meka.classifiers.multilabel

Class PMCC

java.lang.Object

weka.classifiers.AbstractClassifier

weka.classifiers.SingleClassifierEnhancer

meka.classifiers.multilabel.ProblemTransformationMethod

meka.classifiers.multilabel.CC

meka.classifiers.multilabel.MCC

meka.classifiers.multilabel.PMCC

All Implemented Interfaces:

Serializable, Cloneable, MultiLabelClassifier, MultiTargetCapable, MultiLabelDrawable, weka.classifiers.Classifier, weka.core.BatchPredictor, weka.core.CapabilitiesHandler, weka.core.CapabilitiesIgnorer, weka.core.CommandlineRunnable, weka.core.OptionHandler, weka.core.Randomizable, weka.core.RevisionHandler, weka.core.TechnicalInformationHandler

public class PMCC
extends MCC

PMCC.java - Like MCC but creates a population of M chains at training time (from Is candidate chains, using Monte Carlo sampling), and uses this population for inference at test time; If you are looking for a 'more typical' majority-vote ensemble method, use something like EnsembleML or BaggingML with MCC.

NOTE: this implementation was faster, because the chain was only rebuilt from the first node which was different -- this is no longer the case (due to updates to way classifier chains works, using the CNode class).

Version:

Sep 2014

Author:

Jesse Read

See Also:

MCC, Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected CC[]	h
protected double	m_Beta
protected int	m_M
protected int	m_O
protected double[]	w

Fields inherited from class meka.classifiers.multilabel.MCC

m_Is, m_Iy, m_Payoff

Fields inherited from class meka.classifiers.multilabel.CC

confidences, m_Chain, m_R, m_S, nodes

Fields inherited from class meka.classifiers.multilabel.ProblemTransformationMethod

m_InstancesTemplate

Fields inherited from class weka.classifiers.SingleClassifierEnhancer

m_Classifier

Fields inherited from class weka.classifiers.AbstractClassifier

BATCH_SIZE_DEFAULT, m_BatchSize, m_Debug, m_DoNotCheckCapabilities, m_numDecimalPlaces, NUM_DECIMAL_PLACES_DEFAULT

Fields inherited from interface meka.core.MultiLabelDrawable

BayesNet, Newick, NOT_DRAWABLE, TREE

Constructor Summary

Constructors

Constructor and Description
PMCC()

Method Summary

Modifier and Type	Method and Description
String	betaTipText()
protected CC	buildCC(int[] s, weka.core.Instances D) BuildCC - Build a CC of chain-order 's' on dataset 'D'.
void	buildClassifier(weka.core.Instances D)
double[]	distributionForInstance(weka.core.Instance x)
double	getBeta() Get the temperature factor
protected static CC	getClosest(HashMap<String,CC> map, String sequence) GetClosest - returns the 'CC' in 'map' which is built on the sequence most matched to 'sequence'.
int	getM() Get the population size
int	getO() Get the temperature switch
String[]	getOptions()
String	globalInfo() Description to display in the GUI.
Enumeration	listOptions()
static void	main(String[] args)
String	mTipText()
String	oTipText()
static int[]	pi(int[] s, Random r, int t, double beta) pi - proposal distribution; swap elements in s, depending on iteration t (temperature).
protected CC	rebuildCC(CC h_old, int[] s_new, weka.core.Instances D) RebuildCC - rebuild a classifier chain 'h_old' to have a new sequence 's_new'.
void	setBeta(double t) Set the temperature factor
void	setM(int M) Set the population size
void	setO(int t) Set the temperature switch
void	setOptions(String[] options)

Methods inherited from class meka.classifiers.multilabel.MCC

chainIterationsTipText, getChainIterations, getInferenceIterations, getPayoff, getTechnicalInformation, inferenceIterationsTipText, payoff, payoffTipText, setChainIterations, setInferenceIterations, setPayoff

Methods inherited from class meka.classifiers.multilabel.CC

getConfidences, getDefaultSeed, getModel, getSeed, getTransformTemplates, graph, graphType, prepareChain, prepareChain, probabilityForInstance, rebuildClassifier, retrieveChain, sampleForInstance, sampleForInstanceFast, seedTipText, setSeed, toString, transformInstance

Methods inherited from class meka.classifiers.multilabel.ProblemTransformationMethod

defaultClassifierString, evaluation, getCapabilities, getRevision, getTemplate, makeCopies, runClassifier, testCapabilities

Methods inherited from class weka.classifiers.SingleClassifierEnhancer

classifierTipText, defaultClassifierOptions, getClassifier, getClassifierSpec, postExecution, preExecution, setClassifier

Methods inherited from class weka.classifiers.AbstractClassifier

batchSizeTipText, classifyInstance, debugTipText, distributionsForInstances, doNotCheckCapabilitiesTipText, forName, getBatchSize, getDebug, getDoNotCheckCapabilities, getNumDecimalPlaces, implementsMoreEfficientBatchPrediction, makeCopies, makeCopy, numDecimalPlacesTipText, run, runClassifier, setBatchSize, setDebug, setDoNotCheckCapabilities, setNumDecimalPlaces

Methods inherited from class java.lang.Object

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

Methods inherited from interface meka.classifiers.multilabel.MultiLabelClassifier

debugTipText, getDebug, setDebug

Methods inherited from interface weka.classifiers.Classifier

classifyInstance

Field Detail

m_M

protected int m_M

m_O

protected int m_O

m_Beta

protected double m_Beta

h

protected CC[] h

w

protected double[] w

Constructor Detail

PMCC

public PMCC()

Method Detail

getClosest

protected static CC getClosest(HashMap<String,CC> map,
                               String sequence)

GetClosest - returns the 'CC' in 'map' which is built on the sequence most matched to 'sequence'.

rebuildCC

protected CC rebuildCC(CC h_old,
                       int[] s_new,
                       weka.core.Instances D)
                throws Exception

RebuildCC - rebuild a classifier chain 'h_old' to have a new sequence 's_new'.

Throws:

Exception

buildCC

protected CC buildCC(int[] s,
                     weka.core.Instances D)
              throws Exception

BuildCC - Build a CC of chain-order 's' on dataset 'D'.

Throws:

Exception

pi

public static int[] pi(int[] s,
                       Random r,
                       int t,
                       double beta)

pi - proposal distribution; swap elements in s, depending on iteration t (temperature).
TODO - make faster!

Parameters:

s - a chain sequence

r - a random number generator

t - the current iteration

Returns:

s' ~ p(s'|s)

buildClassifier

public void buildClassifier(weka.core.Instances D)
                     throws Exception

Specified by:

buildClassifier in interface weka.classifiers.Classifier

Overrides:

buildClassifier in class MCC

Throws:

Exception

distributionForInstance

public double[] distributionForInstance(weka.core.Instance x)
                                 throws Exception

Specified by:

distributionForInstance in interface weka.classifiers.Classifier

Overrides:

distributionForInstance in class MCC

Throws:

Exception

listOptions

public Enumeration listOptions()

Specified by:

listOptions in interface weka.core.OptionHandler

Overrides:

listOptions in class MCC

setOptions

public void setOptions(String[] options)
                throws Exception

Specified by:

setOptions in interface weka.core.OptionHandler

Overrides:

setOptions in class MCC

Throws:

Exception

getOptions

public String[] getOptions()

Specified by:

getOptions in interface weka.core.OptionHandler

Overrides:

getOptions in class MCC

setBeta

public void setBeta(double t)

Set the temperature factor

getBeta

public double getBeta()

Get the temperature factor

betaTipText

public String betaTipText()

setO

public void setO(int t)

Set the temperature switch

getO

public int getO()

Get the temperature switch

oTipText

public String oTipText()

setM

public void setM(int M)

Set the population size

getM

public int getM()

Get the population size

mTipText

public String mTipText()

globalInfo

public String globalInfo()

Description copied from class: CC

Description to display in the GUI.

Overrides:

globalInfo in class MCC

Returns:

the description

main

public static void main(String[] args)