Documentation
ExoBengal Documentation - ML-Powered Exoplanet Detection
A comprehensive machine learning toolkit for exoplanet detection using NASA Kepler mission data. Train and deploy Random Forest, CNN, k-Nearest Neighbors, and Decision Tree models for planet classification.
Install via pip:
Terminal
$
Quick Start
Get up and running with ExoBengal in minutes
Installation
Python versions, dependencies, and setup
Python Library
DetectExoplanet and ExoParams API reference
Model Artifacts
Pre-trained models, architectures, and performance
Tutorials & Learning
Step-by-step learning path with Jupyter notebooks
API Deployment
Cerebrium cloud API with live endpoints
Data Reference
NASA Exoplanet Archive data format and features
Examples
Real-world use cases and code samples
Why Choose ExoBengal?
Four ML Models
Random Forest, CNN, k-Nearest Neighbors, and Decision Tree classifiers
Pre-trained Artifacts
Ready-to-use models trained on NASA Kepler mission data
ESI Calculation
Automatic Earth Similarity Index for habitability assessment
Cloud API
Production-ready REST API deployed on Cerebrium
Quick Example
from exobengal import DetectExoplanet, ExoParams
# Initialize detector
detector = DetectExoplanet()
# Create Earth-like parameters
params = ExoParams(
period=365.0, prad=1.0, teq=288.0,
srad=1.0, slog_g=4.44, steff=5778,
impact=0.1, duration=5.0, depth=100.0
)
# Make prediction
result = detector.random_forest(params)
print(f"Prediction: {result['prediction']}")
print(f"Probability: {result['probability']:.2%}")
print(f"ESI: {result['ESI']:.3f}")