adams.optimise.genetic.fitnessfunctions
Class AttributeSelection

java.lang.Object
  adams.core.ConsoleObject
      adams.core.option.AbstractOptionHandler
          adams.optimise.AbstractFitnessFunction
              adams.optimise.genetic.fitnessfunctions.AbstractWEKAFitnessFunction
                  adams.optimise.genetic.fitnessfunctions.AttributeSelection

All Implemented Interfaces:

Debuggable, Destroyable, OptionHandler, SizeOfHandler, FitnessFunction, Serializable

public class AttributeSelection
extends AbstractWEKAFitnessFunction

Perform attribute selection using WEKA classification.

Author:

dale

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class adams.optimise.genetic.fitnessfunctions.AbstractWEKAFitnessFunction

AbstractWEKAFitnessFunction.Measure

Field Summary

Fields inherited from class adams.optimise.genetic.fitnessfunctions.AbstractWEKAFitnessFunction

m_Classifier, m_ClassIndex, m_CrossValidationSeed, m_Dataset, m_Folds, m_init, m_Instances, m_Measure, m_OutputDirectory

Fields inherited from class adams.core.option.AbstractOptionHandler

m_DebugLevel, m_OptionManager

Constructor Summary

AttributeSelection()

Method Summary

double	evaluate(OptData opd)
OptData	getDataDef()
String	getRemoveAsString(int[] m_weights) Generates a range string of attributes to keep (= one has to use the inverse matching sense with the Remove filter).
protected int[]	getWeights(OptData opd)
String	globalInfo()
void	newBest(double val, OptData opd) Callback for best measure so far

Methods inherited from class adams.optimise.genetic.fitnessfunctions.AbstractWEKAFitnessFunction

classifierTipText, classIndexTipText, crossValidationSeedTipText, datasetTipText, defineOptions, foldsTipText, getClassifier, getClassIndex, getCrossValidationSeed, getDataset, getFolds, getInstances, getMeasure, getOutputDirectory, init, measureTipText, outputDirectoryTipText, setClassifier, setClassIndex, setCrossValidationSeed, setDataset, setFolds, setInstances, setMeasure, setOutputDirectory

Methods inherited from class adams.core.option.AbstractOptionHandler

cleanUpOptions, debug, debug, debugLevelTipText, destroy, finishInit, getDebugLevel, getOptionManager, initialize, isDebugOn, newOptionManager, reset, setDebugLevel, toCommandLine, toString

Methods inherited from class adams.core.ConsoleObject

getDebugging, getSystemErr, getSystemOut, sizeOf

Methods inherited from class java.lang.Object

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

Constructor Detail

AttributeSelection

public AttributeSelection()

Method Detail

globalInfo

public String globalInfo()

Overrides:

globalInfo in class AbstractWEKAFitnessFunction

getDataDef

public OptData getDataDef()

getWeights

protected int[] getWeights(OptData opd)

evaluate

public double evaluate(OptData opd)

getRemoveAsString

public String getRemoveAsString(int[] m_weights)

Generates a range string of attributes to keep (= one has to use the inverse matching sense with the Remove filter).

Returns:

the range of attributes to keep

newBest

public void newBest(double val,
                    OptData opd)

Callback for best measure so far

Specified by:

newBest in interface FitnessFunction

Overrides:

newBest in class AbstractFitnessFunction