U 4AfP@sdZddlZddlZddlZddlmZmZmZddlm Z m Z m Z m Z ddl mZeeZddiZd d Zd d ZGd dde ZGdddZGdddZdS)zTokenization classes for Bert.N)ListOptionalTuple)PreTrainedTokenizer _is_control_is_punctuation_is_whitespace)logging vocab_filez vocab.txtc CsRt}t|ddd}|}W5QRXt|D]\}}|d}|||<q2|S)z*Loads a vocabulary file into a dictionary.rutf-8encoding ) collections OrderedDictopen readlines enumeraterstrip)r vocabreadertokensindextokenrN/tmp/pip-unpacked-wheel-zw5xktn0/transformers/models/bert/tokenization_bert.py load_vocabs  rcCs|}|sgS|}|S)z@Runs basic whitespace cleaning and splitting on a piece of text.)stripsplit)textrrrrwhitespace_tokenize*s r"c seZdZdZeZd%fd d Zed d Zed dZ ddZ d&ddZ ddZ ddZ ddZd'eeeeeeedddZd(eeeeeeeedfdd Zd)eeeeeeedd d!Zd*eeeeed"d#d$ZZS)+ BertTokenizera Construct a BERT tokenizer. Based on WordPiece. 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`): File containing the vocabulary. do_lower_case (`bool`, *optional*, defaults to `True`): Whether or not to lowercase the input when tokenizing. do_basic_tokenize (`bool`, *optional*, defaults to `True`): Whether or not to do basic tokenization before WordPiece. never_split (`Iterable`, *optional*): Collection of tokens which will never be split during tokenization. Only has an effect when `do_basic_tokenize=True` 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. 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). TN[UNK][SEP][PAD][CLS][MASK]c stj|std|dt||_tdd|jD|_ ||_ |r^t ||| | d|_ t |jt|d|_tjf|||||||| | | d | dS)Nz&Can't find a vocabulary file at path 'z'. To load the vocabulary from a Google pretrained model use `tokenizer = BertTokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`cSsg|]\}}||fqSrr).0tokidsrrr tsz*BertTokenizer.__init__..) do_lower_case never_splittokenize_chinese_chars strip_accents)r unk_token) r-do_basic_tokenizer.r1 sep_token pad_token cls_token mask_tokenr/r0)ospathisfile ValueErrorrrrritems ids_to_tokensr2BasicTokenizerbasic_tokenizerWordpieceTokenizerstrwordpiece_tokenizersuper__init__) selfr r-r2r.r1r3r4r5r6r/r0kwargs __class__rrrC_s:    zBertTokenizer.__init__cCs|jjSN)r>r-rDrrrr-szBertTokenizer.do_lower_casecCs t|jSrH)lenrrIrrr vocab_sizeszBertTokenizer.vocab_sizecCst|jf|jSrH)dictrZadded_tokens_encoderrIrrr get_vocabszBertTokenizer.get_vocabFcCsdg}|jrT|jj||s|jnddD],}||jjkr@||q$||j|7}q$n |j|}|S)N)r.)r2r>tokenizeZall_special_tokensr.appendrA)rDr!Zsplit_special_tokens split_tokensrrrr _tokenizes    zBertTokenizer._tokenizecCs|j||j|jS)z0Converts a token (str) in an id using the vocab.)rgetr1)rDrrrr_convert_token_to_idsz"BertTokenizer._convert_token_to_idcCs|j||jS)z=Converts an index (integer) in a token (str) using the vocab.)r<rRr1)rDrrrr_convert_id_to_tokensz"BertTokenizer._convert_id_to_tokencCsd|dd}|S)z:Converts a sequence of tokens (string) in a single string. z ##)joinreplacer)rDrZ out_stringrrrconvert_tokens_to_stringsz&BertTokenizer.convert_tokens_to_string) token_ids_0 token_ids_1returncCs@|dkr|jg||jgS|jg}|jg}|||||S)a Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. A BERT sequence has the following format: - single sequence: `[CLS] X [SEP]` - pair of sequences: `[CLS] A [SEP] B [SEP]` Args: token_ids_0 (`List[int]`): List of IDs to which the special tokens will be added. token_ids_1 (`List[int]`, *optional*): Optional second list of IDs for sequence pairs. Returns: `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens. N) cls_token_id sep_token_id)rDrZr[clsseprrr build_inputs_with_special_tokenss z.BertTokenizer.build_inputs_with_special_tokens)rZr[already_has_special_tokensr\csf|rtj||ddS|dk rLdgdgt|dgdgt|dgSdgdgt|dgS)a Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding special tokens using the tokenizer `prepare_for_model` method. Args: token_ids_0 (`List[int]`): List of IDs. token_ids_1 (`List[int]`, *optional*): Optional second list of IDs for sequence pairs. already_has_special_tokens (`bool`, *optional*, defaults to `False`): Whether or not the token list is already formatted with special tokens for the model. Returns: `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token. T)rZr[rbNr)rBget_special_tokens_maskrJ)rDrZr[rbrFrrrds.z%BertTokenizer.get_special_tokens_maskcCsV|jg}|jg}|dkr.t|||dgSt|||dgt||dgS)a Create a mask from the two sequences passed to be used in a sequence-pair classification task. A BERT sequence pair mask has the following format: ``` 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 | first sequence | second sequence | ``` If `token_ids_1` is `None`, this method only returns the first portion of the mask (0s). Args: token_ids_0 (`List[int]`): List of IDs. token_ids_1 (`List[int]`, *optional*): Optional second list of IDs for sequence pairs. Returns: `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s). Nrrc)r^r]rJ)rDrZr[r`r_rrr$create_token_type_ids_from_sequencess z2BertTokenizer.create_token_type_ids_from_sequences)save_directoryfilename_prefixr\c Csd}tj|r4tj||r$|dndtd}n|r@|dnd|}t|dddZ}t|jdd d D]<\}}||krt d |d |}| |d |d7}qnW5QRX|fS)Nr-rVr wr rcSs|dS)Nrcr)kvrrrz/BertTokenizer.save_vocabulary..)keyzSaving vocabulary to z\: vocabulary indices are not consecutive. Please check that the vocabulary is not corrupted!rrc) r7r8isdirrWVOCAB_FILES_NAMESrsortedrr;loggerwarningwrite)rDrfrgrr writerrZ token_indexrrrsave_vocabularys"  zBertTokenizer.save_vocabulary) TTNr$r%r&r'r(TN)F)N)NF)N)N)__name__ __module__ __qualname____doc__roZvocab_files_namesrCpropertyr-rKrMrQrSrTrYrintrraboolrdrer@rru __classcell__rrrFrr#3sV)/        r#c@sNeZdZdZdddZdddZdd Zdd d Zd d ZddZ ddZ dS)r=a 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|dkr g}||_t||_||_||_||_dSrH)r-setr.r/r0do_split_on_punc)rDr-r.r/r0rrrrrC3s zBasicTokenizer.__init__cCs|r|jt|n|j}||}|jr4||}td|}t|}g}|D]R}||kr|j r| }|j dk r| |}n|j r| |}| |||qPtd|}|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. NFCFrU)r.unionr~ _clean_textr/_tokenize_chinese_chars unicodedata normalizer"r-lowerr0_run_strip_accentsextend_run_split_on_puncrW)rDr!r.Zunicode_normalized_textZ orig_tokensrPr output_tokensrrrrNCs$       zBasicTokenizer.tokenizecCsBtd|}g}|D]"}t|}|dkr,q||qd|S)z$Strips accents from a piece of text.ZNFDZMnrV)rrcategoryrOrW)rDr!outputcharcatrrrris   z!BasicTokenizer._run_strip_accentscCs|jr|dk r||kr|gSt|}d}d}g}|t|kr||}t|r^||gd}n |rl|gd}|d||d7}q0dd|DS) z&Splits punctuation on a piece of text.NrTFrccSsg|]}d|qS)rV)rW)r)xrrrr,sz5BasicTokenizer._run_split_on_punc..)rlistrJrrO)rDr!r.charsiZstart_new_wordrrrrrrts"    z!BasicTokenizer._run_split_on_punccCsTg}|D]@}t|}||r>|d|||dq||qd|S)z)Adds whitespace around any CJK character.rUrV)ord_is_chinese_charrOrWrDr!rrcprrrrs     z&BasicTokenizer._tokenize_chinese_charscCs|dkr|dks|dkr |dks|dkr0|dks|dkr@|dks|d krP|d ks|d kr`|d ks|d krp|dks|dkr|dkrdSdS)z6Checks whether CP is the codepoint of a CJK character.iNii4iMiiߦii?i@ii iiiiiTFr)rDrrrrrsD  zBasicTokenizer._is_chinese_charcCsXg}|D]D}t|}|dks|dkst|r.qt|rB|dq||qd|S)zBPerforms invalid character removal and whitespace cleanup on text.rirUrV)rrr rOrWrrrrrs  zBasicTokenizer._clean_text)TNTNT)N)N) rvrwrxryrCrNrrrrrrrrrr=s  &  r=c@s"eZdZdZdddZddZdS) r?zRuns WordPiece tokenization.dcCs||_||_||_dSrH)rr1max_input_chars_per_word)rDrr1rrrrrCszWordpieceTokenizer.__init__c Csg}t|D]}t|}t||jkr4||jq d}d}g}|t|krt|}d} ||krd|||} |dkrd| } | |jkr| } q|d8}qX| dkrd}q|| |}q@|r||jq ||q |S)a Tokenizes a piece of text into its word pieces. This uses a greedy longest-match-first algorithm to perform tokenization using the given vocabulary. For example, `input = "unaffable"` wil return as output `["un", "##aff", "##able"]`. Args: text: A single token or whitespace separated tokens. This should have already been passed through *BasicTokenizer*. Returns: A list of wordpiece tokens. FrNrVz##rcT) r"rrJrrOr1rWrr) rDr!rrrZis_badstartZ sub_tokensendZ cur_substrsubstrrrrrNs:       zWordpieceTokenizer.tokenizeN)r)rvrwrxryrCrNrrrrr?s r?)ryrr7rtypingrrrZtokenization_utilsrrrr utilsr Z get_loggerrvrqrorr"r#r=r?rrrrs    j"