I
- Type of individuals to mutatepublic class ParametricSRMutator<I extends NeuralNetIndividual> extends ParametricMutator<I>
Parametric mutator for neural nets, mutate the weights of the neural nets mutated. This implementation uses the "1/5 Success Rule of Rechenberg" method to update alpha values. IMPORTANT NOTE: Parametric mutator works directly with he individuals instead of returning a mutated copy of them. This is for performance reasons. If you want to use another mutator you have to consider that individuals will be changed when you use parametric mutation
Modifier and Type | Field and Description |
---|---|
protected double |
successRatio
Ratio of successful mutations
|
alphaInput, alphaOutput, amplitude, fitDif, initialAlphaInput, initialAlphaOutput, neuronParametricMutators, selective, temperExponent
species
Constructor and Description |
---|
ParametricSRMutator()
Empty Constructor
|
Modifier and Type | Method and Description |
---|---|
protected void |
alphaControlParametersUpdate(double newFitness,
double fitness)
Updates alpha control parameters at the end of each
neuron mutation, if neccesary
|
void |
alphaInit()
Init the values of alpha parameters used in the mutations
|
protected void |
alphaUpdate(double bestFitness)
Updates the values of alpha parameters used in the mutations
|
double |
getSuccessRatio()
Returns the success ratio of last generation
|
configure, getAlphaInput, getAlphaOutput, getAmplitude, getFitDif, getInitialAlphaInput, getInitialAlphaOutput, getTemperExponent, isSelective, mutateNext, prepareMutation, setAmplitude, setFitDif, setInitialAlphaInput, setInitialAlphaOutput, setSelective, setTemperExponent
public double getSuccessRatio()
Returns the success ratio of last generation
public void alphaInit()
Init the values of alpha parameters used in the mutations
alphaInit
in class ParametricMutator<I extends NeuralNetIndividual>
protected void alphaUpdate(double bestFitness)
Updates the values of alpha parameters used in the mutations
alphaUpdate
in class ParametricMutator<I extends NeuralNetIndividual>
bestFitness
- Best fitness of this generationprotected void alphaControlParametersUpdate(double newFitness, double fitness)
Updates alpha control parameters at the end of each neuron mutation, if neccesary
alphaControlParametersUpdate
in class ParametricMutator<I extends NeuralNetIndividual>
newFitness
- Result fitness of the mutationfitness
- Previous fitness befor making the mutation