Skip to content

Latest commit

 

History

History
37 lines (28 loc) · 1.49 KB

README.md

File metadata and controls

37 lines (28 loc) · 1.49 KB
Raw

LangChain RAG Pattern Demo (React, FastAPI, Pinecone DB Vectorstore)

Windows 11 & Python 3.11.0 & Node.js v16.20.2

Overview

This repository contains a demo showcasing the implementation of the RAG (Retrieval Augmented Generation) pattern using Pinecone DB and LangChain. The RAG pattern combines retrieval-based and generative-based approaches to natural language processing, enhancing text generation capabilities.

Key Features

  • Implemented routing
  • Implemented multi queries
  • Implemented query translation : stepback, rewriting, HyDE, sub-query decomposition
  • Implemented adaptive retrieval
  • Implemented reranking
  • Implemented Self-RAG
  • Implemented CRAG
  • Trying to make raptor ..

Requirements

  • Python environment with LangChain installed.
  • Basic knowledge of Vectorstore and natural language processing concepts.

Usage

  1. Follow the steps provided in the README file.

Steps

  1. Step 1 - Create FastAPI to integrate LangChain RAG pattern with web front-end.
  2. Step 2 - Build the React web front-end to ask 'grounded' questions of your data and view relevant documents.
  3. Follow the setup instructions provided in the README file.
  4. Run the demo application and explore the RAG pattern in action.

License

This project is licensed under the MIT License, granting permission for commercial and non-commercial use with proper attribution.

Disclaimer

This demo application is provided for educational and demonstration purposes only. Use at your own risk.