[10163] | 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 |
|
---|
| 22 | using System;
|
---|
[10168] | 23 | using System.Collections;
|
---|
[10163] | 24 | using System.Collections.Generic;
|
---|
[10185] | 25 | using System.Linq;
|
---|
| 26 | using HeuristicLab.Common;
|
---|
[10163] | 27 | using HeuristicLab.Core;
|
---|
[10220] | 28 | using HeuristicLab.Data;
|
---|
[10163] | 29 | using HeuristicLab.Problems.DataAnalysis;
|
---|
[10772] | 30 | using HeuristicLab.Problems.DataAnalysis.Transformations;
|
---|
[10163] | 31 |
|
---|
[10182] | 32 | namespace HeuristicLab.DataPreprocessing {
|
---|
[10550] | 33 |
|
---|
[10163] | 34 | [Item("PreprocessingData", "Represents data used for preprocessing.")]
|
---|
[10978] | 35 | public abstract class PreprocessingData : NamedItem, IPreprocessingData {
|
---|
[10163] | 36 |
|
---|
[10978] | 37 | protected double trainingToTestRatio;
|
---|
| 38 | public IntRange TrainingPartition {
|
---|
| 39 | get { return new IntRange(0, (int)(Rows * trainingToTestRatio)); }
|
---|
| 40 | }
|
---|
| 41 | public IntRange TestPartition {
|
---|
| 42 | get { return new IntRange((int)(Rows * trainingToTestRatio), Rows); }
|
---|
| 43 | }
|
---|
| 44 |
|
---|
| 45 | protected IList<ITransformation> transformations;
|
---|
| 46 | public IList<ITransformation> Transformations {
|
---|
| 47 | get { return transformations; }
|
---|
| 48 | }
|
---|
| 49 |
|
---|
[10740] | 50 | protected IList<IList> variableValues;
|
---|
[10586] | 51 | protected IList<string> variableNames;
|
---|
[10168] | 52 |
|
---|
[10978] | 53 | public IEnumerable<string> VariableNames {
|
---|
| 54 | get { return variableNames; }
|
---|
| 55 | }
|
---|
[10186] | 56 |
|
---|
[10992] | 57 | public IEnumerable<string> GetDoubleVariableNames() {
|
---|
| 58 | var doubleVariableNames = new List<string>();
|
---|
| 59 | for (int i = 0; i < Columns; ++i) {
|
---|
| 60 | if (IsType<double>(i)) {
|
---|
| 61 | doubleVariableNames.Add(variableNames[i]);
|
---|
| 62 | }
|
---|
| 63 | }
|
---|
| 64 | return doubleVariableNames;
|
---|
| 65 | }
|
---|
| 66 |
|
---|
[10978] | 67 | public int Columns {
|
---|
| 68 | get { return variableNames.Count; }
|
---|
| 69 | }
|
---|
[10695] | 70 |
|
---|
[10978] | 71 | public int Rows {
|
---|
| 72 | get { return variableValues.Count > 0 ? variableValues[0].Count : 0; }
|
---|
| 73 | }
|
---|
[10804] | 74 |
|
---|
[10978] | 75 | protected IDictionary<int, IList<int>> selection;
|
---|
| 76 | public IDictionary<int, IList<int>> Selection {
|
---|
| 77 | get { return selection; }
|
---|
| 78 | set {
|
---|
[10992] | 79 | selection = value;
|
---|
| 80 | OnSelectionChanged();
|
---|
| 81 | }
|
---|
| 82 | }
|
---|
[10978] | 83 |
|
---|
[10586] | 84 | protected PreprocessingData(PreprocessingData original, Cloner cloner)
|
---|
[10185] | 85 | : base(original, cloner) {
|
---|
[10550] | 86 | variableValues = CopyVariableValues(original.variableValues);
|
---|
[10548] | 87 | variableNames = new List<string>(original.variableNames);
|
---|
| 88 | trainingToTestRatio = original.trainingToTestRatio;
|
---|
[10842] | 89 | transformations = new List<ITransformation>();
|
---|
[10185] | 90 | }
|
---|
[10187] | 91 |
|
---|
[10978] | 92 | protected PreprocessingData(IDataAnalysisProblemData problemData)
|
---|
[10168] | 93 | : base() {
|
---|
[10786] | 94 | Name = "Preprocessing Data";
|
---|
[10168] | 95 |
|
---|
[10786] | 96 | transformations = new List<ITransformation>();
|
---|
[10978] | 97 | selection = new Dictionary<int, IList<int>>();
|
---|
[10786] | 98 |
|
---|
[10187] | 99 | variableNames = new List<string>(problemData.Dataset.VariableNames);
|
---|
| 100 |
|
---|
[10367] | 101 | int columnIndex = 0;
|
---|
[10740] | 102 | variableValues = new List<IList>();
|
---|
[10185] | 103 | foreach (var variableName in problemData.Dataset.VariableNames) {
|
---|
| 104 | if (problemData.Dataset.IsType<double>(variableName)) {
|
---|
[10740] | 105 | variableValues.Insert(columnIndex, problemData.Dataset.GetDoubleValues(variableName).ToList());
|
---|
[10185] | 106 | } else if (problemData.Dataset.IsType<string>(variableName)) {
|
---|
[10740] | 107 | variableValues.Insert(columnIndex, CreateColumn<string>(problemData.Dataset, columnIndex, x => x));
|
---|
[10185] | 108 | } else if (problemData.Dataset.IsType<DateTime>(variableName)) {
|
---|
[10740] | 109 | variableValues.Insert(columnIndex, CreateColumn<DateTime>(problemData.Dataset, columnIndex, x => DateTime.Parse(x)));
|
---|
[10168] | 110 | } else {
|
---|
[10978] | 111 | throw new ArgumentException("The datatype of column " + variableName + " must be of type double, string or DateTime");
|
---|
[10168] | 112 | }
|
---|
[10367] | 113 | ++columnIndex;
|
---|
[10168] | 114 | }
|
---|
[10185] | 115 |
|
---|
[10235] | 116 | trainingToTestRatio = (double)problemData.TrainingPartition.Size / Math.Max(problemData.Dataset.Rows, double.Epsilon);
|
---|
[10163] | 117 | }
|
---|
| 118 |
|
---|
[10185] | 119 | private static IList CreateColumn<T>(Dataset ds, int column, Func<string, T> selector) {
|
---|
| 120 | var list = new List<T>(ds.Rows);
|
---|
[10341] | 121 | for (int row = 0; row < ds.Rows; ++row) {
|
---|
[10367] | 122 | list.Add(selector(ds.GetValue(row, column)));
|
---|
[10185] | 123 | }
|
---|
| 124 | return list;
|
---|
| 125 | }
|
---|
| 126 |
|
---|
[10740] | 127 | protected IList<IList> CopyVariableValues(IList<IList> original) {
|
---|
[10783] | 128 | var copy = new List<IList>(original);
|
---|
[10740] | 129 | for (int i = 0; i < original.Count; ++i) {
|
---|
[10783] | 130 | copy[i] = (IList)Activator.CreateInstance(original[i].GetType(), original[i]);
|
---|
[10550] | 131 | }
|
---|
| 132 | return copy;
|
---|
| 133 | }
|
---|
| 134 |
|
---|
[10163] | 135 |
|
---|
| 136 | #region IPreprocessingData Members
|
---|
| 137 |
|
---|
[10991] | 138 | public abstract T GetCell<T>(int columnIndex, int rowIndex);
|
---|
[10181] | 139 |
|
---|
[10991] | 140 | public abstract void SetCell<T>(int columnIndex, int rowIndex, T value);
|
---|
[10367] | 141 |
|
---|
[10991] | 142 | public abstract string GetCellAsString(int columnIndex, int rowIndex);
|
---|
[10367] | 143 |
|
---|
[10991] | 144 | public abstract string GetVariableName(int columnIndex);
|
---|
[10547] | 145 |
|
---|
[10991] | 146 | public abstract int GetColumnIndex(string variableName);
|
---|
[10978] | 147 |
|
---|
[10991] | 148 | public abstract bool IsType<T>(int columnIndex);
|
---|
[10978] | 149 |
|
---|
[10367] | 150 | [Obsolete("use the index based variant, is faster")]
|
---|
[10991] | 151 | public abstract IList<T> GetValues<T>(string variableName, bool considerSelection);
|
---|
[10181] | 152 |
|
---|
[10991] | 153 | public abstract IList<T> GetValues<T>(int columnIndex, bool considerSelection);
|
---|
[10367] | 154 |
|
---|
[10991] | 155 | public abstract void SetValues<T>(int columnIndex, IList<T> values);
|
---|
[10181] | 156 |
|
---|
[10991] | 157 | public abstract void InsertRow(int rowIndex);
|
---|
[10163] | 158 |
|
---|
[10991] | 159 | public abstract void DeleteRow(int rowIndex);
|
---|
[10163] | 160 |
|
---|
[10991] | 161 | public abstract void InsertColumn<T>(string variableName, int columnIndex);
|
---|
[10163] | 162 |
|
---|
[10991] | 163 | public abstract void DeleteColumn(int columnIndex);
|
---|
[10367] | 164 |
|
---|
[10991] | 165 | public abstract Dataset ExportToDataset();
|
---|
[10367] | 166 |
|
---|
[10991] | 167 | public abstract void ClearSelection();
|
---|
[10220] | 168 |
|
---|
[10991] | 169 | public abstract event EventHandler SelectionChanged;
|
---|
| 170 | protected abstract void OnSelectionChanged();
|
---|
[10220] | 171 |
|
---|
[10978] | 172 | public event DataPreprocessingChangedEventHandler Changed;
|
---|
[10992] | 173 | protected virtual void OnChanged(DataPreprocessingChangedEventType type, int column, int row) {
|
---|
[10978] | 174 | var listeners = Changed;
|
---|
| 175 | if (listeners != null) listeners(this, new DataPreprocessingChangedEventArgs(type, column, row));
|
---|
[10804] | 176 | }
|
---|
[10220] | 177 | #endregion
|
---|
[10163] | 178 | }
|
---|
| 179 | }
|
---|