`i0ddlZddlmZddlZddlmZddlmZm Z ddl m Z m Z m Z e jeZGdd e Zee d Gd d e ZdS)N)Union)TruncationStrategy)add_end_docstringslogging)ArgumentHandler ChunkPipelinebuild_pipeline_init_argsceZdZdZdZdZdS)%ZeroShotClassificationArgumentHandlerz Handles arguments for zero-shot for text classification by turning each possible label into an NLI premise/hypothesis pair. cnt|trd|dD}|S)Nc^g|]*}||+S)strip).0labels /home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/transformers/pipelines/zero_shot_classification.py zGZeroShotClassificationArgumentHandler._parse_labels..s-TTTekkmmTekkmmTTT,) isinstancestrsplit)selflabelss r _parse_labelsz3ZeroShotClassificationArgumentHandler._parse_labelss9 fc " " UTTc1B1BTTTF rc^t|dkst|dkrtd|dkrtddt|tr|g}g}|D]$|fd|D%||fS)Nrz>You must include at least one label and at least one sequence.z"The provided hypothesis_template "z" 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.c>g|]}|gSr)format)rrhypothesis_templatesequences rrzBZeroShotClassificationArgumentHandler.__call__..'s."e"e"eUZH.A.H.H.O.O#P"e"e"er)len ValueErrorr rrextend)r sequencesrr!sequence_pairsr"s ` @r__call__z.ZeroShotClassificationArgumentHandler.__call__s v;;!  s9~~22]^^ ^  % %fQi 0 04G G Gq5Hqqq  i % % $" I! g gH  ! !"e"e"e"e"e^d"e"e"e f f f fy((rN)__name__ __module__ __qualname____doc__rr(rrrr r s<  )))))rr T) has_tokenizerceZdZdZdZdZdZdZeffd Z e dZ dde j fdZdZdeeeefffd Zdd Zd ZddZxZS)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). FTc ||_tjdi||jdkrtddSdS)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.r) _args_parsersuper__init__ entailment_idloggerwarning)r args_parserkwargs __class__s rr4z'ZeroShotClassificationPipeline.__init__Zsd'""6"""   # # NNk      $ #rc|jjjD]0\}}|dr|cS1dS)Nentailr1)modelconfiglabel2iditemslower startswith)rrinds rr5z,ZeroShotClassificationPipeline.entailment_idcsZ*+4::<<  JE3{{}}''11   rrc b|j}|jj0td|jj|j_ ||||||}nO#t $rB}dt|vr%|||||tj }n|Yd}~nd}~wwxYw|S)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_tokenr6error eos_token ExceptionrrDO_NOT_TRUNCATE) rr'rGrErHr9rFinputses r_parse_and_tokenizez2ZeroShotClassificationPipeline._parse_and_tokenizejs  > # + LL)   (,~'?DN $ ^^#5-% $FF   c!ff$$"'9#1#1A ( $ sA B,*8B''B,c |d%|d|d<tdi}d|vr#|j|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!)getr6r7r2r)rr9preprocess_paramspostprocess_paramss r_sanitize_parametersz3ZeroShotClassificationPipeline._sanitize_parameterss ::m $ $ 0$*=$9F= ! NNU     ' '484E4S4STZ[mTn4o4o 0 1 F * *7=>S7T 3 4 F " "06}0E } - "&888rr&ct|dkrn5t|dkrd|vr |d|d<ntd|tj|fi|S)a Classify the sequence(s) given as inputs. See the [`ZeroShotClassificationPipeline`] documentation for more information. Args: sequences (`str` or `list[str]`): The sequence(s) to classify, will be truncated if the model input is too large. candidate_labels (`str` or `list[str]`): The set of possible class labels to classify each sequence into. Can be a single label, a string of comma-separated labels, or a list of labels. hypothesis_template (`str`, *optional*, defaults to `"This example is {}."`): The template used to turn each label into an NLI-style hypothesis. This template must include a {} or similar syntax for the candidate label to be inserted into the template. For example, the default template is `"This example is {}."` With the candidate label `"sports"`, this would be fed into the model like `" sequence to classify This example is sports . "`. The default template works well in many cases, but it may be worthwhile to experiment with different templates depending on the task setting. multi_label (`bool`, *optional*, defaults to `False`): Whether or not multiple candidate labels can be true. If `False`, the scores are normalized such that the sum of the label likelihoods for each sequence is 1. If `True`, the labels are considered independent and probabilities are normalized for each candidate by doing a softmax of the entailment score vs. the contradiction score. Return: A `dict` or a list of `dict`: Each result comes as a dictionary with the following keys: - **sequence** (`str`) -- The sequence for which this is the output. - **labels** (`list[str]`) -- The labels sorted by order of likelihood. - **scores** (`list[float]`) -- The probabilities for each of the labels. rrrVz%Unable to understand extra arguments )r#r$r3r()rr&argsr9r:s rr(z'ZeroShotClassificationPipeline.__call__szH t99>>  YY!^^ 2& @ @)-aF% & &KTKKLL Luww 44V444rNThis example is {}.c#K||||\}}tt||D]B\}\}}||g} ||d|t |dz kd| VCdS)Nrrcandidate_labelr"is_last)r2 enumerateziprRr#) rrPrVr!r'r&ir` sequence_pair model_inputs r preprocessz)ZeroShotClassificationPipeline.preprocesss$($5$5f>NPc$d$d! 3z;ZeroShotClassificationPipeline._forward..sOOO6!9OOOrpt use_cacheFrar_r) rJmodel_input_namesrIr=forwardcallinspect signature parameters)rrPr`r" model_inputs model_forwardoutputs model_outputss ` r_forwardz'ZeroShotClassificationPipeline._forwards !23*%OOOOdn.NOOO .2n.D.D **$*/ '+M::E E E(-L %$*,,|,, / i(    rcd|Dd|D}|jdkrtjd|D}ntjd|D}|jd}t }||z}|||df}|st dkrd|j} | dkrdnd} |d | | gf} tj| tj| dd z } | d } nM|d |jf} tj| tj| dd z } tt| d }|dfd |D| d|f dS)Ncg|] }|d S)r`rrrvs rrz>ZeroShotClassificationPipeline.postprocess..sTTT7G$56TTTrcg|] }|d S)r"rr{s rrz>ZeroShotClassificationPipeline.postprocess..sFFFWWZ(FFFrrlcfg|].}|d/Slogits)floatnumpyroutputs rrz>ZeroShotClassificationPipeline.postprocess..s5$b$b$b&VH%5%;%;%=%=%C%C%E%E$b$b$brcBg|]}|dSr~)rrs rrz>ZeroShotClassificationPipeline.postprocess..s)$Z$Z$Z&VH%5%;%;%=%=$Z$Z$Zrrr1r.T)keepdims).rc g|] }| Srr)rrdrVs rrz>ZeroShotClassificationPipeline.postprocess.. s===q'*===r)r"rscores) rInp concatenateshaper#reshaper5expsumlistreversedargsorttolist)rrwrUr&rNn num_sequencesreshaped_outputsr5contradiction_identail_contr_logitsr entail_logitstop_indsrVs @r postprocessz*ZeroShotClassificationPipeline.postprocesssTTmTTTFF FFF >T ! !^$b$bTa$b$b$bccFF^$Z$ZM$Z$Z$Z[[F LO  ! !Q !>>=!R*@AA  Z#.//144 .M%2a%7%7rrQ "239I=8Y3Y"Z V/0026:M3N3N3R3RSU`d3R3e3eeFF^FF-S$2D-DEMVM**RVM-B-B-F-FrTX-F-Y-YYF!2!2!4!45566!! ====H===Q[)0022   r)Nr])F)r)r*r+r,_load_processor_load_image_processor_load_feature_extractor_load_tokenizerr r4propertyr5r ONLY_FIRSTrRrZrrrr(rgrxr __classcell__)r:s@rr/r/,s#%%NO!#O#H#H#J#JX'+tPbPm((((T999$+5d3i(+5+5+5+5+5+5Z    $        rr/)rqtypingrrrtokenization_utilsrutilsrrbaser r r get_loggerr)r6r r/rrrrs333333////////JJJJJJJJJJ  H % %)))))O)))<,,4@@@AAb b b b b ]b b BAb b b r