fr.lip6.bag.kernel
Class WPowerBagKernel<V extends MLVector<?>,B extends WeightedBagOfVector<V>>

java.lang.Object
  fr.lip6.kernel.Kernel<B>
      fr.lip6.bag.kernel.WPowerBagKernel<V,B>

Type Parameters:

V - vectors of input space

B - wieghted bags of vectors.

All Implemented Interfaces:

java.io.Serializable

public class WPowerBagKernel<V extends MLVector<?>,B extends WeightedBagOfVector<V>>
extends Kernel<B>

Weighted Power kernel on bags of vectors : K(Bi, Bj) = (sum_Bi(sum_Bj(w_ri*w_sj*k(b_ri,b_sj))^q))^{1/q}

Author:

dpicard

See Also:

Serialized Form

Constructor Summary

WPowerBagKernel(Kernel<V> minorKernel, double power)

Method Summary

double	valueOf(B t1) kernel similarity to zero
double	valueOf(B t1, B t2) compute the kernel similarity between two element of input space

Methods inherited from class fr.lip6.kernel.Kernel

NormalizedValueOf

Methods inherited from class java.lang.Object

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

Constructor Detail

WPowerBagKernel

public WPowerBagKernel(Kernel<V> minorKernel,
                       double power)

Method Detail

valueOf

public double valueOf(B t1,
                      B t2)

Description copied from class: Kernel

compute the kernel similarity between two element of input space

Specified by:

valueOf in class Kernel<B extends WeightedBagOfVector<V>>

Parameters:

t1 - first element

t2 - second element

Returns:

the kernel value

valueOf

public double valueOf(B t1)

Description copied from class: Kernel

kernel similarity to zero

Specified by:

valueOf in class Kernel<B extends WeightedBagOfVector<V>>

Parameters:

t1 - the element to compute the similarity to zero