[13481] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2015 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.Linq;
|
---|
| 23 | using HeuristicLab.Common;
|
---|
| 24 | using HeuristicLab.Core;
|
---|
| 25 | using HeuristicLab.Data;
|
---|
| 26 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
---|
| 27 | using HeuristicLab.EvolutionTracking;
|
---|
| 28 | using HeuristicLab.Optimization;
|
---|
| 29 | using HeuristicLab.Parameters;
|
---|
| 30 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
| 31 |
|
---|
| 32 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Tracking.Analyzers {
|
---|
| 33 | [Item("SymbolicDataAnalysisSchemaFrequencyAnalyzer", "An analyzer which counts schema frequencies in the population.")]
|
---|
| 34 | [StorableClass]
|
---|
| 35 | public class SymbolicDataAnalysisSchemaFrequencyAnalyzer : EvolutionTrackingAnalyzer<ISymbolicExpressionTree> {
|
---|
| 36 | private const string MinimumSchemaLengthParameterName = "MinimumSchemaLength";
|
---|
| 37 |
|
---|
| 38 | [Storable]
|
---|
| 39 | private readonly SymbolicExpressionTreePhenotypicSimilarityCalculator phenotypicSimilarityCalculator;
|
---|
| 40 |
|
---|
| 41 | [Storable]
|
---|
| 42 | private readonly SymbolicExpressionTreeBottomUpSimilarityCalculator genotypicSimilarityCalculator;
|
---|
| 43 |
|
---|
| 44 | [Storable]
|
---|
| 45 | private readonly ISymbolicExpressionTreeNodeEqualityComparer comparer;
|
---|
| 46 | private QueryMatch qm;
|
---|
| 47 |
|
---|
| 48 | public IFixedValueParameter<IntValue> MinimumSchemaLengthParameter {
|
---|
| 49 | get { return (IFixedValueParameter<IntValue>)Parameters[MinimumSchemaLengthParameterName]; }
|
---|
| 50 | }
|
---|
| 51 |
|
---|
| 52 | public SymbolicDataAnalysisSchemaFrequencyAnalyzer() {
|
---|
| 53 | comparer = new SymbolicExpressionTreeNodeEqualityComparer {
|
---|
| 54 | MatchConstantValues = false,
|
---|
| 55 | MatchVariableNames = true,
|
---|
| 56 | MatchVariableWeights = false
|
---|
| 57 | };
|
---|
| 58 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
| 59 | phenotypicSimilarityCalculator = new SymbolicExpressionTreePhenotypicSimilarityCalculator();
|
---|
| 60 | genotypicSimilarityCalculator = new SymbolicExpressionTreeBottomUpSimilarityCalculator { SolutionVariableName = "SymbolicExpressionTree" };
|
---|
| 61 | Parameters.Add(new FixedValueParameter<IntValue>(MinimumSchemaLengthParameterName, new IntValue(10)));
|
---|
| 62 | }
|
---|
| 63 |
|
---|
| 64 | protected SymbolicDataAnalysisSchemaFrequencyAnalyzer(SymbolicDataAnalysisSchemaFrequencyAnalyzer original,
|
---|
| 65 | Cloner cloner) : base(original, cloner) {
|
---|
| 66 | comparer = original.comparer;
|
---|
| 67 | phenotypicSimilarityCalculator = original.phenotypicSimilarityCalculator;
|
---|
| 68 | genotypicSimilarityCalculator = original.genotypicSimilarityCalculator;
|
---|
| 69 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
| 70 | }
|
---|
| 71 |
|
---|
| 72 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 73 | return new SymbolicDataAnalysisSchemaFrequencyAnalyzer(this, cloner);
|
---|
| 74 | }
|
---|
| 75 |
|
---|
| 76 | [StorableConstructor]
|
---|
| 77 | protected SymbolicDataAnalysisSchemaFrequencyAnalyzer(bool deserializing) : base(deserializing) { }
|
---|
| 78 |
|
---|
| 79 |
|
---|
| 80 | [StorableHook(HookType.AfterDeserialization)]
|
---|
| 81 | private void AfterDeserialization() {
|
---|
| 82 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | public override IOperation Apply() {
|
---|
| 86 | int updateInterval = UpdateIntervalParameter.Value.Value;
|
---|
| 87 | IntValue updateCounter = UpdateCounterParameter.ActualValue;
|
---|
| 88 | // if counter does not yet exist then initialize it with update interval
|
---|
| 89 | // to make sure the solutions are analyzed on the first application of this operator
|
---|
| 90 | if (updateCounter == null) {
|
---|
| 91 | updateCounter = new IntValue(updateInterval);
|
---|
| 92 | UpdateCounterParameter.ActualValue = updateCounter;
|
---|
| 93 | }
|
---|
| 94 | //analyze solutions only every 'updateInterval' times
|
---|
| 95 | if (updateCounter.Value != updateInterval) {
|
---|
| 96 | updateCounter.Value++;
|
---|
| 97 | return base.Apply();
|
---|
| 98 | }
|
---|
| 99 | updateCounter.Value = 1;
|
---|
| 100 |
|
---|
| 101 | if (PopulationGraph == null || Generation.Value == 0)
|
---|
| 102 | return base.Apply();
|
---|
| 103 |
|
---|
| 104 | var minimumSchemaLength = MinimumSchemaLengthParameter.Value.Value;
|
---|
| 105 | var vertices = PopulationGraph.GetByRank(Generation.Value).Cast<IGenealogyGraphNode<ISymbolicExpressionTree>>().ToList();
|
---|
| 106 | var formatter = new SymbolicExpressionTreeStringFormatter { Indent = false, AppendNewLines = false };
|
---|
| 107 |
|
---|
| 108 | var schemas = SchemaCreator.GenerateSchemas(vertices, minimumSchemaLength).ToList();
|
---|
| 109 | var trees = vertices.Select(x => x.Data).ToList();
|
---|
| 110 | var qualities = vertices.Select(x => x.Quality).ToList();
|
---|
| 111 | var scopes = ExecutionContext.Scope.SubScopes; // the scopes of all the individuals, needed because the tree evaluated values are stored in the scope and we use them for the phenotypic similarity
|
---|
| 112 |
|
---|
| 113 | var matrix = new DoubleMatrix(schemas.Count, 5) {
|
---|
| 114 | RowNames = schemas.Select(x => formatter.Format(x.Root.GetSubtree(0).GetSubtree(0))),
|
---|
| 115 | ColumnNames = new[] { "Avg. Length", "Avg. Quality", "Relative Frequency", "Avg. Phen. Sim.", "Avg. Gen. Sim" },
|
---|
| 116 | SortableView = true
|
---|
| 117 | };
|
---|
| 118 |
|
---|
| 119 | for (int i = 0; i < schemas.Count; ++i) {
|
---|
| 120 | var schema = schemas[i];
|
---|
| 121 | int count = 0;
|
---|
| 122 | double avgLength = 0;
|
---|
| 123 | double avgQuality = 0;
|
---|
| 124 | var matchingScopes = new ScopeList();
|
---|
| 125 | for (int j = 0; j < trees.Count; ++j) {
|
---|
| 126 | var tree = trees[j];
|
---|
| 127 | if (qm.Match(tree, schema)) {
|
---|
| 128 | count++;
|
---|
| 129 | avgLength += tree.Length;
|
---|
| 130 | avgQuality += qualities[j];
|
---|
| 131 | matchingScopes.Add(scopes[j]);
|
---|
| 132 | }
|
---|
| 133 | }
|
---|
| 134 | avgLength /= count;
|
---|
| 135 | avgQuality /= count;
|
---|
| 136 | double relativeFreq = (double)count / trees.Count;
|
---|
| 137 | double avgPhenotypicSimilarity = SchemaEvaluator.CalculateSimilarity(matchingScopes, phenotypicSimilarityCalculator, true, 4);
|
---|
| 138 | double avgGenotypicSimilarity = SchemaEvaluator.CalculateSimilarity(matchingScopes, genotypicSimilarityCalculator, true, 4);
|
---|
| 139 |
|
---|
| 140 | matrix[i, 0] = avgLength;
|
---|
| 141 | matrix[i, 1] = avgQuality;
|
---|
| 142 | matrix[i, 2] = relativeFreq;
|
---|
| 143 | matrix[i, 3] = avgPhenotypicSimilarity;
|
---|
| 144 | matrix[i, 4] = avgGenotypicSimilarity;
|
---|
| 145 | }
|
---|
| 146 |
|
---|
| 147 | if (Results.ContainsKey("Schema Frequencies")) {
|
---|
| 148 | var result = Results["Schema Frequencies"];
|
---|
| 149 | result.Value = matrix;
|
---|
| 150 | } else {
|
---|
| 151 | var result = new Result("Schema Frequencies", matrix);
|
---|
| 152 | Results.Add(result);
|
---|
| 153 | }
|
---|
| 154 |
|
---|
| 155 | return base.Apply();
|
---|
| 156 | }
|
---|
| 157 | }
|
---|
| 158 | }
|
---|