kpchamp / sindysr3

SINDySR3

Repository for the paper A unified sparse optimization framework to learn parsimonious physics-informed models from data by Kathleen Champion, Peng Zheng, Aleksandr Y. Aravkin, Steven L. Brunton, and J. Nathan Kutz.

This work implements the sparse regularized relaxed regression (SR3) optimization method for sparse identification of nonlinear dynamics (SINDy), including a number of extensions to the algorithm for corrupt data trimming, incorporating physical constraints, and fitting parameterized library functions. Code examples can be found as jupyter notebooks in the examples directory.

Dependencies

This code requires the installation of the PySINDy package.

sindysr3's Issues

Bug report: normalize , n_jobs

Dear developer,

I am writing to report a bug in the SINDySR3 uploaded on GitHub. Two issues are as follow:

• In classes SR3Forcing and SR3Enhanced: 'normalize' should be replaced with 'normalize_columns' as the later is compatible with the pysindy source code.
• In class SINDyForcing: n_jobs is not working so maybe remove or solve the issue.

I hope these help.

Kind regards,
Sina