1 | #region License Information
|
---|
2 | /* HeuristicLab
|
---|
3 | * Copyright (C) 2002-2010 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 | using System;
|
---|
22 | using System.Collections.Generic;
|
---|
23 | using System.ComponentModel;
|
---|
24 | using System.Drawing;
|
---|
25 | using System.Data;
|
---|
26 | using System.Linq;
|
---|
27 | using System.Text;
|
---|
28 | using System.Windows.Forms;
|
---|
29 | using HeuristicLab.MainForm;
|
---|
30 | using HeuristicLab.MainForm.WindowsForms;
|
---|
31 | using HeuristicLab.Data.Views;
|
---|
32 | using HeuristicLab.Data;
|
---|
33 | using HeuristicLab.Problems.DataAnalysis.Evaluators;
|
---|
34 | using HeuristicLab.Problems.DataAnalysis.MultiVariate.TimeSeriesPrognosis.Symbolic;
|
---|
35 |
|
---|
36 | namespace HeuristicLab.Problems.DataAnalysis.MultiVariate.TimeSeriesPrognosis.Views {
|
---|
37 | [Content(typeof(SymbolicTimeSeriesPrognosisSolution), true)]
|
---|
38 | [View("Time Series Prognosis Results View")]
|
---|
39 | public partial class ResultsView : AsynchronousContentView {
|
---|
40 | private List<string> rowNames = new List<string>() {
|
---|
41 | "Mean squared error",
|
---|
42 | "Pearson's R²",
|
---|
43 | "Mean relative error",
|
---|
44 | "Directional symmetry",
|
---|
45 | "Weighted directional symmetry",
|
---|
46 | "Theil's U"
|
---|
47 | };
|
---|
48 |
|
---|
49 | public ResultsView() {
|
---|
50 | InitializeComponent();
|
---|
51 | }
|
---|
52 |
|
---|
53 | public new SymbolicTimeSeriesPrognosisSolution Content {
|
---|
54 | get { return (SymbolicTimeSeriesPrognosisSolution)base.Content; }
|
---|
55 | set { base.Content = value; }
|
---|
56 | }
|
---|
57 |
|
---|
58 | protected override void RegisterContentEvents() {
|
---|
59 | base.RegisterContentEvents();
|
---|
60 | Content.ModelChanged += new EventHandler(Content_ModelChanged);
|
---|
61 | Content.ProblemDataChanged += new EventHandler(Content_ProblemDataChanged);
|
---|
62 | Content.EstimatedValuesChanged += new EventHandler(Content_EstimatedValuesChanged);
|
---|
63 | }
|
---|
64 | protected override void DeregisterContentEvents() {
|
---|
65 | base.DeregisterContentEvents();
|
---|
66 | Content.ModelChanged -= new EventHandler(Content_ModelChanged);
|
---|
67 | Content.ProblemDataChanged -= new EventHandler(Content_ProblemDataChanged);
|
---|
68 | Content.EstimatedValuesChanged -= new EventHandler(Content_EstimatedValuesChanged);
|
---|
69 | }
|
---|
70 |
|
---|
71 | private void Content_ModelChanged(object sender, EventArgs e) {
|
---|
72 | UpdateView();
|
---|
73 | }
|
---|
74 | private void Content_ProblemDataChanged(object sender, EventArgs e) {
|
---|
75 | UpdateView();
|
---|
76 | }
|
---|
77 | private void Content_EstimatedValuesChanged(object sender, EventArgs e) {
|
---|
78 | UpdateView();
|
---|
79 | }
|
---|
80 |
|
---|
81 | protected override void OnContentChanged() {
|
---|
82 | base.OnContentChanged();
|
---|
83 | UpdateView();
|
---|
84 | }
|
---|
85 | private void UpdateView() {
|
---|
86 | if (Content != null) {
|
---|
87 | List<string> targetVariables = Content.ProblemData.TargetVariables.CheckedItems.Select(x => x.Value.Value).ToList();
|
---|
88 | DoubleMatrix matrix = new DoubleMatrix(rowNames.Count, targetVariables.Count() * 2);
|
---|
89 | matrix.RowNames = rowNames;
|
---|
90 | matrix.ColumnNames = targetVariables.SelectMany(x => new List<string>() { x + " (training)", x + " (test)" });
|
---|
91 | matrix.SortableView = false;
|
---|
92 |
|
---|
93 | int trainingStart = Content.ProblemData.TrainingSamplesStart.Value;
|
---|
94 | int trainingEnd = Content.ProblemData.TrainingSamplesEnd.Value;
|
---|
95 | int testStart = Content.ProblemData.TestSamplesStart.Value;
|
---|
96 | int testEnd = Content.ProblemData.TestSamplesEnd.Value;
|
---|
97 | // create a list of time series evaluators for each target variable
|
---|
98 | Dictionary<string, List<IOnlineEvaluator>> trainingEvaluators =
|
---|
99 | new Dictionary<string, List<IOnlineEvaluator>>();
|
---|
100 | Dictionary<string, List<IOnlineEvaluator>> testEvaluators =
|
---|
101 | new Dictionary<string, List<IOnlineEvaluator>>();
|
---|
102 | foreach (string targetVariable in targetVariables) {
|
---|
103 | trainingEvaluators.Add(targetVariable, new List<IOnlineEvaluator>());
|
---|
104 | trainingEvaluators[targetVariable].Add(new OnlineMeanSquaredErrorEvaluator());
|
---|
105 | trainingEvaluators[targetVariable].Add(new OnlinePearsonsRSquaredEvaluator());
|
---|
106 | trainingEvaluators[targetVariable].Add(new OnlineMeanAbsolutePercentageErrorEvaluator());
|
---|
107 | trainingEvaluators[targetVariable].Add(new OnlineDirectionalSymmetryEvaluator());
|
---|
108 | trainingEvaluators[targetVariable].Add(new OnlineWeightedDirectionalSymmetryEvaluator());
|
---|
109 | trainingEvaluators[targetVariable].Add(new OnlineTheilsUStatisticEvaluator());
|
---|
110 |
|
---|
111 | testEvaluators.Add(targetVariable, new List<IOnlineEvaluator>());
|
---|
112 | testEvaluators[targetVariable].Add(new OnlineMeanSquaredErrorEvaluator());
|
---|
113 | testEvaluators[targetVariable].Add(new OnlinePearsonsRSquaredEvaluator());
|
---|
114 | testEvaluators[targetVariable].Add(new OnlineMeanAbsolutePercentageErrorEvaluator());
|
---|
115 | testEvaluators[targetVariable].Add(new OnlineDirectionalSymmetryEvaluator());
|
---|
116 | testEvaluators[targetVariable].Add(new OnlineWeightedDirectionalSymmetryEvaluator());
|
---|
117 | testEvaluators[targetVariable].Add(new OnlineTheilsUStatisticEvaluator());
|
---|
118 | }
|
---|
119 |
|
---|
120 | Evaluate(trainingStart, trainingEnd, trainingEvaluators);
|
---|
121 | Evaluate(testStart, testEnd, testEvaluators);
|
---|
122 |
|
---|
123 | int columnIndex = 0;
|
---|
124 | foreach (string targetVariable in targetVariables) {
|
---|
125 | int rowIndex = 0;
|
---|
126 | // training
|
---|
127 | foreach (var evaluator in trainingEvaluators[targetVariable]) {
|
---|
128 | matrix[rowIndex++, columnIndex] = evaluator.Value;
|
---|
129 | }
|
---|
130 | columnIndex++;
|
---|
131 | // test
|
---|
132 | rowIndex = 0;
|
---|
133 | foreach (var evaluator in testEvaluators[targetVariable]) {
|
---|
134 | matrix[rowIndex++, columnIndex] = evaluator.Value;
|
---|
135 | }
|
---|
136 | columnIndex++;
|
---|
137 | }
|
---|
138 |
|
---|
139 | matrixView.Content = matrix;
|
---|
140 | } else
|
---|
141 | matrixView.Content = null;
|
---|
142 | }
|
---|
143 |
|
---|
144 | private void Evaluate(int start, int end, Dictionary<string, List<IOnlineEvaluator>> evaluators) {
|
---|
145 |
|
---|
146 | for (int row = start; row < end; row++) {
|
---|
147 | if (string.IsNullOrEmpty(Content.ConditionalEvaluationVariable) || Content.ProblemData.Dataset[Content.ConditionalEvaluationVariable, row] != 0) {
|
---|
148 | // prepare evaluators for each target variable for a new prediction window
|
---|
149 | foreach (var entry in evaluators) {
|
---|
150 | double referenceOriginalValue = Content.ProblemData.Dataset[entry.Key, row - 1];
|
---|
151 | foreach (IOnlineTimeSeriesPrognosisEvaluator evaluator in entry.Value.OfType<IOnlineTimeSeriesPrognosisEvaluator>()) {
|
---|
152 | evaluator.StartNewPredictionWindow(referenceOriginalValue);
|
---|
153 | }
|
---|
154 | }
|
---|
155 | List<string> targetVariables = Content.ProblemData.TargetVariables.CheckedItems.Select(x => x.Value.Value).ToList();
|
---|
156 |
|
---|
157 | if (string.IsNullOrEmpty(Content.ConditionalEvaluationVariable) ||
|
---|
158 | Content.ProblemData.Dataset[Content.ConditionalEvaluationVariable, row] > 0) {
|
---|
159 | int timestep = 0;
|
---|
160 | foreach (double[] x in Content.GetPrognosis(row)) {
|
---|
161 | int targetIndex = 0;
|
---|
162 | if (row + timestep < Content.ProblemData.Dataset.Rows) {
|
---|
163 | foreach (var targetVariable in targetVariables) {
|
---|
164 | double originalValue = Content.ProblemData.Dataset[targetVariable, row + timestep];
|
---|
165 | double estimatedValue = x[targetIndex];
|
---|
166 | if (IsValidValue(originalValue) && IsValidValue(estimatedValue)) {
|
---|
167 | foreach (IOnlineEvaluator evaluator in evaluators[targetVariable]) {
|
---|
168 | evaluator.Add(originalValue, estimatedValue);
|
---|
169 | }
|
---|
170 | }
|
---|
171 | targetIndex++;
|
---|
172 | }
|
---|
173 | }
|
---|
174 | timestep++;
|
---|
175 | }
|
---|
176 | }
|
---|
177 | }
|
---|
178 | }
|
---|
179 | }
|
---|
180 |
|
---|
181 | private bool IsValidValue(double d) {
|
---|
182 | return !(double.IsNaN(d) || double.IsInfinity(d));
|
---|
183 | }
|
---|
184 | }
|
---|
185 | }
|
---|