// Starts and executes algorithm
public void Run(int maxRepeat = 9999, double minFitness = 0.999)
{
if (_prototype == null)
{
return;
}
Initialize();
_currentArchiveSize = _populationSize;
// Current generation
int currentGeneration = 0;
int repeat = 0;
double lastBestFit = 0.0;
for (; ;)
{
var best = Result;
if (currentGeneration > 0)
{
var status = string.Format("\rFitness: {0:F6}\t Generation: {1}", best.Fitness, currentGeneration);
Console.Write(status);
// algorithm has reached criteria?
if (best.Fitness > minFitness)
{
FinalizePopulation();
break;
}
double difference = Math.Abs(best.Fitness - lastBestFit);
if (difference <= 0.0000001)
{
++repeat;
}
else
{
repeat = 0;
}
if (repeat > (maxRepeat / 100))
{
++_crossoverProbability;
}
}
CreateParentPopulation();
CreateOffspringPopulation();
MutateOffspringPopulation();
UpdateArchivePopulation();
Configuration.Seed();
++currentGeneration;
}
}
// Starts and executes algorithm
public void Run(int maxRepeat = 9999, double minFitness = 0.999)
{
if (_prototype == null)
{
return;
}
// clear best chromosome group from previous execution
ClearBest();
Initialize(_chromosomes);
// Current generation
int currentGeneration = 0;
int repeat = 0;
double lastBestFit = 0.0;
for (; ;)
{
var best = Result;
var status = string.Format("\rFitness: {0:F6}\t Generation: {1}", best.Fitness, currentGeneration++);
Console.Write(status);
// algorithm has reached criteria?
if (best.Fitness > minFitness)
{
break;
}
var difference = Math.Abs(best.Fitness - lastBestFit);
if (difference <= 0.0000001)
{
++repeat;
}
else
{
repeat = 0;
}
if (repeat > (maxRepeat / 100))
{
ReplaceByGeneration = _replaceByGeneration * 3;
++_crossoverProbability;
}
Replacement(_chromosomes);
Configuration.Seed();
lastBestFit = best.Fitness;
}
}
// Starts and executes algorithm
public void Run(int maxRepeat = 9999, double minFitness = 0.999)
{
if (_prototype == null)
{
return;
}
var population = new List <T>();
Initialize(population);
// Current generation
int currentGeneration = 0;
int repeat = 0;
double lastBestFit = 0.0;
for (; ;)
{
var best = Result;
if (currentGeneration > 0)
{
var status = string.Format("\rFitness: {0:F6}\t Generation: {1}", best.Fitness, currentGeneration);
Console.Write(status);
// algorithm has reached criteria?
if (best.Fitness > minFitness)
{
break;
}
double difference = Math.Abs(best.Fitness - lastBestFit);
if (difference <= 0.0000001)
{
++repeat;
}
else
{
repeat = 0;
}
if (repeat > (maxRepeat / 100))
{
++_crossoverProbability;
}
}
/******************* crossover *****************/
var offspring = Replacement(population);
/******************* mutation *****************/
foreach (var child in offspring)
{
child.Mutation(_mutationSize, _mutationProbability);
}
var totalChromosome = new List <T>(population);
totalChromosome.AddRange(offspring);
/******************* non-dominated sorting *****************/
var front = NonDominatedSorting(totalChromosome);
/******************* selection *****************/
population = Selection(front, totalChromosome);
_populationSize = population.Count;
/******************* comparison *****************/
if (currentGeneration == 0)
{
_chromosomes = population.ToArray();
}
else
{
totalChromosome = new List <T>(population);
totalChromosome.AddRange(_chromosomes);
var newBestFront = NonDominatedSorting(totalChromosome);
_chromosomes = Selection(newBestFront, totalChromosome).ToArray();
lastBestFit = best.Fitness;
}
Configuration.Seed();
++currentGeneration;
}
}