- Implemented 'Bring Your Own Model' (BYOM) support for NER, Relation, and Triplet extraction
- Added NER aggregation strategies (simple, max, average)
- Implemented Relation Extraction via Sequence Classification with entity markers
- Enhanced Triplet Extraction with REBEL post-processing and lazy loading
- Updated all extractors to prioritize runtime options over config defaults
- Added extensive tests and examples (huggingface_demo.py)
- Updated documentation and CHANGELOG
- Implemented high-throughput parallel batch processing across all core extractors (NERExtractor, RelationExtractor, TripletExtractor, EventDetector, SemanticNetworkExtractor) using ThreadPoolExecutor.
- Added max_workers configuration parameter (default: 1) to all extractor extract() methods.
- Implemented parallel processing for large document chunking in _extract_entities_chunked and _extract_relations_chunked.
- Enhanced ProgressTracker to be thread-safe.
- Optimized setUpClass in tests to reduce Groq LLM initialization overhead.
- Updated documentation and usage examples.
- Implemented ML/LLM -> Pattern -> Last Resort fallback chains for NER, Relation, and Triplet extractors to prevent empty results.
- Added provenance metadata (batch_index, document_id) to all extraction schemas (Entity, Relation, Triplet, etc.).
- Unified batch processing API with progress tracking across all extractors.
- Updated documentation (module usage and reference docs) to reflect new features.
- Added robustness and batch provenance tests.
- Reduced code examples in all guide pages (getting-started, quickstart, concepts, modules, examples, use-cases, learning-more)
- Added comprehensive cookbook links with descriptions (topics, difficulty, time, use cases)
- Improved structure and organization across all guide pages
- Updated use-cases.md to only include use cases with corresponding cookbooks
- Removed 'Last Updated: 2024' from all documentation files
- Enhanced navigation with better 'Next Steps' sections
BREAKING CHANGE: Removed build() convenience function from semantic_extract module
- Removed build() function from semantic_extract/__init__.py
- Updated __all__ exports to remove 'build'
- Resolved merge conflicts in named_entity_recognizer.py, relation_extractor.py, triple_extractor.py
- Updated semantic_extract_usage.md with class-based examples
- Updated docs/reference/semantic_extract.md with detailed parameter documentation
- Fixed 01_GraphRAG_Complete.ipynb to use individual extractor classes
- Enhanced 05_Entity_Extraction.ipynb with comprehensive examples (9 sections)
- Enhanced 06_Relation_Extraction.ipynb with complete pipeline examples (9 sections)
Users should now use individual classes (NERExtractor, RelationExtractor, TripleExtractor, etc.)
instead of the build() function for better control and flexibility.
Migration guide available in documentation.
- Added grid‑card layouts, icons, and admonitions to all reference pages
- Removed all mkdocstrings ::: directives that caused alias‑resolution errors
- Fixed export module alias issue (RDFExporter) by cleaning the docs
- Updated README‑style sections for better readability
- Added a PowerShell helper script (remove_mkdocstrings.ps1) for future clean‑ups
