Package weka.classifiers

Class LeaveOneOutByValueGenerator

java.lang.Object

adams.core.logging.LoggingObject

adams.core.logging.CustomLoggingLevelObject

adams.core.option.AbstractOptionHandler

weka.classifiers.AbstractSplitGenerator

weka.classifiers.LeaveOneOutByValueGenerator

All Implemented Interfaces:

Destroyable, GlobalInfoSupporter, LoggingLevelHandler, LoggingSupporter, OptionHandler, Randomizable, SizeOfHandler, SplitGenerator<weka.core.Instances,WekaTrainTestSetContainer>, InstancesViewSupporter, Serializable, Iterator<WekaTrainTestSetContainer>, SplitGenerator

public class LeaveOneOutByValueGenerator
extends AbstractSplitGenerator

Generates train/test split pairs using the unique values from the specified attribute. All values apart from one will be part of the training data and the remainder the test set. This is repeated for each unique value.

The template for the relation name accepts the following placeholders:

Author:

FracPete (fracpete at waikato dot ac dot nz)

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
protected int	m_CurrentPair	the current train/test pair to generate.
protected WekaAttributeIndex	m_Index	the index to get the unique values from.
protected boolean	m_Randomize	whether to randomize the data.
protected String	m_RelationName	the template for the relation name.
protected List<String>	m_UniqueValues	the unique values.
static String	PLACEHOLDER_CURRENTVALUE	the placeholder for the current value in the test set.
static String	PLACEHOLDER_ORIGINAL	the placeholder for the (original) relation name.
static String	PLACEHOLDER_TYPE	the placeholder for "train" or "test" type.

Fields inherited from class weka.classifiers.AbstractSplitGenerator

m_Data, m_Initialized, m_OriginalIndices, m_Seed, m_UseViews

Fields inherited from class adams.core.option.AbstractOptionHandler

m_OptionManager

Fields inherited from class adams.core.logging.LoggingObject

m_Logger, m_LoggingIsEnabled, m_LoggingLevel

Constructor Summary

Constructors

Constructor	Description
LeaveOneOutByValueGenerator()	Initializes the generator.
LeaveOneOutByValueGenerator(weka.core.Instances data, long seed, boolean randomize, WekaAttributeIndex index)	Initializes the generator.

Method Summary

Modifier and Type	Method	Description
protected boolean	canRandomize()	Returns whether randomization is enabled.
protected boolean	checkNext()	Returns true if the iteration has more elements.
protected WekaTrainTestSetContainer	createNext()	Returns the next element in the iteration.
protected String	createRelationName(String relation, String template, String current, boolean train)	Generates a relation name for the current value.
void	defineOptions()	Adds options to the internal list of options.
protected void	doInitializeIterator()	Initializes the iterator, randomizes the data if required.
WekaAttributeIndex	getIndex()	Returns the attribute index to use for grouping.
boolean	getRandomize()	Returns whether to randomize the data.
String	getRelationName()	Returns the relation name template.
String	globalInfo()	Returns a string describing the object.
String	indexTipText()	Returns the tip text for this property.
String	randomizeTipText()	Returns the tip text for this property.
static String	relationNameTipText()	Returns the tiptext for the relation name template.
protected void	reset()	Resets the generator.
void	setIndex(WekaAttributeIndex value)	Sets the attribute index to use for grouping.
void	setRandomize(boolean value)	Sets whether to randomize the data.
void	setRelationName(String value)	Sets the template for the relation name.
String	toString()	Returns a short description of the generator.

Methods inherited from class weka.classifiers.AbstractSplitGenerator

getData, getSeed, getUseViews, hasNext, initialize, initializeIterator, next, randomize, remove, seedTipText, setData, setSeed, setUseViews, useViewsTipText

Methods inherited from class adams.core.option.AbstractOptionHandler

cleanUpOptions, destroy, finishInit, getDefaultLoggingLevel, getOptionManager, loggingLevelTipText, newOptionManager, setLoggingLevel, toCommandLine

Methods inherited from class adams.core.logging.LoggingObject

configureLogger, getLogger, getLoggingLevel, initializeLogging, isLoggingEnabled, sizeOf

Methods inherited from class java.lang.Object

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

Methods inherited from interface adams.core.Destroyable

destroy

Methods inherited from interface java.util.Iterator

forEachRemaining

Methods inherited from interface adams.core.logging.LoggingLevelHandler

getLoggingLevel

Methods inherited from interface adams.core.option.OptionHandler

cleanUpOptions, getOptionManager, toCommandLine

Field Detail

PLACEHOLDER_ORIGINAL

public static final String PLACEHOLDER_ORIGINAL

the placeholder for the (original) relation name.

See Also:

Constant Field Values

PLACEHOLDER_TYPE

public static final String PLACEHOLDER_TYPE

the placeholder for "train" or "test" type.

See Also:

Constant Field Values

PLACEHOLDER_CURRENTVALUE

public static final String PLACEHOLDER_CURRENTVALUE

the placeholder for the current value in the test set.

See Also:

Constant Field Values

m_RelationName

protected String m_RelationName

the template for the relation name.

m_Randomize

protected boolean m_Randomize

whether to randomize the data.

m_Index

protected WekaAttributeIndex m_Index

the index to get the unique values from.

m_UniqueValues

protected transient List<String> m_UniqueValues

the unique values.

m_CurrentPair

protected transient int m_CurrentPair

the current train/test pair to generate.

Constructor Detail

LeaveOneOutByValueGenerator

public LeaveOneOutByValueGenerator()

Initializes the generator.

LeaveOneOutByValueGenerator

public LeaveOneOutByValueGenerator(weka.core.Instances data,
                                   long seed,
                                   boolean randomize,
                                   WekaAttributeIndex index)

Initializes the generator.

Parameters:

data - the full dataset

seed - the seed for randomization

randomize - whether to randomize the data

index - the attribute index

Method Detail

globalInfo

public String globalInfo()

Returns a string describing the object.

Specified by:

globalInfo in interface GlobalInfoSupporter

Specified by:

globalInfo in class AbstractOptionHandler

Returns:

a description suitable for displaying in the gui

defineOptions

public void defineOptions()

Adds options to the internal list of options.

Specified by:

defineOptions in interface OptionHandler

Overrides:

defineOptions in class AbstractSplitGenerator

reset

protected void reset()

Resets the generator.

Overrides:

reset in class AbstractSplitGenerator

setIndex

public void setIndex(WekaAttributeIndex value)

Sets the attribute index to use for grouping.

Parameters:

value - the index

getIndex

public WekaAttributeIndex getIndex()

Returns the attribute index to use for grouping.

Returns:

the index

indexTipText

public String indexTipText()

Returns the tip text for this property.

Returns:

tip text for this property suitable for displaying in the GUI or for listing the options.

setRandomize

public void setRandomize(boolean value)

Sets whether to randomize the data.

Parameters:

value - true if to randomize the data

getRandomize

public boolean getRandomize()

Returns whether to randomize the data.

Returns:

true if to randomize the data

randomizeTipText

public String randomizeTipText()

Returns the tip text for this property.

Returns:

tip text for this property suitable for displaying in the GUI or for listing the options.

setRelationName

public void setRelationName(String value)

Sets the template for the relation name.

Parameters:

value - the template

getRelationName

public String getRelationName()

Returns the relation name template.

Returns:

the template

relationNameTipText

public static String relationNameTipText()

Returns the tiptext for the relation name template.

Returns:

the tiptext

canRandomize

protected boolean canRandomize()

Returns whether randomization is enabled.

Specified by:

canRandomize in class AbstractSplitGenerator

Returns:

true if to randomize

checkNext

protected boolean checkNext()

Returns true if the iteration has more elements. (In other words, returns true if next would return an element rather than throwing an exception.)

Specified by:

checkNext in class AbstractSplitGenerator

Returns:

true if the iterator has more elements.

doInitializeIterator

protected void doInitializeIterator()

Initializes the iterator, randomizes the data if required.

Specified by:

doInitializeIterator in class AbstractSplitGenerator

See Also:

AbstractSplitGenerator.canRandomize()

createRelationName

protected String createRelationName(String relation,
                                    String template,
                                    String current,
                                    boolean train)

Generates a relation name for the current value.

Parameters:

relation - the original relation name

template - the template for the relation name

current - the current value

train - whether train or test set

Returns:

the relation name

createNext

protected WekaTrainTestSetContainer createNext()

Returns the next element in the iteration.

Specified by:

createNext in class AbstractSplitGenerator

Returns:

the next element in the iteration.

Throws:

NoSuchElementException - iteration has no more elements.

toString

public String toString()

Returns a short description of the generator.

Specified by:

toString in interface SplitGenerator<weka.core.Instances,WekaTrainTestSetContainer>

Specified by:

toString in interface SplitGenerator

Overrides:

toString in class AbstractSplitGenerator

Returns:

a short description