weka.classifiers.meta
Class CVParameterSelection

java.lang.Object
  weka.classifiers.Classifier
      weka.classifiers.SingleClassifierEnhancer
          weka.classifiers.RandomizableSingleClassifierEnhancer
              weka.classifiers.meta.CVParameterSelection

All Implemented Interfaces:

java.lang.Cloneable, Drawable, OptionHandler, Randomizable, java.io.Serializable, Summarizable

public class CVParameterSelection

extends RandomizableSingleClassifierEnhancer

implements Drawable, Summarizable

Class for performing parameter selection by cross-validation for any classifier. For more information, see

R. Kohavi (1995). Wrappers for Performance Enhancement and Oblivious Decision Graphs. PhD Thesis. Department of Computer Science, Stanford University.

Valid options are:

-D
Turn on debugging output.

-W classname
Specify the full class name of classifier to perform cross-validation selection on.

-X num
Number of folds used for cross validation (default 10).

-S seed
Random number seed (default 1).

-P "N 1 5 10"
Sets an optimisation parameter for the classifier with name -N, lower bound 1, upper bound 5, and 10 optimisation steps. The upper bound may be the character 'A' or 'I' to substitute the number of attributes or instances in the training data, respectively. This parameter may be supplied more than once to optimise over several classifier options simultaneously.

Options after -- are passed to the designated sub-classifier.

Version:

$Revision: 1.26 $

Author:

Len Trigg (trigg@cs.waikato.ac.nz)

See Also:

Serialized Form

Nested Class Summary

protected class

CVParameterSelection.CVParameter

Field Summary

protected java.lang.String[]	m_BestClassifierOptions The set of all classifier options as determined by cross-validation
protected double	m_BestPerformance The cross-validated performance of the best options
protected java.lang.String[]	m_ClassifierOptions The base classifier options (not including those being set by cross-validation)
protected FastVector	m_CVParams The set of parameters to cross-validate over
protected java.lang.String[]	m_InitOptions The set of all options at initialization time.
protected int	m_NumAttributes The number of attributes in the data
protected int	m_NumFolds The number of folds used in cross-validation
protected int	m_TrainFoldSize The number of instances in a training fold

Fields inherited from class weka.classifiers.RandomizableSingleClassifierEnhancer

m_Seed

Fields inherited from class weka.classifiers.SingleClassifierEnhancer

m_Classifier

Fields inherited from class weka.classifiers.Classifier

m_Debug

Fields inherited from interface weka.core.Drawable

BayesNet, NOT_DRAWABLE, TREE

Constructor Summary

CVParameterSelection()

Method Summary

void	addCVParameter(java.lang.String cvParam) Adds a scheme parameter to the list of parameters to be set by cross-validation
void	buildClassifier(Instances instances) Generates the classifier.
protected java.lang.String[]	createOptions() Create the options array to pass to the classifier.
java.lang.String	CVParametersTipText() Returns the tip text for this property
double[]	distributionForInstance(Instance instance) Predicts the class distribution for the given test instance.
protected void	findParamsByCrossValidation(int depth, Instances trainData, java.util.Random random) Finds the best parameter combination.
java.lang.String	getCVParameter(int index) Gets the scheme paramter with the given index.
java.lang.Object[]	getCVParameters() Get method for CVParameters.
int	getNumFolds() Gets the number of folds for the cross-validation.
java.lang.String[]	getOptions() Gets the current settings of the Classifier.
java.lang.String	globalInfo() Returns a string describing this classifier
java.lang.String	graph() Returns graph describing the classifier (if possible).
int	graphType() Returns the type of graph this classifier represents.
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class.
java.lang.String	numFoldsTipText() Returns the tip text for this property
void	setCVParameters(java.lang.Object[] params) Set method for CVParameters.
void	setNumFolds(int numFolds) Sets the number of folds for the cross-validation.
void	setOptions(java.lang.String[] options) Parses a given list of options.
java.lang.String	toString() Returns description of the cross-validated classifier.
java.lang.String	toSummaryString() Returns a string that summarizes the object.

Methods inherited from class weka.classifiers.RandomizableSingleClassifierEnhancer

getSeed, seedTipText, setSeed

Methods inherited from class weka.classifiers.SingleClassifierEnhancer

classifierTipText, defaultClassifierString, getClassifier, getClassifierSpec, setClassifier

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, forName, getDebug, makeCopies, setDebug

Methods inherited from class java.lang.Object

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

Field Detail

m_ClassifierOptions

protected java.lang.String[] m_ClassifierOptions

The base classifier options (not including those being set by cross-validation)

m_BestClassifierOptions

protected java.lang.String[] m_BestClassifierOptions

The set of all classifier options as determined by cross-validation

m_InitOptions

protected java.lang.String[] m_InitOptions

The set of all options at initialization time. So that getOptions can return this.

m_BestPerformance

protected double m_BestPerformance

The cross-validated performance of the best options

m_CVParams

protected FastVector m_CVParams

The set of parameters to cross-validate over

m_NumAttributes

protected int m_NumAttributes

The number of attributes in the data

m_TrainFoldSize

protected int m_TrainFoldSize

The number of instances in a training fold

m_NumFolds

protected int m_NumFolds

The number of folds used in cross-validation

Constructor Detail

CVParameterSelection

public CVParameterSelection()

Method Detail

createOptions

protected java.lang.String[] createOptions()

Create the options array to pass to the classifier. The parameter values and positions are taken from m_ClassifierOptions and m_CVParams.

Returns:

the options array

findParamsByCrossValidation

protected void findParamsByCrossValidation(int depth,
                                           Instances trainData,
                                           java.util.Random random)
                                    throws java.lang.Exception

Finds the best parameter combination. (recursive for each parameter being optimised).

Parameters:

depth - the index of the parameter to be optimised at this level

Throws:

java.lang.Exception - if an error occurs

globalInfo

public java.lang.String globalInfo()

Returns a string describing this classifier

Returns:

a description of the classifier suitable for displaying in the explorer/experimenter gui

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Overrides:

listOptions in class RandomizableSingleClassifierEnhancer

Returns:

an enumeration of all the available options.

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Parses a given list of options. Valid options are:

-D
Turn on debugging output.

-W classname
Specify the full class name of classifier to perform cross-validation selection on.

-X num
Number of folds used for cross validation (default 10).

-S seed
Random number seed (default 1).

-P "N 1 5 10"
Sets an optimisation parameter for the classifier with name -N, lower bound 1, upper bound 5, and 10 optimisation steps. The upper bound may be the character 'A' or 'I' to substitute the number of attributes or instances in the training data, respectively. This parameter may be supplied more than once to optimise over several classifier options simultaneously.

Options after -- are passed to the designated sub-classifier.

Specified by:

setOptions in interface OptionHandler

Overrides:

setOptions in class RandomizableSingleClassifierEnhancer

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the Classifier.

Specified by:

getOptions in interface OptionHandler

Overrides:

getOptions in class RandomizableSingleClassifierEnhancer

Returns:

an array of strings suitable for passing to setOptions

buildClassifier

public void buildClassifier(Instances instances)
                     throws java.lang.Exception

Generates the classifier.

Specified by:

buildClassifier in class Classifier

Parameters:

instances - set of instances serving as training data

Throws:

java.lang.Exception - if the classifier has not been generated successfully

distributionForInstance

public double[] distributionForInstance(Instance instance)
                                 throws java.lang.Exception

Predicts the class distribution for the given test instance.

Overrides:

distributionForInstance in class Classifier

Parameters:

instance - the instance to be classified

Returns:

the predicted class value

Throws:

java.lang.Exception - if an error occurred during the prediction

addCVParameter

public void addCVParameter(java.lang.String cvParam)
                    throws java.lang.Exception

Adds a scheme parameter to the list of parameters to be set by cross-validation

Parameters:

cvParam - the string representation of a scheme parameter. The format is:
param_char lower_bound upper_bound increment
eg to search a parameter -P from 1 to 10 by increments of 2:
P 1 10 2

Throws:

java.lang.Exception - if the parameter specifier is of the wrong format

getCVParameter

public java.lang.String getCVParameter(int index)

Gets the scheme paramter with the given index.

CVParametersTipText

public java.lang.String CVParametersTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getCVParameters

public java.lang.Object[] getCVParameters()

Get method for CVParameters.

setCVParameters

public void setCVParameters(java.lang.Object[] params)
                     throws java.lang.Exception

Set method for CVParameters.

Throws:

java.lang.Exception

numFoldsTipText

public java.lang.String numFoldsTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

getNumFolds

public int getNumFolds()

Gets the number of folds for the cross-validation.

Returns:

the number of folds for the cross-validation

setNumFolds

public void setNumFolds(int numFolds)
                 throws java.lang.Exception

Sets the number of folds for the cross-validation.

Parameters:

numFolds - the number of folds for the cross-validation

Throws:

java.lang.Exception - if parameter illegal

graphType

public int graphType()

Returns the type of graph this classifier represents.

Specified by:

graphType in interface Drawable

Returns:

the type of graph representing the object

graph

public java.lang.String graph()
                       throws java.lang.Exception

Returns graph describing the classifier (if possible).

Specified by:

graph in interface Drawable

Returns:

the graph of the classifier in dotty format

Throws:

java.lang.Exception - if the classifier cannot be graphed

toString

public java.lang.String toString()

Returns description of the cross-validated classifier.

Returns:

description of the cross-validated classifier as a string

toSummaryString

public java.lang.String toSummaryString()

Description copied from interface: Summarizable

Returns a string that summarizes the object.

Specified by:

toSummaryString in interface Summarizable

Returns:

the object summarized as a string

main

public static void main(java.lang.String[] argv)

Main method for testing this class.

Parameters:

argv - the options