weka.experiment

Class PairedTTester

java.lang.Object

weka.experiment.PairedTTester

All Implemented Interfaces:

Serializable, OptionHandler, RevisionHandler, Tester

Direct Known Subclasses:

PairedCorrectedTTester

public class PairedTTester
extends Object
implements OptionHandler, Tester, RevisionHandler

Calculates T-Test statistics on data stored in a set of instances.

Valid options are:

 -D <index,index2-index4,...>
  Specify list of columns that specify a unique
  dataset.
  First and last are valid indexes. (default none)

 -R <index>
  Set the index of the column containing the run number

 -F <index>
  Set the index of the column containing the fold number

 -G <index1,index2-index4,...>
  Specify list of columns that specify a unique
  'result generator' (eg: classifier name and options).
  First and last are valid indexes. (default none)

 -S <significance level>
  Set the significance level for comparisons (default 0.05)

 -V
  Show standard deviations

 -L
  Produce table comparisons in Latex table format

 -csv
  Produce table comparisons in CSV table format

 -html
  Produce table comparisons in HTML table format

 -significance
  Produce table comparisons with only the significance values

 -gnuplot
  Produce table comparisons output suitable for GNUPlot

Version:

$Revision: 8034 $

Author:

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

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
PairedTTester()

Method Summary

Methods

Modifier and Type	Method and Description
void	assign(Tester tester) retrieves all the settings from the given Tester
PairedStats	calculateStatistics(Instance datasetSpecifier, int resultset1Index, int resultset2Index, int comparisonColumn) Computes a paired t-test comparison for a specified dataset between two resultsets.
boolean	displayResultset(int index) Checks whether the resultset with the given index shall be displayed.
Range	getDatasetKeyColumns() Get the value of DatasetKeyColumns.
int[]	getDisplayedResultsets() Gets the indices of the the datasets that are displayed (if null then all are displayed).
String	getDisplayName() returns the name of the tester
int	getFoldColumn() Get the value of FoldColumn.
Instances	getInstances() Get the value of Instances.
int	getNumDatasets() Gets the number of datasets in the resultsets
int	getNumResultsets() Gets the number of resultsets in the data.
String[]	getOptions() Gets current settings of the PairedTTester.
ResultMatrix	getResultMatrix() Gets the instance that produces the output.
Range	getResultsetKeyColumns() Get the value of ResultsetKeyColumns.
String	getResultsetName(int index) Gets a string descriptive of the specified resultset.
String	getRevision() Returns the revision string.
int	getRunColumn() Get the value of RunColumn.
boolean	getShowStdDevs() Returns true if standard deviations have been requested.
double	getSignificanceLevel() Get the value of SignificanceLevel.
int	getSortColumn() Returns the column to sort on, -1 means the default sorting.
String	getSortColumnName() Returns the name of the column to sort on.
String	getToolTipText() returns a string that is displayed as tooltip on the "perform test" button in the experimenter
String	header(int comparisonColumn) Creates a "header" string describing the current resultsets.
Enumeration	listOptions() Lists options understood by this object.
static void	main(String[] args) Test the class from the command line.
String	multiResultsetFull(int baseResultset, int comparisonColumn) Creates a comparison table where a base resultset is compared to the other resultsets.
String	multiResultsetRanking(int comparisonColumn) returns a ranking of the resultsets
String	multiResultsetSummary(int comparisonColumn) Carries out a comparison between all resultsets, counting the number of datsets where one resultset outperforms the other.
int[][]	multiResultsetWins(int comparisonColumn, int[][] nonSigWin) Carries out a comparison between all resultsets, counting the number of datsets where one resultset outperforms the other.
String	resultsetKey() Creates a key that maps resultset numbers to their descriptions.
void	setDatasetKeyColumns(Range newDatasetKeyColumns) Set the value of DatasetKeyColumns.
void	setDisplayedResultsets(int[] cols) Sets the indicies of the datasets to display (null means all).
void	setFoldColumn(int newFoldColumn) Set the value of FoldColumn.
void	setInstances(Instances newInstances) Set the value of Instances.
void	setOptions(String[] options) Parses a given list of options.
void	setResultMatrix(ResultMatrix matrix) Sets the matrix to use to produce the output.
void	setResultsetKeyColumns(Range newResultsetKeyColumns) Set the value of ResultsetKeyColumns.
void	setRunColumn(int newRunColumn) Set the value of RunColumn.
void	setShowStdDevs(boolean s) Set whether standard deviations are displayed or not.
void	setSignificanceLevel(double newSignificanceLevel) Set the value of SignificanceLevel.
void	setSortColumn(int newSortColumn) Set the column to sort on, -1 means the default sorting.

Methods inherited from class java.lang.Object

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

Constructor Detail

PairedTTester

public PairedTTester()

Method Detail

setResultMatrix

public void setResultMatrix(ResultMatrix matrix)

Sets the matrix to use to produce the output.

Specified by:

setResultMatrix in interface Tester

Parameters:

matrix - the instance to use to produce the output

See Also:

ResultMatrix

getResultMatrix

public ResultMatrix getResultMatrix()

Gets the instance that produces the output.

Specified by:

getResultMatrix in interface Tester

Returns:

the instance to produce the output

setShowStdDevs

public void setShowStdDevs(boolean s)

Set whether standard deviations are displayed or not.

Specified by:

setShowStdDevs in interface Tester

Parameters:

s - true if standard deviations are to be displayed

getShowStdDevs

public boolean getShowStdDevs()

Returns true if standard deviations have been requested.

Specified by:

getShowStdDevs in interface Tester

Returns:

true if standard deviations are to be displayed.

getNumDatasets

public int getNumDatasets()

Gets the number of datasets in the resultsets

Specified by:

getNumDatasets in interface Tester

Returns:

the number of datasets in the resultsets

getNumResultsets

public int getNumResultsets()

Gets the number of resultsets in the data.

Specified by:

getNumResultsets in interface Tester

Returns:

the number of resultsets in the data

getResultsetName

public String getResultsetName(int index)

Gets a string descriptive of the specified resultset.

Specified by:

getResultsetName in interface Tester

Parameters:

index - the index of the resultset

Returns:

a descriptive string for the resultset

displayResultset

public boolean displayResultset(int index)

Checks whether the resultset with the given index shall be displayed.

Specified by:

displayResultset in interface Tester

Parameters:

index - the index of the resultset to check whether it shall be displayed

Returns:

whether the specified resultset is displayed

calculateStatistics

public PairedStats calculateStatistics(Instance datasetSpecifier,
                              int resultset1Index,
                              int resultset2Index,
                              int comparisonColumn)
                                throws Exception

Computes a paired t-test comparison for a specified dataset between two resultsets.

Specified by:

calculateStatistics in interface Tester

Parameters:

datasetSpecifier - the dataset specifier

resultset1Index - the index of the first resultset

resultset2Index - the index of the second resultset

comparisonColumn - the column containing values to compare

Returns:

the results of the paired comparison

Throws:

Exception - if an error occurs

resultsetKey

public String resultsetKey()

Creates a key that maps resultset numbers to their descriptions.

Specified by:

resultsetKey in interface Tester

Returns:

a value of type 'String'

header

public String header(int comparisonColumn)

Creates a "header" string describing the current resultsets.

Specified by:

header in interface Tester

Parameters:

comparisonColumn - a value of type 'int'

Returns:

a value of type 'String'

multiResultsetWins

public int[][] multiResultsetWins(int comparisonColumn,
                         int[][] nonSigWin)
                           throws Exception

Carries out a comparison between all resultsets, counting the number of datsets where one resultset outperforms the other.

Specified by:

multiResultsetWins in interface Tester

Parameters:

comparisonColumn - the index of the comparison column

nonSigWin - for storing the non-significant wins

Returns:

a 2d array where element [i][j] is the number of times resultset j performed significantly better than resultset i.

Throws:

Exception - if an error occurs

multiResultsetSummary

public String multiResultsetSummary(int comparisonColumn)
                             throws Exception

Carries out a comparison between all resultsets, counting the number of datsets where one resultset outperforms the other. The results are summarized in a table.

Specified by:

multiResultsetSummary in interface Tester

Parameters:

comparisonColumn - the index of the comparison column

Returns:

the results in a string

Throws:

Exception - if an error occurs

multiResultsetRanking

public String multiResultsetRanking(int comparisonColumn)
                             throws Exception

returns a ranking of the resultsets

Specified by:

multiResultsetRanking in interface Tester

Parameters:

comparisonColumn - the column to compare with

Returns:

the ranking

Throws:

Exception - if something goes wrong

multiResultsetFull

public String multiResultsetFull(int baseResultset,
                        int comparisonColumn)
                          throws Exception

Creates a comparison table where a base resultset is compared to the other resultsets. Results are presented for every dataset.

Specified by:

multiResultsetFull in interface Tester

Parameters:

baseResultset - the index of the base resultset

comparisonColumn - the index of the column to compare over

Returns:

the comparison table string

Throws:

Exception - if an error occurs

listOptions

public Enumeration listOptions()

Lists options understood by this object.

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of Options.

setOptions

public void setOptions(String[] options)
                throws Exception

Parses a given list of options.

Valid options are:

 -D <index,index2-index4,...>
  Specify list of columns that specify a unique
  dataset.
  First and last are valid indexes. (default none)

 -R <index>
  Set the index of the column containing the run number

 -F <index>
  Set the index of the column containing the fold number

 -G <index1,index2-index4,...>
  Specify list of columns that specify a unique
  'result generator' (eg: classifier name and options).
  First and last are valid indexes. (default none)

 -S <significance level>
  Set the significance level for comparisons (default 0.05)

 -V
  Show standard deviations

 -L
  Produce table comparisons in Latex table format

 -csv
  Produce table comparisons in CSV table format

 -html
  Produce table comparisons in HTML table format

 -significance
  Produce table comparisons with only the significance values

 -gnuplot
  Produce table comparisons output suitable for GNUPlot

Specified by:

setOptions in interface OptionHandler

Parameters:

options - an array containing options to set.

Throws:

Exception - if invalid options are given

getOptions

public String[] getOptions()

Gets current settings of the PairedTTester.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings containing current options.

getResultsetKeyColumns

public Range getResultsetKeyColumns()

Get the value of ResultsetKeyColumns.

Specified by:

getResultsetKeyColumns in interface Tester

Returns:

Value of ResultsetKeyColumns.

setResultsetKeyColumns

public void setResultsetKeyColumns(Range newResultsetKeyColumns)

Set the value of ResultsetKeyColumns.

Specified by:

setResultsetKeyColumns in interface Tester

Parameters:

newResultsetKeyColumns - Value to assign to ResultsetKeyColumns.

getDisplayedResultsets

public int[] getDisplayedResultsets()

Gets the indices of the the datasets that are displayed (if null then all are displayed). The base is always displayed.

Specified by:

getDisplayedResultsets in interface Tester

Returns:

the indices of the datasets to display

setDisplayedResultsets

public void setDisplayedResultsets(int[] cols)

Sets the indicies of the datasets to display (null means all). The base is always displayed.

Specified by:

setDisplayedResultsets in interface Tester

Parameters:

cols - the indices of the datasets to display

getSignificanceLevel

public double getSignificanceLevel()

Get the value of SignificanceLevel.

Specified by:

getSignificanceLevel in interface Tester

Returns:

Value of SignificanceLevel.

setSignificanceLevel

public void setSignificanceLevel(double newSignificanceLevel)

Set the value of SignificanceLevel.

Specified by:

setSignificanceLevel in interface Tester

Parameters:

newSignificanceLevel - Value to assign to SignificanceLevel.

getDatasetKeyColumns

public Range getDatasetKeyColumns()

Get the value of DatasetKeyColumns.

Specified by:

getDatasetKeyColumns in interface Tester

Returns:

Value of DatasetKeyColumns.

setDatasetKeyColumns

public void setDatasetKeyColumns(Range newDatasetKeyColumns)

Set the value of DatasetKeyColumns.

Specified by:

setDatasetKeyColumns in interface Tester

Parameters:

newDatasetKeyColumns - Value to assign to DatasetKeyColumns.

getRunColumn

public int getRunColumn()

Get the value of RunColumn.

Specified by:

getRunColumn in interface Tester

Returns:

Value of RunColumn.

setRunColumn

public void setRunColumn(int newRunColumn)

Set the value of RunColumn.

Specified by:

setRunColumn in interface Tester

Parameters:

newRunColumn - Value to assign to RunColumn.

getFoldColumn

public int getFoldColumn()

Get the value of FoldColumn.

Specified by:

getFoldColumn in interface Tester

Returns:

Value of FoldColumn.

setFoldColumn

public void setFoldColumn(int newFoldColumn)

Set the value of FoldColumn.

Specified by:

setFoldColumn in interface Tester

Parameters:

newFoldColumn - Value to assign to FoldColumn.

getSortColumnName

public String getSortColumnName()

Returns the name of the column to sort on.

Specified by:

getSortColumnName in interface Tester

Returns:

the name of the column to sort on.

getSortColumn

public int getSortColumn()

Returns the column to sort on, -1 means the default sorting.

Specified by:

getSortColumn in interface Tester

Returns:

the column to sort on.

setSortColumn

public void setSortColumn(int newSortColumn)

Set the column to sort on, -1 means the default sorting.

Specified by:

setSortColumn in interface Tester

Parameters:

newSortColumn - the new sort column.

getInstances

public Instances getInstances()

Get the value of Instances.

Specified by:

getInstances in interface Tester

Returns:

Value of Instances.

setInstances

public void setInstances(Instances newInstances)

Set the value of Instances.

Specified by:

setInstances in interface Tester

Parameters:

newInstances - Value to assign to Instances.

assign

public void assign(Tester tester)

retrieves all the settings from the given Tester

Specified by:

assign in interface Tester

Parameters:

tester - the Tester to get the settings from

getToolTipText

public String getToolTipText()

returns a string that is displayed as tooltip on the "perform test" button in the experimenter

Specified by:

getToolTipText in interface Tester

Returns:

the tool tip

getDisplayName

public String getDisplayName()

returns the name of the tester

Specified by:

getDisplayName in interface Tester

Returns:

the display name

getRevision

public String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Returns:

the revision

main

public static void main(String[] args)

Test the class from the command line.

Parameters:

args - contains options for the instance ttests