This Github repository contains codes for the paper Automatically discovering ordinary differential equations from data with sparse regression.
The purpose of this code is to enable users to perform system identification of linear and nonlinear systems automatically. System identification refers to the process of building mathematical models of dynamical systems using measurements of the input and output signals of the system. A dynamical system is a system whose state varies over time and follows sets of ordinary or partial differential equations. We propose a framework that applies sparse regression algorithms with model selection methods to identify the governing equations that describe the data automatically. Upon identification, users can apply the resulting models to generalize and predict the systems existing in their data.
Create_data contains code to allow users to format test results for plotting.
R contains all codes used for implementing our method in R.
We provide code for the bootstrap, lasso and adaptive lasso, and Savitzky-Golay filter.
We also provide all tests run for the Automatically discovering ordinary differential equations from data with sparse regression paper.
Python contains all code and tests we used for SINDy-AIC.
Plot_code contains code for figures shown in the paper.
When using Windows system, please clone to the drive directly like 'C:\' or 'D:\', not the folders in 'C:\' or 'D:\', because some file names are too long to be cloned in Windows. MAC OS and Linux do not have this problem.
We have recently published an R package for ARGOS on CRAN.
The corresponding GitHub page can be found here.