public static Sudoku Eval(Sudoku sudoku, int populationSize, double fitnessThreshold, int generationNb)
{
//creation du chromosome
IChromosome chromosome = new SudokuPermutationsChromosome(sudoku);
var fitness = new SudokuFitness(sudoku);
var selection = new EliteSelection();
var crossover = new UniformCrossover();
var mutation = new UniformMutation();
var population = new Population(populationSize, populationSize, chromosome);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new OrTermination(new ITermination[]
{
new FitnessThresholdTermination(fitnessThreshold),
new GenerationNumberTermination(generationNb)
})
};
ga.Start();
var bestIndividual = ((ISudokuChromosome)ga.Population.BestChromosome);
var solutions = bestIndividual.GetSudokus();
return(solutions[0]);
}
public static Sudoku Eval(Sudoku sudoku, int populationSize, double fitnessThreshold, int generationNb)
{
//creation du chromosome
IChromosome chromosome = new SudokuPermutationsChromosome(sudoku);
//variable qui indique l'erreure
var fitness = new SudokuFitness(sudoku);
//choix de la selection : ici elite
var selection = new EliteSelection();
//var selection = new RouletteWheelSelection();
//var selection = new SelectionException();
//var selection = new TournamentSelection();
//Choix du crossover : ici uniform
var crossover = new UniformCrossover();
//choix de la mutation : ici uniform
var mutation = new UniformMutation();
//creation de la population
var population = new Population(populationSize, populationSize, chromosome);
//création de l'algo génétique
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
//pour mettre fin a l'exec si on atteint le threshold ou le nombre de rep fixé
Termination = new OrTermination(new ITermination[]
{
new FitnessThresholdTermination(fitnessThreshold),
new GenerationNumberTermination(generationNb)
})
};
ga.Start();
//recupération de la meilleure solution
var bestIndividual = ((ISudokuChromosome)ga.Population.BestChromosome);
var solutions = bestIndividual.GetSudokus();
return(solutions[0]);
}
public override IChromosome CreateNew()
{
var toReturn = new SudokuPermutationsChromosome(TargetSudoku);
return(toReturn);
}