https://github.com/laavanjan/rag-document-q-a-with-ollama-and-qwen-2.5
This project is a Retrieval-Augmented Generation (RAG) based Document Q&A system
https://github.com/laavanjan/rag-document-q-a-with-ollama-and-qwen-2.5
Last synced: 7 months ago
JSON representation
This project is a Retrieval-Augmented Generation (RAG) based Document Q&A system
- Host: GitHub
- URL: https://github.com/laavanjan/rag-document-q-a-with-ollama-and-qwen-2.5
- Owner: laavanjan
- License: gpl-3.0
- Created: 2025-02-15T18:13:56.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-15T18:24:56.000Z (8 months ago)
- Last Synced: 2025-02-15T19:26:42.639Z (8 months ago)
- Language: Python
- Size: 2.4 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# RAG Document Q&A with Ollama and Qwen-2.5
## 📌 Overview
This project is a **Retrieval-Augmented Generation (RAG) based Document Q&A system** that leverages:
- **Ollama Embeddings** for vector representations
- **FAISS** for efficient vector search
- **LangChain** for chaining retrieval and generation steps
- **Groq's Qwen-2.5-32B** model for accurate text generation
- **Streamlit** for a user-friendly interface
With this system, users can **upload research papers in PDF format** and ask questions about their content. The model retrieves relevant chunks from the documents and generates precise responses.
---
📸 Example Usage
| Image 1 | Image 2 | Image 3 |
|---------|---------|---------|
|  |  |  |
| Image 4 | Image 5 | Image 6 |
|  |  |  |
## 🚀 Features
✅ Load and process research papers dynamically\
✅ Generate vector embeddings using **Ollama**\
✅ Efficient similarity search with **FAISS**\
✅ Retrieve document chunks relevant to the query\
✅ Generate accurate responses using **Qwen-2.5-32B**\
✅ Streamlit UI for interactive Q&A\
✅ Real-time response time tracking
---
## 📦 Tech Stack
- **Python 3.10**
- **LangChain**
- **FAISS**
- **Ollama Embeddings**
- **Streamlit**
- **Groq API (Qwen-2.5-32B)**
- **PyPDFLoader**
---
## 🔧 Installation & Setup
### 1️⃣ Clone the Repository
```bash
git clone https://github.com/laavanjan/RAG-Document-Q-A-With-Ollama-And-qwen-2.5.git
cd RAG-Document-Q-A-With-Ollama-And-qwen-2.5
```
### 2️⃣ Create a Virtual Environment (Optional but Recommended)
```bash
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
```
### 3️⃣ Install Dependencies
```bash
pip install -r requirements.txt
```
### 4️⃣ Set Up Environment Variables
Create a `.env` file in the project directory and add your **Groq API Key**:
```env
GROQ_KEY=your_api_key_here
```
### 5️⃣ Run the Streamlit App
```bash
streamlit run app.py
```
---
## 📂 Project Structure
```
├── research_papers/ # Folder containing PDF documents
├── app.py # Main Streamlit application
├── requirements.txt # Required dependencies
├── .env # Environment variables (not shared)
└── README.md # This documentation
```
---
## 🎯 How It Works
1️⃣ Upload research papers (PDFs are automatically processed).\
2️⃣ The system extracts text and generates **vector embeddings**.\
3️⃣ User submits a question via Streamlit UI.\
4️⃣ The **retriever fetches the most relevant document chunks**.\
5️⃣ Qwen-2.5-32B generates an answer **based on the retrieved context**.\
6️⃣ Users can explore document similarity search results.
---
## 📸 Screenshots
---
## 🛠 Future Enhancements
🔹 Add support for more document formats (TXT, DOCX)\
🔹 Improve response quality with fine-tuned embeddings\
🔹 Deploy as a web service with **FastAPI**\
🔹 Store and retrieve historical queries for analysis
---
## 🤝 Contribution
Contributions are welcome! Feel free to submit issues and pull requests.
---
## 📜 License
This project is licensed under the **GPL License**.
---
⭐ **If you like this project, don't forget to star the repository!** ⭐