private CombinedOperator CreateLayerOpener() {
var layerOpener = new CombinedOperator() { Name = "Open new Layer if needed" };
var maxLayerReached = new Comparator() { Name = "MaxLayersReached = OpenLayers >= NumberOfLayers" };
var maxLayerReachedBranch = new ConditionalBranch() { Name = "MaxLayersReached?" };
var openNewLayerCalculator = new ExpressionCalculator() { Name = "OpenNewLayer = Generations >= AgeLimits[OpenLayers - 1]" };
var openNewLayerBranch = new ConditionalBranch() { Name = "OpenNewLayer?" };
var layerCreator = new LastLayerCloner() { Name = "Create Layer" };
var updateLayerNumber = new Assigner() { Name = "Layer = OpenLayers" };
var historyWiper = new ResultsHistoryWiper() { Name = "Clear History in Results" };
var createChildrenViaCrossover = new AlpsOffspringSelectionGeneticAlgorithmMainOperator();
var incrEvaluatedSolutionsForNewLayer = new SubScopesCounter() { Name = "Update EvaluatedSolutions" };
var incrOpenLayers = new IntCounter() { Name = "Incr. OpenLayers" };
var newLayerResultsCollector = new ResultsCollector() { Name = "Collect new Layer Results" };
layerOpener.OperatorGraph.InitialOperator = maxLayerReached;
maxLayerReached.LeftSideParameter.ActualName = "OpenLayers";
maxLayerReached.RightSideParameter.ActualName = NumberOfLayersParameter.Name;
maxLayerReached.ResultParameter.ActualName = "MaxLayerReached";
maxLayerReached.Comparison = new Comparison(ComparisonType.GreaterOrEqual);
maxLayerReached.Successor = maxLayerReachedBranch;
maxLayerReachedBranch.ConditionParameter.ActualName = "MaxLayerReached";
maxLayerReachedBranch.FalseBranch = openNewLayerCalculator;
openNewLayerCalculator.CollectedValues.Add(new LookupParameter<IntArray>(AgeLimitsParameter.Name));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter<IntValue>("Generations"));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter<IntValue>(NumberOfLayersParameter.Name));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter<IntValue>("OpenLayers"));
openNewLayerCalculator.ExpressionResultParameter.ActualName = "OpenNewLayer";
openNewLayerCalculator.ExpressionParameter.Value = new StringValue("Generations 1 + AgeLimits OpenLayers 1 - [] >");
openNewLayerCalculator.Successor = openNewLayerBranch;
openNewLayerBranch.ConditionParameter.ActualName = "OpenNewLayer";
openNewLayerBranch.TrueBranch = layerCreator;
layerCreator.NewLayerOperator = updateLayerNumber;
layerCreator.Successor = incrOpenLayers;
updateLayerNumber.LeftSideParameter.ActualName = "Layer";
updateLayerNumber.RightSideParameter.ActualName = "OpenLayers";
updateLayerNumber.Successor = historyWiper;
historyWiper.ResultsParameter.ActualName = "LayerResults";
historyWiper.Successor = createChildrenViaCrossover;
// Maybe use only crossover and no elitism instead of "default operator"
createChildrenViaCrossover.RandomParameter.ActualName = LocalRandomParameter.Name;
createChildrenViaCrossover.EvaluatorParameter.ActualName = EvaluatorParameter.Name;
createChildrenViaCrossover.EvaluatedSolutionsParameter.ActualName = "LayerEvaluatedSolutions";
createChildrenViaCrossover.QualityParameter.ActualName = QualityParameter.Name;
createChildrenViaCrossover.MaximizationParameter.ActualName = MaximizationParameter.Name;
createChildrenViaCrossover.PopulationSizeParameter.ActualName = PopulationSizeParameter.Name;
createChildrenViaCrossover.SelectorParameter.ActualName = SelectorParameter.Name;
createChildrenViaCrossover.CrossoverParameter.ActualName = CrossoverParameter.Name;
createChildrenViaCrossover.MutatorParameter.ActualName = MutatorParameter.ActualName;
createChildrenViaCrossover.MutationProbabilityParameter.ActualName = MutationProbabilityParameter.Name;
createChildrenViaCrossover.ElitesParameter.ActualName = ElitesParameter.Name;
createChildrenViaCrossover.ReevaluateElitesParameter.ActualName = ReevaluateElitesParameter.Name;
createChildrenViaCrossover.ComparisonFactorParameter.ActualName = ComparisonFactorParameter.Name;
createChildrenViaCrossover.SuccessRatioParameter.ActualName = SuccessRatioParameter.Name;
createChildrenViaCrossover.CurrentSuccessRatioParameter.ActualName = "CurrentSuccessRatio";
createChildrenViaCrossover.SelectionPressureParameter.ActualName = "SelectionPressure";
createChildrenViaCrossover.MaximumSelectionPressureParameter.ActualName = MaximumSelectionPressureParameter.Name;
createChildrenViaCrossover.OffspringSelectionBeforeMutationParameter.ActualName = OffspringSelectionBeforeMutationParameter.Name;
createChildrenViaCrossover.FillPopulationWithParentsParameter.ActualName = FillPopulationWithParentsParameter.Name;
createChildrenViaCrossover.AgeParameter.ActualName = AgeParameter.Name;
createChildrenViaCrossover.AgeInheritanceParameter.ActualName = AgeInheritanceParameter.Name;
createChildrenViaCrossover.AgeIncrementParameter.Value = new DoubleValue(0.0);
createChildrenViaCrossover.Successor = incrEvaluatedSolutionsForNewLayer;
incrEvaluatedSolutionsForNewLayer.ValueParameter.ActualName = EvaluatedSolutionsParameter.Name;
incrEvaluatedSolutionsForNewLayer.AccumulateParameter.Value = new BoolValue(true);
incrOpenLayers.ValueParameter.ActualName = "OpenLayers";
incrOpenLayers.Increment = new IntValue(1);
incrOpenLayers.Successor = newLayerResultsCollector;
newLayerResultsCollector.CollectedValues.Add(new ScopeTreeLookupParameter<ResultCollection>("LayerResults", "Result set for each layer", "LayerResults"));
newLayerResultsCollector.CopyValue = new BoolValue(false);
newLayerResultsCollector.Successor = null;
return layerOpener;
}
public AlpsOffspringSelectionGeneticAlgorithmMainLoop()
: base() {
Parameters.Add(new ValueLookupParameter<IRandom>("GlobalRandom", "A pseudo random number generator."));
Parameters.Add(new ValueLookupParameter<IRandom>("LocalRandom", "A pseudo random number generator."));
Parameters.Add(new ValueLookupParameter<IOperator>("Evaluator", "The operator used to evaluate solutions. This operator is executed in parallel, if an engine is used which supports parallelization."));
Parameters.Add(new ValueLookupParameter<IntValue>("EvaluatedSolutions", "The number of times solutions have been evaluated."));
Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("Quality", "The value which represents the quality of a solution."));
Parameters.Add(new ValueLookupParameter<BoolValue>("Maximization", "True if the problem is a maximization problem, otherwise false."));
Parameters.Add(new ValueLookupParameter<IOperator>("Analyzer", "The operator used to analyze all individuals from all layers combined."));
Parameters.Add(new ValueLookupParameter<IOperator>("LayerAnalyzer", "The operator used to analyze each layer."));
Parameters.Add(new ValueLookupParameter<IntValue>("NumberOfLayers", "The number of layers."));
Parameters.Add(new ValueLookupParameter<IntValue>("PopulationSize", "The size of the population of solutions in each layer."));
Parameters.Add(new LookupParameter<IntValue>("CurrentPopulationSize", "The current size of the population."));
Parameters.Add(new ValueLookupParameter<IOperator>("Selector", "The operator used to select solutions for reproduction."));
Parameters.Add(new ValueLookupParameter<IOperator>("Crossover", "The operator used to cross solutions."));
Parameters.Add(new ValueLookupParameter<IOperator>("Mutator", "The operator used to mutate solutions."));
Parameters.Add(new ValueLookupParameter<PercentValue>("MutationProbability", "The probability that the mutation operator is applied on a solution."));
Parameters.Add(new ValueLookupParameter<IntValue>("Elites", "The numer of elite solutions which are kept in each generation."));
Parameters.Add(new ValueLookupParameter<BoolValue>("ReevaluateElites", "Flag to determine if elite individuals should be reevaluated (i.e., if stochastic fitness functions are used.)"));
Parameters.Add(new ValueLookupParameter<DoubleValue>("SuccessRatio", "The ratio of successful to total children that should be achieved."));
Parameters.Add(new ValueLookupParameter<DoubleValue>("ComparisonFactor", "The comparison factor is used to determine whether the offspring should be compared to the better parent, the worse parent or a quality value linearly interpolated between them. It is in the range [0;1]."));
Parameters.Add(new ValueLookupParameter<DoubleValue>("MaximumSelectionPressure", "The maximum selection pressure that terminates the algorithm."));
Parameters.Add(new ValueLookupParameter<BoolValue>("OffspringSelectionBeforeMutation", "True if the offspring selection step should be applied before mutation, false if it should be applied after mutation."));
Parameters.Add(new ValueLookupParameter<BoolValue>("FillPopulationWithParents", "True if the population should be filled with parent individual or false if worse children should be used when the maximum selection pressure is exceeded."));
Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("Age", "The age of individuals."));
Parameters.Add(new ValueLookupParameter<IntValue>("AgeGap", "The frequency of reseeding the lowest layer and scaling factor for the age-limits for the layers."));
Parameters.Add(new ValueLookupParameter<DoubleValue>("AgeInheritance", "A weight that determines the age of a child after crossover based on the older (1.0) and younger (0.0) parent."));
Parameters.Add(new ValueLookupParameter<IntArray>("AgeLimits", "The maximum age an individual is allowed to reach in a certain layer."));
Parameters.Add(new ValueLookupParameter<IntValue>("MatingPoolRange", "The range of sub - populations used for creating a mating pool. (1 = current + previous sub-population)"));
Parameters.Add(new ValueLookupParameter<BoolValue>("ReduceToPopulationSize", "Reduce the CurrentPopulationSize after elder migration to PopulationSize"));
Parameters.Add(new ValueLookupParameter<IOperator>("Terminator", "The termination criteria that defines if the algorithm should continue or stop"));
var variableCreator = new VariableCreator() { Name = "Initialize" };
var initLayerAnalyzerProcessor = new SubScopesProcessor();
var layerVariableCreator = new VariableCreator() { Name = "Initialize Layer" };
var initLayerAnalyzerPlaceholder = new Placeholder() { Name = "LayerAnalyzer (Placeholder)" };
var layerResultCollector = new ResultsCollector() { Name = "Collect layer results" };
var initAnalyzerPlaceholder = new Placeholder() { Name = "Analyzer (Placeholder)" };
var resultsCollector = new ResultsCollector();
var matingPoolCreator = new MatingPoolCreator() { Name = "Create Mating Pools" };
var matingPoolProcessor = new UniformSubScopesProcessor() { Name = "Process Mating Pools" };
var initializeLayer = new Assigner() { Name = "Reset LayerEvaluatedSolutions" };
var mainOperator = new AlpsOffspringSelectionGeneticAlgorithmMainOperator();
var generationsIcrementor = new IntCounter() { Name = "Increment Generations" };
var evaluatedSolutionsReducer = new DataReducer() { Name = "Increment EvaluatedSolutions" };
var eldersEmigrator = CreateEldersEmigrator();
var layerOpener = CreateLayerOpener();
var layerReseeder = CreateReseeder();
var layerAnalyzerProcessor = new UniformSubScopesProcessor();
var layerAnalyzerPlaceholder = new Placeholder() { Name = "LayerAnalyzer (Placeholder)" };
var analyzerPlaceholder = new Placeholder() { Name = "Analyzer (Placeholder)" };
var termination = new TerminationOperator();
OperatorGraph.InitialOperator = variableCreator;
variableCreator.CollectedValues.Add(new ValueParameter<IntValue>("Generations", new IntValue(0)));
variableCreator.CollectedValues.Add(new ValueParameter<IntValue>("OpenLayers", new IntValue(1)));
variableCreator.Successor = initLayerAnalyzerProcessor;
initLayerAnalyzerProcessor.Operators.Add(layerVariableCreator);
initLayerAnalyzerProcessor.Successor = initAnalyzerPlaceholder;
layerVariableCreator.CollectedValues.Add(new ValueParameter<IntValue>("Layer", new IntValue(0)));
layerVariableCreator.CollectedValues.Add(new ValueParameter<ResultCollection>("LayerResults"));
layerVariableCreator.CollectedValues.Add(new ValueParameter<DoubleValue>("SelectionPressure", new DoubleValue(0)));
layerVariableCreator.CollectedValues.Add(new ValueParameter<DoubleValue>("CurrentSuccessRatio", new DoubleValue(0)));
layerVariableCreator.Successor = initLayerAnalyzerPlaceholder;
initLayerAnalyzerPlaceholder.OperatorParameter.ActualName = LayerAnalyzerParameter.Name;
initLayerAnalyzerPlaceholder.Successor = layerResultCollector;
layerResultCollector.ResultsParameter.ActualName = "LayerResults";
layerResultCollector.CollectedValues.Add(new LookupParameter<DoubleValue>("Current Selection Pressure", "Displays the rising selection pressure during a generation.", "SelectionPressure"));
layerResultCollector.CollectedValues.Add(new LookupParameter<DoubleValue>("Current Success Ratio", "Indicates how many successful children were already found during a generation (relative to the population size).", "CurrentSuccessRatio"));
layerResultCollector.Successor = null;
initAnalyzerPlaceholder.OperatorParameter.ActualName = AnalyzerParameter.Name;
initAnalyzerPlaceholder.Successor = resultsCollector;
resultsCollector.CollectedValues.Add(new LookupParameter<IntValue>("Generations"));
resultsCollector.CollectedValues.Add(new ScopeTreeLookupParameter<ResultCollection>("LayerResults", "Result set for each Layer", "LayerResults"));
resultsCollector.CollectedValues.Add(new LookupParameter<IntValue>("OpenLayers"));
resultsCollector.CopyValue = new BoolValue(false);
resultsCollector.Successor = matingPoolCreator;
matingPoolCreator.MatingPoolRangeParameter.Value = null;
matingPoolCreator.MatingPoolRangeParameter.ActualName = MatingPoolRangeParameter.Name;
matingPoolCreator.Successor = matingPoolProcessor;
matingPoolProcessor.Parallel.Value = true;
matingPoolProcessor.Operator = initializeLayer;
matingPoolProcessor.Successor = generationsIcrementor;
initializeLayer.LeftSideParameter.ActualName = "LayerEvaluatedSolutions";
initializeLayer.RightSideParameter.Value = new IntValue(0);
initializeLayer.Successor = mainOperator;
mainOperator.RandomParameter.ActualName = LocalRandomParameter.Name;
mainOperator.EvaluatorParameter.ActualName = EvaluatorParameter.Name;
mainOperator.EvaluatedSolutionsParameter.ActualName = "LayerEvaluatedSolutions";
mainOperator.QualityParameter.ActualName = QualityParameter.Name;
mainOperator.MaximizationParameter.ActualName = MaximizationParameter.Name;
mainOperator.PopulationSizeParameter.ActualName = PopulationSizeParameter.Name;
mainOperator.SelectorParameter.ActualName = SelectorParameter.Name;
mainOperator.CrossoverParameter.ActualName = CrossoverParameter.Name;
mainOperator.MutatorParameter.ActualName = MutatorParameter.ActualName;
mainOperator.MutationProbabilityParameter.ActualName = MutationProbabilityParameter.Name;
mainOperator.ElitesParameter.ActualName = ElitesParameter.Name;
mainOperator.ReevaluateElitesParameter.ActualName = ReevaluateElitesParameter.Name;
mainOperator.ComparisonFactorParameter.ActualName = ComparisonFactorParameter.Name;
mainOperator.SuccessRatioParameter.ActualName = SuccessRatioParameter.Name;
mainOperator.CurrentSuccessRatioParameter.ActualName = "CurrentSuccessRatio";
mainOperator.SelectionPressureParameter.ActualName = "SelectionPressure";
mainOperator.MaximumSelectionPressureParameter.ActualName = MaximumSelectionPressureParameter.Name;
mainOperator.OffspringSelectionBeforeMutationParameter.ActualName = OffspringSelectionBeforeMutationParameter.Name;
mainOperator.FillPopulationWithParentsParameter.ActualName = FillPopulationWithParentsParameter.Name;
mainOperator.AgeParameter.ActualName = AgeParameter.Name;
mainOperator.AgeInheritanceParameter.ActualName = AgeInheritanceParameter.Name;
mainOperator.AgeIncrementParameter.Value = new DoubleValue(1.0);
mainOperator.Successor = null;
generationsIcrementor.ValueParameter.ActualName = "Generations";
generationsIcrementor.Increment = new IntValue(1);
generationsIcrementor.Successor = evaluatedSolutionsReducer;
evaluatedSolutionsReducer.ParameterToReduce.ActualName = "LayerEvaluatedSolutions";
evaluatedSolutionsReducer.TargetParameter.ActualName = EvaluatedSolutionsParameter.Name;
evaluatedSolutionsReducer.ReductionOperation.Value = new ReductionOperation(ReductionOperations.Sum);
evaluatedSolutionsReducer.TargetOperation.Value = new ReductionOperation(ReductionOperations.Sum);
evaluatedSolutionsReducer.Successor = eldersEmigrator;
eldersEmigrator.Successor = layerOpener;
layerOpener.Successor = layerReseeder;
layerReseeder.Successor = layerAnalyzerProcessor;
layerAnalyzerProcessor.Operator = layerAnalyzerPlaceholder;
layerAnalyzerProcessor.Successor = analyzerPlaceholder;
layerAnalyzerPlaceholder.OperatorParameter.ActualName = LayerAnalyzerParameter.Name;
analyzerPlaceholder.OperatorParameter.ActualName = AnalyzerParameter.Name;
analyzerPlaceholder.Successor = termination;
termination.TerminatorParameter.ActualName = TerminatorParameter.Name;
termination.ContinueBranch = matingPoolCreator;
}
private AlpsOffspringSelectionGeneticAlgorithmMainOperator(AlpsOffspringSelectionGeneticAlgorithmMainOperator original, Cloner cloner)
: base(original, cloner)
{
}
private AlpsOffspringSelectionGeneticAlgorithmMainOperator(AlpsOffspringSelectionGeneticAlgorithmMainOperator original, Cloner cloner)
: base(original, cloner) {
}
private CombinedOperator CreateLayerOpener()
{
var layerOpener = new CombinedOperator()
{
Name = "Open new Layer if needed"
};
var maxLayerReached = new Comparator()
{
Name = "MaxLayersReached = OpenLayers >= NumberOfLayers"
};
var maxLayerReachedBranch = new ConditionalBranch()
{
Name = "MaxLayersReached?"
};
var openNewLayerCalculator = new ExpressionCalculator()
{
Name = "OpenNewLayer = Generations >= AgeLimits[OpenLayers - 1]"
};
var openNewLayerBranch = new ConditionalBranch()
{
Name = "OpenNewLayer?"
};
var layerCreator = new LastLayerCloner()
{
Name = "Create Layer"
};
var updateLayerNumber = new Assigner()
{
Name = "Layer = OpenLayers"
};
var historyWiper = new ResultsHistoryWiper()
{
Name = "Clear History in Results"
};
var createChildrenViaCrossover = new AlpsOffspringSelectionGeneticAlgorithmMainOperator();
var incrEvaluatedSolutionsForNewLayer = new SubScopesCounter()
{
Name = "Update EvaluatedSolutions"
};
var incrOpenLayers = new IntCounter()
{
Name = "Incr. OpenLayers"
};
var newLayerResultsCollector = new ResultsCollector()
{
Name = "Collect new Layer Results"
};
layerOpener.OperatorGraph.InitialOperator = maxLayerReached;
maxLayerReached.LeftSideParameter.ActualName = "OpenLayers";
maxLayerReached.RightSideParameter.ActualName = NumberOfLayersParameter.Name;
maxLayerReached.ResultParameter.ActualName = "MaxLayerReached";
maxLayerReached.Comparison = new Comparison(ComparisonType.GreaterOrEqual);
maxLayerReached.Successor = maxLayerReachedBranch;
maxLayerReachedBranch.ConditionParameter.ActualName = "MaxLayerReached";
maxLayerReachedBranch.FalseBranch = openNewLayerCalculator;
openNewLayerCalculator.CollectedValues.Add(new LookupParameter <IntArray>(AgeLimitsParameter.Name));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter <IntValue>("Generations"));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter <IntValue>(NumberOfLayersParameter.Name));
openNewLayerCalculator.CollectedValues.Add(new LookupParameter <IntValue>("OpenLayers"));
openNewLayerCalculator.ExpressionResultParameter.ActualName = "OpenNewLayer";
openNewLayerCalculator.ExpressionParameter.Value = new StringValue("Generations 1 + AgeLimits OpenLayers 1 - [] >");
openNewLayerCalculator.Successor = openNewLayerBranch;
openNewLayerBranch.ConditionParameter.ActualName = "OpenNewLayer";
openNewLayerBranch.TrueBranch = layerCreator;
layerCreator.NewLayerOperator = updateLayerNumber;
layerCreator.Successor = incrOpenLayers;
updateLayerNumber.LeftSideParameter.ActualName = "Layer";
updateLayerNumber.RightSideParameter.ActualName = "OpenLayers";
updateLayerNumber.Successor = historyWiper;
historyWiper.ResultsParameter.ActualName = "LayerResults";
historyWiper.Successor = createChildrenViaCrossover;
// Maybe use only crossover and no elitism instead of "default operator"
createChildrenViaCrossover.RandomParameter.ActualName = LocalRandomParameter.Name;
createChildrenViaCrossover.EvaluatorParameter.ActualName = EvaluatorParameter.Name;
createChildrenViaCrossover.EvaluatedSolutionsParameter.ActualName = "LayerEvaluatedSolutions";
createChildrenViaCrossover.QualityParameter.ActualName = QualityParameter.Name;
createChildrenViaCrossover.MaximizationParameter.ActualName = MaximizationParameter.Name;
createChildrenViaCrossover.PopulationSizeParameter.ActualName = PopulationSizeParameter.Name;
createChildrenViaCrossover.SelectorParameter.ActualName = SelectorParameter.Name;
createChildrenViaCrossover.CrossoverParameter.ActualName = CrossoverParameter.Name;
createChildrenViaCrossover.MutatorParameter.ActualName = MutatorParameter.ActualName;
createChildrenViaCrossover.MutationProbabilityParameter.ActualName = MutationProbabilityParameter.Name;
createChildrenViaCrossover.ElitesParameter.ActualName = ElitesParameter.Name;
createChildrenViaCrossover.ReevaluateElitesParameter.ActualName = ReevaluateElitesParameter.Name;
createChildrenViaCrossover.ComparisonFactorParameter.ActualName = ComparisonFactorParameter.Name;
createChildrenViaCrossover.SuccessRatioParameter.ActualName = SuccessRatioParameter.Name;
createChildrenViaCrossover.CurrentSuccessRatioParameter.ActualName = "CurrentSuccessRatio";
createChildrenViaCrossover.SelectionPressureParameter.ActualName = "SelectionPressure";
createChildrenViaCrossover.MaximumSelectionPressureParameter.ActualName = MaximumSelectionPressureParameter.Name;
createChildrenViaCrossover.OffspringSelectionBeforeMutationParameter.ActualName = OffspringSelectionBeforeMutationParameter.Name;
createChildrenViaCrossover.FillPopulationWithParentsParameter.ActualName = FillPopulationWithParentsParameter.Name;
createChildrenViaCrossover.AgeParameter.ActualName = AgeParameter.Name;
createChildrenViaCrossover.AgeInheritanceParameter.ActualName = AgeInheritanceParameter.Name;
createChildrenViaCrossover.AgeIncrementParameter.Value = new DoubleValue(0.0);
createChildrenViaCrossover.Successor = incrEvaluatedSolutionsForNewLayer;
incrEvaluatedSolutionsForNewLayer.ValueParameter.ActualName = EvaluatedSolutionsParameter.Name;
incrEvaluatedSolutionsForNewLayer.AccumulateParameter.Value = new BoolValue(true);
incrOpenLayers.ValueParameter.ActualName = "OpenLayers";
incrOpenLayers.Increment = new IntValue(1);
incrOpenLayers.Successor = newLayerResultsCollector;
newLayerResultsCollector.CollectedValues.Add(new ScopeTreeLookupParameter <ResultCollection>("LayerResults", "Result set for each layer", "LayerResults"));
newLayerResultsCollector.CopyValue = new BoolValue(false);
newLayerResultsCollector.Successor = null;
return(layerOpener);
}
public AlpsOffspringSelectionGeneticAlgorithmMainLoop()
: base()
{
Parameters.Add(new ValueLookupParameter <IRandom>("GlobalRandom", "A pseudo random number generator."));
Parameters.Add(new ValueLookupParameter <IRandom>("LocalRandom", "A pseudo random number generator."));
Parameters.Add(new ValueLookupParameter <IOperator>("Evaluator", "The operator used to evaluate solutions. This operator is executed in parallel, if an engine is used which supports parallelization."));
Parameters.Add(new ValueLookupParameter <IntValue>("EvaluatedSolutions", "The number of times solutions have been evaluated."));
Parameters.Add(new ScopeTreeLookupParameter <DoubleValue>("Quality", "The value which represents the quality of a solution."));
Parameters.Add(new ValueLookupParameter <BoolValue>("Maximization", "True if the problem is a maximization problem, otherwise false."));
Parameters.Add(new ValueLookupParameter <IOperator>("Analyzer", "The operator used to analyze all individuals from all layers combined."));
Parameters.Add(new ValueLookupParameter <IOperator>("LayerAnalyzer", "The operator used to analyze each layer."));
Parameters.Add(new ValueLookupParameter <IntValue>("NumberOfLayers", "The number of layers."));
Parameters.Add(new ValueLookupParameter <IntValue>("PopulationSize", "The size of the population of solutions in each layer."));
Parameters.Add(new LookupParameter <IntValue>("CurrentPopulationSize", "The current size of the population."));
Parameters.Add(new ValueLookupParameter <IOperator>("Selector", "The operator used to select solutions for reproduction."));
Parameters.Add(new ValueLookupParameter <IOperator>("Crossover", "The operator used to cross solutions."));
Parameters.Add(new ValueLookupParameter <IOperator>("Mutator", "The operator used to mutate solutions."));
Parameters.Add(new ValueLookupParameter <PercentValue>("MutationProbability", "The probability that the mutation operator is applied on a solution."));
Parameters.Add(new ValueLookupParameter <IntValue>("Elites", "The numer of elite solutions which are kept in each generation."));
Parameters.Add(new ValueLookupParameter <BoolValue>("ReevaluateElites", "Flag to determine if elite individuals should be reevaluated (i.e., if stochastic fitness functions are used.)"));
Parameters.Add(new ValueLookupParameter <DoubleValue>("SuccessRatio", "The ratio of successful to total children that should be achieved."));
Parameters.Add(new ValueLookupParameter <DoubleValue>("ComparisonFactor", "The comparison factor is used to determine whether the offspring should be compared to the better parent, the worse parent or a quality value linearly interpolated between them. It is in the range [0;1]."));
Parameters.Add(new ValueLookupParameter <DoubleValue>("MaximumSelectionPressure", "The maximum selection pressure that terminates the algorithm."));
Parameters.Add(new ValueLookupParameter <BoolValue>("OffspringSelectionBeforeMutation", "True if the offspring selection step should be applied before mutation, false if it should be applied after mutation."));
Parameters.Add(new ValueLookupParameter <BoolValue>("FillPopulationWithParents", "True if the population should be filled with parent individual or false if worse children should be used when the maximum selection pressure is exceeded."));
Parameters.Add(new ScopeTreeLookupParameter <DoubleValue>("Age", "The age of individuals."));
Parameters.Add(new ValueLookupParameter <IntValue>("AgeGap", "The frequency of reseeding the lowest layer and scaling factor for the age-limits for the layers."));
Parameters.Add(new ValueLookupParameter <DoubleValue>("AgeInheritance", "A weight that determines the age of a child after crossover based on the older (1.0) and younger (0.0) parent."));
Parameters.Add(new ValueLookupParameter <IntArray>("AgeLimits", "The maximum age an individual is allowed to reach in a certain layer."));
Parameters.Add(new ValueLookupParameter <IntValue>("MatingPoolRange", "The range of sub - populations used for creating a mating pool. (1 = current + previous sub-population)"));
Parameters.Add(new ValueLookupParameter <BoolValue>("ReduceToPopulationSize", "Reduce the CurrentPopulationSize after elder migration to PopulationSize"));
Parameters.Add(new ValueLookupParameter <IOperator>("Terminator", "The termination criteria that defines if the algorithm should continue or stop"));
var variableCreator = new VariableCreator()
{
Name = "Initialize"
};
var initLayerAnalyzerProcessor = new SubScopesProcessor();
var layerVariableCreator = new VariableCreator()
{
Name = "Initialize Layer"
};
var initLayerAnalyzerPlaceholder = new Placeholder()
{
Name = "LayerAnalyzer (Placeholder)"
};
var layerResultCollector = new ResultsCollector()
{
Name = "Collect layer results"
};
var initAnalyzerPlaceholder = new Placeholder()
{
Name = "Analyzer (Placeholder)"
};
var resultsCollector = new ResultsCollector();
var matingPoolCreator = new MatingPoolCreator()
{
Name = "Create Mating Pools"
};
var matingPoolProcessor = new UniformSubScopesProcessor()
{
Name = "Process Mating Pools"
};
var initializeLayer = new Assigner()
{
Name = "Reset LayerEvaluatedSolutions"
};
var mainOperator = new AlpsOffspringSelectionGeneticAlgorithmMainOperator();
var generationsIcrementor = new IntCounter()
{
Name = "Increment Generations"
};
var evaluatedSolutionsReducer = new DataReducer()
{
Name = "Increment EvaluatedSolutions"
};
var eldersEmigrator = CreateEldersEmigrator();
var layerOpener = CreateLayerOpener();
var layerReseeder = CreateReseeder();
var layerAnalyzerProcessor = new UniformSubScopesProcessor();
var layerAnalyzerPlaceholder = new Placeholder()
{
Name = "LayerAnalyzer (Placeholder)"
};
var analyzerPlaceholder = new Placeholder()
{
Name = "Analyzer (Placeholder)"
};
var termination = new TerminationOperator();
OperatorGraph.InitialOperator = variableCreator;
variableCreator.CollectedValues.Add(new ValueParameter <IntValue>("Generations", new IntValue(0)));
variableCreator.CollectedValues.Add(new ValueParameter <IntValue>("OpenLayers", new IntValue(1)));
variableCreator.Successor = initLayerAnalyzerProcessor;
initLayerAnalyzerProcessor.Operators.Add(layerVariableCreator);
initLayerAnalyzerProcessor.Successor = initAnalyzerPlaceholder;
layerVariableCreator.CollectedValues.Add(new ValueParameter <IntValue>("Layer", new IntValue(0)));
layerVariableCreator.CollectedValues.Add(new ValueParameter <ResultCollection>("LayerResults"));
layerVariableCreator.CollectedValues.Add(new ValueParameter <DoubleValue>("SelectionPressure", new DoubleValue(0)));
layerVariableCreator.CollectedValues.Add(new ValueParameter <DoubleValue>("CurrentSuccessRatio", new DoubleValue(0)));
layerVariableCreator.Successor = initLayerAnalyzerPlaceholder;
initLayerAnalyzerPlaceholder.OperatorParameter.ActualName = LayerAnalyzerParameter.Name;
initLayerAnalyzerPlaceholder.Successor = layerResultCollector;
layerResultCollector.ResultsParameter.ActualName = "LayerResults";
layerResultCollector.CollectedValues.Add(new LookupParameter <DoubleValue>("Current Selection Pressure", "Displays the rising selection pressure during a generation.", "SelectionPressure"));
layerResultCollector.CollectedValues.Add(new LookupParameter <DoubleValue>("Current Success Ratio", "Indicates how many successful children were already found during a generation (relative to the population size).", "CurrentSuccessRatio"));
layerResultCollector.Successor = null;
initAnalyzerPlaceholder.OperatorParameter.ActualName = AnalyzerParameter.Name;
initAnalyzerPlaceholder.Successor = resultsCollector;
resultsCollector.CollectedValues.Add(new LookupParameter <IntValue>("Generations"));
resultsCollector.CollectedValues.Add(new ScopeTreeLookupParameter <ResultCollection>("LayerResults", "Result set for each Layer", "LayerResults"));
resultsCollector.CollectedValues.Add(new LookupParameter <IntValue>("OpenLayers"));
resultsCollector.CopyValue = new BoolValue(false);
resultsCollector.Successor = matingPoolCreator;
matingPoolCreator.MatingPoolRangeParameter.Value = null;
matingPoolCreator.MatingPoolRangeParameter.ActualName = MatingPoolRangeParameter.Name;
matingPoolCreator.Successor = matingPoolProcessor;
matingPoolProcessor.Parallel.Value = true;
matingPoolProcessor.Operator = initializeLayer;
matingPoolProcessor.Successor = generationsIcrementor;
initializeLayer.LeftSideParameter.ActualName = "LayerEvaluatedSolutions";
initializeLayer.RightSideParameter.Value = new IntValue(0);
initializeLayer.Successor = mainOperator;
mainOperator.RandomParameter.ActualName = LocalRandomParameter.Name;
mainOperator.EvaluatorParameter.ActualName = EvaluatorParameter.Name;
mainOperator.EvaluatedSolutionsParameter.ActualName = "LayerEvaluatedSolutions";
mainOperator.QualityParameter.ActualName = QualityParameter.Name;
mainOperator.MaximizationParameter.ActualName = MaximizationParameter.Name;
mainOperator.PopulationSizeParameter.ActualName = PopulationSizeParameter.Name;
mainOperator.SelectorParameter.ActualName = SelectorParameter.Name;
mainOperator.CrossoverParameter.ActualName = CrossoverParameter.Name;
mainOperator.MutatorParameter.ActualName = MutatorParameter.ActualName;
mainOperator.MutationProbabilityParameter.ActualName = MutationProbabilityParameter.Name;
mainOperator.ElitesParameter.ActualName = ElitesParameter.Name;
mainOperator.ReevaluateElitesParameter.ActualName = ReevaluateElitesParameter.Name;
mainOperator.ComparisonFactorParameter.ActualName = ComparisonFactorParameter.Name;
mainOperator.SuccessRatioParameter.ActualName = SuccessRatioParameter.Name;
mainOperator.CurrentSuccessRatioParameter.ActualName = "CurrentSuccessRatio";
mainOperator.SelectionPressureParameter.ActualName = "SelectionPressure";
mainOperator.MaximumSelectionPressureParameter.ActualName = MaximumSelectionPressureParameter.Name;
mainOperator.OffspringSelectionBeforeMutationParameter.ActualName = OffspringSelectionBeforeMutationParameter.Name;
mainOperator.FillPopulationWithParentsParameter.ActualName = FillPopulationWithParentsParameter.Name;
mainOperator.AgeParameter.ActualName = AgeParameter.Name;
mainOperator.AgeInheritanceParameter.ActualName = AgeInheritanceParameter.Name;
mainOperator.AgeIncrementParameter.Value = new DoubleValue(1.0);
mainOperator.Successor = null;
generationsIcrementor.ValueParameter.ActualName = "Generations";
generationsIcrementor.Increment = new IntValue(1);
generationsIcrementor.Successor = evaluatedSolutionsReducer;
evaluatedSolutionsReducer.ParameterToReduce.ActualName = "LayerEvaluatedSolutions";
evaluatedSolutionsReducer.TargetParameter.ActualName = EvaluatedSolutionsParameter.Name;
evaluatedSolutionsReducer.ReductionOperation.Value = new ReductionOperation(ReductionOperations.Sum);
evaluatedSolutionsReducer.TargetOperation.Value = new ReductionOperation(ReductionOperations.Sum);
evaluatedSolutionsReducer.Successor = eldersEmigrator;
eldersEmigrator.Successor = layerOpener;
layerOpener.Successor = layerReseeder;
layerReseeder.Successor = layerAnalyzerProcessor;
layerAnalyzerProcessor.Operator = layerAnalyzerPlaceholder;
layerAnalyzerProcessor.Successor = analyzerPlaceholder;
layerAnalyzerPlaceholder.OperatorParameter.ActualName = LayerAnalyzerParameter.Name;
analyzerPlaceholder.OperatorParameter.ActualName = AnalyzerParameter.Name;
analyzerPlaceholder.Successor = termination;
termination.TerminatorParameter.ActualName = TerminatorParameter.Name;
termination.ContinueBranch = matingPoolCreator;
}