README.md

metadata

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:

{
  "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