source: trunk/sources/HeuristicLab.DataPreprocessing/3.4/PreprocessingTransformator.cs @ 15518

Last change on this file since 15518 was 15518, checked in by pfleck, 22 months ago

#2809: merged branch to trunk

File size: 7.3 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2016 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 System.Text;
26using HeuristicLab.Data;
27using HeuristicLab.Problems.DataAnalysis;
28
29namespace HeuristicLab.DataPreprocessing {
30  public class PreprocessingTransformator {
31    private readonly IPreprocessingData preprocessingData;
32
33    private readonly IDictionary<string, IList<double>> originalColumns;
34
35    private readonly IDictionary<string, string> renamedColumns;
36
37    public PreprocessingTransformator(IPreprocessingData preprocessingData) {
38      this.preprocessingData = preprocessingData;
39      originalColumns = new Dictionary<string, IList<double>>();
40      renamedColumns = new Dictionary<string, string>();
41    }
42
43    public bool ApplyTransformations(IEnumerable<ITransformation> transformations, bool preserveColumns, out string errorMsg) {
44      bool success = false;
45      errorMsg = string.Empty;
46      preprocessingData.BeginTransaction(DataPreprocessingChangedEventType.Transformation);
47
48      try {
49        var doubleTransformations = transformations.OfType<Transformation<double>>().ToList();
50
51        if (preserveColumns) {
52          PreserveColumns(doubleTransformations);
53        }
54
55        // all transformations are performed inplace. no creation of new columns for transformations
56        ApplyDoubleTranformationsInplace(doubleTransformations, preserveColumns, out success, out errorMsg);
57
58        if (preserveColumns) {
59          RenameTransformedColumsAndRestorePreservedColumns(doubleTransformations);
60          RenameTransformationColumnParameter(doubleTransformations);
61          InsertCopyColumTransformations(doubleTransformations);
62
63          originalColumns.Clear();
64          renamedColumns.Clear();
65        }
66        // only accept changes if everything was successful
67        if (!success) {
68          preprocessingData.Undo();
69        }
70      } catch (Exception e) {
71        preprocessingData.Undo();
72        if (string.IsNullOrEmpty(errorMsg)) errorMsg = e.Message;
73      } finally {
74        preprocessingData.EndTransaction();
75      }
76
77      return success;
78    }
79
80    private void PreserveColumns(IEnumerable<Transformation<double>> transformations) {
81      foreach (var transformation in transformations) {
82        if (!originalColumns.ContainsKey(transformation.Column)) {
83          int colIndex = preprocessingData.GetColumnIndex(transformation.Column);
84          var originalData = preprocessingData.GetValues<double>(colIndex);
85          originalColumns.Add(transformation.Column, originalData);
86        }
87      }
88    }
89
90    private void ApplyDoubleTranformationsInplace(IEnumerable<Transformation<double>> transformations, bool preserveColumns, out bool success, out string errorMsg) {
91      errorMsg = string.Empty;
92      success = true;
93      foreach (var transformation in transformations) {
94        int colIndex = preprocessingData.GetColumnIndex(transformation.Column);
95
96        var originalData = preprocessingData.GetValues<double>(colIndex);
97
98        string errorMsgPart;
99        bool successPart;
100        var transformedData = ApplyDoubleTransformation(transformation, originalData, out successPart, out errorMsgPart);
101        errorMsg += errorMsgPart + Environment.NewLine;
102
103        if (!successPart) success = false;
104        preprocessingData.SetValues(colIndex, transformedData.ToList());
105        preprocessingData.Transformations.Add(transformation);
106      }
107    }
108
109    private IEnumerable<double> ApplyDoubleTransformation(Transformation<double> transformation, IList<double> data, out bool success, out string errorMsg) {
110      success = transformation.Check(data, out errorMsg);
111      // don't apply when the check fails
112      if (success)
113        return transformation.ConfigureAndApply(data);
114      else
115        return data;
116    }
117
118    private void RenameTransformationColumnParameter(List<Transformation<double>> transformations) {
119      foreach (var transformation in transformations) {
120        var newColumnName = new StringValue(renamedColumns[transformation.Column]);
121        transformation.ColumnParameter.ValidValues.Add(newColumnName);
122        transformation.ColumnParameter.Value = newColumnName;
123      }
124    }
125
126    private void InsertCopyColumTransformations(IList<Transformation<double>> transformations) {
127      foreach (var renaming in renamedColumns) {
128        string oldName = renaming.Key;
129        string newName = renaming.Value;
130
131        var copyTransformation = CreateCopyTransformation(oldName, newName);
132        preprocessingData.Transformations.Insert(0, copyTransformation);
133      }
134    }
135
136    private CopyColumnTransformation CreateCopyTransformation(string oldColumn, string newColumn) {
137      var newColumName = new StringValue(newColumn);
138
139      var copyTransformation = new CopyColumnTransformation();
140      copyTransformation.ColumnParameter.ValidValues.Add(newColumName);
141      copyTransformation.ColumnParameter.Value = newColumName;
142
143      copyTransformation.CopiedColumnNameParameter.Value.Value = oldColumn;
144      return copyTransformation;
145    }
146
147    private void RenameTransformedColumsAndRestorePreservedColumns(IList<Transformation<double>> transformations) {
148      foreach (var column in originalColumns) {
149        int originalColumnIndex = preprocessingData.GetColumnIndex(column.Key);
150        int newColumnIndex = originalColumnIndex + 1;
151        string newColumnName = GetTransformatedColumnName(transformations, column.Key);
152        // save renaming mapping
153        renamedColumns[column.Key] = newColumnName;
154        // create new transformed column
155        preprocessingData.InsertColumn<double>(newColumnName, newColumnIndex);
156        preprocessingData.SetValues(newColumnIndex, preprocessingData.GetValues<double>(originalColumnIndex));
157        // restore old values
158        preprocessingData.SetValues(originalColumnIndex, column.Value);
159      }
160    }
161
162    private string GetTransformatedColumnName(IList<Transformation<double>> transformations, string column) {
163      string suffix = GetTransformationSuffix(transformations, column);
164      return column + "_" + suffix;
165    }
166
167    private string GetTransformationSuffix(IList<Transformation<double>> transformations, string column) {
168      var suffixes = transformations.Where(t => t.Column == column).Select(t => t.ShortName);
169      var builder = new StringBuilder();
170      foreach (var suffix in suffixes) {
171        builder.Append(suffix);
172      }
173      return builder.ToString();
174    }
175  }
176}
Note: See TracBrowser for help on using the repository browser.