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.Text;
|
---|
26 | using System.Threading;
|
---|
27 | using HeuristicLab.Common;
|
---|
28 | using HeuristicLab.Core;
|
---|
29 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
30 | using HeuristicLab.Problems.DataAnalysis;
|
---|
31 |
|
---|
32 | namespace HeuristicLab.Algorithms.DataAnalysis {
|
---|
33 | [StorableClass]
|
---|
34 | internal class M5RuleModel : RegressionModel, IM5MetaModel {
|
---|
35 | internal const string NoCurrentLeafesResultName = "Number of current Leafs";
|
---|
36 |
|
---|
37 | #region Properties
|
---|
38 | [Storable]
|
---|
39 | internal string[] SplitAtts { get; private set; }
|
---|
40 | [Storable]
|
---|
41 | private double[] SplitVals { get; set; }
|
---|
42 | [Storable]
|
---|
43 | private RelOp[] RelOps { get; set; }
|
---|
44 | [Storable]
|
---|
45 | protected IRegressionModel RuleModel { get; set; }
|
---|
46 | [Storable]
|
---|
47 | private IReadOnlyList<string> Variables { get; set; }
|
---|
48 | #endregion
|
---|
49 |
|
---|
50 | #region HLConstructors
|
---|
51 | [StorableConstructor]
|
---|
52 | protected M5RuleModel(bool deserializing) : base(deserializing) { }
|
---|
53 | protected M5RuleModel(M5RuleModel original, Cloner cloner) : base(original, cloner) {
|
---|
54 | if (original.SplitAtts != null) SplitAtts = original.SplitAtts.ToArray();
|
---|
55 | if (original.SplitVals != null) SplitVals = original.SplitVals.ToArray();
|
---|
56 | if (original.RelOps != null) RelOps = original.RelOps.ToArray();
|
---|
57 | RuleModel = cloner.Clone(original.RuleModel);
|
---|
58 | if (original.Variables != null) Variables = original.Variables.ToList();
|
---|
59 | }
|
---|
60 | private M5RuleModel(string target) : base(target) { }
|
---|
61 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
62 | return new M5RuleModel(this, cloner);
|
---|
63 | }
|
---|
64 | #endregion
|
---|
65 |
|
---|
66 | internal static M5RuleModel CreateRuleModel(string target, M5CreationParameters m5CreationParams) {
|
---|
67 | return m5CreationParams.LeafType is ILeafType<IConfidenceRegressionModel> ? new ConfidenceM5RuleModel(target) : new M5RuleModel(target);
|
---|
68 | }
|
---|
69 |
|
---|
70 | #region IRegressionModel
|
---|
71 | public override IEnumerable<string> VariablesUsedForPrediction {
|
---|
72 | get { return Variables; }
|
---|
73 | }
|
---|
74 |
|
---|
75 | public override IEnumerable<double> GetEstimatedValues(IDataset dataset, IEnumerable<int> rows) {
|
---|
76 | if (RuleModel == null) throw new NotSupportedException("M5P has not been built correctly");
|
---|
77 | return RuleModel.GetEstimatedValues(dataset, rows);
|
---|
78 | }
|
---|
79 |
|
---|
80 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
81 | return new RegressionSolution(this, problemData);
|
---|
82 | }
|
---|
83 | #endregion
|
---|
84 |
|
---|
85 | #region IM5Component
|
---|
86 | public void BuildClassifier(IReadOnlyList<int> trainingRows, IReadOnlyList<int> holdoutRows, M5CreationParameters m5CreationParams, CancellationToken cancellation) {
|
---|
87 | Variables = m5CreationParams.AllowedInputVariables.ToList();
|
---|
88 | var tree = M5TreeModel.CreateTreeModel(m5CreationParams.TargetVariable, m5CreationParams);
|
---|
89 | ((IM5MetaModel) tree).BuildClassifier(trainingRows, holdoutRows, m5CreationParams, cancellation);
|
---|
90 | var nodeModel = tree.Root.EnumerateNodes().Where(x => x.IsLeaf).MaxItems(x => x.NumSamples).First();
|
---|
91 |
|
---|
92 | var satts = new List<string>();
|
---|
93 | var svals = new List<double>();
|
---|
94 | var reops = new List<RelOp>();
|
---|
95 |
|
---|
96 | //extract Splits
|
---|
97 | for (var temp = nodeModel; temp.Parent != null; temp = temp.Parent) {
|
---|
98 | satts.Add(temp.Parent.SplitAttr);
|
---|
99 | svals.Add(temp.Parent.SplitValue);
|
---|
100 | reops.Add(temp.Parent.Left == temp ? RelOp.Lessequal : RelOp.Greater);
|
---|
101 | }
|
---|
102 | nodeModel.ToRuleNode();
|
---|
103 | RuleModel = nodeModel.NodeModel;
|
---|
104 | RelOps = reops.ToArray();
|
---|
105 | SplitAtts = satts.ToArray();
|
---|
106 | SplitVals = svals.ToArray();
|
---|
107 | }
|
---|
108 |
|
---|
109 | public void UpdateModel(IReadOnlyList<int> rows, M5UpdateParameters m5UpdateParameters, CancellationToken cancellation) {
|
---|
110 | BuildModel(rows, m5UpdateParameters.Random, m5UpdateParameters.Data, m5UpdateParameters.LeafType, cancellation);
|
---|
111 | }
|
---|
112 | #endregion
|
---|
113 |
|
---|
114 | public bool Covers(IDataset dataset, int row) {
|
---|
115 | return !SplitAtts.Where((t, i) => !RelOps[i].Compare(dataset.GetDoubleValue(t, row), SplitVals[i])).Any();
|
---|
116 | }
|
---|
117 |
|
---|
118 | public string ToCompactString() {
|
---|
119 | var mins = new Dictionary<string, double>();
|
---|
120 | var maxs = new Dictionary<string, double>();
|
---|
121 | for (var i = 0; i < SplitAtts.Length; i++) {
|
---|
122 | var n = SplitAtts[i];
|
---|
123 | var v = SplitVals[i];
|
---|
124 | if (!mins.ContainsKey(n)) mins.Add(n, double.NegativeInfinity);
|
---|
125 | if (!maxs.ContainsKey(n)) maxs.Add(n, double.PositiveInfinity);
|
---|
126 | if (RelOps[i] == RelOp.Lessequal) maxs[n] = Math.Min(maxs[n], v);
|
---|
127 | else mins[n] = Math.Max(mins[n], v);
|
---|
128 | }
|
---|
129 | if (maxs.Count == 0) return "";
|
---|
130 | var s = new StringBuilder();
|
---|
131 | foreach (var key in maxs.Keys)
|
---|
132 | s.Append(string.Format("{0} ∈ [{1:e2}; {2:e2}] && ", key, mins[key], maxs[key]));
|
---|
133 | s.Remove(s.Length - 4, 4);
|
---|
134 | return s.ToString();
|
---|
135 | }
|
---|
136 |
|
---|
137 | #region Helpers
|
---|
138 | private void BuildModel(IReadOnlyList<int> rows, IRandom random, IDataset data, ILeafType<IRegressionModel> leafType, CancellationToken cancellation) {
|
---|
139 | var reducedData = new Dataset(VariablesUsedForPrediction.Concat(new[] {TargetVariable}), VariablesUsedForPrediction.Concat(new[] {TargetVariable}).Select(x => data.GetDoubleValues(x, rows).ToList()));
|
---|
140 | var pd = new RegressionProblemData(reducedData, VariablesUsedForPrediction, TargetVariable);
|
---|
141 | pd.TrainingPartition.Start = 0;
|
---|
142 | pd.TrainingPartition.End = pd.TestPartition.Start = pd.TestPartition.End = reducedData.Rows;
|
---|
143 |
|
---|
144 | int noparams;
|
---|
145 | RuleModel = leafType.BuildModel(pd, random, cancellation, out noparams);
|
---|
146 | cancellation.ThrowIfCancellationRequested();
|
---|
147 | }
|
---|
148 | #endregion
|
---|
149 |
|
---|
150 | [StorableClass]
|
---|
151 | private sealed class ConfidenceM5RuleModel : M5RuleModel, IConfidenceRegressionModel {
|
---|
152 | #region HLConstructors
|
---|
153 | [StorableConstructor]
|
---|
154 | private ConfidenceM5RuleModel(bool deserializing) : base(deserializing) { }
|
---|
155 | private ConfidenceM5RuleModel(ConfidenceM5RuleModel original, Cloner cloner) : base(original, cloner) { }
|
---|
156 | public ConfidenceM5RuleModel(string targetAttr) : base(targetAttr) { }
|
---|
157 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
158 | return new ConfidenceM5RuleModel(this, cloner);
|
---|
159 | }
|
---|
160 | #endregion
|
---|
161 |
|
---|
162 | public IEnumerable<double> GetEstimatedVariances(IDataset dataset, IEnumerable<int> rows) {
|
---|
163 | return ((IConfidenceRegressionModel) RuleModel).GetEstimatedVariances(dataset, rows);
|
---|
164 | }
|
---|
165 |
|
---|
166 | public override IRegressionSolution CreateRegressionSolution(IRegressionProblemData problemData) {
|
---|
167 | return new ConfidenceRegressionSolution(this, problemData);
|
---|
168 | }
|
---|
169 | }
|
---|
170 | }
|
---|
171 |
|
---|
172 | internal enum RelOp {
|
---|
173 | Lessequal,
|
---|
174 | Greater
|
---|
175 | }
|
---|
176 |
|
---|
177 | internal static class RelOpExtentions {
|
---|
178 | public static bool Compare(this RelOp op, double x, double y) {
|
---|
179 | switch (op) {
|
---|
180 | case RelOp.Greater:
|
---|
181 | return x > y;
|
---|
182 | case RelOp.Lessequal:
|
---|
183 | return x <= y;
|
---|
184 | default:
|
---|
185 | throw new ArgumentOutOfRangeException(op.ToString(), op, null);
|
---|
186 | }
|
---|
187 | }
|
---|
188 | }
|
---|
189 | } |
---|