Package io.jenetics

Class Recombinator<G extends Gene<?,G>,C extends Comparable<? super C>>

All Implemented Interfaces:
Direct Known Subclasses:
CombineAlterer, Crossover

public abstract class Recombinator<G extends Gene<?,G>,C extends Comparable<? super C>> extends AbstractAlterer<G,C>

An enhanced genetic algorithm (EGA) combine elements of existing solutions in order to create a new solution, with some of the properties of each parent. Recombination creates a new chromosome by combining parts of two (or more) parent chromosomes. This combination of chromosomes can be made by selecting one or more crossover points, splitting these chromosomes on the selected points, and merge those portions of different chromosomes to form new ones.

The recombination probability P(r) determines the probability that a given individual (genotype, not gene) of a population is selected for recombination. The (mean) number of changed individuals depend on the concrete implementation and can be vary from P(r)·NG to P(r)·NG·OR, where OR is the order of the recombination, which is the number of individuals involved int the recombine(io.jenetics.util.MSeq<io.jenetics.Phenotype<G, C>>, int[], long) method.

  • Constructor Details

    • Recombinator

      protected Recombinator(double probability, int order)
      Constructs an alterer with a given recombination probability.
      probability - The recombination probability.
      order - the number of individuals involved in the recombine(MSeq, int[], long) step
      IllegalArgumentException - if the probability is not in the valid range of [0, 1] or the given order is smaller than two.
  • Method Details

    • order

      public int order()
      Return the number of individuals involved in the recombine(MSeq, int[], long) step.
      the number of individuals involved in the recombination step.
    • alter

      public final AltererResult<G,C> alter(Seq<Phenotype<G,C>> population, long generation)
      Description copied from interface: Alterer
      Alters (recombine) a given population. If the population is empty, nothing is altered. The altered population is part of the returned AlterResult object.
      population - The Population to be altered. If the population is null or empty, nothing is altered.
      generation - the date of birth (generation) of the altered phenotypes.
      the alter-result object, which contains the altered population and the alteration counts
    • recombine

      protected abstract int recombine(MSeq<Phenotype<G,C>> population, int[] individuals, long generation)
      Recombination template method. This method is called 0 to n times. It is guaranteed that this method is only called by one thread.
      population - the population to recombine
      individuals - the array with the indexes of the individuals which are involved in the recombination step. The length of the array is order(). The first individual is the primary individual.
      generation - the current generation.
      the number of genes that has been altered.