Package org.artoflife.gp

Art Of Life's Genetic Programming core is an abstract implementation that defines a number of simple concepts.

See:
Description

Interface Summary

PopulationListener	Implement this to receive notification of changes in a Population.
Reproducer	Creates new Individuals using a Selector to choose fit parents from a Population, and an implemented reproduction strategy that may include cloning, mutation or crossover.

Class Summary

Chromosome	An element of an Individual's genotype, has a unique id to ensure compatibility during crossover, and can be mutated.
Fitness	Fitness is a measure of an Individual's fitness (a float) within a Population, with higher values "better" than lower.
Individual	An Individual has a genotype containing a number of Chromosomes, a Fitness used during selection from a Population and a Phenotype resulting from expression of the genotype.
IndividualFitnessComparator	A java.util.Comparator that compares two Individuals based on their Fitness.
Phenotype	The entity that is changed by a genome's Chromosomes during expression.
Population	A Population contains a number of Individuals, has an ideal size that will be maintained during a breeding cycle, can add and remove Individuals and populates new generations using Reproducers.
PopulationEvent	Received by PopulationListeners when a Population changes.
Selector	Used to select Individuals from a Population on the basis of Individuals' Fitnesses.
World	A World contains a number of Populations and can be serialized so that it carries on where it left upon re-loading.

Exception Summary

IncompatibleChromosomeException	Thrown when attempting crossover between Chromosomes that are incompatible.
IncompatibleGenotypeException	Thrown when trying to breed incompatible individuals, or add an incompatible individual to a population.

Package org.artoflife.gp Description

Art Of Life's Genetic Programming core is an abstract implementation that defines a number of simple concepts. This implementation is targetted at the problem of evolving images that will be judged for fitness by a human observer, and aims to be a simple object-oriented approach rather than a highly optimized one.