Free cookie consent management tool by TermsFeed Policy Generator

source: trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationProblemData.cs @ 8573

Last change on this file since 8573 was 8554, checked in by mkommend, 12 years ago

#1915:

  • Corrected class names and class values caching in ClassificationProblemData
  • Removed caching of classification penalties
  • Corrected AccuracyMaximizationThresholdCalculator (retrieving of penalties was switched)
  • Added asserts for the achieved accuracy to the classification sample unit test
File size: 18.7 KB
RevLine 
[5559]1#region License Information
2/* HeuristicLab
[7259]3 * Copyright (C) 2002-2012 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
[5559]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
22using System;
23using System.Collections.Generic;
24using System.Linq;
25using HeuristicLab.Common;
[5601]26using HeuristicLab.Core;
27using HeuristicLab.Data;
28using HeuristicLab.Parameters;
[5559]29using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
30
31namespace HeuristicLab.Problems.DataAnalysis {
32  [StorableClass]
[5601]33  [Item("ClassificationProblemData", "Represents an item containing all data defining a classification problem.")]
[7134]34  public class ClassificationProblemData : DataAnalysisProblemData, IClassificationProblemData, IStorableContent {
[6666]35    protected const string TargetVariableParameterName = "TargetVariable";
36    protected const string ClassNamesParameterName = "ClassNames";
37    protected const string ClassificationPenaltiesParameterName = "ClassificationPenalties";
[7266]38    protected const int MaximumNumberOfClasses = 100;
39    protected const int InspectedRowsToDetermineTargets = 2000;
[5601]40
[7134]41    public string Filename { get; set; }
42
[5559]43    #region default data
44    private static string[] defaultVariableNames = new string[] { "sample", "clump thickness", "cell size", "cell shape", "marginal adhesion", "epithelial cell size", "bare nuclei", "chromatin", "nucleoli", "mitoses", "class" };
45    private static double[,] defaultData = new double[,]{
46     {1000025,5,1,1,1,2,1,3,1,1,2      },
47     {1002945,5,4,4,5,7,10,3,2,1,2     },
48     {1015425,3,1,1,1,2,2,3,1,1,2      },
49     {1016277,6,8,8,1,3,4,3,7,1,2      },
50     {1017023,4,1,1,3,2,1,3,1,1,2      },
51     {1017122,8,10,10,8,7,10,9,7,1,4   },
52     {1018099,1,1,1,1,2,10,3,1,1,2     },
53     {1018561,2,1,2,1,2,1,3,1,1,2      },
54     {1033078,2,1,1,1,2,1,1,1,5,2      },
55     {1033078,4,2,1,1,2,1,2,1,1,2      },
56     {1035283,1,1,1,1,1,1,3,1,1,2      },
57     {1036172,2,1,1,1,2,1,2,1,1,2      },
58     {1041801,5,3,3,3,2,3,4,4,1,4      },
59     {1043999,1,1,1,1,2,3,3,1,1,2      },
60     {1044572,8,7,5,10,7,9,5,5,4,4     },
61     {1047630,7,4,6,4,6,1,4,3,1,4      },
62     {1048672,4,1,1,1,2,1,2,1,1,2      },
63     {1049815,4,1,1,1,2,1,3,1,1,2      },
64     {1050670,10,7,7,6,4,10,4,1,2,4    },
65     {1050718,6,1,1,1,2,1,3,1,1,2      },
66     {1054590,7,3,2,10,5,10,5,4,4,4    },
67     {1054593,10,5,5,3,6,7,7,10,1,4    },
68     {1056784,3,1,1,1,2,1,2,1,1,2      },
69     {1057013,8,4,5,1,2,2,7,3,1,4      },
70     {1059552,1,1,1,1,2,1,3,1,1,2      },
71     {1065726,5,2,3,4,2,7,3,6,1,4      },
72     {1066373,3,2,1,1,1,1,2,1,1,2      },
73     {1066979,5,1,1,1,2,1,2,1,1,2      },
74     {1067444,2,1,1,1,2,1,2,1,1,2      },
75     {1070935,1,1,3,1,2,1,1,1,1,2      },
76     {1070935,3,1,1,1,1,1,2,1,1,2      },
77     {1071760,2,1,1,1,2,1,3,1,1,2      },
78     {1072179,10,7,7,3,8,5,7,4,3,4     },
79     {1074610,2,1,1,2,2,1,3,1,1,2      },
80     {1075123,3,1,2,1,2,1,2,1,1,2      },
81     {1079304,2,1,1,1,2,1,2,1,1,2      },
82     {1080185,10,10,10,8,6,1,8,9,1,4   },
83     {1081791,6,2,1,1,1,1,7,1,1,2      },
84     {1084584,5,4,4,9,2,10,5,6,1,4     },
85     {1091262,2,5,3,3,6,7,7,5,1,4      },
86     {1096800,6,6,6,9,6,4,7,8,1,2      },
87     {1099510,10,4,3,1,3,3,6,5,2,4     },
88     {1100524,6,10,10,2,8,10,7,3,3,4   },
89     {1102573,5,6,5,6,10,1,3,1,1,4     },
90     {1103608,10,10,10,4,8,1,8,10,1,4  },
91     {1103722,1,1,1,1,2,1,2,1,2,2      },
92     {1105257,3,7,7,4,4,9,4,8,1,4      },
93     {1105524,1,1,1,1,2,1,2,1,1,2      },
94     {1106095,4,1,1,3,2,1,3,1,1,2      },
95     {1106829,7,8,7,2,4,8,3,8,2,4      },
96     {1108370,9,5,8,1,2,3,2,1,5,4      },
97     {1108449,5,3,3,4,2,4,3,4,1,4      },
98     {1110102,10,3,6,2,3,5,4,10,2,4    },
99     {1110503,5,5,5,8,10,8,7,3,7,4     },
100     {1110524,10,5,5,6,8,8,7,1,1,4     },
101     {1111249,10,6,6,3,4,5,3,6,1,4     },
102     {1112209,8,10,10,1,3,6,3,9,1,4    },
103     {1113038,8,2,4,1,5,1,5,4,4,4      },
104     {1113483,5,2,3,1,6,10,5,1,1,4     },
105     {1113906,9,5,5,2,2,2,5,1,1,4      },
106     {1115282,5,3,5,5,3,3,4,10,1,4     },
107     {1115293,1,1,1,1,2,2,2,1,1,2      },
108     {1116116,9,10,10,1,10,8,3,3,1,4   },
109     {1116132,6,3,4,1,5,2,3,9,1,4      },
110     {1116192,1,1,1,1,2,1,2,1,1,2      },
111     {1116998,10,4,2,1,3,2,4,3,10,4    },
112     {1117152,4,1,1,1,2,1,3,1,1,2      },
113     {1118039,5,3,4,1,8,10,4,9,1,4     },
114     {1120559,8,3,8,3,4,9,8,9,8,4      },
115     {1121732,1,1,1,1,2,1,3,2,1,2      },
116     {1121919,5,1,3,1,2,1,2,1,1,2      },
117     {1123061,6,10,2,8,10,2,7,8,10,4   },
118     {1124651,1,3,3,2,2,1,7,2,1,2      },
119     {1125035,9,4,5,10,6,10,4,8,1,4    },
120     {1126417,10,6,4,1,3,4,3,2,3,4     },
121     {1131294,1,1,2,1,2,2,4,2,1,2      },
122     {1132347,1,1,4,1,2,1,2,1,1,2      },
123     {1133041,5,3,1,2,2,1,2,1,1,2      },
124     {1133136,3,1,1,1,2,3,3,1,1,2      },
125     {1136142,2,1,1,1,3,1,2,1,1,2      },
126     {1137156,2,2,2,1,1,1,7,1,1,2      },
127     {1143978,4,1,1,2,2,1,2,1,1,2      },
128     {1143978,5,2,1,1,2,1,3,1,1,2      },
129     {1147044,3,1,1,1,2,2,7,1,1,2      },
130     {1147699,3,5,7,8,8,9,7,10,7,4     },
131     {1147748,5,10,6,1,10,4,4,10,10,4  },
132     {1148278,3,3,6,4,5,8,4,4,1,4      },
133     {1148873,3,6,6,6,5,10,6,8,3,4     },
134     {1152331,4,1,1,1,2,1,3,1,1,2      },
135     {1155546,2,1,1,2,3,1,2,1,1,2      },
136     {1156272,1,1,1,1,2,1,3,1,1,2      },
137     {1156948,3,1,1,2,2,1,1,1,1,2      },
138     {1157734,4,1,1,1,2,1,3,1,1,2      },
139     {1158247,1,1,1,1,2,1,2,1,1,2      },
140     {1160476,2,1,1,1,2,1,3,1,1,2      },
141     {1164066,1,1,1,1,2,1,3,1,1,2      },
142     {1165297,2,1,1,2,2,1,1,1,1,2      },
143     {1165790,5,1,1,1,2,1,3,1,1,2      },
144     {1165926,9,6,9,2,10,6,2,9,10,4    },
145     {1166630,7,5,6,10,5,10,7,9,4,4    },
146     {1166654,10,3,5,1,10,5,3,10,2,4   },
147     {1167439,2,3,4,4,2,5,2,5,1,4      },
148     {1167471,4,1,2,1,2,1,3,1,1,2      },
149     {1168359,8,2,3,1,6,3,7,1,1,4      },
150     {1168736,10,10,10,10,10,1,8,8,8,4 },
151     {1169049,7,3,4,4,3,3,3,2,7,4      },
152     {1170419,10,10,10,8,2,10,4,1,1,4  },
153     {1170420,1,6,8,10,8,10,5,7,1,4    },
154     {1171710,1,1,1,1,2,1,2,3,1,2      },
155     {1171710,6,5,4,4,3,9,7,8,3,4      },
156     {1171795,1,3,1,2,2,2,5,3,2,2      },
157     {1171845,8,6,4,3,5,9,3,1,1,4      },
158     {1172152,10,3,3,10,2,10,7,3,3,4   },
159     {1173216,10,10,10,3,10,8,8,1,1,4  },
160     {1173235,3,3,2,1,2,3,3,1,1,2      },
161     {1173347,1,1,1,1,2,5,1,1,1,2      },
162     {1173347,8,3,3,1,2,2,3,2,1,2      },
163     {1173509,4,5,5,10,4,10,7,5,8,4    },
164     {1173514,1,1,1,1,4,3,1,1,1,2      },
165     {1173681,3,2,1,1,2,2,3,1,1,2      },
166     {1174057,1,1,2,2,2,1,3,1,1,2      },
167     {1174057,4,2,1,1,2,2,3,1,1,2      },
168     {1174131,10,10,10,2,10,10,5,3,3,4 },
169     {1174428,5,3,5,1,8,10,5,3,1,4     },
170     {1175937,5,4,6,7,9,7,8,10,1,4     },
171     {1176406,1,1,1,1,2,1,2,1,1,2      },
172     {1176881,7,5,3,7,4,10,7,5,5,4        }
173};
[6672]174    private static readonly Dataset defaultDataset;
175    private static readonly IEnumerable<string> defaultAllowedInputVariables;
176    private static readonly string defaultTargetVariable;
[6666]177
[6672]178    private static readonly ClassificationProblemData emptyProblemData;
[6666]179    public static ClassificationProblemData EmptyProblemData {
180      get { return EmptyProblemData; }
181    }
182
[5559]183    static ClassificationProblemData() {
184      defaultDataset = new Dataset(defaultVariableNames, defaultData);
185      defaultDataset.Name = "Wisconsin classification problem";
186      defaultDataset.Description = "subset from to ..";
187
188      defaultAllowedInputVariables = defaultVariableNames.Except(new List<string>() { "sample", "class" });
189      defaultTargetVariable = "class";
[6666]190
191      var problemData = new ClassificationProblemData();
192      problemData.Parameters.Clear();
193      problemData.Name = "Empty Classification ProblemData";
194      problemData.Description = "This ProblemData acts as place holder before the correct problem data is loaded.";
195      problemData.isEmpty = true;
196
197      problemData.Parameters.Add(new FixedValueParameter<Dataset>(DatasetParameterName, "", new Dataset()));
198      problemData.Parameters.Add(new FixedValueParameter<ReadOnlyCheckedItemList<StringValue>>(InputVariablesParameterName, ""));
199      problemData.Parameters.Add(new FixedValueParameter<IntRange>(TrainingPartitionParameterName, "", (IntRange)new IntRange(0, 0).AsReadOnly()));
200      problemData.Parameters.Add(new FixedValueParameter<IntRange>(TestPartitionParameterName, "", (IntRange)new IntRange(0, 0).AsReadOnly()));
201      problemData.Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>()));
202      problemData.Parameters.Add(new FixedValueParameter<StringMatrix>(ClassNamesParameterName, "", new StringMatrix(0, 0).AsReadOnly()));
203      problemData.Parameters.Add(new FixedValueParameter<DoubleMatrix>(ClassificationPenaltiesParameterName, "", (DoubleMatrix)new DoubleMatrix(0, 0).AsReadOnly()));
204      emptyProblemData = problemData;
[5559]205    }
206    #endregion
207
[5601]208    #region parameter properties
[8121]209    public IConstrainedValueParameter<StringValue> TargetVariableParameter {
210      get { return (IConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
[5601]211    }
212    public IFixedValueParameter<StringMatrix> ClassNamesParameter {
213      get { return (IFixedValueParameter<StringMatrix>)Parameters[ClassNamesParameterName]; }
214    }
215    public IFixedValueParameter<DoubleMatrix> ClassificationPenaltiesParameter {
216      get { return (IFixedValueParameter<DoubleMatrix>)Parameters[ClassificationPenaltiesParameterName]; }
217    }
218    #endregion
219
[5649]220    #region properties
[5559]221    public string TargetVariable {
[5601]222      get { return TargetVariableParameter.Value.Value; }
223    }
[5559]224
[8554]225    private List<double> classValuesCache;
226    private List<double> ClassValuesCache {
[5601]227      get {
[8554]228        if (classValuesCache == null) {
229          classValuesCache = Dataset.GetDoubleValues(TargetVariableParameter.Value.Value).Distinct().OrderBy(x => x).ToList();
[5559]230        }
[8554]231        return classValuesCache;
[5559]232      }
233    }
[8554]234    public IEnumerable<double> ClassValues {
235      get { return ClassValuesCache; }
[5559]236    }
237    public int Classes {
[8554]238      get { return ClassValuesCache.Count; }
[5559]239    }
240
[8554]241    private List<string> classNamesCache;
242    private List<string> ClassNamesCache {
[5601]243      get {
[8554]244        if (classNamesCache == null) {
245          classNamesCache = new List<string>();
[5601]246          for (int i = 0; i < ClassNamesParameter.Value.Rows; i++)
[8554]247            classNamesCache.Add(ClassNamesParameter.Value[i, 0]);
[5601]248        }
[8554]249        return classNamesCache;
[5559]250      }
251    }
[8554]252    public IEnumerable<string> ClassNames {
253      get { return ClassNamesCache; }
[5559]254    }
255    #endregion
256
257
258    [StorableConstructor]
259    protected ClassificationProblemData(bool deserializing) : base(deserializing) { }
[5601]260    [StorableHook(HookType.AfterDeserialization)]
261    private void AfterDeserialization() {
262      RegisterParameterEvents();
263    }
[5559]264
[5601]265    protected ClassificationProblemData(ClassificationProblemData original, Cloner cloner)
266      : base(original, cloner) {
267      RegisterParameterEvents();
[5559]268    }
[6666]269    public override IDeepCloneable Clone(Cloner cloner) {
270      if (this == emptyProblemData) return emptyProblemData;
271      return new ClassificationProblemData(this, cloner);
272    }
[5559]273
[5601]274    public ClassificationProblemData() : this(defaultDataset, defaultAllowedInputVariables, defaultTargetVariable) { }
[8528]275
276    public ClassificationProblemData(IClassificationProblemData classificationProblemData)
277      : this(classificationProblemData.Dataset, classificationProblemData.AllowedInputVariables, classificationProblemData.TargetVariable) {
278      TrainingPartition.Start = classificationProblemData.TrainingPartition.Start;
279      TrainingPartition.End = classificationProblemData.TrainingPartition.End;
280      TestPartition.Start = classificationProblemData.TestPartition.Start;
281      TestPartition.End = classificationProblemData.TestPartition.End;
282    }
283
[5559]284    public ClassificationProblemData(Dataset dataset, IEnumerable<string> allowedInputVariables, string targetVariable)
285      : base(dataset, allowedInputVariables) {
[6186]286      var validTargetVariableValues = CheckVariablesForPossibleTargetVariables(dataset).Select(x => new StringValue(x).AsReadOnly()).ToList();
287      var target = validTargetVariableValues.Where(x => x.Value == targetVariable).DefaultIfEmpty(validTargetVariableValues.First()).First();
288
289      Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>(validTargetVariableValues), target));
[5847]290      Parameters.Add(new FixedValueParameter<StringMatrix>(ClassNamesParameterName, ""));
291      Parameters.Add(new FixedValueParameter<DoubleMatrix>(ClassificationPenaltiesParameterName, ""));
[5559]292
[5601]293      ResetTargetVariableDependentMembers();
294      RegisterParameterEvents();
[5559]295    }
296
[6186]297    private static IEnumerable<string> CheckVariablesForPossibleTargetVariables(Dataset dataset) {
[6223]298      int maxSamples = Math.Min(InspectedRowsToDetermineTargets, dataset.Rows);
[6740]299      var validTargetVariables = (from v in dataset.DoubleVariables
300                                  let distinctValues = dataset.GetDoubleValues(v)
[6654]301                                    .Take(maxSamples)
302                                    .Distinct()
303                                    .Count()
304                                  where distinctValues < MaximumNumberOfClasses
305                                  select v).ToArray();
[6186]306
307      if (!validTargetVariables.Any())
[6223]308        throw new ArgumentException("Import of classification problem data was not successful, because no target variable was found." +
309          " A target variable must have at most " + MaximumNumberOfClasses + " distinct values to be applicable to classification.");
[6186]310      return validTargetVariables;
311    }
312
313
[5601]314    private void ResetTargetVariableDependentMembers() {
[6654]315      DeregisterParameterEvents();
[5559]316
[5601]317      ((IStringConvertibleMatrix)ClassNamesParameter.Value).Columns = 1;
[8554]318      ((IStringConvertibleMatrix)ClassNamesParameter.Value).Rows = ClassValuesCache.Count;
[5601]319      for (int i = 0; i < Classes; i++)
[8554]320        ClassNamesParameter.Value[i, 0] = "Class " + ClassValuesCache[i];
[5601]321      ClassNamesParameter.Value.ColumnNames = new List<string>() { "ClassNames" };
322      ClassNamesParameter.Value.RowNames = ClassValues.Select(s => "ClassValue: " + s);
[5559]323
[5601]324      ((IStringConvertibleMatrix)ClassificationPenaltiesParameter.Value).Rows = Classes;
325      ((IStringConvertibleMatrix)ClassificationPenaltiesParameter.Value).Columns = Classes;
326      ClassificationPenaltiesParameter.Value.RowNames = ClassNames.Select(name => "Actual " + name);
327      ClassificationPenaltiesParameter.Value.ColumnNames = ClassNames.Select(name => "Estimated " + name);
328      for (int i = 0; i < Classes; i++) {
329        for (int j = 0; j < Classes; j++) {
330          if (i != j) ClassificationPenaltiesParameter.Value[i, j] = 1;
331          else ClassificationPenaltiesParameter.Value[i, j] = 0;
[5559]332        }
333      }
[5601]334      RegisterParameterEvents();
[5559]335    }
336
337    public string GetClassName(double classValue) {
[8554]338      if (!ClassValuesCache.Contains(classValue)) throw new ArgumentException();
339      int index = ClassValuesCache.IndexOf(classValue);
340      return ClassNamesCache[index];
[5559]341    }
342    public double GetClassValue(string className) {
[8554]343      if (!ClassNamesCache.Contains(className)) throw new ArgumentException();
344      int index = ClassNamesCache.IndexOf(className);
345      return ClassValuesCache[index];
[5559]346    }
347    public void SetClassName(double classValue, string className) {
[8554]348      if (!ClassValuesCache.Contains(classValue)) throw new ArgumentException();
349      int index = ClassValuesCache.IndexOf(classValue);
[5601]350      ClassNamesParameter.Value[index, 0] = className;
[8554]351      // updating of class names cache is not necessary here as the parameter value fires a changed event which updates the cache
[5559]352    }
353
354    public double GetClassificationPenalty(string correctClassName, string estimatedClassName) {
355      return GetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName));
356    }
357    public double GetClassificationPenalty(double correctClassValue, double estimatedClassValue) {
[8554]358      int correctClassIndex = ClassValuesCache.IndexOf(correctClassValue);
359      int estimatedClassIndex = ClassValuesCache.IndexOf(estimatedClassValue);
360      return ClassificationPenaltiesParameter.Value[correctClassIndex, estimatedClassIndex];
[5559]361    }
362    public void SetClassificationPenalty(string correctClassName, string estimatedClassName, double penalty) {
363      SetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName), penalty);
364    }
365    public void SetClassificationPenalty(double correctClassValue, double estimatedClassValue, double penalty) {
[8554]366      int correctClassIndex = ClassValuesCache.IndexOf(correctClassValue);
367      int estimatedClassIndex = ClassValuesCache.IndexOf(estimatedClassValue);
[5601]368
369      ClassificationPenaltiesParameter.Value[correctClassIndex, estimatedClassIndex] = penalty;
[5559]370    }
371
[5601]372    #region events
373    private void RegisterParameterEvents() {
374      TargetVariableParameter.ValueChanged += new EventHandler(TargetVariableParameter_ValueChanged);
375      ClassNamesParameter.Value.Reset += new EventHandler(Parameter_ValueChanged);
376      ClassNamesParameter.Value.ItemChanged += new EventHandler<EventArgs<int, int>>(MatrixParameter_ItemChanged);
[5559]377    }
[6654]378    private void DeregisterParameterEvents() {
[5601]379      TargetVariableParameter.ValueChanged -= new EventHandler(TargetVariableParameter_ValueChanged);
380      ClassNamesParameter.Value.Reset -= new EventHandler(Parameter_ValueChanged);
381      ClassNamesParameter.Value.ItemChanged -= new EventHandler<EventArgs<int, int>>(MatrixParameter_ItemChanged);
[5559]382    }
[5601]383
384    private void TargetVariableParameter_ValueChanged(object sender, EventArgs e) {
[8554]385      classValuesCache = null;
386      classNamesCache = null;
[5601]387      ResetTargetVariableDependentMembers();
388      OnChanged();
389    }
390    private void Parameter_ValueChanged(object sender, EventArgs e) {
[8554]391      classNamesCache = null;
[5601]392      OnChanged();
393    }
394    private void MatrixParameter_ItemChanged(object sender, EventArgs<int, int> e) {
[8554]395      classNamesCache = null;
[5601]396      OnChanged();
397    }
398    #endregion
[5559]399  }
400}
Note: See TracBrowser for help on using the repository browser.