Class GeneticOptionsBase

Base class for common Genetic algorithm options.

Inheritance
System.Object
GeneticOptionsBase
GeneticOptimizerOptions
GeneticSimplificationOptions
Implements
System.ComponentModel.INotifyPropertyChanged
Inherited Members
System.Object.Equals(System.Object)
System.Object.Equals(System.Object, System.Object)
System.Object.GetHashCode()
System.Object.GetType()
System.Object.MemberwiseClone()
System.Object.ReferenceEquals(System.Object, System.Object)
System.Object.ToString()
Namespace: BayesServer.Optimization.Genetic
Assembly: BayesServer.Optimization.dll
Syntax
public abstract class GeneticOptionsBase : IOptimizerOptions, INotifyPropertyChanged

Properties

Cancellation

Gets of sets the instance implementing ICancellation, used for cancellation.

Declaration
[Browsable(false)]
public ICancellation Cancellation { get; set; }
Property Value
Type Description
ICancellation
See Also

CausalEffectKind

Gets or sets the kind of causal effect to optimize. e.g. Total/Direct.

Declaration
[Browsable(false)]
public CausalEffectKind CausalEffectKind { get; set; }
Property Value
Type Description
CausalEffectKind
Remarks

Not all algorithms support direct effect calculations.

CrossoverProbability

The probability of parents being crossed. Must be in the range [0,1].

Declaration
[OverflowProperty(true)]
public double CrossoverProbability { get; set; }
Property Value
Type Description
System.Double

InferenceFactory

Used to create one or more inference engines, used by the algorithm to determine the fitness of possible solutions.

Declaration
[Browsable(false)]
public IInferenceFactory InferenceFactory { get; set; }
Property Value
Type Description
IInferenceFactory

MaximumConcurrency

Gets or sets the maximum number of inference engines used during optimization.

Declaration
[OverflowProperty(true)]
public int? MaximumConcurrency { get; set; }
Property Value
Type Description
System.Nullable<System.Int32>
Remarks

During optimization, multiple inference engines may be used in parallel. However each inference engine has its own memory requirements for inference, and so this parameter allows the number to be limited, to avoid excessive memory consumption. The amount of memory used per inference engine, depends on the Network and also the evidence.

MutationProbability

The probability of genes being mutated. Must be in the range [0,1].

Declaration
[OverflowProperty(true)]
public double MutationProbability { get; set; }
Property Value
Type Description
System.Double

PopulationSize

Gets or sets the number of chromosomes in each generation.

Declaration
public int PopulationSize { get; set; }
Property Value
Type Description
System.Int32

Progress

Gets of sets the instance implementing IOptimizerProgress, used for progress notifications.

Declaration
[Browsable(false)]
public IOptimizerProgress Progress { get; set; }
Property Value
Type Description
IOptimizerProgress
See Also

QueryDistributions

Determines which additional queries, if any, should be calculated by the inference engine when evaluating the fitness of a solution. For example, additional queries maybe referenced by function nodes.

Declaration
[Browsable(false)]
public IList<QueryDistribution> QueryDistributions { get; }
Property Value
Type Description
System.Collections.Generic.IList<QueryDistribution>

QueryFunctions

Determines which additional functions, if any, should be calculated by the inference engine when evaluating the fitness of a solution. For example, functions that are referenced by a function node optimization target/objective.

Declaration
[Browsable(false)]
public IList<QueryFunctionOutput> QueryFunctions { get; }
Property Value
Type Description
System.Collections.Generic.IList<QueryFunctionOutput>

QueryLogLikelihood

Determines whether the log-likelihood should be calculated by the inference engine when evaluating the fitness of a solution. For example, the log-likelihood maybe referenced by function nodes.

Declaration
[Browsable(false)]
public bool? QueryLogLikelihood { get; set; }
Property Value
Type Description
System.Nullable<System.Boolean>

Seed

The seed for the random number generator used by the Genetic Algorithm.

Declaration
[OverflowProperty(true)]
public int? Seed { get; set; }
Property Value
Type Description
System.Nullable<System.Int32>

Stopping

Gets or sets the instance implementing IStop used for early stopping.

Declaration
[Browsable(false)]
public IStop Stopping { get; set; }
Property Value
Type Description
IStop

The instance used for stopping.
Remarks

Stopping is different to cancellation, as stopping will still return an objective value from the optimization process, albeit having performed fewer iterations.

Termination

Termination options.

Declaration
[TypeConverter(typeof(ExpandableObjectConverter))]
[OverflowProperty(true)]
public GeneticTerminationOptions Termination { get; }
Property Value
Type Description
GeneticTerminationOptions

Methods

RaisePropertyChanged(String)

Raises a PropertyChanged event.

Declaration
protected void RaisePropertyChanged(string name)
Parameters
Type Name Description
System.String name

The name of the property that has changed.

Events

PropertyChanged

Declaration
public event PropertyChangedEventHandler PropertyChanged
Event Type
Type Description
System.ComponentModel.PropertyChangedEventHandler

Implements

System.ComponentModel.INotifyPropertyChanged