This major release upgrades support to the newest version of Scikit Learn, but
also introduces breaking changes to the API of the FracRidgeRegressorCV
class.
- Allow flexible upgrading of sklearn. (#31)
- made coef.squeeze in fracridge less eager (#40)
- added lapack import & updated too long string; fixes #33 (#34)
- Optimizes the computation to use best option for order of operations. (#30)
- MAINT: Support version of Python up to 3.9 (#29)
- BF: Propagate keyword arguments in the CV estimator. (#26)
- DOC: Explains how to upgrade the software to newer releases. (#27)
- Single regressor (#25)
Fixes a bug in cases where the number of targets is larger than the number of samples in each target.
This version includes changes to the documentation, to refer to the published paper.
This micro version fixes a few small things in the documentation.
- Doc fixes (#15)
This version substantially overhauls the documentation and adds a cross-validation estimator.
- Documentation overhaul. (#12)
- FracRidgeCV estimator (#13)
This version adds more documentation
This is the first major release of the software, in tandem with the appearance of the preprint about the method.