org.deidentifier.arx.aggregates

Class ClassificationConfigurationRandomForest

java.lang.Object

org.deidentifier.arx.ARXClassificationConfiguration<ClassificationConfigurationRandomForest>

org.deidentifier.arx.aggregates.ClassificationConfigurationRandomForest

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable

public class ClassificationConfigurationRandomForest
extends ARXClassificationConfiguration<ClassificationConfigurationRandomForest>
implements java.io.Serializable, java.lang.Cloneable

Configuration for Random Forest classifiers

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	ClassificationConfigurationRandomForest.SplitRule Split rule for the decision tree

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_MAXMIMUM_NUMBER_OF_LEAF_NODES Default value
static int	DEFAULT_MINIMUM_SIZE_OF_LEAF_NODES Default value
static int	DEFAULT_NUMBER_OF_TREES Default value
static int	DEFAULT_NUMBER_OF_VARIABLES_TO_SPLIT Default value = 0: sqrt(#features) seems to provide good results
static ClassificationConfigurationRandomForest.SplitRule	DEFAULT_SPLIT_RULE Split rule
static double	DEFAULT_SUBSAMPLE 1.0 = sampling with replacement; <1.0 = sampling without replacement

Fields inherited from class org.deidentifier.arx.ARXClassificationConfiguration

DEFAULT_DETERMINISTIC, DEFAULT_MAX_RECORDS, DEFAULT_NUMBER_OF_FOLDS, DEFAULT_VECTOR_LENGTH

Method Summary

Modifier and Type	Method and Description
ClassificationConfigurationRandomForest	clone()
static ClassificationConfigurationRandomForest	create() Returns a new instance
int	getMaximumNumberOfLeafNodes()
int	getMinimumSizeOfLeafNodes()
int	getNumberOfTrees()
int	getNumberOfVariablesToSplit()
ClassificationConfigurationRandomForest.SplitRule	getSplitRule()
double	getSubsample()
void	parse(ARXClassificationConfiguration<?> config) Parses another configuration
ClassificationConfigurationRandomForest	setMaximumNumberOfLeafNodes(int maximumNumberOfLeafNodes)
ClassificationConfigurationRandomForest	setMinimumSizeOfLeafNodes(int minimumSizeOfLeafNodes)
ClassificationConfigurationRandomForest	setNumberOfTrees(int numberOfTrees)
ClassificationConfigurationRandomForest	setNumberOfVariablesToSplit(int numberOfVariablesToSplit)
ClassificationConfigurationRandomForest	setSplitRule(ClassificationConfigurationRandomForest.SplitRule splitRule)
ClassificationConfigurationRandomForest	setSubsample(double subsample)

Methods inherited from class org.deidentifier.arx.ARXClassificationConfiguration

createLogisticRegression, createNaiveBayes, createRandomForest, getMaxRecords, getNumFolds, getSeed, getVectorLength, isDeterministic, isModified, setDeterministic, setMaxRecords, setModified, setNumFolds, setSeed, setUnmodified, setVectorLength

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_NUMBER_OF_TREES

public static final int DEFAULT_NUMBER_OF_TREES

Default value

See Also:

Constant Field Values

DEFAULT_NUMBER_OF_VARIABLES_TO_SPLIT

public static final int DEFAULT_NUMBER_OF_VARIABLES_TO_SPLIT

Default value = 0: sqrt(#features) seems to provide good results

See Also:

Constant Field Values

DEFAULT_MINIMUM_SIZE_OF_LEAF_NODES

public static final int DEFAULT_MINIMUM_SIZE_OF_LEAF_NODES

Default value

See Also:

Constant Field Values

DEFAULT_MAXMIMUM_NUMBER_OF_LEAF_NODES

public static final int DEFAULT_MAXMIMUM_NUMBER_OF_LEAF_NODES

Default value

See Also:

Constant Field Values

DEFAULT_SUBSAMPLE

public static final double DEFAULT_SUBSAMPLE

1.0 = sampling with replacement; <1.0 = sampling without replacement

See Also:

Constant Field Values

DEFAULT_SPLIT_RULE

public static final ClassificationConfigurationRandomForest.SplitRule DEFAULT_SPLIT_RULE

Split rule

Method Detail

create

public static ClassificationConfigurationRandomForest create()

Returns a new instance

Returns:

clone

public ClassificationConfigurationRandomForest clone()

Specified by:

clone in class ARXClassificationConfiguration<ClassificationConfigurationRandomForest>

getMaximumNumberOfLeafNodes

public int getMaximumNumberOfLeafNodes()

Returns:

the maximumNumberOfLeafNodes

getMinimumSizeOfLeafNodes

public int getMinimumSizeOfLeafNodes()

Returns:

the minimumSizeOfLeafNodes

getNumberOfTrees

public int getNumberOfTrees()

Returns:

the numberOfTrees

getNumberOfVariablesToSplit

public int getNumberOfVariablesToSplit()

Returns:

the numberOfVariablesToSplit

getSplitRule

public ClassificationConfigurationRandomForest.SplitRule getSplitRule()

Returns:

the splitRule

getSubsample

public double getSubsample()

Returns:

the subsample

parse

public void parse(ARXClassificationConfiguration<?> config)

Description copied from class: ARXClassificationConfiguration

Parses another configuration

Overrides:

parse in class ARXClassificationConfiguration<ClassificationConfigurationRandomForest>

setMaximumNumberOfLeafNodes

public ClassificationConfigurationRandomForest setMaximumNumberOfLeafNodes(int maximumNumberOfLeafNodes)

Parameters:

maximumNumberOfLeafNodes - the maximumNumberOfLeafNodes to set

setMinimumSizeOfLeafNodes

public ClassificationConfigurationRandomForest setMinimumSizeOfLeafNodes(int minimumSizeOfLeafNodes)

Parameters:

minimumSizeOfLeafNodes - the minimumSizeOfLeafNodes to set

setNumberOfTrees

public ClassificationConfigurationRandomForest setNumberOfTrees(int numberOfTrees)

Parameters:

numberOfTrees - the numberOfTrees to set

setNumberOfVariablesToSplit

public ClassificationConfigurationRandomForest setNumberOfVariablesToSplit(int numberOfVariablesToSplit)

Parameters:

numberOfVariablesToSplit - the numberOfVariablesToSplit to set

setSplitRule

public ClassificationConfigurationRandomForest setSplitRule(ClassificationConfigurationRandomForest.SplitRule splitRule)

Parameters:

splitRule - the splitRule to set

setSubsample

public ClassificationConfigurationRandomForest setSubsample(double subsample)

Parameters:

subsample - the subsample to set