pl.edu.agh.age.compute.ea.variation.mutation

Class StochasticMutate<T,S extends Solution<io.vavr.collection.Array<T>>>

java.lang.Object

pl.edu.agh.age.compute.ea.variation.mutation.StochasticMutate<T,S>

Type Parameters:

R - the representation type of the solution to be mutated

All Implemented Interfaces:

io.vavr.Function1<S,S>, java.util.function.Function<S,S>, Mutation<S>

Direct Known Subclasses:

BinaryMutate, DoubleStochasticMutate, IntegerStochasticMutate

public abstract class StochasticMutate<T,S extends Solution<io.vavr.collection.Array<T>>>
extends java.lang.Object
implements Mutation<S>

Abstract implementation of IMutateSolution. Features are not mutated independently. Instead, a random subset of features is first selected (the size of this subset is proportional to the chance to mutate). Then these features are mutated.

Concrete subclasses are supposed to provide the actual mutation computation.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static interface	io.vavr.Lambda.Memoized

Field Summary

Fields

Modifier and Type	Field and Description
private double	chanceToMutate
private static double	DEFAULT_CHANCE_TO_MUTATE
private DoubleRandomGenerator	doubleRand
private IntRandomGenerator	intRand

Fields inherited from interface io.vavr.Function1

serialVersionUID

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	StochasticMutate(DoubleRandomGenerator doubleRand, IntRandomGenerator intRand)
protected	StochasticMutate(DoubleRandomGenerator doubleRand, IntRandomGenerator intRand, double chanceToMutate)

Method Summary

Modifier and Type	Method and Description
protected abstract io.vavr.collection.Array<T>	doMutate(io.vavr.collection.Array<T> representation, int index) Mutate the representation at the given index.
default boolean	isMemoized()
S	mutate(S solution)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface pl.edu.agh.age.compute.ea.variation.mutation.Mutation

apply

Methods inherited from interface io.vavr.Function1

andThen, arity, compose, constant, curried, identity, lift, liftTry, memoized, narrow, of, partial, reversed, tupled

Field Detail

DEFAULT_CHANCE_TO_MUTATE

private static final double DEFAULT_CHANCE_TO_MUTATE

See Also:

Constant Field Values

doubleRand

private final DoubleRandomGenerator doubleRand

intRand

private final IntRandomGenerator intRand

chanceToMutate

private final double chanceToMutate

Constructor Detail

StochasticMutate

protected StochasticMutate(DoubleRandomGenerator doubleRand,
                           IntRandomGenerator intRand)

StochasticMutate

protected StochasticMutate(DoubleRandomGenerator doubleRand,
                           IntRandomGenerator intRand,
                           double chanceToMutate)

Method Detail

mutate

public final S mutate(S solution)

Specified by:

mutate in interface Mutation<S extends Solution<io.vavr.collection.Array<T>>>

doMutate

protected abstract io.vavr.collection.Array<T> doMutate(io.vavr.collection.Array<T> representation,
                                                        int index)

Mutate the representation at the given index.

This method purpose is to allow efficient unboxing in case of representations of primitives.

Parameters:

representation - the representation to be mutated

index - the index at which mutation should occur

Returns:

the new representation of the solution

isMemoized

public boolean isMemoized()