1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2012 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.Core;
|
---|
27 | using HeuristicLab.Parameters;
|
---|
28 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
29 | using HeuristicLab.Problems.Instances;
|
---|
30 |
|
---|
31 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
|
---|
32 | [Item("Symbolic Regression Problem (single objective)", "Represents a single objective symbolic regression problem.")]
|
---|
33 | [StorableClass]
|
---|
34 | [Creatable("Problems")]
|
---|
35 | public class SymbolicRegressionSingleObjectiveProblem : SymbolicDataAnalysisSingleObjectiveProblem<IRegressionProblemData, ISymbolicRegressionSingleObjectiveEvaluator, ISymbolicDataAnalysisSolutionCreator>, IRegressionProblem,
|
---|
36 | IProblemInstanceConsumer<RegressionData>, IProblemInstanceExporter<RegressionData> {
|
---|
37 | private const double PunishmentFactor = 10;
|
---|
38 | private const int InitialMaximumTreeDepth = 8;
|
---|
39 | private const int InitialMaximumTreeLength = 25;
|
---|
40 | private const string EstimationLimitsParameterName = "EstimationLimits";
|
---|
41 | private const string EstimationLimitsParameterDescription = "The limits for the estimated value that can be returned by the symbolic regression model.";
|
---|
42 |
|
---|
43 | #region parameter properties
|
---|
44 | public IFixedValueParameter<DoubleLimit> EstimationLimitsParameter {
|
---|
45 | get { return (IFixedValueParameter<DoubleLimit>)Parameters[EstimationLimitsParameterName]; }
|
---|
46 | }
|
---|
47 | #endregion
|
---|
48 | #region properties
|
---|
49 | public DoubleLimit EstimationLimits {
|
---|
50 | get { return EstimationLimitsParameter.Value; }
|
---|
51 | }
|
---|
52 | #endregion
|
---|
53 | [StorableConstructor]
|
---|
54 | protected SymbolicRegressionSingleObjectiveProblem(bool deserializing) : base(deserializing) { }
|
---|
55 | protected SymbolicRegressionSingleObjectiveProblem(SymbolicRegressionSingleObjectiveProblem original, Cloner cloner) : base(original, cloner) { }
|
---|
56 | public override IDeepCloneable Clone(Cloner cloner) { return new SymbolicRegressionSingleObjectiveProblem(this, cloner); }
|
---|
57 |
|
---|
58 | public SymbolicRegressionSingleObjectiveProblem()
|
---|
59 | : base(new RegressionProblemData(), new SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator(), new SymbolicDataAnalysisExpressionTreeCreator()) {
|
---|
60 | Parameters.Add(new FixedValueParameter<DoubleLimit>(EstimationLimitsParameterName, EstimationLimitsParameterDescription));
|
---|
61 |
|
---|
62 | EstimationLimitsParameter.Hidden = true;
|
---|
63 |
|
---|
64 | Maximization.Value = true;
|
---|
65 | MaximumSymbolicExpressionTreeDepth.Value = InitialMaximumTreeDepth;
|
---|
66 | MaximumSymbolicExpressionTreeLength.Value = InitialMaximumTreeLength;
|
---|
67 |
|
---|
68 | SymbolicExpressionTreeGrammarParameter.ValueChanged += (o, e) => ConfigureGrammarSymbols();
|
---|
69 |
|
---|
70 | ConfigureGrammarSymbols();
|
---|
71 | InitializeOperators();
|
---|
72 | UpdateEstimationLimits();
|
---|
73 | }
|
---|
74 |
|
---|
75 | private void ConfigureGrammarSymbols() {
|
---|
76 | var grammar = SymbolicExpressionTreeGrammar as TypeCoherentExpressionGrammar;
|
---|
77 | if (grammar != null) grammar.ConfigureAsDefaultRegressionGrammar();
|
---|
78 | }
|
---|
79 |
|
---|
80 | private void InitializeOperators() {
|
---|
81 | Operators.Add(new SymbolicRegressionSingleObjectiveTrainingBestSolutionAnalyzer());
|
---|
82 | Operators.Add(new SymbolicRegressionSingleObjectiveValidationBestSolutionAnalyzer());
|
---|
83 | Operators.Add(new SymbolicRegressionSingleObjectiveOverfittingAnalyzer());
|
---|
84 | Operators.Add(new SymbolicRegressionSingleObjectiveTrainingParetoBestSolutionAnalyzer());
|
---|
85 | Operators.Add(new SymbolicRegressionSingleObjectiveValidationParetoBestSolutionAnalyzer());
|
---|
86 |
|
---|
87 | ParameterizeOperators();
|
---|
88 | }
|
---|
89 |
|
---|
90 | private void UpdateEstimationLimits() {
|
---|
91 | if (ProblemData.TrainingIndizes.Any()) {
|
---|
92 | var targetValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndizes).ToList();
|
---|
93 | var mean = targetValues.Average();
|
---|
94 | var range = targetValues.Max() - targetValues.Min();
|
---|
95 | EstimationLimits.Upper = mean + PunishmentFactor * range;
|
---|
96 | EstimationLimits.Lower = mean - PunishmentFactor * range;
|
---|
97 | } else {
|
---|
98 | EstimationLimits.Upper = double.MaxValue;
|
---|
99 | EstimationLimits.Lower = double.MinValue;
|
---|
100 | }
|
---|
101 | }
|
---|
102 |
|
---|
103 | protected override void OnProblemDataChanged() {
|
---|
104 | base.OnProblemDataChanged();
|
---|
105 | UpdateEstimationLimits();
|
---|
106 | }
|
---|
107 |
|
---|
108 | protected override void ParameterizeOperators() {
|
---|
109 | base.ParameterizeOperators();
|
---|
110 | if (Parameters.ContainsKey(EstimationLimitsParameterName)) {
|
---|
111 | var operators = Parameters.OfType<IValueParameter>().Select(p => p.Value).OfType<IOperator>().Union(Operators);
|
---|
112 | foreach (var op in operators.OfType<ISymbolicDataAnalysisBoundedOperator>()) {
|
---|
113 | op.EstimationLimitsParameter.ActualName = EstimationLimitsParameter.Name;
|
---|
114 | }
|
---|
115 | }
|
---|
116 | }
|
---|
117 |
|
---|
118 | public override void ImportProblemDataFromFile(string fileName) {
|
---|
119 | RegressionProblemData problemData = RegressionProblemData.ImportFromFile(fileName);
|
---|
120 | ProblemData = problemData;
|
---|
121 | }
|
---|
122 |
|
---|
123 | public void Load(RegressionData data) {
|
---|
124 | Name = data.Name;
|
---|
125 | Description = data.Description;
|
---|
126 | Dataset dataset = new Dataset(data.InputVariables, data.Values);
|
---|
127 | ProblemData = new RegressionProblemData(dataset, data.AllowedInputVariables, data.TargetVariable);
|
---|
128 | ProblemData.TrainingPartition.Start = data.TrainingPartitionStart;
|
---|
129 | ProblemData.TrainingPartition.End = data.TrainingPartitionEnd;
|
---|
130 | ProblemData.TestPartition.Start = data.TestPartitionStart;
|
---|
131 | ProblemData.TestPartition.End = data.TestPartitionEnd;
|
---|
132 | OnReset();
|
---|
133 | }
|
---|
134 |
|
---|
135 | public RegressionData Export() {
|
---|
136 | if (!ProblemData.InputVariables.Count.Equals(ProblemData.Dataset.DoubleVariables.Count()))
|
---|
137 | throw new ArgumentException("Not all input variables are double variables! (Export only works with double variables)");
|
---|
138 |
|
---|
139 | RegressionData regData = new RegressionData();
|
---|
140 | regData.Name = Name;
|
---|
141 | regData.Description = Description;
|
---|
142 | regData.TargetVariable = ProblemData.TargetVariable;
|
---|
143 | regData.InputVariables = ProblemData.InputVariables.Select(x => x.Value).ToArray();
|
---|
144 | regData.AllowedInputVariables = ProblemData.AllowedInputVariables.ToArray();
|
---|
145 | regData.TrainingPartitionStart = ProblemData.TrainingPartition.Start;
|
---|
146 | regData.TrainingPartitionEnd = ProblemData.TrainingPartition.End;
|
---|
147 | regData.TestPartitionStart = ProblemData.TestPartition.Start;
|
---|
148 | regData.TestPartitionEnd = ProblemData.TestPartition.End;
|
---|
149 |
|
---|
150 | List<List<double>> data = new List<List<double>>();
|
---|
151 | foreach (var variable in ProblemData.Dataset.DoubleVariables) {
|
---|
152 | data.Add(ProblemData.Dataset.GetDoubleValues(variable).ToList());
|
---|
153 | }
|
---|
154 | regData.Values = Transformer.Transformation(data);
|
---|
155 |
|
---|
156 | return regData;
|
---|
157 | }
|
---|
158 | }
|
---|
159 | }
|
---|