public class LAGDHillClimber extends HillClimber
-L <nr of look ahead steps> Look Ahead Depth
-G <nr of good operations> Nr of Good Operations
-P <nr of parents> Maximum number of parents
-R Use arc reversal operation. (default false)
-N Initial structure is empty (instead of Naive Bayes)
-mbc Applies a Markov Blanket correction to the network structure, after a network structure is learned. This ensures that all nodes in the network are part of the Markov blanket of the classifier node.
-S [BAYES|MDL|ENTROPY|AIC|CROSS_CLASSIC|CROSS_BAYES] Score type (BAYES, BDeu, MDL, ENTROPY and AIC)
TAGS_SCORE_TYPE
m_bInitAsNaiveBayes, m_bMarkovBlanketClassifier, m_nMaxNrOfParents
Constructor and Description |
---|
LAGDHillClimber() |
Modifier and Type | Method and Description |
---|---|
protected weka.classifiers.bayes.net.search.local.HillClimber.Operation |
getAntiOperation(weka.classifiers.bayes.net.search.local.HillClimber.Operation oOperation)
getAntiOperation determines the Operation, which is needed to cancel oOperation
|
protected weka.classifiers.bayes.net.search.local.HillClimber.Operation[] |
getGoodOperations(BayesNet bayesNet,
Instances instances,
int nrOfGoodOperations)
getGoodOperations determines the nrOfGoodOperations best Operations, which are considered for
the calculation of an optimal operationsequence
|
int |
getMaxNrOfParents()
Gets the max number of parents.
|
int |
getNrOfGoodOperations()
Gets the number of "good operations"
|
int |
getNrOfLookAheadSteps()
Gets the number of look-ahead steps
|
protected weka.classifiers.bayes.net.search.local.HillClimber.Operation[] |
getOptimalOperations(BayesNet bayesNet,
Instances instances,
int nrOfLookAheadSteps,
int nrOfGoodOperations)
getOptimalOperations determines an optimal operationsequence in respect of the parameters
nrOfLookAheadSteps and nrOfGoodOperations
|
String[] |
getOptions()
Gets the current settings of the search algorithm.
|
String |
getRevision()
Returns the revision string.
|
String |
globalInfo()
This will return a string describing the search algorithm.
|
Enumeration |
listOptions()
Returns an enumeration describing the available options.
|
protected void |
lookAheadInGoodDirectionsSearch(BayesNet bayesNet,
Instances instances,
int nrOfLookAheadSteps,
int nrOfGoodOperations)
lookAheadInGoodDirectionsSearch determines the network structure/graph of the network
with best score according to LAGD Hill Climbing
|
String |
nrOfGoodOperationsTipText() |
String |
nrOfLookAheadStepsTipText() |
protected void |
search(BayesNet bayesNet,
Instances instances)
search determines the network structure/graph of the network
|
void |
setMaxNrOfParents(int nMaxNrOfParents)
Sets the max number of parents
|
void |
setNrOfGoodOperations(int nNrOfGoodOperations)
Sets the number of "good operations"
|
void |
setNrOfLookAheadSteps(int nNrOfLookAheadSteps)
Sets the number of look-ahead steps
|
void |
setOptions(String[] options)
Parses a given list of options.
|
getInitAsNaiveBayes, getUseArcReversal, setInitAsNaiveBayes, setUseArcReversal, useArcReversalTipText
buildStructure, calcNodeScore, calcScoreOfCounts, calcScoreOfCounts2, calcScoreWithExtraParent, calcScoreWithMissingParent, getMarkovBlanketClassifier, getScoreType, logScore, markovBlanketClassifierTipText, scoreTypeTipText, setMarkovBlanketClassifier, setScoreType
addArcMakesSense, doMarkovBlanketCorrection, initAsNaiveBayesTipText, isArc, maxNrOfParentsTipText, reverseArcMakesSense, toString
protected void search(BayesNet bayesNet, Instances instances) throws Exception
search
in class HillClimber
bayesNet
- the networkinstances
- the data to useException
- if something goes wrongprotected void lookAheadInGoodDirectionsSearch(BayesNet bayesNet, Instances instances, int nrOfLookAheadSteps, int nrOfGoodOperations) throws Exception
bayesNet
- the networkinstances
- the data to usenrOfLookAheadSteps
- nrOfGoodOperations
- Exception
- if something goes wrongprotected weka.classifiers.bayes.net.search.local.HillClimber.Operation getAntiOperation(weka.classifiers.bayes.net.search.local.HillClimber.Operation oOperation) throws Exception
oOperation
- Operation to cancelException
- if something goes wrongprotected weka.classifiers.bayes.net.search.local.HillClimber.Operation[] getGoodOperations(BayesNet bayesNet, Instances instances, int nrOfGoodOperations) throws Exception
bayesNet
- Bayes network to apply operation oninstances
- data set to learn fromnrOfGoodOperations
- number of good operations to considerException
- if something goes wrongprotected weka.classifiers.bayes.net.search.local.HillClimber.Operation[] getOptimalOperations(BayesNet bayesNet, Instances instances, int nrOfLookAheadSteps, int nrOfGoodOperations) throws Exception
bayesNet
- Bayes network to apply operation oninstances
- data set to learn fromnrOfLookAheadSteps
- number of lood ahead steps to usenrOfGoodOperations
- number of good operations to considerException
- if something goes wrongpublic void setMaxNrOfParents(int nMaxNrOfParents)
setMaxNrOfParents
in class HillClimber
nMaxNrOfParents
- the max number of parentspublic int getMaxNrOfParents()
getMaxNrOfParents
in class HillClimber
public void setNrOfLookAheadSteps(int nNrOfLookAheadSteps)
nNrOfLookAheadSteps
- the number of look-ahead stepspublic int getNrOfLookAheadSteps()
public void setNrOfGoodOperations(int nNrOfGoodOperations)
nNrOfGoodOperations
- the number of "good operations"public int getNrOfGoodOperations()
public Enumeration listOptions()
listOptions
in interface OptionHandler
listOptions
in class HillClimber
public void setOptions(String[] options) throws Exception
Valid options are:
-L <nr of look ahead steps> Look Ahead Depth
-G <nr of good operations> Nr of Good Operations
-P <nr of parents> Maximum number of parents
-R Use arc reversal operation. (default false)
-N Initial structure is empty (instead of Naive Bayes)
-mbc Applies a Markov Blanket correction to the network structure, after a network structure is learned. This ensures that all nodes in the network are part of the Markov blanket of the classifier node.
-S [BAYES|MDL|ENTROPY|AIC|CROSS_CLASSIC|CROSS_BAYES] Score type (BAYES, BDeu, MDL, ENTROPY and AIC)
setOptions
in interface OptionHandler
setOptions
in class HillClimber
options
- the list of options as an array of stringsException
- if an option is not supportedpublic String[] getOptions()
getOptions
in interface OptionHandler
getOptions
in class HillClimber
public String globalInfo()
globalInfo
in class HillClimber
public String nrOfLookAheadStepsTipText()
public String nrOfGoodOperationsTipText()
public String getRevision()
getRevision
in interface RevisionHandler
getRevision
in class HillClimber
Copyright © 2015 University of Waikato, Hamilton, NZ. All rights reserved.