com.bayesserver.learning.parameters

Class DistributionSpecification

java.lang.Object

com.bayesserver.learning.parameters.DistributionSpecification

public final class DistributionSpecification
extends Object

Identifies a node's distribution to learn, and options for learning. Note that temporal nodes and noisy nodes can have multiple distributions, and each distribution is identified by temporal order.

Constructor Summary

Constructors

Constructor and Description
DistributionSpecification(Node node) Initializes a new instance of the DistributionSpecification class.
DistributionSpecification(Node node, int order) Initializes a new instance of the DistributionSpecification class.
DistributionSpecification(Node node, NodeDistributionKey key) Initializes a new instance of the DistributionSpecification class.

Method Summary

Modifier and Type	Method and Description
DiscretePriorMethod	getDiscretePriorMethod() Gets the type of discrete prior to use for this distribution.
Boolean	getInitialize() Gets a flag indicating whether the distribution should be initialized.
NodeDistributionKey	getKey() Gets the order/related node of the distribution.
Node	getNode() Gets the Node this distribution specification refers to.
int	getOrder() Gets the order of the distribution, for temporal nodes.
void	setDiscretePriorMethod(DiscretePriorMethod value) Sets the type of discrete prior to use for this distribution.
void	setInitialize(Boolean value) Sets a flag indicating whether the distribution should be initialized.

Methods inherited from class java.lang.Object

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

Constructor Detail

DistributionSpecification

public DistributionSpecification(Node node)

Initializes a new instance of the DistributionSpecification class.

Parameters:

node - The node the distribution refers to.

DistributionSpecification

public DistributionSpecification(Node node,
                                 int order)

Initializes a new instance of the DistributionSpecification class.

Parameters:

node - The node the distribution refers to.

order - The order of the distribution, for temporal nodes.

DistributionSpecification

public DistributionSpecification(Node node,
                                 NodeDistributionKey key)

Initializes a new instance of the DistributionSpecification class.

Parameters:

node - The node the distribution refers to.

key - Properties that identify the distribution, such as the temporal order.

Method Detail

getInitialize

public Boolean getInitialize()

Gets a flag indicating whether the distribution should be initialized. When nullInitializationOptions.getInitializeDistributions() will be used to determine whether initialization is performed.

Returns:

When true initialize the distribution, when false do not initialize the distributions, and when null use the value from InitializationOptions.getInitializeDistributions().

setInitialize

public void setInitialize(Boolean value)

Sets a flag indicating whether the distribution should be initialized. When nullInitializationOptions.getInitializeDistributions() will be used to determine whether initialization is performed.

Parameters:

value - When true initialize the distribution, when false do not initialize the distributions, and when null use the value from InitializationOptions.getInitializeDistributions().

getDiscretePriorMethod

public DiscretePriorMethod getDiscretePriorMethod()

Gets the type of discrete prior to use for this distribution. When nullcom.bayesserver.learning.parameters.PriorOptions#getDiscretePriorMethod will be used to determine the type of prior.

setDiscretePriorMethod

public void setDiscretePriorMethod(DiscretePriorMethod value)

Sets the type of discrete prior to use for this distribution. When nullcom.bayesserver.learning.parameters.PriorOptions#getDiscretePriorMethod will be used to determine the type of prior.

getNode

public Node getNode()

Gets the Node this distribution specification refers to.

getOrder

public int getOrder()

Gets the order of the distribution, for temporal nodes. For non temporal nodes this value will always be zero.

getKey

public NodeDistributionKey getKey()

Gets the order/related node of the distribution. For non temporal nodes this value will always be zero.