf}|jvrj|St|}|s#|dS t|fddd}|jvr4ny|\}}g}d}|t|krz|||} Wnty\|||dYn?w|||| | }|||kr|t|dkr||d|kr| |||d 7}n | |||d7}|t|ksBt|}|}t|dkrnt|}q$d |}|d krd }|j|<|S) Nr2Tcsj|tdS)Ninf)rlgetfloat)pairrrrr0sz(OpenAIGPTTokenizer.bpe..keyrr3rz z ) rYrmrGminrlr9index ValueErrorr&r.r() rr*rErFZbigramfirstsecondZnew_wordr;jrrrrbpe&sN     ,     zOpenAIGPTTokenizer.bpecCsg}|jdur!|j|}|D]}|t||dq|S|t||}|D]}|t||j dq-|S)zTokenize a string.Nr) r^r]r+r&r8rr rNrr$)rrr)r*rrr _tokenizeRs  "zOpenAIGPTTokenizer._tokenizecCs|j||j|jS)z0Converts a token (str) in an id using the vocab.)rerwr[)rr*rrr_convert_token_to_idasz'OpenAIGPTTokenizer._convert_token_to_idcCs|j||jS)z0Converts an id in a token (BPE) using the vocab.)rgrwr[)rrrrr_convert_id_to_tokenesz'OpenAIGPTTokenizer._convert_id_to_tokencCsd|dd}|S)z:Converts a sequence of tokens (string) in a single string.r,rur)r(rKr )rrZ out_stringrrrconvert_tokens_to_stringisz+OpenAIGPTTokenizer.convert_tokens_to_stringNsave_directoryfilename_prefixreturnc CsVtj|std|ddStj||r|dndtd}tj||r,|dndtd}t|ddd }|t j |j d d d d dWdn1sTwYd}t|ddd =}|dt |j dddD]!\}} || krtd|d| }|d|d|d7}qsWd||fS1swY||fS)NzVocabulary path (z) should be a directoryrHr,r r wrRrSr}TF)indent sort_keys ensure_asciirXrz#version: 0.2 cSs|dS)Nr3r)kvrrrrzsz4OpenAIGPTTokenizer.save_vocabulary..r{zSaving vocabulary to zZ: BPE merge indices are not consecutive. Please check that the tokenizer is not corrupted!rr3)ospathisdirr`errorr(VOCAB_FILES_NAMESrbwritercdumpsresortedrlrfra) rrrr Z merge_filefrwriterZ bpe_tokensZ token_indexrrrsave_vocabularyns8        z"OpenAIGPTTokenizer.save_vocabulary)rPr)r@rArBrCrZvocab_files_namesZmodel_input_namesrpropertyrrsrtrrrrrstrrrr __classcell__rrrprrOs   ,(rO)rCrcrrLr"typingrrZtokenization_utilsrrrrutilsr Z get_loggerr@r`rrrrGrNrO__all__rrrrs(   "  "