1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2010 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.Data;
|
---|
28 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
---|
29 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Analyzers;
|
---|
30 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Creators;
|
---|
31 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Interfaces;
|
---|
32 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Symbols;
|
---|
33 | using HeuristicLab.Parameters;
|
---|
34 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
35 | using HeuristicLab.PluginInfrastructure;
|
---|
36 | using HeuristicLab.Problems.DataAnalysis.Regression.Symbolic.Analyzers;
|
---|
37 | using HeuristicLab.Problems.DataAnalysis.Symbolic;
|
---|
38 | using HeuristicLab.Problems.DataAnalysis.Symbolic.Symbols;
|
---|
39 |
|
---|
40 | namespace HeuristicLab.Problems.DataAnalysis.Classification {
|
---|
41 | [Item("Classification Problem", "Represents a classfication problem.")]
|
---|
42 | [StorableClass]
|
---|
43 | [Creatable("Problems")]
|
---|
44 | public class SymbolicClassificationProblem : SingleObjectiveClassificationProblem<ISymbolicClassificationEvaluator, ISymbolicExpressionTreeCreator>, IStorableContent {
|
---|
45 | private const string SymbolicExpressionTreeInterpreterParameterName = "SymbolicExpressionTreeInterpreter";
|
---|
46 | private const string FunctionTreeGrammarParameterName = "FunctionTreeGrammar";
|
---|
47 | private const string MaxExpressionLengthParameterName = "MaxExpressionLength";
|
---|
48 | private const string MaxExpressionDepthParameterName = "MaxExpressionDepth";
|
---|
49 | private const string UpperEstimationLimitParameterName = "UpperEstimationLimit";
|
---|
50 | private const string LowerEstimationLimitParameterName = "LowerEstimationLimit";
|
---|
51 | private const string MaxFunctionDefiningBranchensParameterName = "MaxFunctionDefiningBranches";
|
---|
52 | private const string MaxFunctionArgumentsParameterName = "MaxFunctionArguments";
|
---|
53 |
|
---|
54 | #region properties
|
---|
55 | public string Filename { get; set; }
|
---|
56 |
|
---|
57 | public ISymbolicExpressionTreeInterpreter SymbolicExpressionTreeInterpreter {
|
---|
58 | get { return SymbolicExpressionTreeInterpreterParameter.Value; }
|
---|
59 | protected set { SymbolicExpressionTreeInterpreterParameter.Value = value; }
|
---|
60 | }
|
---|
61 | public IValueParameter<ISymbolicExpressionTreeInterpreter> SymbolicExpressionTreeInterpreterParameter {
|
---|
62 | get { return (IValueParameter<ISymbolicExpressionTreeInterpreter>)Parameters[SymbolicExpressionTreeInterpreterParameterName]; }
|
---|
63 | }
|
---|
64 |
|
---|
65 | public ISymbolicExpressionGrammar FunctionTreeGrammar {
|
---|
66 | get { return (ISymbolicExpressionGrammar)FunctionTreeGrammarParameter.Value; }
|
---|
67 | protected set { FunctionTreeGrammarParameter.Value = value; }
|
---|
68 | }
|
---|
69 | public IValueParameter<ISymbolicExpressionGrammar> FunctionTreeGrammarParameter {
|
---|
70 | get { return (IValueParameter<ISymbolicExpressionGrammar>)Parameters[FunctionTreeGrammarParameterName]; }
|
---|
71 | }
|
---|
72 |
|
---|
73 | public IntValue MaxExpressionLength {
|
---|
74 | get { return MaxExpressionLengthParameter.Value; }
|
---|
75 | protected set { MaxExpressionLengthParameter.Value = value; }
|
---|
76 | }
|
---|
77 | public IValueParameter<IntValue> MaxExpressionLengthParameter {
|
---|
78 | get { return (IValueParameter<IntValue>)Parameters[MaxExpressionLengthParameterName]; }
|
---|
79 | }
|
---|
80 |
|
---|
81 | public IntValue MaxExpressionDepth {
|
---|
82 | get { return MaxExpressionDepthParameter.Value; }
|
---|
83 | protected set { MaxExpressionDepthParameter.Value = value; }
|
---|
84 | }
|
---|
85 | public ValueParameter<IntValue> MaxExpressionDepthParameter {
|
---|
86 | get { return (ValueParameter<IntValue>)Parameters[MaxExpressionDepthParameterName]; }
|
---|
87 | }
|
---|
88 |
|
---|
89 | public DoubleValue UpperEstimationLimit {
|
---|
90 | get { return UpperEstimationLimitParameter.Value; }
|
---|
91 | protected set { UpperEstimationLimitParameter.Value = value; }
|
---|
92 | }
|
---|
93 | public IValueParameter<DoubleValue> UpperEstimationLimitParameter {
|
---|
94 | get { return (IValueParameter<DoubleValue>)Parameters[UpperEstimationLimitParameterName]; }
|
---|
95 | }
|
---|
96 |
|
---|
97 | public DoubleValue LowerEstimationLimit {
|
---|
98 | get { return LowerEstimationLimitParameter.Value; }
|
---|
99 | protected set { LowerEstimationLimitParameter.Value = value; }
|
---|
100 | }
|
---|
101 | public IValueParameter<DoubleValue> LowerEstimationLimitParameter {
|
---|
102 | get { return (IValueParameter<DoubleValue>)Parameters[LowerEstimationLimitParameterName]; }
|
---|
103 | }
|
---|
104 |
|
---|
105 | public IntValue MaxFunctionDefiningBranches {
|
---|
106 | get { return MaxFunctionDefiningBranchesParameter.Value; }
|
---|
107 | protected set { MaxFunctionDefiningBranchesParameter.Value = value; }
|
---|
108 | }
|
---|
109 | public IValueParameter<IntValue> MaxFunctionDefiningBranchesParameter {
|
---|
110 | get { return (IValueParameter<IntValue>)Parameters[MaxFunctionDefiningBranchensParameterName]; }
|
---|
111 | }
|
---|
112 |
|
---|
113 | public IntValue MaxFunctionArguments {
|
---|
114 | get { return MaxFunctionArgumentsParameter.Value; }
|
---|
115 | protected set { MaxFunctionArgumentsParameter.Value = value; }
|
---|
116 | }
|
---|
117 | public IValueParameter<IntValue> MaxFunctionArgumentsParameter {
|
---|
118 | get { return (IValueParameter<IntValue>)Parameters[MaxFunctionArgumentsParameterName]; }
|
---|
119 | }
|
---|
120 |
|
---|
121 | public DoubleValue PunishmentFactor {
|
---|
122 | get { return new DoubleValue(10.0); }
|
---|
123 | }
|
---|
124 | public IntValue TrainingSamplesStart { get { return new IntValue(ClassificationProblemData.TrainingSamplesStart.Value); } }
|
---|
125 | public IntValue TrainingSamplesEnd {
|
---|
126 | get { return new IntValue((ClassificationProblemData.TrainingSamplesStart.Value + ClassificationProblemData.TrainingSamplesEnd.Value) / 2); }
|
---|
127 | }
|
---|
128 | public IntValue ValidationSamplesStart { get { return TrainingSamplesEnd; } }
|
---|
129 | public IntValue ValidationSamplesEnd { get { return new IntValue(ClassificationProblemData.TrainingSamplesEnd.Value); } }
|
---|
130 | public IntValue TestSamplesStart { get { return ClassificationProblemData.TestSamplesStart; } }
|
---|
131 | public IntValue TestSamplesEnd { get { return ClassificationProblemData.TestSamplesEnd; } }
|
---|
132 | #endregion
|
---|
133 |
|
---|
134 | [StorableConstructor]
|
---|
135 | protected SymbolicClassificationProblem(bool deserializing) : base(deserializing) { }
|
---|
136 | public SymbolicClassificationProblem()
|
---|
137 | : base() {
|
---|
138 | Parameters.Add(new ValueParameter<ISymbolicExpressionTreeInterpreter>(SymbolicExpressionTreeInterpreterParameterName, "The interpreter that should be used to evaluate the symbolic expression tree."));
|
---|
139 | Parameters.Add(new ValueParameter<ISymbolicExpressionGrammar>(FunctionTreeGrammarParameterName, "The grammar that should be used for symbolic regression models."));
|
---|
140 | Parameters.Add(new ValueParameter<IntValue>(MaxExpressionLengthParameterName, "Maximal length of the symbolic expression."));
|
---|
141 | Parameters.Add(new ValueParameter<IntValue>(MaxExpressionDepthParameterName, "Maximal depth of the symbolic expression."));
|
---|
142 | Parameters.Add(new ValueParameter<DoubleValue>(UpperEstimationLimitParameterName, "The upper limit for the estimated value that can be returned by the symbolic regression model."));
|
---|
143 | Parameters.Add(new ValueParameter<DoubleValue>(LowerEstimationLimitParameterName, "The lower limit for the estimated value that can be returned by the symbolic regression model."));
|
---|
144 | Parameters.Add(new ValueParameter<IntValue>(MaxFunctionDefiningBranchensParameterName, "Maximal number of automatically defined functions."));
|
---|
145 | Parameters.Add(new ValueParameter<IntValue>(MaxFunctionArgumentsParameterName, "Maximal number of arguments of automatically defined functions."));
|
---|
146 |
|
---|
147 | SolutionCreator = new ProbabilisticTreeCreator();
|
---|
148 | Evaluator = new SymbolicClassifacitionMeanSquaredErrorEvaluator();
|
---|
149 | ParameterizeSolutionCreator();
|
---|
150 | Maximization = new BoolValue(false);
|
---|
151 | FunctionTreeGrammar = new GlobalSymbolicExpressionGrammar(new FullFunctionalExpressionGrammar());
|
---|
152 | SymbolicExpressionTreeInterpreter = new SimpleArithmeticExpressionInterpreter();
|
---|
153 | MaxExpressionLength = new IntValue(100);
|
---|
154 | MaxExpressionDepth = new IntValue(10);
|
---|
155 | MaxFunctionDefiningBranches = new IntValue(0);
|
---|
156 | MaxFunctionArguments = new IntValue(0);
|
---|
157 |
|
---|
158 | InitializeOperators();
|
---|
159 | RegisterParameterEvents();
|
---|
160 |
|
---|
161 | UpdateEstimationLimits();
|
---|
162 | ParameterizeEvaluator();
|
---|
163 | ParameterizeSolutionCreator();
|
---|
164 | ParameterizeGrammar();
|
---|
165 | ParameterizeOperators();
|
---|
166 | ParameterizeAnalyzers();
|
---|
167 | }
|
---|
168 |
|
---|
169 |
|
---|
170 | [StorableHook(HookType.AfterDeserialization)]
|
---|
171 | private void AfterDeserialization() {
|
---|
172 |
|
---|
173 | }
|
---|
174 |
|
---|
175 | private void RegisterParameterEvents() {
|
---|
176 | SolutionCreator.SymbolicExpressionTreeParameter.ActualNameChanged += new EventHandler(SolutionCreator_SymbolicExpressionTreeParameter_ActualNameChanged);
|
---|
177 | FunctionTreeGrammarParameter.ValueChanged += new EventHandler(FunctionTreeGrammarParameter_ValueChanged);
|
---|
178 |
|
---|
179 | MaxFunctionArgumentsParameter.ValueChanged += new EventHandler(ArchitectureParameter_ValueChanged);
|
---|
180 | MaxFunctionDefiningBranchesParameter.ValueChanged += new EventHandler(ArchitectureParameter_ValueChanged);
|
---|
181 | MaxFunctionArgumentsParameter.Value.ValueChanged += new EventHandler(ArchitectureParameterValue_ValueChanged);
|
---|
182 | MaxFunctionDefiningBranchesParameter.Value.ValueChanged += new EventHandler(ArchitectureParameterValue_ValueChanged);
|
---|
183 | }
|
---|
184 |
|
---|
185 | protected override void OnEvaluatorChanged() {
|
---|
186 | ParameterizeEvaluator();
|
---|
187 | ParameterizeAnalyzers();
|
---|
188 | base.OnEvaluatorChanged();
|
---|
189 | }
|
---|
190 |
|
---|
191 | protected override void OnSolutionCreatorChanged() {
|
---|
192 | ParameterizeSolutionCreator();
|
---|
193 | SolutionCreator.SymbolicExpressionTreeParameter.ActualNameChanged += new EventHandler(SolutionCreator_SymbolicExpressionTreeParameter_ActualNameChanged);
|
---|
194 | base.OnSolutionCreatorChanged();
|
---|
195 | }
|
---|
196 | private void SolutionCreator_SymbolicExpressionTreeParameter_ActualNameChanged(object sender, System.EventArgs e) {
|
---|
197 | ParameterizeEvaluator();
|
---|
198 | ParameterizeOperators();
|
---|
199 | ParameterizeAnalyzers();
|
---|
200 | }
|
---|
201 |
|
---|
202 | protected override void OnClassificationProblemDataChanged() {
|
---|
203 | ParameterizeAnalyzers();
|
---|
204 | ParameterizeGrammar();
|
---|
205 | ParameterizeEvaluator();
|
---|
206 | UpdateEstimationLimits();
|
---|
207 | base.OnClassificationProblemDataChanged();
|
---|
208 | }
|
---|
209 |
|
---|
210 | private void FunctionTreeGrammarParameter_ValueChanged(object sender, System.EventArgs e) {
|
---|
211 | if (!(FunctionTreeGrammar is GlobalSymbolicExpressionGrammar))
|
---|
212 | FunctionTreeGrammar = new GlobalSymbolicExpressionGrammar(FunctionTreeGrammar);
|
---|
213 | OnGrammarChanged();
|
---|
214 | }
|
---|
215 | protected virtual void OnGrammarChanged() {
|
---|
216 | ParameterizeGrammar();
|
---|
217 | }
|
---|
218 |
|
---|
219 | private void ArchitectureParameter_ValueChanged(object sender, EventArgs e) {
|
---|
220 | MaxFunctionArgumentsParameter.Value.ValueChanged += new EventHandler(ArchitectureParameterValue_ValueChanged);
|
---|
221 | MaxFunctionDefiningBranchesParameter.Value.ValueChanged += new EventHandler(ArchitectureParameterValue_ValueChanged);
|
---|
222 | OnArchitectureParameterChanged();
|
---|
223 | }
|
---|
224 | private void ArchitectureParameterValue_ValueChanged(object sender, EventArgs e) {
|
---|
225 | OnArchitectureParameterChanged();
|
---|
226 | }
|
---|
227 | protected virtual void OnArchitectureParameterChanged() {
|
---|
228 | ParameterizeGrammar();
|
---|
229 | }
|
---|
230 |
|
---|
231 | protected virtual void InitializeOperators() {
|
---|
232 | Operators.AddRange(ApplicationManager.Manager.GetInstances<ISymbolicExpressionTreeOperator>().OfType<IOperator>());
|
---|
233 | Operators.Add(new MinAverageMaxSymbolicExpressionTreeSizeAnalyzer());
|
---|
234 | Operators.Add(new SymbolicRegressionVariableFrequencyAnalyzer());
|
---|
235 | Operators.Add(new ValidationBestSymbolicClassificationSolutionAnalyzer());
|
---|
236 | }
|
---|
237 |
|
---|
238 | #region operator parameterization
|
---|
239 | private void UpdateEstimationLimits() {
|
---|
240 | if (TrainingSamplesStart.Value < TrainingSamplesEnd.Value &&
|
---|
241 | ClassificationProblemData.Dataset.VariableNames.Contains(ClassificationProblemData.TargetVariable.Value)) {
|
---|
242 | var targetValues = ClassificationProblemData.Dataset.GetVariableValues(ClassificationProblemData.TargetVariable.Value, TrainingSamplesStart.Value, TrainingSamplesEnd.Value);
|
---|
243 | var mean = targetValues.Average();
|
---|
244 | var range = targetValues.Max() - targetValues.Min();
|
---|
245 | UpperEstimationLimit = new DoubleValue(mean + PunishmentFactor.Value * range);
|
---|
246 | LowerEstimationLimit = new DoubleValue(mean - PunishmentFactor.Value * range);
|
---|
247 | }
|
---|
248 | }
|
---|
249 |
|
---|
250 | private void ParameterizeEvaluator() {
|
---|
251 | if (Evaluator != null) {
|
---|
252 | Evaluator.SymbolicExpressionTreeParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
253 | Evaluator.RegressionProblemDataParameter.ActualName = ClassificationProblemDataParameter.Name;
|
---|
254 | Evaluator.SamplesStartParameter.Value = TrainingSamplesStart;
|
---|
255 | Evaluator.SamplesEndParameter.Value = TrainingSamplesEnd;
|
---|
256 | }
|
---|
257 | }
|
---|
258 |
|
---|
259 | private void ParameterizeGrammar() {
|
---|
260 | List<LaggedVariable> laggedSymbols = FunctionTreeGrammar.Symbols.OfType<LaggedVariable>().ToList();
|
---|
261 | foreach (Symbol symbol in laggedSymbols)
|
---|
262 | FunctionTreeGrammar.RemoveSymbol(symbol);
|
---|
263 | foreach (var varSymbol in FunctionTreeGrammar.Symbols.OfType<HeuristicLab.Problems.DataAnalysis.Symbolic.Symbols.Variable>()) {
|
---|
264 | varSymbol.VariableNames = ClassificationProblemData.InputVariables.CheckedItems.Select(x => x.Value.Value);
|
---|
265 | }
|
---|
266 | var globalGrammar = FunctionTreeGrammar as GlobalSymbolicExpressionGrammar;
|
---|
267 | if (globalGrammar != null) {
|
---|
268 | globalGrammar.MaxFunctionArguments = MaxFunctionArguments.Value;
|
---|
269 | globalGrammar.MaxFunctionDefinitions = MaxFunctionDefiningBranches.Value;
|
---|
270 | }
|
---|
271 | }
|
---|
272 |
|
---|
273 | private void ParameterizeSolutionCreator() {
|
---|
274 | SolutionCreator.SymbolicExpressionGrammarParameter.ActualName = FunctionTreeGrammarParameter.Name;
|
---|
275 | SolutionCreator.MaxTreeHeightParameter.ActualName = MaxExpressionDepthParameter.Name;
|
---|
276 | SolutionCreator.MaxTreeSizeParameter.ActualName = MaxExpressionLengthParameter.Name;
|
---|
277 | SolutionCreator.MaxFunctionArgumentsParameter.ActualName = MaxFunctionArgumentsParameter.Name;
|
---|
278 | SolutionCreator.MaxFunctionDefinitionsParameter.ActualName = MaxFunctionDefiningBranchesParameter.Name;
|
---|
279 | }
|
---|
280 |
|
---|
281 | private void ParameterizeOperators() {
|
---|
282 | foreach (ISymbolicExpressionTreeOperator op in Operators.OfType<ISymbolicExpressionTreeOperator>()) {
|
---|
283 | op.MaxTreeHeightParameter.ActualName = MaxExpressionDepthParameter.Name;
|
---|
284 | op.MaxTreeSizeParameter.ActualName = MaxExpressionLengthParameter.Name;
|
---|
285 | op.SymbolicExpressionGrammarParameter.ActualName = FunctionTreeGrammarParameter.Name;
|
---|
286 | }
|
---|
287 | foreach (ISymbolicExpressionTreeCrossover op in Operators.OfType<ISymbolicExpressionTreeCrossover>()) {
|
---|
288 | op.ParentsParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
289 | op.ChildParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
290 | }
|
---|
291 | foreach (ISymbolicExpressionTreeManipulator op in Operators.OfType<ISymbolicExpressionTreeManipulator>()) {
|
---|
292 | op.SymbolicExpressionTreeParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
293 | }
|
---|
294 | foreach (ISymbolicExpressionTreeArchitectureManipulator op in Operators.OfType<ISymbolicExpressionTreeArchitectureManipulator>()) {
|
---|
295 | op.MaxFunctionArgumentsParameter.ActualName = MaxFunctionArgumentsParameter.Name;
|
---|
296 | op.MaxFunctionDefinitionsParameter.ActualName = MaxFunctionDefiningBranchesParameter.Name;
|
---|
297 | }
|
---|
298 | }
|
---|
299 |
|
---|
300 | private void ParameterizeAnalyzers() {
|
---|
301 | foreach (ISymbolicRegressionAnalyzer analyzer in Operators.OfType<ISymbolicRegressionAnalyzer>()) {
|
---|
302 | analyzer.SymbolicExpressionTreeParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
303 | var bestValidationSolutionAnalyzer = analyzer as ValidationBestSymbolicClassificationSolutionAnalyzer;
|
---|
304 | if (bestValidationSolutionAnalyzer != null) {
|
---|
305 | bestValidationSolutionAnalyzer.ClassificationProblemDataParameter.ActualName = ClassificationProblemDataParameter.Name;
|
---|
306 | bestValidationSolutionAnalyzer.UpperEstimationLimitParameter.ActualName = UpperEstimationLimitParameter.Name;
|
---|
307 | bestValidationSolutionAnalyzer.LowerEstimationLimitParameter.ActualName = LowerEstimationLimitParameter.Name;
|
---|
308 | bestValidationSolutionAnalyzer.SymbolicExpressionTreeInterpreterParameter.ActualName = SymbolicExpressionTreeInterpreterParameter.Name;
|
---|
309 | bestValidationSolutionAnalyzer.SymbolicExpressionTreeParameter.ActualName = SolutionCreator.SymbolicExpressionTreeParameter.ActualName;
|
---|
310 | bestValidationSolutionAnalyzer.ValidationSamplesStartParameter.Value = ValidationSamplesStart;
|
---|
311 | bestValidationSolutionAnalyzer.ValidationSamplesEndParameter.Value = ValidationSamplesEnd;
|
---|
312 | }
|
---|
313 | var varFreqAnalyzer = analyzer as SymbolicRegressionVariableFrequencyAnalyzer;
|
---|
314 | if (varFreqAnalyzer != null) {
|
---|
315 | varFreqAnalyzer.ProblemDataParameter.ActualName = ClassificationProblemDataParameter.Name;
|
---|
316 | }
|
---|
317 | }
|
---|
318 | }
|
---|
319 | #endregion
|
---|
320 | }
|
---|
321 | }
|
---|