Free cookie consent management tool by TermsFeed Policy Generator

source: branches/GP-MoveOperators/HeuristicLab.Problems.DataAnalysis.Symbolic/3.4/Analyzers/SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer.cs @ 9666

Last change on this file since 9666 was 7259, checked in by swagner, 13 years ago

Updated year of copyrights to 2012 (#1716)

File size: 7.7 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2012 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.Collections.Generic;
23using System.Linq;
24using HeuristicLab.Common;
25using HeuristicLab.Core;
26using HeuristicLab.Data;
27using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
28using HeuristicLab.Operators;
29using HeuristicLab.Optimization;
30using HeuristicLab.Parameters;
31using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
32using System;
33
34namespace HeuristicLab.Problems.DataAnalysis.Symbolic {
35  /// <summary>
36  /// An operator that analyzes the training best symbolic data analysis solution for multi objective symbolic data analysis problems.
37  /// </summary>
38  [Item("SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer", "An operator that analyzes the training best symbolic data analysis solution for multi objective symbolic data analysis problems.")]
39  [StorableClass]
40  public abstract class SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer<T> : SymbolicDataAnalysisMultiObjectiveAnalyzer
41    where T : class, ISymbolicDataAnalysisSolution {
42    private const string TrainingBestSolutionsParameterName = "Best training solutions";
43    private const string TrainingBestSolutionQualitiesParameterName = "Best training solution qualities";
44
45    #region parameter properties
46    public ILookupParameter<ItemList<T>> TrainingBestSolutionsParameter {
47      get { return (ILookupParameter<ItemList<T>>)Parameters[TrainingBestSolutionsParameterName]; }
48    }
49    public ILookupParameter<ItemList<DoubleArray>> TrainingBestSolutionQualitiesParameter {
50      get { return (ILookupParameter<ItemList<DoubleArray>>)Parameters[TrainingBestSolutionQualitiesParameterName]; }
51    }
52    #endregion
53    #region properties
54    public ItemList<T> TrainingBestSolutions {
55      get { return TrainingBestSolutionsParameter.ActualValue; }
56      set { TrainingBestSolutionsParameter.ActualValue = value; }
57    }
58    public ItemList<DoubleArray> TrainingBestSolutionQualities {
59      get { return TrainingBestSolutionQualitiesParameter.ActualValue; }
60      set { TrainingBestSolutionQualitiesParameter.ActualValue = value; }
61    }
62    #endregion
63
64    [StorableConstructor]
65    protected SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer(bool deserializing) : base(deserializing) { }
66    protected SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer(SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer<T> original, Cloner cloner) : base(original, cloner) { }
67    public SymbolicDataAnalysisMultiObjectiveTrainingBestSolutionAnalyzer()
68      : base() {
69      Parameters.Add(new LookupParameter<ItemList<T>>(TrainingBestSolutionsParameterName, "The training best (Pareto-optimal) symbolic data analysis solutions."));
70      Parameters.Add(new LookupParameter<ItemList<DoubleArray>>(TrainingBestSolutionQualitiesParameterName, "The qualities of the training best (Pareto-optimal) solutions."));
71    }
72
73    public override IOperation Apply() {
74      var results = ResultCollection;
75      // create empty parameter and result values
76      if (TrainingBestSolutions == null) {
77        TrainingBestSolutions = new ItemList<T>();
78        TrainingBestSolutionQualities = new ItemList<DoubleArray>();
79        results.Add(new Result(TrainingBestSolutionQualitiesParameter.Name, TrainingBestSolutionQualitiesParameter.Description, TrainingBestSolutionQualities));
80        results.Add(new Result(TrainingBestSolutionsParameter.Name, TrainingBestSolutionsParameter.Description, TrainingBestSolutions));
81      }
82
83      IList<double[]> trainingBestQualities = TrainingBestSolutionQualities
84        .Select(x => x.ToArray())
85        .ToList();
86
87      #region find best trees
88      IList<int> nonDominatedIndexes = new List<int>();
89      ISymbolicExpressionTree[] tree = SymbolicExpressionTree.ToArray();
90      List<double[]> qualities = Qualities.Select(x => x.ToArray()).ToList();
91      bool[] maximization = Maximization.ToArray();
92      List<double[]> newNonDominatedQualities = new List<double[]>();
93      for (int i = 0; i < tree.Length; i++) {
94        if (IsNonDominated(qualities[i], trainingBestQualities, maximization) &&
95          IsNonDominated(qualities[i], qualities, maximization)) {
96          if (!newNonDominatedQualities.Contains(qualities[i], new DoubleArrayComparer())) {
97            newNonDominatedQualities.Add(qualities[i]);
98            nonDominatedIndexes.Add(i);
99          }
100        }
101      }
102      #endregion
103      #region update Pareto-optimal solution archive
104      if (nonDominatedIndexes.Count > 0) {
105        ItemList<DoubleArray> nonDominatedQualities = new ItemList<DoubleArray>();
106        ItemList<T> nonDominatedSolutions = new ItemList<T>();
107        // add all new non-dominated solutions to the archive
108        foreach (var index in nonDominatedIndexes) {
109          T solution = CreateSolution(tree[index], qualities[index]);
110          nonDominatedSolutions.Add(solution);
111          nonDominatedQualities.Add(new DoubleArray(qualities[index]));
112        }
113        // add old non-dominated solutions only if they are not dominated by one of the new solutions
114        for (int i = 0; i < trainingBestQualities.Count; i++) {
115          if (IsNonDominated(trainingBestQualities[i], newNonDominatedQualities, maximization)) {
116            if (!newNonDominatedQualities.Contains(trainingBestQualities[i], new DoubleArrayComparer())) {
117              nonDominatedSolutions.Add(TrainingBestSolutions[i]);
118              nonDominatedQualities.Add(TrainingBestSolutionQualities[i]);
119            }
120          }
121        }
122
123        results[TrainingBestSolutionsParameter.Name].Value = nonDominatedSolutions;
124        results[TrainingBestSolutionQualitiesParameter.Name].Value = nonDominatedQualities;
125      }
126      #endregion
127      return base.Apply();
128    }
129
130    private class DoubleArrayComparer : IEqualityComparer<double[]> {
131      public bool Equals(double[] x, double[] y) {
132        if (y.Length != x.Length) throw new ArgumentException();
133        for (int i = 0; i < x.Length;i++ ) {
134          if (!x[i].IsAlmost(y[i])) return false;
135        }
136        return true;
137      }
138
139      public int GetHashCode(double[] obj) {
140        int c = obj.Length;
141        for (int i = 0; i < obj.Length; i++)
142          c ^= obj[i].GetHashCode();
143        return c;
144      }
145    }
146
147    protected abstract T CreateSolution(ISymbolicExpressionTree bestTree, double[] bestQuality);
148
149    private bool IsNonDominated(double[] point, IList<double[]> points, bool[] maximization) {
150      foreach (var refPoint in points) {
151        bool refPointDominatesPoint = true;
152        for (int i = 0; i < point.Length; i++) {
153          refPointDominatesPoint &= IsBetterOrEqual(refPoint[i], point[i], maximization[i]);
154        }
155        if (refPointDominatesPoint) return false;
156      }
157      return true;
158    }
159    private bool IsBetterOrEqual(double lhs, double rhs, bool maximization) {
160      if (maximization) return lhs > rhs;
161      else return lhs < rhs;
162    }
163  }
164}
Note: See TracBrowser for help on using the repository browser.