Class GeneticAlgorithm
java.lang.Object
org.apache.commons.math3.genetics.GeneticAlgorithm
Implementation of a genetic algorithm. All factors that govern the operation
of the algorithm can be configured for a specific problem.
- Since:
- 2.0
-
Constructor Summary
ConstructorsConstructorDescriptionGeneticAlgorithm(CrossoverPolicy crossoverPolicy, double crossoverRate, MutationPolicy mutationPolicy, double mutationRate, SelectionPolicy selectionPolicy) Create a new genetic algorithm. -
Method Summary
Modifier and TypeMethodDescriptionevolve(Population initial, StoppingCondition condition) Evolve the given population.Returns the crossover policy.doubleReturns the crossover rate.intReturns the number of generations evolved to reachStoppingConditionin the last run.Returns the mutation policy.doubleReturns the mutation rate.static RandomGeneratorReturns the (static) random generator.Returns the selection policy.nextGeneration(Population current) Evolve the given population into the next generation.static voidsetRandomGenerator(RandomGenerator random) Set the (static) random generator.
-
Constructor Details
-
GeneticAlgorithm
public GeneticAlgorithm(CrossoverPolicy crossoverPolicy, double crossoverRate, MutationPolicy mutationPolicy, double mutationRate, SelectionPolicy selectionPolicy) throws OutOfRangeException Create a new genetic algorithm.- Parameters:
crossoverPolicy- TheCrossoverPolicycrossoverRate- The crossover rate as a percentage (0-1 inclusive)mutationPolicy- TheMutationPolicymutationRate- The mutation rate as a percentage (0-1 inclusive)selectionPolicy- TheSelectionPolicy- Throws:
OutOfRangeException- if the crossover or mutation rate is outside the [0, 1] range
-
-
Method Details
-
setRandomGenerator
Set the (static) random generator.- Parameters:
random- random generator
-
getRandomGenerator
Returns the (static) random generator.- Returns:
- the static random generator shared by GA implementation classes
-
evolve
Evolve the given population. Evolution stops when the stopping condition is satisfied. Updates thegenerationsEvolvedproperty with the number of generations evolved before the StoppingCondition is satisfied.- Parameters:
initial- the initial, seed population.condition- the stopping condition used to stop evolution.- Returns:
- the population that satisfies the stopping condition.
-
nextGeneration
Evolve the given population into the next generation.- Get nextGeneration population to fill from
currentgeneration, using its nextGeneration method - Loop until new generation is filled:
- Apply configured SelectionPolicy to select a pair of parents
from
current - With probability =
getCrossoverRate(), apply configuredCrossoverPolicyto parents - With probability =
getMutationRate(), apply configuredMutationPolicyto each of the offspring - Add offspring individually to nextGeneration, space permitting
- Return nextGeneration
- Parameters:
current- the current population.- Returns:
- the population for the next generation.
- Get nextGeneration population to fill from
-
getCrossoverPolicy
Returns the crossover policy.- Returns:
- crossover policy
-
getCrossoverRate
public double getCrossoverRate()Returns the crossover rate.- Returns:
- crossover rate
-
getMutationPolicy
Returns the mutation policy.- Returns:
- mutation policy
-
getMutationRate
public double getMutationRate()Returns the mutation rate.- Returns:
- mutation rate
-
getSelectionPolicy
Returns the selection policy.- Returns:
- selection policy
-
getGenerationsEvolved
public int getGenerationsEvolved()Returns the number of generations evolved to reachStoppingConditionin the last run.- Returns:
- number of generations evolved
- Since:
- 2.1
-