1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2017 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 |
|
---|
22 | using System;
|
---|
23 | using System.Collections.Generic;
|
---|
24 | using System.Linq;
|
---|
25 | using HeuristicLab.Common;
|
---|
26 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
27 | using HeuristicLab.Problems.DataAnalysis;
|
---|
28 |
|
---|
29 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
30 | //mulitdimensional extension of http://www2.stat.duke.edu/~tjl13/s101/slides/unit6lec3H.pdf
|
---|
31 | [StorableClass]
|
---|
32 | public class DampenedLinearModel : RegressionModel, IConfidenceRegressionModel {
|
---|
33 | [Storable]
|
---|
34 | private IConfidenceRegressionModel Model;
|
---|
35 | [Storable]
|
---|
36 | private double Min;
|
---|
37 | [Storable]
|
---|
38 | private double Max;
|
---|
39 | [Storable]
|
---|
40 | private double Dampening;
|
---|
41 |
|
---|
42 | [StorableConstructor]
|
---|
43 | private DampenedLinearModel(bool deserializing) : base(deserializing) { }
|
---|
44 | private DampenedLinearModel(DampenedLinearModel original, Cloner cloner) : base(original, cloner) {
|
---|
45 | Model = cloner.Clone(original.Model);
|
---|
46 | Min = original.Min;
|
---|
47 | Max = original.Max;
|
---|
48 | Dampening = original.Dampening;
|
---|
49 | }
|
---|
50 | public DampenedLinearModel(IConfidenceRegressionModel model, IRegressionProblemData pd, double dampening) : base(model.TargetVariable) {
|
---|
51 | Model = model;
|
---|
52 | Min = pd.TargetVariableTrainingValues.Min();
|
---|
53 | Max = pd.TargetVariableTrainingValues.Max();
|
---|
54 | Dampening = dampening;
|
---|
55 | }
|
---|
56 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
57 | return new DampenedLinearModel(this, cloner);
|
---|
58 | }
|
---|
59 | public override IEnumerable<string> VariablesUsedForPrediction {
|
---|
60 | get { return Model.VariablesUsedForPrediction; }
|
---|
61 | }
|
---|
62 | public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
|
---|
63 | var slow = Sigmoid(-Dampening);
|
---|
64 | var shigh = Sigmoid(Dampening);
|
---|
65 | foreach (var x in Model.GetEstimatedValues(dataset, rows)) {
|
---|
66 | var y = Rescale(x, Min, Max, -Dampening, Dampening);
|
---|
67 | y = Sigmoid(y);
|
---|
68 | y = Rescale(y, slow, shigh, Min, Max);
|
---|
69 | yield return y;
|
---|
70 | }
|
---|
71 | }
|
---|
72 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
73 | return new ConfidenceRegressionSolution(this, problemData);
|
---|
74 | }
|
---|
75 | public IEnumerable<double> GetEstimatedVariances(IDataset dataset, IEnumerable<int> rows) {
|
---|
76 | return Model.GetEstimatedVariances(dataset, rows);
|
---|
77 | }
|
---|
78 |
|
---|
79 | private static double Rescale(double x, double oMin, double oMax, double nMin, double nMax) {
|
---|
80 | var d = oMax - oMin;
|
---|
81 | var nd = nMax - nMin;
|
---|
82 | if (d.IsAlmost(0)) {
|
---|
83 | d = 1;
|
---|
84 | nMin += nd / 2;
|
---|
85 | nd = 0;
|
---|
86 | }
|
---|
87 | return ((x - oMin) / d) * nd + nMin;
|
---|
88 | }
|
---|
89 | private static double Sigmoid(double x) {
|
---|
90 | return 1 / (1 + Math.Exp(-x));
|
---|
91 | }
|
---|
92 | }
|
---|
93 | } |
---|