Free cookie consent management tool by TermsFeed Policy Generator

source: branches/GBT-trunkintegration/HeuristicLab.Algorithms.DataAnalysis/3.4/GradientBoostedTrees/GradientBoostedTreesModel.cs @ 12660

Last change on this file since 12660 was 12660, checked in by gkronber, 7 years ago

#2261: minor change if(!rows.Any()) return ...

File size: 3.5 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2015 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 * and the BEACON Center for the Study of Evolution in Action.
5 *
6 * This file is part of HeuristicLab.
7 *
8 * HeuristicLab is free software: you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation, either version 3 of the License, or
11 * (at your option) any later version.
12 *
13 * HeuristicLab is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
16 * GNU General Public License for more details.
17 *
18 * You should have received a copy of the GNU General Public License
19 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
20 */
21#endregion
22
23using System;
24using System.Collections.Generic;
25using System.Linq;
26using HeuristicLab.Common;
27using HeuristicLab.Core;
28using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
29using HeuristicLab.Problems.DataAnalysis;
30
31namespace HeuristicLab.Algorithms.DataAnalysis {
32  [StorableClass]
33  [Item("Gradient boosted tree model", "")]
34  // this is essentially a collection of weighted regression models
35  public sealed class GradientBoostedTreesModel : NamedItem, IRegressionModel {
36    [Storable]
37    private readonly IList<IRegressionModel> models;
38    public IEnumerable<IRegressionModel> Models { get { return models; } }
39
40    [Storable]
41    private readonly IList<double> weights;
42    public IEnumerable<double> Weights { get { return weights; } }
43
44    [StorableConstructor]
45    private GradientBoostedTreesModel(bool deserializing) : base(deserializing) { }
46    private GradientBoostedTreesModel(GradientBoostedTreesModel original, Cloner cloner)
47      : base(original, cloner) {
48      this.weights = new List<double>(original.weights);
49      this.models = new List<IRegressionModel>(original.models.Select(m => cloner.Clone(m)));
50    }
51    public GradientBoostedTreesModel(IEnumerable<IRegressionModel> models, IEnumerable<double> weights)
52      : base("Gradient boosted tree model", string.Empty) {
53      this.models = new List<IRegressionModel>(models);
54      this.weights = new List<double>(weights);
55
56      if (this.models.Count != this.weights.Count) throw new ArgumentException();
57    }
58
59    public override IDeepCloneable Clone(Cloner cloner) {
60      return new GradientBoostedTreesModel(this, cloner);
61    }
62
63    public IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
64      // allocate target array go over all models and add up weighted estimation for each row
65      if (!rows.Any()) return Enumerable.Empty<double>(); // return immediately if rows is empty. This prevents multiple iteration over lazy rows enumerable.
66                                                          // (which essentially looks up indexes in a dictionary)
67      var res = new double[rows.Count()];
68      for (int i = 0; i < models.Count; i++) {
69        var w = weights[i];
70        var m = models[i];
71        int r = 0;
72        foreach (var est in m.GetEstimatedValues(dataset, rows)) {
73          res[r++] += w * est;
74        }
75      }
76      return res;
77    }
78
79    public IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
80      return new RegressionSolution(this, (IRegressionProblemData)problemData.Clone());
81    }
82  }
83}
Note: See TracBrowser for help on using the repository browser.