pykoop v1.0.5

This release features two quality of life improvements: a better lifting function interface for use outside of scikit-learn, and improved trajectory prediction functionality. More importantly, the docs have been reorganized, the unit tests are no longer a mess, and some Jupyter notebook examples have been added to binder.

Full changelog: v1.0.4...v1.0.5

New features

• Added lift(), lift_state(), lift_input(), retract(), retract_state(), and retract_input() helper methods to KoopmanPipeline and all Koopman lifting functions. These functions provide a more convenient way to use a fit Koopman model outside of scikit-learn (e.g. in control applications) (#61)
• Added predict_trajectory() as a replacement for predict_multistep(), which is now deprecated. This new function provides a more convenient interface for use outside of scikit-learn, and also supports global Koopman predictions, where states are not retracted and re-lifted between timesteps (#65).

Enhancements

• Overhauled organization of Sphinx docs (#66)
• Updated examples and added binder links to Juypter notebooks (#70, #71).
• Refactored unit tests, and enabled remote testing and doctests in CI (#67).

Bug fixes

• Fixed a serious bug in predict_multistep() where only the first episode was scored (#65).
• Allowed force quitting LMI regressor using ^C twice (#54).
• Stopped doctests from failing due to floating point comparisons (#58).

pykoop v1.0.4

Small release to adjust LMI strictness to match cited arXiv preprint.

pykoop v1.0.3

No actual code changes. No Zenodo archive was generated because of errors in CITATION.cff.

pykoop v1.0.2

No actual code changes, only continuous-integration workflows and citation metadata.

pykoop v1.0.1

v1.0.0 of pykoop was not usable because of a missing __init__.py in a sub-package. All fixed now!