📄 RAG System with Query Expansion (Ollama + ChromaDB)

This project is a Retrieval-Augmented Generation (RAG) application built with Streamlit, Ollama, and ChromaDB.
It allows users to upload PDF documents, store them as embeddings, expand user queries using an LLM, and generate citation-backed answers.

---

🚀 Features

• 📂 Upload and process PDF documents
• ✂️ Chunking with RecursiveCharacterTextSplitter
• 🔎 Query Expansion using LLM
• 📚 Vector storage with ChromaDB (persistent)
• 🤖 Answer generation with citations
• 🧠 Local LLM inference using Ollama
• 🖥️ Interactive Streamlit UI

---

🛠️ Tech Stack

• Python
• Streamlit
• LangChain
• Ollama
• ChromaDB
• all-MiniLM embeddings
• Gemma / LLaMA models

---

📂 Project Structure

.
├── app.py
├── data/ # Uploaded PDFs
├── chromadb/ # Persistent vector store
├── requirements.txt
└── README.md

---

⚙️ Prerequisites

1️⃣ Install Ollama Download and install Ollama:
👉 https://ollama.com

Pull required models:

ollama pull gemma3
ollama pull all-minilm

2️⃣ Install dependencies

pip install -r requirements.txt

▶️ How to Run the App

streamlit run app.py

🧪 How to Use

Upload PDF files from the sidebar

Click Initialize System

Click Process Documents

Enter a question

Click Search

View answers with citations and sources