Basic RAG Chatbot (PDF upload)

This project is a simple Streamlit-based RAG (Retrieval-Augmented Generation) chatbot. You upload PDF files, it embeds text using OpenAI embeddings, stores vectors in FAISS, and answers questions with OpenAI responses.

Requirements

• Python 3.11 (strongly recommended for binary wheel compatibility on Windows)
• OpenAI API key

Setup

1. Create and activate virtualenv:

cd "C:\Users\sande\Desktop\VSCode Projects\RAG Chatbot"
py -3.11 -m venv .venv
Set-ExecutionPolicy Bypass -Scope Process -Force
. .\.venv\Scripts\Activate.ps1

2. Install dependencies:

python -m pip install --upgrade pip setuptools wheel
python -m pip install -r requirements.txt

3. Run app:

python -m streamlit run app.py --server.port 8501 --server.headless true

4. In the browser:

• Enter OPENAI_API_KEY in sidebar
• Upload PDF(s)
• Ask questions

Notes

• Works best with searchable text PDFs (not pure scanned images). For OCR PDFs, pre-process with OCR first.
• For production, add persistent storage and error handling.