package jsat.distributions.kernels;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Collections;
import java.util.List;
import jsat.DataSet;
import jsat.classifiers.ClassificationDataSet;
import jsat.distributions.Distribution;
import jsat.distributions.Exponential;
import jsat.distributions.LogUniform;
import jsat.distributions.Uniform;
import jsat.linear.Vec;
import jsat.linear.distancemetrics.EuclideanDistance;
import jsat.math.OnLineStatistics;
import jsat.text.GreekLetters;
import jsat.utils.DoubleList;
import jsat.utils.IntList;
import jsat.utils.ListUtils;
/**
* Provides a kernel for the Radial Basis Function, which is of the form
*
* k(x, y) = exp(-||x-y||2/(2*σ2))
*
* @author Edward Raff
*/
public class RBFKernel extends BaseL2Kernel
{
private static final long serialVersionUID = -6733691081172950067L;
private double sigma;
private double sigmaSqrd2Inv;
/**
* Creates a new RBF kernel with σ = 1
*/
public RBFKernel()
{
this(1.0);
}
/**
* Creates a new RBF kernel
* @param sigma the sigma parameter
*/
public RBFKernel(double sigma)
{
setSigma(sigma);
}
@Override
public double eval(Vec a, Vec b)
{
if(a == b)//Same refrence means dist of 0, exp(0) = 1
return 1;
return Math.exp(-Math.pow(a.pNormDist(2, b),2) * sigmaSqrd2Inv);
}
@Override
public double eval(int a, int b, List extends Vec> trainingSet, List cache)
{
if(a == b)
return 1;
return Math.exp(-getSqrdNorm(a, b, trainingSet, cache)* sigmaSqrd2Inv);
}
@Override
public double eval(int a, Vec b, List qi, List extends Vec> vecs, List cache)
{
return Math.exp(-getSqrdNorm(a, b, qi, vecs, cache)* sigmaSqrd2Inv);
}
/**
* Sets the sigma parameter, which must be a positive value
* @param sigma the sigma value
*/
public void setSigma(double sigma)
{
if(sigma <= 0)
throw new IllegalArgumentException("Sigma must be a positive constant, not " + sigma);
this.sigma = sigma;
this.sigmaSqrd2Inv = 0.5/(sigma*sigma);
}
public double getSigma()
{
return sigma;
}
@Override
public String toString()
{
return "RBF Kernel( " + GreekLetters.sigma +" = " + sigma +")";
}
@Override
public RBFKernel clone()
{
return new RBFKernel(sigma);
}
/**
* Another common (equivalent) form of the RBF kernel is k(x, y) =
* exp(-γ||x-y||2). This method converts the σ value
* used by this class to the equivalent γ value.
* @param sigma the value of σ
* @return the equivalent γ value.
*/
public static double sigmaToGamma(double sigma)
{
if(sigma <= 0 || Double.isNaN(sigma) || Double.isInfinite(sigma))
throw new IllegalArgumentException("sigma must be positive, not " + sigma);
return 1/(2*sigma*sigma);
}
/**
* Another common (equivalent) form of the RBF kernel is k(x, y) =
* exp(-γ||x-y||2). This method converts the γ value
* equivalent σ value used by this class.
* @param gamma the value of γ
* @return the equivalent σ value
*/
public static double gammToSigma(double gamma)
{
if(gamma <= 0 || Double.isNaN(gamma) || Double.isInfinite(gamma))
throw new IllegalArgumentException("gamma must be positive, not " + gamma);
return 1/Math.sqrt(2*gamma);
}
/**
* Guess the distribution to use for the kernel width term
* {@link #setSigma(double) σ} in the RBF kernel.
*
* @param d the data set to get the guess for
* @return the guess for the σ parameter in the RBF Kernel
*/
public static Distribution guessSigma(DataSet d)
{
return GeneralRBFKernel.guessSigma(d, new EuclideanDistance());
}
@Override
public boolean normalized()
{
return true;
}
}
Ads help maintain this website.