This is a python implementation of the SYNDy algorithm introduced by Brunton et al. of University of Washington, Seattle. See paper.pdf for the details of the method. You can also watch a presentation of it here.
The algorithm is used to identify governing equations for nonlinear systems for which we have data, but no mathematical model.
A jupyter notebook showing an example of deducing a math model purely from data.
This file includes all the implementation details of the algorithm as well as a demo example in the main method.
Implementation of the classical SIR model which is used as an example to demonstrate how SYNDy works.
The PNAS publication introducing the algorithm.