Package edu.cmu.tetrad.bayes

Class BayesPm

java.lang.Object

edu.cmu.tetrad.bayes.BayesPm

All Implemented Interfaces:

VariableSource, Pm, TetradSerializable, Serializable

public final class BayesPm
extends Object
implements Pm, VariableSource

Implements a discrete Bayes parametric model--that is, a DAG together with a map from the nodes in the graph to a set of discrete variables, specifying the number of categories for each variable and the name of each category for each variable. This is all the information one needs to know in order to determine the parametric form of a Bayes net up to actual values of parameters. Specific values for the Bayes net are stored in a BayesIM object (see).

Version:

$Id: $Id

Author:

josephramsey

See Also:

• Dag
• BayesIm
• Serialized Form

Constructor Summary

Constructors

Constructor	Description
BayesPm(BayesPm bayesPm)	Copy constructor.
BayesPm(Graph graph)	Construct a new BayesPm using the given DAG, assigning each variable two values named "value1" and "value2" unless nodes are discrete variables with categories already defined.
BayesPm(Graph graph, int lowerBound, int upperBound)	Constructs a new BayesPm from the given DAG, assigning each variable a random number of values between lowerBound and upperBound.
BayesPm(Graph graph, BayesPm oldBayesPm)	Constructs a new BayesPm using a given DAG, using as much information from the old BayesPm as possible.
BayesPm(Graph graph, BayesPm oldBayesPm, int lowerBound, int upperBound)	Constructs a new BayesPm from the given DAG, using as much information from the old BayesPm as possible.

Method Summary

Modifier and Type	Method	Description
boolean	equals(Object o)
String	getCategory(Node node, int index)	Returns the index'th value for the given node.
int	getCategoryIndex(Node node, String category)	Returns the index of the given category for the given node.
Graph	getDag()	Returns the DAG.
List<Node>	getMeasuredNodes()	Returns the measured nodes.
Node	getNode(int index)	Returns the node at the given index.
Node	getNode(String nodeName)	Returns the node by the given name.
int	getNodeIndex()	Returns the node index.
int	getNumCategories(Node node)	Returns the number of values for the given node.
int	getNumNodes()	Returns the number of nodes.
static List<String>	getParameterNames()	Returns the parameter names.
Node	getVariable(Node node)	Returns the variable for the given node.
List<String>	getVariableNames()	Returns the variable names.
List<Node>	getVariables()	getVariables.
static BayesPm	serializableInstance()	Generates a simple exemplar of this class to test serialization.
void	setCategories(Node node, List<String> categories)	Sets the number of values for the given node to the given number.
void	setNumCategories(Node node, int numCategories)	Sets the number of values for the given node to the given number.
String	toString()	Prints out the list of values for each node.

Methods inherited from class java.lang.Object

getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Details

BayesPm

public BayesPm(Graph graph)

Construct a new BayesPm using the given DAG, assigning each variable two values named "value1" and "value2" unless nodes are discrete variables with categories already defined.

Parameters:

graph - Ibid.

BayesPm

public BayesPm(Graph graph,
 BayesPm oldBayesPm)

Constructs a new BayesPm using a given DAG, using as much information from the old BayesPm as possible.

Parameters:

graph - Ibid.

oldBayesPm - Ibid.

BayesPm

public BayesPm(Graph graph,
 int lowerBound,
 int upperBound)

Constructs a new BayesPm from the given DAG, assigning each variable a random number of values between lowerBound and upperBound. Uses a fixed number of values if lowerBound == upperBound. The values are named "value1" ... "valuen".

Parameters:

graph - Ibid.

lowerBound - Ibid.

upperBound - Ibid.

BayesPm

public BayesPm(Graph graph,
 BayesPm oldBayesPm,
 int lowerBound,
 int upperBound)

Constructs a new BayesPm from the given DAG, using as much information from the old BayesPm as possible. For variables not in the old BayesPm, assigns each variable a random number of values between lowerBound and upperBound. Uses a fixed number of values if lowerBound == upperBound. The values are named "value1" ... "valuen".

Parameters:

graph - Ibid.

oldBayesPm - Ibid.

lowerBound - Ibid.

upperBound - Ibid.

BayesPm

public BayesPm(BayesPm bayesPm)

Copy constructor.

Parameters:

bayesPm - Ibid.

Method Details

serializableInstance

public static BayesPm serializableInstance()

Generates a simple exemplar of this class to test serialization.

Returns:

Ibid.

getParameterNames

public static List<String> getParameterNames()

Returns the parameter names.

Returns:

Ibid.

getDag

public Graph getDag()

Returns the DAG.

Returns:

the DAG as a Graph.

getNumCategories

public int getNumCategories(Node node)

Returns the number of values for the given node.

Parameters:

node - Ibid.

Returns:

the number of values for the given node.

getCategory

public String getCategory(Node node,
 int index)

Returns the index'th value for the given node.

Parameters:

node - Ibid.

index - Ibid.

Returns:

the index'th value for the given node.

getCategoryIndex

public int getCategoryIndex(Node node,
 String category)

Returns the index of the given category for the given node.

Parameters:

node - Ibid.

category - Ibid.

Returns:

the index of the given category for the given node.

setNumCategories

public void setNumCategories(Node node,
 int numCategories)

Sets the number of values for the given node to the given number.

Parameters:

node - Ibid.

numCategories - Ibid.

equals

public boolean equals(Object o)

Will return true if the argument is a BayesPm with the same graph and variables.

Overrides:

equals in class Object

setCategories

public void setCategories(Node node,
 List<String> categories)

Sets the number of values for the given node to the given number.

Parameters:

node - Ibid.

categories - Ibid.

getVariables

public List<Node> getVariables()

getVariables.

Specified by:

getVariables in interface VariableSource

Returns:

a List object

getVariableNames

public List<String> getVariableNames()

Returns the variable names.

Specified by:

getVariableNames in interface VariableSource

Returns:

Ibid.

getVariable

public Node getVariable(Node node)

Returns the variable for the given node.

Parameters:

node - Ibid.

Returns:

Ibid.

getMeasuredNodes

public List<Node> getMeasuredNodes()

Returns the measured nodes.

Returns:

the list of measured variableNodes.

toString

public String toString()

Prints out the list of values for each node.

Overrides:

toString in class Object

Returns:

Ibid.

getNode

public Node getNode(String nodeName)

Returns the node by the given name.

Parameters:

nodeName - Ibid.

Returns:

Ibid.

getNode

public Node getNode(int index)

Returns the node at the given index.

Parameters:

index - Ibid.

Returns:

Ibid.

getNodeIndex

public int getNodeIndex()

Returns the node index.

Returns:

-1.

getNumNodes

public int getNumNodes()

Returns the number of nodes.

Returns:

Ibid.