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.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.Parameters;
|
---|
30 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
31 |
|
---|
32 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
|
---|
33 | [Item("Constant Optimization Evaluator", "Calculates Pearson R² of a symbolic regression solution and optimizes the constant used.")]
|
---|
34 | [StorableClass]
|
---|
35 | public class SymbolicRegressionConstantOptimizationEvaluator : SymbolicRegressionSingleObjectiveEvaluator {
|
---|
36 | private const string ConstantOptimizationIterationsParameterName = "ConstantOptimizationIterations";
|
---|
37 | private const string ConstantOptimizationImprovementParameterName = "ConstantOptimizationImprovement";
|
---|
38 | private const string ConstantOptimizationProbabilityParameterName = "ConstantOptimizationProbability";
|
---|
39 | private const string ConstantOptimizationRowsPercentageParameterName = "ConstantOptimizationRowsPercentage";
|
---|
40 | private const string UpdateConstantsInTreeParameterName = "UpdateConstantsInSymbolicExpressionTree";
|
---|
41 | private const string UpdateVariableWeightsParameterName = "Update Variable Weights";
|
---|
42 |
|
---|
43 | public IFixedValueParameter<IntValue> ConstantOptimizationIterationsParameter {
|
---|
44 | get { return (IFixedValueParameter<IntValue>)Parameters[ConstantOptimizationIterationsParameterName]; }
|
---|
45 | }
|
---|
46 | public IFixedValueParameter<DoubleValue> ConstantOptimizationImprovementParameter {
|
---|
47 | get { return (IFixedValueParameter<DoubleValue>)Parameters[ConstantOptimizationImprovementParameterName]; }
|
---|
48 | }
|
---|
49 | public IFixedValueParameter<PercentValue> ConstantOptimizationProbabilityParameter {
|
---|
50 | get { return (IFixedValueParameter<PercentValue>)Parameters[ConstantOptimizationProbabilityParameterName]; }
|
---|
51 | }
|
---|
52 | public IFixedValueParameter<PercentValue> ConstantOptimizationRowsPercentageParameter {
|
---|
53 | get { return (IFixedValueParameter<PercentValue>)Parameters[ConstantOptimizationRowsPercentageParameterName]; }
|
---|
54 | }
|
---|
55 | public IFixedValueParameter<BoolValue> UpdateConstantsInTreeParameter {
|
---|
56 | get { return (IFixedValueParameter<BoolValue>)Parameters[UpdateConstantsInTreeParameterName]; }
|
---|
57 | }
|
---|
58 | public IFixedValueParameter<BoolValue> UpdateVariableWeightsParameter {
|
---|
59 | get { return (IFixedValueParameter<BoolValue>)Parameters[UpdateVariableWeightsParameterName]; }
|
---|
60 | }
|
---|
61 |
|
---|
62 |
|
---|
63 | public IntValue ConstantOptimizationIterations {
|
---|
64 | get { return ConstantOptimizationIterationsParameter.Value; }
|
---|
65 | }
|
---|
66 | public DoubleValue ConstantOptimizationImprovement {
|
---|
67 | get { return ConstantOptimizationImprovementParameter.Value; }
|
---|
68 | }
|
---|
69 | public PercentValue ConstantOptimizationProbability {
|
---|
70 | get { return ConstantOptimizationProbabilityParameter.Value; }
|
---|
71 | }
|
---|
72 | public PercentValue ConstantOptimizationRowsPercentage {
|
---|
73 | get { return ConstantOptimizationRowsPercentageParameter.Value; }
|
---|
74 | }
|
---|
75 | public bool UpdateConstantsInTree {
|
---|
76 | get { return UpdateConstantsInTreeParameter.Value.Value; }
|
---|
77 | set { UpdateConstantsInTreeParameter.Value.Value = value; }
|
---|
78 | }
|
---|
79 |
|
---|
80 | public bool UpdateVariableWeights {
|
---|
81 | get { return UpdateVariableWeightsParameter.Value.Value; }
|
---|
82 | set { UpdateVariableWeightsParameter.Value.Value = value; }
|
---|
83 | }
|
---|
84 |
|
---|
85 | public override bool Maximization {
|
---|
86 | get { return true; }
|
---|
87 | }
|
---|
88 |
|
---|
89 | [StorableConstructor]
|
---|
90 | protected SymbolicRegressionConstantOptimizationEvaluator(bool deserializing) : base(deserializing) { }
|
---|
91 | protected SymbolicRegressionConstantOptimizationEvaluator(SymbolicRegressionConstantOptimizationEvaluator original, Cloner cloner)
|
---|
92 | : base(original, cloner) {
|
---|
93 | }
|
---|
94 | public SymbolicRegressionConstantOptimizationEvaluator()
|
---|
95 | : base() {
|
---|
96 | Parameters.Add(new FixedValueParameter<IntValue>(ConstantOptimizationIterationsParameterName, "Determines how many iterations should be calculated while optimizing the constant of a symbolic expression tree (0 indicates other or default stopping criterion).", new IntValue(10), true));
|
---|
97 | Parameters.Add(new FixedValueParameter<DoubleValue>(ConstantOptimizationImprovementParameterName, "Determines the relative improvement which must be achieved in the constant optimization to continue with it (0 indicates other or default stopping criterion).", new DoubleValue(0), true) { Hidden = true });
|
---|
98 | Parameters.Add(new FixedValueParameter<PercentValue>(ConstantOptimizationProbabilityParameterName, "Determines the probability that the constants are optimized", new PercentValue(1), true));
|
---|
99 | Parameters.Add(new FixedValueParameter<PercentValue>(ConstantOptimizationRowsPercentageParameterName, "Determines the percentage of the rows which should be used for constant optimization", new PercentValue(1), true));
|
---|
100 | Parameters.Add(new FixedValueParameter<BoolValue>(UpdateConstantsInTreeParameterName, "Determines if the constants in the tree should be overwritten by the optimized constants.", new BoolValue(true)) { Hidden = true });
|
---|
101 | Parameters.Add(new FixedValueParameter<BoolValue>(UpdateVariableWeightsParameterName, "Determines if the variable weights in the tree should be optimized.", new BoolValue(true)) { Hidden = true });
|
---|
102 | }
|
---|
103 |
|
---|
104 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
105 | return new SymbolicRegressionConstantOptimizationEvaluator(this, cloner);
|
---|
106 | }
|
---|
107 |
|
---|
108 | [StorableHook(HookType.AfterDeserialization)]
|
---|
109 | private void AfterDeserialization() {
|
---|
110 | if (!Parameters.ContainsKey(UpdateConstantsInTreeParameterName))
|
---|
111 | Parameters.Add(new FixedValueParameter<BoolValue>(UpdateConstantsInTreeParameterName, "Determines if the constants in the tree should be overwritten by the optimized constants.", new BoolValue(true)));
|
---|
112 | if (!Parameters.ContainsKey(UpdateVariableWeightsParameterName))
|
---|
113 | Parameters.Add(new FixedValueParameter<BoolValue>(UpdateVariableWeightsParameterName, "Determines if the variable weights in the tree should be optimized.", new BoolValue(true)));
|
---|
114 | }
|
---|
115 |
|
---|
116 | public override IOperation InstrumentedApply() {
|
---|
117 | var solution = SymbolicExpressionTreeParameter.ActualValue;
|
---|
118 | double quality;
|
---|
119 | if (RandomParameter.ActualValue.NextDouble() < ConstantOptimizationProbability.Value) {
|
---|
120 | IEnumerable<int> constantOptimizationRows = GenerateRowsToEvaluate(ConstantOptimizationRowsPercentage.Value);
|
---|
121 | quality = OptimizeConstants(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, ProblemDataParameter.ActualValue,
|
---|
122 | constantOptimizationRows, ApplyLinearScalingParameter.ActualValue.Value, ConstantOptimizationIterations.Value, updateVariableWeights: UpdateVariableWeights, lowerEstimationLimit: EstimationLimitsParameter.ActualValue.Lower, upperEstimationLimit: EstimationLimitsParameter.ActualValue.Upper, updateConstantsInTree: UpdateConstantsInTree);
|
---|
123 |
|
---|
124 | if (ConstantOptimizationRowsPercentage.Value != RelativeNumberOfEvaluatedSamplesParameter.ActualValue.Value) {
|
---|
125 | var evaluationRows = GenerateRowsToEvaluate();
|
---|
126 | quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows, ApplyLinearScalingParameter.ActualValue.Value);
|
---|
127 | }
|
---|
128 | } else {
|
---|
129 | var evaluationRows = GenerateRowsToEvaluate();
|
---|
130 | quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows, ApplyLinearScalingParameter.ActualValue.Value);
|
---|
131 | }
|
---|
132 | QualityParameter.ActualValue = new DoubleValue(quality);
|
---|
133 |
|
---|
134 | return base.InstrumentedApply();
|
---|
135 | }
|
---|
136 |
|
---|
137 | public override double Evaluate(IExecutionContext context, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows) {
|
---|
138 | SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
|
---|
139 | EstimationLimitsParameter.ExecutionContext = context;
|
---|
140 | ApplyLinearScalingParameter.ExecutionContext = context;
|
---|
141 |
|
---|
142 | // Pearson R² evaluator is used on purpose instead of the const-opt evaluator,
|
---|
143 | // because Evaluate() is used to get the quality of evolved models on
|
---|
144 | // different partitions of the dataset (e.g., best validation model)
|
---|
145 | double r2 = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
|
---|
146 |
|
---|
147 | SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
|
---|
148 | EstimationLimitsParameter.ExecutionContext = null;
|
---|
149 | ApplyLinearScalingParameter.ExecutionContext = null;
|
---|
150 |
|
---|
151 | return r2;
|
---|
152 | }
|
---|
153 |
|
---|
154 | public static double OptimizeConstants(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows, bool applyLinearScaling, int maxIterations, bool updateVariableWeights = true, double lowerEstimationLimit = double.MinValue, double upperEstimationLimit = double.MaxValue, bool updateConstantsInTree = true) {
|
---|
155 |
|
---|
156 | string[] variableNames;
|
---|
157 | int[] lags;
|
---|
158 | double[] constants;
|
---|
159 |
|
---|
160 | TreeToAutoDiffTermConverter.ParametricFunction func;
|
---|
161 | TreeToAutoDiffTermConverter.ParametricFunctionGradient func_grad;
|
---|
162 | if (!TreeToAutoDiffTermConverter.TryTransformToAutoDiff(tree, updateVariableWeights, out variableNames, out lags, out constants, out func, out func_grad))
|
---|
163 | throw new NotSupportedException("Could not optimize constants of symbolic expression tree due to not supported symbols used in the tree.");
|
---|
164 | if (variableNames.Length == 0) return 0.0;
|
---|
165 |
|
---|
166 | //extract inital constants
|
---|
167 | double[] c = new double[constants.Length + 2];
|
---|
168 | c[0] = 0.0;
|
---|
169 | c[1] = 1.0;
|
---|
170 | Array.Copy(constants, 0, c, 2, constants.Length);
|
---|
171 | double[] originalConstants = (double[])c.Clone();
|
---|
172 | double originalQuality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(interpreter, tree, lowerEstimationLimit, upperEstimationLimit, problemData, rows, applyLinearScaling);
|
---|
173 |
|
---|
174 | alglib.lsfitstate state;
|
---|
175 | alglib.lsfitreport rep;
|
---|
176 | int info;
|
---|
177 |
|
---|
178 | // TODO: refactor
|
---|
179 | IDataset ds = problemData.Dataset;
|
---|
180 | double[,] x = new double[rows.Count(), variableNames.Length];
|
---|
181 | int row = 0;
|
---|
182 | foreach (var r in rows) {
|
---|
183 | for (int col = 0; col < variableNames.Length; col++) {
|
---|
184 | int lag = lags[col];
|
---|
185 | x[row, col] = ds.GetDoubleValue(variableNames[col], r + lag);
|
---|
186 | }
|
---|
187 | row++;
|
---|
188 | }
|
---|
189 | double[] y = ds.GetDoubleValues(problemData.TargetVariable, rows).ToArray();
|
---|
190 | int n = x.GetLength(0);
|
---|
191 | int m = x.GetLength(1);
|
---|
192 | int k = c.Length;
|
---|
193 |
|
---|
194 | alglib.ndimensional_pfunc function_cx_1_func = CreatePFunc(func);
|
---|
195 | alglib.ndimensional_pgrad function_cx_1_grad = CreatePGrad(func_grad);
|
---|
196 |
|
---|
197 | try {
|
---|
198 | alglib.lsfitcreatefg(x, y, c, n, m, k, false, out state);
|
---|
199 | alglib.lsfitsetcond(state, 0.0, 0.0, maxIterations);
|
---|
200 | //alglib.lsfitsetgradientcheck(state, 0.001);
|
---|
201 | alglib.lsfitfit(state, function_cx_1_func, function_cx_1_grad, null, null);
|
---|
202 | alglib.lsfitresults(state, out info, out c, out rep);
|
---|
203 | } catch (ArithmeticException) {
|
---|
204 | return originalQuality;
|
---|
205 | } catch (alglib.alglibexception) {
|
---|
206 | return originalQuality;
|
---|
207 | }
|
---|
208 |
|
---|
209 | //info == -7 => constant optimization failed due to wrong gradient
|
---|
210 | if (info != -7) UpdateConstants(tree, c.Skip(2).ToArray(), updateVariableWeights);
|
---|
211 | var quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(interpreter, tree, lowerEstimationLimit, upperEstimationLimit, problemData, rows, applyLinearScaling);
|
---|
212 |
|
---|
213 | if (!updateConstantsInTree) UpdateConstants(tree, originalConstants.Skip(2).ToArray(), updateVariableWeights);
|
---|
214 | if (originalQuality - quality > 0.001 || double.IsNaN(quality)) {
|
---|
215 | UpdateConstants(tree, originalConstants.Skip(2).ToArray(), updateVariableWeights);
|
---|
216 | return originalQuality;
|
---|
217 | }
|
---|
218 | return quality;
|
---|
219 | }
|
---|
220 |
|
---|
221 | private static void UpdateConstants(ISymbolicExpressionTree tree, double[] constants, bool updateVariableWeights) {
|
---|
222 | int i = 0;
|
---|
223 | foreach (var node in tree.Root.IterateNodesPrefix().OfType<SymbolicExpressionTreeTerminalNode>()) {
|
---|
224 | ConstantTreeNode constantTreeNode = node as ConstantTreeNode;
|
---|
225 | VariableTreeNode variableTreeNode = node as VariableTreeNode;
|
---|
226 | if (constantTreeNode != null)
|
---|
227 | constantTreeNode.Value = constants[i++];
|
---|
228 | else if (updateVariableWeights && variableTreeNode != null)
|
---|
229 | variableTreeNode.Weight = constants[i++];
|
---|
230 | }
|
---|
231 | }
|
---|
232 |
|
---|
233 | private static alglib.ndimensional_pfunc CreatePFunc(TreeToAutoDiffTermConverter.ParametricFunction func) {
|
---|
234 | return (double[] c, double[] x, ref double fx, object o) => {
|
---|
235 | fx = func(c, x);
|
---|
236 | };
|
---|
237 | }
|
---|
238 |
|
---|
239 | private static alglib.ndimensional_pgrad CreatePGrad(TreeToAutoDiffTermConverter.ParametricFunctionGradient func_grad) {
|
---|
240 | return (double[] c, double[] x, ref double fx, double[] grad, object o) => {
|
---|
241 | var tupel = func_grad(c, x);
|
---|
242 | fx = tupel.Item2;
|
---|
243 | Array.Copy(tupel.Item1, grad, grad.Length);
|
---|
244 | };
|
---|
245 | }
|
---|
246 |
|
---|
247 | public static bool CanOptimizeConstants(ISymbolicExpressionTree tree) {
|
---|
248 | return TreeToAutoDiffTermConverter.IsCompatible(tree);
|
---|
249 | }
|
---|
250 | }
|
---|
251 | }
|
---|