Class LinearCombinationLearningCurve

  • All Implemented Interfaces:
    AnalyticalLearningCurve, LearningCurve

    public class LinearCombinationLearningCurve
    extends java.lang.Object
    implements AnalyticalLearningCurve
    The LinearCombinationLearningCurve consists of the actual linear combination function that describes the learning curve, as well as the derivative of this function. The derivative is used in order to calculate the saturation point.
    • Method Detail

      • getCurveValue

        public double getCurveValue​(double x)
        Description copied from interface: LearningCurve
        Calculates or looks-up the curves value at a given point.
        Specified by:
        getCurveValue in interface LearningCurve
        Parameters:
        x - The x value of the point.
        Returns:
        The y value of the curve at the given x value.
      • getSaturationPoint

        public double getSaturationPoint​(double epsilon)
        Description copied from interface: AnalyticalLearningCurve
        Calculated or search a saturation point with a tolerance of epsilon.
        Specified by:
        getSaturationPoint in interface AnalyticalLearningCurve
        Parameters:
        epsilon - Epsilon value(must be > 0) that is a tolerance the saturation point can deviate.
        Returns:
        Saturation point with tolerated deviation.
      • getDerivativeCurveValue

        public double getDerivativeCurveValue​(double x)
        Description copied from interface: AnalyticalLearningCurve
        Calculates or looks-up the value of the derivative of the learning point at a given point.
        Specified by:
        getDerivativeCurveValue in interface AnalyticalLearningCurve
        Parameters:
        x - The x value of the point.
        Returns:
        The y value of the derivative at the given x value.
      • getConvergenceValue

        public double getConvergenceValue()
        Description copied from interface: AnalyticalLearningCurve
        Calculates or looks-up the value the learning curve converges to. In other words, the convergence value is the value that would be achieved if sufficiently much data would be available.
        Specified by:
        getConvergenceValue in interface AnalyticalLearningCurve
        Returns: