Releases: SciPhi-AI/R2R
v0.2.11
up (#405) * up * bump package
v0.2.10
What's New in the Latest Release
User & Document Management Enhancements
- Enhanced document management with synchronized document info, efficient document chunk fetching, and revamped document ingestion.
- Added new endpoints for deleting users, deleting documents, and updating documents to streamline management.
Hybrid Search Implementation
- Introduced hybrid search combining semantic and keyword search for improved accuracy and relevance over ingested files.
Analytics Integration
- Added comprehensive analytics to monitor and analyze application performance and usage.
Standardized API Endpoints
- Standardized and cleaned up API endpoints to ensure consistency and reliability across the application.
v0.2.04
What's Changed
- Added and patched config serialization and loader.
- Built the builder and added R2R builder.
- Fixed streaming issues.
- Added ingestion version updates and multi-search feature.
- Refreshed documentation and README.
- Added R2R app architecture diagrams.
- Fixed image links for Vercel deployment.
- Refactored logging and run ID management.
- Cleaned up images in documentation.
- Handled errors for local development.
Full Changelog: v0.2.03...v0.2.04
v0.2.03
Summary:
This pull iteration implements minor fixes including boolean value corrections, document ID updates, removal of unused config parameters, and enhancements in error handling and logging across the R2R project.
Key points:
- Corrected boolean values for streaming in various bash commands.
- Updated document IDs in example scripts for consistency.
- Removed sampling_fraction from configuration files.
- Enhanced error handling and logging in pipeline and pipe classes.
- Introduced async context managers for better run ID and run information management.
v0.2.02
Minor changes to v0.2.01 which had the following release notes -
Summary:
This pull iteration implements minor fixes including boolean value corrections, document ID updates, removal of unused config parameters, and enhancements in error handling and logging across the R2R project.
Key points:
- Corrected boolean values for streaming in various bash commands.
- Updated document IDs in example scripts for consistency.
- Removed sampling_fraction from configuration files.
- Enhanced error handling and logging in pipeline and pipe classes.
- Introduced async context managers for better run ID and run information management.
v0.2.01
Summary:
This pull iteration implements minor fixes including boolean value corrections, document ID updates, removal of unused config parameters, and enhancements in error handling and logging across the R2R project.
Key points:
- Corrected boolean values for streaming in various bash commands.
- Updated document IDs in example scripts for consistency.
- Removed sampling_fraction from configuration files.
- Enhanced error handling and logging in pipeline and pipe classes.
- Introduced async context managers for better run ID and run information management.
V2 Framework Overhaul
About
This commit represents a major overhaul of the R2R framework. Many changes have been made to improve modularity and default capabilities.
Key Features
- 🔧 Build: Use the framework to build arbitrary asynchronous pipelines.
- 🚀 Deploy: Instantly launch production-ready asynchronous RAG pipelines with streaming capabilities.
- 🧩 Customize: Tailor your multimodal pipeline with intuitive configuration files.
- 🔌 Extend: Enhance your pipeline with custom code integrations.
- 🤖 OSS: Benefit from a framework developed by the open-source community, designed to simplify RAG deployment.
v0.1.35
Feature/web search pipeline (#293) * web search pipeline * web search pipeline * re-merge logic
v0.1.34
update lock file
v0.1.33
Update pyproject.toml