a  d<$ @sUdZddlmZmZmZddlmZmZddlm Z e e Z ddiZ dZdZd Zd Zd Zd Zd ZedededededediZeeefed<ddeDZeeefed<GdddeZdS)z Tokenization classes for CANINE.)DictListOptional) AddedTokenPreTrainedTokenizer)loggingznielsr/canine-siiiiiiz[CLS]z[SEP]z[BOS]z[MASK]z[PAD]z [RESERVED]SPECIAL_CODEPOINTScCsi|]\}}||qSr ).0 codepointnamer r w/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/transformers/models/canine/tokenization_canine.py <rSPECIAL_CODEPOINTS_BY_NAMEcs$eZdZdZeZeeeeeeeeee ee ddffdd Z e e dddZeeed d d Zee d d dZe edddZddZd ee eee ee dddZd!ee eee eee dfdd Zd"ee eee ee dddZd#eeedddZZS)$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. Fr c st|trt|dddn|}t|tr4t|dddn|}t|trPt|dddn|}t|trlt|dddn|}t|trt|dddn|}t|trt|dddn|}tjf||||||||d| i|_tD]\} } | |j| <qdd|jD|_t |_ t |j|_ dS)NF)lstriprstripT) bos_token eos_token sep_token cls_token pad_token mask_tokenadd_prefix_spacemodel_max_lengthcSsi|]\}}||qSr r )r rr r r rrvsz,CanineTokenizer.__init__..) isinstancestrrsuper__init__Z_special_codepointsr itemsZ_special_codepoint_stringsUNICODE_VOCAB_SIZE_unicode_vocab_sizelenZ_num_special_tokens) selfrrrrrrrrkwargsr r __class__r rr!Os4    zCanineTokenizer.__init__)returncCs|jS)N)r$)r&r r r vocab_size}szCanineTokenizer.vocab_size)textr*cCst|S)z5Tokenize a string (i.e. perform character splitting).)list)r&r,r r r _tokenizeszCanineTokenizer._tokenize)tokenr*cCs2z t|WSty,td|dYn0dS)zaConverts a token (i.e. a Unicode character) in an id (i.e. its integer Unicode code point value).zinvalid token: ''N)ord TypeError ValueError)r&r/r r r_convert_token_to_ids  z$CanineTokenizer._convert_token_to_id)indexr*cCsBz|tvrt|WSt|WSty<td|Yn0dS)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: N)r chrr2r3)r&r5r r r_convert_id_to_tokens    z$CanineTokenizer._convert_id_to_tokencCs d|S)N)join)r&tokensr r rconvert_tokens_to_stringsz(CanineTokenizer.convert_tokens_to_stringN) token_ids_0 token_ids_1r*cCs4|jg}|jg}|||}|dur0|||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_idr&r<r=sepclsresultr r r build_inputs_with_special_tokenss   z0CanineTokenizer.build_inputs_with_special_tokens)r<r=already_has_special_tokensr*csT|rtj||ddSdgdgt|dg}|durP|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)r<r=rErN)r get_special_tokens_maskr%)r&r<r=rErCr(r rrGsz'CanineTokenizer.get_special_tokens_maskcCsH|jg}|jg}t|||dg}|durD|t||dg7}|S)a Create a mask from the two sequences passed to be used in a sequence-pair classification task. A CANINE 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). rNrF)r>r?r%r@r r r$create_token_type_ids_from_sequencess z4CanineTokenizer.create_token_type_ids_from_sequences)save_directoryfilename_prefixcCsdS)Nr r )r&rIrJr r rsave_vocabularyszCanineTokenizer.save_vocabulary)N)NF)N)N)__name__ __module__ __qualname____doc__&PRETRAINED_POSITIONAL_EMBEDDINGS_SIZESZmax_model_input_sizesr6CLSSEPPADMASKr!propertyintr+rrr.r4r7r;rrDboolrGrHrK __classcell__r r r(rr?sB .   rN)rOtypingrrrZtokenization_utilsrrutilsrZ get_loggerrLloggerrPr#rSrQrRZBOSrTZRESERVEDr rVr__annotations__r"rrr r r rs,   "