Free cookie consent management tool by TermsFeed Policy Generator

source: stable/HeuristicLab.DataPreprocessing/3.4/PreprocessingTransformator.cs @ 15231

Last change on this file since 15231 was 15142, checked in by gkronber, 7 years ago

#2697: merged r14843 (resolving conflicts in csproj file for HL.Algorithms.DataAnalysis because MCTS has been removed)

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 ITransactionalPreprocessingData 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 = (ITransactionalPreprocessingData)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      }
71      catch (Exception e) {
72        preprocessingData.Undo();
73        if (string.IsNullOrEmpty(errorMsg)) errorMsg = e.Message;
74      }
75      finally {
76        preprocessingData.EndTransaction();
77      }
78
79      return success;
80    }
81
82    private void PreserveColumns(IEnumerable<Transformation<double>> transformations) {
83      foreach (var transformation in transformations) {
84        if (!originalColumns.ContainsKey(transformation.Column)) {
85          int colIndex = preprocessingData.GetColumnIndex(transformation.Column);
86          var originalData = preprocessingData.GetValues<double>(colIndex);
87          originalColumns.Add(transformation.Column, originalData);
88        }
89      }
90    }
91
92    private void ApplyDoubleTranformationsInplace(IEnumerable<Transformation<double>> transformations, bool preserveColumns, out bool success, out string errorMsg) {
93      errorMsg = string.Empty;
94      success = true;
95      foreach (var transformation in transformations) {
96        int colIndex = preprocessingData.GetColumnIndex(transformation.Column);
97
98        var originalData = preprocessingData.GetValues<double>(colIndex);
99
100        string errorMsgPart;
101        bool successPart;
102        var transformedData = ApplyDoubleTransformation(transformation, originalData, out successPart, out errorMsgPart);
103        errorMsg += errorMsgPart + Environment.NewLine;
104
105        if (!successPart) success = false;
106        preprocessingData.SetValues(colIndex, transformedData.ToList());
107        preprocessingData.Transformations.Add(transformation);
108      }
109    }
110
111    private IEnumerable<double> ApplyDoubleTransformation(Transformation<double> transformation, IList<double> data, out bool success, out string errorMsg) {
112      success = transformation.Check(data, out errorMsg);
113      // don't apply when the check fails
114      if (success)
115        return transformation.ConfigureAndApply(data);
116      else
117        return data;
118    }
119
120    private void RenameTransformationColumnParameter(List<Transformation<double>> transformations) {
121      foreach (var transformation in transformations) {
122        var newColumnName = new StringValue(renamedColumns[transformation.Column]);
123        transformation.ColumnParameter.ValidValues.Add(newColumnName);
124        transformation.ColumnParameter.Value = newColumnName;
125      }
126    }
127
128    private void InsertCopyColumTransformations(IList<Transformation<double>> transformations) {
129      foreach (var renaming in renamedColumns) {
130        string oldName = renaming.Key;
131        string newName = renaming.Value;
132
133        var copyTransformation = CreateCopyTransformation(oldName, newName);
134        preprocessingData.Transformations.Insert(0, copyTransformation);
135      }
136    }
137
138    private CopyColumnTransformation CreateCopyTransformation(string oldColumn, string newColumn) {
139      var newColumName = new StringValue(newColumn);
140
141      var copyTransformation = new CopyColumnTransformation();
142      copyTransformation.ColumnParameter.ValidValues.Add(newColumName);
143      copyTransformation.ColumnParameter.Value = newColumName;
144
145      copyTransformation.CopiedColumnNameParameter.Value.Value = oldColumn;
146      return copyTransformation;
147    }
148
149    private void RenameTransformedColumsAndRestorePreservedColumns(IList<Transformation<double>> transformations) {
150      foreach (var column in originalColumns) {
151        int originalColumnIndex = preprocessingData.GetColumnIndex(column.Key);
152        int newColumnIndex = originalColumnIndex + 1;
153        string newColumnName = GetTransformatedColumnName(transformations, column.Key);
154        // save renaming mapping
155        renamedColumns[column.Key] = newColumnName;
156        // create new transformed column
157        preprocessingData.InsertColumn<double>(newColumnName, newColumnIndex);
158        preprocessingData.SetValues(newColumnIndex, preprocessingData.GetValues<double>(originalColumnIndex));
159        // restore old values
160        preprocessingData.SetValues(originalColumnIndex, column.Value);
161      }
162    }
163
164    private string GetTransformatedColumnName(IList<Transformation<double>> transformations, string column) {
165      string suffix = GetTransformationSuffix(transformations, column);
166      return column + "_" + suffix;
167    }
168
169    private string GetTransformationSuffix(IList<Transformation<double>> transformations, string column) {
170      var suffixes = transformations.Where(t => t.Column == column).Select(t => t.ShortName);
171      var builder = new StringBuilder();
172      foreach (var suffix in suffixes) {
173        builder.Append(suffix);
174      }
175      return builder.ToString();
176    }
177  }
178}
Note: See TracBrowser for help on using the repository browser.