PDF-Chat-Mistral7B-RAG

Document pdf based Question-Answering With RAG+Mistral

This system integrates the Retrieval-Augmented Generation (RAG) framework with the Mistral model to deliver advanced document question-answering capabilities. RAG combines the strengths of information retrieval and generative modeling, allowing it to fetch relevant context from PDF documents and generate precise answers. Mistral, a state-of-the-art language model, powers the generative component, ensuring high-quality responses. The system is designed for scenarios where users need fast, accurate answers from extensive documents, making it ideal for research, legal, or educational purposes.

Instructions to Run Document PDF-Based Question-Answering with RAG + Mistral:

1. Prerequisites:

Python (>=3.8)
GPU (optional but recommended for faster inference)
Libraries: PyTorch, Hugging Face Transformers, FAISS, PyMuPDF (for PDF processing)

2. Setup Environment:

Install required dependencies:
pip install torch transformers faiss-cpu pymupdf sentence-transformers

3.If using a GPU, install GPU-specific PyTorch:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

4. Clone the Repository:

Clone or download the project files:
git clone https://github.com/ravigithubshankar/PDF-Chat-Mistral7B-RAG.git
cd PDF-Chat-Mistral7B-RAG

5. Start the RAG + Mistral question-answering system::

  python3 app.py

6.Output:

The system will retrieve relevant document excerpts, generate a contextually appropriate answer, and display it in the console or API response.