public void Run()
{
var selection = new EliteSelection();
var crossover = new TwoPointCrossover();
var mutation = new MyMutation();
var fitness = new MyProblemFitness(pointList.ToArray());
var chromosome = new MyProblemChromosome(pointList.Count);
var population = new Population(100, 100, chromosome);
ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new FitnessStagnationTermination(100),
MutationProbability = 0.5f, //KS:变异概率50%
CrossoverProbability = 0.5f, //KS:交配概率50%
};
int index = 0; //KS:计算代数
ga.GenerationRan += delegate
{
var bestChromosome = ga.Population.BestChromosome;
Console.Write("Index: " + index);
Console.Write(", Fitness: {0}", bestChromosome.Fitness);
Console.Write(", Genes: {0}", string.Join("-", bestChromosome.GetGenes()));
Console.WriteLine();
index++;
};
Console.WriteLine("GA running...");
Stopwatch SW = new Stopwatch();
SW.Start();
ga.Start();//调用GA线程
SW.Stop();
//Console.Write(", Time: {0}", SW.ElapsedMilliseconds);
//Console.WriteLine("Best solution found has {0} fitness.", ga.BestChromosome.Fitness);
fitness.Evaluate(ga.BestChromosome); //KS: 评价基因的优劣
startPoint = new Point(0, (double)fitness.lr.Alpha);
endPoint = new Point(screenWidth, (double)(fitness.lr.Alpha + screenWidth * fitness.lr.Beta));
}
public void Run(ref double Minfitness, ref Point startPoint, ref Point endPoint)
{
if (linePointArr.Length == 0)
{
return;
}
else
{
var selection = new EliteSelection();
var crossover = new TwoPointCrossover();
var mutation = new MyMutation();
fitness = new MyProblemFitness2(linePointArr, startPoint, endPoint);
var chromosome = new MyProblemChromosome2(linePointArr.Length < 3 ? 3 : linePointArr.Length);
var population = new Population(linePointArr.Length + 3, linePointArr.Length + 3, chromosome);
ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new TimeEvolvingTermination(TimeSpan.FromMilliseconds(100)),
MutationProbability = 2f, //KS:变异概率50%
CrossoverProbability = 0.5f, //KS:交配概率50%
};
int index = 0; //KS:计算代数
ga.GenerationRan += delegate
{
var bestChromosome = ga.Population.BestChromosome;
/* Console.Write("Index: " + index);
* Console.Write(", Fitness: {0}", bestChromosome.Fitness);
*
* Console.Write(", Genes: {0}", string.Join("-", bestChromosome.GetGenes()));
*
* Console.WriteLine();*/
index++;
};
/* Console.WriteLine("GA running...");
* Stopwatch SW = new Stopwatch();
* SW.Start();
*/
ga.Start();//调用GA线程
// SW.Stop();
//Console.Write(", Time: {0}", SW.ElapsedMilliseconds);
Console.WriteLine("Best solution found has {0} fitness : {1}.", string.Join("-", ga.Population.BestChromosome.GetGenes()), ga.Population.BestChromosome.Fitness);
}
fitness.Evaluate(ga.Population.BestChromosome); //KS: 评价基因的优劣
Minfitness = ga.Population.BestChromosome.Fitness.Value;
startPoint = fitness.start.ToPoint();
endPoint = fitness.end.ToPoint();
/*
* if (ga.Population.BestChromosome.Fitness > Minfitness)
* {
* Minfitness = ga.Population.BestChromosome.Fitness.Value;
* startPoint = fitness.start.ToPoint();
* endPoint = fitness.end.ToPoint();
* }
* else
* {
* Minfitness = math.lerp((float)Minfitness, (float)ga.Population.BestChromosome.Fitness, 0.5f);
* }
*/
}
