README.md

---
title: NASA Space Apps - Exoplanet Classification
emoji: 🪐
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.49.0
app_file: app.py
pinned: false
short_description: Exoplanet classification using NASA Kepler/TESS data
license: apache-2.0
---

# 🪐 NASA Space Apps - Exoplanet Classification

This Space provides an API for exoplanet classification using ensemble machine learning models trained on NASA Kepler and TESS mission data.

## Features

- **Ensemble Model**: Uses multiple ML algorithms for robust predictions
- **Preprocessing Pipeline**: Includes feature imputation, scaling, and variance-based selection
- **API Endpoint**: RESTful API for integration with web applications
- **Interactive Interface**: Gradio-based UI for testing predictions

## Usage

### Web Interface
Use the interface above to input comma-separated feature values and get predictions.

### API Endpoint
Send POST requests to `/api/predict` with JSON data:

```json
{
  "data": ["1.2,3.4,5.6,7.8,9.1,2.3,4.5,6.7"]
}
```

## Model Details

The model uses an ensemble approach combining multiple algorithms and includes:
- Feature imputation for handling missing values
- StandardScaler for feature normalization
- VarianceThreshold for feature selection
- Ensemble classifier for final predictions

## Data Source

Models trained on NASA Kepler and TESS exoplanet datasets with features including:
- Stellar properties
- Orbital characteristics  
- Transit photometry measurements
- Statistical derived features