Free cookie consent management tool by TermsFeed Policy Generator

source: branches/PausableBasicAlgorithm/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolutionBase.cs @ 13371

Last change on this file since 13371 was 12641, checked in by gkronber, 9 years ago

#2392: fixed all warnings by using PearsonsRCalculator instead of PearsonsRSquaredCalculator

File size: 16.7 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.Data;
26using HeuristicLab.Optimization;
27using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
28
29namespace HeuristicLab.Problems.DataAnalysis {
30  [StorableClass]
31  public abstract class RegressionSolutionBase : DataAnalysisSolution, IRegressionSolution {
32    protected const string TrainingMeanSquaredErrorResultName = "Mean squared error (training)";
33    protected const string TestMeanSquaredErrorResultName = "Mean squared error (test)";
34    protected const string TrainingMeanAbsoluteErrorResultName = "Mean absolute error (training)";
35    protected const string TestMeanAbsoluteErrorResultName = "Mean absolute error (test)";
36    protected const string TrainingSquaredCorrelationResultName = "Pearson's R² (training)";
37    protected const string TestSquaredCorrelationResultName = "Pearson's R² (test)";
38    protected const string TrainingRelativeErrorResultName = "Average relative error (training)";
39    protected const string TestRelativeErrorResultName = "Average relative error (test)";
40    protected const string TrainingNormalizedMeanSquaredErrorResultName = "Normalized mean squared error (training)";
41    protected const string TestNormalizedMeanSquaredErrorResultName = "Normalized mean squared error (test)";
42    protected const string TrainingRootMeanSquaredErrorResultName = "Root mean squared error (training)";
43    protected const string TestRootMeanSquaredErrorResultName = "Root mean squared error (test)";
44
45    // BackwardsCompatibility3.3
46    #region Backwards compatible code, remove with 3.5
47    private const string TrainingMeanErrorResultName = "Mean error (training)";
48    private const string TestMeanErrorResultName = "Mean error (test)";
49    #endregion
50
51
52    protected const string TrainingMeanSquaredErrorResultDescription = "Mean of squared errors of the model on the training partition";
53    protected const string TestMeanSquaredErrorResultDescription = "Mean of squared errors of the model on the test partition";
54    protected const string TrainingMeanAbsoluteErrorResultDescription = "Mean of absolute errors of the model on the training partition";
55    protected const string TestMeanAbsoluteErrorResultDescription = "Mean of absolute errors of the model on the test partition";
56    protected const string TrainingSquaredCorrelationResultDescription = "Squared Pearson's correlation coefficient of the model output and the actual values on the training partition";
57    protected const string TestSquaredCorrelationResultDescription = "Squared Pearson's correlation coefficient of the model output and the actual values on the test partition";
58    protected const string TrainingRelativeErrorResultDescription = "Average of the relative errors of the model output and the actual values on the training partition";
59    protected const string TestRelativeErrorResultDescription = "Average of the relative errors of the model output and the actual values on the test partition";
60    protected const string TrainingNormalizedMeanSquaredErrorResultDescription = "Normalized mean of squared errors of the model on the training partition";
61    protected const string TestNormalizedMeanSquaredErrorResultDescription = "Normalized mean of squared errors of the model on the test partition";
62    protected const string TrainingRootMeanSquaredErrorResultDescription = "Root mean of squared errors of the model on the training partition";
63    protected const string TestRootMeanSquaredErrorResultDescription = "Root mean of squared errors of the model on the test partition";
64
65    public new IRegressionModel Model {
66      get { return (IRegressionModel)base.Model; }
67      protected set { base.Model = value; }
68    }
69
70    public new IRegressionProblemData ProblemData {
71      get { return (IRegressionProblemData)base.ProblemData; }
72      set { base.ProblemData = value; }
73    }
74
75    public abstract IEnumerable<double> EstimatedValues { get; }
76    public abstract IEnumerable<double> EstimatedTrainingValues { get; }
77    public abstract IEnumerable<double> EstimatedTestValues { get; }
78    public abstract IEnumerable<double> GetEstimatedValues(IEnumerable<int> rows);
79
80    #region Results
81    public double TrainingMeanSquaredError {
82      get { return ((DoubleValue)this[TrainingMeanSquaredErrorResultName].Value).Value; }
83      private set { ((DoubleValue)this[TrainingMeanSquaredErrorResultName].Value).Value = value; }
84    }
85    public double TestMeanSquaredError {
86      get { return ((DoubleValue)this[TestMeanSquaredErrorResultName].Value).Value; }
87      private set { ((DoubleValue)this[TestMeanSquaredErrorResultName].Value).Value = value; }
88    }
89    public double TrainingMeanAbsoluteError {
90      get { return ((DoubleValue)this[TrainingMeanAbsoluteErrorResultName].Value).Value; }
91      private set { ((DoubleValue)this[TrainingMeanAbsoluteErrorResultName].Value).Value = value; }
92    }
93    public double TestMeanAbsoluteError {
94      get { return ((DoubleValue)this[TestMeanAbsoluteErrorResultName].Value).Value; }
95      private set { ((DoubleValue)this[TestMeanAbsoluteErrorResultName].Value).Value = value; }
96    }
97    public double TrainingRSquared {
98      get { return ((DoubleValue)this[TrainingSquaredCorrelationResultName].Value).Value; }
99      private set { ((DoubleValue)this[TrainingSquaredCorrelationResultName].Value).Value = value; }
100    }
101    public double TestRSquared {
102      get { return ((DoubleValue)this[TestSquaredCorrelationResultName].Value).Value; }
103      private set { ((DoubleValue)this[TestSquaredCorrelationResultName].Value).Value = value; }
104    }
105    public double TrainingRelativeError {
106      get { return ((DoubleValue)this[TrainingRelativeErrorResultName].Value).Value; }
107      private set { ((DoubleValue)this[TrainingRelativeErrorResultName].Value).Value = value; }
108    }
109    public double TestRelativeError {
110      get { return ((DoubleValue)this[TestRelativeErrorResultName].Value).Value; }
111      private set { ((DoubleValue)this[TestRelativeErrorResultName].Value).Value = value; }
112    }
113    public double TrainingNormalizedMeanSquaredError {
114      get { return ((DoubleValue)this[TrainingNormalizedMeanSquaredErrorResultName].Value).Value; }
115      private set { ((DoubleValue)this[TrainingNormalizedMeanSquaredErrorResultName].Value).Value = value; }
116    }
117    public double TestNormalizedMeanSquaredError {
118      get { return ((DoubleValue)this[TestNormalizedMeanSquaredErrorResultName].Value).Value; }
119      private set { ((DoubleValue)this[TestNormalizedMeanSquaredErrorResultName].Value).Value = value; }
120    }
121    public double TrainingRootMeanSquaredError {
122      get { return ((DoubleValue)this[TrainingRootMeanSquaredErrorResultName].Value).Value; }
123      private set { ((DoubleValue)this[TrainingRootMeanSquaredErrorResultName].Value).Value = value; }
124    }
125    public double TestRootMeanSquaredError {
126      get { return ((DoubleValue)this[TestRootMeanSquaredErrorResultName].Value).Value; }
127      private set { ((DoubleValue)this[TestRootMeanSquaredErrorResultName].Value).Value = value; }
128    }
129
130    // BackwardsCompatibility3.3
131    #region Backwards compatible code, remove with 3.5
132    private double TrainingMeanError {
133      get {
134        if (!ContainsKey(TrainingMeanErrorResultName)) return double.NaN;
135        return ((DoubleValue)this[TrainingMeanErrorResultName].Value).Value;
136      }
137      set {
138        if (ContainsKey(TrainingMeanErrorResultName))
139          ((DoubleValue)this[TrainingMeanErrorResultName].Value).Value = value;
140      }
141    }
142    private double TestMeanError {
143      get {
144        if (!ContainsKey(TestMeanErrorResultName)) return double.NaN;
145        return ((DoubleValue)this[TestMeanErrorResultName].Value).Value;
146      }
147      set {
148        if (ContainsKey(TestMeanErrorResultName))
149          ((DoubleValue)this[TestMeanErrorResultName].Value).Value = value;
150      }
151    }
152    #endregion
153    #endregion
154
155    [StorableConstructor]
156    protected RegressionSolutionBase(bool deserializing) : base(deserializing) { }
157    protected RegressionSolutionBase(RegressionSolutionBase original, Cloner cloner)
158      : base(original, cloner) {
159    }
160    protected RegressionSolutionBase(IRegressionModel model, IRegressionProblemData problemData)
161      : base(model, problemData) {
162      Add(new Result(TrainingMeanSquaredErrorResultName, TrainingMeanSquaredErrorResultDescription, new DoubleValue()));
163      Add(new Result(TestMeanSquaredErrorResultName, TestMeanSquaredErrorResultDescription, new DoubleValue()));
164      Add(new Result(TrainingMeanAbsoluteErrorResultName, TrainingMeanAbsoluteErrorResultDescription, new DoubleValue()));
165      Add(new Result(TestMeanAbsoluteErrorResultName, TestMeanAbsoluteErrorResultDescription, new DoubleValue()));
166      Add(new Result(TrainingSquaredCorrelationResultName, TrainingSquaredCorrelationResultDescription, new DoubleValue()));
167      Add(new Result(TestSquaredCorrelationResultName, TestSquaredCorrelationResultDescription, new DoubleValue()));
168      Add(new Result(TrainingRelativeErrorResultName, TrainingRelativeErrorResultDescription, new PercentValue()));
169      Add(new Result(TestRelativeErrorResultName, TestRelativeErrorResultDescription, new PercentValue()));
170      Add(new Result(TrainingNormalizedMeanSquaredErrorResultName, TrainingNormalizedMeanSquaredErrorResultDescription, new DoubleValue()));
171      Add(new Result(TestNormalizedMeanSquaredErrorResultName, TestNormalizedMeanSquaredErrorResultDescription, new DoubleValue()));
172      Add(new Result(TrainingRootMeanSquaredErrorResultName, TrainingRootMeanSquaredErrorResultDescription, new DoubleValue()));
173      Add(new Result(TestRootMeanSquaredErrorResultName, TestRootMeanSquaredErrorResultDescription, new DoubleValue()));
174    }
175
176    [StorableHook(HookType.AfterDeserialization)]
177    private void AfterDeserialization() {
178      // BackwardsCompatibility3.4
179      #region Backwards compatible code, remove with 3.5
180      if (!ContainsKey(TrainingMeanAbsoluteErrorResultName)) {
181        OnlineCalculatorError errorState;
182        Add(new Result(TrainingMeanAbsoluteErrorResultName, "Mean of absolute errors of the model on the training partition", new DoubleValue()));
183        double trainingMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices), out errorState);
184        TrainingMeanAbsoluteError = errorState == OnlineCalculatorError.None ? trainingMAE : double.NaN;
185      }
186
187      if (!ContainsKey(TestMeanAbsoluteErrorResultName)) {
188        OnlineCalculatorError errorState;
189        Add(new Result(TestMeanAbsoluteErrorResultName, "Mean of absolute errors of the model on the test partition", new DoubleValue()));
190        double testMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices), out errorState);
191        TestMeanAbsoluteError = errorState == OnlineCalculatorError.None ? testMAE : double.NaN;
192      }
193
194      if (!ContainsKey(TrainingRootMeanSquaredErrorResultName)) {
195        OnlineCalculatorError errorState;
196        Add(new Result(TrainingRootMeanSquaredErrorResultName, TrainingRootMeanSquaredErrorResultDescription, new DoubleValue()));
197        double trainingMSE = OnlineMeanSquaredErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices), out errorState);
198        TrainingRootMeanSquaredError = errorState == OnlineCalculatorError.None ? Math.Sqrt(trainingMSE) : double.NaN;
199      }
200
201      if (!ContainsKey(TestRootMeanSquaredErrorResultName)) {
202        OnlineCalculatorError errorState;
203        Add(new Result(TestRootMeanSquaredErrorResultName, TestRootMeanSquaredErrorResultDescription, new DoubleValue()));
204        double testMSE = OnlineMeanSquaredErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices), out errorState);
205        TestRootMeanSquaredError = errorState == OnlineCalculatorError.None ? Math.Sqrt(testMSE) : double.NaN;
206      }
207      #endregion
208    }
209
210    protected override void RecalculateResults() {
211      CalculateRegressionResults();
212    }
213
214    protected void CalculateRegressionResults() {
215      IEnumerable<double> estimatedTrainingValues = EstimatedTrainingValues; // cache values
216      IEnumerable<double> originalTrainingValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices);
217      IEnumerable<double> estimatedTestValues = EstimatedTestValues; // cache values
218      IEnumerable<double> originalTestValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices);
219
220      OnlineCalculatorError errorState;
221      double trainingMSE = OnlineMeanSquaredErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
222      TrainingMeanSquaredError = errorState == OnlineCalculatorError.None ? trainingMSE : double.NaN;
223      double testMSE = OnlineMeanSquaredErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
224      TestMeanSquaredError = errorState == OnlineCalculatorError.None ? testMSE : double.NaN;
225
226      double trainingMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
227      TrainingMeanAbsoluteError = errorState == OnlineCalculatorError.None ? trainingMAE : double.NaN;
228      double testMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
229      TestMeanAbsoluteError = errorState == OnlineCalculatorError.None ? testMAE : double.NaN;
230
231      double trainingR = OnlinePearsonsRCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
232      TrainingRSquared = errorState == OnlineCalculatorError.None ? trainingR*trainingR : double.NaN;
233      double testR = OnlinePearsonsRCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
234      TestRSquared = errorState == OnlineCalculatorError.None ? testR*testR : double.NaN;
235
236      double trainingRelError = OnlineMeanAbsolutePercentageErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
237      TrainingRelativeError = errorState == OnlineCalculatorError.None ? trainingRelError : double.NaN;
238      double testRelError = OnlineMeanAbsolutePercentageErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
239      TestRelativeError = errorState == OnlineCalculatorError.None ? testRelError : double.NaN;
240
241      double trainingNMSE = OnlineNormalizedMeanSquaredErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
242      TrainingNormalizedMeanSquaredError = errorState == OnlineCalculatorError.None ? trainingNMSE : double.NaN;
243      double testNMSE = OnlineNormalizedMeanSquaredErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
244      TestNormalizedMeanSquaredError = errorState == OnlineCalculatorError.None ? testNMSE : double.NaN;
245
246      TrainingRootMeanSquaredError = Math.Sqrt(TrainingMeanSquaredError);
247      TestRootMeanSquaredError = Math.Sqrt(TestMeanSquaredError);
248
249      // BackwardsCompatibility3.3
250      #region Backwards compatible code, remove with 3.5
251      if (ContainsKey(TrainingMeanErrorResultName)) {
252        double trainingME = OnlineMeanErrorCalculator.Calculate(originalTrainingValues, estimatedTrainingValues, out errorState);
253        TrainingMeanError = errorState == OnlineCalculatorError.None ? trainingME : double.NaN;
254      }
255      if (ContainsKey(TestMeanErrorResultName)) {
256        double testME = OnlineMeanErrorCalculator.Calculate(originalTestValues, estimatedTestValues, out errorState);
257        TestMeanError = errorState == OnlineCalculatorError.None ? testME : double.NaN;
258      }
259      #endregion
260    }
261  }
262}
Note: See TracBrowser for help on using the repository browser.