- 03 Mar, 2025 13 commits
-
-
myhloli authored
- Upgrade twine and install additional dependencies - Add id, keyring, packaging, readme-renderer, requests, requests-toolbelt, rfc3986, rich, and urllib3 to ensure successful package publishing
-
myhloli authored
- Update README.md and README_zh-CN.md with the latest changes - Add details about bug fixes in version1.2.1 - Include improvements for full-width to half-width conversion, caption matching, and formula span issues
-
myhloli authored
- Increase batch ratio to 8 for GPU memory >=16GB - Improve inference performance on systems with higher GPU memory
-
Xiaomeng Zhao authored
fix: caption match
-
Xiaomeng Zhao authored
refactor(pre_proc): allow interline equations to be associated with text blocks
-
myhloli authored
- Update OCR dictionary merge logic to include text blocks when processing interline equations - This change improves the handling of equations that may be embedded within text content
-
icecraft authored
-
Xiaomeng Zhao authored
perf(mfr): improve Math Formula Recognition by sorting images by area
-
myhloli authored
- Simplify batch ratio logic for GPU memory >= 16GB - Remove unnecessary conditions for 20GB and 40GB memory
-
myhloli authored
- Simplify batch ratio logic for GPU memory >= 16GB - Remove unnecessary conditions for 20GB and 40GB memory
-
myhloli authored
- Sort detected images by area before processing to enhance MFR accuracy - Implement stable sorting to maintain original order of images with equal
-
myhloli authored
- Comment out @measure_time decorator for txt_spans_extract_v2 and sort_lines_by_model functions - Remove logger.info for page_process_time - Comment out PerformanceStats.print_stats call
-
myhloli authored
- Add performance_stats module to measure and print execution time statistics - Implement measure_time decorator to track execution time of key functions - Remove multi-threading in pdf parsing for better resource management - Optimize pdf parsing logic for improved performance
-
- 28 Feb, 2025 1 commit
-
-
myhloli authored
- Add ThreadPoolExecutor to process PDF pages in parallel - Create separate function for page processing to improve readability and maintainability - Include error handling for individual page processing tasks - Log total page processing time for performance monitoring
-
- 27 Feb, 2025 2 commits
-
-
Xiaomeng Zhao authored
refactor(ocr_mkcontent): optimize full-width character handling
-
myhloli authored
- Update condition to only convert full-width letters and numbers - Remove separate case for full-width space
-
- 26 Feb, 2025 5 commits
-
-
Xiaomeng Zhao authored
-
icecraft authored
-
Xiaomeng Zhao authored
refactor(magic_pdf): remove bfloat16 support checks and usage
-
myhloli authored
- Replace complex device selection logic with a single line using torch.device - Remove redundant checks and imports for better readability and maintainability
-
myhloli authored
- Remove supports_bfloat16 variable and related checks - Remove model.bfloat16() call for LayoutLMv3ForTokenClassification - Simplify device selection logic
-
- 25 Feb, 2025 5 commits
-
-
Xiaomeng Zhao authored
perf(model): optimize batch analyze process
-
myhloli authored
- 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
-
myhloli authored
- Move batch model initialization outside the loop - Collect page dimensions before analyzing- Update page info dictionary structure - Add null dimensions for non-analyzed pages
-
Xiaomeng Zhao authored
docs(ascend): update Ascend NPU acceleration documentation
-
myhloli authored
- 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
-
- 24 Feb, 2025 11 commits
-
-
Xiaomeng Zhao authored
docs(README): update release notes for version 1.2.0
-
myhloli authored
- 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
-
myhloli authored
-
myhloli authored
- 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
-
Xiaomeng Zhao authored
feat(pre_proc): add block type compatibility check for span allocation
-
myhloli authored
- 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
-
Xiaomeng Zhao authored
fix(llm_aided): update prompt
-
myhloli authored
-
myhloli authored
- 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
-
Xiaomeng Zhao authored
fix #1747
-
Xiaomeng Zhao authored
Updata ext.py is_pdf function to support the pdf with Chinese characters and special characters
-
- 23 Feb, 2025 2 commits
-
-
Xiaomeng Zhao authored
chore(magic_pdf): enhance license logging information
-
myhloli authored
- Add license ID information to the log for better traceability - Improve logging format to include both license ID and expiration date
-
- 22 Feb, 2025 1 commit
-
-
Nathan Dahlberg authored
-