public static void Main(string[] args)
{
if (args.Length < 1)
{
Console.Error.WriteLine("You have to at least specify the data file path");
return;
}
var mutation = new UniformMutation(MutationProbability);
var selection = new KTournamentSelection(TournamentSize);
var crossover = new ArithmeticCrossover();
var fitnessFunction = new FitnessFunction(args[0]);
IGeneticAlgorithm <DecimalArrayChromosome> geneticAlgorithm;
if (args.Length == 2 && args[1].ToLower() == "gen")
{
geneticAlgorithm = new GenerationGeneticAlgorithm(mutation, selection, crossover, fitnessFunction,
IterationLimit, FitnessTerminator, PopulationSize);
}
else
{
geneticAlgorithm = new EliminationGeneticAlgorithm(mutation, selection, crossover, fitnessFunction,
IterationLimit, FitnessTerminator, PopulationSize);
}
var optimum = geneticAlgorithm.FindOptimum();
Console.WriteLine();
Console.WriteLine(optimum);
}
private static void Run()
{
var dataset = new Dataset();
var nn = new NeuralNetwork(dataset, 8); //8, 4
const int populationSize = 10;
const int iterations = 1_000_000;
var crossovers = new ICrossover[]
{
new BlxDoubleCrossover(),
new ArithmeticCrossover(),
new DiscreteUniformRecombination(),
new OnePointCrossover(),
new SimpleArithmeticRecombination()
};
var mutations = new IMutation[]
{
new GaussianAdditiveMutation(mutationProbability: 0.02, deviation: 0.2),
new GaussianAdditiveMutation(mutationProbability: 0.02, deviation: 0.4),
new GaussianSwapMutation(mutationProbability: 0.02, deviation: 0.5)
};
var desirability = new[] { 18.0, 4.0, 2.0 };
var combinedCrossover = new CombinedCrossover(crossovers);
var combinedMutation = new CombinedMutation(mutations, desirability);
var selection =
new KTournamentSelection(nn.MeanSquareError, combinedCrossover, combinedMutation, k: 3);
var parametersCount = nn.GetNumberOfRequiredParameters();
var parameters = new double[parametersCount];
var doubleGa = new DoubleGa(nn, parameters, parametersCount, populationSize, iterations);
var bestNetworkParameters = doubleGa.FindBestIndividual(speedRun: true, selection).Representation;
var success = 0;
Console.WriteLine($"\nPrediction | Actual");
foreach (var sample in dataset)
{
var prediction = nn.CalculateOutput(sample.X, sample.Y, bestNetworkParameters);
var A = prediction[0] < 0.5 ? 0 : 1;
var B = prediction[1] < 0.5 ? 0 : 1;
var C = prediction[2] < 0.5 ? 0 : 1;
Console.WriteLine($"[{A} {B} {C}] | [{sample.A} {sample.B} {sample.C}]");
if (A == sample.A && B == sample.B && C == sample.C)
{
success++;
}
}
Console.WriteLine(
$"Prediction rate: {success}/{dataset.DatasetCount()} = {(double) success / dataset.DatasetCount() * 100}%");
}
public static void Main(string[] args)
{
var dataset = new Dataset.Dataset(DatasetFilePath);
var ann = new ANN(Architecture);
var fitness = new FitnessFunction(ann, dataset);
var mutation = new GaussMutation(MutationThreshold, MutationProbability, Sigma1, Sigma2);
var selection = new KTournamentSelection <DoubleArrayChromosome>(TournamentSize);
var crossovers = new List <ICrossover <DoubleArrayChromosome> >()
{
new ArithmeticCrossover(),
new HeuristicCrossover(),
new UniformCrossover()
};
var geneticAlgorithm =
new EliminationGeneticAlgorithm(mutation, selection, crossovers, fitness, IterationLimit, ErrorLimit, PopulationSize);
var optimum = geneticAlgorithm.FindOptimum();
var correctClassification = 0;
foreach (var sample in dataset)
{
var classification = ann.CalculateOutput(sample.Input, optimum.Values);
var correct = true;
for (int i = 0; i < classification.Length; i++)
{
classification[i] = classification[i] < 0.5 ? 0 : 1;
if (Math.Abs(classification[i] - sample.Classification[i]) > 10e-9)
{
correct = false;
}
}
Console.WriteLine(classification[0] + " " + classification[1] + " " + classification[2] +
" <=> " + sample.Classification[0] + " " + sample.Classification[1] + " " + sample.Classification[2] + " ");
if (correct)
{
correctClassification++;
}
}
Console.WriteLine("Correct => " + correctClassification + ", Total => " + dataset.Count());
ann.WriteNeuronLayerParametersToFile(ParametersFilePath, 1, optimum.Values);
}
public static void Main(string[] args)
{
FitnessFunction.Rides = ParseData(out int fleetSize, out int bonus, out int steps);
Chromosome.FleetSize = fleetSize;
FitnessFunction.NumberOfSteps = steps;
FitnessFunction.Bonus = bonus;
var mutation = new UniformMutation(MutationProbability);
var selection = new KTournamentSelection(TournamentSize);
var crossover = new UniformCrossover();
var fitnessFunction = new FitnessFunction();
var geneticAlgorithm = new EliminationGeneticAlgorithm(mutation, selection, crossover, fitnessFunction,
IterationLimit, FitnessTerminator, PopulationSize, FitnessFunction.Rides.Count);
var optimum = geneticAlgorithm.FindOptimum();
WriteOutput(optimum);
}
public static void Main(string[] args)
{
var busStops = new List <BusStop>();
var students = new List <Student>();
ParseFromFile(busStops, students);
var mutation = new StudentBusMutation(MutationProbability);
var selection = new KTournamentSelection(TournamentSize);
var crossover = new StudentBusCrossover();
var fitnessFunction = new FitnessFunction.FitnessFunction();
var geneticAlgorithm = new EliminationGeneticAlgorithm(mutation, selection, crossover, fitnessFunction,
IterationLimit, FitnessTerminator, PopulationSize, students, busStops, InstanceFilePath);
var optimum = geneticAlgorithm.FindOptimum();
Console.WriteLine();
Console.WriteLine(optimum.Fitness);
optimum.Students = optimum.Students.OrderBy(x => x.Id).ToList();
HelperMethods.WriteToFile(optimum.ToString(), InstanceResultFilePath);
}
