source: stable/HeuristicLab.DataPreprocessing/3.4/Implementations/PreprocessingData.cs @ 13881

Last change on this file since 13881 was 13881, checked in by mkommend, 5 years ago

#2536: Merged r13427 and r13539 into stable.

File size: 7.4 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2015 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;
24using System.Collections.Generic;
25using System.Linq;
26using HeuristicLab.Common;
27using HeuristicLab.Core;
28using HeuristicLab.Data;
29using HeuristicLab.Problems.DataAnalysis;
30
31namespace HeuristicLab.DataPreprocessing {
32
33  [Item("PreprocessingData", "Represents data used for preprocessing.")]
34  public abstract class PreprocessingData : NamedItem, IPreprocessingData {
35
36    public IntRange TrainingPartition { get; set; }
37    public IntRange TestPartition { get; set; }
38
39    protected IList<ITransformation> transformations;
40    public IList<ITransformation> Transformations {
41      get { return transformations; }
42    }
43
44    protected IList<IList> variableValues;
45    protected IList<string> variableNames;
46
47    public IEnumerable<string> VariableNames {
48      get { return variableNames; }
49    }
50
51    public IEnumerable<string> GetDoubleVariableNames() {
52      var doubleVariableNames = new List<string>();
53      for (int i = 0; i < Columns; ++i) {
54        if (VariableHasType<double>(i)) {
55          doubleVariableNames.Add(variableNames[i]);
56        }
57      }
58      return doubleVariableNames;
59    }
60
61    public int Columns {
62      get { return variableNames.Count; }
63    }
64
65    public int Rows {
66      get { return variableValues.Count > 0 ? variableValues[0].Count : 0; }
67    }
68
69    protected IDictionary<int, IList<int>> selection;
70    public IDictionary<int, IList<int>> Selection {
71      get { return selection; }
72      set {
73        selection = value;
74        OnSelectionChanged();
75      }
76    }
77
78    protected PreprocessingData(PreprocessingData original, Cloner cloner)
79      : base(original, cloner) {
80      variableValues = CopyVariableValues(original.variableValues);
81      variableNames = new List<string>(original.variableNames);
82      TrainingPartition = (IntRange)original.TrainingPartition.Clone(cloner);
83      TestPartition = (IntRange)original.TestPartition.Clone(cloner);
84      transformations = new List<ITransformation>();
85
86      RegisterEventHandler();
87    }
88
89    protected PreprocessingData(IDataAnalysisProblemData problemData)
90      : base() {
91      Name = "Preprocessing Data";
92
93      transformations = new List<ITransformation>();
94      selection = new Dictionary<int, IList<int>>();
95
96      Dataset dataset = (Dataset)problemData.Dataset;
97      variableNames = new List<string>(problemData.Dataset.VariableNames);
98
99      int columnIndex = 0;
100      variableValues = new List<IList>();
101      foreach (var variableName in problemData.Dataset.VariableNames) {
102        if (dataset.VariableHasType<double>(variableName)) {
103          variableValues.Insert(columnIndex, dataset.GetDoubleValues(variableName).ToList());
104        } else if (dataset.VariableHasType<string>(variableName)) {
105          variableValues.Insert(columnIndex, dataset.GetStringValues(variableName).ToList());
106        } else if (dataset.VariableHasType<DateTime>(variableName)) {
107          variableValues.Insert(columnIndex, dataset.GetDateTimeValues(variableName).ToList());
108        } else {
109          throw new ArgumentException("The datatype of column " + variableName + " must be of type double, string or DateTime");
110        }
111        ++columnIndex;
112      }
113
114      TrainingPartition = new IntRange(problemData.TrainingPartition.Start, problemData.TrainingPartition.End);
115      TestPartition = new IntRange(problemData.TestPartition.Start, problemData.TestPartition.End);
116
117      RegisterEventHandler();
118    }
119
120    private void RegisterEventHandler() {
121      Changed += (s, e) => {
122        switch (e.Type) {
123          case DataPreprocessingChangedEventType.DeleteRow:
124            CheckPartitionRanges();
125            break;
126          case DataPreprocessingChangedEventType.Any:
127            CheckPartitionRanges();
128            break;
129          case DataPreprocessingChangedEventType.Transformation:
130            CheckPartitionRanges();
131            break;
132        }
133      };
134    }
135
136    private void CheckPartitionRanges() {
137      int maxRowIndex = Math.Max(0, Rows - 1);
138      TrainingPartition.Start = Math.Min(TrainingPartition.Start, maxRowIndex);
139      TrainingPartition.End = Math.Min(TrainingPartition.End, maxRowIndex);
140      TestPartition.Start = Math.Min(TestPartition.Start, maxRowIndex);
141      TestPartition.End = Math.Min(TestPartition.End, maxRowIndex);
142    }
143
144    protected IList<IList> CopyVariableValues(IList<IList> original) {
145      var copy = new List<IList>(original);
146      for (int i = 0; i < original.Count; ++i) {
147        copy[i] = (IList)Activator.CreateInstance(original[i].GetType(), original[i]);
148      }
149      return copy;
150    }
151
152
153    #region IPreprocessingData Members
154
155    public abstract T GetCell<T>(int columnIndex, int rowIndex);
156
157    public abstract void SetCell<T>(int columnIndex, int rowIndex, T value);
158
159    public abstract string GetCellAsString(int columnIndex, int rowIndex);
160
161    public abstract string GetVariableName(int columnIndex);
162
163    public abstract int GetColumnIndex(string variableName);
164
165    public abstract bool VariableHasType<T>(int columnIndex);
166
167    [Obsolete("use the index based variant, is faster")]
168    public abstract IList<T> GetValues<T>(string variableName, bool considerSelection);
169
170    public abstract IList<T> GetValues<T>(int columnIndex, bool considerSelection);
171
172    public abstract void SetValues<T>(int columnIndex, IList<T> values);
173
174    public abstract bool SetValue(string value, int columnIndex, int rowIndex);
175
176    public abstract bool Validate(string value, out string errorMessage, int columnIndex);
177
178    public abstract bool AreAllStringColumns(IEnumerable<int> columnIndices);
179
180    public abstract void DeleteRowsWithIndices(IEnumerable<int> rows);
181
182    public abstract void InsertRow(int rowIndex);
183
184    public abstract void DeleteRow(int rowIndex);
185
186    public abstract void InsertColumn<T>(string variableName, int columnIndex);
187
188    public abstract void DeleteColumn(int columnIndex);
189
190    public abstract void RenameColumn(int columnIndex, string name);
191    public abstract void RenameColumns(IList<string> list);
192
193    public abstract Dataset ExportToDataset();
194
195    public abstract void ClearSelection();
196
197    public abstract event EventHandler SelectionChanged;
198    protected abstract void OnSelectionChanged();
199
200    public event DataPreprocessingChangedEventHandler Changed;
201    protected virtual void OnChanged(DataPreprocessingChangedEventType type, int column, int row) {
202      var listeners = Changed;
203      if (listeners != null) listeners(this, new DataPreprocessingChangedEventArgs(type, column, row));
204    }
205    #endregion
206  }
207}
Note: See TracBrowser for help on using the repository browser.