public void Elitism()
{
population.OrderByDescending(rrs => rrs.fitness).ToList();
individual temp = population[0];
population.Clear();
population.Add(temp);
firstPop();
}
static void Copychild(individual source, individual destination)
{
// 去新建一个对象 按照常规失去了作用域
// 这个是错误的，无法计算出应有的结果数值
// destination = new individual();
for (int j = 0; j < DNAsize; j++)
{
destination.num[j] = source.num[j];
}
destination.selectP = source.selectP;
destination.fitness = source.fitness;
}
private void ExchangeGeneticMaterial(individual P1, individual P2, out individual F1, out individual F2)
{
int chromossomesize = CHROMOSSOME_SIZE / 2;
int indexX = random.Next(chromossomesize);
int indexY = chromossomesize + random.Next(chromossomesize);
int[] chromosome_F1 = new int[CHROMOSSOME_SIZE];
int[] chromosome_F2 = new int[CHROMOSSOME_SIZE];
for (int i1 = 0, i2 = chromossomesize; i1 < chromossomesize; i1++, i2++)
{
chromosome_F1[i1] = (i1 < indexX) ? P1.chromosome[i1] : P2.chromosome[i1];
chromosome_F2[i1] = (i1 < indexX) ? P2.chromosome[i1] : P1.chromosome[i1];
chromosome_F1[i2] = (i2 < indexY) ? P1.chromosome[i2] : P2.chromosome[i2];
chromosome_F2[i2] = (i2 < indexY) ? P2.chromosome[i2] : P1.chromosome[i2];
}
F1 = new individual(chromosome_F1);
F2 = new individual(chromosome_F2);
}
//
static void Main(string[] args)
{
Console.WriteLine("遗传算法");
//传入种群数，DNA 数 迭代次数
int NumberGenerations = 200;
individual[] Population = null;
GeneticAlgorithm ga = new GeneticAlgorithm();
//生成初始化的种群。。。
//生成一系列的DNA序列数,返回种群
Console.WriteLine("初始化: ");
individual[] pop = ga.InitialPopulation(Population);
// translate DNA
//根据迭代次数去迭代计算
int count = 0;
for (int iiii = 0; iiii < NumberGenerations; iiii++)
{
ga.GetFitness(pop);
//这里得到的也是一个引用的类型
// 这个会交叉引用会变数值的，有可能
pop = ga.PopProduce(pop);
//去得到每次的函数的数值
Console.WriteLine("第{0}次迭代", count);
count++;
for (int m = 0; m < ga.Popsize; m++)
{
//去得到函数值
// most fitted DNA is
double variableX = ga.TranslateDNA(pop[m]);
//we define the fitness is the value of the function
double functionValue = ga.Function(variableX);
Console.WriteLine("{0}, {1}, {2}, {3}", variableX, functionValue, pop[m].fitness, pop[m].selectP);
}
//给每个群体一个适应度
// 根据 fitness 和轮盘转法确定出优秀的群体信息
// 去copy 一份
individual[] PopCopy = new individual[popsize];
for (int iii = 0; iii < popsize; iii++)
{
PopCopy[iii] = new individual();
}
//这个依旧是引用，不是复制了
Copy(pop, PopCopy);
//父代的信息已经保存到了 parentcopy了
individual child = new individual();
// 选择交叉和基因变异
// 新建一个child的对象的类
individual[] Child = new individual[popsize];
// 进行初始化
for (int ii = 0; ii < popsize; ii++)
{
Child[ii] = new individual();
}
for (int j = 0; j < popsize; j++)
{
child = ga.CrossOver(pop[j], PopCopy);
child = ga.Mutation(child);
//对象引用了而不是复制了
//进行复制 child
// parentcopy[j] = child;
// 会丢失一段DNA的信息
Copychild(child, Child[j]);
}
pop = Child;
}
// 去打印出最大的数值，这样才是比较好的了。。。
double maxValue = 0;
for (int i = 0; i < popsize; i++)
{
//如何去处理一下
if (maxValue < pop[i].fitness)
{
maxValue = pop[i].fitness;
}
}
Console.WriteLine("最大的数值{0}", maxValue);
}