1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2016 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.Linq;
|
---|
24 | using HeuristicLab.Common;
|
---|
25 | using HeuristicLab.Core;
|
---|
26 | using HeuristicLab.Data;
|
---|
27 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
---|
28 | using HeuristicLab.EvolutionTracking;
|
---|
29 | using HeuristicLab.Parameters;
|
---|
30 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
31 |
|
---|
32 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic {
|
---|
33 | [Item("UpdateQualityOperator", "Put the estimated values of the tree in the scope to be used by the phenotypic similarity calculator")]
|
---|
34 | [StorableClass]
|
---|
35 | public class UpdateQualityOperator : EvolutionTrackingOperator<ISymbolicExpressionTree> {
|
---|
36 | private const string ProblemDataParameterName = "ProblemData";
|
---|
37 | private const string InterpreterParameterName = "SymbolicExpressionTreeInterpreter";
|
---|
38 | private const string EstimationLimitsParameterName = "EstimationLimits";
|
---|
39 | private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
|
---|
40 | private const string ScaleEstimatedValuesParameterName = "ScaleEstimatedValues";
|
---|
41 |
|
---|
42 | public ILookupParameter<IRegressionProblemData> ProblemDataParameter {
|
---|
43 | get { return (ILookupParameter<IRegressionProblemData>)Parameters[ProblemDataParameterName]; }
|
---|
44 | }
|
---|
45 | public ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter> InterpreterParameter {
|
---|
46 | get { return (ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>)Parameters[InterpreterParameterName]; }
|
---|
47 | }
|
---|
48 | public ILookupParameter<DoubleLimit> EstimationLimitsParameter {
|
---|
49 | get { return (ILookupParameter<DoubleLimit>)Parameters[EstimationLimitsParameterName]; }
|
---|
50 | }
|
---|
51 | public ILookupParameter<ISymbolicExpressionTree> SymbolicExpressionTreeParameter {
|
---|
52 | get { return (ILookupParameter<ISymbolicExpressionTree>)Parameters[SymbolicExpressionTreeParameterName]; }
|
---|
53 | }
|
---|
54 | public ILookupParameter<BoolValue> ScaleEstimatedValuesParameter {
|
---|
55 | get { return (ILookupParameter<BoolValue>)Parameters[ScaleEstimatedValuesParameterName]; }
|
---|
56 | }
|
---|
57 |
|
---|
58 | public UpdateQualityOperator() {
|
---|
59 | Parameters.Add(new LookupParameter<IRegressionProblemData>(ProblemDataParameterName));
|
---|
60 | Parameters.Add(new LookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>(InterpreterParameterName));
|
---|
61 | Parameters.Add(new LookupParameter<DoubleLimit>(EstimationLimitsParameterName));
|
---|
62 | Parameters.Add(new LookupParameter<ISymbolicExpressionTree>(SymbolicExpressionTreeParameterName));
|
---|
63 | Parameters.Add(new LookupParameter<BoolValue>(ScaleEstimatedValuesParameterName));
|
---|
64 | }
|
---|
65 |
|
---|
66 | [StorableConstructor]
|
---|
67 | protected UpdateQualityOperator(bool deserializing) : base(deserializing) { }
|
---|
68 |
|
---|
69 | protected UpdateQualityOperator(UpdateQualityOperator original, Cloner cloner) : base(original, cloner) {
|
---|
70 | }
|
---|
71 |
|
---|
72 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
73 | return new UpdateQualityOperator(this, cloner);
|
---|
74 | }
|
---|
75 |
|
---|
76 | public override IOperation Apply() {
|
---|
77 | var tree = SymbolicExpressionTreeParameter.ActualValue;
|
---|
78 | FixParentLinks(tree);
|
---|
79 | var problemData = ProblemDataParameter.ActualValue;
|
---|
80 | var estimationLimits = EstimationLimitsParameter.ActualValue;
|
---|
81 | var interpreter = InterpreterParameter.ActualValue;
|
---|
82 |
|
---|
83 | var estimatedValues = interpreter.GetSymbolicExpressionTreeValues(tree, problemData.Dataset, problemData.TrainingIndices).ToArray();
|
---|
84 | var targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, problemData.TrainingIndices).ToArray();
|
---|
85 |
|
---|
86 | if (estimatedValues.Length != targetValues.Length)
|
---|
87 | throw new ArgumentException("Number of elements in target and estimated values enumeration do not match.");
|
---|
88 |
|
---|
89 | var linearScalingCalculator = new OnlineLinearScalingParameterCalculator();
|
---|
90 |
|
---|
91 | for (int i = 0; i < estimatedValues.Length; ++i) {
|
---|
92 | var estimated = estimatedValues[i];
|
---|
93 | var target = targetValues[i];
|
---|
94 | if (!double.IsNaN(estimated) && !double.IsInfinity(estimated))
|
---|
95 | linearScalingCalculator.Add(estimated, target);
|
---|
96 | }
|
---|
97 | double alpha = linearScalingCalculator.Alpha;
|
---|
98 | double beta = linearScalingCalculator.Beta;
|
---|
99 | if (linearScalingCalculator.ErrorState != OnlineCalculatorError.None) {
|
---|
100 | alpha = 0.0;
|
---|
101 | beta = 1.0;
|
---|
102 | }
|
---|
103 | var scaled = estimatedValues.Select(x => x * beta + alpha).LimitToRange(estimationLimits.Lower, estimationLimits.Upper).ToArray();
|
---|
104 | OnlineCalculatorError error;
|
---|
105 | var r = OnlinePearsonsRCalculator.Calculate(targetValues, scaled, out error);
|
---|
106 | if (error != OnlineCalculatorError.None) r = double.NaN;
|
---|
107 |
|
---|
108 | var r2 = r * r;
|
---|
109 |
|
---|
110 | var variables = ExecutionContext.Scope.Variables;
|
---|
111 |
|
---|
112 | ((DoubleValue)variables["Quality"].Value).Value = r2;
|
---|
113 | GenealogyGraph.GetByContent(tree).Quality = r2;
|
---|
114 |
|
---|
115 | var scaleEstimatedValues = ScaleEstimatedValuesParameter.ActualValue;
|
---|
116 | if (!scaleEstimatedValues.Value)
|
---|
117 | scaled = estimatedValues.LimitToRange(estimationLimits.Lower, estimationLimits.Upper).ToArray();
|
---|
118 |
|
---|
119 | if (variables.ContainsKey("EstimatedValues")) {
|
---|
120 | variables["EstimatedValues"].Value = new DoubleArray(scaled);
|
---|
121 | } else {
|
---|
122 | variables.Add(new Core.Variable("EstimatedValues", new DoubleArray(scaled)));
|
---|
123 | }
|
---|
124 | return base.Apply();
|
---|
125 | }
|
---|
126 |
|
---|
127 | private static void FixParentLinks(ISymbolicExpressionTree tree) {
|
---|
128 | foreach (var node in tree.IterateNodesPrefix().Where(x => x.SubtreeCount > 0)) {
|
---|
129 | foreach (var s in node.Subtrees) {
|
---|
130 | if (s.Parent != node)
|
---|
131 | s.Parent = node;
|
---|
132 | }
|
---|
133 | }
|
---|
134 | }
|
---|
135 | }
|
---|
136 | }
|
---|