fr.cnes.sirius.patrius.math.optim.univariate

Class SimpleUnivariateValueChecker

java.lang.Object

fr.cnes.sirius.patrius.math.optim.AbstractConvergenceChecker<UnivariatePointValuePair>

fr.cnes.sirius.patrius.math.optim.univariate.SimpleUnivariateValueChecker

All Implemented Interfaces:

ConvergenceChecker<UnivariatePointValuePair>

public class SimpleUnivariateValueChecker
extends AbstractConvergenceChecker<UnivariatePointValuePair>

Simple implementation of the AbstractConvergenceChecker interface that uses only objective function values. Convergence is considered to have been reached if either the relative difference between the objective function values is smaller than a threshold or if either the absolute difference between the objective function values is smaller than another threshold.
The converged method will also return true if the number of iterations has been set (see this constructor).

Since:

3.1

Version:

$Id: SimpleUnivariateValueChecker.java 18108 2017-10-04 06:45:27Z bignon $

Constructor Summary

Constructors

Constructor and Description
SimpleUnivariateValueChecker(double relativeThreshold, double absoluteThreshold) Build an instance with specified thresholds.
SimpleUnivariateValueChecker(double relativeThreshold, double absoluteThreshold, int maxIter) Builds an instance with specified thresholds.

Method Summary

Methods

Modifier and Type	Method and Description
boolean	converged(int iteration, UnivariatePointValuePair previous, UnivariatePointValuePair current) Check if the optimization algorithm has converged considering the last two points.

Methods inherited from class fr.cnes.sirius.patrius.math.optim.AbstractConvergenceChecker

getAbsoluteThreshold, getRelativeThreshold

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SimpleUnivariateValueChecker

public SimpleUnivariateValueChecker(double relativeThreshold,
                            double absoluteThreshold)

Build an instance with specified thresholds. In order to perform only relative checks, the absolute tolerance must be set to a negative value. In order to perform only absolute checks, the relative tolerance must be set to a negative value.

Parameters:

relativeThreshold - relative tolerance threshold

absoluteThreshold - absolute tolerance threshold

SimpleUnivariateValueChecker

public SimpleUnivariateValueChecker(double relativeThreshold,
                            double absoluteThreshold,
                            int maxIter)

Builds an instance with specified thresholds. In order to perform only relative checks, the absolute tolerance must be set to a negative value. In order to perform only absolute checks, the relative tolerance must be set to a negative value.

Parameters:

relativeThreshold - relative tolerance threshold

absoluteThreshold - absolute tolerance threshold

maxIter - Maximum iteration count.

Throws:

NotStrictlyPositiveException - if maxIter <= 0.

Since:

3.1

Method Detail

converged

public boolean converged(int iteration,
                UnivariatePointValuePair previous,
                UnivariatePointValuePair current)

Check if the optimization algorithm has converged considering the last two points. This method may be called several time from the same algorithm iteration with different points. This can be detected by checking the iteration number at each call if needed. Each time this method is called, the previous and current point correspond to points with the same role at each iteration, so they can be compared. As an example, simplex-based algorithms call this method for all points of the simplex, not only for the best or worst ones.

Specified by:

converged in interface ConvergenceChecker<UnivariatePointValuePair>

Specified by:

converged in class AbstractConvergenceChecker<UnivariatePointValuePair>

Parameters:

iteration - Index of current iteration

previous - Best point in the previous iteration.

current - Best point in the current iteration.

Returns:

true if the algorithm has converged.