java.lang.Object

io.jenetics.ext.moea.Pareto

public final class Pareto extends Object

Low-level utility methods for doing pareto-optimal calculations. These methods are mostly for users who want to extend the existing MOEA classes.

Since:: 4.1
Version:: 4.1

Method Summary

Modifier and Type

Method

Description

static <T> double[]

crowdingDistance(BaseSeq<? extends Vec<T>> set)

The crowding distance value of a solution provides an estimate of the density of solutions surrounding that solution.

static <T> double[]

crowdingDistance(BaseSeq<? extends T> set, ElementComparator<? super T> comparator, ElementDistance<? super T> distance, ToIntFunction<? super T> dimension)

The crowding distance value of a solution provides an estimate of the density of solutions surrounding that solution.

static int

dominance(double[] u, double[] v)

Calculates the Pareto Dominance of the two vectors u and v.

static int

dominance(int[] u, int[] v)

Calculates the Pareto Dominance of the two vectors u and v.

static int

dominance(long[] u, long[] v)

Calculates the Pareto Dominance of the two vectors u and v.

static <C extends Comparable<? super C>> int

dominance(C[] u, C[] v)

Calculates the Pareto Dominance of the two vectors u and v.

static <T> int

dominance(T[] u, T[] v, Comparator<? super T> comparator)

Calculates the Pareto Dominance of the two vectors u and v.

static <V> int

dominance(V u, V v, int dimensions, ElementComparator<? super V> comparator)

Calculates the Pareto Dominance of the two vectors u and v.

static <T> ISeq<Vec<T>>

front(BaseSeq<? extends Vec<T>> set)

Return the elements, from the given input set, which are part of the pareto front.

static <T> ISeq<T>

front(BaseSeq<? extends T> set, Comparator<? super T> dominance)

Return the elements, from the given input set, which are part of the pareto front.

static <T> int[]

rank(BaseSeq<? extends Vec<T>> set)

Calculates the non-domination rank of the given input set, using the natural order of the elements as dominance measure.

static <T> int[]

rank(BaseSeq<? extends T> set, Comparator<? super T> dominance)

Calculates the non-domination rank of the given input set, using the given dominance comparator.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details
- crowdingDistance
  
  public static <T> double[] crowdingDistance(BaseSeq<? extends Vec<T>> set)
  
  The crowding distance value of a solution provides an estimate of the density of solutions surrounding that solution. The crowding distance value of a particular solution is the average distance of its two neighboring solutions.
  Type Parameters:
  
  T - the vector type
  
  Parameters:
  
  set - the point set used for calculating the crowding distance
  
  Returns:
  
  the crowded distances of the set points
  
  Throws:
  
  NullPointerException - if the input set is null
  
  IllegalArgumentException - if set.get(0).length() < 2
  
  See Also:
  
  crowdingDistance(BaseSeq, ElementComparator, ElementDistance, ToIntFunction)
  
  API Note:
  
  Calculating the crowding distance has a time complexity of O(d*n*log(n)), where d is the number of dimensions and n the set size.
- crowdingDistance
  
  public static <T> double[] crowdingDistance(BaseSeq<? extends T> set, ElementComparator<? super T> comparator, ElementDistance<? super T> distance, ToIntFunction<? super T> dimension)
  
  The crowding distance value of a solution provides an estimate of the density of solutions surrounding that solution. The crowding distance value of a particular solution is the average distance of its two neighboring solutions.
  Type Parameters:
  
  T - the vector type
  
  Parameters:
  
  set - the point set used for calculating the crowding distance
  
  comparator - the comparator which defines the (total) order of the vector elements of T
  
  distance - the distance of two vector elements
  
  dimension - the dimension of vector type T
  
  Returns:
  
  the crowded distances of the set points
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  See Also:
  
  crowdingDistance(BaseSeq)
  
  API Note:
  
  Calculating the crowding distance has a time complexity of O(d*n*log(n)), where d is the number of dimensions and n the set size.
- rank
  
  public static <T> int[] rank(BaseSeq<? extends Vec<T>> set)
  
  Calculates the non-domination rank of the given input set, using the natural order of the elements as dominance measure.
  
  Type Parameters:
  
  T - the element type
  
  Parameters:
  
  set - the input set
  
  Returns:
  
  the non-domination rank of the given input set
  
  API Note:
  
  Calculating the rank has a time complexity of O(n^2, where n the set size.
  Reference: Kalyanmoy Deb, Associate Member, IEEE, Amrit Pratap, Sameer Agarwal, and T. Meyarivan. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II, IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, VOL. 6, NO. 2, APRIL 2002.
- rank
  
  public static <T> int[] rank(BaseSeq<? extends T> set, Comparator<? super T> dominance)
  
  Calculates the non-domination rank of the given input set, using the given dominance comparator.
  
  Type Parameters:
  
  T - the element type
  
  Parameters:
  
  set - the input set
  
  dominance - the dominance comparator used
  
  Returns:
  
  the non-domination rank of the given input set
  
  API Note:
  
  Calculating the rank has a time and space complexity of O(n^2, where n the set size.
  Reference: Kalyanmoy Deb, Associate Member, IEEE, Amrit Pratap, Sameer Agarwal, and T. Meyarivan. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II, IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, VOL. 6, NO. 2, APRIL 2002.
- front
  
  public static <T> ISeq<Vec<T>> front(BaseSeq<? extends Vec<T>> set)
  
  Return the elements, from the given input set, which are part of the pareto front. The Comparable interface defines the dominance measure of the elements, used for calculating the pareto front.
  Reference: E. Zitzler and L. Thiele. Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach, IEEE Transactions on Evolutionary Computation, vol. 3, no. 4, pp. 257-271, 1999.
  
  Type Parameters:
  
  T - the element type
  
  Parameters:
  
  set - the input set
  
  Returns:
  
  the elements which are part of the pareto set
  
  Throws:
  
  NullPointerException - if one of the arguments is null
- front
  
  public static <T> ISeq<T> front(BaseSeq<? extends T> set, Comparator<? super T> dominance)
  
  Return the elements, from the given input set, which are part of the pareto front.
  Reference: E. Zitzler and L. Thiele. Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach, IEEE Transactions on Evolutionary Computation, vol. 3, no. 4, pp. 257-271, 1999.
  
  Type Parameters:
  
  T - the element type
  
  Parameters:
  
  set - the input set
  
  dominance - the dominance comparator used
  
  Returns:
  
  the elements which are part of the pareto set
  
  Throws:
  
  NullPointerException - if one of the arguments is null
- dominance
  
  public static <C extends Comparable<? super C>> int dominance(C[] u, C[] v)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  Type Parameters:
  
  C - the element type of vector u and v
  
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  IllegalArgumentException - if u.length != v.length
  
  See Also:
  
  Vec.dominance(Comparable[], Comparable[])
- dominance
  
  public static <T> int dominance(T[] u, T[] v, Comparator<? super T> comparator)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  Type Parameters:
  
  T - the element type of vector u and v
  
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  comparator - the element comparator which is used for calculating the dominance
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  IllegalArgumentException - if u.length != v.length
  
  See Also:
  
  Vec.dominance(Object[], Object[], Comparator)
- dominance
  
  public static int dominance(int[] u, int[] v)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  IllegalArgumentException - if u.length != v.length
  
  See Also:
  
  Vec.dominance(int[], int[])
- dominance
  
  public static int dominance(long[] u, long[] v)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  IllegalArgumentException - if u.length != v.length
  
  See Also:
  
  Vec.dominance(long[], long[])
- dominance
  
  public static int dominance(double[] u, double[] v)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  IllegalArgumentException - if u.length != v.length
  
  See Also:
  
  Vec.dominance(double[], double[])
- dominance
  
  public static <V> int dominance(V u, V v, int dimensions, ElementComparator<? super V> comparator)
  
  Calculates the Pareto Dominance of the two vectors u and v.
  
  Type Parameters:
  
  V - the vector type
  
  Parameters:
  
  u - the first vector
  
  v - the second vector
  
  dimensions - the number of vector elements
  
  comparator - the comparator used for comparing the vector elements
  
  Returns:
  
  1 if u ≻ v, -1 if v ≻ u and 0 otherwise
  
  Throws:
  
  NullPointerException - if one of the arguments is null
  
  Since:
  
  5.2

Class Pareto

Method Summary

Methods inherited from class java.lang.Object

Method Details

crowdingDistance

crowdingDistance

rank

rank

front

front

dominance

dominance

dominance

dominance

dominance

dominance