README.md

RAG Pipeline with LangChain and Google Generative AI

This project demonstrates a Retrieval-Augmented Generation (RAG) pipeline using LangChain, ChromaDB, and Google Generative AI. It scrapes course content from a website, splits the text, embeds it, and enables question-answering over the retrieved context.

Features

• Web scraping using LangChain Community's WebBaseLoader
• Document splitting with RecursiveCharacterTextSplitter
• Embedding generation via Google Generative AI
• Vector storage and retrieval using ChromaDB
• RAG pipeline for concise question answering
• Prompt debugging with custom print function

Setup

1. Install dependencies:

   pip install langchain_community langchainhub chromadb langchain langchain-google-genai langchain-openai

2. Set your Google API key:

   • If using Google Colab, store your key in Colab's userdata.
   • Otherwise, set GOOGLE_API_KEY in your environment.

Usage

Open RAG.ipynb and run the cells sequentially:

• Scrape course data from https://www.educosys.com/course/genai
• Split and embed documents
• Store embeddings in ChromaDB
• Query the RAG pipeline with your questions

Example

rag_chain.invoke("Is there any free courses?")

Project Structure

• RAG.ipynb: Main notebook containing all code for scraping, embedding, and RAG pipeline.

License

This project is for