Package org.biojavax.ga.functions

GA functions A genetic algorithm requires a number of functions.

See: Description

Interface Summary

Interface	Description
CrossOverFunction	Crosses two chromosomes.
FitnessFunction	Calculates the fitness of an Organism in a Population of Organisms
GACross	Holds the results of a CrossOver event, objects of this type are made by CrossOverFunctions
GACrossResult	Holds the results of a CrossOver event, objects of this type are made by CrossOverFunctions
MutationFunction	A class that mutates a SymbolList
SelectionFunction	Selects Organisms for Replication and returns the offspring.

Class Summary

Class	Description
AbstractCrossOverFunction	Abstract implementation of CrossOverFunction.
AbstractMutationFunction	Abstract implementation of MutationFunction all custom implementations should inherit from here.
CrossOverFunction.NoCross	A place holder CrossOverFunction that doesn't perform cross overs
MutationFunction.NoMutation	Place Holder class that doesn't mutate its SymbolLists
OrderCrossover	This does a 2-point-crossover on two chromosomes keeping the Symbols in each chromosome constant.
ProportionalSelection	A Selection function that determines the proportion of individuals in a new population proportionally to their fitness.
SelectionFunction.SelectAll
SelectionFunction.Threshold	Selects individuals who's fitness exceeds a threshold value.
SimpleCrossOverFunction	Simple Implementation of the CrossOverFunction interface
SimpleGACrossResult	Simple implementation of the GACross interface.
SimpleMutationFunction	Simple no frills Implementation of the MutationFunction interface This class is final, custom implementations should extend AbstractMutationFunction
SwapMutationFunction	This class does a sort of mutation by exchanging two positions on the chromosome.
TournamentSelection	Tournament Selection chooses the best organisms from n random subsets of a given population.

Package org.biojavax.ga.functions Description

GA functions

A genetic algorithm requires a number of functions. This package provides the interfaces for those functions and simple implementations. By implementing, mixing and matching these functions you can create highly customized genetic algorithms.

A GA requires (in alphabetical order): a CrossOverFunction to govern the behaivour of 'chromosome' crossovers, a FitnessFunction to determine the fitness of each organism after each iteration, a MutationFuntion to govern mutation behaivour, and a SelectionFunction to select organisms for the next round of replication.