static void Main(string[] args)
{
// trying to make the word "abracadabra"
const string target = "dog";
var evaluator = new FitnessEvaluator(target);
var random = new Random();
var generator = new Generator(random);
const int Total = 100;
var population = new Population <string>(Total, generator, evaluator);
var mutator = new Mutator(state =>
{
var count = state.Chromosomes.Length / 20;
return(count > 0 ? count : 1);
}, evaluator);
var crossOver = new SinglePointCrossOver(random, evaluator,
state =>
{
return(state.Chromosomes.Length / 10);
});
var selector = new UniqueChromosomeSelector(evaluator, 10, Total);
var killer = new ChromosomeKiller(state =>
{
if (state.Chromosomes.Length < 5)
{
return(0);
}
return(state.Chromosomes.Length / 5);
});
var solution = new Solution();
var geneticAlgorithm = new GeneticAlgorithmBuilder <string>()
.SetPopulation(population.Chromosomes)
.SetMutationRate(0.05)
.SetMutator(mutator)
.SetCrossOver(crossOver)
.SetKiller(killer)
.Build();
for (var i = 0; i < 9999; i++)
{
var state = geneticAlgorithm.Step();
if (state.Best.Genes == target)
{
Console.WriteLine("success!");
}
}
}
public void Run()
{
// var populations = new int[]{30,50,100,200,500,1000,3000};
// Population and tournament size box plot data made with following alogorithm
var populationSizes = new int[] { 30, 50, 100, 200, 500, 1000, 3000 };
for (int j = 0; j < 10; j++)
{
string result = "";
for (int i = 0; i < populationSizes.Length; i++)
{
var populationSize = populationSizes[i];
var problem = new FunctionMinimizationFloatingPointProblem(new F6(), 2, -50, 150);
// var populationSize = 500;
var tournamentSize = 3;
var selectionOperator = new TournamentSelectionOperator <FloatingPointChromosome>(tournamentSize, problem);
var crossoverOperator = new HeuristicCrossoverOperator(0.8);
var mutationOperator = new GaussianMutationOperator(0.05);
var elitismRate = 1.0 / populationSize;
var ga = new GeneticAlgorithmBuilder <FloatingPointChromosome>(
problem,
crossoverOperator,
mutationOperator
)
.SetGeneticAlgorithmVariant(GeneticAlgorithmVariant.Generational)
.SetPopulationSize(populationSize)
.SetElitismRate(1.0 / populationSize)
.Build();
// new GeneticAlgorithmRunner<FloatingPointChromosome>(ga).Run();
ga.Run();
result += ga.BestIndividual.Fitness;
if (i != populationSizes.Length - 1)
{
result += ",";
}
}
Console.WriteLine(result);
}
}
static void Main(string[] args)
{
// trying to make the word "abracadabra"
var evaluator = new FitnessEvaluator("abracadabra");
var random = new Random();
var generator = new Generator(random);
const int Total = 100;
var population = new Population <string>(Total, generator, evaluator);
var fitnessDict = new Dictionary <Chromosome <string>, double>();
var mutator = new Mutator();
var crossOver = new SinglePointCrossOver(random, evaluator);
var selector = new Selector(evaluator, 10, Total);
var killer = new ChromosomeKiller(5);
var solution = new Solution();
var geneticAlgorithm = new GeneticAlgorithmBuilder <string>()
.SetPopulation(population)
.SetFitnessEvaluator(evaluator)
.SetGenerations(150000)
.SetMutationRate(0.05)
.SetMutator(mutator)
.SetCrossOver(crossOver)
.SetSelection(selector)
.SetKiller(killer)
.SetSolution(solution)
.Build();
geneticAlgorithm.BeforeRun += GeneticAlgorithm_BeforeRun;
geneticAlgorithm.AfterRun += GeneticAlgorithm_AfterRun;
geneticAlgorithm.OnSolution += GeneticAlgorithm_OnSolution;
geneticAlgorithm.Run();
}
