Free cookie consent management tool by TermsFeed Policy Generator

source: branches/HeuristicLab.Problems.GaussianProcessTuning/HeuristicLab.Problems.Instances.DataAnalysis.GaussianProcessRegression/GaussianProcess2dPeriodic.cs @ 9214

Last change on this file since 9214 was 9214, checked in by gkronber, 12 years ago

#1967: added a benchmark function

File size: 3.8 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.Algorithms.DataAnalysis;
26using HeuristicLab.Data;
27using HeuristicLab.Random;
28
29namespace HeuristicLab.Problems.Instances.DataAnalysis {
30  public class GaussianProcess2dPeriodic : ArtificialRegressionDataDescriptor {
31
32    public override string Name {
33      get {
34        return "Gaussian Process 2d periodic";
35      }
36    }
37    public override string Description {
38      get { return ""; }
39    }
40    protected override string TargetVariable { get { return "Y"; } }
41    protected override string[] VariableNames { get { return new string[] { "X1", "X2", "Y" }; } }
42    protected override string[] AllowedInputVariables { get { return new string[] { "X1", "X2" }; } }
43    protected override int TrainingPartitionStart { get { return 0; } }
44    protected override int TrainingPartitionEnd { get { return 20 * 20; } }
45    protected override int TestPartitionStart { get { return 20 * 20; } }
46    protected override int TestPartitionEnd { get { return 2 * (20 * 20); } }
47
48    protected override List<List<double>> GenerateValues() {
49      List<List<double>> independentTrainingData = new List<List<double>>();
50      List<List<double>> independentTestData = new List<List<double>>();
51      for (int i = 0; i < AllowedInputVariables.Count(); i++) {
52        independentTrainingData.Add(ValueGenerator.GenerateSteps(0, 0.99, 1.0 / 20).ToList());
53        independentTestData.Add(ValueGenerator.GenerateSteps(0.005, 1, 1.0 / 20).ToList());
54      }
55
56
57      var trainingData = ValueGenerator.GenerateAllCombinationsOfValuesInLists(independentTrainingData);
58      var testData = ValueGenerator.GenerateAllCombinationsOfValuesInLists(independentTestData);
59      List<List<double>> data = new List<List<double>>();
60      foreach (var e in trainingData) {
61        data.Add(e.ToList());
62      }
63      int j = 0;
64      foreach (var e in testData) {
65        data[j].AddRange(e);
66        j++;
67      }
68
69      var covarianceFunction = new CovarianceSum();
70      var m1 = new CovarianceMask();
71      m1.SelectedDimensionsParameter.Value = new IntArray(new int[] { 0 });
72      m1.CovarianceFunctionParameter.Value = new CovariancePeriodic();
73      var m2 = new CovarianceMask();
74      m2.SelectedDimensionsParameter.Value = new IntArray(new int[] { 1 });
75      m2.CovarianceFunctionParameter.Value = new CovariancePeriodic();
76
77      covarianceFunction.Terms.Add(m1);
78      covarianceFunction.Terms.Add(m2);
79      covarianceFunction.Terms.Add(new CovarianceNoise());
80      var cov =
81        covarianceFunction.GetParameterizedCovarianceFunction(
82          Enumerable.Repeat(0.0, covarianceFunction.GetNumberOfParameters(2) - 1)
83          .Concat(new double[] { Math.Log(Math.Sqrt(0.01)) })
84          .ToArray(),
85          null);
86
87      var mt = new MersenneTwister(31415);
88      var target = Util.SampleGaussianProcess(mt, cov, data);
89      data.Add(target);
90
91      return data;
92    }
93  }
94}
Note: See TracBrowser for help on using the repository browser.