source: stable/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationSolutionBase.cs @ 11170

Last change on this file since 11170 was 11170, checked in by ascheibe, 4 years ago

#2115 updated copyright year in stable branch

File size: 6.5 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2014 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.Data;
26using HeuristicLab.Optimization;
27using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
28
29namespace HeuristicLab.Problems.DataAnalysis {
30  [StorableClass]
31  public abstract class ClassificationSolutionBase : DataAnalysisSolution, IClassificationSolution {
32    private const string TrainingAccuracyResultName = "Accuracy (training)";
33    private const string TestAccuracyResultName = "Accuracy (test)";
34    private const string TrainingNormalizedGiniCoefficientResultName = "Normalized Gini Coefficient (training)";
35    private const string TestNormalizedGiniCoefficientResultName = "Normalized Gini Coefficient (test)";
36
37    public new IClassificationModel Model {
38      get { return (IClassificationModel)base.Model; }
39      protected set { base.Model = value; }
40    }
41
42    public new IClassificationProblemData ProblemData {
43      get { return (IClassificationProblemData)base.ProblemData; }
44      set { base.ProblemData = value; }
45    }
46
47    #region Results
48    public double TrainingAccuracy {
49      get { return ((DoubleValue)this[TrainingAccuracyResultName].Value).Value; }
50      private set { ((DoubleValue)this[TrainingAccuracyResultName].Value).Value = value; }
51    }
52    public double TestAccuracy {
53      get { return ((DoubleValue)this[TestAccuracyResultName].Value).Value; }
54      private set { ((DoubleValue)this[TestAccuracyResultName].Value).Value = value; }
55    }
56    public double TrainingNormalizedGiniCoefficient {
57      get { return ((DoubleValue)this[TrainingNormalizedGiniCoefficientResultName].Value).Value; }
58      protected set { ((DoubleValue)this[TrainingNormalizedGiniCoefficientResultName].Value).Value = value; }
59    }
60    public double TestNormalizedGiniCoefficient {
61      get { return ((DoubleValue)this[TestNormalizedGiniCoefficientResultName].Value).Value; }
62      protected set { ((DoubleValue)this[TestNormalizedGiniCoefficientResultName].Value).Value = value; }
63    }
64    #endregion
65
66    [StorableConstructor]
67    protected ClassificationSolutionBase(bool deserializing) : base(deserializing) { }
68    protected ClassificationSolutionBase(ClassificationSolutionBase original, Cloner cloner)
69      : base(original, cloner) {
70    }
71    protected ClassificationSolutionBase(IClassificationModel model, IClassificationProblemData problemData)
72      : base(model, problemData) {
73      Add(new Result(TrainingAccuracyResultName, "Accuracy of the model on the training partition (percentage of correctly classified instances).", new PercentValue()));
74      Add(new Result(TestAccuracyResultName, "Accuracy of the model on the test partition (percentage of correctly classified instances).", new PercentValue()));
75      Add(new Result(TrainingNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the training partition.", new DoubleValue()));
76      Add(new Result(TestNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the test partition.", new DoubleValue()));
77    }
78
79    [StorableHook(HookType.AfterDeserialization)]
80    private void AfterDeserialization() {
81      if (!this.ContainsKey(TrainingNormalizedGiniCoefficientResultName))
82        Add(new Result(TrainingNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the training partition.", new DoubleValue()));
83      if (!this.ContainsKey(TestNormalizedGiniCoefficientResultName))
84        Add(new Result(TestNormalizedGiniCoefficientResultName, "Normalized Gini coefficient of the model on the test partition.", new DoubleValue()));
85    }
86
87    protected void CalculateClassificationResults() {
88      double[] estimatedTrainingClassValues = EstimatedTrainingClassValues.ToArray(); // cache values
89      double[] originalTrainingClassValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices).ToArray();
90      double[] estimatedTestClassValues = EstimatedTestClassValues.ToArray(); // cache values
91      double[] originalTestClassValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices).ToArray();
92
93      OnlineCalculatorError errorState;
94      double trainingAccuracy = OnlineAccuracyCalculator.Calculate(originalTrainingClassValues, estimatedTrainingClassValues, out errorState);
95      if (errorState != OnlineCalculatorError.None) trainingAccuracy = double.NaN;
96      double testAccuracy = OnlineAccuracyCalculator.Calculate(originalTestClassValues, estimatedTestClassValues, out errorState);
97      if (errorState != OnlineCalculatorError.None) testAccuracy = double.NaN;
98
99      TrainingAccuracy = trainingAccuracy;
100      TestAccuracy = testAccuracy;
101
102      double trainingNormalizedGini = NormalizedGiniCalculator.Calculate(originalTrainingClassValues, estimatedTrainingClassValues, out errorState);
103      if (errorState != OnlineCalculatorError.None) trainingNormalizedGini = double.NaN;
104      double testNormalizedGini = NormalizedGiniCalculator.Calculate(originalTestClassValues, estimatedTestClassValues, out errorState);
105      if (errorState != OnlineCalculatorError.None) testNormalizedGini = double.NaN;
106
107      TrainingNormalizedGiniCoefficient = trainingNormalizedGini;
108      TestNormalizedGiniCoefficient = testNormalizedGini;
109    }
110
111    public abstract IEnumerable<double> EstimatedClassValues { get; }
112    public abstract IEnumerable<double> EstimatedTrainingClassValues { get; }
113    public abstract IEnumerable<double> EstimatedTestClassValues { get; }
114
115    public abstract IEnumerable<double> GetEstimatedClassValues(IEnumerable<int> rows);
116
117    protected override void RecalculateResults() {
118      CalculateClassificationResults();
119    }
120  }
121}
Note: See TracBrowser for help on using the repository browser.