Free cookie consent management tool by TermsFeed Policy Generator

source: branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/WeightCalculators/NeighbourhoodWeightCalculator.cs @ 7531

Last change on this file since 7531 was 7531, checked in by sforsten, 12 years ago

#1776:

  • 2 more strategies have been implemented
  • major changes in the inheritance have been made to make it possible to add strategies which don't use a voting strategy with weights
  • ClassificationEnsembleSolutionEstimatedClassValuesView doesn't currently show the confidence (has been removed for test purpose)
File size: 3.7 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;
23using System.Collections.Generic;
24using System.Linq;
25using HeuristicLab.Common;
26using HeuristicLab.Core;
27using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
28
29namespace HeuristicLab.Problems.DataAnalysis {
30  /// <summary>
31  ///
32  /// </summary>
33  [StorableClass]
34  [Item("NeighbourhoodWeightCalculator", "")]
35  public class NeighbourhoodWeightCalculator : DiscriminantClassificationWeightCalculator {
36
37    public NeighbourhoodWeightCalculator()
38      : base() {
39    }
40
41    [StorableConstructor]
42    protected NeighbourhoodWeightCalculator(bool deserializing) : base(deserializing) { }
43    protected NeighbourhoodWeightCalculator(NeighbourhoodWeightCalculator original, Cloner cloner)
44      : base(original, cloner) {
45    }
46
47    public override IDeepCloneable Clone(Cloner cloner) {
48      return new NeighbourhoodWeightCalculator(this, cloner);
49    }
50
51    protected override IEnumerable<double> DiscriminantCalculateWeights(ItemCollection<IDiscriminantFunctionClassificationSolution> discriminantSolutions) {
52      List<List<double>> estimatedTrainingValEnumerators = new List<List<double>>();
53      List<List<double>> estimatedTrainingClassValEnumerators = new List<List<double>>();
54      foreach (var solution in discriminantSolutions) {
55        estimatedTrainingValEnumerators.Add(solution.EstimatedTrainingValues.ToList());
56        estimatedTrainingClassValEnumerators.Add(solution.EstimatedTrainingClassValues.ToList());
57      }
58
59      List<double> weights = Enumerable.Repeat<double>(0, discriminantSolutions.Count()).ToList<double>();
60
61      IClassificationProblemData problemData = discriminantSolutions.ElementAt(0).ProblemData;
62      List<double> targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable).ToList();
63      List<double> trainingVal = GetValues(targetValues, problemData.TrainingIndizes).ToList();
64
65      double pointAvg, help;
66      int count;
67      for (int point = 0; point < estimatedTrainingClassValEnumerators.First().Count; point++) {
68        pointAvg = 0.0;
69        count = 0;
70        for (int solution = 0; solution < estimatedTrainingClassValEnumerators.Count; solution++) {
71          if (estimatedTrainingClassValEnumerators[solution][point].Equals(targetValues[point])) {
72            pointAvg += estimatedTrainingValEnumerators[solution][point];
73            count++;
74          }
75        }
76        pointAvg /= (double)count;
77        for (int solution = 0; solution < estimatedTrainingClassValEnumerators.Count; solution++) {
78          if (estimatedTrainingClassValEnumerators[solution][point].Equals(targetValues[point])) {
79            weights[solution] += 0.5;
80            help = Math.Abs(estimatedTrainingValEnumerators[solution][point] - 0.5);
81            weights[solution] += help < 0.5 ? 0.5 - help : 0.0;
82          }
83        }
84      }
85      return weights;
86    }
87  }
88}
Note: See TracBrowser for help on using the repository browser.