Module es.upm.fi.cig.multictbnc

Package es.upm.fi.cig.multictbnc.learning.structure.hybrid.PC

Class CTPCHybridAlgorithm

java.lang.Object

es.upm.fi.cig.multictbnc.learning.structure.constraintlearning.PC.CTPC

es.upm.fi.cig.multictbnc.learning.structure.hybrid.PC.CTPCHybridAlgorithm

All Implemented Interfaces:

StructureLearningAlgorithm

public class CTPCHybridAlgorithm
extends CTPC

Implements the restriction phase (CTPC algorithm) of the hybrid structure learning algorithm.

Constructor Summary

Constructors

Constructor	Description
CTPCHybridAlgorithm(int maxSizeSepSet, double sigTimeTransitionHypothesis, double sigStateToStateTransitionHypothesis)	Initialises the algorithm by proving a significance level.

Method Summary

Modifier and Type	Method	Description
boolean[][]	learnInitialStructure(PGM<? extends Node> pgm, List<Integer> idxFeatureVariables)	Learns the initial structure of a given PGM.
protected void	learnParentSetNode(PGM<? extends Node> pgm, int idxNode, boolean[][] adjacencyMatrix)	Learns the parent set of a node.

Methods inherited from class es.upm.fi.cig.multictbnc.learning.structure.constraintlearning.PC.CTPC

addSepSetAndNodeAsParents, addSepSetAsParents, buildCompleteStructure, getIdentifier, getIdxFeatureVariables, getIdxParentsNode, getParametersAlgorithm, learn, learn, learn, retrieveParametersAndSuffStatistics, testNullStateToStateTransitionHypForGivenSepSet, testNullTimeToTransitionHypForGivenSepSet

Methods inherited from class java.lang.Object

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

Constructor Details

CTPCHybridAlgorithm

public CTPCHybridAlgorithm(int maxSizeSepSet,
 double sigTimeTransitionHypothesis,
 double sigStateToStateTransitionHypothesis)

Initialises the algorithm by proving a significance level.

Parameters:

maxSizeSepSet - maximum separating set size

sigTimeTransitionHypothesis - significance level

sigStateToStateTransitionHypothesis - significance level

Method Details

learnInitialStructure

public boolean[][] learnInitialStructure(PGM<? extends Node> pgm,
 List<Integer> idxFeatureVariables)

Learns the initial structure of a given PGM. Indexes of feature variables whose parent sets will be learnt need to be specified.

Parameters:

pgm - probabilistic graphical model

idxFeatureVariables - indexes of feature variables

Returns:

initial adjacency matrix

learnParentSetNode

protected void learnParentSetNode(PGM<? extends Node> pgm,
 int idxNode,
 boolean[][] adjacencyMatrix)
                           throws ErroneousValueException

Description copied from class: CTPC

Learns the parent set of a node.

Overrides:

learnParentSetNode in class CTPC

Parameters:

pgm - probabilistic graphical model that contains the node

idxNode - index of the node whose parent set is being learnt

adjacencyMatrix - current adjacency matrix

Throws:

ErroneousValueException - if a provided parameter is erroneous for the requested task