GitLab

docs(README): update table recognition default setting and accelerate

docs(README): update table recognition default setting and accelerate options

- Enable table recognition by default in both English and Chinese README files
- Add MPS (Metal Performance Shaders) as an acceleration option for MacOS in the Chinese README
- Update the description of acceleration options in the Chinese README

docs: update readme

feat(README): add MPS acceleration support for Apple silicon

- Simplify and clarify the description of MinerU Desktop
- Make the language more inviting and straightforward for users

- Added information about MPS acceleration support for Apple silicon chips- Updated README files to include MPS-related content and configuration instructions
- Improved language and formatting consistency across both English and Chinese README files

fix(device): enable MPS support and fix related issues

- Add MPS support for Apple Silicon devices
- Implement empty_cache() for MPS devices
- Set PYTORCH_ENABLE_MPS_FALLBACK environment variable
- Adjust MFR model device allocation for MPS

fix(docs): update pip install command in README files

- Correct the release date for version 1.0.0 in both English and Chinese README files
- Ensure consistency across documentation for the official release

feat(header): update homepage link and add client download link 

feat(header): update homepage link and add client download link

- Update homepage link to point to https://mineru.org.cn/home
- Add client download link for users to easily access the app- Change globe icon to home icon for homepage

- Update homepage link to point to https://mineru.org.cn/home
- Add client download link for users to easily access the app- Change globe icon to home icon for homepage

build(docker): update dependencies across all docker images

build(docker): update dependencies across all docker images

- Update fast-langdetect from 0.2.0 to >=0.2.3,<0.3.0
- Specify version for rapid_table: ==0.3.0

Dev

fix: update issue template for version1.0.x and add npu option

- Update the bug report template to include version 1.0.x
- Add 'npu' as a new option for hardware backend

docs: reorganize inference acceleration sections in README 

docs: reorganize inference acceleration sections in README

Reordered sections related to inference acceleration using CUDA and CANN to provide a clearer structure in both English and Chinese README files.

fix(language): enhance language detection and text processing

fix(language): enhance language detection and text processing

- Improve language detection by removing newline characters from the input text
- Add error handling and fallback mechanism to deal with text containing control characters

Dev

feat(README): update for 1.0.0 release and add automatic language identification feature

- Update README.md and README_zh-CN.md for1.0.0 release
- Add information about new automatic language identification feature
- Include link to Ascend NPU Acceleration documentation
- Highlight improvements in API interface and compatibility

- Change fast-langdetect dependency from version 0.2.0 to >=0.2.3

Dev

- Change the default value of the language dropdown from an empty string to 'auto'
- This ensures a language is always selected by default, improving

- Remove conditional logic for OCR engine selection
- Always use RapidOCR as the OCR engine
- Simplify the __init__ method by removing unused code

- Update clear_bu.add() to exclude 'language' input

Dev