Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/najmaelboutaheri/scientific-chatboot-generator
This project demonstrates a Streamlit-based web application for knowledge retrieval using a Retrieval-Augmented Generation (RAG) approach.
https://github.com/najmaelboutaheri/scientific-chatboot-generator
Last synced: 17 days ago
JSON representation
This project demonstrates a Streamlit-based web application for knowledge retrieval using a Retrieval-Augmented Generation (RAG) approach.
- Host: GitHub
- URL: https://github.com/najmaelboutaheri/scientific-chatboot-generator
- Owner: najmaelboutaheri
- License: mit
- Created: 2024-10-07T16:28:06.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2024-12-26T09:26:50.000Z (29 days ago)
- Last Synced: 2024-12-26T10:19:52.522Z (29 days ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 13.1 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: License
Awesome Lists containing this project
README
# RAG-Based Scientific Chatbot
This project implements a Retrieval-Augmented Generation (RAG) Scientific Chatbot that provides insightful and contextually relevant responses to queries based on scientific documents. The chatbot leverages a language model to retrieve and synthesize information from a pre-constructed vector database of scientific knowledge.
## Features
- **Retrieval-Augmented Generation (RAG)**: Combines information from a set of relevant documents to answer questions.
- **Experiment Tracking**: Follow ongoing experiments and log updates.
## Architecture
- **Frontend**: User interface for interaction.
- **Backend**: Handles data retrieval and processing using RAG architecture.
- **Database**: Stores scientific documents and experiment data.
- **Deployment**: The Streamlit app is deployed on an **Amazon EC2 instance** for scalable access.
## Requirements
- Python 3.7 or higher
- Streamlit (for the frontend)
- Transformers (for the language model)
- FAISS (for efficient similarity search)
- EC2 Gpu based instance to deploy the app and interact with the chatbot
- Install Nvidia
## Setup within EC2 insatnce
1. Clone the repository:
```bash
git clone https://github.com/your_username/rag-scientific-chatbot.git
cd rag-scientific-chatbot
```
2. Create a virtual environment (optional but recommended):
```bash
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`
```
3. Install the required packages:
```bash
pip install -r requirements.txt
```
4. Set up the environment variables if necessary (specify any required environment variables).
## Steps Taken throughtout the deployment phase:
1. Find the EC2 Instance’s Public DNS or IP Address and verify that the type is **g4dn.xlarge**
2. Connect to the EC2 Instance via SSH
```bash
ssh -i /path/to/your-key.pem ec2-user@your-public-dns
```
3. Accept the Security Warning : yes
4. You may need to set Permissions for the Key Pair as follows:
```bash
chmod 400 /path/to/your-key.pem
```
5. You need to install CUDA Toolkit to utilize GPU resources for PyTorch.
```bash
nvcc --version # Check if nvcc is installed
sudo apt install nvidia-cuda-toolkit # Install CUDA Toolkit
```
6. After cloning the Project Repository inside the EC2 instance and installing the required packages you need to run the streamlit app file ```streamlit run app.py``` and acess the **ngrok tunells** links or you can sart the application with **EC2 instance's public IP address**
7. Interact with the chatbot through the provided interface.
8. The retrieved documents:
## Contributing
If you'd like to contribute to this project, please fork the repository and submit a pull request. For major changes, please open an issue first to discuss what you would like to change.
1. Fork the Project
2. Create your Feature Branch (`git checkout -b feature/YourFeature`)
3. Commit your Changes (`git commit -m 'Add some feature'`)
4. Push to the Branch (`git push origin feature/YourFeature`)
5. Open a Pull Request
## Contact
For further information or questions, feel free to reach out:
- **Email**: [email protected]