U 9%e;h@sddlZddlZddlmZmZmZmZddlZddl m Z ddl m Z m Z mZmZddlmZmZmZmZerddlZddlmZerdd lmZGd d d eZGd d d e Ze edGdddeZeZdS)N)ListOptionalTupleUnion)BasicTokenizer) ExplicitEnumadd_end_docstringsis_tf_availableis_torch_available)PIPELINE_INIT_ARGSArgumentHandler ChunkPipelineDataset)/TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMES),MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMESc@s*eZdZdZeeeefdddZdS)"TokenClassificationArgumentHandlerz5 Handles arguments for token classification. inputscKs|dk r4t|ttfr4t|dkr4t|}t|}nDt|trJ|g}d}n.tdk r\t|tsht|tjrp|dfStd| d}|rt|trt|dtr|g}t||krtd||fS)Nrr zAt least one input is required.offset_mappingz;offset_mapping should have the same batch size as the input) isinstancelisttuplelenstrrtypes GeneratorType ValueErrorget)selfrkwargsZ batch_sizerr"j/var/www/html/Darija-Ai-API/env/lib/python3.8/site-packages/transformers/pipelines/token_classification.py__call__s "    z+TokenClassificationArgumentHandler.__call__N)__name__ __module__ __qualname____doc__rrrr$r"r"r"r#rsrc@s$eZdZdZdZdZdZdZdZdS)AggregationStrategyzDAll the valid aggregation strategies for TokenClassificationPipelinenonesimplefirstZaveragemaxN) r%r&r'r(NONESIMPLEFIRSTAVERAGEMAXr"r"r"r#r)3s r)a ignore_labels (`List[str]`, defaults to `["O"]`): A list of labels to ignore. grouped_entities (`bool`, *optional*, defaults to `False`): DEPRECATED, use `aggregation_strategy` instead. Whether or not to group the tokens corresponding to the same entity together in the predictions or not. stride (`int`, *optional*): If stride is provided, the pipeline is applied on all the text. The text is split into chunks of size model_max_length. Works only with fast tokenizers and `aggregation_strategy` different from `NONE`. The value of this argument defines the number of overlapping tokens between chunks. In other words, the model will shift forward by `tokenizer.model_max_length - stride` tokens each step. aggregation_strategy (`str`, *optional*, defaults to `"none"`): The strategy to fuse (or not) tokens based on the model prediction. - "none" : Will simply not do any aggregation and simply return raw results from the model - "simple" : Will attempt to group entities following the default schema. (A, B-TAG), (B, I-TAG), (C, I-TAG), (D, B-TAG2) (E, B-TAG2) will end up being [{"word": ABC, "entity": "TAG"}, {"word": "D", "entity": "TAG2"}, {"word": "E", "entity": "TAG2"}] Notice that two consecutive B tags will end up as different entities. On word based languages, we might end up splitting words undesirably : Imagine Microsoft being tagged as [{"word": "Micro", "entity": "ENTERPRISE"}, {"word": "soft", "entity": "NAME"}]. Look for FIRST, MAX, AVERAGE for ways to mitigate that and disambiguate words (on languages that support that meaning, which is basically tokens separated by a space). These mitigations will only work on real words, "New york" might still be tagged with two different entities. - "first" : (works only on word based models) Will use the `SIMPLE` strategy except that words, cannot end up with different tags. Words will simply use the tag of the first token of the word when there is ambiguity. - "average" : (works only on word based models) Will use the `SIMPLE` strategy except that words, cannot end up with different tags. scores will be averaged first across tokens, and then the maximum label is applied. - "max" : (works only on word based models) Will use the `SIMPLE` strategy except that words, cannot end up with different tags. Word entity will simply be the token with the maximum score. c sleZdZdZdZeffdd Zd'eeeeee ee e e e fee dddZ eee efd fd d Zd(d d ZddZe jdfddZddZeejejee e e e feje e edddZe ee e edddZe ee edddZe ee e edddZe eedd d!Zee eefd"d#d$Ze ee edd%d&ZZ S))TokenClassificationPipelineuv Named Entity Recognition pipeline using any `ModelForTokenClassification`. See the [named entity recognition examples](../task_summary#named-entity-recognition) for more information. Example: ```python >>> from transformers import pipeline >>> token_classifier = pipeline(model="Jean-Baptiste/camembert-ner", aggregation_strategy="simple") >>> sentence = "Je m'appelle jean-baptiste et je vis à montréal" >>> tokens = token_classifier(sentence) >>> tokens [{'entity_group': 'PER', 'score': 0.9931, 'word': 'jean-baptiste', 'start': 12, 'end': 26}, {'entity_group': 'LOC', 'score': 0.998, 'word': 'montréal', 'start': 38, 'end': 47}] >>> token = tokens[0] >>> # Start and end provide an easy way to highlight words in the original text. >>> sentence[token["start"] : token["end"]] ' jean-baptiste' >>> # Some models use the same idea to do part of speech. >>> syntaxer = pipeline(model="vblagoje/bert-english-uncased-finetuned-pos", aggregation_strategy="simple") >>> syntaxer("My name is Sarah and I live in London") [{'entity_group': 'PRON', 'score': 0.999, 'word': 'my', 'start': 0, 'end': 2}, {'entity_group': 'NOUN', 'score': 0.997, 'word': 'name', 'start': 3, 'end': 7}, {'entity_group': 'AUX', 'score': 0.994, 'word': 'is', 'start': 8, 'end': 10}, {'entity_group': 'PROPN', 'score': 0.999, 'word': 'sarah', 'start': 11, 'end': 16}, {'entity_group': 'CCONJ', 'score': 0.999, 'word': 'and', 'start': 17, 'end': 20}, {'entity_group': 'PRON', 'score': 0.999, 'word': 'i', 'start': 21, 'end': 22}, {'entity_group': 'VERB', 'score': 0.998, 'word': 'live', 'start': 23, 'end': 27}, {'entity_group': 'ADP', 'score': 0.999, 'word': 'in', 'start': 28, 'end': 30}, {'entity_group': 'PROPN', 'score': 0.999, 'word': 'london', 'start': 31, 'end': 37}] ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial) This token recognition pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"ner"` (for predicting the classes of tokens in a sequence: person, organisation, location or miscellaneous). The models that this pipeline can use are models that have been fine-tuned on a token classification task. See the up-to-date list of available models on [huggingface.co/models](https://huggingface.co/models?filter=token-classification). sequencescs<tj||||jdkr tnttdd|_||_dS)NtfF)Z do_lower_case) super__init__Zcheck_model_type frameworkrrrZ_basic_tokenizer _args_parser)r Z args_parserargsr! __class__r"r#r7s z$TokenClassificationPipeline.__init__N)grouped_entitiesignore_subwordsaggregation_strategyrstridec CsJi}|dk r||d<i}|dk s(|dk r|r8|r8tj}n|rH|sHtj}ntj}|dk rhtd|d|dk rtd|d|dk rt|trt|}|tjtj tj hkr|j j st d||d<|dk r||d<|dk r@||j jkrt d|tjkrt d |d n(|j j r8d d |d } | |d <nt d|i|fS)Nrzl`grouped_entities` is deprecated and will be removed in version v5.0.0, defaulted to `aggregation_strategy="z "` instead.zk`ignore_subwords` is deprecated and will be removed in version v5.0.0, defaulted to `aggregation_strategy="z{Slow tokenizers cannot handle subwords. Please set the `aggregation_strategy` option to `"simple"` or use a fast tokenizer.r? ignore_labelszl`stride` must be less than `tokenizer.model_max_length` (or even lower if the tokenizer adds special tokens)zI`stride` was provided to process all the text but `aggregation_strategy="z&"`, please select another one instead.T)Zreturn_overflowing_tokenspaddingr@tokenizer_paramszm`stride` was provided to process all the text but you're using a slow tokenizer. Please use a fast tokenizer.)r)r0r/r.warningswarnrrupperr2r1 tokenizeris_fastrmodel_max_length) r rAr=r>r?rr@preprocess_paramsZpostprocess_paramsrCr"r"r#_sanitize_parameterssh           z0TokenClassificationPipeline._sanitize_parametersrc s.|j|f|\}}|r||d<tj|f|S)a Classify each token of the text(s) given as inputs. Args: inputs (`str` or `List[str]`): One or several texts (or one list of texts) for token classification. Return: A list or a list of list of `dict`: Each result comes as a list of dictionaries (one for each token in the corresponding input, or each entity if this pipeline was instantiated with an aggregation_strategy) with the following keys: - **word** (`str`) -- The token/word classified. This is obtained by decoding the selected tokens. If you want to have the exact string in the original sentence, use `start` and `end`. - **score** (`float`) -- The corresponding probability for `entity`. - **entity** (`str`) -- The entity predicted for that token/word (it is named *entity_group* when *aggregation_strategy* is not `"none"`. - **index** (`int`, only present when `aggregation_strategy="none"`) -- The index of the corresponding token in the sentence. - **start** (`int`, *optional*) -- The index of the start of the corresponding entity in the sentence. Only exists if the offsets are available within the tokenizer - **end** (`int`, *optional*) -- The index of the end of the corresponding entity in the sentence. Only exists if the offsets are available within the tokenizer r)r9r6r$)r rr!Z_inputsrr;r"r#r$sz$TokenClassificationPipeline.__call__c +s|di}|jjr$|jjdkr$dnd}|j|f|j|d|jjd|}|ddt|d}t|D]v|jdkrfd d |D}nfd d |D}|dk r||d <dkr|nd|d <|dk|d<|VqjdS)NrCrTF)Zreturn_tensors truncationZreturn_special_tokens_maskZreturn_offsets_mappingZoverflow_to_sample_mapping input_idsr5cs"i|]\}}|t|dqSr)r5Z expand_dims.0kvir"r# sz:TokenClassificationPipeline.preprocess..cs i|]\}}||dqSrN)Z unsqueezerOrSr"r#rU srsentencer is_last)poprGrIr8rHrrangeitems) r rVrrJrCrLr num_chunks model_inputsr"rSr# preprocesss.     z&TokenClassificationPipeline.preprocesscCs|d}|dd}|d}|d}|jdkrF|jf|d}n&|jf|}t|trd|dn|d}|||||d|S) Nspecial_tokens_maskrrVrWr5rlogits)r_r^rrVrW)rXr8modelrdict)r r\r^rrVrWr_outputr"r"r#_forwards       z$TokenClassificationPipeline._forwardc s"dkrdgg}|D]}|dd}|dd}|dd}|ddk rZ|ddnd} |dd} tj|dd d } t|| } | | jdd d } |jd kr|}| dk r| nd} |||| | | |}|||}fd d |D}||qt |}|dkr| |}|S)NOr_rrVrMrr^T)axisZkeepdimsr5cs0g|](}|ddkr|ddkr|qS)entityN entity_group)rrPrgrAr"r# Dsz;TokenClassificationPipeline.postprocess..r ) numpynpr-expsumr8gather_pre_entities aggregateextendraggregate_overlapping_entities)r Z all_outputsr?rAZ all_entitiesZ model_outputsr_rVrMrr^ZmaxesZ shifted_expscores pre_entitiesr=entitiesr[r"rjr# postprocess*sB        z'TokenClassificationPipeline.postprocesscCst|dkr|St|ddd}g}|d}|D]}|d|dkrT|dkrnnL|d|d}|d|d}||kr|}q||kr|d|dkr|}q0|||}q0|||S)NrcSs|dS)Nstartr")xr"r"r#SzLTokenClassificationPipeline.aggregate_overlapping_entities..keyrxendscore)rsortedappend)r rvZaggregated_entitiesZprevious_entityrgZcurrent_lengthZprevious_lengthr"r"r#rsPs" $  z:TokenClassificationPipeline.aggregate_overlapping_entities)rVrMrtrr^r?returncCs:g}t|D]&\}} ||r q |jt||} |dk r ||\} } t| tsn|jdkrn| } | } || | } t|jddrt|jjj ddrt | t | k}n@|t j t j t jhkrtdt| dkod|| d| dk}t|||jjkr| } d }n d} d} d }| | | | ||d }||q |S) zTFuse various numpy arrays into dicts with all the information needed for aggregationNpt _tokenizerZcontinuing_subword_prefixz?Tokenizer does not support real words, using fallback heuristicr r F)wordrtrxr~index is_subword) enumeraterGZconvert_ids_to_tokensintrr8itemgetattrrr`rr)r0r1r2rDrE UserWarningZ unk_token_idr)r rVrMrtrr^r?ruidxZ token_scoresrZ start_indZend_indZword_refr pre_entityr"r"r#rpdsV        z/TokenClassificationPipeline.gather_pre_entities)rur?rcCs|tjtjhkrng}|D]R}|d}|d|}|jjj|||d|d|d|dd}||qn |||}|tjkr|S| |S)Nrtrrrxr~)rgrrrrxr~) r)r.r/argmaxr`configid2labelraggregate_wordsgroup_entities)r rur?rvr entity_idxrrgr"r"r#rqs"     z%TokenClassificationPipeline.aggregate)rvr?rc Cs|jdd|D}|tjkrL|dd}|}||}|jjj|}n|tjkrt |ddd}|d}|}||}|jjj|}nT|tj krt dd|D}t j |dd } | } |jjj| }| | }ntd ||||dd |d d d} | S)NcSsg|] }|dqSrr"rir"r"r#rksz>TokenClassificationPipeline.aggregate_word..rrtcSs |dS)Nrt)r-)rgr"r"r#rzr{z.r|cSsg|] }|dqS)rtr"rir"r"r#rks)rfzInvalid aggregation_strategyrxrer~)rgrrrxr~)rGconvert_tokens_to_stringr)r0rr`rrr2r-r1rmstacknanmeanr) r rvr?rrtrrrgZ max_entityZaverage_scoresrZ new_entityr"r"r#aggregate_words4       z*TokenClassificationPipeline.aggregate_wordcCs|tjtjhkrtdg}d}|D]@}|dkr8|g}q$|drL||q$|||||g}q$|dk r|||||S)z Override tokens from a given word that disagree to force agreement on word boundaries. Example: micro|soft| com|pany| B-ENT I-NAME I-ENT I-ENT will be rewritten with first strategy as microsoft| company| B-ENT I-ENT z;NONE and SIMPLE strategies are invalid for word aggregationNr)r)r.r/rrr)r rvr?Z word_entitiesZ word_grouprgr"r"r#rs" z+TokenClassificationPipeline.aggregate_words)rvrcCsj|dddd}tdd|D}dd|D}|t||j||dd|dd d }|S) z Group together the adjacent tokens with the same entity predicted. Args: entities (`dict`): The entities predicted by the pipeline. rrg-recSsg|] }|dqS)rr"rir"r"r#rkszBTokenClassificationPipeline.group_sub_entities..cSsg|] }|dqSrr"rir"r"r#rksrxr~)rhrrrxr~)splitrmrZmeanrGr)r rvrgrttokensrhr"r"r#group_sub_entitiess   z.TokenClassificationPipeline.group_sub_entities) entity_namercCsH|drd}|dd}n$|dr8d}|dd}nd}|}||fS)NzB-BrzI-I) startswith)r rbitagr"r"r#get_tags  z#TokenClassificationPipeline.get_tagc Csg}g}|D]n}|s ||q ||d\}}||dd\}}||krd|dkrd||q ||||g}q |r||||S)z Find and group together the adjacent tokens with the same entity predicted. Args: entities (`dict`): The entities predicted by the pipeline. rgrer)rrr) r rvZ entity_groupsZentity_group_disaggrgrrZlast_biZlast_tagr"r"r#rs  z*TokenClassificationPipeline.group_entities)NNNNNN)N)!r%r&r'r(Zdefault_input_namesrr7rboolr)rrrrKrrr$r]rcr.rwrsrmZndarrayrarprqrrrrr __classcell__r"r"r;r#r3=sF$$  I & >r3)rrDtypingrrrrrlrmZmodels.bert.tokenization_bertrutilsrr r r baser rrrZ tensorflowr5Zmodels.auto.modeling_tf_autorZmodels.auto.modeling_autorrr)r3Z NerPipeliner"r"r"r#s,    #^