GeneticAlgorithm

An easy-to-use C# library for implementing a genetic algorithm. What you need to do is create a setting class that overrides GeneticAlgorithmSettings's methods.

public class MySettings : GeneticAlgorithmSettings<T>
{
  public override float MutationRate { get => 0.001f; set => MutationRate = value; }
  public override uint PopulationSize { get => 50; protected set => PopulationSize = value; }
  public override uint NumberOfParentsForCrossover { get => 2; protected set => NumberOfParentsForCrossover = value; }
   
  public override T Crossover(params T[] t) { ... }
  public override float Fitness(T t) { ... }
  public override T Mutate(T t) { ... }
  public override T RandomElement() { ... }
}

T can be any object you want.

Than you can use the settings like this to run the genetic algorithm.

GeneticAlgorithmSettings<T> settings = new MySettings();
GeneticAlgorithm<T> algo = new GeneticAlgorithm<T>(settings);

while(condition)
{
  algo.Step();
}

Console.Write(algo.Fittest());

Docs

GeneticAlgorithmSettings

Attribute	Description
float MutationRate	Set the probability of a mutation. This is the only attribute editable outside the class.
int PopulationSize	Set the initial population size. This attribute is not editable outside the class.
int NumberOfParentsForCrossover	Set the numbers of random parents that a crossovers needs. This attribute is not editable outside the class.

Method	Description
T Crossover(T[] t)	Specify how to perform the crossover. Parameter array t has to be lenght equal to NumberOfParentsForCrossover.
float Fitness(T t)	Specify how to calculate the fitness function for an element t.
T Mutate(T t)	Specify how to perform the mutation of an element t.
T RandomElement()	Specify how to generate a random element t.

GeneticAlgorithm

Method	Description
T Fittest()	Returns the fittest element in the current population.
void Step()	Performs a new generation.