franken.rf.heads

This module contains random feature implementations for different kernels

Base Class

Base class for random-feature heads

Approximations to the classical Gaussian (or RBF) kernel

`franken.rf.heads.OrthogonalRFF`	Orthogonal Random Fourier Features by Yu et al. for approximating the Gaussian kernel.
`franken.rf.heads.MultiScaleOrthogonalRFF`	A multi-scale version of `franken.rf.heads.OrthogonalRFF` which splits the available random features among multiple length-scales.
`franken.rf.heads.BiasedOrthogonalRFF`

`franken.rf.heads.Linear`	Implements the linear kernel.
`franken.rf.heads.RandomFeaturesHead`	Base class for random-feature heads
`franken.rf.heads.TensorSketch`	TensorSketch random-features approximation of the polynomial kernel