C# Класс Encog.Neural.Flat.Train.Prop.TrainFlatNetworkResilient

Наследование: TrainFlatNetworkProp

Показать файл Открыть проект Примеры использования класса

Открытые методы

Метод	Описание
InitOthers ( ) : void	Not needed for this training type.
TrainFlatNetworkResilient ( FlatNetwork flat, IMLDataSet trainingSet ) : System	Tran a network using RPROP.
TrainFlatNetworkResilient ( FlatNetwork network, IMLDataSet training, double zeroTolerance, double initialUpdate, double maxStep ) : System	Construct a resilient trainer for flat networks.
UpdateWeight ( double gradients, double lastGradient, int index ) : double	Calculate the amount to change the weight by.
UpdateWeightMinus ( double gradients, double lastGradient, int index ) : double
UpdateWeightPlus ( double gradients, double lastGradient, int index ) : double
UpdateiWeightMinus ( double gradients, double lastGradient, int index ) : double
UpdateiWeightPlus ( double gradients, double lastGradient, int index ) : double

Приватные методы

Метод	Описание
Sign ( double v ) : int	Determine the sign of the value.

Описание методов

InitOthers() публичный Метод

Not needed for this training type.

public InitOthers ( ) : void
Результат	void

TrainFlatNetworkResilient() публичный Метод

Tran a network using RPROP.

public TrainFlatNetworkResilient ( FlatNetwork flat, IMLDataSet trainingSet ) : System
flat	Encog.Neural.Flat.FlatNetwork	The network to train.
trainingSet	IMLDataSet	The training data to use.
Результат	System

TrainFlatNetworkResilient() публичный Метод

Construct a resilient trainer for flat networks.

public TrainFlatNetworkResilient ( FlatNetwork network, IMLDataSet training, double zeroTolerance, double initialUpdate, double maxStep ) : System
network	Encog.Neural.Flat.FlatNetwork	The network to train.
training	IMLDataSet	The training data to use.
zeroTolerance	double	How close a number should be to zero to be counted as zero.
initialUpdate	double	The initial update value.
maxStep	double	The maximum step value.
Результат	System

UpdateWeight() публичный Метод

Calculate the amount to change the weight by.

public UpdateWeight ( double gradients, double lastGradient, int index ) : double
gradients	double	The gradients.
lastGradient	double	The last gradients.
index	int	The index to update.
Результат	double

UpdateWeightMinus() публичный Метод

public UpdateWeightMinus ( double gradients, double lastGradient, int index ) : double
gradients	double
lastGradient	double
index	int
Результат	double

UpdateWeightPlus() публичный Метод

public UpdateWeightPlus ( double gradients, double lastGradient, int index ) : double
gradients	double
lastGradient	double
index	int
Результат	double

UpdateiWeightMinus() публичный Метод

public UpdateiWeightMinus ( double gradients, double lastGradient, int index ) : double
gradients	double
lastGradient	double
index	int
Результат	double

UpdateiWeightPlus() публичный Метод

public UpdateiWeightPlus ( double gradients, double lastGradient, int index ) : double
gradients	double
lastGradient	double
index	int
Результат	double