[16955] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2018 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.IO;
|
---|
| 25 | using System.Linq;
|
---|
| 26 | using HeuristicLab.Analysis.FitnessLandscape;
|
---|
| 27 | using HeuristicLab.Core;
|
---|
| 28 | using HeuristicLab.Encodings.PermutationEncoding;
|
---|
| 29 | using HeuristicLab.Problems.Instances;
|
---|
| 30 | using HeuristicLab.Problems.Instances.QAPLIB;
|
---|
| 31 | using HeuristicLab.Random;
|
---|
| 32 | using ProtoBuf;
|
---|
| 33 |
|
---|
| 34 | namespace WalkExporter {
|
---|
| 35 | class RandomWalk {
|
---|
| 36 | public static (Knowledgebase training, Knowledgebase test) GetKnowledgeBases(Experiment experiment, int length) {
|
---|
| 37 | var training = new Knowledgebase() { Problems = new List<ProblemInstanceDescriptor>() };
|
---|
| 38 | var test = new Knowledgebase() { Problems = new List<ProblemInstanceDescriptor>() };
|
---|
| 39 | foreach (var trial in experiment.Trials) {
|
---|
| 40 | foreach (var desc in AnalyzeEachWalk(trial, length)) {
|
---|
| 41 | if (training.Problems.Count == test.Problems.Count)
|
---|
| 42 | training.Problems.Add(desc);
|
---|
| 43 | else test.Problems.Add(desc);
|
---|
| 44 | }
|
---|
| 45 | }
|
---|
| 46 | return (training, test);
|
---|
| 47 | }
|
---|
| 48 |
|
---|
| 49 | private static IEnumerable<ProblemInstanceDescriptor> AnalyzeEachWalk(Exploration trial, int length) {
|
---|
| 50 | var instance = trial.Problem;
|
---|
| 51 | var dim = trial.Dimension;
|
---|
| 52 | foreach (var walk in trial.Walks) {
|
---|
| 53 | var trail = walk.QualityTrail.Take(length).ToArray();
|
---|
| 54 | var clen = RuggednessCalculator.CalculateCorrelationLength(trail, out double[] acf);
|
---|
| 55 | var ia = new InformationAnalysis(trail);
|
---|
| 56 | var desc = new ProblemInstanceDescriptor() {
|
---|
| 57 | Name = instance,
|
---|
| 58 | Class = Util.GetGeneratorClass(instance),
|
---|
| 59 | Dimension = dim,
|
---|
| 60 | DescriptionEffort = length * 4.0 / dim,
|
---|
| 61 | };
|
---|
| 62 | desc.Features = new List<KeyValue>();
|
---|
| 63 | desc.Features.Add(new KeyValue { Key = "AC1", ContinuousValue = acf[1] });
|
---|
| 64 | desc.Features.Add(new KeyValue { Key = "CorrelationLength", DiscreteValue = clen });
|
---|
| 65 | foreach (var f in ia.GetFeatures()) {
|
---|
| 66 | if (f.Item2 is double d) desc.Features.Add(new KeyValue { Key = f.Item1, ContinuousValue = d });
|
---|
| 67 | else if (f.Item2 is int i) desc.Features.Add(new KeyValue { Key = f.Item1, DiscreteValue = i });
|
---|
| 68 | }
|
---|
| 69 | yield return desc;
|
---|
| 70 | }
|
---|
| 71 | }
|
---|
| 72 |
|
---|
| 73 | public static Experiment PerformExperiment() {
|
---|
| 74 | var experiment = new Experiment() { Algorithm = "RandomWalk", Trials = new List<Exploration>() };
|
---|
| 75 | foreach (var dimension in new[] { 20, 30, 40 }) {
|
---|
| 76 | var provider = new OneSizeInstanceProvider(dimension);
|
---|
| 77 | foreach (var desc in provider.GetDataDescriptors()) {
|
---|
| 78 | var qapData = provider.LoadData(desc);
|
---|
| 79 | var exploration = new Exploration() { Problem = qapData.Name, Dimension = qapData.Dimension, Walks = new List<Walk>() };
|
---|
| 80 | for (var r = 0; r < 2; r++) {
|
---|
| 81 | var walk = RandomWalk.Run(new MersenneTwister((uint)(r + 13)), qapData).Take((int)Math.Pow(2, 18)).ToList();
|
---|
| 82 | exploration.Walks.Add(new Walk() { QualityTrail = walk });
|
---|
| 83 | }
|
---|
| 84 | experiment.Trials.Add(exploration);
|
---|
| 85 | }
|
---|
| 86 | }
|
---|
| 87 | return experiment;
|
---|
| 88 | }
|
---|
| 89 |
|
---|
| 90 | public static IEnumerable<double> Run(IRandom random, QAPData qap) {
|
---|
| 91 | var sol = new Permutation(PermutationTypes.Absolute, qap.Dimension, random);
|
---|
| 92 | var fit = Util.Evaluate(sol, qap);
|
---|
| 93 | yield return fit;
|
---|
| 94 | while (true) {
|
---|
| 95 | var z1 = random.Next(qap.Dimension);
|
---|
| 96 | var z2 = (z1 + random.Next(1, qap.Dimension)) % qap.Dimension;
|
---|
| 97 | var move = Util.EvaluateSwap2Diff(sol, z1, z2, qap);
|
---|
| 98 | fit += move;
|
---|
| 99 | yield return fit;
|
---|
| 100 | sol.Swap(z1, z2);
|
---|
| 101 | }
|
---|
| 102 | }
|
---|
| 103 |
|
---|
| 104 | ///////////////////////////////////////////////////////////////////////////////////
|
---|
| 105 | /// CONFINED RANDOM WALK ///
|
---|
| 106 | ///////////////////////////////////////////////////////////////////////////////////
|
---|
| 107 |
|
---|
| 108 |
|
---|
| 109 | public static void ConfinedRandomWalkAnalysis(QAPData qapData) {
|
---|
| 110 | Exploration exploration = null;
|
---|
| 111 | if (File.Exists($"confinedrandwalk_{qapData.Name}.buf")) {
|
---|
| 112 | using (var file = File.OpenRead($"confinedrandwalk_{qapData.Name}.buf")) {
|
---|
| 113 | exploration = Serializer.Deserialize<Exploration>(file);
|
---|
| 114 | }
|
---|
| 115 | } else {
|
---|
| 116 | exploration = PerformCondinedRandomwWalkExploration(qapData, 100);
|
---|
| 117 | using (var file = File.Create($"confinedrandwalk_{qapData.Name}.buf")) {
|
---|
| 118 | Serializer.Serialize(file, exploration);
|
---|
| 119 | }
|
---|
| 120 | }
|
---|
| 121 |
|
---|
| 122 | using (var writer = File.CreateText($"confinedrandwalk_{qapData.Name}.csv")) {
|
---|
| 123 | var headers = new[] { "Run", "Algorithm Name", "Problem Name", "Dimension", "Ld(Iterations)", "Iterations", "Effort",
|
---|
| 124 | "AC1", "CorrelationLength", "InformationContent", "DensityBasinInformation", "PartialInformationContent",
|
---|
| 125 | "InformationStability", "Diversity", "Regularity", "TotalEntropy", "SymmetricInformationContent",
|
---|
| 126 | "SymmetricDensityBasinInformation", "SymmetricTotalEntropy", "PeakInformationContent", "PeakDensityBasinInformation",
|
---|
| 127 | "PeakTotalEntropy", "PeakSymmetricInformationContent", "PeakSymmetricDensityBasinInformation", "PeakSymmetricTotalEntropy" };
|
---|
| 128 | var order = headers.Select((v, i) => new { Index = i, Header = v }).ToDictionary(x => x.Header, x => x.Index);
|
---|
| 129 | writer.WriteLine(string.Format(string.Join(";", Enumerable.Range(0, headers.Length).Select(x => "{" + x + "}")), headers));
|
---|
| 130 | foreach (var exp in Enumerable.Range(7, 18 - 6)) {
|
---|
| 131 | var length = (int)Math.Pow(2, exp);
|
---|
| 132 | var run = 0;
|
---|
| 133 | foreach (var desc in AnalyzeEachWalk(exploration, length)) {
|
---|
| 134 | var features = string.Join(";", desc.Features.OrderBy(x => order[x.Key]).Select(x => x.GetNumericValue().ToString()));
|
---|
| 135 | writer.WriteLine(string.Format("R{0};Confined Random Walk;{5};{6};{1};{2};{3};{4}", run, exp, length, length * 4.0 / qapData.Dimension, features, qapData.Name, qapData.Dimension));
|
---|
| 136 | run++;
|
---|
| 137 | }
|
---|
| 138 | }
|
---|
| 139 | }
|
---|
| 140 | }
|
---|
| 141 |
|
---|
| 142 | private static Exploration PerformCondinedRandomwWalkExploration(QAPData qapData, int repetitions) {
|
---|
| 143 | var exploration = new Exploration() { Problem = qapData.Name, Dimension = qapData.Dimension, Walks = new List<Walk>() };
|
---|
| 144 | for (var r = 0; r < repetitions; r++) {
|
---|
| 145 | var walk = RunConfined(new MersenneTwister((uint)r), qapData, qapData.Dimension / 5).Take((int)Math.Pow(2, 18)).ToList();
|
---|
| 146 | exploration.Walks.Add(new Walk() { QualityTrail = walk });
|
---|
| 147 | }
|
---|
| 148 | return exploration;
|
---|
| 149 | }
|
---|
| 150 |
|
---|
| 151 | public static IEnumerable<double> RunConfined(IRandom random, QAPData qap, int distance) {
|
---|
| 152 | var sol = new Permutation(PermutationTypes.Absolute, qap.Dimension, random);
|
---|
| 153 | var anchor = (Permutation)sol.Clone();
|
---|
| 154 | var fit = Util.Evaluate(sol, qap);
|
---|
| 155 | var dist = 0;
|
---|
| 156 | yield return fit;
|
---|
| 157 |
|
---|
| 158 | while (true) {
|
---|
| 159 | var (j, k, deltaDist) = MoveConfined(random, sol, anchor, dist, distance);
|
---|
| 160 | dist += deltaDist;
|
---|
| 161 | var move = Util.EvaluateSwap2Diff(sol, j, k, qap);
|
---|
| 162 | fit += move;
|
---|
| 163 | yield return fit;
|
---|
| 164 | sol.Swap(j, k);
|
---|
| 165 | }
|
---|
| 166 | }
|
---|
| 167 |
|
---|
| 168 | private static (int j, int k, int deltaDist) MoveConfined(IRandom random, Permutation current, Permutation anchor, int dist, int maxDist) {
|
---|
| 169 | var evalSolPerMove = 4.0 / current.Length;
|
---|
| 170 | var orderJ = Enumerable.Range(0, current.Length).Shuffle(random);
|
---|
| 171 | var orderK = Enumerable.Range(0, current.Length).Shuffle(random);
|
---|
| 172 | foreach (var j in orderJ) {
|
---|
| 173 | if (dist == maxDist && current[j] == anchor[j]) continue;
|
---|
| 174 | foreach (var k in orderK) {
|
---|
| 175 | if (j == k) continue;
|
---|
| 176 | var distChange = 0;
|
---|
| 177 | if (current[j] != anchor[j] && current[k] == anchor[j]) distChange--;
|
---|
| 178 | if (current[k] != anchor[k] && current[j] == anchor[k]) distChange--;
|
---|
| 179 | if (current[j] == anchor[j]) distChange++;
|
---|
| 180 | if (current[k] == anchor[k]) distChange++;
|
---|
| 181 | if (dist + distChange > maxDist) continue;
|
---|
| 182 |
|
---|
| 183 | return (j, k, distChange);
|
---|
| 184 | }
|
---|
| 185 | }
|
---|
| 186 | return (-1, -1, 0);
|
---|
| 187 | }
|
---|
| 188 | }
|
---|
| 189 | }
|
---|