unbbayes.prs.hybridbn
Class CNNormalDistribution

java.lang.Object
  unbbayes.prs.hybridbn.CNNormalDistribution

public class CNNormalDistribution
extends Object

This class represents the continuous node normal distribution.

Author:

Rommel Carvalho (rommel.carvalho@gmail.com)

Nested Class Summary

protected class

CNNormalDistribution.NormalDistributionFunction Follows SumOf(k[i] * CPND[i]) + N(mean, variance), for all continuous node (i).

Field Summary

protected ContinuousNode	cNode
protected List<Node>	continuousParentList
protected List<Node>	discreteParentList
protected int[]	factors
protected CNNormalDistribution.NormalDistributionFunction[]	ndfList

Constructor Summary

CNNormalDistribution(ContinuousNode cNode)

Method Summary

protected void	calculateFactors() Calculate the factors necessary to transform the linear coordinate into a multidimensional one (which is the the state for each possible discrete parent node).
int	functionSize() Returns the number of normal distribution functions.
double	getConstantAt(int constantIndex, int index)
double	getConstantAt(int constantIndex, int[] multidimensionalCoord)
int	getConstantListSize()
List<Node>	getContinuousParentList()
List<Node>	getDiscreteParentList()
int	getLinearCoord(int[] multidimensionalCoord) Get the linear coordinate from the multidimensional one.
double	getMean(int index)
double	getMean(int[] multidimensionalCoord)
int[]	getMultidimensionalCoord(int linearCoord) Get the multidimensional coordinate from the linear one.
double	getVariance(int index)
double	getVariance(int[] multidimensionalCoord)
void	refreshParents() Must be called when there is some change in the parents.
void	setConstantAt(int constantIndex, double value, int index) Set the constant that multiplies the continuous parent node at index to the given value for the given combination of discrete parent node's states, which is the multidimensional coordinate.
void	setConstantAt(int constantIndex, double value, int[] multidimensionalCoord) Set the constant that multiplies the continuous parent node at index to the given value for the given combination of discrete parent node's states, which is the multidimensional coordinate.
void	setMean(double mean, int index) Set the normal distribution mean for the given combination of discrete parent node's states, which is represented by the given index.
void	setMean(double mean, int[] multidimensionalCoord) Set the normal distribution mean for the given combination of discrete parent node's states, which is the multidimensional coordinate.
void	setVariance(double variance, int index) Set the normal distribution variance for the given combination of discrete parent node's states, which is represented by the given index.
void	setVariance(double variance, int[] multidimensionalCoord) Set the normal distribution variance for the given combination of discrete parent node's states, which is the multidimensional coordinate.

Methods inherited from class java.lang.Object

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

Field Detail

cNode

protected ContinuousNode cNode

ndfList

protected CNNormalDistribution.NormalDistributionFunction[] ndfList

discreteParentList

protected List<Node> discreteParentList

continuousParentList

protected List<Node> continuousParentList

factors

protected int[] factors

Constructor Detail

CNNormalDistribution

public CNNormalDistribution(ContinuousNode cNode)

Method Detail

getDiscreteParentList

public List<Node> getDiscreteParentList()

getContinuousParentList

public List<Node> getContinuousParentList()

refreshParents

public void refreshParents()

Must be called when there is some change in the parents. The order of discrete parents list and continuous parents list is always ascendent by its name. This is important to know to which parent the respective index refers.

setConstantAt

public void setConstantAt(int constantIndex,
                          double value,
                          int[] multidimensionalCoord)

Set the constant that multiplies the continuous parent node at index to the given value for the given combination of discrete parent node's states, which is the multidimensional coordinate.

Parameters:

constantIndex - The continuous parent node index.

value - The new value for the constant.

multidimensionalCoord - The multidimensional coordinate which is the state associated with each possible discrete parent node.

setConstantAt

public void setConstantAt(int constantIndex,
                          double value,
                          int index)

Set the constant that multiplies the continuous parent node at index to the given value for the given combination of discrete parent node's states, which is the multidimensional coordinate.

Parameters:

constantIndex - The continuous parent node index.

value - The new value for the constant.

index - The index which is the state associated with each possible discrete parent node.

getConstantAt

public double getConstantAt(int constantIndex,
                            int[] multidimensionalCoord)

getConstantAt

public double getConstantAt(int constantIndex,
                            int index)

getConstantListSize

public int getConstantListSize()

setMean

public void setMean(double mean,
                    int[] multidimensionalCoord)

Set the normal distribution mean for the given combination of discrete parent node's states, which is the multidimensional coordinate.

Parameters:

mean - The normal distribution mean.

multidimensionalCoord - The multidimensional coordinate which is the state associated with each possible discrete parent node.

setMean

public void setMean(double mean,
                    int index)

Set the normal distribution mean for the given combination of discrete parent node's states, which is represented by the given index.

Parameters:

mean - The normal distribution mean.

index - The index which is the state associated with each possible discrete parent node.

getMean

public double getMean(int[] multidimensionalCoord)

getMean

public double getMean(int index)

setVariance

public void setVariance(double variance,
                        int[] multidimensionalCoord)

Set the normal distribution variance for the given combination of discrete parent node's states, which is the multidimensional coordinate.

Parameters:

variance - The normal distribution variance.

multidimensionalCoord - The multidimensional coordinate which is the state associated with each possible discrete parent node.

setVariance

public void setVariance(double variance,
                        int index)

Set the normal distribution variance for the given combination of discrete parent node's states, which is represented by the given index.

Parameters:

variance - The normal distribution variance.

index - The index which is the state associated with each possible discrete parent node.

getVariance

public double getVariance(int[] multidimensionalCoord)

getVariance

public double getVariance(int index)

calculateFactors

protected void calculateFactors()

Calculate the factors necessary to transform the linear coordinate into a multidimensional one (which is the the state for each possible discrete parent node). FactorForNode[i + 1] = ProductOf(NumberOfStates[i]), for all previous discrete parent nodes (i).

getLinearCoord

public final int getLinearCoord(int[] multidimensionalCoord)

Get the linear coordinate from the multidimensional one. LinearCoord = SumOf(StateOf[i] * FactorOf[i]), for all possible discrete parent nodes (i).

Parameters:

multidimensionalCoord - Multidimensional coordinate (represented by the state for each discrete parent node).

Returns:

The corresponding linear coordinate.

getMultidimensionalCoord

public final int[] getMultidimensionalCoord(int linearCoord)

Get the multidimensional coordinate from the linear one.

Parameters:

linearCoord - The linear coordinate.

Returns:

The corresponding multidimensional coordinate.

functionSize

public int functionSize()

Returns the number of normal distribution functions.

Returns:

The number of normal distribution functions.