source: branches/symbreg-factors-2650/HeuristicLab.Algorithms.DataAnalysis/3.4/NearestNeighbour/NearestNeighbourRegression.cs @ 14542

Last change on this file since 14542 was 14542, checked in by gkronber, 3 years ago

#2650: merged r14504:14533 from trunk to branch

File size: 5.1 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2016 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.Threading;
24using HeuristicLab.Common;
25using HeuristicLab.Core;
26using HeuristicLab.Data;
27using HeuristicLab.Optimization;
28using HeuristicLab.Parameters;
29using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
30using HeuristicLab.Problems.DataAnalysis;
31
32namespace HeuristicLab.Algorithms.DataAnalysis {
33  /// <summary>
34  /// Nearest neighbour regression data analysis algorithm.
35  /// </summary>
36  [Item("Nearest Neighbour Regression (kNN)", "Nearest neighbour regression data analysis algorithm (wrapper for ALGLIB).")]
37  [Creatable(CreatableAttribute.Categories.DataAnalysisRegression, Priority = 150)]
38  [StorableClass]
39  public sealed class NearestNeighbourRegression : FixedDataAnalysisAlgorithm<IRegressionProblem> {
40    private const string KParameterName = "K";
41    private const string NearestNeighbourRegressionModelResultName = "Nearest neighbour regression solution";
42    private const string WeightsParameterName = "Weights";
43
44    #region parameter properties
45    public IFixedValueParameter<IntValue> KParameter {
46      get { return (IFixedValueParameter<IntValue>)Parameters[KParameterName]; }
47    }
48
49    public IValueParameter<DoubleArray> WeightsParameter {
50      get { return (IValueParameter<DoubleArray>)Parameters[WeightsParameterName]; }
51    }
52    #endregion
53    #region properties
54    public int K {
55      get { return KParameter.Value.Value; }
56      set {
57        if (value <= 0) throw new ArgumentException("K must be larger than zero.", "K");
58        else KParameter.Value.Value = value;
59      }
60    }
61
62    public DoubleArray Weights {
63      get { return WeightsParameter.Value; }
64      set { WeightsParameter.Value = value; }
65    }
66    #endregion
67
68    [StorableConstructor]
69    private NearestNeighbourRegression(bool deserializing) : base(deserializing) { }
70    private NearestNeighbourRegression(NearestNeighbourRegression original, Cloner cloner)
71      : base(original, cloner) {
72    }
73    public NearestNeighbourRegression()
74      : base() {
75      Parameters.Add(new FixedValueParameter<IntValue>(KParameterName, "The number of nearest neighbours to consider for regression.", new IntValue(3)));
76      Parameters.Add(new OptionalValueParameter<DoubleArray>(WeightsParameterName, "Optional: use weights to specify individual scaling values for all features. If not set the weights are calculated automatically (each feature is scaled to unit variance)"));
77      Problem = new RegressionProblem();
78    }
79
80    [StorableHook(HookType.AfterDeserialization)]
81    private void AfterDeserialization() {
82      // BackwardsCompatibility3.3
83      #region Backwards compatible code, remove with 3.4
84      if (!Parameters.ContainsKey(WeightsParameterName)) {
85        Parameters.Add(new OptionalValueParameter<DoubleArray>(WeightsParameterName, "Optional: use weights to specify individual scaling values for all features. If not set the weights are calculated automatically (each feature is scaled to unit variance)"));
86      }
87      #endregion
88    }
89
90    public override IDeepCloneable Clone(Cloner cloner) {
91      return new NearestNeighbourRegression(this, cloner);
92    }
93
94    #region nearest neighbour
95    protected override void Run(CancellationToken cancellationToken) {
96      double[] weights = null;
97      if (Weights != null) weights = Weights.CloneAsArray();
98      var solution = CreateNearestNeighbourRegressionSolution(Problem.ProblemData, K, weights);
99      Results.Add(new Result(NearestNeighbourRegressionModelResultName, "The nearest neighbour regression solution.", solution));
100    }
101
102    public static IRegressionSolution CreateNearestNeighbourRegressionSolution(IRegressionProblemData problemData, int k, double[] weights = null) {
103      var clonedProblemData = (IRegressionProblemData)problemData.Clone();
104      return new NearestNeighbourRegressionSolution(Train(problemData, k, weights), clonedProblemData);
105    }
106
107    public static INearestNeighbourModel Train(IRegressionProblemData problemData, int k, double[] weights = null) {
108      return new NearestNeighbourModel(problemData.Dataset,
109        problemData.TrainingIndices,
110        k,
111        problemData.TargetVariable,
112        problemData.AllowedInputVariables,
113        weights);
114    }
115    #endregion
116  }
117}
Note: See TracBrowser for help on using the repository browser.