Poster
in
Workshop: CODEML: Championing Open-source DEvelopment in Machine Learning
skglm: Improving scikit-learn for regularized Generalized Linear Models
Mathurin Massias · Badr MOUFAD · Quentin Bertrand
We introduce skglm, an open-source Python package for regularized Generalized Linear Models (GLMs). It solves many limitations of scikit-learn, that have impaired the use of GLMs by practitioners. Thanks to its composable nature, skglm supports combining datafits, penalties, and solvers to fit a wide range of models, many of them not included in scikit-learn (e.g. Group Lasso and variants). It uses state-of-the-art algorithms to solve problems involving high-dimensional datasets, providing large speed-ups compared to existing implementations, and unlocking new applications. It is fully compliant with the scikit-learn API and acts as a drop-in replacement for its estimators. Finally, it abides by the standards of open source development and is integrated in the scikit-learn-contrib GitHub organization.