com.gregdennis.drej
Class GaussianKernel

java.lang.Object
  com.gregdennis.drej.GaussianKernel

All Implemented Interfaces:

Kernel

public class GaussianKernel
extends java.lang.Object
implements Kernel

A Gaussian kernel of the following form:

K(x1, x2) = exp(‒γ² · ∥x1 - x2∥²)

Author:

Greg Dennis (gdennis@mit.edu)

Constructor Summary

GaussianKernel(double gamma)
Construct a Gaussian kernel with the specified value for γ.

Method Summary

double	eval(GVector x1, GVector x2) Evaluates the kernel function at the specified points.
double	gamma() Returns the value of γ.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

GaussianKernel

public GaussianKernel(double gamma)

Construct a Gaussian kernel with the specified value for γ.

Method Detail

gamma

public double gamma()

Returns the value of γ.

eval

public double eval(GVector x1,
                   GVector x2)

Description copied from interface: Kernel

Evaluates the kernel function at the specified points.

Specified by:

eval in interface Kernel