Introduction

ExoGen AI is a full-stack platform that combines physics-based validation with deep learning to analyze stellar light curves from NASA's Kepler mission and identify potential exoplanets.

The system integrates a FastAPI backend for model inference and training with a Next.js frontend for data upload, visualization, and result interaction.

ExoGen AI Backend Setup

Quick Start

1. Clone the repository

   git clone https://github.com/alexanderabesteh/ExoGenAI.git
   cd ExoGenAI/backend

2. Install dependencies

   pip install -r requirements.txt

3. Download required files

   • Trained model:
     Download best_model.pt from https://drive.google.com/drive/folders/1O8w9Ls2u_D0gkZ8Trnpj9tayAbRcH3Gc?usp=sharing and place it in:

     backend/checkpoints/best_model.pt
     

   • Dataset:
     Download from https://drive.google.com/drive/folders/1O8w9Ls2u_D0gkZ8Trnpj9tayAbRcH3Gc?usp=sharing.(labelled time-series data) Place these files:

     backend/data/exoTrain.csv
     backend/data/exoTest.csv
     

4. Start API server

   uvicorn api:app --host 0.0.0.0 --port 8000

5. Test the API

   • Health check:
     http://localhost:8000/health
   • Documentation (Swagger):
     http://localhost:8000/docs

API Endpoint

POST /predict-with-plot

• Upload: CSV file with time and flux columns.
• Returns:
  • Prediction
  • Probabilities
  • plot_data

Troubleshooting

• No GPU detected?
  Edit api.py line 38:
  Change device='cuda' to device='cpu'.

• Port conflict?
  Start the server with a different port:

  uvicorn api:app --host 0.0.0.0 --port 8001

---

ExoGen AI Frontend Setup

Quick Start

1. Navigate to the frontend directory

   cd ../frontend

2. Install dependencies

   npm install

3. Start the development server

   npm run dev

4. Open your browser and navigate to http://localhost:3000

5. Upload a CSV file and view results

ExoGen AI Full Stack

Quick Start

Ensure both backend and frontend are set up as per the instructions above.

1. Start the backend server

   cd ../backend
   uvicorn api:app --host 0.0.0.0 --port 8000

2. In a new terminal, start the frontend server

   cd ../frontend
   npm run dev

3. Open your browser and navigate to http://localhost:3000
4. Upload a CSV file and view results