4673 Commits

Author SHA1 Message Date
Xiaomeng Zhao
146f655c5f Update mineru/backend/vlm/vlm_magic_model.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-12-30 17:02:17 +08:00
myhloli
05b6ed3d8d feat: enhance logging by adding dynamic log level configuration and performance metrics 2025-12-30 16:43:24 +08:00
myhloli
466b85ba3f refactor: remove unused import and update help text for device mode option 2025-12-30 11:12:46 +08:00
myhloli
6f9ef69b34 fix: update Dockerfile to clarify GPU architecture compatibility and specify mineru version 2025-12-30 11:02:22 +08:00
myhloli
55f6731aa6 docs: add usage tip for switching model source in README_zh-CN.md 2025-12-30 10:48:43 +08:00
myhloli
edf422b4f7 fix: update hardware configuration requirement wording in documentation 2025-12-30 10:41:30 +08:00
Xiaomeng Zhao
22c7a84c19 更新 index.md 2025-12-30 03:31:37 +08:00
Xiaomeng Zhao
b02f60d772 Update 'High Config Requirements' to 'High Hardware Requirements' 2025-12-30 02:35:44 +08:00
Xiaomeng Zhao
f9f00cd2ee Merge branch 'opendatalab:dev' into dev 2025-12-30 02:29:35 +08:00
myhloli
f7fc7bd928 refactor: fix formatting in hybrid document analysis function 2025-12-29 19:05:33 +08:00
myhloli
037e5f2460 refactor: enhance hybrid backend logic and improve span processing 2025-12-29 19:03:30 +08:00
myhloli
7750d864ed refactor: streamline OCR processing and enable VLM OCR configuration 2025-12-29 17:09:08 +08:00
myhloli
997aab7c55 fix: enhance caption handling to include images and improve gap detection logic 2025-12-28 22:23:39 +08:00
myhloli
190b4ea472 fix: extend special handling for captions and footnotes to include images 2025-12-28 20:08:17 +08:00
myhloli
0dd4c4c4e4 refactor: improve CJK language handling and hyphen management in text processing 2025-12-27 01:00:50 +08:00
myhloli
e54e0c3001 fix: correct hyphen handling based on next line's span case 2025-12-27 00:25:16 +08:00
myhloli
b1aefabbea feat: add bbox_center_distance function and refactor category tying by index 2025-12-26 23:39:57 +08:00
myhloli
0032421167 docs: update memory and disk space requirements in README files for clarity 2025-12-26 17:25:47 +08:00
myhloli
efc428115c refactor: remove vllm engine references and streamline backend choice handling in Gradio app 2025-12-26 16:50:53 +08:00
myhloli
661aebdb2b docs: update Dockerfile comments for GPU architecture compatibility 2025-12-26 16:38:11 +08:00
myhloli
3c4334a37f docs: update installation instructions for lightweight client with vlm-http-client and hybrid-http-client modes 2025-12-26 15:57:46 +08:00
myhloli
5f751d44fb refactor: enhance hyphen handling in text processing for western contexts 2025-12-26 15:46:29 +08:00
myhloli
7496def7a5 fix: update changelog links in README files for accuracy 2025-12-26 14:17:34 +08:00
myhloli
deba6a991f feat: add changelog section to documentation and create changelog file 2025-12-26 11:56:17 +08:00
myhloli
9a355fca02 refactor: remove unnecessary environment variable from Docker run command in VastAI.md 2025-12-26 10:47:56 +08:00
myhloli
bf61e022d8 refactor: update backend options and enhance documentation for hybrid parsing methods 2025-12-25 19:32:06 +08:00
myhloli
984b303dfa refactor: update default backend to hybrid-auto-engine and enhance documentation for parsing options 2025-12-25 19:17:08 +08:00
myhloli
b2c126ef8a refactor: update comments for clarity in hybrid_model_output_to_middle_json.py 2025-12-25 15:46:24 +08:00
myhloli
136cc2fc3b refactor: remove redundant GPU memory allocation message in pipeline_analyze.py 2025-12-25 14:20:58 +08:00
Xiaomeng Zhao
f4eb59c954 Merge pull request #4244 from myhloli/dev
Dev
2025-12-25 01:19:54 +08:00
myhloli
726e0de6fe refactor: simplify backend choices in client.py 2025-12-24 17:28:01 +08:00
myhloli
edd1656851 refactor: streamline backend choice handling in update_interface function 2025-12-24 17:21:49 +08:00
myhloli
7f6f7d9d97 refactor: improve Markdown and API handling in gradio_app.py 2025-12-24 17:07:28 +08:00
myhloli
5f516ea7dc refactor: add support for hybrid backend in parse directory structure 2025-12-24 16:38:01 +08:00
myhloli
eeea4f38e3 refactor: update GPU model support information in Docker deployment documentation 2025-12-24 14:56:41 +08:00
myhloli
88822c7918 refactor: remove unused GPU capability checks and simplify batch ratio calculation 2025-12-24 14:48:28 +08:00
myhloli
0e4c9aee00 refactor: enhance batch ratio calculation based on GPU compute capability 2025-12-24 14:00:13 +08:00
Xiaomeng Zhao
76f7f778cd Merge pull request #4241 from myhloli/dev
Dev
2025-12-24 02:11:49 +08:00
Xiaomeng Zhao
41d5b4843a Merge branch 'opendatalab:dev' into dev 2025-12-24 02:08:48 +08:00
myhloli
7b02d8fbf6 remove unuse file 2025-12-23 19:05:55 +08:00
myhloli
6d7d1c3b0c refactor: expand OCR text conditions for category assignment in analysis scripts 2025-12-23 18:36:13 +08:00
myhloli
e2a06bbb0a refactor: add environment variable check to control pipeline enablement in OCR processing 2025-12-23 18:12:12 +08:00
myhloli
6cecafd99d refactor: add environment variable check to control pipeline enablement in OCR processing 2025-12-23 18:11:24 +08:00
myhloli
a58cb06a6d refactor: update batch ratio documentation for clarity and adjust memory thresholds 2025-12-23 18:01:25 +08:00
myhloli
361f949d4a refactor: update batch ratio documentation for clarity and adjust memory thresholds 2025-12-23 17:59:58 +08:00
myhloli
914770d651 refactor: enhance batch ratio documentation and adjust GPU memory thresholds 2025-12-23 17:40:06 +08:00
myhloli
447ffcd32f refactor: implement dynamic batch ratio based on GPU memory and environment variable 2025-12-23 17:00:02 +08:00
Xiaomeng Zhao
135eaf0c4f Merge pull request #4239 from myhloli/dev
Dev
2025-12-23 16:32:17 +08:00
myhloli
408d94ed58 refactor: add VastAI support documentation to index and create new VastAI.md file 2025-12-23 16:31:12 +08:00
myhloli
e48e1619f9 refactor: improve language support descriptions in OCR input options 2025-12-23 11:32:42 +08:00