RoPWR: Robust Piecewise Regression¶

The RoPWR library implements several mathematical programming formulations to compute the optimal continuous/discontinuous piecewise polynomial regression given a list of split points. It supports several monotonic constraints, objective functions and regularizations. The library is written in Python and relies on cvxpy (ECOS, OSQP, SCS and HIGHS solvers) to solve the underlying optimization problems. Other formulations are solved using a direct approach.