NonGaussianRBFKernelFunctor Class Reference

Performs an RBF kernel evaluation that better suit sample distribution with high Kurtosis. More...

#include <otbNonGaussianRBFKernelFunctor.h>

Detailed Description

Performs an RBF kernel evaluation that better suit sample distribution with high Kurtosis.

It is of kind $\exp\left( - \gamma \sum_i | x_i^\alpha - y_i^\alpha |^\beta \right)$ where $0 \leqslant \alpha \leqslant 1$ and $0 \leqslant \beta \leqslant 2$ .

Variables to be instanciated (through

SetValue

) are: Alpha (def=1), Beta (def=2) and Gamma (def 1.0).

The documentation for this class was generated from the following file:

otbNonGaussianRBFKernelFunctor.h