RAG Application with Langchain and Streamlit

Short description: Streamlit RAG app using Langchain, Groq API, and HF embeddings over local PDFs. Developer: Dr. Partha Majumder

This Streamlit application implements a Retrieval-Augmented Generation (RAG) system using Langchain, Groq API, and Hugging Face embeddings. It processes PDF documents from a specified directory, creates vector embeddings, and allows users to query the documents for accurate responses.

Features:

• Document Embedding: Converts PDFs into vector embeddings using Hugging Face's all-MiniLM-L6-v2 model.
• Query Interface: Users can input queries to retrieve relevant information from the documents.
• Similarity Search: Displays document chunks with similarity scores for transparency.

Usage:

1. Place PDFs in the data directory.
2. Click "Embed The Documents" to generate embeddings.
3. Enter your query to get responses.

Dependencies: streamlit, langchain, huggingface_hub, faiss, python-dotenv.