public static Population GeneratePopulation(IRandom rand, int populationSize, int numberOfChromosomes, int geneLength, int minChromosomeLength, int maxChromosomeLength, List<Chromosome> geneBlacklist, List<Chromosome> geneWhitelist, Func<IOrganism, IOrganism> correctionFunction)
{
Population candidates = new Population();
RandomChromosomeGenerator generator = new RandomChromosomeGenerator(rand, new ChromosomePart(rand) {
GeneLength = geneLength,
GeneBlacklist = geneBlacklist,
GeneWhitelist = geneWhitelist,
MinimumChromosomeLength = minChromosomeLength,
MaximumChromosomeLegnth = maxChromosomeLength
});
for (int organismId = 0; organismId <= populationSize - 1; organismId++) {
IOrganism organism = new Organism();
for (int chromosomeId = 0; chromosomeId <= numberOfChromosomes - 1; chromosomeId++) {
organism.Chromosomes.Add(generator.GetChromosome());
}
if (correctionFunction != null) {
organism = correctionFunction(organism);
}
candidates.Add(organism);
}
return candidates;
}
public GeneticAlgorithm(IRandom rand, Population candidates, Func<IOrganism, double> fitnessFunction, ISelector selector, IMutator mutator, ICrossover crossLinker, double elitismProportion, int maximumGenerations)
{
this.rand = rand;
this.candidates = candidates;
this.fitnessFunction = fitnessFunction;
this.selector = selector;
this.mutator = mutator;
this.crossLinker = crossLinker;
this.elitismProportion = elitismProportion;
this._generationCount = 0;
this.generationInformation = new Dictionary<int, double>();
this.initialCount = candidates.Count;
this.maximumGenerations = maximumGenerations;
this.candidates.CalculateFitnesses(this.fitnessFunction);
}
private void CreateNextGeneration()
{
Population nextGeneration = new Population();
this.selector.SetCandidates(this.candidates);
for (int i = 0; i <= Convert.ToInt32((this.candidates.Count / 2) * (1 - this.elitismProportion)) - 1; i++) {
this.CreateOffspring(nextGeneration);
}
// Elitism involves bringing the top X% of the current into the next generation
if (this.elitismProportion > 0) {
int numberToKeep = Convert.ToInt32(this.elitismProportion * this.initialCount);
this.candidates.GetBest(this.elitismProportion).ToList().ForEach(c => nextGeneration.Add(c));
}
this.candidates = nextGeneration;
this.candidates.CalculateFitnesses(this.fitnessFunction);
this.generationInformation.Add(this.GenerationCount, this.candidates.TotalFitness / this.candidates.Count);
this._generationCount += 1;
}
public void SetCandidates(Population candidates)
{
this.candidates = candidates;
}
private void CreateOffspring(Population nextGeneration)
{
var parents = this.selector.PickCandidates(2).Take(2).ToList();
var offspring = this.crossLinker.CrossLink(parents[0], parents[1]);
this.mutator.Mutate(offspring.Item1);
this.mutator.Mutate(offspring.Item2);
if (this.encodingCorrector != null) {
offspring = new Tuple<IOrganism, IOrganism>(this.encodingCorrector(offspring.Item1), this.encodingCorrector(offspring.Item2));
}
nextGeneration.Add(offspring.Item1);
nextGeneration.Add(offspring.Item2);
}
public GeneticAlgorithm(IRandom rand, Population candidates, Func<IOrganism, double> fitnessFunction)
: this(rand, candidates, fitnessFunction, new LinearRankingSelector(rand), new Mutator(rand, 0.05), new OnePointCrossover(rand, 0.7), 0.1, 15000)
{
this.mutator.AddMutator(new BitFlipMutation(rand));
}