pt.tumba.ngram.svm
Class SVRQ

java.lang.Object
  pt.tumba.ngram.svm.Kernel
      pt.tumba.ngram.svm.SVRQ

 class SVRQ
extends Kernel

Inner class representing a Kernel matrix for Support Vector regression.

Field Summary

private float[][]	buffer
private Cache	cache
private int[]	index
private int	l
private int	next_buffer
private byte[]	sign

Constructor Summary

SVRQ(SVMProblem prob, SVMParameter param)

Method Summary

(package private) float[]	getQ(int i, int len) Return one column of the Kernel matrix.
(package private) void	swapIndex(int i, int j) Swap two nodes of the Kernel matrix.

Methods inherited from class pt.tumba.ngram.svm.Kernel

dot, kernelFunction, kernelFunction

Methods inherited from class java.lang.Object

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

Field Detail

l

private final int l

cache

private final Cache cache

sign

private final byte[] sign

index

private final int[] index

next_buffer

private int next_buffer

buffer

private float[][] buffer

Constructor Detail

SVRQ

SVRQ(SVMProblem prob,
     SVMParameter param)

Method Detail

swapIndex

void swapIndex(int i,
               int j)

Description copied from class: Kernel

Swap two nodes of the Kernel matrix.

Overrides:

swapIndex in class Kernel

getQ

float[] getQ(int i,
             int len)

Description copied from class: Kernel

Return one column of the Kernel matrix.

Specified by:

getQ in class Kernel

Parameters:

i - The column of the kernel matrix.

len - The nu

Returns:

One column of the Kernel matrix.