org.encog.neural.neat.training.opp
Class NEATMutateWeights
java.lang.Object
org.encog.neural.neat.training.opp.NEATMutation
org.encog.neural.neat.training.opp.NEATMutateWeights
- All Implemented Interfaces:
- EvolutionaryOperator
public class NEATMutateWeights
- extends NEATMutation
Mutate the weights of a genome. A method is select the links for mutation.
Another method should also be provided for the actual mutation.
-----------------------------------------------------------------------------
http://www.cs.ucf.edu/~kstanley/ Encog's NEAT implementation was drawn from
the following three Journal Articles. For more complete BibTeX sources, see
NEATNetwork.java.
Evolving Neural Networks Through Augmenting Topologies
Generating Large-Scale Neural Networks Through Discovering Geometric
Regularities
Automatic feature selection in neuroevolution
Methods inherited from class org.encog.neural.neat.training.opp.NEATMutation |
chooseRandomNeuron, createLink, getElementPos, getOwner, init, isDuplicateLink, isNeuronNeeded, obtainGenome, offspringProduced, parentsNeeded, removeNeuron |
NEATMutateWeights
public NEATMutateWeights(SelectLinks theLinkSelection,
MutateLinkWeight theWeightMutation)
- Construct a weight mutation operator.
- Parameters:
theLinkSelection
- The method used to choose the links for mutation.theWeightMutation
- The method used to actually mutate the weights.
getLinkSelection
public SelectLinks getLinkSelection()
- Returns:
- The method used to select links for mutation.
getWeightMutation
public MutateLinkWeight getWeightMutation()
- Returns:
- The method used to mutate the weights.
performOperation
public void performOperation(Random rnd,
Genome[] parents,
int parentIndex,
Genome[] offspring,
int offspringIndex)
- Perform the evolutionary operation.
- Parameters:
rnd
- A random number generator.parents
- The parents.parentIndex
- The index into the parents array.offspring
- The offspring.offspringIndex
- An index into the offspring array.
toString
public String toString()
-
- Overrides:
toString
in class Object
Copyright © 2014. All Rights Reserved.