C# Class Encog.Engine.Network.Train.Prop.TrainFlatNetworkOpenCL

Train a flat network using OpenCL.
Inheritance: ITrainFlatNetwork
Afficher le fichier Open project: encog/encog-silverlight-core Class Usage Examples

Méthodes publiques

Méthode Description
FinishTraining ( ) : void
Iteration ( ) : void

Perform a single iteration.
Iteration ( int iterations ) : void
LearnBPROP ( double learningRate, double momentum ) : void

Learn using backpropagation.
LearnManhattan ( double learningRate ) : void

Learn using the Manhattan update rule.
LearnRPROP ( ) : void

Learn using RPROP. Use default max step and initial update.
LearnRPROP ( double initialUpdate, double maxStep ) : void

Learn using RPROP with a custom initial update and max step.
TrainFlatNetworkOpenCL ( FlatNetwork network, IEngineDataSet training, OpenCLTrainingProfile profile ) : Encog.Engine

Train a flat network multithreaded.

Private Methods

Méthode Description
CallKernel ( int start, int size, bool learn, int iterations ) : void

Call the kernel.
GetOptions ( String learningType ) : String>.IDictionary

Get the learning properties.

Method Details

FinishTraining() public méthode

public FinishTraining ( ) : void
Résultat void

Iteration() public méthode

Perform a single iteration.
public Iteration ( ) : void
Résultat void

Iteration() public méthode

public Iteration ( int iterations ) : void
iterations int
Résultat void

LearnBPROP() public méthode

Learn using backpropagation.
public LearnBPROP ( double learningRate, double momentum ) : void
learningRate double The learning rate.
momentum double The momentum.
Résultat void

LearnManhattan() public méthode

Learn using the Manhattan update rule.
public LearnManhattan ( double learningRate ) : void
learningRate double The learning rate.
Résultat void

LearnRPROP() public méthode

Learn using RPROP. Use default max step and initial update.
public LearnRPROP ( ) : void
Résultat void

LearnRPROP() public méthode

Learn using RPROP with a custom initial update and max step.
public LearnRPROP ( double initialUpdate, double maxStep ) : void
initialUpdate double The initial update value.
maxStep double The max step.
Résultat void

TrainFlatNetworkOpenCL() public méthode

Train a flat network multithreaded.
public TrainFlatNetworkOpenCL ( FlatNetwork network, IEngineDataSet training, OpenCLTrainingProfile profile ) : Encog.Engine
network Encog.Engine.Network.Flat.FlatNetwork The network to train.
training IEngineDataSet The training data to use.
profile OpenCLTrainingProfile The OpenCL training profile.
Résultat Encog.Engine