I am a researcher on System Identification community regarding NARMAX models (Nonlinear Autoregressive Moving Avarage with eXogenous inputs) and I am here to share a python package that I made public just now.
It is a working in progress but I'm really happy with it. There are many improvements to be done (concerning the code, docs, tests) but we hope that the community can help us with all that!
SysIdentpy is a library for modeling nonlinear dynamical data using nonlinear autoregressive models known as NARMAX models. One can consider these models a generalization of ARMAX models for nonlinear data, but NARMAX models are not, however, a simple extension of ARMAX models.
I'll share the repository on comments.
Many additional features are already in progress. It will be a pleasure to have the collaboration of all of you to make the library more and more complete.