GitLab

refactor(magic_pdf): remove bfloat16 support checks and usage

refactor(magic_pdf): remove bfloat16 support checks and usage

- Replace complex device selection logic with a single line using torch.device
- Remove redundant checks and imports for better readability and maintainability

- Remove supports_bfloat16 variable and related checks
- Remove model.bfloat16() call for LayoutLMv3ForTokenClassification
- Simplify device selection logic

perf(model): optimize batch analyze process

perf(model): optimize batch analyze process

- Implement full_to_half function to convert full-width characters to half-width
- Apply conversion to span content before merging paragraphs
- Improve text processing for better readability and consistency

- Move batch model initialization outside the loop
- Collect page dimensions before analyzing- Update page info dictionary structure
- Add null dimensions for non-analyzed pages

docs(ascend): update Ascend NPU acceleration documentation

docs(ascend): update Ascend NPU acceleration documentation

- Add information about OS and CANN version compatibility
- Include details on high-performance mode and its requirements
- Update paddlepaddle-gpu installation instructions for CUDA acceleration
- Remove unnecessary empty line from changelog

docs(README): update release notes for version 1.2.0

docs(README): update release notes for version 1.2.0

- Update English and Chinese README files with the changelog for version 1.2.0
- Include details on performance optimizations, parsing improvements, and bug fixes
- Highlight specific enhancements for PDF document classification, watermark handling, and layout matching

- Update English and Chinese README files with the changelog for version 1.2.0
- Include details on performance optimizations, parsing improvements, and bug fixes
- Highlight specific enhancements for PDF document classification, watermark handling, and layout matching

feat(pre_proc): add block type compatibility check for span allocation

- Introduce span_block_type_compatible function to check compatibility between span and block types
- Update fill_spans_in_blocks function to use the new compatibility check
- Improve accuracy of span allocation to blocks based on content type

fix(llm_aided): update prompt

- Update the logic for determining `end_page_id` to handle negative values
- This change ensures proper behavior when `end_page_id` is set to -1 or other negative values

fix #1747

Updata ext.py is_pdf function to support the pdf with Chinese characters and special characters

chore(magic_pdf): enhance license logging information

- Add license ID information to the log for better traceability
- Improve logging format to include both license ID and expiration date

Determine whether the file name with a.pdf extension supports the inclusion of special characters or Chinese characters.

fix(model): handle import errors and improve exception logging

- Add ImportError handling to silence known import-related exceptions
- Improve generic exception handling to log error messages- Maintain existing specific exception handlers for license-related issues

feat(model_init): implement license verification for Ascend plugin

- Add license verification logic for Ascend plugin
- Handle different license-related exceptions with appropriate error messages
- Log success message with license expiration date if verification passes
- Fall back to CPU model if license verification fails or plugin is not available

refactor(magic_pdf): improve title optimization process

- Update instructions for AI-generated titles optimization
- Use ast.literal_eval() instead of json.loads() for parsing completion content
- Refactor variable names and logging for better code readability- Add error handling for JSON decoding issues

docs(windows): add numpy version limit for CUDA installation

fix: ut

Fix/caption match

- Add numpy version limit (<2.0.0) in the pip installation command for Windows CUDA acceleration
- Update both English and Chinese versions of the README