Free cookie consent management tool by TermsFeed Policy Generator

source: branches/DataAnalysis Refactoring/HeuristicLab.Problems.DataAnalysis/3.4/ClassificationProblemData.cs @ 5593

Last change on this file since 5593 was 5579, checked in by mkommend, 14 years ago

#1418: Adapted new ProblemData classes.
File size: 12.8 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2011 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
22using System;
23using System.Collections.Generic;
24using System.Linq;
25using HeuristicLab.Common;
26using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
27
28namespace HeuristicLab.Problems.DataAnalysis {
29  [StorableClass]
30  public class ClassificationProblemData : DataAnalysisProblemData, IClassificationProblemData {
31    #region default data
32    private static string[] defaultVariableNames = new string[] { "sample", "clump thickness", "cell size", "cell shape", "marginal adhesion", "epithelial cell size", "bare nuclei", "chromatin", "nucleoli", "mitoses", "class" };
33    private static double[,] defaultData = new double[,]{
34     {1000025,5,1,1,1,2,1,3,1,1,2      },
35     {1002945,5,4,4,5,7,10,3,2,1,2     },
36     {1015425,3,1,1,1,2,2,3,1,1,2      },
37     {1016277,6,8,8,1,3,4,3,7,1,2      },
38     {1017023,4,1,1,3,2,1,3,1,1,2      },
39     {1017122,8,10,10,8,7,10,9,7,1,4   },
40     {1018099,1,1,1,1,2,10,3,1,1,2     },
41     {1018561,2,1,2,1,2,1,3,1,1,2      },
42     {1033078,2,1,1,1,2,1,1,1,5,2      },
43     {1033078,4,2,1,1,2,1,2,1,1,2      },
44     {1035283,1,1,1,1,1,1,3,1,1,2      },
45     {1036172,2,1,1,1,2,1,2,1,1,2      },
46     {1041801,5,3,3,3,2,3,4,4,1,4      },
47     {1043999,1,1,1,1,2,3,3,1,1,2      },
48     {1044572,8,7,5,10,7,9,5,5,4,4     },
49     {1047630,7,4,6,4,6,1,4,3,1,4      },
50     {1048672,4,1,1,1,2,1,2,1,1,2      },
51     {1049815,4,1,1,1,2,1,3,1,1,2      },
52     {1050670,10,7,7,6,4,10,4,1,2,4    },
53     {1050718,6,1,1,1,2,1,3,1,1,2      },
54     {1054590,7,3,2,10,5,10,5,4,4,4    },
55     {1054593,10,5,5,3,6,7,7,10,1,4    },
56     {1056784,3,1,1,1,2,1,2,1,1,2      },
57     {1057013,8,4,5,1,2,2,7,3,1,4      },
58     {1059552,1,1,1,1,2,1,3,1,1,2      },
59     {1065726,5,2,3,4,2,7,3,6,1,4      },
60     {1066373,3,2,1,1,1,1,2,1,1,2      },
61     {1066979,5,1,1,1,2,1,2,1,1,2      },
62     {1067444,2,1,1,1,2,1,2,1,1,2      },
63     {1070935,1,1,3,1,2,1,1,1,1,2      },
64     {1070935,3,1,1,1,1,1,2,1,1,2      },
65     {1071760,2,1,1,1,2,1,3,1,1,2      },
66     {1072179,10,7,7,3,8,5,7,4,3,4     },
67     {1074610,2,1,1,2,2,1,3,1,1,2      },
68     {1075123,3,1,2,1,2,1,2,1,1,2      },
69     {1079304,2,1,1,1,2,1,2,1,1,2      },
70     {1080185,10,10,10,8,6,1,8,9,1,4   },
71     {1081791,6,2,1,1,1,1,7,1,1,2      },
72     {1084584,5,4,4,9,2,10,5,6,1,4     },
73     {1091262,2,5,3,3,6,7,7,5,1,4      },
74     {1096800,6,6,6,9,6,4,7,8,1,2      },
75     {1099510,10,4,3,1,3,3,6,5,2,4     },
76     {1100524,6,10,10,2,8,10,7,3,3,4   },
77     {1102573,5,6,5,6,10,1,3,1,1,4     },
78     {1103608,10,10,10,4,8,1,8,10,1,4  },
79     {1103722,1,1,1,1,2,1,2,1,2,2      },
80     {1105257,3,7,7,4,4,9,4,8,1,4      },
81     {1105524,1,1,1,1,2,1,2,1,1,2      },
82     {1106095,4,1,1,3,2,1,3,1,1,2      },
83     {1106829,7,8,7,2,4,8,3,8,2,4      },
84     {1108370,9,5,8,1,2,3,2,1,5,4      },
85     {1108449,5,3,3,4,2,4,3,4,1,4      },
86     {1110102,10,3,6,2,3,5,4,10,2,4    },
87     {1110503,5,5,5,8,10,8,7,3,7,4     },
88     {1110524,10,5,5,6,8,8,7,1,1,4     },
89     {1111249,10,6,6,3,4,5,3,6,1,4     },
90     {1112209,8,10,10,1,3,6,3,9,1,4    },
91     {1113038,8,2,4,1,5,1,5,4,4,4      },
92     {1113483,5,2,3,1,6,10,5,1,1,4     },
93     {1113906,9,5,5,2,2,2,5,1,1,4      },
94     {1115282,5,3,5,5,3,3,4,10,1,4     },
95     {1115293,1,1,1,1,2,2,2,1,1,2      },
96     {1116116,9,10,10,1,10,8,3,3,1,4   },
97     {1116132,6,3,4,1,5,2,3,9,1,4      },
98     {1116192,1,1,1,1,2,1,2,1,1,2      },
99     {1116998,10,4,2,1,3,2,4,3,10,4    },
100     {1117152,4,1,1,1,2,1,3,1,1,2      },
101     {1118039,5,3,4,1,8,10,4,9,1,4     },
102     {1120559,8,3,8,3,4,9,8,9,8,4      },
103     {1121732,1,1,1,1,2,1,3,2,1,2      },
104     {1121919,5,1,3,1,2,1,2,1,1,2      },
105     {1123061,6,10,2,8,10,2,7,8,10,4   },
106     {1124651,1,3,3,2,2,1,7,2,1,2      },
107     {1125035,9,4,5,10,6,10,4,8,1,4    },
108     {1126417,10,6,4,1,3,4,3,2,3,4     },
109     {1131294,1,1,2,1,2,2,4,2,1,2      },
110     {1132347,1,1,4,1,2,1,2,1,1,2      },
111     {1133041,5,3,1,2,2,1,2,1,1,2      },
112     {1133136,3,1,1,1,2,3,3,1,1,2      },
113     {1136142,2,1,1,1,3,1,2,1,1,2      },
114     {1137156,2,2,2,1,1,1,7,1,1,2      },
115     {1143978,4,1,1,2,2,1,2,1,1,2      },
116     {1143978,5,2,1,1,2,1,3,1,1,2      },
117     {1147044,3,1,1,1,2,2,7,1,1,2      },
118     {1147699,3,5,7,8,8,9,7,10,7,4     },
119     {1147748,5,10,6,1,10,4,4,10,10,4  },
120     {1148278,3,3,6,4,5,8,4,4,1,4      },
121     {1148873,3,6,6,6,5,10,6,8,3,4     },
122     {1152331,4,1,1,1,2,1,3,1,1,2      },
123     {1155546,2,1,1,2,3,1,2,1,1,2      },
124     {1156272,1,1,1,1,2,1,3,1,1,2      },
125     {1156948,3,1,1,2,2,1,1,1,1,2      },
126     {1157734,4,1,1,1,2,1,3,1,1,2      },
127     {1158247,1,1,1,1,2,1,2,1,1,2      },
128     {1160476,2,1,1,1,2,1,3,1,1,2      },
129     {1164066,1,1,1,1,2,1,3,1,1,2      },
130     {1165297,2,1,1,2,2,1,1,1,1,2      },
131     {1165790,5,1,1,1,2,1,3,1,1,2      },
132     {1165926,9,6,9,2,10,6,2,9,10,4    },
133     {1166630,7,5,6,10,5,10,7,9,4,4    },
134     {1166654,10,3,5,1,10,5,3,10,2,4   },
135     {1167439,2,3,4,4,2,5,2,5,1,4      },
136     {1167471,4,1,2,1,2,1,3,1,1,2      },
137     {1168359,8,2,3,1,6,3,7,1,1,4      },
138     {1168736,10,10,10,10,10,1,8,8,8,4 },
139     {1169049,7,3,4,4,3,3,3,2,7,4      },
140     {1170419,10,10,10,8,2,10,4,1,1,4  },
141     {1170420,1,6,8,10,8,10,5,7,1,4    },
142     {1171710,1,1,1,1,2,1,2,3,1,2      },
143     {1171710,6,5,4,4,3,9,7,8,3,4      },
144     {1171795,1,3,1,2,2,2,5,3,2,2      },
145     {1171845,8,6,4,3,5,9,3,1,1,4      },
146     {1172152,10,3,3,10,2,10,7,3,3,4   },
147     {1173216,10,10,10,3,10,8,8,1,1,4  },
148     {1173235,3,3,2,1,2,3,3,1,1,2      },
149     {1173347,1,1,1,1,2,5,1,1,1,2      },
150     {1173347,8,3,3,1,2,2,3,2,1,2      },
151     {1173509,4,5,5,10,4,10,7,5,8,4    },
152     {1173514,1,1,1,1,4,3,1,1,1,2      },
153     {1173681,3,2,1,1,2,2,3,1,1,2      },
154     {1174057,1,1,2,2,2,1,3,1,1,2      },
155     {1174057,4,2,1,1,2,2,3,1,1,2      },
156     {1174131,10,10,10,2,10,10,5,3,3,4 },
157     {1174428,5,3,5,1,8,10,5,3,1,4     },
158     {1175937,5,4,6,7,9,7,8,10,1,4     },
159     {1176406,1,1,1,1,2,1,2,1,1,2      },
160     {1176881,7,5,3,7,4,10,7,5,5,4        }
161};
162    private static Dataset defaultDataset;
163    private static IEnumerable<string> defaultAllowedInputVariables;
164    private static string defaultTargetVariable;
165    static ClassificationProblemData() {
166      defaultDataset = new Dataset(defaultVariableNames, defaultData);
167      defaultDataset.Name = "Wisconsin classification problem";
168      defaultDataset.Description = "subset from to ..";
169
170      defaultAllowedInputVariables = defaultVariableNames.Except(new List<string>() { "sample", "class" });
171      defaultTargetVariable = "class";
172    }
173    #endregion
174
175    #region propeties
176    [Storable]
177    private string targetVariable;
178    public string TargetVariable {
179      get { return targetVariable; }
180      set {
181        if (!Dataset.VariableNames.Contains(value))
182          throw new ArgumentException(string.Format("The target variable {0} is not present in the dataset", value));
183        if (targetVariable != value) {
184          targetVariable = value;
185
186          InitializeTargetVariableDependentMembers();
187          OnChanged();
188        }
189      }
190    }
191
192    private List<double> classValues;
193    [Storable]
194    public IEnumerable<double> ClassValues {
195      get { return classValues; }
196      private set { classValues = new List<double>(value); }
197    }
198
199    public int Classes {
200      get { return classValues.Count; }
201    }
202
203    private Dictionary<double, string> classValueClassNameMapping;
204    private Dictionary<string, double> classNameClassValueMapping;
205    [Storable(Name = "ClassNames")]
206    private IEnumerable<KeyValuePair<double, string>> classValueClassNameMappingStorable {
207      get { return classValueClassNameMapping; }
208      set {
209        classValueClassNameMapping = new Dictionary<double, string>();
210        foreach (var mapping in value)
211          classValueClassNameMapping[mapping.Key] = mapping.Value;
212
213        classNameClassValueMapping = classValueClassNameMapping.ToDictionary(x => x.Value, x => x.Key);
214      }
215    }
216    public IEnumerable<string> ClassNames {
217      get { return classValueClassNameMapping.Values; }
218    }
219
220    [Storable]
221    private Dictionary<Tuple<double, double>, double> misclassificationMatrix;
222    #endregion
223
224
225    [StorableConstructor]
226    protected ClassificationProblemData(bool deserializing) : base(deserializing) { }
227    protected ClassificationProblemData(ClassificationProblemData original, Cloner cloner) : base(original, cloner) { }
228    public override IDeepCloneable Clone(Cloner cloner) { return new ClassificationProblemData(this, cloner); }
229
230    public ClassificationProblemData()
231      : this(defaultDataset, defaultAllowedInputVariables, defaultTargetVariable) {
232    }
233
234    public ClassificationProblemData(Dataset dataset, IEnumerable<string> allowedInputVariables, string targetVariable)
235      : base(dataset, allowedInputVariables) {
236      TargetVariable = targetVariable;
237
238      InitializeTargetVariableDependentMembers();
239    }
240
241    private void InitializeTargetVariableDependentMembers() {
242      classValues = Dataset.GetEnumeratedVariableValues(targetVariable).Distinct().ToList();
243      classValues.Sort();
244
245      classValueClassNameMapping = new Dictionary<double, string>();
246      foreach (double classValue in ClassValues) {
247        classValueClassNameMapping.Add(classValue, "Class " + classValue);
248      }
249      classNameClassValueMapping = classValueClassNameMapping.ToDictionary(x => x.Value, x => x.Key);
250      OnClassNamesChanged();
251
252      //set default misclassification matrix
253      misclassificationMatrix = new Dictionary<Tuple<double, double>, double>();
254      foreach (double classValue1 in ClassValues) {
255        foreach (double classValue2 in ClassValues) {
256          double penalty = 1.0;
257          if (classValue1 == classValue2) penalty = 0.0;
258          SetClassificationPenalty(classValue1, classValue1, penalty);
259        }
260      }
261      OnClassificationPenaltyChanged();
262    }
263
264    public string GetClassName(double classValue) {
265      return classValueClassNameMapping[classValue];
266    }
267    public double GetClassValue(string className) {
268      return classNameClassValueMapping[className];
269    }
270    public void SetClassName(double classValue, string className) {
271      if (!classValues.Contains(classValue)) throw new ArgumentException();
272      if (classValueClassNameMapping[classValue] != className) {
273        classValueClassNameMapping[classValue] = className;
274        classNameClassValueMapping = classValueClassNameMapping.ToDictionary(x => x.Value, x => x.Key);
275      }
276    }
277
278    public double GetClassificationPenalty(string correctClassName, string estimatedClassName) {
279      return GetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName));
280    }
281    public double GetClassificationPenalty(double correctClassValue, double estimatedClassValue) {
282      var key = Tuple.Create(correctClassValue, estimatedClassValue);
283      return misclassificationMatrix[key];
284    }
285    public void SetClassificationPenalty(string correctClassName, string estimatedClassName, double penalty) {
286      SetClassificationPenalty(GetClassValue(correctClassName), GetClassValue(estimatedClassName), penalty);
287    }
288    public void SetClassificationPenalty(double correctClassValue, double estimatedClassValue, double penalty) {
289      var key = Tuple.Create(correctClassValue, estimatedClassValue);
290      if (!misclassificationMatrix.ContainsKey(key) || misclassificationMatrix[key] != penalty) {
291        misclassificationMatrix[key] = penalty;
292        OnClassificationPenaltyChanged();
293      }
294    }
295
296    public event EventHandler ClassNamesChanged;
297    protected virtual void OnClassNamesChanged() {
298      var handler = ClassNamesChanged;
299      if (handler != null) handler(this, EventArgs.Empty);
300    }
301    public event EventHandler ClassificationPenaltyChanged;
302    protected virtual void OnClassificationPenaltyChanged() {
303      var handler = ClassificationPenaltyChanged;
304      if (handler != null) handler(this, EventArgs.Empty);
305    }
306  }
307}
Note: See TracBrowser for help on using the repository browser.