weka.classifiers.functions.supportVector
Class NormalizedPolyKernel

java.lang.Object
  extended by weka.classifiers.functions.supportVector.Kernel
      extended by weka.classifiers.functions.supportVector.CachedKernel
          extended by weka.classifiers.functions.supportVector.PolyKernel
              extended by weka.classifiers.functions.supportVector.NormalizedPolyKernel
All Implemented Interfaces:
java.io.Serializable

public class NormalizedPolyKernel
extends PolyKernel

The normalized polynomial kernel. K(x,y) = /sqrt() where = PolyKernel(x,y)

Version:
$Revision: 1.1.2.2 $
Author:
Eibe Frank (eibe@cs.waikato.ac.nz)
See Also:
Serialized Form

Constructor Summary
NormalizedPolyKernel(Instances dataset, int cacheSize, double exponent, boolean lowerOrder)
          Creates a new NormalizedPolyKernel instance.
 
Method Summary
 double eval(int id1, int id2, Instance inst1)
          Redefines the eval function of PolyKernel.
 
Methods inherited from class weka.classifiers.functions.supportVector.CachedKernel
clean, numCacheHits, numEvals
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

NormalizedPolyKernel

public NormalizedPolyKernel(Instances dataset,
                            int cacheSize,
                            double exponent,
                            boolean lowerOrder)
Creates a new NormalizedPolyKernel instance.

Parameters:
dataset - the training dataset used.
cacheSize - the size of the cache (a prime number)
Method Detail

eval

public double eval(int id1,
                   int id2,
                   Instance inst1)
            throws java.lang.Exception
Redefines the eval function of PolyKernel.

Overrides:
eval in class CachedKernel
Parameters:
id1 - the index of the first instance in the dataset
id2 - the index of the second instance in the dataset
Returns:
the result of the kernel function
Throws:
java.lang.Exception