source: trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/GaussianProcess/CovarianceFunctions/CovarianceSum.cs @ 13784

Last change on this file since 13784 was 13784, checked in by pfleck, 5 years ago

#2591 Made the creation of a GaussianProcessModel faster by avoiding additional iterators during calculation of the hyperparameter gradients.
The gradients of the hyperparameters are now calculated in one sweep and returned as IList, instead of returning an iterator (with yield return).
This avoids a large amount of Move-calls of the iterator, especially for covariance functions with a lot of hyperparameters.
Besides, the signature of the CovarianceGradientFunctionDelegate is changed, to return an IList instead of an IEnumerable to avoid unnececary ToList or ToArray calls.

File size: 3.6 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 System.Linq;
25using HeuristicLab.Common;
26using HeuristicLab.Core;
27using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
28
29namespace HeuristicLab.Algorithms.DataAnalysis {
30  [StorableClass]
31  [Item(Name = "CovarianceSum",
32    Description = "Sum covariance function for Gaussian processes.")]
33  public sealed class CovarianceSum : Item, ICovarianceFunction {
34    [Storable]
35    private ItemList<ICovarianceFunction> terms;
36
37    [Storable]
38    private int numberOfVariables;
39    public ItemList<ICovarianceFunction> Terms {
40      get { return terms; }
41    }
42
43    [StorableConstructor]
44    private CovarianceSum(bool deserializing)
45      : base(deserializing) {
46    }
47
48    private CovarianceSum(CovarianceSum original, Cloner cloner)
49      : base(original, cloner) {
50      this.terms = cloner.Clone(original.terms);
51      this.numberOfVariables = original.numberOfVariables;
52    }
53
54    public CovarianceSum()
55      : base() {
56      this.terms = new ItemList<ICovarianceFunction>();
57    }
58
59    public override IDeepCloneable Clone(Cloner cloner) {
60      return new CovarianceSum(this, cloner);
61    }
62
63    public int GetNumberOfParameters(int numberOfVariables) {
64      this.numberOfVariables = numberOfVariables;
65      return terms.Select(t => t.GetNumberOfParameters(numberOfVariables)).Sum();
66    }
67
68    public void SetParameter(double[] p) {
69      int offset = 0;
70      foreach (var t in terms) {
71        var numberOfParameters = t.GetNumberOfParameters(numberOfVariables);
72        t.SetParameter(p.Skip(offset).Take(numberOfParameters).ToArray());
73        offset += numberOfParameters;
74      }
75    }
76
77    public ParameterizedCovarianceFunction GetParameterizedCovarianceFunction(double[] p, int[] columnIndices) {
78      if (terms.Count == 0) throw new ArgumentException("at least one term is necessary for the product covariance function.");
79      var functions = new List<ParameterizedCovarianceFunction>();
80      foreach (var t in terms) {
81        var numberOfParameters = t.GetNumberOfParameters(numberOfVariables);
82        functions.Add(t.GetParameterizedCovarianceFunction(p.Take(numberOfParameters).ToArray(), columnIndices));
83        p = p.Skip(numberOfParameters).ToArray();
84      }
85
86      var sum = new ParameterizedCovarianceFunction();
87      sum.Covariance = (x, i, j) => functions.Select(e => e.Covariance(x, i, j)).Sum();
88      sum.CrossCovariance = (x, xt, i, j) => functions.Select(e => e.CrossCovariance(x, xt, i, j)).Sum();
89      sum.CovarianceGradient = (x, i, j) => {
90        var g = new List<double>(functions.Sum(e => e.CovarianceGradient(x, i, j).Count));
91        foreach (var e in functions)
92          g.AddRange(e.CovarianceGradient(x, i, j));
93        return g;
94      };
95      return sum;
96    }
97  }
98}
Note: See TracBrowser for help on using the repository browser.