Package org.hipparchus.optim

Class AbstractConvergenceChecker<PAIR>

java.lang.Object

org.hipparchus.optim.AbstractConvergenceChecker<PAIR>

Type Parameters:

PAIR - Type of (point, value) pair.

All Implemented Interfaces:

ConvergenceChecker<PAIR>

Direct Known Subclasses:

SimplePointChecker, SimpleUnivariateValueChecker, SimpleValueChecker, SimpleVectorValueChecker

public abstract class AbstractConvergenceChecker<PAIR>
extends Object
implements ConvergenceChecker<PAIR>

Base class for all convergence checker implementations.

Constructor Summary

Constructors

Constructor	Description
AbstractConvergenceChecker(double relativeThreshold, double absoluteThreshold)	Build an instance with a specified thresholds.

Method Summary

Modifier and Type	Method	Description
abstract boolean	converged(int iteration, PAIR previous, PAIR current)	Check if the optimization algorithm has converged.
double	getAbsoluteThreshold()
double	getRelativeThreshold()

Methods inherited from class java.lang.Object

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

Constructor Detail

AbstractConvergenceChecker

public AbstractConvergenceChecker(double relativeThreshold,
                                  double absoluteThreshold)

Build an instance with a specified thresholds.

Parameters:

relativeThreshold - relative tolerance threshold

absoluteThreshold - absolute tolerance threshold

Method Detail

getRelativeThreshold

public double getRelativeThreshold()

Returns:

the relative threshold.

getAbsoluteThreshold

public double getAbsoluteThreshold()

Returns:

the absolute threshold.

converged

public abstract boolean converged(int iteration,
                                  PAIR previous,
                                  PAIR current)

Check if the optimization algorithm has converged.

Specified by:

converged in interface ConvergenceChecker<PAIR>

Parameters:

iteration - Current iteration.

previous - Best point in the previous iteration.

current - Best point in the current iteration.

Returns:

true if the algorithm is considered to have converged.