edu.udo.cs.yale.operator.learner.kernel.evosvm
Class PSOSVMOptimization

java.lang.Object
  edu.udo.cs.yale.tools.math.optimization.ec.pso.PSOOptimization
      edu.udo.cs.yale.operator.learner.kernel.evosvm.PSOSVMOptimization

All Implemented Interfaces:

Optimization

public class PSOSVMOptimization
extends PSOOptimization

PSO approach for SVM optimization. Currently only classification problems are supported.

Version:

$Id: PSOSVMOptimization.java,v 1.3 2006/04/05 08:57:26 ingomierswa Exp $

Author:

Ingo Mierswa

Field Summary

private double	c This parameter indicates the weight of errors.
private DataTable	dataTable
private double	epsilon This parameter indicates the weight of errors for regression.
private ExampleSet	exampleSet The training example set.
private static double	IS_ZERO Number smaller than this number are regarded as zero.
private Kernel	kernel The used kernel function.
private OptimizationFunction	optimizationFunction This function is to maximize.
private SimplePlotterDialog	plotter
private boolean	showPlot Indicates if a convergence plot should be drawn.
private double[]	ys The label values.

Constructor Summary

PSOSVMOptimization(ExampleSet exampleSet, Kernel kernel, double c, int maxIterations, int generationsWithoutImprovement, int popSize, double inertiaWeight, double localWeight, double globalWeight, boolean dynamicInertiaWeight, boolean showPlot, RandomGenerator random)
Creates a new evolutionary SVM optimization.

Method Summary

PerformanceVector	evaluateIndividual(double[] individual) Evaluates the individuals of the given population.
EvoSVMModel	getModel(double[] alphas) Returns a model containing all support vectors, i.e. the examples with non-zerp alphas.
void	nextIteration() This method is invoked after each evaluation.

Methods inherited from class edu.udo.cs.yale.tools.math.optimization.ec.pso.PSOOptimization

createInitialPopulation, getBestFitnessEver, getBestFitnessInGeneration, getBestPerformanceEver, getBestValuesEver, getGeneration, optimize, setMaxValue, setMinValue

Methods inherited from class java.lang.Object

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

Field Detail

IS_ZERO

private static final double IS_ZERO

Number smaller than this number are regarded as zero.

See Also:

Constant Field Values

exampleSet

private ExampleSet exampleSet

The training example set.

kernel

private Kernel kernel

The used kernel function.

c

private double c

This parameter indicates the weight of errors.

epsilon

private double epsilon

This parameter indicates the weight of errors for regression.

ys

private double[] ys

The label values.

optimizationFunction

private OptimizationFunction optimizationFunction

This function is to maximize.

showPlot

private boolean showPlot

Indicates if a convergence plot should be drawn.

dataTable

private DataTable dataTable

plotter

private SimplePlotterDialog plotter

Constructor Detail

PSOSVMOptimization

public PSOSVMOptimization(ExampleSet exampleSet,
                          Kernel kernel,
                          double c,
                          int maxIterations,
                          int generationsWithoutImprovement,
                          int popSize,
                          double inertiaWeight,
                          double localWeight,
                          double globalWeight,
                          boolean dynamicInertiaWeight,
                          boolean showPlot,
                          RandomGenerator random)

Creates a new evolutionary SVM optimization.

Method Detail

nextIteration

public void nextIteration()

Description copied from class: PSOOptimization

This method is invoked after each evaluation. The default implementation does nothing but subclasses might implement this method to support online plotting or logging.

Overrides:

nextIteration in class PSOOptimization

evaluateIndividual

public PerformanceVector evaluateIndividual(double[] individual)

Evaluates the individuals of the given population.

Specified by:

evaluateIndividual in class PSOOptimization

getModel

public EvoSVMModel getModel(double[] alphas)

Returns a model containing all support vectors, i.e. the examples with non-zerp alphas.