myhloli 7d5850e3ce feat(model): add xycut algorithm for block sorting ...

- Implement xycut algorithm to sort blocks when layoutreader fails
- Add recursive_xy_cut function to perform the xycut algorithm- Update pdf_parse_union_core_v2.py to use xycut when layoutreader fails
- Modify draw_bbox.py to handle cases where layoutreader fails to sort blocks

2024-11-08 00:49:20 +08:00

..

__init__.py

目录重构

2024-03-01 16:07:51 +08:00

boxbase.py

feat(model): add support for DocLayout-YOLO model

2024-10-23 17:07:39 +08:00

calc_span_stats.py

目录重构

2024-03-01 16:07:51 +08:00

clean_memory.py

refactor(memory management): remove unused clean_memory function

2024-09-29 10:47:54 +08:00

commons.py

add version_name to middle json

2024-06-04 11:15:52 +08:00

config_reader.py

feat: add [figure | table] match [caption | footnote] match algorithm v2

2024-10-24 10:50:04 +08:00

Constants.py

Merge branch 'dev' into dev-table-model-update

2024-10-28 16:36:32 +08:00

convert_utils.py

中间态dict结构调整

2024-04-15 18:51:58 +08:00

coordinate_transform.py

将fix缩放倍率的bbox写入model_list

2024-04-17 16:54:33 +08:00

detect_language_from_model.py

语言检测逻辑移动到parse流程

2024-04-11 18:07:45 +08:00

draw_bbox.py

feat(model): add xycut algorithm for block sorting

2024-11-08 00:49:20 +08:00

drop_reason.py

add new drop scene

2024-05-10 18:33:31 +08:00

drop_tag.py

block type 字段名修复

2024-04-22 16:58:02 +08:00

hash_utils.py

实现parse_ocr_pdf api，切图逻辑s3使用平铺地址，本地使用层级地址，删除预设s3_image_save_path

2024-04-10 15:21:33 +08:00

json_compressor.py

目录重构

2024-03-01 16:07:51 +08:00

language.py

feat(language): add FT LANG cache directory setup

2024-07-23 20:09:05 +08:00

local_math.py

refactor(magic_pdf): replace math module with local_math

2024-07-23 20:09:57 +08:00

MakeContentConfig.py

feat(ocr_mkcontent): support drop reason in none_with_reason modeEnable the NONE_WITH_REASON drop mode in para_to_standard_format_v2 by updating the

2024-09-18 15:47:36 +08:00

markdown_utils.py

make markdown时特殊符号转义

2024-03-14 18:49:54 +08:00

ModelBlockTypeEnum.py

io modules

2024-04-19 15:29:11 +08:00

nlp_utils.py

目录重构

2024-03-01 16:07:51 +08:00

ocr_content_type.py

feat(list&index block): detect and merge list and index blocks

2024-10-15 00:50:48 +08:00

path_utils.py

update:cleanup requirements.txt

2024-06-28 14:41:12 +08:00

pdf_check.py

update check invalid_chars algorithm to improve accuracy

2024-06-20 11:18:15 +08:00

pdf_image_tools.py

Merge branch 'master' into master

2024-04-16 21:16:29 +08:00

safe_filename.py

目录重构

2024-03-01 16:07:51 +08:00

textbase.py

目录重构

2024-03-01 16:07:51 +08:00

version.py

feat(pdf_parse): improve span filtering and add new block types

2024-11-01 15:21:16 +08:00

vis_utils.py

目录重构

2024-03-01 16:07:51 +08:00