o wi@sdZddlZddlmZddlmZddlmZddlm Z ddl m Z m Z m Z dd lmZmZGd d d e d d ZGddde ZdgZdS)z Processor class for TrOCR. N)contextmanager)Union) BatchFeature) ImageInput)ProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInputc@seZdZiZdS)TrOCRProcessorKwargsN)__name__ __module__ __qualname__ _defaultsrrg/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/transformers/models/trocr/processing_trocr.pyr sr F)totalc seZdZdZddgZdZdZdfdd Z    dd ed e e e e e e e fd e ed efd dZddZddZeddZeddZeddZZS)TrOCRProcessora Constructs a TrOCR processor which wraps a vision image processor and a TrOCR tokenizer into a single processor. [`TrOCRProcessor`] offers all the functionalities of [`ViTImageProcessor`/`DeiTImageProcessor`] and [`RobertaTokenizer`/`XLMRobertaTokenizer`]. See the [`~TrOCRProcessor.__call__`] and [`~TrOCRProcessor.decode`] for more information. Args: image_processor ([`ViTImageProcessor`/`DeiTImageProcessor`], *optional*): An instance of [`ViTImageProcessor`/`DeiTImageProcessor`]. The image processor is a required input. tokenizer ([`RobertaTokenizer`/`XLMRobertaTokenizer`], *optional*): An instance of [`RobertaTokenizer`/`XLMRobertaTokenizer`]. The tokenizer is a required input. image_processor tokenizerAutoImageProcessor AutoTokenizerNc srd}d|vrtdt|d}|dur|n|}|dur!td|dur)tdt|||j|_d|_ dS)Nfeature_extractorzhThe `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor` instead.z)You need to specify an `image_processor`.z"You need to specify a `tokenizer`.F) warningswarn FutureWarningpop ValueErrorsuper__init__rcurrent_processor_in_target_context_manager)selfrrkwargsr __class__rrr 4s  zTrOCRProcessor.__init__imagestextr$returnc Ks|jr |j|fi|S|dur|durtd|jtfd|jji|}|dur4|j|fi|d}|durC|j|fi|d}|durI|S|durO|S|d|d<|S)a When used in normal mode, this method forwards all its arguments to AutoImageProcessor's [`~AutoImageProcessor.__call__`] and returns its output. If used in the context [`~TrOCRProcessor.as_target_processor`] this method forwards all its arguments to TrOCRTokenizer's [`~TrOCRTokenizer.__call__`]. Please refer to the docstring of the above two methods for more information. NzBYou need to specify either an `images` or `text` input to process.tokenizer_init_kwargs images_kwargs text_kwargs input_idslabels)r"r!r _merge_kwargsr r init_kwargsr) r#r'r(audiovideosr$ output_kwargsinputs encodingsrrr__call__Hs* zTrOCRProcessor.__call__cO|jj|i|S)z This method forwards all its arguments to TrOCRTokenizer's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r batch_decoder#argsr$rrrr8pzTrOCRProcessor.batch_decodecOr7)z This method forwards all its arguments to TrOCRTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )rdecoder9rrrr<wr;zTrOCRProcessor.decodeccs2tdd|_|j|_dV|j|_d|_dS)z Temporarily sets the tokenizer for processing the input. Useful for encoding the labels when fine-tuning TrOCR. z`as_target_processor` is deprecated and will be removed in v5 of Transformers. You can process your labels by using the argument `text` of the regular `__call__` method (either in the same call as your images inputs, or in a separate call.TNF)rrr"rr!rr#rrras_target_processor~s z"TrOCRProcessor.as_target_processorcCtdt|jS)Nzg`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.)rrrimage_processor_classr=rrrfeature_extractor_class z&TrOCRProcessor.feature_extractor_classcCr?)Nz[`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.)rrrrr=rrrrrBz TrOCRProcessor.feature_extractor)NN)NNNN)r rr__doc__ attributesr@tokenizer_classr rrr r listr r rr6r8r<rr>propertyrAr __classcell__rrr%rr!s6 (  r)rCr contextlibrtypingrimage_processing_utilsr image_utilsrprocessing_utilsrrr tokenization_utils_baser r r r__all__rrrrs     ~