o i @sUdZddlmZddlmZmZddlmZee Z dZ dZ dZ dZd Zd Zd Ze d ed edede dediZeeefed<ddeDZeeefed<GdddeZdgZdS)z Tokenization classes for CANINE.)Optional) AddedTokenPreTrainedTokenizer)loggingiiiiiiz[CLS]z[SEP]z[BOS]z[MASK]z[PAD]z [RESERVED]SPECIAL_CODEPOINTScCi|]\}}||qSr ).0 codepointnamer r k/home/ubuntu/veenaModal/venv/lib/python3.10/site-packages/transformers/models/canine/tokenization_canine.py 7rSPECIAL_CODEPOINTS_BY_NAMEc seZdZdZeeeeeeeeeeeeddffdd Z e de fddZ d d Z d edeefd d Zdede fddZde defddZddZ d"dee deee dee fddZ d#dee deee dedee ffdd Zd"dedeefd d!ZZS)$CanineTokenizera Construct a CANINE tokenizer (i.e. a character splitter). It turns text into a sequence of characters, and then converts each character into its Unicode code point. [`CanineTokenizer`] inherits from [`PreTrainedTokenizer`]. Refer to superclass [`PreTrainedTokenizer`] for usage examples and documentation concerning parameters. Args: model_max_length (`int`, *optional*, defaults to 2048): The maximum sentence length the model accepts. Fic st|tr t|dddn|}t|trt|dddn|}t|tr(t|dddn|}t|tr6t|dddn|}t|trDt|dddn|}t|trRt|dddn|}i|_tD] \} } | |j| <q[dd|jD|_t|_t |j|_ t j d||||||||d| dS)NF)lstriprstripTcSrr r )r r r r r r rcs z,CanineTokenizer.__init__..) bos_token eos_token sep_token cls_token pad_token mask_tokenadd_prefix_spacemodel_max_lengthr ) isinstancestrr_special_codepointsritems_special_codepoint_stringsUNICODE_VOCAB_SIZE_unicode_vocab_sizelen_num_special_tokenssuper__init__) selfrrrrrrrrkwargsr r  __class__r r r&Hs4    zCanineTokenizer.__init__returncCs|jSN)r")r'r r r vocab_sizevszCanineTokenizer.vocab_sizecCs$ddt|jD}||j|S)NcSsi|]}t||qSr )chr)r ir r r r{rz-CanineTokenizer.get_vocab..)ranger-updateadded_tokens_encoder)r'vocabr r r get_vocabzs zCanineTokenizer.get_vocabtextcCst|S)z5Tokenize a string (i.e. perform character splitting).)list)r'r5r r r _tokenizeszCanineTokenizer._tokenizetokencCs*zt|WStytd|dw)zaConverts a token (i.e. a Unicode character) in an id (i.e. its integer Unicode code point value).zinvalid token: '')ord TypeError ValueError)r'r8r r r _convert_token_to_ids   z$CanineTokenizer._convert_token_to_idindexcCs:z|tvr t|WSt|WStytd|w)z Converts a Unicode code point (integer) in a token (str). In case it's a special code point, convert to human-readable format. z invalid id: )rr.r;r<)r'r>r r r _convert_id_to_tokens   z$CanineTokenizer._convert_id_to_tokencCs d|S)N)join)r'tokensr r r convert_tokens_to_strings z(CanineTokenizer.convert_tokens_to_stringN token_ids_0 token_ids_1cCs4|jg}|jg}|||}|dur|||7}|S)a Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and adding special tokens. A CANINE 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) sep_token_id cls_token_id)r'rDrEsepclsresultr r r build_inputs_with_special_tokenss   z0CanineTokenizer.build_inputs_with_special_tokensalready_has_special_tokenscsT|r tj||ddSdgdgt|dg}|dur(|dgt|dg7}|S)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)rDrErLrN)r%get_special_tokens_maskr#)r'rDrErLrJr)r r rNsz'CanineTokenizer.get_special_tokens_masksave_directoryfilename_prefixcCsdS)Nr r )r'rOrPr r r save_vocabularyszCanineTokenizer.save_vocabularyr,)NF)__name__ __module__ __qualname____doc__r.CLSSEPPADMASKr&propertyintr-r4rr6r7r=r?rCrrKboolrNrQ __classcell__r r r)r r:sJ.     rN)rUtypingrtokenization_utilsrrutilsr get_loggerrRloggerr!rXrVrWBOSrYRESERVEDrdictr[r__annotations__rrr__all__r r r r s.    "