COMBSS: Best Subset Selection for Generalised Linear Models
combss is a Python package implementing COMBSS (Continuous Optimisation for Best Subset Selection) for generalised linear models. COMBSS reformulates the NP-hard discrete subset selection problem as a continuous optimisation over the hypercube \([0,1]^p\), making it scalable to high-dimensional settings with \(p \gg n\).
Supported model types
Linear regression (continuous response)
Binary logistic regression (two-class classification)
Multinomial logistic regression (multi-class, \(C > 2\))
Installation
pip install combss
References
Moka, Liquet, Zhu & Muller (2024). COMBSS: best subset selection via continuous optimization. Statistics and Computing.
Mathur, Liquet, Muller & Moka (2026). Parsimonious Subset Selection for Generalized Linear Models with Biomedical Applications. arXiv preprint.