public OptimizationState(Tuple <double, double> objectiveMaxValues, List <Tuple <double, double> > optimalParetoFront)
{
this.objectiveMaxValues = objectiveMaxValues;
this.optimalParetoFront = optimalParetoFront;
dominationComparer = new DefaultDominationComparer();
SetNewResult();
}
private static void Lab8BiObjectiveBinaryGA(IEvaluation <bool, Tuple <double, double> > evaluation, int?seed)
{
RunningTimeStopCondition stopCondition = new RunningTimeStopCondition(5);
DefaultDominationComparer dominationComparer = new DefaultDominationComparer();
BinaryRandomGenerator generator = new BinaryRandomGenerator(evaluation.pcConstraint, seed);
OnePointCrossover crossover = new OnePointCrossover(0.5, seed);
BinaryBitFlipMutation mutation = new BinaryBitFlipMutation(1.0 / evaluation.iSize, evaluation, seed);
SampleBiObjectiveSelection selection = new SampleBiObjectiveSelection(dominationComparer, seed);
BiObjective.GeneticAlgorithm <bool> ga = new BiObjective.GeneticAlgorithm <bool>(evaluation, stopCondition, generator, selection, crossover, mutation, 100, seed);
ga.Run();
ReportBiObjectiveOptimizationResult(ga.Result);
}
private static Experiment <bool, Tuple <double, double>, BiObjective.OptimizationResult <bool> > PrepareExperiment9(int method, int evaluationType, int selectionMethod, double mutationProb, int populationSize, int?seed = null)
{
IEvaluation <bool, Tuple <double, double> > evaluation;
Dictionary <string, string> info = new Dictionary <string, string>();
DefaultDominationComparer dominationComparer = new DefaultDominationComparer();
switch (evaluationType)
{
case 0:
evaluation = new CBinaryZeroMaxOneMaxEvaluation(100);
info["problem"] = "ZeroMaxOneMax";
break;
case 1:
evaluation = new CBinaryTrapInvTrapEvaluation(5, 100);
info["problem"] = "Trap5InvTrap5";
break;
case 2:
evaluation = new CBinaryLOTZEvaluation(10);
info["problem"] = "LOTZ";
break;
case 3:
evaluation = new CBinaryMOMaxCutEvaluation(EBinaryBiObjectiveMaxCutInstance.maxcut_instance_100);
info["problem"] = "MaxCut";
break;
case 4:
evaluation = new CBinaryMOKnapsackEvaluation(EBinaryBiObjectiveKnapsackInstance.knapsack_100);
info["problem"] = "MOKnapsack";
break;
default:
throw new ArgumentOutOfRangeException();
}
info["variables"] = evaluation.iSize.ToString();
RunningTimeStopCondition stopCondition = new RunningTimeStopCondition(30);
BinaryRandomGenerator generator = new BinaryRandomGenerator(evaluation.pcConstraint, seed);
OnePointCrossover crossover = new OnePointCrossover(0.5, seed);
BinaryBitFlipMutation mutation = new BinaryBitFlipMutation(1.0 / evaluation.iSize, evaluation, seed);
ASelection <Tuple <double, double> > selection;
switch (selectionMethod)
{
case 0:
selection = new SampleBiObjectiveSelection(dominationComparer, seed);
info["selection_method"] = "Sample";
break;
case 1:
selection = new NSGA2Selection(10, evaluation.tMaxValue, dominationComparer, true, seed);
info["selection_method"] = "NSGA2";
break;
default:
throw new ArgumentOutOfRangeException();
}
mutationProb = mutationProb / evaluation.iSize;
info["mutation_prob"] = mutationProb.ToString();
info["population"] = populationSize.ToString();
BiObjective.GeneticAlgorithm <bool> ga;
switch (method)
{
case 0:
ga = new BiObjective.GeneticAlgorithm <bool>(evaluation, stopCondition, generator, selection, crossover, mutation, populationSize, seed);
info["method"] = "GA";
break;
case 1:
ga = new BiObjective.ClusteringGeneticAlgorithm <bool>(evaluation, stopCondition, generator, selection, crossover, mutation, populationSize, populationSize / 10, seed);
info["method"] = "ClusteringGA";
break;
case 2:
ga = new BiObjective.ClusteringGeneticAlgorithm2 <bool>(evaluation, stopCondition, generator, selection, crossover, mutation, populationSize, populationSize / 10, seed);
info["method"] = "ClusteringGA2";
break;
default:
throw new ArgumentOutOfRangeException();
}
Experiment <bool, Tuple <double, double>, BiObjective.OptimizationResult <bool> > experiment = new Experiment <bool, Tuple <double, double>, BiObjective.OptimizationResult <bool> >(ga, info, ExperimentFinished);
return(experiment);
}
