PyMVPA LibSVM-based kernels

These kernels do not currently have the ability to run the calculations, so they are not translateable to other kernel types. They are implemented solely to standardize the interface between other kernel machines.

Inheritance diagram of mvpa2.kernels.libsvm


Kernel(\*args, \*\*kwargs) Abstract class which calculates a kernel function between datasets
LSKernel(\*args, \*\*kwargs) A Kernel object which dictates how LibSVM will calculate the kernel
LinearLSKernel(\*args, \*\*kwargs) A simple Linear kernel: K(a,b) = a*b.T
Parameter(default[, constraints, ro, index, ...]) This class shall serve as a representation of a parameter.
PolyLSKernel(\*\*kwargs) Polynomial kernel: K(a,b) = (gamma*a*b.T + coef0)**degree
RbfLSKernel(\*\*kwargs) Radial Basis Function kernel (aka Gaussian):
SigmoidLSKernel(\*\*kwargs) Sigmoid kernel: K(a,b) = tanh(gamma*a*b.T + coef0)