protected IBreeding <Synapse> NewBreeding(System.Random random)
{
ISelection <Synapse> selector;
if (roulteWheelSelection)
{
selector = new RouletteWheelSelection <Synapse>(random, 2);
}
else
{
selector = new EliteSelection <Synapse>(2);
}
var crossover = new SinglePointCrossover <Synapse>(random);
var mutator = new FixedNeuralNetMutation(
random,
geneMutChance,
mutRange);
return(new BreedingBase <Synapse>(
poplLne - (int)(poplLne * reinsertPart),
selector,
crossover,
mutator));
}
public void Start_TplManySlowFitness_Timeout()
{
var taskExecutor = new TplTaskExecutor();
taskExecutor.Timeout = TimeSpan.FromMilliseconds(1000);
var selection = new RouletteWheelSelection();
var crossover = new OnePointCrossover(1);
var mutation = new UniformMutation();
var chromosome = new ChromosomeStub();
var target = new GeneticAlgorithm(new TplPopulation(100, 150, chromosome),
new FitnessStub()
{
SupportsParallel = true, ParallelSleep = 1500
}, selection, crossover, mutation);
target.OperatorsStrategy = new TplOperatorsStrategy();
target.TaskExecutor = taskExecutor;
Assert.Catch <TimeoutException>(() =>
{
target.Start();
}, "The fitness evaluation reached the 00:00:01 timeout.");
Assert.IsFalse(target.IsRunning);
Assert.AreEqual(GeneticAlgorithmState.Stopped, target.State);
}
public void CheckIfAllAreSelected()
{
var selection = new RouletteWheelSelection();
var generationMng = new GenerationManagerKeepLast();
var genomes = new List <GenomeBase>(50);
for (int i = 0; i < genomes.Count; i++)
{
var newGenome = new GenomeBase
{
Fitness = i - 25
};
genomes.Add(newGenome);
}
generationMng.RegisterNewGeneration(new Generation(genomes.ToArray(), 0));
var roulette = new RouletteWheelSelection();
roulette.Prepare(generationMng.GetGenomes().ToArray(), null, null, genomes.Count);
var selected = roulette.Select(genomes.Count).ToArray();
Assert.True(selected.Count() == genomes.Count);
Assert.False(selected.Any(x => x == null));
foreach (var genome in selected)
{
Assert.True(selected.Count(x => x == genome) == 1);
}
}
public void Start_InvalidFitnessEvaluateResult_Exception()
{
var selection = new RouletteWheelSelection();
var crossover = new OnePointCrossover(1);
var mutation = new UniformMutation();
var chromosome = new ChromosomeStub();
var fitness = MockRepository.GenerateMock <IFitness>();
fitness.Expect(f => f.Evaluate(null)).IgnoreArguments().Return(1.1);
var target = new GeneticAlgorithm(
new Population(20, 20, chromosome),
fitness, selection, crossover, mutation);
ExceptionAssert.IsThrowing(new FitnessException(fitness, "The {0}.Evaluate returns a fitness with value 1.1. The fitness value should be between 0.0 and 1.0.".With(fitness.GetType())), () => {
target.Start();
});
fitness = MockRepository.GenerateMock <IFitness>();
fitness.Expect(f => f.Evaluate(null)).IgnoreArguments().Return(-0.1);
target = new GeneticAlgorithm(
new Population(20, 20, chromosome),
fitness, selection, crossover, mutation);
ExceptionAssert.IsThrowing(new FitnessException(fitness, "The {0}.Evaluate returns a fitness with value -0.1. The fitness value should be between 0.0 and 1.0.".With(fitness.GetType())), () =>
{
target.Start();
});
}
public ISelection RouletteWheel()
{
var target = new RouletteWheelSelection();
target.SelectChromosomes(_chromosomesNumber, _generation);
return(target);
}
public void SelectChromosomes_Generation_ChromosomesZeroFitness()
{
var target = new RouletteWheelSelection();
var c1 = Substitute.ForPartsOf <ChromosomeBase>(2);
c1.Fitness = 0;
var c2 = Substitute.ForPartsOf <ChromosomeBase>(2);
c2.Fitness = 0;
var c3 = Substitute.ForPartsOf <ChromosomeBase>(2);
c3.Fitness = 0;
var c4 = Substitute.ForPartsOf <ChromosomeBase>(2);
c4.Fitness = 0;
var generation = new Generation(1, new List <IChromosome>()
{
c1, c2, c3, c4
});
var actual = target.SelectChromosomes(2, generation);
Assert.AreEqual(0, actual.Count);
}
public void OnCollisionEnter2D(Collision2D collision)
{
NotifyObservers("UpdateScore");
Instantiate(RouletteWheelSelection.GetRandomByWeight(posiblesPowerUPS, weights), this.transform.position, this.transform.rotation);
Instantiate(explosion, this.transform.position, this.transform.rotation);
_enemyPool.ReturnEnemyToPool(this);
}
public void Start_ParallelManySlowFitness_Timeout()
{
var taskExecutor = new SmartThreadPoolTaskExecutor();
taskExecutor.MinThreads = 100;
taskExecutor.MaxThreads = 100;
taskExecutor.Timeout = TimeSpan.FromMilliseconds(1000);
var selection = new RouletteWheelSelection();
var crossover = new OnePointCrossover(1);
var mutation = new UniformMutation();
var chromosome = new ChromosomeStub();
var target = new GeneticAlgorithm(new Population(100, 150, chromosome),
new FitnessStub()
{
SupportsParallel = true, ParallelSleep = 1500
}, selection, crossover, mutation);
target.TaskExecutor = taskExecutor;
ExceptionAssert.IsThrowing(new TimeoutException("The fitness evaluation rech the 00:00:01 timeout."), () =>
{
target.Start();
});
Assert.IsFalse(target.IsRunning);
Assert.AreEqual(GeneticAlgorithmState.Stopped, target.State);
}
private BreedingBase <int> NewBreeding(
int poplLen,
double partToReinsert,
double geneMutChance,
int mutRange)
{
BreedingBase <int> result;
ISelection <int> selector;
ICrossover <int> crossover;
IMutation <int> mutator;
//selector = new EliteSelection<int>(2, partToBeElites: 0.5d);
selector = new RouletteWheelSelection <int>(RandomInst, 2);
crossover = new SinglePointCrossover <int>(RandomInst, 1);
mutator = new ES_Mutation(RandomInst, geneMutChance, mutRange);
result = new BreedingBase <int>(
poplLen - (int)(partToReinsert * poplLen),
selector,
crossover,
mutator
);
return(result);
}
public void ItHasAValidConstructor()
{
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(_pool, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
parentSelection.GetParents();
}
public void ItConstructsProperlyIfAllFitnessScoresAreAboveOrEqualToZero()
{
var genome = GATestHelper.GetTravelingSalesmanPopulation();
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(genome, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
}
public static GeneticSharp.Domain.GeneticAlgorithm MakeDefault(Func <double[], double> func, Tuple <double, double>[] minmax)
{
var population = new Population(20, 40, chromosome.CreateFloatChromosone(minmax));
var fitness = new FuncFitness((c) =>
{
var fc = c as FloatingPointChromosome;
var values = fc.ToFloatingPoints();
return(func(values));
});
var selection = new RouletteWheelSelection();
var crossover = new UniformCrossover(0.5f);
var mutation = new FlipBitMutation();
var ga = new GeneticSharp.Domain.GeneticAlgorithm(
population,
fitness,
selection,
crossover,
mutation);
var termination = new GenerationNumberTermination(1000);
ga.Termination = termination;
Console.WriteLine("Generation: (x1, y1), (x2, y2) = distance");
return(ga);
}
private void Start()
{
var fitness = new TspFitness(m_numberOfCities);
var chromosome = new TspChromosome(m_numberOfCities);
// This operators are classic genetic algorithm operators that lead to a good solution on TSP,
// but you can try others combinations and see what result you get.
var crossover = new OrderedCrossover();
var mutation = new ReverseSequenceMutation();
var selection = new RouletteWheelSelection();
var population = new Population(50, 100, chromosome);
m_ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
m_ga.Termination = new TimeEvolvingTermination(System.TimeSpan.FromHours(1));
// The fitness evaluation of whole population will be running on parallel.
m_ga.TaskExecutor = new ParallelTaskExecutor
{
MinThreads = 100,
MaxThreads = 200
};
// Everty time a generation ends, we log the best solution.
m_ga.GenerationRan += delegate
{
var distance = ((TspChromosome)m_ga.BestChromosome).Distance;
Debug.Log($"Generation: {m_ga.GenerationsNumber} - Distance: ${distance}");
};
DrawCities();
// Starts the genetic algorithm in a separate thread.
m_gaThread = new Thread(() => m_ga.Start());
m_gaThread.Start();
}
private Species GetKey(Vector3 worldPos)
{
var sqr = radius * radius;
var nears = lifes
.Where(v => (v.transform.position - worldPos).sqrMagnitude < sqr);
var count = nears.Count();
int modelIndex = -1;
int colorIndex = -1;
if (count > 0 && !Interfered())
{
var modelTabs = new Tabulator <int, float>();
foreach (var n in nears.Select(v => v.CurrentKey))
{
modelTabs[n.model]++;
}
var max = modelTabs.Values.Max();
int valueIndex;
if (RouletteWheelSelection.Sample(out valueIndex, 100, max, modelTabs.ValuesAsList))
{
modelIndex = modelTabs.KeysAsList[valueIndex];
}
var colors = nears.Where(v => v.CurrentKey.model == modelIndex);
var colorCount = colors.Count();
if (colorCount > 0 && !Interfered())
{
var colorTabs = new Tabulator <int, float>();
foreach (var n in colors.Select(v => v.CurrentKey))
{
colorTabs[n.color]++;
}
var maxColor = colorTabs.Values.Max();
int colorTabIndex;
if (RouletteWheelSelection.Sample(out colorTabIndex, 100, maxColor, colorTabs.ValuesAsList))
{
colorIndex = colorTabs.KeysAsList[colorTabIndex];
}
}
}
if (modelIndex < 0)
{
modelIndex = Random.Range(0, LifeStrage.Instance.models.Length);
}
var model = LifeStrage.Instance.GetModel(modelIndex);
if (colorIndex < 0)
{
colorIndex = Random.Range(0, model.colors.Length);
}
var spec = new Species(modelIndex, colorIndex);
Debug.LogFormat("Chosen key={0} neighbors={1}", spec, count);
return(spec);
}
public void SelectChromosomes_NullGeneration_Exception()
{
var target = new RouletteWheelSelection();
ExceptionAssert.IsThrowing(new ArgumentNullException("generation"), () => {
target.SelectChromosomes(2, null);
});
}
public void ItFailsIfANegativeFitnessScoreIsPassed()
{
var genome = GATestHelper.GetTravelingSalesmanPopulation();
genome[0].FitnessScore = -1;
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(genome, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
}
public void ItCanGetAValidParent()
{
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(_pool, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
var parent = parentSelection.GetParent(0.22);
Assert.IsNotNull(parent);
}
public void SelectChromosomes_NullGeneration_Exception()
{
var target = new RouletteWheelSelection();
var actual = Assert.Catch <ArgumentNullException>(() =>
{
target.SelectChromosomes(2, null);
});
Assert.AreEqual("generation", actual.ParamName);
}
public void ItCantReturnTheSameParentTwice()
{
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(_pool, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
for (int i = 0; i < 1000; i++)
{
var parent = parentSelection.GetParents();
Assert.IsFalse(ReferenceEquals(parent.Father, parent.Mother));
}
}
public void ItCanReturnParents()
{
var parentSelection = new RouletteWheelSelection();
parentSelection.Setup(_pool, GATestHelper.GetTravelingSalesmanDefaultConfiguration());
var parent = parentSelection.GetParents();
Assert.IsNotNull(parent);
Assert.IsNotNull(parent.Father);
Assert.IsNotNull(parent.Mother);
}
static void Main(string[] args)
{
ICrossOperator crossOperator = new OnePointCross();
IMutationOperator mutationOperator = new Mutatation();
ISelectionOperator selectionOperator = new RouletteWheelSelection();
Func <double, double> function = FitnessFunction;
GeneticAlgorithm algorithm = new GeneticAlgorithm(20, 5, crossOperator, mutationOperator, selectionOperator, function);
for (int i = 0; i < 10; i++)
{
Console.WriteLine(algorithm.RunSimulation(100).Fitness);
}
Console.ReadLine();
}
public void Initialize(int numberOfCities, int areaWidth, int areaHeight)
{
Stop();
Fitness = new TspFitness(numberOfCities, areaWidth, areaHeight);
var chromosome = new TspChromosome(numberOfCities);
// This operators are classic genetic algorithm operators that lead to a good solution on TSP,
// but you can try others combinations and see what result you get.
var crossover = new OrderedCrossover();
var mutation = new ReverseSequenceMutation();
var selection = new RouletteWheelSelection();
var population = new Population(50, 100, chromosome);
_ga = new GeneticAlgorithm(population, Fitness, selection, crossover, mutation);
}
public void Start_FitnessEvaluationFailed_FitnessException()
{
var selection = new RouletteWheelSelection();
var crossover = new OnePointCrossover(1);
var mutation = new UniformMutation();
var chromosome = new ChromosomeStub();
var fitness = new FuncFitness((c) => throw new Exception("TEST"));
var target = new GeneticAlgorithm(new Population(100, 150, chromosome), fitness, selection, crossover, mutation);
Assert.Catch <FitnessException>(target.Start);
Assert.IsFalse(target.IsRunning);
Assert.AreEqual(GeneticAlgorithmState.Stopped, target.State);
}
public void SelectChromosomes_InvalidNumber_Exception()
{
var target = new RouletteWheelSelection();
ExceptionAssert.IsThrowing(new ArgumentOutOfRangeException("number", "The number of selected chromosomes should be at least 2."), () => {
target.SelectChromosomes(-1, null);
});
ExceptionAssert.IsThrowing(new ArgumentOutOfRangeException("number", "The number of selected chromosomes should be at least 2."), () => {
target.SelectChromosomes(0, null);
});
ExceptionAssert.IsThrowing(new ArgumentOutOfRangeException("number", "The number of selected chromosomes should be at least 2."), () => {
target.SelectChromosomes(1, null);
});
}
public static void NextGen()
{
Genome c1 = new Genome(false);
Genome c2 = new Genome(false);
RouletteWheelSelection select = new RouletteWheelSelection();
CrossOver cross = new CrossOver();
cross.SinglePointCrossOver(select.RouletteWheelSelect(), select.RouletteWheelSelect(), ref c1, ref c2);
Mutate(ref c1);
Mutate(ref c2);
genome.Add(c1);
genome.Add(c2);
if (genome.Count == DNAattributes.PopulationSize)
{
DNAattributes.pop = genome;
}
}
private void GenerateLevelGS()
{
Level = new LevelElement[LevelSize];
for (var i = 0; i < nChunks; i++)
{
var dcParams = DifficultyCurveParams;
var difficulty = DifficultyCurve(i, dcParams.a, dcParams.b, dcParams.k);
var eParams = EvolutionParams;
var chunkSize = LevelSize / nChunks;
var fitness = new LevelChunkFitness(difficulty);
var chromosome = new LevelChunkChromosome(chunkSize);
var crossover = new OnePointCrossover(chunkSize / 2);
var mutation = new UniformMutation(allGenesMutable: true);
var selection = new RouletteWheelSelection();
var population = new Population(50, 100, chromosome);
var m_ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation)
{
Termination = new TimeEvolvingTermination(new TimeSpan(0, 0, 1))
};
m_ga.TaskExecutor = new ParallelTaskExecutor
{
MinThreads = 2,
MaxThreads = 5
};
m_ga.Start();
// Everty time a generation ends, we log the best solution.
/*m_ga.GenerationRan += delegate
* {
* var f = m_ga.BestChromosome.Fitness;
* Debug.Log($"Generation: {m_ga.GenerationsNumber} - Fitness: ${f}");
* };*/
var genes = m_ga.BestChromosome.GetGenes();
for (int j = 0; j < chunkSize; j++)
{
Level[i * chunkSize + j] = (LevelElement)genes[j].Value;
}
Debug.Log($"Fitness: {m_ga.BestChromosome.Fitness}");
}
}
public void SelectChromosomes_InvalidNumber_Exception()
{
var target = new RouletteWheelSelection();
Assert.Catch <ArgumentOutOfRangeException>(() =>
{
target.SelectChromosomes(-1, null);
}, "The number of selected chromosomes should be at least 2.");
Assert.Catch <ArgumentOutOfRangeException>(() =>
{
target.SelectChromosomes(0, null);
}, "The number of selected chromosomes should be at least 2.");
Assert.Catch <ArgumentOutOfRangeException>(() =>
{
target.SelectChromosomes(1, null);
}, "The number of selected chromosomes should be at least 2.");
}
public void SelectChromosomes_NullFitness_Exception()
{
var target = new RouletteWheelSelection();
var generation = new Generation(1, new List <IChromosome>
{
new TspChromosome(10),
new TspChromosome(10),
new TspChromosome(10),
new TspChromosome(10),
new TspChromosome(10)
});
var actual = Assert.Catch <SelectionException>(() =>
{
target.SelectChromosomes(2, generation);
});
Assert.AreEqual("RouletteWheelSelection: There are chromosomes with null fitness.", actual.Message);
}
public IChromosome Solve(List <Span> spans)
{
var selection = new RouletteWheelSelection();
var crossover = new TwoPointCrossover();
var mutation = new UniformMutation();
var fitness = new Fitness();
var chromosome = new Chromosome(spans);
var population = new Population(100, 100, chromosome);
var ga = new GeneticAlgorithm(population, fitness, selection, crossover, mutation);
ga.Termination = new GenerationNumberTermination(100);
ga.CrossoverProbability = 0.8F;
ga.MutationProbability = 0.1F;
ga.Start();
return(ga.BestChromosome);
}
protected override GeneticAlgorithm CreateGA()
{
var size = (int)Camera.main.orthographicSize - 1;
m_fitness = new TspFitness(m_numberOfCities, -size, size, -size, size);
var chromosome = new TspChromosome(m_numberOfCities);
var crossover = new OrderedCrossover();
var mutation = new ReverseSequenceMutation();
var selection = new RouletteWheelSelection();
var population = new Population(50, 100, chromosome);
var ga = new GeneticAlgorithm(population, m_fitness, selection, crossover, mutation);
ga.Termination = new TimeEvolvingTermination(System.TimeSpan.FromDays(1));
ga.TaskExecutor = new ParallelTaskExecutor
{
MinThreads = 100,
MaxThreads = 200
};
return(ga);
}
