C# Класс Encog.Neural.NEAT.Training.NEATTraining

Implements NEAT genetic training. NeuroEvolution of Augmenting Topologies (NEAT) is a genetic algorithm for the generation of evolving artificial neural networks. It was developed by Ken Stanley while at The University of Texas at Austin. http://www.cs.ucf.edu/~kstanley/

Наследование: Encog.ML.Genetic.GeneticAlgorithm, IMLTrain

Показать файл Открыть проект Примеры использования класса

Открытые методы

Метод	Описание
AddNeuronID ( long nodeID, IList vec ) : void	Add the specified neuron id.
AddStrategy ( IStrategy strategy ) : void
AdjustCompatibilityThreshold ( ) : void	Adjust the compatibility threshold.
AdjustSpeciesScore ( ) : void	Adjust each species score.
Crossover ( NEATGenome mom, NEATGenome dad ) : NEATGenome	Perform a cross over.
FinishTraining ( ) : void
Iteration ( ) : void	Perform one training iteration.
Iteration ( int count ) : void
NEATTraining ( ICalculateScore calculateScore, IPopulation population ) : System	Construct neat training with an existing population.
NEATTraining ( ICalculateScore calculateScore, int inputCount, int outputCount, int populationSize ) : System	Construct a neat trainer with a new population. The new population is created from the specified parameters.
Pause ( ) : TrainingContinuation
ResetAndKill ( ) : void	Reset counts and kill genomes with worse scores.
Resume ( TrainingContinuation state ) : void
SortAndRecord ( ) : void	Sort the genomes.
SpeciateAndCalculateSpawnLevels ( ) : void	Determine the species.
TournamentSelection ( int numComparisons ) : NEATGenome	Select a gene using a tournament.

Приватные методы

Метод	Описание
Init ( ) : void	Init the training.

Описание методов

AddNeuronID() публичный Метод

Add the specified neuron id.

public AddNeuronID ( long nodeID, IList vec ) : void
nodeID	long	The neuron to add.
vec	IList	The list to add to.
Результат	void

AddStrategy() публичный Метод

public AddStrategy ( IStrategy strategy ) : void
strategy	IStrategy
Результат	void

AdjustCompatibilityThreshold() публичный Метод

Adjust the compatibility threshold.

public AdjustCompatibilityThreshold ( ) : void
Результат	void

AdjustSpeciesScore() публичный Метод

Adjust each species score.

public AdjustSpeciesScore ( ) : void
Результат	void

Crossover() публичный Метод

Perform a cross over.

public Crossover ( NEATGenome mom, NEATGenome dad ) : NEATGenome
mom	NEATGenome	The mother genome.
dad	NEATGenome	The father genome.
Результат	NEATGenome

FinishTraining() публичный Метод

public FinishTraining ( ) : void
Результат	void

Iteration() публичный Метод

Perform one training iteration.

public Iteration ( ) : void
Результат	void

Iteration() публичный Метод

public Iteration ( int count ) : void
count	int
Результат	void

NEATTraining() публичный Метод

Construct neat training with an existing population.

public NEATTraining ( ICalculateScore calculateScore, IPopulation population ) : System
calculateScore	ICalculateScore	The score object to use.
population	IPopulation	The population to use.
Результат	System

NEATTraining() публичный Метод

Construct a neat trainer with a new population. The new population is created from the specified parameters.

public NEATTraining ( ICalculateScore calculateScore, int inputCount, int outputCount, int populationSize ) : System
calculateScore	ICalculateScore	The score calculation object.
inputCount	int	The input neuron count.
outputCount	int	The output neuron count.
populationSize	int	The population size.
Результат	System

Pause() публичный Метод

public Pause ( ) : TrainingContinuation
Результат	Encog.Neural.Networks.Training.Propagation.TrainingContinuation

ResetAndKill() публичный Метод

Reset counts and kill genomes with worse scores.

public ResetAndKill ( ) : void
Результат	void

Resume() публичный Метод

public Resume ( TrainingContinuation state ) : void
state	Encog.Neural.Networks.Training.Propagation.TrainingContinuation
Результат	void

SortAndRecord() публичный Метод

Sort the genomes.

public SortAndRecord ( ) : void
Результат	void

SpeciateAndCalculateSpawnLevels() публичный Метод

Determine the species.

public SpeciateAndCalculateSpawnLevels ( ) : void
Результат	void

TournamentSelection() публичный Метод

Select a gene using a tournament.

public TournamentSelection ( int numComparisons ) : NEATGenome
numComparisons	int	The number of compares to do.
Результат	NEATGenome