Open-source implementations of the methods in my papers. Each package ships with documentation and tests, and the Python and MATLAB versions share the same functions, options, and conventions so results can be reproduced in either environment.
Cuts a nonnegative measure into scale strata with no threshold to tune and no group count to declare. Given a spectrum, a set of weights, or any distribution of magnitudes, SPINE finds where one scale ends and the next begins. The Python package and the MATLAB reference implementation share the same interface.
pip install entropy-spineA data generator for gene regulatory networks with Hill-function kinetics. Every dataset comes with the signed regulatory network, every kinetic parameter, the initial conditions, the seeds, and the solver settings that produced it, so causal discovery and network inference methods can be scored against a known answer. Available as a Python package and a MATLAB toolbox.
pip install grnlabRecovers the governing equations of a dynamical system from data by selecting terms with an information-theoretic criterion rather than a sparsity penalty, which makes it robust to noise and to outliers. The toolbox behind the Editor's Pick paper in Chaos.
A Python package and MATLAB toolbox for building, analyzing, and synchronizing complex networks, including the essential synchronization backbone and information-theoretic dependency measures from my network papers.