Free cookie consent management tool by TermsFeed Policy Generator

source: branches/2708_ScopedAlgorithms/HeuristicLab.Encodings.BinaryVectorEncoding/3.3/Crossovers/NPointCrossover.cs @ 16123

Last change on this file since 16123 was 14429, checked in by abeham, 8 years ago

#2701, #2708: Made a new branch from ProblemRefactoring and removed ScopedBasicAlgorithm branch (which becomes MemPR branch)

File size: 7.2 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2015 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 *
5 * This file is part of HeuristicLab.
6 *
7 * HeuristicLab is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
11 *
12 * HeuristicLab is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License
18 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
19 */
20#endregion
21
22using System;
23using System.Collections.Generic;
24using HeuristicLab.Common;
25using HeuristicLab.Core;
26using HeuristicLab.Data;
27using HeuristicLab.Optimization;
28using HeuristicLab.Parameters;
29using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
30
31namespace HeuristicLab.Encodings.BinaryVectorEncoding {
32  /// <summary>
33  /// N point crossover for binary vectors.
34  /// </summary>
35  /// <remarks>
36  /// It is implemented as described in Eiben, A.E. and Smith, J.E. 2003. Introduction to Evolutionary Computation. Natural Computing Series, Springer-Verlag Berlin Heidelberg..
37  /// </remarks>
38  [Item("NPointCrossover", "N point crossover for binary vectors. It is implemented as described in Eiben, A.E. and Smith, J.E. 2003. Introduction to Evolutionary Computation. Natural Computing Series, Springer-Verlag Berlin Heidelberg.")]
39  [StorableClass]
40  public sealed class NPointCrossover : BinaryVectorCrossover {
41    /// <summary>
42    /// Number of crossover points.
43    /// </summary>
44    public IValueLookupParameter<IntValue> NParameter {
45      get { return (IValueLookupParameter<IntValue>)Parameters["N"]; }
46    }
47
48    [StorableConstructor]
49    private NPointCrossover(bool deserializing) : base(deserializing) { }
50    private NPointCrossover(NPointCrossover original, Cloner cloner) : base(original, cloner) { }
51    /// <summary>
52    /// Initializes a new instance of <see cref="NPointCrossover"/>
53    /// </summary>
54    public NPointCrossover()
55      : base() {
56      Parameters.Add(new ValueLookupParameter<IntValue>("N", "Number of crossover points", new IntValue(2)));
57    }
58
59    public override IDeepCloneable Clone(Cloner cloner) {
60      return new NPointCrossover(this, cloner);
61    }
62
63    /// <summary>
64    /// Performs a N point crossover at randomly chosen positions of the two
65    /// given parent binary vectors.
66    /// </summary>
67    /// <exception cref="ArgumentException">Thrown when the value for N is invalid or when the parent vectors are of different length.</exception>
68    /// <param name="random">A random number generator.</param>
69    /// <param name="parent1">The first parent for crossover.</param>
70    /// <param name="parent2">The second parent for crossover.</param>
71    /// <param name="n">Number of crossover points.</param>
72    /// <returns>The newly created binary vector, resulting from the N point crossover.</returns>
73    public static BinaryVector Apply(IRandom random, BinaryVector parent1, BinaryVector parent2, int n) {
74      if (parent1.Length != parent2.Length)
75        throw new ArgumentException("NPointCrossover: The parents are of different length.");
76
77      if (n > parent1.Length)
78        throw new ArgumentException("NPointCrossover: There cannot be more breakpoints than the size of the parents.");
79
80      if (n < 1)
81        throw new ArgumentException("NPointCrossover: N cannot be < 1.");
82
83      int length = parent1.Length;
84      bool[] result = new bool[length];
85      int[] breakpoints = new int[n];
86
87      //choose break points
88      List<int> breakpointPool = new List<int>();
89
90      for (int i = 0; i < length; i++)
91        breakpointPool.Add(i);
92
93      for (int i = 0; i < n; i++) {
94        int index = random.Next(breakpointPool.Count);
95        breakpoints[i] = breakpointPool[index];
96        breakpointPool.RemoveAt(index);
97      }
98
99      Array.Sort(breakpoints);
100
101      //perform crossover
102      int arrayIndex = 0;
103      int breakPointIndex = 0;
104      bool firstParent = true;
105
106      while (arrayIndex < length) {
107        if (breakPointIndex < breakpoints.Length &&
108          arrayIndex == breakpoints[breakPointIndex]) {
109          breakPointIndex++;
110          firstParent = !firstParent;
111        }
112
113        if (firstParent)
114          result[arrayIndex] = parent1[arrayIndex];
115        else
116          result[arrayIndex] = parent2[arrayIndex];
117
118        arrayIndex++;
119      }
120
121      return new BinaryVector(result);
122    }
123
124    /// <summary>
125    /// Performs a N point crossover at a randomly chosen position of two
126    /// given parent binary vectors.
127    /// </summary>
128    /// <exception cref="ArgumentException">Thrown if there are not exactly two parents.</exception>
129    /// <exception cref="InvalidOperationException">
130    /// Thrown when the N parameter could not be found.</description></item>
131    /// </exception>
132    /// <param name="random">A random number generator.</param>
133    /// <param name="parents">An array containing the two binary vectors that should be crossed.</param>
134    /// <returns>The newly created binary vector, resulting from the N point crossover.</returns>
135    protected override BinaryVector Cross(IRandom random, ItemArray<BinaryVector> parents) {
136      if (parents.Length != 2) throw new ArgumentException("ERROR in NPointCrossover: The number of parents is not equal to 2");
137
138      if (NParameter.ActualValue == null) throw new InvalidOperationException("NPointCrossover: Parameter " + NParameter.ActualName + " could not be found.");
139
140      return Apply(random, parents[0], parents[1], NParameter.ActualValue.Value);
141    }
142  }
143
144  [Item("N-point Crossover", "", ExcludeGenericTypeInfo = true)]
145  [StorableClass]
146  public sealed class NPointCrossover<TContext> : ParameterizedNamedItem, IBinaryCrossover<TContext>
147    where TContext : IMatingContext<BinaryVector>, IStochasticContext {
148
149    [Storable]
150    private IValueParameter<IntValue> nparameter;
151    public int N {
152      get { return nparameter.Value.Value; }
153      set {
154        if (value < 1) throw new ArgumentException("Cannot set N to less than 1.");
155        nparameter.Value.Value = value;
156      }
157    }
158
159    [StorableConstructor]
160    private NPointCrossover(bool deserializing) : base(deserializing) { }
161    private NPointCrossover(NPointCrossover<TContext> original, Cloner cloner)
162      : base(original, cloner) {
163      nparameter = cloner.Clone(original.nparameter);
164    }
165    public NPointCrossover() {
166      Parameters.Add(nparameter = new ValueParameter<IntValue>("N", "The number of crossover points.", new IntValue(1)));
167    }
168
169
170    public override IDeepCloneable Clone(Cloner cloner) {
171      return new NPointCrossover<TContext>(this, cloner);
172    }
173
174    public void Cross(TContext context) {
175      context.Child.Solution = NPointCrossover.Apply(context.Random, context.Parents.Item1.Solution, context.Parents.Item2.Solution, N);
176    }
177  }
178}
Note: See TracBrowser for help on using the repository browser.