C# Class Encog.Neural.Networks.Training.PNN.GlobalMinimumSearch

Mostrar archivo Open project: encog/encog-silverlight-core Class Usage Examples

Public Methods

Method Description
Brentmin ( int maxIterations, double maxError, double eps, double tol, ICalculationCriteria network, double y ) : double

Use the "Brent Method" to find a better minimum.

FindBestRange ( double low, double high, int numberOfPoints, bool useLog, double minError, ICalculationCriteria network ) : void

Find the best common gamma. Use the same gamma for all kernels. This is a crude brute-force search. The range found should be refined using the "Brent Method", also provided in this class.

Method Details

Brentmin() public method

Use the "Brent Method" to find a better minimum.
public Brentmin ( int maxIterations, double maxError, double eps, double tol, ICalculationCriteria network, double y ) : double
maxIterations int THe maximum number of iterations.
maxError double We can stop if we reach this error.
eps double The approximate machine precision.
tol double Brent's tolerance, must be >= sqrt( eps )
network ICalculationCriteria The network to obtain the error from.
y double The error at x2.
return double

FindBestRange() public method

Find the best common gamma. Use the same gamma for all kernels. This is a crude brute-force search. The range found should be refined using the "Brent Method", also provided in this class.
public FindBestRange ( double low, double high, int numberOfPoints, bool useLog, double minError, ICalculationCriteria network ) : void
low double The low gamma to begin the search with.
high double The high gamma to end the search with.
numberOfPoints int If you do set this to negative, set x2 and y2 to the correct values.
useLog bool Should we progress "logarithmically" from low to high.
minError double We are done if the error is below this.
network ICalculationCriteria The network to evaluate.
return void