Free cookie consent management tool by TermsFeed Policy Generator

source: stable/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/GaussianProcessRegressionModelCreator.cs @ 17513

Last change on this file since 17513 was 17181, checked in by swagner, 5 years ago

#2875: Merged r17180 from trunk to stable

File size: 3.9 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 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.Linq;
24using HeuristicLab.Common;
25using HeuristicLab.Core;
26using HeuristicLab.Data;
27using HeuristicLab.Encodings.RealVectorEncoding;
28using HeuristicLab.Parameters;
29using HEAL.Attic;
30using HeuristicLab.Problems.DataAnalysis;
31
32namespace HeuristicLab.Algorithms.DataAnalysis {
33  [StorableType("0C20B1D2-9A58-4E77-9300-F5D76650DC19")]
34  [Item(Name = "GaussianProcessRegressionModelCreator",
35    Description = "Creates a Gaussian process model for regression given the data, the hyperparameters, a mean function, and a covariance function.")]
36  public sealed class GaussianProcessRegressionModelCreator : GaussianProcessModelCreator, IGaussianProcessRegressionModelCreator {
37    private const string ProblemDataParameterName = "ProblemData";
38
39    #region Parameter Properties
40    public ILookupParameter<IRegressionProblemData> ProblemDataParameter {
41      get { return (ILookupParameter<IRegressionProblemData>)Parameters[ProblemDataParameterName]; }
42    }
43    #endregion
44
45    #region Properties
46    private IRegressionProblemData ProblemData {
47      get { return ProblemDataParameter.ActualValue; }
48    }
49    #endregion
50    [StorableConstructor]
51    private GaussianProcessRegressionModelCreator(StorableConstructorFlag _) : base(_) { }
52    private GaussianProcessRegressionModelCreator(GaussianProcessRegressionModelCreator original, Cloner cloner) : base(original, cloner) { }
53    public GaussianProcessRegressionModelCreator()
54      : base() {
55      Parameters.Add(new LookupParameter<IRegressionProblemData>(ProblemDataParameterName, "The regression problem data for the Gaussian process model."));
56    }
57
58    public override IDeepCloneable Clone(Cloner cloner) {
59      return new GaussianProcessRegressionModelCreator(this, cloner);
60    }
61
62    public override IOperation Apply() {
63      try {
64        var model = Create(ProblemData, Hyperparameter.ToArray(), MeanFunction, CovarianceFunction, ScaleInputValues);
65        ModelParameter.ActualValue = model;
66        NegativeLogLikelihoodParameter.ActualValue = new DoubleValue(model.NegativeLogLikelihood);
67        NegativeLogPseudoLikelihoodParameter.ActualValue = new DoubleValue(model.LooCvNegativeLogPseudoLikelihood);
68        HyperparameterGradientsParameter.ActualValue = new RealVector(model.HyperparameterGradients);
69        return base.Apply();
70      }
71      catch (ArgumentException) { }
72      catch (alglib.alglibexception) { }
73      NegativeLogLikelihoodParameter.ActualValue = new DoubleValue(1E300);
74      HyperparameterGradientsParameter.ActualValue = new RealVector(Hyperparameter.Count());
75      return base.Apply();
76    }
77
78    public static IGaussianProcessModel Create(IRegressionProblemData problemData, double[] hyperparameter, IMeanFunction meanFunction, ICovarianceFunction covarianceFunction, bool scaleInputs = true) {
79      return new GaussianProcessModel(problemData.Dataset, problemData.TargetVariable, problemData.AllowedInputVariables, problemData.TrainingIndices, hyperparameter, meanFunction, covarianceFunction, scaleInputs);
80    }
81  }
82}
Note: See TracBrowser for help on using the repository browser.