pysindy
A package for the sparse identification of nonlinear dynamical systems from data
What is pysindy?
A package for the sparse identification of nonlinear dynamical systems from data
Overview of pysindy
What it is
PySINDy is a package for system identification, primarily revolving around the method of Sparse Identification of Nonlinear Dynamical systems (SINDy) introduced in Brunton et al. (2016a). It also includes other methods from related literature. System identification refers to the process of using measurement data to infer the governing dynamics. Once discovered, these equations can make predictions about future states, can inform control inputs, or can enable theoretical study using analytical techniques. The resulting models are inherently interpretable and generalizable.
Key features
- Implements the SINDy method and other related system identification methods.
- Uses measurement data to infer interpretable and generalizable dynamical models.
- Supports multiple optimizers via optional extras: convex optimizer SR3 (cvxpy), branch-and-bound optimizer MIOSR (miosr), and Bayesian regression optimizer SBR (sbr).
- Provides a variety of tutorials and examples, starting with generating data and fitting models.
- Integrates with the derivative package for differentiating and smoothing data.
- Offers a Python API with a SINDy class for easy model fitting and printing.
Use cases
- Discovering governing equations from time-series measurement data.
- Making predictions about future states of a dynamical system.
- Informing control inputs for system manipulation.
- Enabling theoretical analysis of discovered models.
- Applications in physics, biology, engineering, and other fields where interpretable models are needed.
Getting started
The preferred way to install is with pip or conda: pip install pysindy You may have to add the --user option. PySINDy also provides several extras, e.g. pip install pysindy[miosr]. For a quick example, see the repository README which demonstrates fitting a model to data from a simple differential equation system.
License and pricing
The license is not stated in the metadata (NOASSERTION). The package is listed as open source and free to use.
Based on the project's public README and repository metadata.