// Genetic Algorithm maxA testing method
public static void main( String[] args )
{
// Initializing the population (we chose 500 genes for the population,
// but you can play with the population size to try different approaches)
GeneticAlgorithm population = new GeneticAlgorithm(POPULATION_SIZE);
int generationCount = 0;
// For the sake of this sample, evolution goes on forever.
// If you wish the evolution to halt (for instance, after a number of
// generations is reached or the maximum fitness has been achieved),
// this is the place to make any such checks
while(true){
// --- evaluate current generation:
population.evaluateGeneration();
// --- print results here:
// we choose to print the average fitness,
// as well as the maximum and minimum fitness
// as part of our progress monitoring
float avgFitness=0f;
float minFitness=float.PositiveInfinity;
float maxFitness=float.NegativeInfinity;
string bestIndividual="";
string worstIndividual="";
for(int i = 0; i < population.size(); i++){
float currFitness = population.getGene(i).getFitness();
avgFitness += currFitness;
if(currFitness < minFitness){
minFitness = currFitness;
worstIndividual = population.getGene(i).getPhenotype();
}
if(currFitness > maxFitness){
maxFitness = currFitness;
bestIndividual = population.getGene(i).getPhenotype();
}
}
if(population.size()>0){ avgFitness = avgFitness/population.size(); }
string output = "Generation: " + generationCount;
output += "\t AvgFitness: " + avgFitness;
output += "\t MinFitness: " + minFitness + " (" + worstIndividual +")";
output += "\t MaxFitness: " + maxFitness + " (" + bestIndividual +")";
Debug.Log(output);
// produce next generation:
population.produceNextGeneration();
generationCount++;
}
}
