package jsat.linear.distancemetrics;
import java.util.List;
import jsat.classifiers.knn.NearestNeighbour;
import jsat.distributions.kernels.*;
import jsat.linear.Vec;
/**
* Creates a distance metric from a given kernel trick.
* For the distance metric to be valid, the kernel used
* must be positive definite.
*
* @author Edward Raff
*/
public class KernelDistance implements DistanceMetric
{
private static final long serialVersionUID = -1553315486668768024L;
private KernelTrick kf;
/**
* Creates a distane metric from the given kernel. For the metric to be valid, the kernel must be positive definite. This means that
*
* ∀ ci ∈ ℜ , xi ∈ ℜd
* ∑i, j = 1m ci cj K(xi, xj) ≥ 0
* @param kf
*/
public KernelDistance(KernelTrick kf)
{
this.kf = kf;
}
/**
* Returns the square of the distance function expanded as kernel methods.
*
* d2(x,y) = K(x,x) - 2*K(x,y) + K(y,y)
*
*
* Special Notes:
* The use of {@link RBFKernel} or {@link PolynomialKernel} of degree 1
* in the {@link NearestNeighbour} classifier will degenerate into the
* normal nearest neighbor algorithm.
*
* @param a the first vector
* @param b the second vector
* @return the distance metric based on a kernel function
*/
@Override
public double dist(Vec a, Vec b)
{
return kf.eval(a, a) - 2*kf.eval(a, b) + kf.eval(b, b);
}
@Override
public boolean isSymmetric()
{
return true;
}
@Override
public boolean isSubadditive()
{
return true;
}
@Override
public boolean isIndiscemible()
{
return true;
}
@Override
public double metricBound()
{
return Double.POSITIVE_INFINITY;
}
@Override
public String toString()
{
return "Kernel (" + kf + ") Distance";
}
@Override
public KernelDistance clone()
{
return new KernelDistance(kf);
}
//TODO modify CacheAccerleratedKernel to work with lists...
@Override
public boolean supportsAcceleration()
{
return false;
}
@Override
public List getAccelerationCache(List extends Vec> vecs, boolean parallel)
{
return null;
}
@Override
public double dist(int a, int b, List extends Vec> vecs, List cache)
{
return dist(vecs.get(a), vecs.get(b));
}
@Override
public double dist(int a, Vec b, List extends Vec> vecs, List cache)
{
return dist(vecs.get(a), b);
}
@Override
public List getQueryInfo(Vec q)
{
return null;
}
@Override
public double dist(int a, Vec b, List qi, List extends Vec> vecs, List cache)
{
return dist(vecs.get(a), b);
}
}
Ads help maintain this website.