C# Class Encog.MathUtil.Error.ErrorCalculation

Calculate the error of a neural network. Encog currently supports three error calculation modes. See ErrorCalculationMode for more info.
Afficher le fichier Open project: encog/encog-silverlight-core Class Usage Examples

Méthodes publiques

Méthode Description
Calculate ( ) : double

Returns the root mean square error for a complete training set.
CalculateMSE ( ) : double

Calculate the error with MSE.
CalculateRMS ( ) : double

Calculate the error with RMS.
Reset ( ) : void

Reset the error accumulation to zero.
UpdateError ( double actual, double ideal ) : void

Update the error with single values.
UpdateError ( double actual, double ideal, double significance ) : void

Called to update for each number that should be checked.

Method Details

Calculate() public méthode

Returns the root mean square error for a complete training set.
public Calculate ( ) : double
Résultat double

CalculateMSE() public méthode

Calculate the error with MSE.
public CalculateMSE ( ) : double
Résultat double

CalculateRMS() public méthode

Calculate the error with RMS.
public CalculateRMS ( ) : double
Résultat double

Reset() public méthode

Reset the error accumulation to zero.
public Reset ( ) : void
Résultat void

UpdateError() public méthode

Update the error with single values.
public UpdateError ( double actual, double ideal ) : void
actual double The actual value.
ideal double The ideal value.
Résultat void

UpdateError() public méthode

Called to update for each number that should be checked.
public UpdateError ( double actual, double ideal, double significance ) : void
actual double The actual number.
ideal double The ideal number.
significance double The significance of this error, 1.0 is the baseline.
Résultat void