sg0ddlZddlmZmZddlZddlmZddlm Z m Z ddl m Z m Z mZe jeZGdd e Ze ed Gd d e Zy)N)ListUnion)TruncationStrategy)add_end_docstringslogging)ArgumentHandler ChunkPipelinebuild_pipeline_init_argsceZdZdZdZdZy)%ZeroShotClassificationArgumentHandlerz Handles arguments for zero-shot for text classification by turning each possible label into an NLI premise/hypothesis pair. ct|tr=|jdDcgc]#}|js|j%}}|Scc}w)N,) isinstancestrsplitstrip)selflabelslabels b/var/www/html/venv/lib/python3.12/site-packages/transformers/pipelines/zero_shot_classification.py _parse_labelsz3ZeroShotClassificationArgumentHandler._parse_labelssA fc "17c1BTekkmekkmTFT Us AAc \t|dk(st|dk(r td|j|d|k(rtdj|t|tr|g}g}|D]2}|j |Dcgc]}||j|gc}4||fScc}w)Nrz>You must include at least one label and at least one sequence.zThe provided hypothesis_template "{}" was not able to be formatted with the target labels. Make sure the passed template includes formatting syntax such as {{}} where the label should go.)len ValueErrorformatrrextend)r sequencesrhypothesis_templatesequence_pairssequencers r__call__z.ZeroShotClassificationArgumentHandler.__call__s v;! s9~2]^ ^  % %fQi 04G Gw&,-   i %" I! gH  ! !^d"eUZH.A.H.H.O#P"e f gy((#fsB) N)__name__ __module__ __qualname____doc__rr#rrrs  )r)rT) has_tokenizerceZdZdZeffd ZedZddejfdZ dZ de e ee fffd Zd d Zd Zd d ZxZS)ZeroShotClassificationPipelinea NLI-based zero-shot classification pipeline using a `ModelForSequenceClassification` trained on NLI (natural language inference) tasks. Equivalent of `text-classification` pipelines, but these models don't require a hardcoded number of potential classes, they can be chosen at runtime. It usually means it's slower but it is **much** more flexible. Any combination of sequences and labels can be passed and each combination will be posed as a premise/hypothesis pair and passed to the pretrained model. Then, the logit for *entailment* is taken as the logit for the candidate label being valid. Any NLI model can be used, but the id of the *entailment* label must be included in the model config's :attr:*~transformers.PretrainedConfig.label2id*. Example: ```python >>> from transformers import pipeline >>> oracle = pipeline(model="facebook/bart-large-mnli") >>> oracle( ... "I have a problem with my iphone that needs to be resolved asap!!", ... candidate_labels=["urgent", "not urgent", "phone", "tablet", "computer"], ... ) {'sequence': 'I have a problem with my iphone that needs to be resolved asap!!', 'labels': ['urgent', 'phone', 'computer', 'not urgent', 'tablet'], 'scores': [0.504, 0.479, 0.013, 0.003, 0.002]} >>> oracle( ... "I have a problem with my iphone that needs to be resolved asap!!", ... candidate_labels=["english", "german"], ... ) {'sequence': 'I have a problem with my iphone that needs to be resolved asap!!', 'labels': ['english', 'german'], 'scores': [0.814, 0.186]} ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial) This NLI pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"zero-shot-classification"`. The models that this pipeline can use are models that have been fine-tuned on an NLI task. See the up-to-date list of available models on [huggingface.co/models](https://huggingface.co/models?search=nli). c|||_t||i||jdk(rtj dyy)NzFailed to determine 'entailment' label id from the label2id mapping in the model config. Setting to -1. Define a descriptive label2id mapping in the model config to ensure correct outputs.) _args_parsersuper__init__ entailment_idloggerwarning)r args_parserargskwargs __class__s rr1z'ZeroShotClassificationPipeline.__init__WsB' $)&)    # NNk  $r)c|jjjjD](\}}|j j ds&|cSy)Nentailr.)modelconfiglabel2iditemslower startswith)rrinds rr2z,ZeroShotClassificationPipeline.entailment_id`sL**++44::< JE3{{}''1  r)Tc |j}|jj:tj d|jj |j_ |j|||||}|S#t $r?}dt|vr%|j||||tj}n|Yd}~|Sd}~wwxYw)ze Parse arguments and tokenize only_first so that hypothesis (label) is not truncated NzfTokenizer was not supporting padding necessary for zero-shot, attempting to use `pad_token=eos_token`)add_special_tokensreturn_tensorspadding truncationz too short) framework tokenizer pad_tokenr3error eos_token ExceptionrrDO_NOT_TRUNCATE) rr!rErCrFr7rDinputses r_parse_and_tokenizez2ZeroShotClassificationPipeline._parse_and_tokenizegs  >> # # + LL) (,~~'?'?DNN $ ^^#5-% $F2 % c!f$"'9#1#1AA ( % sA66 B>?4B99B>c |jdd|d|d<tjdi}d|vr!|jj |d|d<d|vr|d|d<i}d|vr|d|d<|i|fS)N multi_class multi_labelzThe `multi_class` argument has been deprecated and renamed to `multi_label`. `multi_class` will be removed in a future version of Transformers.candidate_labelsr )getr3r4r/r)rr7preprocess_paramspostprocess_paramss r_sanitize_parametersz3ZeroShotClassificationPipeline._sanitize_parameterss ::mT * 6$*=$9F= ! NNU   '484E4E4S4STZ[mTn4o 0 1 F *7=>S7T 3 4 F "06}0E } - "&888r)rct|dk(rn)t|dk(r d|vr |d|d<ntd|t| |fi|S)a Classify the sequence(s) given as inputs. See the [`ZeroShotClassificationPipeline`] documentation for more information. Args: sequences (`str` or `List[str]`): The sequence(s) to classify, will be truncated if the model input is too large. candidate_labels (`str` or `List[str]`): The set of possible class labels to classify each sequence into. Can be a single label, a string of comma-separated labels, or a list of labels. hypothesis_template (`str`, *optional*, defaults to `"This example is {}."`): The template used to turn each label into an NLI-style hypothesis. This template must include a {} or similar syntax for the candidate label to be inserted into the template. For example, the default template is `"This example is {}."` With the candidate label `"sports"`, this would be fed into the model like `" sequence to classify This example is sports . "`. The default template works well in many cases, but it may be worthwhile to experiment with different templates depending on the task setting. multi_label (`bool`, *optional*, defaults to `False`): Whether or not multiple candidate labels can be true. If `False`, the scores are normalized such that the sum of the label likelihoods for each sequence is 1. If `True`, the labels are considered independent and probabilities are normalized for each candidate by doing a softmax of the entailment score vs. the contradiction score. Return: A `dict` or a list of `dict`: Each result comes as a dictionary with the following keys: - **sequence** (`str`) -- The sequence for which this is the output. - **labels** (`List[str]`) -- The labels sorted by order of likelihood. - **scores** (`List[float]`) -- The probabilities for each of the labels. rr rTz%Unable to understand extra arguments )rrr0r#)rrr6r7r8s rr#z'ZeroShotClassificationPipeline.__call__s^H t9>  Y!^ 2& @)-aF% &DTFKL Lw 4V44r)c#K|j|||\}}tt||D]6\}\}}|j|g} ||d|t |dz k(d| 8yw)Nrr candidate_labelr"is_last)r/ enumerateziprPr) rrNrTr r!rir\ sequence_pair model_inputs r preprocessz)ZeroShotClassificationPipeline.preprocesss$($5$5f>NPc$d! 3>T !^^Ta$b&VH%5%;%;%=%C%C%E$bcF^^M$Z&VH%5%;%;%=$Z[F LLO  !Q !>>=!R*@A #./14 ..M%2a%7rQ "239I=8Y3Y"Z VV/0266:M3N3R3RSU`d3R3eeFF^F-S$2D2D-DEMVVM*RVVM-B-F-FrTX-F-YYF!2!2!456!! 4<=q'*=Q[)002  1UF$b$Z*>s G- G2(G7 G< H)NzThis example is {}.)F)r$r%r&r'rr1propertyr2r ONLY_FIRSTrPrXrrrr#rcrsr __classcell__)r8s@rr,r,.sp%N$I#J'+tPbPmPm(T9$+5d3i(+5Z $ r)r,)rjtypingrrr{rxtokenization_utilsrutilsrrbaser r r get_loggerr$r3rr,r(r)rrse3/JJ   H %)O)@,4@A] ]] B] r)