v0.2.1
microsoft/graphragv0.2.1Aug 6, 2024by AlonsoGuevara
AI Summary
A patch release addressing stability and configuration issues. Fixes include JSON parsing errors, vector store defaults, and community report generation. It also adds configuration options like encoding models and history tracking.
Key Highlights
- Fixed Local Search breaking when loading Embeddings input
- Fixed missing community reports and refactored community context builder
- Added encoding model support to entity/claim extraction and text chunking configs
- Added history tracking to the LLM with user prompts
- Fixed a bug that erased the vector database and added a config file path parameter
New Features
- Added default columns for vector store at create_pipeline_config
- Added encoding-model to entity/claim extraction config
- Added encoding-model to text chunking config
- Added user prompt to history-tracking llm
Full Release Notes
Download from [pypi](https://pypi.org/project/graphrag/0.2.1/) ## What's Changed - patch: Added default columns for vector store at create_pipeline_config. No change for other cases. - patch: Change json parsing error in the map step of global search to warning - patch: Fix Local Search breaking when loading Embeddings input. Defaulting overwrite to True as in the rest of the vector store config - patch: Fix json parsing when LLM returns faulty responses - patch: Fix missing community reports and refactor community context builder - patch: Fixed a bug that erased the vector database, added a new parameter to specify the config file path, and updated the documentation accordingly. - patch: Try parsing json before even repairing - patch: Update Prompt Tuning meta prompts with finer examples - patch: Update default entity extraction and gleaning prompts to reduce hallucinations - patch: add encoding-model to entity/claim extraction config - patch: add encoding-model to text chunking config - patch: add user prompt to history-tracking llm - patch: update config reader to allow for zero gleans - patch: update config-reader to allow for empty chunk-by arrays - patch: update history-tracking LLm to use 'assistant' instead of 'system' in output history. - patch: use history argument in hash key computation; add history input to cache data **Full Changelog**: https://github.com/microsoft/graphrag/compare/v0.2.0...v0.2.1