This repository provides a straightforward tutorial for building a Retrieval-Augmented Generation (RAG) application using Langchain. It is designed for developers and researchers looking to implement custom document Q&A systems, offering a practical guide to integrating document loading, embedding, vector storage, and LLM querying.

How It Works

The application leverages Langchain's orchestration capabilities to build a RAG pipeline. It processes documents, generates embeddings using an LLM, stores these embeddings in a ChromaDB vector store, and then retrieves relevant document chunks to augment LLM prompts for contextually accurate answers.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Install markdown dependencies: pip install "unstructured[md]"
• Create database: python create_database.py
• Query database: python query_data.py "How does Alice meet the Mad Hatter?"
• Requires an OpenAI API key set as an environment variable.
• MacOS users may need to install onnxruntime via conda install onnxruntime -c conda-forge.
• Windows users require Microsoft C++ Build Tools.
• Official tutorial video: RAG+Langchain Python Project: Easy AI/Chat For Your Docs

Highlighted Details

• Demonstrates a complete RAG workflow from data ingestion to querying.
• Utilizes ChromaDB for efficient vector storage and retrieval.
• Integrates with OpenAI for LLM capabilities and embeddings.

Maintenance & Community

No specific information on maintainers, community channels, or roadmap is provided in the README.

Licensing & Compatibility

The repository does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The setup process includes platform-specific workarounds for dependency installation, indicating potential fragility. The reliance on OpenAI API keys limits its use to those with an OpenAI account.