package jsat.distributions.kernels;
import java.util.List;
import jsat.DataSet;
import jsat.distributions.Distribution;
import jsat.linear.Vec;
import jsat.linear.distancemetrics.EuclideanDistance;
/**
* Provides an implementation of the Rational Quadratic Kernel, which is of the
* form:
* k(x, y) = 1 - ||x-y||2 / (||x-y||2 + c)
*
* @author Edward Raff
*/
public class RationalQuadraticKernel extends BaseL2Kernel
{
private static final long serialVersionUID = 6773399185851115840L;
private double c;
/**
* Creates a new RQ Kernel
* @param c the positive additive coefficient
*/
public RationalQuadraticKernel(double c)
{
this.c = c;
}
/**
* Sets the positive additive coefficient
* @param c the positive additive coefficient
*/
public void setC(double c)
{
if(c <= 0 || Double.isNaN(c) || Double.isInfinite(c))
throw new IllegalArgumentException("coefficient must be in (0, Inf), not " + c);
this.c = c;
}
/**
* Returns the positive additive coefficient
* @return the positive additive coefficient
*/
public double getC()
{
return c;
}
@Override
public double eval(Vec a, Vec b)
{
double dist = Math.pow(a.pNormDist(2, b), 2);
return 1-dist/(dist+c);
}
@Override
public double eval(int a, int b, List extends Vec> trainingSet, List cache)
{
double dist = getSqrdNorm(a, b, trainingSet, cache);
return 1-dist/(dist+c);
}
@Override
public double eval(int a, Vec b, List qi, List extends Vec> vecs, List cache)
{
double dist = getSqrdNorm(a, b, qi, vecs, cache);
return 1-dist/(dist+c);
}
@Override
public RationalQuadraticKernel clone()
{
return new RationalQuadraticKernel(c);
}
/**
* Guess the distribution to use for the C parameter.
*
* @param d the data set to get the guess for
* @return the guess for the C parameter
* @see #setC(double)
*/
public static Distribution guessC(DataSet d)
{
//TODO come up with a better estiamte
return RBFKernel.guessSigma(d);//suprisingly this seens to work well
}
@Override
public boolean normalized()
{
return true;
}
}
Ads help maintain this website.