rh/$pdZddlZddlmZmZddlmZddlmZm Z m Z m Z m Z ddl mZGdd eZd gZy) z Processor class for LayoutXLM. N)OptionalUnion)ProcessorMixin) BatchEncodingPaddingStrategyPreTokenizedInput TextInputTruncationStrategy) TensorTypec(eZdZdZddgZdZdZd!fd Z d"dee e e e e e fde ee e e fd e ee e e e e e e fd e ee e e e e fd ed eeeefd eeeefde e de de e de ede edededededede eeefdef&dZdZdZdZedZedZed ZxZS)#LayoutXLMProcessoran Constructs a LayoutXLM processor which combines a LayoutXLM image processor and a LayoutXLM tokenizer into a single processor. [`LayoutXLMProcessor`] offers all the functionalities you need to prepare data for the model. It first uses [`LayoutLMv2ImageProcessor`] to resize document images to a fixed size, and optionally applies OCR to get words and normalized bounding boxes. These are then provided to [`LayoutXLMTokenizer`] or [`LayoutXLMTokenizerFast`], which turns the words and bounding boxes into token-level `input_ids`, `attention_mask`, `token_type_ids`, `bbox`. Optionally, one can provide integer `word_labels`, which are turned into token-level `labels` for token classification tasks (such as FUNSD, CORD). Args: image_processor (`LayoutLMv2ImageProcessor`, *optional*): An instance of [`LayoutLMv2ImageProcessor`]. The image processor is a required input. tokenizer (`LayoutXLMTokenizer` or `LayoutXLMTokenizerFast`, *optional*): An instance of [`LayoutXLMTokenizer`] or [`LayoutXLMTokenizerFast`]. The tokenizer is a required input. image_processor tokenizerLayoutLMv2ImageProcessor)LayoutXLMTokenizerLayoutXLMTokenizerFastc d|vr+tjdt|jd}||n}| t d| t dt |||y)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`.)warningswarn FutureWarningpop ValueErrorsuper__init__)selfrrkwargsr __class__s /var/www/html/ai-insurance-compliance-backend/venv/lib/python3.12/site-packages/transformers/models/layoutxlm/processing_layoutxlm.pyrzLayoutXLMProcessor.__init__3sq & ( MM  !' +> ? -<-H/N_  "HI I  AB B )4text text_pairboxes word_labelsadd_special_tokenspadding truncation max_lengthstridepad_to_multiple_ofreturn_token_type_idsreturn_attention_maskreturn_overflowing_tokensreturn_special_tokens_maskreturn_offsets_mapping return_lengthverbosereturn_tensorsreturnc T|jjr | td|jjr | td|dur|dur td|j||}|0|jjr|t|tr|g}|d}|j did ||n|dd ||ndd ||n|d d |d |d|d|d| d| d| d| d| d|d|d|d|d|d||}|j d}|dur|j||d}||d<|S)a This method first forwards the `images` argument to [`~LayoutLMv2ImagePrpcessor.__call__`]. In case [`LayoutLMv2ImagePrpcessor`] was initialized with `apply_ocr` set to `True`, it passes the obtained words and bounding boxes along with the additional arguments to [`~LayoutXLMTokenizer.__call__`] and returns the output, together with resized `images`. In case [`LayoutLMv2ImagePrpcessor`] was initialized with `apply_ocr` set to `False`, it passes the words (`text`/``text_pair`) and `boxes` specified by the user along with the additional arguments to [`~LayoutXLMTokenizer.__call__`] and returns the output, together with resized `images``. Please refer to the docstring of the above two methods for more information. NzdYou cannot provide bounding boxes if you initialized the image processor with apply_ocr set to True.zaYou cannot provide word labels if you initialized the image processor with apply_ocr set to True.TFzKYou cannot return overflowing tokens without returning the offsets mapping.)imagesr3wordsr"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2r3 pixel_valuesoverflow_to_sample_mappingimage)r apply_ocrr isinstancestrrrget_overflowing_images)rr6r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2r3rfeaturesencoded_inputss r __call__zLayoutXLMProcessor.__call__DsD    ) )u/@v     ) ){/Fs  % ,1G51Pjk k''vn'U   4 4 > >9CT$$v )I' )x/@ #,#8id !,%(72C $  2    " "   2 #8 #8 '@ (B $: (! "# $*' .n- $ ,00Hd9efF"(wr!cg}|D]}|j||t|t|k7r#tdt|dt||S)Nz`Expected length of images to be the same as the length of `overflow_to_sample_mapping`, but got z and )appendlenr)rr6r9images_with_overflow sample_idxs r r?z)LayoutXLMProcessor.get_overflowing_imagessy!4 rr r rrBr?rIrMpropertyrSrVr __classcell__)rs@r rrs(&$[1J6FO5(_cQUIMCG#'5:;?$(,00404*/+0',#;?)TI0$y/4HYCZZ[TE"3T:K5L"LMN T d49otDcO/DDEF T eDItDI$>?@ T!TtS/12T$%778TSMTT%SMT (~T (~T$(T %)!T"!%#T$%T&'T(!sJ!78)T, -Tl $<6@@**$$r!r)r[rtypingrrprocessing_utilsrtokenization_utils_baserrr r r utilsr r__all__r;r!r rhs7".wwm$m$` r!