S^(h]?zdZddlZddlmZmZmZddlmZddlm Z m Z m Z m Z ddl mZd d lmZmZmZe j&eZd d d ZGddeZGddeZej2dgdZej2dgdZdZe eGddZe eGddeeZgdZy)zTokenization classes for DPR.N)ListOptionalUnion) BatchEncoding) TensorTypeadd_end_docstringsadd_start_docstringslogging)BertTokenizerFast)DPRContextEncoderTokenizerDPRQuestionEncoderTokenizerDPRReaderTokenizerz vocab.txtztokenizer.json) vocab_filetokenizer_fileceZdZdZeZeZy)DPRContextEncoderTokenizerFastal Construct a "fast" DPRContextEncoder tokenizer (backed by HuggingFace's *tokenizers* library). [`DPRContextEncoderTokenizerFast`] is identical to [`BertTokenizerFast`] and runs end-to-end tokenization: punctuation splitting and wordpiece. Refer to superclass [`BertTokenizerFast`] for usage examples and documentation concerning parameters. N)__name__ __module__ __qualname____doc__VOCAB_FILES_NAMESvocab_files_namesrslow_tokenizer_classk/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/models/dpr/tokenization_dpr_fast.pyrrs*5rrceZdZdZeZeZy)DPRQuestionEncoderTokenizerFastao Constructs a "fast" DPRQuestionEncoder tokenizer (backed by HuggingFace's *tokenizers* library). [`DPRQuestionEncoderTokenizerFast`] is identical to [`BertTokenizerFast`] and runs end-to-end tokenization: punctuation splitting and wordpiece. Refer to superclass [`BertTokenizerFast`] for usage examples and documentation concerning parameters. N)rrrrrrrrrrrr!r!-s*6rr!DPRSpanPrediction span_scorerelevance_scoredoc_id start_index end_indextextDPRReaderOutput) start_logits end_logitsrelevance_logitsa Return a dictionary with the token ids of the input strings and other information to give to `.decode_best_spans`. It converts the strings of a question and different passages (title and text) in a sequence of IDs (integers), using the tokenizer and vocabulary. The resulting `input_ids` is a matrix of size `(n_passages, sequence_length)` with the format: [CLS] [SEP] [SEP] Args: questions (`str` or `List[str]`): The questions to be encoded. You can specify one question for many passages. In this case, the question will be duplicated like `[questions] * n_passages`. Otherwise you have to specify as many questions as in `titles` or `texts`. titles (`str` or `List[str]`): The passages titles to be encoded. This can be a string or a list of strings if there are several passages. texts (`str` or `List[str]`): The passages texts to be encoded. This can be a string or a list of strings if there are several passages. padding (`bool`, `str` or [`~utils.PaddingStrategy`], *optional*, defaults to `False`): Activates and controls padding. Accepts the following values: - `True` or `'longest'`: Pad to the longest sequence in the batch (or no padding if only a single sequence if provided). - `'max_length'`: Pad to a maximum length specified with the argument `max_length` or to the maximum acceptable input length for the model if that argument is not provided. - `False` or `'do_not_pad'` (default): No padding (i.e., can output a batch with sequences of different lengths). truncation (`bool`, `str` or [`~tokenization_utils_base.TruncationStrategy`], *optional*, defaults to `False`): Activates and controls truncation. Accepts the following values: - `True` or `'longest_first'`: Truncate to a maximum length specified with the argument `max_length` or to the maximum acceptable input length for the model if that argument is not provided. This will truncate token by token, removing a token from the longest sequence in the pair if a pair of sequences (or a batch of pairs) is provided. - `'only_first'`: Truncate to a maximum length specified with the argument `max_length` or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the first sequence of a pair if a pair of sequences (or a batch of pairs) is provided. - `'only_second'`: Truncate to a maximum length specified with the argument `max_length` or to the maximum acceptable input length for the model if that argument is not provided. This will only truncate the second sequence of a pair if a pair of sequences (or a batch of pairs) is provided. - `False` or `'do_not_truncate'` (default): No truncation (i.e., can output batch with sequence lengths greater than the model maximum admissible input size). max_length (`int`, *optional*): Controls the maximum length to use by one of the truncation/padding parameters. If left unset or set to `None`, this will use the predefined model maximum length if a maximum length is required by one of the truncation/padding parameters. If the model has no specific maximum input length (like XLNet) truncation/padding to a maximum length will be deactivated. return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors instead of list of python integers. Acceptable values are: - `'tf'`: Return TensorFlow `tf.constant` objects. - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return Numpy `np.ndarray` objects. return_attention_mask (`bool`, *optional*): Whether or not to return the attention mask. If not set, will return the attention mask according to the specific tokenizer's default, defined by the `return_outputs` attribute. [What are attention masks?](../glossary#attention-mask) Return: `Dict[str, List[List[int]]]`: A dictionary with the following keys: - `input_ids`: List of token ids to be fed to a model. - `attention_mask`: List of indices specifying which tokens should be attended to by the model. ceZdZ ddeedeedeeefdeeefdeedeeeefdeede ffd Z dd e d e d ed edede e f dZde ede ed edede e f dZxZS)CustomDPRReaderTokenizerMixintitlestextspadding truncation max_lengthreturn_tensorsreturn_attention_maskreturnc ||t||f|||||d| S||||n|} t||| f|||||d| St|ts|n|g}t|ts|n|g}t |} t|ts|n|g| z}t |t |k(s Jdt |dt |dt|||ddd} t||dddd} dt | | Dcgc]\}}| |r||zd|n||zc}}i}|durFg}|dD]7}|j |Dcgc]}t||jk7c}9||d <|j|||| Scc}}wcc}w) N)r2r3r4r5r6z2There should be as many titles than texts but got z titles and z texts.F)r2r3 input_ids)add_special_tokensr2r3attention_mask)r2r4r5) super__call__ isinstancestrlenzipappendint pad_token_idpad)self questionsr0r1r2r3r4r5r6kwargs text_pair n_passagesencoded_question_and_titles encoded_textsencoded_question_and_title encoded_textencoded_inputsr;r9input_id __class__s rr=z&CustomDPRReaderTokenizerMixin.__call__s) >em7#%%-&;  ^u}"'-UI7#  %%-&;    *&#6VH's3%[ %/ 3%?Ii[S]E] 6{c%j( @V \Z]^cZdYeel m (',g&6y&RWdi&6&jkv&w#(5RWdi(jkvw ADD_an@o =. )j,l:KZH/,>?  ! -N+K8 f %%Zc&dhs8t7H7H+H'I&de f/=N+ ,xxJ_mxnn'es =E<:F reader_input reader_output num_spansmax_answer_lengthnum_spans_per_passagecv|d}|dd\}}} t| } tt| d| j} g} | D]} t || }|j |j ddz}|d|jk(r|j |j}n t|}|j|| |||| |||| }|D]V\}}||z }||z }| jt|| ||| |z| | | |||j|||dz Xt| |k\sn| d|S) a` Get the span predictions for the extractive Q&A model. Returns: *List* of *DPRReaderOutput* sorted by descending *(relevance_score, span_score)*. Each *DPRReaderOutput* is a *Tuple* with: - **span_score**: `float` that corresponds to the score given by the reader for this span compared to other spans in the same passage. It corresponds to the sum of the start and end logits of the span. - **relevance_score**: `float` that corresponds to the score of the each passage to answer the question, compared to all the other passages. It corresponds to the output of the QA classifier of the DPRReader. - **doc_id**: `int` the id of the passage. - ***start_index**: `int` the start index of the span (inclusive). - **end_index**: `int` the end index of the span (inclusive). Examples: ```python >>> from transformers import DPRReader, DPRReaderTokenizer >>> tokenizer = DPRReaderTokenizer.from_pretrained("facebook/dpr-reader-single-nq-base") >>> model = DPRReader.from_pretrained("facebook/dpr-reader-single-nq-base") >>> encoded_inputs = tokenizer( ... questions=["What is love ?"], ... titles=["Haddaway"], ... texts=["'What Is Love' is a song recorded by the artist Haddaway"], ... return_tensors="pt", ... ) >>> outputs = model(**encoded_inputs) >>> predicted_spans = tokenizer.decode_best_spans(encoded_inputs, outputs) >>> print(predicted_spans[0].text) # best span a song ```r9NrT)reversekeyr r)r+r,rU top_spansr#) r@sortedrange __getitem__listindex sep_token_idrD_get_best_spansrBr"decode)rFrRrSrTrUrVr9r+r,r-rJ sorted_docsnbest_spans_predictionsr& sequence_idspassage_offset sequence_len best_spansr'r(s rdecode_best_spansz/CustomDPRReaderTokenizerMixin.decode_best_spanssN!- 5B2A5F2 j"2)* U:.BRB^B^_ 9;! F & 12L)//0A0A1EINB4#4#44+11$2C2CD "<0 --)&1.N%f-n\J"3/ .J +5 & Y~- ^+ '..%#/#7 #DzRXGYZcGd#d(8(@%$/"+![[kIPQM)RS   *+y8; <'z 22rr+r,r[c g}t|D]9\ }t| |zD] \}}|j |zf||zf";t|dd}g} |D]o\\ } ksJd d d z dz} | |ks Jd| d |t fd | DrL| j ft | |k(sn| S| S) z Finds the best answer span for the extractive Q&A model for one passage. It returns the best span by descending `span_score` order and keeping max `top_spans` spans. Spans longer that `max_answer_length` are ignored. c |dS)Nrr)xs rz?CustomDPRReaderTokenizerMixin._get_best_spans..s adrT)rYrXzWrong span indices: [:]rzSpan is too long: z > c3K|]9\}}|cxkxr |cxkxrkncxs|cxkxr cxkxr|knc;yw)Nr).0prev_start_indexprev_end_indexr(r's r z@CustomDPRReaderTokenizerMixin._get_best_spans.."sN7%~/N>NYNR#{QiQ>QRs?A) enumeraterBr\anyr@)rFr+r,rUr[scores start_score answer_length end_scorechosen_span_intervalsscorelengthr(r's @@rrbz-CustomDPRReaderTokenizerMixin._get_best_spans sF(1,(? e $K,5j{]nOn6o,p e( y [=-HI;YbKbcd e eNDA "/5  + $[)e)+ _/D[MQRS\R]]^-_ _+,q0F.. c2DVHCPaOb0c c.;P  ! ( (+y)A B()Y6$$ %$r)NNFFNNN)@)rrrrr?rboolrCrrr=r*rr"rjrb __classcell__)rQs@rr/r/sD !%#$)',$(;?048o 8o} 8o tSy! 8o $)$ 8oSM8o!sJ!788o (~8o 8o|!#%& J3#J3'J3 J3  J3 # J3  J3X%3i%I% %  %  %rr/c eZdZdZeZddgZeZy)DPRReaderTokenizerFasta Constructs a "fast" DPRReader tokenizer (backed by HuggingFace's *tokenizers* library). [`DPRReaderTokenizerFast`] is almost identical to [`BertTokenizerFast`] and runs end-to-end tokenization: punctuation splitting and wordpiece. The difference is that is has three inputs strings: question, titles and texts that are combined to be fed to the [`DPRReader`] model. Refer to superclass [`BertTokenizerFast`] for usage examples and documentation concerning parameters. r9r;N) rrrrrrmodel_input_namesrrrrrrr/s! *$&67-rr)rr!r) r collectionstypingrrrtokenization_utils_baserutilsrr r r bert.tokenization_bert_fastr tokenization_dprrrr get_loggerrloggerrrr! namedtupler"r*CUSTOM_DPR_READER_DOCSTRINGr/r__all__rrrrs$((4RR;ii   H %#.BRS 6%6 6 7&7 7+K**h)+(():