Open
0 of 2 issues completedDescription
It would be an amazing addition to the documentation and worth of the package to have a notebook describing each approximator available in shapiq. This describtion can be theoretical (describing the mechanism behind the method) but also make the ideas behind it accessible for people unfarmiliar with the details. A benchmark of all of the approximators can then be found in a seperat overarching notebook.
Of course it would make sensitive group similar approximators together. (I am looking at you SVARM/SVARM-IQ or the kernelshap ones, KernelSHAP/Kernelshapiq/RegressionFSII)
Sub-issues
Metadata
Assignees
Projects
Status
💡Idea