o ij@sdZddlZddlmZddlmZmZddlmZddl m Z ddl m Z m Z mZdd lmZmZGd d d e d d ZGddde ZdgZdS)z Processor class for TrOCR. N)contextmanager)OptionalUnion) BatchFeature) ImageInput)ProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInputc@seZdZiZdS)TrOCRProcessorKwargsN)__name__ __module__ __qualname__ _defaultsrr^/home/ubuntu/.local/lib/python3.10/site-packages/transformers/models/trocr/processing_trocr.pyr sr F)totalc seZdZdZddgZdZdZdfdd Z    dd ee d e e e e e e e fd eed efd dZeddZeddZeddZeddZZS)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|}t|||j|_d|_dS)Nfeature_extractorzhThe `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor` instead.F) warningswarn FutureWarningpopsuper__init__rcurrent_processor_in_target_context_manager)selfrrkwargsr __class__rrr 4s  zTrOCRProcessor.__init__imagestextr$returnc Ks|jr |j|fi|S|dur|durtd|jtfd|jji|}|dur4|j|fi|d}|durC|j|fi|d}|durI|S|durO|S|d|d<|S)a When used in normal mode, this method forwards all its arguments to AutoImageProcessor's [`~AutoImageProcessor.__call__`] and returns its output. If used in the context [`~TrOCRProcessor.as_target_processor`] this method forwards all its arguments to TrOCRTokenizer's [`~TrOCRTokenizer.__call__`]. Please refer to the docstring of the above two methods for more information. NzBYou need to specify either an `images` or `text` input to process.tokenizer_init_kwargs images_kwargs text_kwargs input_idslabels)r"r! ValueError _merge_kwargsr r init_kwargsr) r#r'r(audiovideosr$ output_kwargsinputs encodingsrrr__call__Ds* zTrOCRProcessor.__call__cCs|jj}|dgS)Nr.)rmodel_input_names)r#image_processor_input_namesrrrr8ls z TrOCRProcessor.model_input_namesccs2tdd|_|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_processorqs 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:rrrrr?z TrOCRProcessor.feature_extractor)NN)NNNN)rrr__doc__ attributesr=tokenizer_classr rrrr r listr r rr7propertyr8rr;r>r __classcell__rrr%rr!s6 (   r)r@r contextlibrtypingrrimage_processing_utilsr image_utilsrprocessing_utilsrr r tokenization_utils_baser r r r__all__rrrrs    q