PDF Analysis Tool using Retrieval-Augmented Generation (RAG)

This project is designed to perform pdf analysis using a Retrieval-Augmented Generation (RAG) model. The system extracts relevant information from a pdf in PDF format based on user queries. It uses FAISS for efficient vector-based retrieval and Google's Generative AI Model for embedding the content.

Features

• Extracts text content from PDF pdfs.
• Splits pdf content into meaningful chunks.
• Embeds the text chunks using a pre-trained generative model.
• Uses FAISS to build an index of embeddings for fast retrieval.
• Retrieves and displays relevant sections of the pdf based on user queries.

Requirements

To run this project, you will need the following libraries:

• PyPDF2 for PDF text extraction
• langchain for text splitting
• google-generativeai for embeddings from Google's generative model
• torch for handling embeddings as tensors
• faiss for creating a fast search index
• numpy for numerical computations

Install the necessary libraries using pip:

pip install PyPDF2 langchain google-generativeai torch faiss-cpu numpy psutil