AI and the World's Hard Problems · Astronomy & space · arXiv · 2026-10-23
SpecLess: A Multi Survey Python Package for Machine Learning Classification of Galaxy Morphology and Activity from Photometry
Galaxy classification from broadband photometry does not always transfer well when the same galaxies are measured by a different survey. The authors present SpecLess, an open-source Python package with survey-specific XGBoost classifiers for galaxy morphology and activity, reaching macro F1 scores of about 0.93 (SDSS) and 0.92 (KiDS) for morphology and about 0.76 (SDSS) and 0.74 (DESI) for activity. They also quantify how performance drops, and the drop depends on the direction of transfer between surveys.