MatCoupLy: Learning coupled matrix factorizations with Python

Coupled matrix factorization (CMF) models jointly decompose a collection of matrices with one shared mode. For interpretable decompositions, constraints are often needed, and variations constrained CMF have been used in various fields, including data mining, chemometrics remote sensing. Although such broadly used, there is lack easy-to-use, documented, open-source implementations for fitting CMFs user-specified on all modes. We address this need MatCoupLy, Python package that implements state-of-the-art algorithm PARAFAC2 supports any proximable constraint This paper outlines the functionality three examples demonstrating flexibility extendibility package.