Package com.bayesserver.optimization

Class DesignVariable

java.lang.Object

com.bayesserver.optimization.DesignVariable

public final class DesignVariable
extends Object

Specifies on or more inputs to the optimization algorithm.

Constructor Summary

Constructors

Constructor	Description
DesignVariable(Variable variable, Double lowerBound, Double upperBound, boolean allowMissing)	Initializes a new instance of the com.bayesserver.optization.DesignVariable class, automatically generating the necessary design states.
DesignVariable(Variable variable, Double lowerBound, Double upperBound, boolean allowMissing, InterventionType interventionType)	Initializes a new instance of the DesignVariable class, automatically generating the necessary design states.
DesignVariable(Variable variable, List<DesignState> designStates, boolean allowMissing)	Initializes a new instance of the DesignVariable class.
DesignVariable(Variable variable, List<DesignState> designStates, DesignEvidenceKind evidenceKind, boolean allowMissing, InterventionType interventionType)	Initializes a new instance of the DesignVariable class.

Method Summary

Modifier and Type	Method	Description
boolean	getAllowMissing()	Determines whether the optimizer can consider missing values (evidence not set) on this variable.
List<DesignState>	getDesignStates()	Gets the design states, one for each state in the variable.
DesignEvidenceKind	getEvidenceKind()	Determines whether the optimizer uses hard or soft/virtual evidence for this variable.
InterventionType	getInterventionType()	Determines the evidence intervention type for this variable.
Variable	getVariable()	Gets the variable these options refer to.
void	setAllowMissing(boolean value)	Determines whether the optimizer can consider missing values (evidence not set) on this variable.
void	setEvidenceKind(DesignEvidenceKind value)	Determines whether the optimizer uses hard or soft/virtual evidence for this variable.
void	setInterventionType(InterventionType value)	Determines the evidence intervention type for this variable.

Methods inherited from class java.lang.Object

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

Constructor Detail

DesignVariable

public DesignVariable(Variable variable,
                      Double lowerBound,
                      Double upperBound,
                      boolean allowMissing)

Initializes a new instance of the com.bayesserver.optization.DesignVariable class, automatically generating the necessary design states.

Parameters:

variable - The optimizer input variable.

lowerBound - Optional lower bound for the variable value if continuous, or the variable states if discrete.

upperBound - Optional upper bound for the variable value if continuous, or the variable states if discrete.

allowMissing - Determines whether the algorithm is allowed to consider having no evidence set on this variable.

DesignVariable

public DesignVariable(Variable variable,
                      Double lowerBound,
                      Double upperBound,
                      boolean allowMissing,
                      InterventionType interventionType)

Initializes a new instance of the DesignVariable class, automatically generating the necessary design states.

Parameters:

variable - The optimizer input variable.

lowerBound - Optional lower bound for the variable value if continuous, or the variable states if discrete.

upperBound - Optional upper bound for the variable value if continuous, or the variable states if discrete.

allowMissing - Determines whether the algorithm is allowed to consider having no evidence set on this variable.

interventionType - Determines whether the evidence should be an intervention or not.

DesignVariable

public DesignVariable(Variable variable,
                      List<DesignState> designStates,
                      boolean allowMissing)

Initializes a new instance of the DesignVariable class.

Parameters:

variable - The optimizer input variable.

designStates - Options for each variable state.

allowMissing - Determines whether the algorithm is allowed to consider having no evidence set on this variable.

DesignVariable

public DesignVariable(Variable variable,
                      List<DesignState> designStates,
                      DesignEvidenceKind evidenceKind,
                      boolean allowMissing,
                      InterventionType interventionType)

Initializes a new instance of the DesignVariable class.

Parameters:

variable - The optimizer input variable.

designStates - Options for each variable state.

evidenceKind - Determines whether to consider hard or soft evidence.

allowMissing - Determines whether the algorithm is allowed to consider having no evidence set on this variable.

interventionType - Determines whether the evidence should be an intervention or not.

Method Detail

getEvidenceKind

public DesignEvidenceKind getEvidenceKind()

Determines whether the optimizer uses hard or soft/virtual evidence for this variable.

setEvidenceKind

public void setEvidenceKind(DesignEvidenceKind value)

Determines whether the optimizer uses hard or soft/virtual evidence for this variable.

getVariable

public Variable getVariable()

Gets the variable these options refer to.

getInterventionType

public InterventionType getInterventionType()

Determines the evidence intervention type for this variable.

setInterventionType

public void setInterventionType(InterventionType value)

Determines the evidence intervention type for this variable.

getDesignStates

public List<DesignState> getDesignStates()

Gets the design states, one for each state in the variable. Note that continuous and function variables have a single state.

getAllowMissing

public boolean getAllowMissing()

Determines whether the optimizer can consider missing values (evidence not set) on this variable.

setAllowMissing

public void setAllowMissing(boolean value)

Determines whether the optimizer can consider missing values (evidence not set) on this variable.