Python Symbolic regression Through Algebraic Representations (pySTAR) provides tools for generating mathematical (symbolic) models that best fit some input data. The pySTAR toolbox allows for the definition of the form of the surrogate model and the regression parameter values simultaneously.

If you use this code please cite: "Sarwar, O. 2022, Algorithms for Interpretable High-Dimensional Regression, Carnegie Mellon University."

The pySTAR toolkit is installed as part of the IDAES Integrated Platform (IDAES-IP). Please see the README of the IDAES-PSE repository and/or the full IDAES install instructions for details.

This work was conducted as part of the Institute for the Design of Advanced Energy Systems (IDAES) with support through the Crosscutting Research Program within the U.S. Department of Energy’s Office of Fossil Energy and Carbon Management (FECM).