Documentation of 'org.jquantlib.math.optimization.EndCriteria' Java class
EndCriteria
org.jquantlib.math.optimization

Class EndCriteria



  • public class EndCriteria
    extends java.lang.Object
    Criteria to end optimization process

    • maximum number of iterations AND minimum number of iterations around stationary point
    • x (independent variable) stationary point
    • y=f(x) (dependent variable) stationary point
    • stationary gradient
    • Constructor Detail

      • EndCriteria

        public EndCriteria(int maxIterations,
                           int maxStationaryStateIterations,
                           double rootEpsilon,
                           double functionEpsilon,
                           double gradientNormEpsilon)
    • Method Detail

      • checkMaxIterations

        public boolean checkMaxIterations(int iteration,
                                          EndCriteria.Type ecType)
        Test if the number of iteration is below MaxIterations
      • checkStationaryPoint

        public boolean checkStationaryPoint(double xOld,
                                            double xNew,
                                            int statStateIterations,
                                            EndCriteria.Type ecType)
        Test if the root variation is below rootEpsilon
      • checkStationaryFunctionValue

        public boolean checkStationaryFunctionValue(double fxOld,
                                                    double fxNew,
                                                    int statStateIterations,
                                                    EndCriteria.Type ecType)
        Test if the function variation is below functionEpsilon
      • checkStationaryFunctionAccuracy

        public boolean checkStationaryFunctionAccuracy(double f,
                                                       boolean positiveOptimization,
                                                       EndCriteria.Type ecType)
        Test if the function value is below functionEpsilon
      • checkZeroGradientNorm

        public boolean checkZeroGradientNorm(double gradientNorm,
                                             EndCriteria.Type ecType)
        Test if the gradient norm value is below gradientNormEpsilon
      • get

        public boolean get(int iteration,
                           int statStateIterations,
                           boolean positiveOptimization,
                           double fold,
                           double normgold,
                           double fnew,
                           double normgnew,
                           EndCriteria.Type ecType)
        Test if the number of iterations is not too big and if a minimum point is not reached
      • getMaxIterations

        public final int getMaxIterations()
      • getMaxStationaryStateIterations

        public final int getMaxStationaryStateIterations()
      • getRootEpsilon

        public final double getRootEpsilon()
      • getFunctionEpsilon

        public final double getFunctionEpsilon()
      • getGradientNormEpsilon

        public final double getGradientNormEpsilon()

DataMelt 3.0 © DataMelt by jWork.ORG

You see the box below because you did not login.