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 System.Threading;
|
---|
26 | using HeuristicLab.Common;
|
---|
27 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
28 | using HeuristicLab.Problems.DataAnalysis;
|
---|
29 |
|
---|
30 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
31 | [StorableClass]
|
---|
32 | public class RegressionNodeModel : RegressionModel {
|
---|
33 | #region Properties
|
---|
34 | public double PruningStrength = double.NaN;
|
---|
35 |
|
---|
36 | [Storable]
|
---|
37 | private IReadOnlyList<string> Variables {
|
---|
38 | get {
|
---|
39 | if (IsLeaf && Model == null) return new List<string>();
|
---|
40 | if (IsLeaf) return Model.VariablesUsedForPrediction.ToList();
|
---|
41 | var set = new HashSet<string> {SplitAttribute};
|
---|
42 | var vl = Left.Variables;
|
---|
43 | var vr = Right.Variables;
|
---|
44 | for (var i = 0; i < vl.Count; i++) set.Add(vl[i]);
|
---|
45 | for (var i = 0; i < vr.Count; i++) set.Add(vr[i]);
|
---|
46 | return set.ToList();
|
---|
47 | }
|
---|
48 | }
|
---|
49 | [Storable]
|
---|
50 | internal int NumSamples { get; private set; }
|
---|
51 | [Storable]
|
---|
52 | internal bool IsLeaf { get; private set; }
|
---|
53 | [Storable]
|
---|
54 | internal IRegressionModel Model { get; private set; }
|
---|
55 |
|
---|
56 | [Storable]
|
---|
57 | public string SplitAttribute { get; private set; }
|
---|
58 | [Storable]
|
---|
59 | public double SplitValue { get; private set; }
|
---|
60 | [Storable]
|
---|
61 | public RegressionNodeModel Left { get; private set; }
|
---|
62 | [Storable]
|
---|
63 | public RegressionNodeModel Right { get; private set; }
|
---|
64 | [Storable]
|
---|
65 | public RegressionNodeModel Parent { get; private set; }
|
---|
66 | #endregion
|
---|
67 |
|
---|
68 | #region HLConstructors
|
---|
69 | [StorableConstructor]
|
---|
70 | protected RegressionNodeModel(bool deserializing) : base(deserializing) { }
|
---|
71 | protected RegressionNodeModel(RegressionNodeModel original, Cloner cloner) : base(original, cloner) {
|
---|
72 | IsLeaf = original.IsLeaf;
|
---|
73 | Model = cloner.Clone(original.Model);
|
---|
74 | SplitValue = original.SplitValue;
|
---|
75 | SplitAttribute = original.SplitAttribute;
|
---|
76 | Left = cloner.Clone(original.Left);
|
---|
77 | Right = cloner.Clone(original.Right);
|
---|
78 | Parent = cloner.Clone(original.Parent);
|
---|
79 | NumSamples = original.NumSamples;
|
---|
80 | }
|
---|
81 | private RegressionNodeModel(string targetAttr) : base(targetAttr) {
|
---|
82 | IsLeaf = true;
|
---|
83 | }
|
---|
84 | private RegressionNodeModel(RegressionNodeModel parent) : this(parent.TargetVariable) {
|
---|
85 | Parent = parent;
|
---|
86 | IsLeaf = true;
|
---|
87 | }
|
---|
88 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
89 | return new RegressionNodeModel(this, cloner);
|
---|
90 | }
|
---|
91 | public static RegressionNodeModel CreateNode(string targetAttr, RegressionTreeParameters regressionTreeParams) {
|
---|
92 | return regressionTreeParams.LeafModel.ProvidesConfidence ? new ConfidenceRegressionNodeModel(targetAttr) : new RegressionNodeModel(targetAttr);
|
---|
93 | }
|
---|
94 | private static RegressionNodeModel CreateNode(RegressionNodeModel parent, RegressionTreeParameters regressionTreeParams) {
|
---|
95 | return regressionTreeParams.LeafModel.ProvidesConfidence ? new ConfidenceRegressionNodeModel(parent) : new RegressionNodeModel(parent);
|
---|
96 | }
|
---|
97 | #endregion
|
---|
98 |
|
---|
99 | #region RegressionModel
|
---|
100 | public override IEnumerable<string> VariablesUsedForPrediction {
|
---|
101 | get { return Variables; }
|
---|
102 | }
|
---|
103 | public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
|
---|
104 | if (!IsLeaf) return rows.Select(row => GetEstimatedValue(dataset, row));
|
---|
105 | if (Model == null) throw new NotSupportedException("The model has not been built correctly");
|
---|
106 | return Model.GetEstimatedValues(dataset, rows);
|
---|
107 | }
|
---|
108 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
109 | return new RegressionSolution(this, problemData);
|
---|
110 | }
|
---|
111 | #endregion
|
---|
112 |
|
---|
113 | internal void Split(RegressionTreeParameters regressionTreeParams, string splitAttribute, double splitValue, int numSamples) {
|
---|
114 | NumSamples = numSamples;
|
---|
115 | SplitAttribute = splitAttribute;
|
---|
116 | SplitValue = splitValue;
|
---|
117 | Left = CreateNode(this, regressionTreeParams);
|
---|
118 | Right = CreateNode(this, regressionTreeParams);
|
---|
119 | IsLeaf = false;
|
---|
120 | }
|
---|
121 |
|
---|
122 | internal void ToLeaf() {
|
---|
123 | IsLeaf = true;
|
---|
124 | Right = null;
|
---|
125 | Left = null;
|
---|
126 | }
|
---|
127 |
|
---|
128 | internal void SetLeafModel(IRegressionModel model) {
|
---|
129 | Model = model;
|
---|
130 | }
|
---|
131 |
|
---|
132 | internal IEnumerable<RegressionNodeModel> EnumerateNodes() {
|
---|
133 | var queue = new Queue<RegressionNodeModel>();
|
---|
134 | queue.Enqueue(this);
|
---|
135 | while (queue.Count != 0) {
|
---|
136 | var cur = queue.Dequeue();
|
---|
137 | yield return cur;
|
---|
138 | if (cur.Left == null && cur.Right == null) continue;
|
---|
139 | if (cur.Left != null) queue.Enqueue(cur.Left);
|
---|
140 | if (cur.Right != null) queue.Enqueue(cur.Right);
|
---|
141 | }
|
---|
142 | }
|
---|
143 |
|
---|
144 | #region Helpers
|
---|
145 | private double GetEstimatedValue(IDataset dataset, int row) {
|
---|
146 | if (!IsLeaf) return (dataset.GetDoubleValue(SplitAttribute, row) <= SplitValue ? Left : Right).GetEstimatedValue(dataset, row);
|
---|
147 | if (Model == null) throw new NotSupportedException("The model has not been built correctly");
|
---|
148 | return Model.GetEstimatedValues(dataset, new[] {row}).First();
|
---|
149 | }
|
---|
150 | #endregion
|
---|
151 |
|
---|
152 | [StorableClass]
|
---|
153 | private sealed class ConfidenceRegressionNodeModel : RegressionNodeModel, IConfidenceRegressionModel {
|
---|
154 | #region HLConstructors
|
---|
155 | [StorableConstructor]
|
---|
156 | private ConfidenceRegressionNodeModel(bool deserializing) : base(deserializing) { }
|
---|
157 | private ConfidenceRegressionNodeModel(ConfidenceRegressionNodeModel original, Cloner cloner) : base(original, cloner) { }
|
---|
158 | public ConfidenceRegressionNodeModel(string targetAttr) : base(targetAttr) { }
|
---|
159 | public ConfidenceRegressionNodeModel(RegressionNodeModel parent) : base(parent) { }
|
---|
160 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
161 | return new ConfidenceRegressionNodeModel(this, cloner);
|
---|
162 | }
|
---|
163 | #endregion
|
---|
164 |
|
---|
165 | public IEnumerable<double> GetEstimatedVariances(IDataset dataset, IEnumerable<int> rows) {
|
---|
166 | return IsLeaf ? ((IConfidenceRegressionModel)Model).GetEstimatedVariances(dataset, rows) : rows.Select(row => GetEstimatedVariance(dataset, row));
|
---|
167 | }
|
---|
168 |
|
---|
169 | private double GetEstimatedVariance(IDataset dataset, int row) {
|
---|
170 | if (!IsLeaf)
|
---|
171 | return ((IConfidenceRegressionModel)(dataset.GetDoubleValue(SplitAttribute, row) <= SplitValue ? Left : Right)).GetEstimatedVariances(dataset, row.ToEnumerable()).Single();
|
---|
172 | return ((IConfidenceRegressionModel)Model).GetEstimatedVariances(dataset, new[] {row}).First();
|
---|
173 | }
|
---|
174 |
|
---|
175 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
176 | return new ConfidenceRegressionSolution(this, problemData);
|
---|
177 | }
|
---|
178 | }
|
---|
179 | }
|
---|
180 | } |
---|