Commit történet

Szerző SHA1 Üzenet Dátum
  myhloli 99192002dd refactor: improve character span calculation and sorting logic 6 hónapja
  myhloli 546be00aac refactor: update OCR score handling to filter low-confidence results 6 hónapja
  myhloli 3334157f15 refactor: clean up unused OCR area calculation and update demo PDF path 6 hónapja
  myhloli 236a6033f1 refactor: improve block processing logic and enhance span handling 6 hónapja
  myhloli 6f2c3ad89e refactor: enhance title block processing and improve markdown generation 6 hónapja
  myhloli 7d4ce0c380 refactor: add LLM-aided title optimization and improve config handling 6 hónapja
  myhloli d2de6d801a refactor: update text span extraction to use new version and improve character handling 6 hónapja
  myhloli 1ed61cb5d6 refactor: update OCR span extraction logic and improve PDF page processing 6 hónapja
  myhloli 1f2ea493cd refactor: change default value of enable_ocr_det_batch to False in BatchAnalyze constructor 6 hónapja
  myhloli f211554137 refactor: improve text processing by adding ligature and unicode replacement functions 6 hónapja
  myhloli 76e1a7c1b7 refactor: enhance markdown generation by introducing pipeline_union_make and improving latex delimiter handling 6 hónapja
  myhloli 9ded9e6bad refactor: simplify UnimernetModel constructor by removing unused cfg_path parameter 6 hónapja
  myhloli 51393aa814 refactor: update union_make import and adjust middle JSON structure for consistency 6 hónapja
  myhloli 6b1df41947 refactor: optimize OCR batch processing and enhance image cropping logic 6 hónapja
  myhloli 73f8503514 refactor: optimize OCR batch processing and enhance image cropping logic 6 hónapja
  myhloli 101b12a10a refactor: improve image handling by transitioning from NumPy arrays to PIL images in cropping functions 6 hónapja
  myhloli a9abb4e201 refactor: enhance OCR processing and paragraph splitting in document analysis pipeline 6 hónapja
  myhloli 7a22bfeebe refactor: enhance image margin cropping and processing for improved handling of PIL and NumPy images 6 hónapja
  myhloli bd2c3d120a refactor: update OCR handling and adjust root directory path for model loading 6 hónapja
  myhloli 38ace5dc61 refactor: streamline document analysis and enhance image handling in processing pipeline 6 hónapja
  myhloli 6833882585 refactor: enhance language support and improve document parsing for multiple files 6 hónapja
  myhloli 0f21495a06 refactor: enhance block processing and sorting utilities for improved span management 6 hónapja
  myhloli ae7b0a6eba refactor: implement block preprocessing utilities for improved bounding box management 6 hónapja
  myhloli 8f1f9abec5 refactor: enhance bounding box utilities and add configuration reader for S3 integration 6 hónapja
  myhloli 7285ea9285 refactor: improve document analysis by integrating image loading and enhancing data handling 6 hónapja
  myhloli ea5cb65a1f refactor: enhance document parsing by supporting multiple PDF files and improving method organization 6 hónapja
  myhloli 0a899f1af8 feat: add batch processing for OCR detection and implement new client and common utilities 6 hónapja
  myhloli cbba27b4f5 refactor: reorganize project structure and update import paths 6 hónapja
  Xiaomeng Zhao 3027c677c9 Merge pull request #11 from johnking0099/refactor-mineru2 6 hónapja
  Jin Zhen Jiang 8e55a52693 feat: add mineru-vlm backend. 6 hónapja