Package org.moeaframework.core.indicator

Class PISAHypervolume

java.lang.Object

org.moeaframework.core.indicator.NormalizedIndicator

org.moeaframework.core.indicator.PISAHypervolume

All Implemented Interfaces:

Indicator

public class PISAHypervolume
extends NormalizedIndicator

Hypervolume indicator as implemented by the PISA library. This version is computationally expensive and is being replaced by WFGNormalizedHypervolume.

Field Summary

Fields inherited from class org.moeaframework.core.indicator.NormalizedIndicator

problem

Constructor Summary

Constructors

Constructor	Description
PISAHypervolume(Problem problem, double[] minimum, double[] maximum)	Constructs a hypervolume evaluator for the specified problem using the given minimum and maximum bounds.
PISAHypervolume(Problem problem, NondominatedPopulation referenceSet)	Constructs a hypervolume evaluator for the specified problem and reference set.
PISAHypervolume(Problem problem, NondominatedPopulation referenceSet, double[] referencePoint)	Constructs a hypervolume evaluator for the specified problem using the given reference set and reference point.

Method Summary

Modifier and Type	Method	Description
static double	calculateHypervolume(List<Solution> population, int numberOfSolutions, int numberOfObjectives)	The internal, un-normalized hypervolume calculation.
double	evaluate(NondominatedPopulation approximationSet)	Returns the value of this unary quality indicator given the specified non-dominated population.
protected static void	invert(Problem problem, Solution solution)	Inverts the objective values since this hypervolume algorithm operates on maximization problems.

Methods inherited from class org.moeaframework.core.indicator.NormalizedIndicator

getNormalizedReferenceSet, normalize

Methods inherited from class java.lang.Object

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

Constructor Details

PISAHypervolume

public PISAHypervolume(Problem problem,
 NondominatedPopulation referenceSet)

Constructs a hypervolume evaluator for the specified problem and reference set. See Hypervolume.getNormalizer(Problem, NondominatedPopulation) for details on configuring normalization.

Parameters:

problem - the problem

referenceSet - the reference set

PISAHypervolume

public PISAHypervolume(Problem problem,
 NondominatedPopulation referenceSet,
 double[] referencePoint)

Constructs a hypervolume evaluator for the specified problem using the given reference set and reference point.

Parameters:

problem - the problem

referenceSet - the reference set

referencePoint - the reference point

PISAHypervolume

public PISAHypervolume(Problem problem,
 double[] minimum,
 double[] maximum)

Constructs a hypervolume evaluator for the specified problem using the given minimum and maximum bounds.

Parameters:

problem - the problem

minimum - the minimum bounds of the set

maximum - the maximum bounds of the set

Method Details

invert

protected static void invert(Problem problem,
 Solution solution)

Inverts the objective values since this hypervolume algorithm operates on maximization problems.

Parameters:

problem - the problem

solution - the solution to be inverted

calculateHypervolume

public static double calculateHypervolume(List<Solution> population,
 int numberOfSolutions,
 int numberOfObjectives)

The internal, un-normalized hypervolume calculation. While this method is public, we do not encourage its use since incorrect arguments can cause unexpected behavior. Instead, use the PISAHypervolume(org.moeaframework.core.Problem,org.moeaframework.core.NondominatedPopulation) constructor to create a normalizing version of the hypervolume calculation.

Parameters:

population - the population

numberOfSolutions - the number of solutions

numberOfObjectives - the number of objectives

Returns:

the hypervolume metric

evaluate

public double evaluate(NondominatedPopulation approximationSet)

Description copied from interface: Indicator

Returns the value of this unary quality indicator given the specified non-dominated population.

Parameters:

approximationSet - the non-dominated population to be evaluated

Returns:

the value of this quality indicator given the specified non-dominated population