edu.udo.cs.yale.operator.learner.kernel
Class LibSVMLearner

java.lang.Object
  edu.udo.cs.yale.operator.Operator
      edu.udo.cs.yale.operator.learner.AbstractLearner
          edu.udo.cs.yale.operator.learner.kernel.LibSVMLearner

All Implemented Interfaces:

ConfigurationListener, Learner

public class LibSVMLearner
extends AbstractLearner

Applies the libsvm learner by Chih-Chung Chang and Chih-Jen Lin. The SVM is a powerful method for both classification and regression. This operator supports the SVM types C-SVC and nu-SVC for classification tasks and epsilon-SVR and nu-SVR for regression tasks. Supports also multiclass learning and probability estimation based on Platt scaling for proper confidence values after applying the learned model on a classification data set.

Version:

$Id: LibSVMLearner.java,v 1.28 2006/09/30 00:05:31 ingomierswa Exp $

Author:

Ingo Mierswa

Keywords:

SVM

Field Summary

static java.lang.String[]	KERNEL_TYPES The different kernel types implemented by the LibSVM package.
private double	lastFitness The fitness of the last learned model.
private int	lastNumberSV The number of SVs for the last learned model.
static java.lang.String[]	SVM_TYPES The different SVM types implemented by the LibSVM package.

Constructor Summary

LibSVMLearner(OperatorDescription description)

Method Summary

private double	getFitness(double[] alphas, double[] ys, libsvm.Kernel kernel)
PerformanceVector	getOptimizationPerformance() Returns the optimization performance of the best result.
private libsvm.svm_parameter	getParameters(ExampleSet exampleSet) Creates a LibSVM parameter object based on the user defined parameters.
java.util.List<ParameterType>	getParameterTypes() Returns a list of ParameterTypes describing the parameters of this operator.
private libsvm.svm_problem	getProblem(ExampleSet exampleSet) Creates a support vector problem for the LibSVM.
Model	learn(ExampleSet exampleSet) Learns a new SVM model with the LibSVM package.
protected static libsvm.svm_node[]	makeNodes(Example e, FastExample2SparseTransform ripper) Creates a data node row for the LibSVM (sparse format, i.e. each node keeps the index and the value if not default).
boolean	shouldDeliverOptimizationPerformance() Returns the value of the corresponding parameter.
boolean	supportsCapability(LearnerCapability lc) Checks for Learner capabilities.

Methods inherited from class edu.udo.cs.yale.operator.learner.AbstractLearner

apply, checkLearnerCapabilities, getEstimatedPerformance, getInputClasses, getInputDescription, getOutputClasses, getWeights, shouldCalculateWeights, shouldEstimatePerformance

Methods inherited from class edu.udo.cs.yale.operator.Operator

addError, addValue, addWarning, apply, checkDeprecations, checkIO, checkProperties, clearErrorList, cloneOperator, createExperimentTree, createExperimentTree, createFromXML, createMarkedExperimentTree, delete, experimentFinished, experimentStarts, getAddOnlyAdditionalOutput, getApplyCount, getDeliveredOutputClasses, getDeprecationInfo, getDesiredInputClasses, getErrorList, getExperiment, getInnerOperatorsXML, getInput, getInput, getInput, getIOContainerForInApplyLoopBreakpoint, getName, getNumberOfSteps, getOperatorClassName, getOperatorDescription, getParameter, getParameterAsBoolean, getParameterAsColor, getParameterAsDouble, getParameterAsFile, getParameterAsInt, getParameterAsString, getParameterList, getParameters, getParameterType, getParent, getStartTime, getStatus, getUserDescription, getValue, getValues, getXML, hasBreakpoint, hasBreakpoint, hasInput, inApplyLoop, isEnabled, isParameterSet, logMessage, performAdditionalChecks, register, remove, rename, resume, setBreakpoint, setEnabled, setExperiment, setInput, setListParameter, setOperatorParameters, setParameter, setParameters, setParent, setUserDescription, toString, writeXML

Methods inherited from class java.lang.Object

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

Methods inherited from interface edu.udo.cs.yale.operator.learner.Learner

getName

Field Detail

SVM_TYPES

public static final java.lang.String[] SVM_TYPES

The different SVM types implemented by the LibSVM package.

KERNEL_TYPES

public static final java.lang.String[] KERNEL_TYPES

The different kernel types implemented by the LibSVM package.

lastFitness

private double lastFitness

The fitness of the last learned model.

lastNumberSV

private int lastNumberSV

The number of SVs for the last learned model.

Constructor Detail

LibSVMLearner

public LibSVMLearner(OperatorDescription description)

Method Detail

shouldDeliverOptimizationPerformance

public boolean shouldDeliverOptimizationPerformance()

Returns the value of the corresponding parameter.

Overrides:

shouldDeliverOptimizationPerformance in class AbstractLearner

getOptimizationPerformance

public PerformanceVector getOptimizationPerformance()
                                             throws OperatorException

Returns the optimization performance of the best result. This method must be called after training, not before.

Overrides:

getOptimizationPerformance in class AbstractLearner

Throws:

OperatorException

getFitness

private double getFitness(double[] alphas,
                          double[] ys,
                          libsvm.Kernel kernel)

supportsCapability

public boolean supportsCapability(LearnerCapability lc)

Description copied from interface: Learner

Checks for Learner capabilities. Should return true if the given capability is supported.

makeNodes

protected static libsvm.svm_node[] makeNodes(Example e,
                                             FastExample2SparseTransform ripper)

Creates a data node row for the LibSVM (sparse format, i.e. each node keeps the index and the value if not default).

getProblem

private libsvm.svm_problem getProblem(ExampleSet exampleSet)

Creates a support vector problem for the LibSVM.

getParameters

private libsvm.svm_parameter getParameters(ExampleSet exampleSet)
                                    throws UndefinedParameterError

Creates a LibSVM parameter object based on the user defined parameters. If gamma is set to zero, it will be overwritten by 1 divided by the number of attributes.

Throws:

UndefinedParameterError

learn

public Model learn(ExampleSet exampleSet)
            throws OperatorException

Learns a new SVM model with the LibSVM package.

Throws:

OperatorException

getParameterTypes

public java.util.List<ParameterType> getParameterTypes()

Description copied from class: Operator

Returns a list of ParameterTypes describing the parameters of this operator. The default implementation returns an empty list if no input objects can be retained and special parameters for those input objects which can be prevented from being consumed.

Overrides:

getParameterTypes in class Operator