| 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.Collections.Generic;
|
|---|
| 23 | using System.Linq;
|
|---|
| 24 | using HeuristicLab.Analysis;
|
|---|
| 25 | using HeuristicLab.Core;
|
|---|
| 26 | using HeuristicLab.Data;
|
|---|
| 27 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
|---|
| 28 | using HeuristicLab.Operators;
|
|---|
| 29 | using HeuristicLab.Optimization;
|
|---|
| 30 | using HeuristicLab.Parameters;
|
|---|
| 31 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
|---|
| 32 | using HeuristicLab.Problems.DataAnalysis.MultiVariate.Regression.Symbolic.Evaluators;
|
|---|
| 33 | using HeuristicLab.Problems.DataAnalysis.Symbolic;
|
|---|
| 34 | using HeuristicLab.Problems.DataAnalysis.MultiVariate.Regression.Symbolic.Interfaces;
|
|---|
| 35 | using HeuristicLab.Common;
|
|---|
| 36 |
|
|---|
| 37 | namespace HeuristicLab.Problems.DataAnalysis.MultiVariate.Regression.Symbolic.Analyzers {
|
|---|
| 38 | /// <summary>
|
|---|
| 39 | /// An operator that analyzes the validation best scaled symbolic vector regression solution.
|
|---|
| 40 | /// </summary>
|
|---|
| 41 | [Item("ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer", "An operator that analyzes the validation best scaled symbolic vector regression solution.")]
|
|---|
| 42 | [StorableClass]
|
|---|
| 43 | public sealed class ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer : SingleSuccessorOperator, IAnalyzer {
|
|---|
| 44 | private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
|
|---|
| 45 | private const string ScaledSymbolicExpressionTreeParameterName = "ScaledSymbolicExpressionTree";
|
|---|
| 46 | private const string SymbolicExpressionTreeInterpreterParameterName = "SymbolicExpressionTreeInterpreter";
|
|---|
| 47 | private const string ProblemDataParameterName = "ProblemData";
|
|---|
| 48 | private const string ValidationSamplesStartParameterName = "ValidationSamplesStart";
|
|---|
| 49 | private const string ValidationSamplesEndParameterName = "ValidationSamplesEnd";
|
|---|
| 50 | private const string EvaluatorParameterName = "Evaluator";
|
|---|
| 51 | private const string MaximizationParameterName = "Maximization";
|
|---|
| 52 | private const string UpperEstimationLimitParameterName = "UpperEstimationLimit";
|
|---|
| 53 | private const string LowerEstimationLimitParameterName = "LowerEstimationLimit";
|
|---|
| 54 | private const string AlphaParameterName = "Alpha";
|
|---|
| 55 | private const string BetaParameterName = "Beta";
|
|---|
| 56 | private const string BestSolutionParameterName = "Best solution (validation)";
|
|---|
| 57 | private const string BestSolutionQualityParameterName = "Best solution quality (validation)";
|
|---|
| 58 | private const string CurrentBestValidationQualityParameterName = "Current best validation quality";
|
|---|
| 59 | private const string BestSolutionQualityValuesParameterName = "Validation Quality";
|
|---|
| 60 | private const string ResultsParameterName = "Results";
|
|---|
| 61 | private const string BestKnownQualityParameterName = "BestKnownQuality";
|
|---|
| 62 |
|
|---|
| 63 | #region parameter properties
|
|---|
| 64 | public ScopeTreeLookupParameter<SymbolicExpressionTree> SymbolicExpressionTreeParameter {
|
|---|
| 65 | get { return (ScopeTreeLookupParameter<SymbolicExpressionTree>)Parameters[SymbolicExpressionTreeParameterName]; }
|
|---|
| 66 | }
|
|---|
| 67 | public ScopeTreeLookupParameter<DoubleArray> AlphaParameter {
|
|---|
| 68 | get { return (ScopeTreeLookupParameter<DoubleArray>)Parameters[AlphaParameterName]; }
|
|---|
| 69 | }
|
|---|
| 70 | public ScopeTreeLookupParameter<DoubleArray> BetaParameter {
|
|---|
| 71 | get { return (ScopeTreeLookupParameter<DoubleArray>)Parameters[BetaParameterName]; }
|
|---|
| 72 | }
|
|---|
| 73 | public IValueLookupParameter<ISymbolicExpressionTreeInterpreter> SymbolicExpressionTreeInterpreterParameter {
|
|---|
| 74 | get { return (IValueLookupParameter<ISymbolicExpressionTreeInterpreter>)Parameters[SymbolicExpressionTreeInterpreterParameterName]; }
|
|---|
| 75 | }
|
|---|
| 76 | public IValueLookupParameter<MultiVariateDataAnalysisProblemData> ProblemDataParameter {
|
|---|
| 77 | get { return (IValueLookupParameter<MultiVariateDataAnalysisProblemData>)Parameters[ProblemDataParameterName]; }
|
|---|
| 78 | }
|
|---|
| 79 | public IValueLookupParameter<IntValue> ValidationSamplesStartParameter {
|
|---|
| 80 | get { return (IValueLookupParameter<IntValue>)Parameters[ValidationSamplesStartParameterName]; }
|
|---|
| 81 | }
|
|---|
| 82 | public IValueLookupParameter<IntValue> ValidationSamplesEndParameter {
|
|---|
| 83 | get { return (IValueLookupParameter<IntValue>)Parameters[ValidationSamplesEndParameterName]; }
|
|---|
| 84 | }
|
|---|
| 85 | public IValueLookupParameter<ISingleObjectiveSymbolicVectorRegressionEvaluator> EvaluatorParameter {
|
|---|
| 86 | get { return (IValueLookupParameter<ISingleObjectiveSymbolicVectorRegressionEvaluator>)Parameters[EvaluatorParameterName]; }
|
|---|
| 87 | }
|
|---|
| 88 | public IValueLookupParameter<BoolValue> MaximizationParameter {
|
|---|
| 89 | get { return (IValueLookupParameter<BoolValue>)Parameters[MaximizationParameterName]; }
|
|---|
| 90 | }
|
|---|
| 91 | public IValueLookupParameter<DoubleArray> UpperEstimationLimitParameter {
|
|---|
| 92 | get { return (IValueLookupParameter<DoubleArray>)Parameters[UpperEstimationLimitParameterName]; }
|
|---|
| 93 | }
|
|---|
| 94 | public IValueLookupParameter<DoubleArray> LowerEstimationLimitParameter {
|
|---|
| 95 | get { return (IValueLookupParameter<DoubleArray>)Parameters[LowerEstimationLimitParameterName]; }
|
|---|
| 96 | }
|
|---|
| 97 | public ILookupParameter<SymbolicExpressionTree> BestSolutionParameter {
|
|---|
| 98 | get { return (ILookupParameter<SymbolicExpressionTree>)Parameters[BestSolutionParameterName]; }
|
|---|
| 99 | }
|
|---|
| 100 | public ILookupParameter<DoubleValue> BestSolutionQualityParameter {
|
|---|
| 101 | get { return (ILookupParameter<DoubleValue>)Parameters[BestSolutionQualityParameterName]; }
|
|---|
| 102 | }
|
|---|
| 103 | public ILookupParameter<ResultCollection> ResultsParameter {
|
|---|
| 104 | get { return (ILookupParameter<ResultCollection>)Parameters[ResultsParameterName]; }
|
|---|
| 105 | }
|
|---|
| 106 | public ILookupParameter<DoubleValue> BestKnownQualityParameter {
|
|---|
| 107 | get { return (ILookupParameter<DoubleValue>)Parameters[BestKnownQualityParameterName]; }
|
|---|
| 108 | }
|
|---|
| 109 | #endregion
|
|---|
| 110 | #region properties
|
|---|
| 111 | public MultiVariateDataAnalysisProblemData ProblemData {
|
|---|
| 112 | get { return ProblemDataParameter.ActualValue; }
|
|---|
| 113 | }
|
|---|
| 114 | public ItemArray<DoubleArray> Alpha {
|
|---|
| 115 | get { return AlphaParameter.ActualValue; }
|
|---|
| 116 | }
|
|---|
| 117 | public ItemArray<DoubleArray> Beta {
|
|---|
| 118 | get { return BetaParameter.ActualValue; }
|
|---|
| 119 | }
|
|---|
| 120 | public DoubleArray LowerEstimationLimit {
|
|---|
| 121 | get { return LowerEstimationLimitParameter.ActualValue; }
|
|---|
| 122 | }
|
|---|
| 123 | public DoubleArray UpperEstimationLimit {
|
|---|
| 124 | get { return UpperEstimationLimitParameter.ActualValue; }
|
|---|
| 125 | }
|
|---|
| 126 | public ISingleObjectiveSymbolicVectorRegressionEvaluator Evaluator {
|
|---|
| 127 | get { return EvaluatorParameter.ActualValue; }
|
|---|
| 128 | }
|
|---|
| 129 | public BoolValue Maximization {
|
|---|
| 130 | get { return MaximizationParameter.ActualValue; }
|
|---|
| 131 | }
|
|---|
| 132 | public DoubleValue BestSolutionQuality {
|
|---|
| 133 | get { return BestSolutionQualityParameter.ActualValue; }
|
|---|
| 134 | }
|
|---|
| 135 | #endregion
|
|---|
| 136 | [StorableConstructor]
|
|---|
| 137 | protected ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer(bool deserializing) : base(deserializing) { }
|
|---|
| 138 | protected ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer(ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer original, Cloner cloner)
|
|---|
| 139 | : base(original, cloner) {
|
|---|
| 140 | }
|
|---|
| 141 | public ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer()
|
|---|
| 142 | : base() {
|
|---|
| 143 | Parameters.Add(new ScopeTreeLookupParameter<SymbolicExpressionTree>(SymbolicExpressionTreeParameterName, "The symbolic expression trees to analyze."));
|
|---|
| 144 | Parameters.Add(new ScopeTreeLookupParameter<DoubleArray>(AlphaParameterName, "The alpha parameter for linear scaling."));
|
|---|
| 145 | Parameters.Add(new ScopeTreeLookupParameter<DoubleArray>(BetaParameterName, "The beta parameter for linear scaling."));
|
|---|
| 146 | Parameters.Add(new ValueLookupParameter<ISymbolicExpressionTreeInterpreter>(SymbolicExpressionTreeInterpreterParameterName, "The interpreter that should be used for the analysis of symbolic expression trees."));
|
|---|
| 147 | Parameters.Add(new ValueLookupParameter<MultiVariateDataAnalysisProblemData>(ProblemDataParameterName, "The problem data for which the symbolic expression tree is a solution."));
|
|---|
| 148 | Parameters.Add(new ValueLookupParameter<IntValue>(ValidationSamplesStartParameterName, "The first index of the validation partition of the data set."));
|
|---|
| 149 | Parameters.Add(new ValueLookupParameter<IntValue>(ValidationSamplesEndParameterName, "The last index of the validation partition of the data set."));
|
|---|
| 150 | Parameters.Add(new ValueLookupParameter<ISingleObjectiveSymbolicVectorRegressionEvaluator>(EvaluatorParameterName, "The evaluator which should be used to evaluate the solution on the validation set."));
|
|---|
| 151 | Parameters.Add(new ValueLookupParameter<BoolValue>(MaximizationParameterName, "The direction of optimization."));
|
|---|
| 152 | Parameters.Add(new ValueLookupParameter<DoubleArray>(UpperEstimationLimitParameterName, "The upper estimation limit that was set for the evaluation of the symbolic expression trees."));
|
|---|
| 153 | Parameters.Add(new ValueLookupParameter<DoubleArray>(LowerEstimationLimitParameterName, "The lower estimation limit that was set for the evaluation of the symbolic expression trees."));
|
|---|
| 154 | Parameters.Add(new LookupParameter<SymbolicExpressionTree>(BestSolutionParameterName, "The best symbolic regression solution."));
|
|---|
| 155 | Parameters.Add(new LookupParameter<DoubleValue>(BestSolutionQualityParameterName, "The quality of the best symbolic regression solution."));
|
|---|
| 156 | Parameters.Add(new LookupParameter<ResultCollection>(ResultsParameterName, "The result collection where the best symbolic regression solution should be stored."));
|
|---|
| 157 | Parameters.Add(new LookupParameter<DoubleValue>(BestKnownQualityParameterName, "The best known (validation) quality achieved on the data set."));
|
|---|
| 158 | }
|
|---|
| 159 |
|
|---|
| 160 | public override IDeepCloneable Clone(Cloner cloner) {
|
|---|
| 161 | return new ValidationBestScaledSymbolicVectorRegressionSolutionAnalyzer(this, cloner);
|
|---|
| 162 | }
|
|---|
| 163 |
|
|---|
| 164 | public override IOperation Apply() {
|
|---|
| 165 | var trees = SymbolicExpressionTreeParameter.ActualValue;
|
|---|
| 166 | IEnumerable<SymbolicExpressionTree> scaledTrees;
|
|---|
| 167 | if (Alpha.Length == trees.Length) {
|
|---|
| 168 | scaledTrees = from i in Enumerable.Range(0, trees.Length)
|
|---|
| 169 | select SymbolicVectorRegressionSolutionLinearScaler.Scale(trees[i], Beta[i].ToArray(), Alpha[i].ToArray());
|
|---|
| 170 | } else {
|
|---|
| 171 | scaledTrees = trees;
|
|---|
| 172 | }
|
|---|
| 173 | IEnumerable<string> selectedTargetVariables = from item in ProblemData.TargetVariables.CheckedItems
|
|---|
| 174 | select item.Value.Value;
|
|---|
| 175 | var interpreter = SymbolicExpressionTreeInterpreterParameter.ActualValue;
|
|---|
| 176 | int validationStart = ValidationSamplesStartParameter.ActualValue.Value;
|
|---|
| 177 | int validationEnd = ValidationSamplesEndParameter.ActualValue.Value;
|
|---|
| 178 | IEnumerable<int> rows = Enumerable.Range(validationStart, validationEnd - validationStart);
|
|---|
| 179 | SymbolicExpressionTree bestTree = null;
|
|---|
| 180 | double bestQuality = Maximization.Value ? double.NegativeInfinity : double.PositiveInfinity;
|
|---|
| 181 | foreach (var tree in scaledTrees) {
|
|---|
| 182 | // calculate quality on validation set
|
|---|
| 183 | double quality = Evaluator.Evaluate(tree, interpreter, ProblemData, selectedTargetVariables, rows, LowerEstimationLimit, UpperEstimationLimit);
|
|---|
| 184 | if ((Maximization.Value && quality > bestQuality) ||
|
|---|
| 185 | (!Maximization.Value && quality < bestQuality)) {
|
|---|
| 186 | bestQuality = quality;
|
|---|
| 187 | bestTree = tree;
|
|---|
| 188 | }
|
|---|
| 189 | }
|
|---|
| 190 | bool newBest =
|
|---|
| 191 | BestSolutionQualityParameter.ActualValue == null ||
|
|---|
| 192 | (Maximization.Value && bestQuality > BestSolutionQuality.Value) ||
|
|---|
| 193 | (!Maximization.Value && bestQuality < BestSolutionQuality.Value);
|
|---|
| 194 | if (newBest) {
|
|---|
| 195 | var bestSolution = bestTree;
|
|---|
| 196 |
|
|---|
| 197 | //bestSolution.Name = BestSolutionParameterName;
|
|---|
| 198 | //solution.Description = "Best solution on validation partition found over the whole run.";
|
|---|
| 199 |
|
|---|
| 200 | BestSolutionParameter.ActualValue = bestSolution;
|
|---|
| 201 | BestSolutionQualityParameter.ActualValue = new DoubleValue(bestQuality);
|
|---|
| 202 | }
|
|---|
| 203 |
|
|---|
| 204 | // update results
|
|---|
| 205 | var results = ResultsParameter.ActualValue;
|
|---|
| 206 | if (!results.ContainsKey(BestSolutionQualityValuesParameterName)) {
|
|---|
| 207 | results.Add(new Result(BestSolutionParameterName, BestSolutionParameter.ActualValue));
|
|---|
| 208 | results.Add(new Result(BestSolutionQualityValuesParameterName, new DataTable(BestSolutionQualityValuesParameterName, BestSolutionQualityValuesParameterName)));
|
|---|
| 209 | results.Add(new Result(BestSolutionQualityParameterName, new DoubleValue()));
|
|---|
| 210 | results.Add(new Result(CurrentBestValidationQualityParameterName, new DoubleValue()));
|
|---|
| 211 | }
|
|---|
| 212 | results[BestSolutionParameterName].Value = BestSolutionParameter.ActualValue;
|
|---|
| 213 | results[BestSolutionQualityParameterName].Value = new DoubleValue(BestSolutionQualityParameter.ActualValue.Value);
|
|---|
| 214 | results[CurrentBestValidationQualityParameterName].Value = new DoubleValue(bestQuality);
|
|---|
| 215 |
|
|---|
| 216 | DataTable validationValues = (DataTable)results[BestSolutionQualityValuesParameterName].Value;
|
|---|
| 217 | AddValue(validationValues, BestSolutionQualityParameter.ActualValue.Value, BestSolutionQualityParameterName, BestSolutionQualityParameterName);
|
|---|
| 218 | AddValue(validationValues, bestQuality, CurrentBestValidationQualityParameterName, CurrentBestValidationQualityParameterName);
|
|---|
| 219 |
|
|---|
| 220 | return base.Apply();
|
|---|
| 221 | }
|
|---|
| 222 |
|
|---|
| 223 | private static void AddValue(DataTable table, double data, string name, string description) {
|
|---|
| 224 | DataRow row;
|
|---|
| 225 | table.Rows.TryGetValue(name, out row);
|
|---|
| 226 | if (row == null) {
|
|---|
| 227 | row = new DataRow(name, description);
|
|---|
| 228 | row.Values.Add(data);
|
|---|
| 229 | table.Rows.Add(row);
|
|---|
| 230 | } else {
|
|---|
| 231 | row.Values.Add(data);
|
|---|
| 232 | }
|
|---|
| 233 | }
|
|---|
| 234 | }
|
|---|
| 235 | }
|
|---|
