Dark from Light 🌌

"Unveiling the invisible universe, one galaxy at a time" A Python package that classifies galaxies into groups and estimates dark matter halo masses using Random Forest regressors trained on large volume cosmological simulations: EAGLE and IllustrisTNG.

Features

• Galaxy Group Classification: Automatically allocate galaxies into groups
• Dark Matter Mass Estimation: Predict halo masses from observable properties (photometric/spectroscopic stellar mass estimates, sky coordinates and redhsift)
• Pre-trained Models: Ready-to-use Random Forest regressors trained on simulation data (mapping between electromagnetic observables and halo properties)
• Multiple Halo Mass - Stellar Mass Relations: Choice among models trained on EAGLE, IllustrisTNG and hybrid approach (mean prediction between the two simulations)

Quick Start

from dark_from_light import allocate_galaxy_groups, HybridRegressionModel

# Classify galaxies into groups
allocate_galaxy_groups(galaxy_data)

# Estimate dark matter halo masses
model = HybridRegressionModel()
halo_mass = model.predict_group(group_total_stellar_mass)

Turn stellar observations into cosmic insights.