1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2012 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.Collections;
|
---|
26 | using HeuristicLab.Common;
|
---|
27 | using HeuristicLab.Core;
|
---|
28 | using HeuristicLab.Data;
|
---|
29 | using HeuristicLab.Encodings.DecisionList;
|
---|
30 | using HeuristicLab.Optimization.Operators.LCS;
|
---|
31 | using HeuristicLab.Parameters;
|
---|
32 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
33 | using HeuristicLab.Problems.DataAnalysis;
|
---|
34 |
|
---|
35 | namespace HeuristicLab.Problems.DecisionListClassification {
|
---|
36 | [StorableClass]
|
---|
37 | [Item("DecisionListClassificationProblemData", "")]
|
---|
38 | public class DecisionListClassificationProblemData : ParameterizedNamedItem, IDecisionListClassificationProblemData {
|
---|
39 |
|
---|
40 | #region default data
|
---|
41 | public static string[] defaultVariableNames = new string[] { "a", "b", "c", "d", "e", "f", "g" };
|
---|
42 | public static double[,] defaultData = new double[,]{
|
---|
43 | {0,0,1,1,0,0,0},
|
---|
44 | {0,1,1,1,0,0,0},
|
---|
45 | {0,0,1,0,0,0,1},
|
---|
46 | {1,0,1,0,1,1,0}
|
---|
47 | };
|
---|
48 | #endregion
|
---|
49 |
|
---|
50 | #region parameter properites
|
---|
51 | public IFixedValueParameter<Dataset> DatasetParameter {
|
---|
52 | get { return (IFixedValueParameter<Dataset>)Parameters["Dataset"]; }
|
---|
53 | }
|
---|
54 | public IFixedValueParameter<ReadOnlyCheckedItemList<StringValue>> ConditionVariablesParameter {
|
---|
55 | get { return (IFixedValueParameter<ReadOnlyCheckedItemList<StringValue>>)Parameters["ConditionVariables"]; }
|
---|
56 | }
|
---|
57 | public IConstrainedValueParameter<StringValue> TargetVariableParameter {
|
---|
58 | get { return (IConstrainedValueParameter<StringValue>)Parameters["TargetVariable"]; }
|
---|
59 | }
|
---|
60 | public IFixedValueParameter<IntRange> TrainingPartitionParameter {
|
---|
61 | get { return (IFixedValueParameter<IntRange>)Parameters["TrainingPartition"]; }
|
---|
62 | }
|
---|
63 | public IFixedValueParameter<IntRange> TestPartitionParameter {
|
---|
64 | get { return (IFixedValueParameter<IntRange>)Parameters["TestPartition"]; }
|
---|
65 | }
|
---|
66 | public IValueParameter<Rule> SampleRuleParameter {
|
---|
67 | get { return (IValueParameter<Rule>)Parameters["SampleRule"]; }
|
---|
68 | }
|
---|
69 | public IFixedValueParameter<IntValue> MaxIntervalsParameter {
|
---|
70 | get { return (IFixedValueParameter<IntValue>)Parameters["MaxIntervals"]; }
|
---|
71 | }
|
---|
72 | #endregion
|
---|
73 |
|
---|
74 | #region properties
|
---|
75 | public Dataset Dataset {
|
---|
76 | get { return DatasetParameter.Value; }
|
---|
77 | }
|
---|
78 | public ICheckedItemList<StringValue> ConditionVariables {
|
---|
79 | get { return ConditionVariablesParameter.Value; }
|
---|
80 | }
|
---|
81 | public StringValue TargetVariable {
|
---|
82 | get { return TargetVariableParameter.Value; }
|
---|
83 | }
|
---|
84 | public IEnumerable<string> AllowedConditionVariables {
|
---|
85 | get { return ConditionVariables.CheckedItems.Select(x => x.Value.Value); }
|
---|
86 | }
|
---|
87 | //public IEnumerable<string> AllowedTargetVariables {
|
---|
88 | // get { return ActionVariables.CheckedItems.Select(x => x.Value.Value); }
|
---|
89 | //}
|
---|
90 | public IntRange TrainingPartition {
|
---|
91 | get { return TrainingPartitionParameter.Value; }
|
---|
92 | }
|
---|
93 | public IntRange TestPartition {
|
---|
94 | get { return TestPartitionParameter.Value; }
|
---|
95 | }
|
---|
96 | public IEnumerable<int> TrainingIndices {
|
---|
97 | get {
|
---|
98 | return Enumerable.Range(TrainingPartition.Start, Math.Max(0, TrainingPartition.End - TrainingPartition.Start))
|
---|
99 | .Where(IsTrainingSample);
|
---|
100 | }
|
---|
101 | }
|
---|
102 | public IEnumerable<int> TestIndices {
|
---|
103 | get {
|
---|
104 | return Enumerable.Range(TestPartition.Start, Math.Max(0, TestPartition.End - TestPartition.Start))
|
---|
105 | .Where(IsTestSample);
|
---|
106 | }
|
---|
107 | }
|
---|
108 | public bool IsTrainingSample(int index) {
|
---|
109 | return index >= 0 && index < Dataset.Rows &&
|
---|
110 | TrainingPartition.Start <= index && index < TrainingPartition.End &&
|
---|
111 | (index < TestPartition.Start || TestPartition.End <= index);
|
---|
112 | }
|
---|
113 | public bool IsTestSample(int index) {
|
---|
114 | return index >= 0 && index < Dataset.Rows &&
|
---|
115 | TestPartition.Start <= index && index < TestPartition.End;
|
---|
116 | }
|
---|
117 | public int Classes {
|
---|
118 | get { return SampleRuleParameter.Value.Action.Possibilities; }
|
---|
119 | }
|
---|
120 | #endregion
|
---|
121 |
|
---|
122 | [StorableHook(HookType.AfterDeserialization)]
|
---|
123 | private void AfterDeserialization() {
|
---|
124 | RegisterParameterEvents();
|
---|
125 | }
|
---|
126 | [StorableConstructor]
|
---|
127 | protected DecisionListClassificationProblemData(bool deserializing) : base(deserializing) { }
|
---|
128 | protected DecisionListClassificationProblemData(DecisionListClassificationProblemData original, Cloner cloner)
|
---|
129 | : base(original, cloner) {
|
---|
130 | RegisterParameterEvents();
|
---|
131 | }
|
---|
132 | public DecisionListClassificationProblemData(Dataset dataset, IEnumerable<string> allowedConditionVariables, string targetVariable) {
|
---|
133 | if (dataset == null) throw new ArgumentNullException("The dataset must not be null.");
|
---|
134 | if (allowedConditionVariables == null) throw new ArgumentNullException("The allowedActionVariables must not be null.");
|
---|
135 |
|
---|
136 | if (allowedConditionVariables.Except(dataset.DoubleVariables).Any())
|
---|
137 | throw new ArgumentException("All allowed condition variables must be present in the dataset and of type double.");
|
---|
138 |
|
---|
139 | //var validTargetVariableValues = CheckVariablesForPossibleTargetVariables(dataset).Select(x => new StringValue(x).AsReadOnly()).ToList();
|
---|
140 | var validTargetVariableValues = dataset.VariableNames.Select(x => new StringValue(x));
|
---|
141 | var target = validTargetVariableValues.Where(x => x.Value == targetVariable).DefaultIfEmpty(validTargetVariableValues.First()).First();
|
---|
142 |
|
---|
143 | var conditionVariables = new CheckedItemList<StringValue>(dataset.DoubleVariables.Select(x => new StringValue(x)));
|
---|
144 | foreach (StringValue x in conditionVariables) {
|
---|
145 | conditionVariables.SetItemCheckedState(x, allowedConditionVariables.Contains(x.Value));
|
---|
146 | }
|
---|
147 |
|
---|
148 | int trainingPartitionStart = 0;
|
---|
149 | int trainingPartitionEnd = dataset.Rows / 2;
|
---|
150 | int testPartitionStart = dataset.Rows / 2;
|
---|
151 | int testPartitionEnd = dataset.Rows;
|
---|
152 |
|
---|
153 | Parameters.Add(new FixedValueParameter<Dataset>("Dataset", "", dataset));
|
---|
154 | Parameters.Add(new ConstrainedValueParameter<StringValue>("TargetVariable", new ItemSet<StringValue>(validTargetVariableValues), target));
|
---|
155 | Parameters.Add(new FixedValueParameter<ReadOnlyCheckedItemList<StringValue>>("ConditionVariables", "", conditionVariables.AsReadOnly()));
|
---|
156 | Parameters.Add(new FixedValueParameter<IntRange>("TrainingPartition", "", new IntRange(trainingPartitionStart, trainingPartitionEnd)));
|
---|
157 | Parameters.Add(new FixedValueParameter<IntRange>("TestPartition", "", new IntRange(testPartitionStart, testPartitionEnd)));
|
---|
158 | Parameters.Add(new FixedValueParameter<IntValue>("MaxIntervals", "", new IntValue(5)));
|
---|
159 | Parameters.Add(new ValueParameter<Rule>("SampleRule", "", CreateSampleRule(dataset, conditionVariables.CheckedItems.Select(x => x.Value.Value), target.Value)));
|
---|
160 |
|
---|
161 | ((ValueParameter<Dataset>)DatasetParameter).ReactOnValueToStringChangedAndValueItemImageChanged = false;
|
---|
162 |
|
---|
163 | RegisterParameterEvents();
|
---|
164 | }
|
---|
165 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
166 | return new DecisionListClassificationProblemData(this, cloner);
|
---|
167 | }
|
---|
168 |
|
---|
169 | private Rule CreateSampleRule(Dataset dataset, IEnumerable<string> conditionVariables, string target) {
|
---|
170 | IEnumerable<HeuristicLab.Encodings.DecisionList.IVariable> condition = GetConditionOfDataSet(dataset, conditionVariables);
|
---|
171 | IAction action = GetActionOfDataset(dataset, target);
|
---|
172 | return new Rule(condition, action);
|
---|
173 | }
|
---|
174 |
|
---|
175 | private IAction GetActionOfDataset(DataAnalysis.Dataset dataset, string target) {
|
---|
176 | IAction action = null;
|
---|
177 | var variableValues = dataset.GetValues(target);
|
---|
178 | if (variableValues is List<string>) {
|
---|
179 | action = new StringAction(target, (variableValues as List<string>).Distinct().ToList());
|
---|
180 | } else if (variableValues is List<double>) {
|
---|
181 | var doubleValues = (variableValues as List<double>).Distinct();
|
---|
182 | if (doubleValues.All(x => x % 1 == 0)) {
|
---|
183 | action = new IntAction(target, doubleValues.Select(x => Convert.ToInt32(x)).ToList());
|
---|
184 | } else {
|
---|
185 | throw new ArgumentException("The target variable contains double values. Therefore it can't be used as target");
|
---|
186 | }
|
---|
187 | } else {
|
---|
188 | throw new ArgumentException("There is no matching variable type for the values in the dataset");
|
---|
189 | }
|
---|
190 | return action;
|
---|
191 | }
|
---|
192 |
|
---|
193 | private IEnumerable<Encodings.DecisionList.IVariable> GetConditionOfDataSet(DataAnalysis.Dataset dataset, IEnumerable<string> conditionVariables) {
|
---|
194 | var condition = new List<HeuristicLab.Encodings.DecisionList.IVariable>();
|
---|
195 | foreach (var variableName in conditionVariables) {
|
---|
196 | var variableValues = dataset.GetValues(variableName);
|
---|
197 | HeuristicLab.Encodings.DecisionList.IVariable variable;
|
---|
198 | if (variableValues is List<string>) {
|
---|
199 | variable = new StringVariable(variableName, (variableValues as List<string>).Distinct().ToList());
|
---|
200 | } else if (variableValues is List<double>) {
|
---|
201 | var doubleValues = (variableValues as List<double>).Distinct();
|
---|
202 | if (doubleValues.All(x => x % 1 == 0)) {
|
---|
203 | // ToList call is necessary, because otherwise it wouldn't be possible to serialize it
|
---|
204 | variable = new IntVariable(variableName, doubleValues.Select(x => Convert.ToInt32(x)).ToList());
|
---|
205 | } else {
|
---|
206 | variable = new DoubleVariable(variableName, MaxIntervalsParameter.Value.Value);
|
---|
207 | }
|
---|
208 | } else {
|
---|
209 | throw new ArgumentException("There is no matching variable type for the values in the dataset");
|
---|
210 | }
|
---|
211 | condition.Add(variable);
|
---|
212 | }
|
---|
213 | return condition;
|
---|
214 | }
|
---|
215 |
|
---|
216 | public IEnumerable<IGAssistInput> FetchInput(IEnumerable<int> rows) {
|
---|
217 | foreach (var row in rows) {
|
---|
218 | yield return FetchInput(row);
|
---|
219 | }
|
---|
220 | }
|
---|
221 |
|
---|
222 | protected IDictionary<int, DecisionListInput> fetchInputCache = new Dictionary<int, DecisionListInput>();
|
---|
223 | public IGAssistInput FetchInput(int row) {
|
---|
224 | if (!fetchInputCache.ContainsKey(row)) {
|
---|
225 | DecisionListInput input = new DecisionListInput();
|
---|
226 | var variableNames = SampleRuleParameter.Value.Variables.Keys.ToList();
|
---|
227 | variableNames.Add(SampleRuleParameter.Value.Action.VariableName);
|
---|
228 | foreach (var variableName in variableNames) {
|
---|
229 | input.InputDictionary.Add(variableName, Dataset.GetValue(row, variableName));
|
---|
230 | }
|
---|
231 | fetchInputCache.Add(row, input);
|
---|
232 | }
|
---|
233 | return fetchInputCache[row];
|
---|
234 | }
|
---|
235 |
|
---|
236 | public IEnumerable<IGAssistNiche> FetchAction(IEnumerable<int> rows) {
|
---|
237 | foreach (var row in rows) {
|
---|
238 | yield return FetchAction(row);
|
---|
239 | }
|
---|
240 | }
|
---|
241 | protected IDictionary<int, IAction> fetchActionCache = new Dictionary<int, IAction>();
|
---|
242 | public IGAssistNiche FetchAction(int row) {
|
---|
243 | if (!fetchActionCache.ContainsKey(row)) {
|
---|
244 | var action = (IAction)SampleRuleParameter.Value.Action.Clone();
|
---|
245 | action.SetTo(Dataset.GetValue(row, action.VariableName));
|
---|
246 | fetchActionCache.Add(row, action);
|
---|
247 | }
|
---|
248 | return fetchActionCache[row];
|
---|
249 | }
|
---|
250 |
|
---|
251 | protected IList<IGAssistNiche> possibleNiches;
|
---|
252 | public IEnumerable<IGAssistNiche> GetPossibleNiches() {
|
---|
253 | if (possibleNiches == null) {
|
---|
254 | possibleNiches = new List<IGAssistNiche>();
|
---|
255 | for (int i = 0; i < Dataset.Rows; i++) {
|
---|
256 | var action = FetchAction(i);
|
---|
257 | if (!possibleNiches.Any(x => x.SameNiche(action))) {
|
---|
258 | possibleNiches.Add(action);
|
---|
259 | }
|
---|
260 | }
|
---|
261 | }
|
---|
262 | return possibleNiches;
|
---|
263 | }
|
---|
264 |
|
---|
265 | public event EventHandler Changed;
|
---|
266 |
|
---|
267 | #region events
|
---|
268 | private void RegisterParameterEvents() {
|
---|
269 | ConditionVariablesParameter.ValueChanged += new EventHandler(VariablesChanged);
|
---|
270 | ConditionVariablesParameter.Value.CheckedItemsChanged += new CollectionItemsChangedEventHandler<IndexedItem<StringValue>>(VariablesChanged);
|
---|
271 | TargetVariableParameter.ValueChanged += new EventHandler(VariablesChanged);
|
---|
272 | }
|
---|
273 | private void DeregisterParameterEvents() {
|
---|
274 | TargetVariableParameter.ValueChanged += new EventHandler(VariablesChanged);
|
---|
275 | ConditionVariablesParameter.Value.CheckedItemsChanged += new CollectionItemsChangedEventHandler<IndexedItem<StringValue>>(VariablesChanged);
|
---|
276 | ConditionVariablesParameter.ValueChanged += new EventHandler(VariablesChanged);
|
---|
277 | }
|
---|
278 | private void Value_CheckedItemsChanged(object sender, CollectionItemsChangedEventArgs<IndexedItem<StringValue>> e) {
|
---|
279 | VariablesChanged();
|
---|
280 | }
|
---|
281 | private void VariablesChanged(object sender, EventArgs e) {
|
---|
282 | VariablesChanged();
|
---|
283 | }
|
---|
284 |
|
---|
285 | private void VariablesChanged() {
|
---|
286 | SampleRuleParameter.Value = CreateSampleRule(Dataset, AllowedConditionVariables, TargetVariable.Value);
|
---|
287 | }
|
---|
288 | #endregion
|
---|
289 |
|
---|
290 | #region IDataAnalysisProblemData Members
|
---|
291 | public bool IsEmpty {
|
---|
292 | get { return true; }
|
---|
293 | }
|
---|
294 | public ICheckedItemList<StringValue> InputVariables {
|
---|
295 | get { return ConditionVariables; }
|
---|
296 | }
|
---|
297 | public IEnumerable<string> AllowedInputVariables {
|
---|
298 | get { return AllowedConditionVariables; }
|
---|
299 | }
|
---|
300 | #endregion
|
---|
301 | }
|
---|
302 | }
|
---|