public IActionResult GenerateRandomState(int dim, int populationSize, int elitism, int crossoverProb, int mutationProb, int nmbrGenerations)
{
int[][] Array = (_IFunctions.GenerateArray(dim));
GeneticVM VM = new GeneticVM()
{
array = _IFunctions.array2Dto1D(Array, dim),
dimension = dim,
heuristic = _IFunctions.Heuristic(Array, dim),
crossoverProb = crossoverProb,
elitism = elitism,
mutationProb = mutationProb,
nmbrGenerations = nmbrGenerations,
populationSize = populationSize,
dimensions = new List <SelectListItem>()
{
new SelectListItem {
Text = "4x4", Value = "4"
},
new SelectListItem {
Text = "5x5", Value = "5"
},
new SelectListItem {
Text = "6x6", Value = "6"
},
new SelectListItem {
Text = "7x7", Value = "7"
},
new SelectListItem {
Text = "8x8", Value = "8"
},
new SelectListItem {
Text = "9x9", Value = "9"
},
new SelectListItem {
Text = "10x10", Value = "10"
},
new SelectListItem {
Text = "11x11", Value = "11"
},
new SelectListItem {
Text = "12x12", Value = "12"
},
}
};
return(View("Index", VM));
}
public IActionResult Index()
{
int[] a = new int[16] {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
};
GeneticVM VM = new GeneticVM()
{
array = a,
dimension = 4,
heuristic = _IFunctions.Heuristic(_IFunctions.array1Dto2D(a, 4), 4),
dimensions = new List <SelectListItem>()
{
new SelectListItem {
Text = "4x4", Value = "4"
},
new SelectListItem {
Text = "5x5", Value = "5"
},
new SelectListItem {
Text = "6x6", Value = "6"
},
new SelectListItem {
Text = "7x7", Value = "7"
},
new SelectListItem {
Text = "8x8", Value = "8"
},
new SelectListItem {
Text = "9x9", Value = "9"
},
new SelectListItem {
Text = "10x10", Value = "10"
},
new SelectListItem {
Text = "11x11", Value = "11"
},
new SelectListItem {
Text = "12x12", Value = "12"
},
},
};
return(View(VM));
}
public IActionResult GeneticA(string stringArray, int dim, int populationSize, int elitism, double crossoverProb, double mutationProb, int nmbrGenerations)
{
int[] array = _IFunctions.StringToInt(stringArray, dim);
int[][] Array = _IFunctions.array1Dto2D(array, dim);
List <State> chromosomes = new List <State>();
chromosomes.Add(new State(dim)); //adding initial state to chromosomes/random states
chromosomes[0].niz = Array;
chromosomes[0].h = _IFunctions.Heuristic(Array, dim);
for (int i = 1; i < populationSize; i++) //adding rest of random chromosomes/random states
{
chromosomes.Add(new State(dim));
chromosomes[i].niz = _IFunctions.GenerateArray(dim);
chromosomes[i].h = _IFunctions.Heuristic(chromosomes[i].niz, dim);
}
bool promjena;
State bestChromosome;
int brojac = 0;
while (nmbrGenerations > 0)//do until we go through all generations
{
promjena = true;
while (promjena) //sort chromosomes by h
{
promjena = false;
for (int i = 0; i < populationSize - 1; i++)
{
if (chromosomes[i].h > chromosomes[i + 1].h)
{
State s = chromosomes[i];
chromosomes[i] = chromosomes[i + 1];
chromosomes[i + 1] = s;
promjena = true;
}
}
}
bestChromosome = chromosomes[0];
if (bestChromosome.h == 0) //if we find solution
{
break;
}
List <State> chromsToAdd = new List <State>();
for (int i = 0; i < populationSize * elitism / 100; i++)
{
chromsToAdd.Add(chromosomes[i]); //adding elite chromosomes to the next generation
}
for (int i = chromsToAdd.Count() / 2; i < populationSize / 2; i++) //adding the rest of chromosomes to the next generation
{
int k = rouletteSelection(chromosomes, populationSize); //roulette Selection- select a parent
State C1 = chromosomes[k];
int n;
do
{
n = rouletteSelection(chromosomes, populationSize); //select second parent
} while (n == k); //restriction not to have same chromosome as two parents
State C2 = chromosomes[n];
State C1c = C1;
State C2c = C2;
if (new Random().NextDouble() < crossoverProb)//if the propability is high enough we do the crossover between two parents
{
Crossover(C1c, C2c, dim);
}
State C1m = C1c;
State C2m = C2c;
if (new Random().NextDouble() < mutationProb)//if the propability is high enough we do the mutation of each of the parents
{
Mutation(C1m, dim);
Mutation(C2m, dim);
}
chromsToAdd.Add(C1m);
chromsToAdd.Add(C2m);
}
brojac++;
chromosomes = chromsToAdd;
nmbrGenerations--;
}
if (chromosomes[0].h != 0)
{
promjena = true;
while (promjena) //because on the end of while loop we don't sort the final generation, so we do it here to
{
promjena = false;
for (int i = 0; i < populationSize - 1; i++)
{
if (chromosomes[i].h > chromosomes[i + 1].h)
{
State s = chromosomes[i];
chromosomes[i] = chromosomes[i + 1];
chromosomes[i + 1] = s;
promjena = true;
}
}
}
}
GeneticVM VM = new GeneticVM()
{
array = _IFunctions.array2Dto1D(chromosomes[0].niz, dim),
dimension = dim,
heuristic = chromosomes[0].h,
crossoverProb = crossoverProb,
elitism = elitism,
mutationProb = mutationProb,
nmbrGenerations = nmbrGenerations,
populationSize = populationSize,
dimensions = new List <SelectListItem>()
{
new SelectListItem {
Text = "4x4", Value = "4"
},
new SelectListItem {
Text = "5x5", Value = "5"
},
new SelectListItem {
Text = "6x6", Value = "6"
},
new SelectListItem {
Text = "7x7", Value = "7"
},
new SelectListItem {
Text = "8x8", Value = "8"
},
new SelectListItem {
Text = "9x9", Value = "9"
},
new SelectListItem {
Text = "10x10", Value = "10"
},
new SelectListItem {
Text = "11x11", Value = "11"
},
new SelectListItem {
Text = "12x12", Value = "12"
},
},
counter = brojac
};
return(PartialView("GeneticDone", VM));
}
