1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 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 HEAL.Attic;
|
---|
25 | using HeuristicLab.Common;
|
---|
26 | using HeuristicLab.Core;
|
---|
27 | using HeuristicLab.Data;
|
---|
28 | using HeuristicLab.Encodings.IntegerVectorEncoding;
|
---|
29 | using HeuristicLab.Optimization;
|
---|
30 | using HeuristicLab.Parameters;
|
---|
31 | using HeuristicLab.Problems.Instances;
|
---|
32 | using HeuristicLab.Problems.Instances.Types;
|
---|
33 |
|
---|
34 | using DoubleVector = MathNet.Numerics.LinearAlgebra.Vector<double>;
|
---|
35 |
|
---|
36 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic.SegmentOptimization {
|
---|
37 | [Item("Segment Optimization Problem (SOP)", "")]
|
---|
38 | [Creatable(CreatableAttribute.Categories.CombinatorialProblems, Priority = 1200)]
|
---|
39 | [StorableType("64107939-34A7-4530-BFAB-8EA1C321BF6F")]
|
---|
40 | public class SegmentOptimizationProblem : SingleObjectiveBasicProblem<IntegerVectorEncoding>, IProblemInstanceConsumer<SOPData> {
|
---|
41 |
|
---|
42 | [StorableType("63243591-5A56-41A6-B079-122B83583993")]
|
---|
43 | public enum Aggregation {
|
---|
44 | Sum,
|
---|
45 | Mean,
|
---|
46 | StandardDeviation
|
---|
47 | }
|
---|
48 |
|
---|
49 | public override bool Maximization => false;
|
---|
50 |
|
---|
51 | [Storable]
|
---|
52 | private IValueParameter<DoubleMatrix> dataParameter;
|
---|
53 | public IValueParameter<DoubleMatrix> DataParameter {
|
---|
54 | get { return dataParameter; }
|
---|
55 | }
|
---|
56 | [Storable]
|
---|
57 | private IValueParameter<IntRange> knownBoundsParameter;
|
---|
58 | public IValueParameter<IntRange> KnownBoundsParameter {
|
---|
59 | get { return knownBoundsParameter; }
|
---|
60 | }
|
---|
61 | [Storable]
|
---|
62 | private IValueParameter<EnumValue<Aggregation>> aggregationParameter;
|
---|
63 | public IValueParameter<EnumValue<Aggregation>> AggregationParameter {
|
---|
64 | get { return aggregationParameter; }
|
---|
65 | }
|
---|
66 |
|
---|
67 | public SegmentOptimizationProblem() {
|
---|
68 | Encoding = new IntegerVectorEncoding("bounds");
|
---|
69 |
|
---|
70 | Parameters.Add(dataParameter = new ValueParameter<DoubleMatrix>("Data", ""));
|
---|
71 | Parameters.Add(knownBoundsParameter = new ValueParameter<IntRange>("Known Bounds", ""));
|
---|
72 | Parameters.Add(aggregationParameter = new ValueParameter<EnumValue<Aggregation>>("Aggregation Function", ""));
|
---|
73 |
|
---|
74 | RegisterEventHandlers();
|
---|
75 |
|
---|
76 | #region Default Instance
|
---|
77 | Load(new SOPData() {
|
---|
78 | Data = ToNdimArray(Enumerable.Range(1, 50).Select(x => (double)x * x).ToArray()),
|
---|
79 | Lower = 20, Upper = 30,
|
---|
80 | Aggregation = "mean"
|
---|
81 | });
|
---|
82 | #endregion
|
---|
83 | }
|
---|
84 | private SegmentOptimizationProblem(SegmentOptimizationProblem original, Cloner cloner)
|
---|
85 | : base(original, cloner) {
|
---|
86 | dataParameter = cloner.Clone(original.dataParameter);
|
---|
87 | knownBoundsParameter = cloner.Clone(original.knownBoundsParameter);
|
---|
88 | aggregationParameter = cloner.Clone(original.aggregationParameter);
|
---|
89 |
|
---|
90 | RegisterEventHandlers();
|
---|
91 | }
|
---|
92 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
93 | return new SegmentOptimizationProblem(this, cloner);
|
---|
94 | }
|
---|
95 |
|
---|
96 | [StorableConstructor]
|
---|
97 | private SegmentOptimizationProblem(StorableConstructorFlag _) : base(_) { }
|
---|
98 | [StorableHook(HookType.AfterDeserialization)]
|
---|
99 | private void AfterDeserialization() {
|
---|
100 | if (Parameters.ContainsKey("Data Vector") && Parameters["Data Vector"] is ValueParameter<DoubleArray> arrayParameter) {
|
---|
101 | Parameters.Remove(arrayParameter);
|
---|
102 | var array = arrayParameter.Value;
|
---|
103 | var matrix = new DoubleMatrix(1, array.Length);
|
---|
104 | for (int i = 0; i < array.Length; i++) matrix[0, i] = array[i];
|
---|
105 | Parameters.Add(dataParameter = new ValueParameter<DoubleMatrix>("Data", "", matrix));
|
---|
106 | }
|
---|
107 |
|
---|
108 | RegisterEventHandlers();
|
---|
109 | }
|
---|
110 |
|
---|
111 | private void RegisterEventHandlers() {
|
---|
112 | dataParameter.ValueChanged += DataChanged;
|
---|
113 | knownBoundsParameter.ValueChanged += KnownBoundsChanged;
|
---|
114 | aggregationParameter.Value.ValueChanged += AggregationFunctionChanged;
|
---|
115 | }
|
---|
116 | private void DataChanged(object sender, EventArgs eventArgs) {
|
---|
117 | Encoding.Bounds = new IntMatrix(new[,] { { 0, DataParameter.Value.Columns } });
|
---|
118 | }
|
---|
119 | private void KnownBoundsChanged(object sender, EventArgs e) {
|
---|
120 | }
|
---|
121 | private void AggregationFunctionChanged(object sender, EventArgs eventArgs) {
|
---|
122 | }
|
---|
123 |
|
---|
124 | public override double Evaluate(Individual individual, IRandom random) {
|
---|
125 | var data = DataParameter.Value;
|
---|
126 | var knownBounds = KnownBoundsParameter.Value;
|
---|
127 | var aggregation = aggregationParameter.Value.Value;
|
---|
128 |
|
---|
129 | var solution = individual.IntegerVector(Encoding.Name);
|
---|
130 | var bounds = new IntRange(solution.Min(), solution.Max());
|
---|
131 |
|
---|
132 | double target = BoundedAggregation(data, knownBounds, aggregation);
|
---|
133 | double prediction = BoundedAggregation(data, bounds, aggregation);
|
---|
134 |
|
---|
135 | return Math.Pow(target - prediction, 2);
|
---|
136 | }
|
---|
137 |
|
---|
138 | public override void Analyze(Individual[] individuals, double[] qualities, ResultCollection results, IRandom random) {
|
---|
139 | var orderedIndividuals = individuals.Zip(qualities, (i, q) => new { Individual = i, Quality = q }).OrderBy(z => z.Quality);
|
---|
140 | var best = Maximization ? orderedIndividuals.Last().Individual.IntegerVector(Encoding.Name) : orderedIndividuals.First().Individual.IntegerVector(Encoding.Name);
|
---|
141 |
|
---|
142 | var bounds = new IntRange(best.Min(), best.Max());
|
---|
143 |
|
---|
144 | var data = DataParameter.Value;
|
---|
145 | var knownBounds = KnownBoundsParameter.Value;
|
---|
146 | var aggregation = aggregationParameter.Value.Value;
|
---|
147 |
|
---|
148 | double target = BoundedAggregation(data, knownBounds, aggregation);
|
---|
149 | double prediction = BoundedAggregation(data, bounds, aggregation);
|
---|
150 | double diff = target - prediction;
|
---|
151 |
|
---|
152 | results.AddOrUpdateResult("Bounds", bounds);
|
---|
153 |
|
---|
154 | results.AddOrUpdateResult("AggValue Diff", new DoubleValue(diff));
|
---|
155 | results.AddOrUpdateResult("AggValue Squared Diff", new DoubleValue(Math.Pow(diff, 2)));
|
---|
156 |
|
---|
157 | results.AddOrUpdateResult("Lower Diff", new IntValue(knownBounds.Start - bounds.Start));
|
---|
158 | results.AddOrUpdateResult("Upper Diff", new IntValue(knownBounds.End - bounds.End));
|
---|
159 | results.AddOrUpdateResult("Length Diff", new IntValue(knownBounds.Size - bounds.Size));
|
---|
160 | }
|
---|
161 |
|
---|
162 | private static double BoundedAggregation(DoubleArray data, IntRange bounds, Aggregation aggregation) {
|
---|
163 | var matrix = new DoubleMatrix(1, data.Length);
|
---|
164 | for (int i = 0; i < data.Length; i++) matrix[0, i] = data[i];
|
---|
165 | return BoundedAggregation(matrix, bounds, aggregation);
|
---|
166 | }
|
---|
167 |
|
---|
168 | private static double BoundedAggregation(DoubleMatrix data, IntRange bounds, Aggregation aggregation) {
|
---|
169 | if (bounds.Size == 0) {
|
---|
170 | return 0;
|
---|
171 | }
|
---|
172 |
|
---|
173 | var resultValues = new double[data.Rows];
|
---|
174 | var array = new double[data.Columns];
|
---|
175 | for (int row = 0; row < data.Rows; row++) {
|
---|
176 | for (int i = 0; i < array.Length; i++) array[i] = data[row, i];
|
---|
177 |
|
---|
178 | var vector = DoubleVector.Build.DenseOfArray(array);
|
---|
179 | var segment = vector.SubVector(bounds.Start, bounds.Size);
|
---|
180 |
|
---|
181 | switch (aggregation) {
|
---|
182 | case Aggregation.Sum:
|
---|
183 | resultValues[row] = segment.Sum();
|
---|
184 | break;
|
---|
185 | case Aggregation.Mean:
|
---|
186 | resultValues[row] = segment.Average();
|
---|
187 | break;
|
---|
188 | case Aggregation.StandardDeviation:
|
---|
189 | resultValues[row] = segment.StandardDeviationPop();
|
---|
190 | break;
|
---|
191 | default:
|
---|
192 | throw new NotImplementedException();
|
---|
193 | }
|
---|
194 | }
|
---|
195 |
|
---|
196 | return resultValues.Average();
|
---|
197 | }
|
---|
198 |
|
---|
199 | public void Load(SOPData data) {
|
---|
200 | DataParameter.Value = new DoubleMatrix(data.Data);
|
---|
201 | KnownBoundsParameter.Value = new IntRange(data.Lower, data.Upper);
|
---|
202 | switch (data.Aggregation.ToLower()) {
|
---|
203 | case "sum":
|
---|
204 | AggregationParameter.Value.Value = Aggregation.Sum;
|
---|
205 | break;
|
---|
206 | case "mean":
|
---|
207 | case "avg":
|
---|
208 | AggregationParameter.Value.Value = Aggregation.Mean;
|
---|
209 | break;
|
---|
210 | case "standarddeviation":
|
---|
211 | case "std":
|
---|
212 | case "sd":
|
---|
213 | AggregationParameter.Value.Value = Aggregation.StandardDeviation;
|
---|
214 | break;
|
---|
215 | default:
|
---|
216 | throw new NotSupportedException();
|
---|
217 | }
|
---|
218 |
|
---|
219 | Encoding.Length = 2;
|
---|
220 | Encoding.Bounds = new IntMatrix(new[,] { { 0, DataParameter.Value.Columns } });
|
---|
221 |
|
---|
222 | BestKnownQuality = 0;
|
---|
223 |
|
---|
224 | Name = data.Name;
|
---|
225 | Description = data.Description;
|
---|
226 | }
|
---|
227 |
|
---|
228 | public static T[,] ToNdimArray<T>(T[] array) {
|
---|
229 | var matrix = new T[1, array.Length];
|
---|
230 | for (int i = 0; i < array.Length; i++)
|
---|
231 | matrix[0, i] = array[i];
|
---|
232 | return matrix;
|
---|
233 | }
|
---|
234 | }
|
---|
235 | } |
---|