9iTXSSKrSSKJrJrJrJrJrJr SSKr SSK J r SSK J r SSKJr SSKJr SS KJr SS KJr SS KJr SS KJr S SKJr S SKJr S SKJr S SK J!r! S SK"J#r# SSK$J%r%J&r& SSK'J(r( SSK J)r) \RT"SS\55r+\ "S5"SS\55r,g)N)AnyDictListOptionalTupleUnion)logging)pipeline_requires_extra)ImageBatchSampler) ReadImage) DetResult) benchmark) HPIConfig)PaddlePredictorOption)(AutoParallelImageSimpleInferencePipeline) BasePipeline) CropByBoxes)DocPreprocessorResult) OCRResult)SingleTableRecognitionResultTableRecognitionResult)get_table_recognition_res)get_neighbor_boxes_idxc!P^\rSrSrSrS'S\S\S\S\S\ \ \\\ 4\ 4S S4 U4S jjjr S \ \S \ \S \ \S\ \S \4 SjrS\S\S\S \4SjrS\R(S\S \\\R(44SjrSrSrS(S\R(S\S\S\S\S\S \4SjjrS)S\ \\\\R(\\R(4S \ \S \ \S \ \S\ \S\ \S\ \S\ \S \ \S!\ \S"\ \S#\ \S$\ \S\S\ \S \4 S%jjr S&r!U=r"$)*_TableRecognitionPipeline&zTable Recognition PipelineNconfigdevice pp_optionuse_hpip hpi_configreturnc>[T U]X#XES9 URSS5UlUR(a:URS05RSSS05nUR U5UlURSS5UlUR (a:URS 05RS S S 05nURU5UlURS 05RS S S05nURU5Ul URSS5Ul UR(a;URS05RSSS05n UR U 5Ul O'URS05RSS5Ul [5Ul[SS9Ul[#SS9Ulg)aQInitializes the layout parsing pipeline. Args: config (Dict): Configuration dictionary containing various settings. device (str, optional): Device to run the predictions on. Defaults to None. pp_option (PaddlePredictorOption, optional): PaddlePredictor options. Defaults to None. use_hpip (bool, optional): Whether to use the high-performance inference plugin (HPIP) by default. Defaults to False. hpi_config (Optional[Union[Dict[str, Any], HPIConfig]], optional): The default high-performance inference configuration dictionary. Defaults to None. )r"r#r$r%use_doc_preprocessorT SubPipelinesDocPreprocessorpipeline_config_errorz+config error for doc_preprocessor_pipeline!use_layout_detection SubModulesLayoutDetectionmodel_config_errorz"config error for layout_det_model!TableStructureRecognitionz'config error for table_structure_model! use_ocr_model GeneralOCRz&config error for general_ocr_pipeline!Nr) batch_sizeBGR)format)super__init__getr(create_pipelinedoc_preprocessor_pipeliner, create_modellayout_det_modeltable_structure_modelr1general_ocr_pipelinegeneral_ocr_config_bakr_crop_by_boxesr batch_samplerr img_reader) selfr!r"r#r$r%doc_preprocessor_configlayout_det_configtable_structure_configgeneral_ocr_config __class__s v/var/www/html/land-doc-ocr/venv/lib/python3.13/site-packages/paddlex/inference/pipelines/table_recognition/pipeline.pyr7"_TableRecognitionPipeline.__init__*s*   %+JJ/Et$L!  $ $&,jj&D&H&H!+-Z' # .2-A-A'.D *%+JJ/Et$L!  $ $ & < < @ @!%'KL! %)$5$56G$HD !!'L"!=!A!A ' !#L M" &*%6%67M%N"#ZZ>   !'NB!?!C!C(*RS" )-(<(<=O(PD %*0**^R*H*L*Ld+D '*m.!<#51use_doc_orientation_classifyuse_doc_unwarpingr,r1cUcUc URnOUSLdUSLaSnOSnUc URnUc URn[UUUS9$)a Get the model settings based on the provided parameters or default values. Args: use_doc_orientation_classify (Optional[bool]): Whether to use document orientation classification. use_doc_unwarping (Optional[bool]): Whether to use document unwarping. use_layout_detection (Optional[bool]): Whether to use layout detection. use_ocr_model (Optional[bool]): Whether to use OCR model. Returns: dict: A dictionary containing the model settings. TF)r(r,r1)r(r,r1dict)rCrLrMr,r1r(s rIget_model_settings,_TableRecognitionPipeline.get_model_settingsnsq& ( /4E4M#'#<#< +t37HD7P'+$',$  '#'#<#<   ..M!5!5'  rKmodel_settingsoverall_ocr_reslayout_det_rescUS(a(UR(d[R"S5 gUS(aBUb[R"S5 gUR(d[R"S5 gUS(aDUb[R"S5 gUR(d[R"S 5 gg Uc[R"S 5 gg ) a Check if the input parameters are valid based on the initialized models. Args: model_settings (Dict): A dictionary containing input parameters. overall_ocr_res (OCRResult): Overall OCR result obtained after running the OCR pipeline. The overall OCR result with convert_points_to_boxes information. layout_det_res (DetResult): The layout detection result. Returns: bool: True if all required models are initialized according to input parameters, False otherwise. r(zRSet use_doc_preprocessor, but the models for doc preprocessor are not initialized.Fr,z^The layout detection model has already been initialized, please set use_layout_detection=FalsezRSet use_layout_detection, but the models for layout detection are not initialized.r1zLThe OCR models have already been initialized, please set use_ocr_model=Falsez>Set use_ocr_model, but the models for OCR are not initialized.z7Set use_ocr_model=False, but no OCR results were found.T)r(r errorr,r1)rCrRrSrTs rIcheck_model_settings_valid4_TableRecognitionPipeline.check_model_settings_valids$ 0 1$:S:S MMd  0 1) t,, h / ** b%% T && WXrK image_array input_paramscUS(a/USnUSn[URUUUS95SnUSnXV4$0nUnXV4$)as Preprocess the document image based on input parameters. Args: image_array (np.ndarray): The input image array. input_params (dict): Dictionary containing preprocessing parameters. Returns: tuple[DocPreprocessorResult, np.ndarray]: A tuple containing the preprocessing result dictionary and the processed image array. r(rLrMrLrMr output_img)listr:)rCrYrZrLrMdoc_preprocessor_resdoc_preprocessor_images rIpredict_doc_preprocessor_res6_TableRecognitionPipeline.predict_doc_preprocessor_ress . /+78V+W ( ,-@ A #'..1M&7/$  $ &:,%G "$;;$& %0 "#;;rKc [U[5(dUR5n/n[[ U55HynX$Vs/sHn[ R "U5PM snupgp[URXU 2Xh2SS2455Sn URSRU S55 M{ U$s snfa< Splits OCR bounding boxes by table cells and retrieves text. Args: ori_img (ndarray): The original image from which text regions will be extracted. cells_bboxes (list or ndarray): Detected cell bounding boxes to extract text from. Returns: list: A list containing the recognized texts from each cell. Nr rec_texts isinstancer^tolistrangelenmathceilr>appendjoin rCori_img cells_bboxes texts_listikx1y1x2y2rec_tes rIsplit_ocr_bboxes_by_table_cells9_TableRecognitionPipeline.split_ocr_bboxes_by_table_cells,--'..0L s<()A4@ODOqdiilODNBB$33GrE25!O4LMNqQF   bggf[&9: ; * E B;c [U[5(dUR5n/n[[ U55HynX$Vs/sHn[ R "U5PM snupgp[URXU 2Xh2SS2455Sn URSRU S55 M{ U$s snfrdrgrps rIr{r|r}r~ table_box use_ocr_results_with_table_cellsflag_find_nei_textcell_sort_by_y_projectionc h[URU55SnUS:Xa8USnUV s/sHoSU SU SU S/PM nn URX5n O/n [UUUU UUS9n Sn U(a4[ US U5n [ U 5S:aU HnXS US -- n M XS 'U $s sn f) a Predict table recognition results from an image array, layout detection results, and OCR results. Args: image_array (np.ndarray): The input image represented as a numpy array. overall_ocr_res (OCRResult): Overall OCR result obtained after running the OCR pipeline. The overall OCR results containing text recognition information. table_box (list): The table box coordinates. use_ocr_results_with_table_cells (bool): whether to use OCR results with cells. flag_find_nei_text (bool): Whether to find neighboring text. cell_sort_by_y_projection (bool): Whether to sort the matched OCR boxes by y-projection. Returns: SingleTableRecognitionResult: single table recognition result. rTbboxrr rre rec_boxesrfz; neighbor_texts)r^r=r{rrrk)rCrYrSrrrrtable_structure_predtable_cells_resultrectcells_texts_listsingle_table_recognition_res neighbor_textmatch_idx_listidxs rI$predict_single_table_recognition_res>_TableRecognitionPipeline.predict_single_table_recognition_res s. $D$>$>{$KLQO +t 3!5f!= AS"ASa$q'47DG4AS " $CC   " '@    ,&? ( $ 3 ,iN>"Q&)C![%A#%F%MMM*9F%56++3"sB/inputtext_det_limit_side_lentext_det_limit_typetext_det_threshtext_det_box_threshtext_det_unclip_ratiotext_rec_score_threshc +# URUUUU5nUcSnURUXg5(dSS0v [URU55GHunnUR UR 5SnUS(a[ URUUUS95SnOSU0nUSnUS (a"[ URUUU U U U U S 95SnO8US :Xa2URS:wdeURUR5Ul/nS nUS (dSUcP0nURSSunnSSUS - US - /nURUUUUSUS9nUUS'URU5 US - nOUS (a[ URU55SnUSHjnUSR5S;dMUR!UU/5nUSnUSnURUSUUUUS9nUUS'URU5 US - nMl UR"SUR$SUUUUUS.n['U5v GM g7f)a This function predicts the layout parsing result for the given input. Args: input (Union[str, list[str], np.ndarray, list[np.ndarray]]): The input image(s) of pdf(s) to be processed. use_layout_detection (bool): Whether to use layout detection. use_doc_orientation_classify (bool): Whether to use document orientation classification. use_doc_unwarping (bool): Whether to use document unwarping. overall_ocr_res (OCRResult): The overall OCR result with convert_points_to_boxes information. It will be used if it is not None and use_ocr_model is False. layout_det_res (DetResult): The layout detection result. It will be used if it is not None and use_layout_detection is False. use_ocr_results_with_table_cells (bool): whether to use OCR results with cells. cell_sort_by_y_projection (bool): Whether to sort the matched OCR boxes by y-projection. **kwargs: Additional keyword arguments. Returns: TableRecognitionResult: The predicted table recognition result. NFrVz0the input params for model settings are invalid!rr(r\r]r1)rrrrrrTrr,r)rrtable_region_idboxeslabel)tableboximgr) input_path page_indexr_rTrStable_res_listrR)rPrW enumeraterArB instancesr^r:r>r?r9shaperrnr<lowerr@ input_paths page_indexesr) rCrrLrMr,r1rSrTrrrrrrrrkwargsrRimg_id batch_datarYr_r`rr img_height img_widthrsingle_table_rec_resbox_info crop_img_infosingle_img_ress rIpredict!_TableRecognitionPipeline.predictUsN00 (     % ,(- %.. O  NO O"+D,>,>u,E"F FJ//**>*>?BK45'+22#5Q*;3(  ($)5k'B$%9,%G "o."&--.0G,?(7,?.C.C. # #2T922d:::,0,@,@//-) NO!"89n>T!#(>(D(DRa(H% I9q=*q.A '+'P'P*#4',.G (Q($;J$%67%%&:;1$!"89%)--.DE&&N!/w 7H(..0I=(,(;(;K((T (5a(8 $1%$8  EE -e 4 / ) @:S F-CR,->?&--.BC'1,!!8&)44Q7(55a8(<"0#2"0"0N)8 8i#Gs F0H?6B H?) r@rAr:r?r>rBr<r=r(r,r1)NNFN)FTF)NNNNNNNNNNNNFN)#__name__ __module__ __qualname____firstlineno____doc__rstrrboolrrrrr7rOrPrrrWnpndarrayrrrar{r^rrrintfloatrr__static_attributes__ __classcell__)rHs@rIrr&s$ +/AE B2B2B2) B2  B2 U4S>9#<=> B2 B2B2H% &.tn% $D>% 'tn %  ~ %  % N55#5" 5  5n<::<59< $bjj0 1<<6@27#'*/3,ZZ3,#3, 3, +/ 3, ! 3,$(3, &3,p8<,0/3(,/3.215-1+//315151648!I9S$s)RZZbjj1AABI9'/tnI9$D> I9 'tn I9  ~ I9"),I9!+I9"*#I9&c]I9"%I9&e_I9 (I9 (I9+/I9 $,D>!I9$ %I9I9rKrocrc0\rSrSrS/r\S5rSrSrg)TableRecognitionPipelineitable_recognitionc[$)N)r)rCs rI _pipeline_cls&TableRecognitionPipeline._pipeline_clss((rKcg)Nr)rCr!s rI_get_batch_size(TableRecognitionPipeline._get_batch_sizesrKrN) rrrrentitiespropertyrrrrrKrIrrs!#$H ))rKr)-rltypingrrrrrrnumpyrutilsr utils.depsr common.batch_samplerr common.readerrmodels.object_detection.resultrutils.benchmarkr utils.hpirutils.pp_optionr _parallelrbaser componentsrdoc_preprocessor.resultr ocr.resultrresultrr!table_recognition_post_processingrr time_methodsrrrrKrIrs ::25&7("4@$;"HH) w9 w9w9t G rK