Free cookie consent management tool by TermsFeed Policy Generator

source: trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationProblemData.cs @ 11130

Last change on this file since 11130 was 11114, checked in by mkommend, 10 years ago

#2206: Merged data preprocessing into the trunk.
File size: 22.7 KB
RevLine 
[5559]1#region License Information
2/* HeuristicLab
[9456]3 * Copyright (C) 2002-2013 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; }
[10540]223      set {
224        if (value == null) throw new ArgumentNullException("targetVariable", "The provided value for the targetVariable is null.");
225        if (value == TargetVariable) return;
226
227
228        var matchingParameterValue = TargetVariableParameter.ValidValues.FirstOrDefault(v => v.Value == value);
229        if (matchingParameterValue == null) throw new ArgumentException("The provided value is not valid as the targetVariable.", "targetVariable");
230        TargetVariableParameter.Value = matchingParameterValue;
231      }
[5601]232    }
[5559]233
[8554]234    private List<double> classValuesCache;
235    private List<double> ClassValuesCache {
[5601]236      get {
[8554]237        if (classValuesCache == null) {
238          classValuesCache = Dataset.GetDoubleValues(TargetVariableParameter.Value.Value).Distinct().OrderBy(x => x).ToList();
[5559]239        }
[8554]240        return classValuesCache;
[5559]241      }
242    }
[8554]243    public IEnumerable<double> ClassValues {
244      get { return ClassValuesCache; }
[5559]245    }
246    public int Classes {
[8554]247      get { return ClassValuesCache.Count; }
[5559]248    }
249
[8554]250    private List<string> classNamesCache;
251    private List<string> ClassNamesCache {
[5601]252      get {
[8554]253        if (classNamesCache == null) {
254          classNamesCache = new List<string>();
[5601]255          for (int i = 0; i < ClassNamesParameter.Value.Rows; i++)
[8554]256            classNamesCache.Add(ClassNamesParameter.Value[i, 0]);
[5601]257        }
[8554]258        return classNamesCache;
[5559]259      }
260    }
[8554]261    public IEnumerable<string> ClassNames {
262      get { return ClassNamesCache; }
[5559]263    }
264    #endregion
265
266
267    [StorableConstructor]
268    protected ClassificationProblemData(bool deserializing) : base(deserializing) { }
[5601]269    [StorableHook(HookType.AfterDeserialization)]
270    private void AfterDeserialization() {
271      RegisterParameterEvents();
272    }
[5559]273
[5601]274    protected ClassificationProblemData(ClassificationProblemData original, Cloner cloner)
275      : base(original, cloner) {
276      RegisterParameterEvents();
[5559]277    }
[6666]278    public override IDeepCloneable Clone(Cloner cloner) {
279      if (this == emptyProblemData) return emptyProblemData;
280      return new ClassificationProblemData(this, cloner);
281    }
[5559]282
[5601]283    public ClassificationProblemData() : this(defaultDataset, defaultAllowedInputVariables, defaultTargetVariable) { }
[8528]284
285    public ClassificationProblemData(IClassificationProblemData classificationProblemData)
286      : this(classificationProblemData.Dataset, classificationProblemData.AllowedInputVariables, classificationProblemData.TargetVariable) {
287      TrainingPartition.Start = classificationProblemData.TrainingPartition.Start;
288      TrainingPartition.End = classificationProblemData.TrainingPartition.End;
289      TestPartition.Start = classificationProblemData.TestPartition.Start;
290      TestPartition.End = classificationProblemData.TestPartition.End;
[8716]291
292      for (int i = 0; i < classificationProblemData.ClassNames.Count(); i++)
293        ClassNamesParameter.Value[i, 0] = classificationProblemData.ClassNames.ElementAt(i);
[8717]294
295      for (int i = 0; i < Classes; i++) {
296        for (int j = 0; j < Classes; j++) {
[8745]297          ClassificationPenaltiesParameter.Value[i, j] = classificationProblemData.GetClassificationPenalty(ClassValuesCache[i], ClassValuesCache[j]);
[8717]298        }
299      }
[8528]300    }
301
[11114]302    public ClassificationProblemData(Dataset dataset, IEnumerable<string> allowedInputVariables, string targetVariable, IEnumerable<ITransformation> transformations = null)
303      : base(dataset, allowedInputVariables, transformations ?? Enumerable.Empty<ITransformation>()) {
[6186]304      var validTargetVariableValues = CheckVariablesForPossibleTargetVariables(dataset).Select(x => new StringValue(x).AsReadOnly()).ToList();
305      var target = validTargetVariableValues.Where(x => x.Value == targetVariable).DefaultIfEmpty(validTargetVariableValues.First()).First();
306
307      Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>(validTargetVariableValues), target));
[5847]308      Parameters.Add(new FixedValueParameter<StringMatrix>(ClassNamesParameterName, ""));
309      Parameters.Add(new FixedValueParameter<DoubleMatrix>(ClassificationPenaltiesParameterName, ""));
[5559]310
[8802]311      RegisterParameterEvents();
[5601]312      ResetTargetVariableDependentMembers();
[5559]313    }
314
[8877]315    public static IEnumerable<string> CheckVariablesForPossibleTargetVariables(Dataset dataset) {
[6223]316      int maxSamples = Math.Min(InspectedRowsToDetermineTargets, dataset.Rows);
[6740]317      var validTargetVariables = (from v in dataset.DoubleVariables
318                                  let distinctValues = dataset.GetDoubleValues(v)
[6654]319                                    .Take(maxSamples)
320                                    .Distinct()
321                                    .Count()
[9449]322                                  where distinctValues <= MaximumNumberOfClasses
[6654]323                                  select v).ToArray();
[6186]324
325      if (!validTargetVariables.Any())
[6223]326        throw new ArgumentException("Import of classification problem data was not successful, because no target variable was found." +
327          " A target variable must have at most " + MaximumNumberOfClasses + " distinct values to be applicable to classification.");
[6186]328      return validTargetVariables;
329    }
330
331
[5601]332    private void ResetTargetVariableDependentMembers() {
[6654]333      DeregisterParameterEvents();
[5559]334
[5601]335      ((IStringConvertibleMatrix)ClassNamesParameter.Value).Columns = 1;
[8554]336      ((IStringConvertibleMatrix)ClassNamesParameter.Value).Rows = ClassValuesCache.Count;
[5601]337      for (int i = 0; i < Classes; i++)
[8554]338        ClassNamesParameter.Value[i, 0] = "Class " + ClassValuesCache[i];
[5601]339      ClassNamesParameter.Value.ColumnNames = new List<string>() { "ClassNames" };
340      ClassNamesParameter.Value.RowNames = ClassValues.Select(s => "ClassValue: " + s);
[5559]341
[5601]342      ((IStringConvertibleMatrix)ClassificationPenaltiesParameter.Value).Rows = Classes;
343      ((IStringConvertibleMatrix)ClassificationPenaltiesParameter.Value).Columns = Classes;
344      ClassificationPenaltiesParameter.Value.RowNames = ClassNames.Select(name => "Actual " + name);
345      ClassificationPenaltiesParameter.Value.ColumnNames = ClassNames.Select(name => "Estimated " + name);
346      for (int i = 0; i < Classes; i++) {
347        for (int j = 0; j < Classes; j++) {
348          if (i != j) ClassificationPenaltiesParameter.Value[i, j] = 1;
349          else ClassificationPenaltiesParameter.Value[i, j] = 0;
[5559]350        }
351      }
[5601]352      RegisterParameterEvents();
[5559]353    }
354
355    public string GetClassName(double classValue) {
[8554]356      if (!ClassValuesCache.Contains(classValue)) throw new ArgumentException();
357      int index = ClassValuesCache.IndexOf(classValue);
358      return ClassNamesCache[index];
[5559]359    }
360    public double GetClassValue(string className) {
[8554]361      if (!ClassNamesCache.Contains(className)) throw new ArgumentException();
362      int index = ClassNamesCache.IndexOf(className);
363      return ClassValuesCache[index];
[5559]364    }
365    public void SetClassName(double classValue, string className) {
[8554]366      if (!ClassValuesCache.Contains(classValue)) throw new ArgumentException();
367      int index = ClassValuesCache.IndexOf(classValue);
[5601]368      ClassNamesParameter.Value[index, 0] = className;
[8554]369      // updating of class names cache is not necessary here as the parameter value fires a changed event which updates the cache
[5559]370    }
371
372    public double GetClassificationPenalty(string correctClassName, string estimatedClassName) {
373      return GetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName));
374    }
375    public double GetClassificationPenalty(double correctClassValue, double estimatedClassValue) {
[8554]376      int correctClassIndex = ClassValuesCache.IndexOf(correctClassValue);
377      int estimatedClassIndex = ClassValuesCache.IndexOf(estimatedClassValue);
378      return ClassificationPenaltiesParameter.Value[correctClassIndex, estimatedClassIndex];
[5559]379    }
380    public void SetClassificationPenalty(string correctClassName, string estimatedClassName, double penalty) {
381      SetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName), penalty);
382    }
383    public void SetClassificationPenalty(double correctClassValue, double estimatedClassValue, double penalty) {
[8554]384      int correctClassIndex = ClassValuesCache.IndexOf(correctClassValue);
385      int estimatedClassIndex = ClassValuesCache.IndexOf(estimatedClassValue);
[5601]386
387      ClassificationPenaltiesParameter.Value[correctClassIndex, estimatedClassIndex] = penalty;
[5559]388    }
389
[5601]390    #region events
391    private void RegisterParameterEvents() {
392      TargetVariableParameter.ValueChanged += new EventHandler(TargetVariableParameter_ValueChanged);
393      ClassNamesParameter.Value.Reset += new EventHandler(Parameter_ValueChanged);
[8716]394      ClassNamesParameter.Value.ItemChanged += new EventHandler<EventArgs<int, int>>(Parameter_ValueChanged);
[8717]395      ClassificationPenaltiesParameter.Value.ItemChanged += new EventHandler<EventArgs<int, int>>(Parameter_ValueChanged);
396      ClassificationPenaltiesParameter.Value.Reset += new EventHandler(Parameter_ValueChanged);
[5559]397    }
[6654]398    private void DeregisterParameterEvents() {
[5601]399      TargetVariableParameter.ValueChanged -= new EventHandler(TargetVariableParameter_ValueChanged);
400      ClassNamesParameter.Value.Reset -= new EventHandler(Parameter_ValueChanged);
[8716]401      ClassNamesParameter.Value.ItemChanged -= new EventHandler<EventArgs<int, int>>(Parameter_ValueChanged);
[8717]402      ClassificationPenaltiesParameter.Value.ItemChanged -= new EventHandler<EventArgs<int, int>>(Parameter_ValueChanged);
403      ClassificationPenaltiesParameter.Value.Reset -= new EventHandler(Parameter_ValueChanged);
[5559]404    }
[5601]405
406    private void TargetVariableParameter_ValueChanged(object sender, EventArgs e) {
[8554]407      classValuesCache = null;
408      classNamesCache = null;
[5601]409      ResetTargetVariableDependentMembers();
410      OnChanged();
411    }
412    private void Parameter_ValueChanged(object sender, EventArgs e) {
[8554]413      classNamesCache = null;
[8717]414      ClassificationPenaltiesParameter.Value.RowNames = ClassNames.Select(name => "Actual " + name);
415      ClassificationPenaltiesParameter.Value.ColumnNames = ClassNames.Select(name => "Estimated " + name);
[5601]416      OnChanged();
417    }
418    #endregion
[10540]419
420    protected override bool IsProblemDataCompatible(IDataAnalysisProblemData problemData, out string errorMessage) {
421      if (problemData == null) throw new ArgumentNullException("problemData", "The provided problemData is null.");
422      IClassificationProblemData classificationProblemData = problemData as IClassificationProblemData;
423      if (classificationProblemData == null)
424        throw new ArgumentException("The problem data is no classification problem data. Instead a " + problemData.GetType().GetPrettyName() + " was provided.", "problemData");
425
426      var returnValue = base.IsProblemDataCompatible(classificationProblemData, out errorMessage);
427      //check targetVariable
428      if (classificationProblemData.InputVariables.All(var => var.Value != TargetVariable)) {
429        errorMessage = string.Format("The target variable {0} is not present in the new problem data.", TargetVariable)
430                       + Environment.NewLine + errorMessage;
431        return false;
432      }
433
434      var newClassValues = classificationProblemData.Dataset.GetDoubleValues(TargetVariable).Distinct().OrderBy(x => x);
435      if (!newClassValues.SequenceEqual(ClassValues)) {
436        errorMessage = errorMessage + string.Format("The class values differ in the provided classification problem data.");
437        return false;
438      }
439
440      return returnValue;
441    }
442
443    public override void AdjustProblemDataProperties(IDataAnalysisProblemData problemData) {
444      if (problemData == null) throw new ArgumentNullException("problemData", "The provided problemData is null.");
445      ClassificationProblemData classificationProblemData = problemData as ClassificationProblemData;
446      if (classificationProblemData == null)
447        throw new ArgumentException("The problem data is not a classification problem data. Instead a " + problemData.GetType().GetPrettyName() + " was provided.", "problemData");
448
449      base.AdjustProblemDataProperties(problemData);
450      TargetVariable = classificationProblemData.TargetVariable;
451      for (int i = 0; i < classificationProblemData.ClassNames.Count(); i++)
452        ClassNamesParameter.Value[i, 0] = classificationProblemData.ClassNames.ElementAt(i);
453
454      for (int i = 0; i < Classes; i++) {
455        for (int j = 0; j < Classes; j++) {
456          ClassificationPenaltiesParameter.Value[i, j] = classificationProblemData.GetClassificationPenalty(ClassValuesCache[i], ClassValuesCache[j]);
457        }
458      }
459    }
[5559]460  }
461}
Note: See TracBrowser for help on using the repository browser.