o Zh@sdZddlmZmZmZddlmZddlmZm Z m Z m Z ddl m Z mZmZddlmZGdd d ed d ZGd d d e d d ZeeZGddde ZdgZdS)z! Processor class for Pix2Struct. )ListOptionalUnion) BatchFeature) ImagesKwargsProcessingKwargsProcessorMixinUnpack) BatchEncodingPreTokenizedInput TextInput)loggingc@s:eZdZUeeed<eeeee ee efed<dS)Pix2StructImagesKwargs max_patches header_textN) __name__ __module__ __qualname__rint__annotations__rr r rrrc/var/www/auris/lib/python3.10/site-packages/transformers/models/pix2struct/processing_pix2struct.pyrs $rF)totalc @s8eZdZUeed<dddddddddd ddidZd S) Pix2StructProcessorKwargs images_kwargsTFr) add_special_tokenspaddingZstrideZreturn_overflowing_tokensZreturn_special_tokens_maskZreturn_offsets_mappingreturn_token_type_idsZ return_lengthverboseri) text_kwargsrN)rrrrr _defaultsrrrrr s   rc seZdZdZddgZdZdZfddZ    dd ee e e e e e fd e e d eeeffd d ZddZddZeddZZS)Pix2StructProcessora Constructs a PIX2STRUCT processor which wraps a BERT tokenizer and PIX2STRUCT image processor into a single processor. [`Pix2StructProcessor`] offers all the functionalities of [`Pix2StructImageProcessor`] and [`T5TokenizerFast`]. See the docstring of [`~Pix2StructProcessor.__call__`] and [`~Pix2StructProcessor.decode`] for more information. Args: image_processor (`Pix2StructImageProcessor`): An instance of [`Pix2StructImageProcessor`]. The image processor is a required input. tokenizer (Union[`T5TokenizerFast`, `T5Tokenizer`]): An instance of ['T5TokenizerFast`] or ['T5Tokenizer`]. The tokenizer is a required input. image_processor tokenizerZPix2StructImageProcessor)Z T5TokenizerZT5TokenizerFastcsd|_t||dS)NF)rsuper__init__)selfr#r$ __class__rrr&JszPix2StructProcessor.__init__Ntextkwargsreturnc Ks\|dur |dur td|jtfd|jji|}|ddd}|durG|jjsG|dur/|nd|dd<|j|_|jdd|i|d}|S|jjsW|j|fi|d} n|d d ||j|fi|d} |dur|jjs|durx|nd |dd<|jdd|i|d}d |vr|d |d <d |vr|d |d<nd}|dur| || S)a This method uses [`Pix2StructImageProcessor.preprocess`] method to prepare image(s) for the model, and [`T5TokenizerFast.__call__`] to prepare text for the model. Please refer to the docstring of the above two methods for more information. Nz*You have to specify either images or text.Ztokenizer_init_kwargsr rTr*rrFZattention_maskZdecoder_attention_maskZ input_idsZdecoder_input_idsr) ValueErrorZ _merge_kwargsrr$Z init_kwargspopr#Zis_vqaZcurrent_processor setdefaultupdate) r'Zimagesr*ZaudioZvideosr+Z output_kwargsr text_encodingZencoding_image_processorrrr__call__NsB   zPix2StructProcessor.__call__cO|jj|i|S)z This method forwards all its arguments to Pix2StructTokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r$ batch_decoder'argsr+rrrr4z Pix2StructProcessor.batch_decodecOr3)z This method forwards all its arguments to Pix2StructTokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r$decoder5rrrr8r7zPix2StructProcessor.decodecCs"|jj}|jj}tt||S)N)r$model_input_namesr#listdictfromkeys)r'Ztokenizer_input_namesZimage_processor_input_namesrrrr9sz%Pix2StructProcessor.model_input_names)NNNN)rrr__doc__ attributesZimage_processor_classZtokenizer_classr&rr r rr rr rr2r4r8propertyr9 __classcell__rrr(rr"7s*   :r"N)r=typingrrrZfeature_extraction_utilsrZprocessing_utilsrrr r Ztokenization_utils_baser r r utilsrrrZ get_loggerrloggerr"__all__rrrrs    f