Metadata-Version: 2.1
Name: brdata-rag-tools
Version: 0.1.5.1
Summary: Improve development of retrieval augmented generation (RAG) applications at the BR AI + Automation Lab.
Author-email: Marco Lehner <marco.lehner@br.de>
Project-URL: Homepage, https://github.com/br-data/rag-tools-library
Project-URL: Issues, https://github.com/br-data/rag-tools-library/issues
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: aenum>=3.1.15
Requires-Dist: annotated-types>=0.6.0
Requires-Dist: anyio>=4.2.0
Requires-Dist: certifi<=2023.11.17
Requires-Dist: charset-normalizer<=3.3.2
Requires-Dist: distro<=1.9.0
Requires-Dist: exceptiongroup<=1.2.0
Requires-Dist: faiss-cpu<=1.7.4
Requires-Dist: greenlet<=3.0.3
Requires-Dist: h11<=0.14.0
Requires-Dist: httpcore<=1.0.2
Requires-Dist: httpx<=0.26.0
Requires-Dist: idna<=3.6
Requires-Dist: numpy<=1.26.3
Requires-Dist: openai<=1.7.0
Requires-Dist: pgvector>=0.2.4
Requires-Dist: psycopg2-binary<=2.9.9
Requires-Dist: pydantic<=2.5.3
Requires-Dist: pydantic_core<=2.14.6
Requires-Dist: regex<=2023.12.25
Requires-Dist: requests<=2.31.0
Requires-Dist: sniffio<=1.3.0
Requires-Dist: SQLAlchemy<=2.0.25
Requires-Dist: tiktoken<=0.5.2
Requires-Dist: tqdm<=4.66.1
Requires-Dist: typing_extensions<=4.9.0
Requires-Dist: urllib3<=2.1.0

# rag-tools-library
Library to support common tasks in retrieval augmented generation (RAG).

This library is in a very early stage and all the documentation is AI generated.

## Tutorial and Documentation

You find a brief tutorial and the documentation under [br-data.github.io/rag-tools-library](https://br-data.github.io/rag-tools-library/).

## Roadmap

- [ ] Add Google Bison to available LLMs
- [x] Add an offline database alternative
  - [x] FAISS and SQLite
- [x] Allow users to register their own LLMs 
- [x] Allow users to register their own Embedding models
- [ ] Support Semantic Scholar endpoint to generate embeddings for scientific papers.
- [x] Support chat functionality; e.g. let the user give feedback on the result to the LLM.

# Deployment

Run the `build_and_deploy.sh` script in the root folder. Once prompted for the username, pass `__token__` and the pypi API 
token you've received. If you don't have an API token and feel like you should, feel free to contact the maintainers.

# Contact

Marco Lehner

[marco.lehner@br.de](mailto:marco.lehner@br.de)
