Free cookie consent management tool by TermsFeed Policy Generator

source: branches/DataPreprocessing/HeuristicLab.DataPreprocessing/3.4/Implementations/SearchLogic.cs @ 10978

Last change on this file since 10978 was 10978, checked in by mleitner, 10 years ago

Refactoring

File size: 5.6 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2013 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;
26
27namespace HeuristicLab.DataPreprocessing {
28  public class SearchLogic : ISearchLogic {
29    private readonly ITransactionalPreprocessingData preprocessingData;
30
31    private Dictionary<int, IList<int>> MissingValueIndicies { get; set; }
32    private Dictionary<int, IList> ValuesWithoutNaN { get; set; }
33
34    public SearchLogic(ITransactionalPreprocessingData thePreprocessingData) {
35      preprocessingData = thePreprocessingData;
36
37      MissingValueIndicies = new Dictionary<int, IList<int>>();
38      ValuesWithoutNaN = new Dictionary<int, IList>();
39
40      preprocessingData.Changed += preprocessingData_Changed;
41    }
42
43    void preprocessingData_Changed(object sender, DataPreprocessingChangedEventArgs e)
44    {
45      switch (e.Type) {
46        case DataPreprocessingChangedEventType.DeleteColumn:
47        case DataPreprocessingChangedEventType.ChangeColumn:
48          MissingValueIndicies.Remove(e.Column);
49          ValuesWithoutNaN.Remove(e.Column);
50          break;
51        case DataPreprocessingChangedEventType.AddColumn:
52          //cache does not need to be updated, will be calculated the first time it is requested
53          break;
54        case DataPreprocessingChangedEventType.DeleteRow:
55        case DataPreprocessingChangedEventType.AddRow:
56        case DataPreprocessingChangedEventType.ChangeItem:
57        case DataPreprocessingChangedEventType.Any:
58        case DataPreprocessingChangedEventType.Transformation:
59        default:
60          MissingValueIndicies = new Dictionary<int, IList<int>>();
61          ValuesWithoutNaN = new Dictionary<int, IList>();
62          break;
63      }
64    }
65
66    public IDictionary<int, IList<int>> GetMissingValueIndices() {
67      var dic = new Dictionary<int, IList<int>>();
68      for (int i = 0; i < preprocessingData.Columns; ++i) {
69        dic.Add(i, GetMissingValueIndices(i));
70      }
71      return dic;
72    }
73
74    public bool IsMissingValue(int columnIndex, int rowIndex) {
75      if (preprocessingData.IsType<double>(columnIndex)) {
76        return double.IsNaN(preprocessingData.GetCell<double>(columnIndex, rowIndex));
77      } else if (preprocessingData.IsType<string>(columnIndex)) {
78        return string.IsNullOrEmpty(preprocessingData.GetCell<string>(columnIndex, rowIndex));
79      } else if (preprocessingData.IsType<DateTime>(columnIndex)) {
80        return preprocessingData.GetCell<DateTime>(columnIndex, rowIndex).Equals(DateTime.MinValue);
81      } else {
82        throw new ArgumentException("cell in column " + columnIndex + " and row index " + rowIndex + " contains a non supported type.");
83      }
84    }
85
86    public IList<int> GetMissingValueIndices(int columnIndex) {
87      if (!MissingValueIndicies.ContainsKey(columnIndex)){       
88          if (preprocessingData.IsType<double>(columnIndex)) {
89            MissingValueIndicies[columnIndex] = GetMissingValueIndices<double>(columnIndex);
90          } else if (preprocessingData.IsType<string>(columnIndex)) {
91            MissingValueIndicies[columnIndex] = GetMissingValueIndices<string>(columnIndex);
92          } else if (preprocessingData.IsType<DateTime>(columnIndex)) {
93            MissingValueIndicies[columnIndex] = GetMissingValueIndices<DateTime>(columnIndex);
94          } else {
95            throw new ArgumentException("column " + columnIndex + " contains a non supported type.");
96          }
97      }
98
99      return MissingValueIndicies[columnIndex];
100   }
101    private IList<int> GetMissingValueIndices<T>(int columnIndex) {
102      List<int> missingIndices = new List<int>();
103     
104      for(int row = 0; row < preprocessingData.Rows; ++row) {
105        if (IsMissingValue(columnIndex, row)) {
106          missingIndices.Add(row);
107        }
108      }
109
110      return missingIndices;
111    }
112
113    public IEnumerable<T> GetValuesWithoutNaN<T>(int columnIndex, bool considerSelection)
114    {
115      if (considerSelection) {     
116        var selectedRows =  preprocessingData.Selection[columnIndex];
117       
118        List<T> values = new List<T>();
119        foreach (var rowIdx in selectedRows) {
120          if (!IsMissingValue(columnIndex, rowIdx)) {
121            values.Add(preprocessingData.GetCell<T>(columnIndex, rowIdx));
122          }
123        }
124        return values;
125      } else {
126        if (!ValuesWithoutNaN.ContainsKey(columnIndex)) {
127          List<T> values = new List<T>();
128
129          for (int row = 0; row < preprocessingData.Rows; ++row) {
130            if (!IsMissingValue(columnIndex, row)) {
131              values.Add(preprocessingData.GetCell<T>(columnIndex, row));
132            }
133          }
134
135          ValuesWithoutNaN[columnIndex] = values;
136        }
137        return (IEnumerable<T>)ValuesWithoutNaN[columnIndex];
138      }
139    }
140  }
141}
Note: See TracBrowser for help on using the repository browser.