o shI@sddlZddlZddlmZmZddlmZddlmZddl m Z m Z m Z m Z ddlmZmZe r=ddlZdd lmZe rFdd lmZe eZGd d d ejZe ed dGdddeZe ed dGdddeZe ed dGdddeZdS)N)AnyUnion)GenerationConfig)TruncationStrategy)add_end_docstringsis_tf_availableis_torch_availablelogging)Pipelinebuild_pipeline_init_args)/TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING_NAMES),MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING_NAMESc@seZdZdZdZdS) ReturnTyperr N)__name__ __module__ __qualname__TENSORSTEXTrri/var/www/html/alpaca_bot/venv/lib/python3.10/site-packages/transformers/pipelines/text2text_generation.pyrsrT) has_tokenizerc seZdZdZdZdZdZdZdZe dddZ dZ fdd Z dd d Z d ededefddZddZdeeeefdedeeeefffdd ZejfddZddZejdfddZZS) Text2TextGenerationPipelinea9 Pipeline for text to text generation using seq2seq models. Unless the model you're using explicitly sets these generation parameters in its configuration files (`generation_config.json`), the following default values will be used: - max_new_tokens: 256 - num_beams: 4 Example: ```python >>> from transformers import pipeline >>> generator = pipeline(model="mrm8488/t5-base-finetuned-question-generation-ap") >>> generator( ... "answer: Manuel context: Manuel has created RuPERTa-base with the support of HF-Transformers and Google" ... ) [{'generated_text': 'question: Who created the RuPERTa-base?'}] ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial). You can pass text generation parameters to this pipeline to control stopping criteria, decoding strategy, and more. Learn more about text generation parameters in [Text generation strategies](../generation_strategies) and [Text generation](text_generation). This Text2TextGenerationPipeline pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"text2text-generation"`. The models that this pipeline can use are models that have been fine-tuned on a translation task. See the up-to-date list of available models on [huggingface.co/models](https://huggingface.co/models?filter=text2text-generation). For a list of available parameters, see the [following documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation#transformers.generation.GenerationMixin.generate) Usage: ```python text2text_generator = pipeline("text2text-generation") text2text_generator("question: What is 42 ? context: 42 is the answer to life, the universe and everything") ```TF)max_new_tokens num_beams generatedcs4tj|i|||jdkrtdStdS)Ntf)super__init__check_model_type frameworkrrselfargskwargs __class__rrr!Ts z$Text2TextGenerationPipeline.__init__Nc Ksi}|dur ||d<|} i} |dur|dur|rtjntj}|dur&|| d<|dur.|| d<|durK|jj|dd} t| dkrEtd| d|d <|jdurU|j| d <|j durd|j| d <|j | d <|| | fS) N truncation return_typeclean_up_tokenization_spacesF)add_special_tokensr zStopping on a multiple token sequence is not yet supported on transformers. The first token of the stop sequence will be used as the stop sequence string in the interim.r eos_token_idassistant_model tokenizerassistant_tokenizer) rrrr0encodelenwarningswarnr/r1) r%return_tensors return_textr+r,r* stop_sequencegenerate_kwargspreprocess_paramsforward_paramspostprocess_paramsstop_sequence_idsrrr_sanitize_parameters]s0         z0Text2TextGenerationPipeline._sanitize_parameters input_length min_length max_lengthcCsdS)j Checks whether there might be something wrong with given input with regard to the model. Trr%r?r@rArrr check_inputssz(Text2TextGenerationPipeline.check_inputscs|jdur|jndt|dtr*|jjdurtdfdd|dDf}d}nt|dtr;|df}d}n td|dd |j||||jd }d |vrW|d =|S) NrzOPlease make sure that the tokenizer has a pad_token_id when using a batch inputcsg|]}|qSrr).0argprefixrr zCText2TextGenerationPipeline._parse_and_tokenize..TFz `args[0]`: zI have the wrong format. The should be either of type `str` or type `list`)paddingr*r6token_type_ids) rI isinstancelistr0 pad_token_id ValueErrorstr TypeErrorr#)r%r*r&rLinputsrrHr_parse_and_tokenizes  z/Text2TextGenerationPipeline._parse_and_tokenizer&r'returncsZtj|i|}t|dtr+tdd|dDr+tdd|Dr+dd|DS|S)a Generate the output text(s) using text(s) given as inputs. Args: args (`str` or `list[str]`): Input text for the encoder. return_tensors (`bool`, *optional*, defaults to `False`): Whether or not to include the tensors of predictions (as token indices) in the outputs. return_text (`bool`, *optional*, defaults to `True`): Whether or not to include the decoded texts in the outputs. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the potential extra spaces in the text output. truncation (`TruncationStrategy`, *optional*, defaults to `TruncationStrategy.DO_NOT_TRUNCATE`): The truncation strategy for the tokenization within the pipeline. `TruncationStrategy.DO_NOT_TRUNCATE` (default) will never truncate, but it is sometimes desirable to truncate the input to fit the model's max_length instead of throwing an error down the line. generate_kwargs: Additional keyword arguments to pass along to the generate method of the model (see the generate method corresponding to your framework [here](./text_generation)). Return: A list or a list of list of `dict`: Each result comes as a dictionary with the following keys: - **generated_text** (`str`, present when `return_text=True`) -- The generated text. - **generated_token_ids** (`torch.Tensor` or `tf.Tensor`, present when `return_tensors=True`) -- The token ids of the generated text. rcss|]}t|tVqdS)N)rNrR)rFelrrr sz7Text2TextGenerationPipeline.__call__..css|] }t|dkVqdS)r N)r3rFresrrrrXscSsg|]}|dqS)rrrYrrrrJrKz8Text2TextGenerationPipeline.__call__..)r __call__rNrOall)r%r&r'resultr(rrr[s z$Text2TextGenerationPipeline.__call__cKs|j|fd|i|}|S)Nr*)rU)r%rTr*r'rrr preprocesssz&Text2TextGenerationPipeline.preprocesscKs|jdkr |dj\}}n|jdkrt|d\}}|||d|jj|d|jjd|vr9|j|d<|j j d i||}|jd}|jdkrc|j |||g|jddR}d |iS|jdkrzt ||||g|jddR}d |iS) Npt input_idsrr@rAgeneration_configrr output_idsr) r#shapernumpyrDgetrar@rAmodelgeneratereshape)r% model_inputsr9in_br?rbout_brrr_forwards&     " $z$Text2TextGenerationPipeline._forwardcCshg}|ddD])}|tjkr|jd|i}n|tjkr,|jd|jj|d|di}||q|S)Nrbr _token_ids_textT)skip_special_tokensr,)rr return_namerr0decodeappend)r% model_outputsr+r,recordsrbrecordrrr postprocesss   z'Text2TextGenerationPipeline.postprocess)NNNNNN)rrr__doc___pipeline_calls_generate_load_processor_load_image_processor_load_feature_extractor_load_tokenizerr_default_generation_configrpr!r>intrDrUrrRrOrdictr[rDO_NOT_TRUNCATEr^rlrrrv __classcell__rrr(rrs4)   *2&rcs>eZdZdZdZfddZdedededefd d ZZ S) SummarizationPipelinea Summarize news articles and other documents. This summarizing pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"summarization"`. The models that this pipeline can use are models that have been fine-tuned on a summarization task, which is currently, '*bart-large-cnn*', '*google-t5/t5-small*', '*google-t5/t5-base*', '*google-t5/t5-large*', '*google-t5/t5-3b*', '*google-t5/t5-11b*'. See the up-to-date list of available models on [huggingface.co/models](https://huggingface.co/models?filter=summarization). For a list of available parameters, see the [following documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation#transformers.generation.GenerationMixin.generate) Unless the model you're using explicitly sets these generation parameters in its configuration files (`generation_config.json`), the following default values will be used: - max_new_tokens: 256 - num_beams: 4 Usage: ```python # use bart in pytorch summarizer = pipeline("summarization") summarizer("An apple a day, keeps the doctor away", min_length=5, max_length=20) # use t5 in tf summarizer = pipeline("summarization", model="google-t5/t5-base", tokenizer="google-t5/t5-base", framework="tf") summarizer("An apple a day, keeps the doctor away", min_length=5, max_length=20) ```summaryctj|i|S)a Summarize the text(s) given as inputs. Args: documents (*str* or `list[str]`): One or several articles (or one list of articles) to summarize. return_text (`bool`, *optional*, defaults to `True`): Whether or not to include the decoded texts in the outputs return_tensors (`bool`, *optional*, defaults to `False`): Whether or not to include the tensors of predictions (as token indices) in the outputs. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the potential extra spaces in the text output. generate_kwargs: Additional keyword arguments to pass along to the generate method of the model (see the generate method corresponding to your framework [here](./text_generation)). Return: A list or a list of list of `dict`: Each result comes as a dictionary with the following keys: - **summary_text** (`str`, present when `return_text=True`) -- The summary of the corresponding input. - **summary_token_ids** (`torch.Tensor` or `tf.Tensor`, present when `return_tensors=True`) -- The token ids of the summary. r r[r$r(rrr[szSummarizationPipeline.__call__r?r@rArVc CsR||krtd|d|d||kr'td|d|d|ddd Sd S) rBzYour min_length=z' must be inferior than your max_length=.zYour max_length is set to z , but your input_length is only z. Since this is a summarization task, where outputs shorter than the input are typically wanted, you might consider decreasing max_length manually, e.g. summarizer('...', max_length=r)NloggerwarningrCrrrrD1sz"SummarizationPipeline.check_inputs) rrrrwrpr[r~boolrDrrrr(rrs  "rcs`eZdZdZdZdededefddZejddd fd d Z dfd d Z fddZ Z S)TranslationPipelinea Translates from one language to another. This translation pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"translation_xx_to_yy"`. The models that this pipeline can use are models that have been fine-tuned on a translation task. See the up-to-date list of available models on [huggingface.co/models](https://huggingface.co/models?filter=translation). For a list of available parameters, see the [following documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation#transformers.generation.GenerationMixin.generate) Unless the model you're using explicitly sets these generation parameters in its configuration files (`generation_config.json`), the following default values will be used: - max_new_tokens: 256 - num_beams: 4 Usage: ```python en_fr_translator = pipeline("translation_en_to_fr") en_fr_translator("How old are you?") ``` translationr?r@rAcCs(|d|krtd|d|ddS)Ng?zYour input_length: z" is bigger than 0.9 * max_length: z`. You might consider increasing your max_length manually, e.g. translator('...', max_length=400)TrrCrrrrD\s z TranslationPipeline.check_inputsN)r*src_langtgt_langcs:t|jddr|jj||j|||dStj|d|iS)N_build_translation_inputs)r6r*rrr*)getattrr0rr#r rU)r%r*rrr&r(rrr^ds  zTranslationPipeline.preprocessc  stjdi|\}}}|dur||d<|dur||d<|durD|durD|d|j}|d}|rDt|dkrD|d|d<|d|d<|||fS) Nrrtask_rr r)r r>rersplitr3) r%rrr'r:r;r<ritemsr(rrr>ls    z(TranslationPipeline._sanitize_parameterscr)a Translate the text(s) given as inputs. Args: args (`str` or `list[str]`): Texts to be translated. return_tensors (`bool`, *optional*, defaults to `False`): Whether or not to include the tensors of predictions (as token indices) in the outputs. return_text (`bool`, *optional*, defaults to `True`): Whether or not to include the decoded texts in the outputs. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the potential extra spaces in the text output. src_lang (`str`, *optional*): The language of the input. Might be required for multilingual models. Will not have any effect for single pair translation models tgt_lang (`str`, *optional*): The language of the desired output. Might be required for multilingual models. Will not have any effect for single pair translation models generate_kwargs: Additional keyword arguments to pass along to the generate method of the model (see the generate method corresponding to your framework [here](./text_generation)). Return: A list or a list of list of `dict`: Each result comes as a dictionary with the following keys: - **translation_text** (`str`, present when `return_text=True`) -- The translation. - **translation_token_ids** (`torch.Tensor` or `tf.Tensor`, present when `return_tensors=True`) -- The token ids of the translation. rr$r(rrr[|szTranslationPipeline.__call__)NN) rrrrwrpr~rDrrr^r>r[rrrr(rr@sr)enumr4typingrr generationrtokenization_utilsrutilsrrr r baser r tensorflowrmodels.auto.modeling_tf_autormodels.auto.modeling_autor get_loggerrrEnumrrrrrrrrs*       Z J