Sci-BOt: Driving Scientific Optimization using Natural Language

A monorepo containing a Next.js frontend and FastAPI backend for scientific optimization using Bayesian optimization and active learning, enhanced by LLMs.

Project Structure

sci-llm/
├── frontend/                # Next.js frontend application
│   ├── src/                # Source code
│   │   ├── app/           # Next.js app router pages
│   │   ├── components/    # React components
│   │   ├── lib/           # Utilities and API client
│   │   ├── hooks/         # React hooks
│   │   └── __tests__/     # Frontend tests
│   ├── public/            # Static assets
│   └── package.json       # Frontend dependencies
├── backend/               # FastAPI backend application
│   ├── app/              # Main application
│   │   ├── api/          # API routes
│   │   ├── core/         # Core business logic
│   │   ├── schemas/      # Pydantic schemas
│   │   └── utils/        # Utility functions
│   ├── tests/            # Backend tests
│   ├── main.py           # Application entry point
│   └── requirements.txt   # Python dependencies
├── dataset/              # Sample datasets and configurations
│   ├── data_format_example.csv  # Example data format
│   ├── json_example.json        # Example configuration
│   └── test_problems.txt        # Test problem descriptions
├── active_learning/      # Active learning module
├── pyproject.toml        # Python project configuration
└── package.json          # Root dependencies

Prerequisites

• Node.js 18+ and npm
• Python 3.9+
• Poetry (optional) or pip

Setup

1. Install root dependencies:

   npm install

2. Set up backend:

   cd backend
   python -m venv venv
   source venv/bin/activate  # On Windows: .\venv\Scripts\activate
   pip install -r requirements.txt

3. Set up frontend:

   cd frontend
   npm install

4. Create .env files:

   • Backend (.env):

     OPENAI_API_KEY=your_key_here
     

   • Frontend (.env.local):

     NEXT_PUBLIC_API_URL=http://localhost:8000
     

Development

Run both frontend and backend in development mode:

npm run dev

Or run them separately:

• Frontend only: npm run frontend
• Backend only: npm run backend

Features

• Modern Next.js frontend with TypeScript and Tailwind CSS
• FastAPI backend with async support
• Active Learning and Bayesian Optimization
• LLM integration for configuration generation
• Real-time data visualization
• Type-safe API communication

Testing

• Backend: cd backend && pytest
• Frontend: cd frontend && npm test

Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request