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 |
|
---|
22 | using System;
|
---|
23 | using System.Collections.Generic;
|
---|
24 | using System.ComponentModel;
|
---|
25 | using System.Drawing;
|
---|
26 | using System.Linq;
|
---|
27 | using System.Text;
|
---|
28 | using System.Windows.Forms;
|
---|
29 | using HeuristicLab.Common;
|
---|
30 | using HeuristicLab.Core;
|
---|
31 | using HeuristicLab.Core.Views;
|
---|
32 | using HeuristicLab.Data;
|
---|
33 | using HeuristicLab.MainForm;
|
---|
34 | using HeuristicLab.Optimization;
|
---|
35 | using HeuristicLab.Problems.DataAnalysis;
|
---|
36 | using HeuristicLab.Visualization;
|
---|
37 | using Ellipse = HeuristicLab.Visualization.Ellipse;
|
---|
38 | using Rectangle = HeuristicLab.Visualization.Rectangle;
|
---|
39 |
|
---|
40 | namespace HeuristicLab.VariableInteractionNetworks.Views {
|
---|
41 | [View("Variable Interaction Network")]
|
---|
42 | [Content(typeof(RunCollection), IsDefaultView = false)]
|
---|
43 |
|
---|
44 | public sealed partial class RunCollectionVariableInteractionNetworkView : ItemView {
|
---|
45 | public RunCollectionVariableInteractionNetworkView() {
|
---|
46 | InitializeComponent();
|
---|
47 | ConfigureNodeShapes();
|
---|
48 | }
|
---|
49 |
|
---|
50 | public new RunCollection Content {
|
---|
51 | get { return (RunCollection)base.Content; }
|
---|
52 | set {
|
---|
53 | if (value != null && value != Content) {
|
---|
54 | base.Content = value;
|
---|
55 | }
|
---|
56 | }
|
---|
57 | }
|
---|
58 |
|
---|
59 | private VariableInteractionNetwork variableInteractionNetwork;
|
---|
60 |
|
---|
61 | private static void AssertSameProblemData(RunCollection runs) {
|
---|
62 | IDataset dataset = null;
|
---|
63 | IRegressionProblemData problemData = null;
|
---|
64 | foreach (var run in runs) {
|
---|
65 | var solution = (IRegressionSolution)run.Results.Values.Single(x => x is IRegressionSolution);
|
---|
66 | var ds = solution.ProblemData.Dataset;
|
---|
67 |
|
---|
68 | if (solution.ProblemData == problemData) continue;
|
---|
69 | if (ds == dataset) continue;
|
---|
70 | if (problemData == null) {
|
---|
71 | problemData = solution.ProblemData;
|
---|
72 | continue;
|
---|
73 | }
|
---|
74 | if (dataset == null) {
|
---|
75 | dataset = ds;
|
---|
76 | continue;
|
---|
77 | }
|
---|
78 |
|
---|
79 | if (problemData.TrainingPartition.Start != solution.ProblemData.TrainingPartition.Start || problemData.TrainingPartition.End != solution.ProblemData.TrainingPartition.End)
|
---|
80 | throw new InvalidOperationException("The runs must share the same data.");
|
---|
81 |
|
---|
82 | if (!ds.DoubleVariables.SequenceEqual(dataset.DoubleVariables))
|
---|
83 | throw new InvalidOperationException("The runs must share the same data.");
|
---|
84 |
|
---|
85 | foreach (var v in ds.DoubleVariables) {
|
---|
86 | var values1 = (IList<double>)ds.GetReadOnlyDoubleValues(v);
|
---|
87 | var values2 = (IList<double>)dataset.GetReadOnlyDoubleValues(v);
|
---|
88 |
|
---|
89 | if (values1.Count != values2.Count)
|
---|
90 | throw new InvalidOperationException("The runs must share the same data.");
|
---|
91 |
|
---|
92 | if (!values1.SequenceEqual(values2))
|
---|
93 | throw new InvalidOperationException("The runs must share the same data.");
|
---|
94 | }
|
---|
95 | }
|
---|
96 | }
|
---|
97 |
|
---|
98 | private static RegressionEnsembleSolution CreateEnsembleSolution(IEnumerable<IRun> runs) {
|
---|
99 | var solutions = runs.Select(x => x.Results.Values.Single(v => v is IRegressionSolution)).Cast<IRegressionSolution>();
|
---|
100 | return new RegressionEnsembleSolution(new RegressionEnsembleModel(solutions.Select(x => x.Model)), solutions.First().ProblemData);
|
---|
101 | }
|
---|
102 |
|
---|
103 | public static Dictionary<string, Tuple<IEnumerable<IRun>, Dictionary<string, double>>> CalculateVariableImpactsOnline(RunCollection runs, bool useBest) {
|
---|
104 | AssertSameProblemData(runs);
|
---|
105 | var solution = (IRegressionSolution)runs.First().Results.Values.Single(x => x is IRegressionSolution);
|
---|
106 | var dataset = (Dataset)solution.ProblemData.Dataset;
|
---|
107 | var originalValues = dataset.DoubleVariables.ToDictionary(x => x, x => dataset.GetReadOnlyDoubleValues(x).ToList());
|
---|
108 | var md = dataset.ToModifiable();
|
---|
109 | var medians = new Dictionary<string, List<double>>();
|
---|
110 | foreach (var v in dataset.DoubleVariables) {
|
---|
111 | var median = dataset.GetDoubleValues(v, solution.ProblemData.TrainingIndices).Median();
|
---|
112 | medians[v] = Enumerable.Repeat(median, originalValues[v].Count).ToList();
|
---|
113 | }
|
---|
114 |
|
---|
115 | var targetImpacts = new Dictionary<string, Tuple<IEnumerable<IRun>, Dictionary<string, double>>>();
|
---|
116 |
|
---|
117 | if (useBest) {
|
---|
118 | // build network using only the best run for each target
|
---|
119 | } else {
|
---|
120 | var groups = runs.GroupBy(run => {
|
---|
121 | var sol = (IRegressionSolution)run.Results.Values.Single(x => x is IRegressionSolution);
|
---|
122 | return Concatenate(sol.ProblemData.AllowedInputVariables) + sol.ProblemData.TargetVariable;
|
---|
123 | });
|
---|
124 |
|
---|
125 | foreach (var group in groups) {
|
---|
126 | // calculate average impacts
|
---|
127 | var averageImpacts = new Dictionary<string, double>();
|
---|
128 | solution = (IRegressionSolution)group.First().Results.Values.Single(x => x is IRegressionSolution);
|
---|
129 | foreach (var run in group) {
|
---|
130 | var sol = (IRegressionSolution)run.Results.Values.Single(v => v is IRegressionSolution);
|
---|
131 |
|
---|
132 | DoubleLimit estimationLimits = null;
|
---|
133 | if (run.Parameters.ContainsKey("EstimationLimits")) {
|
---|
134 | estimationLimits = (DoubleLimit)run.Parameters["EstimationLimits"];
|
---|
135 | }
|
---|
136 | var impacts = CalculateImpacts(sol, md, originalValues, medians, estimationLimits);
|
---|
137 | // var impacts = RegressionSolutionVariableImpactsCalculator.CalculateImpacts(sol).ToDictionary(x => x.Item1, x => x.Item2);
|
---|
138 | foreach (var pair in impacts) {
|
---|
139 | if (averageImpacts.ContainsKey(pair.Key))
|
---|
140 | averageImpacts[pair.Key] += pair.Value;
|
---|
141 | else {
|
---|
142 | averageImpacts[pair.Key] = pair.Value;
|
---|
143 | }
|
---|
144 | }
|
---|
145 | }
|
---|
146 | var count = group.Count();
|
---|
147 | var keys = averageImpacts.Keys.ToList();
|
---|
148 | foreach (var v in keys) {
|
---|
149 | averageImpacts[v] /= count;
|
---|
150 | }
|
---|
151 |
|
---|
152 | targetImpacts[solution.ProblemData.TargetVariable] = new Tuple<IEnumerable<IRun>, Dictionary<string, double>>(group, averageImpacts);
|
---|
153 | }
|
---|
154 | }
|
---|
155 | return targetImpacts;
|
---|
156 | }
|
---|
157 |
|
---|
158 | private static Dictionary<string, double> CalculateImpacts(IRegressionSolution solution, ModifiableDataset dataset,
|
---|
159 | Dictionary<string, List<double>> originalValues, Dictionary<string, List<double>> medianValues, DoubleLimit estimationLimits = null) {
|
---|
160 | var impacts = new Dictionary<string, double>();
|
---|
161 |
|
---|
162 | var model = solution.Model;
|
---|
163 | var pd = solution.ProblemData;
|
---|
164 |
|
---|
165 | var rows = pd.TrainingIndices.ToList();
|
---|
166 | var targetValues = pd.Dataset.GetDoubleValues(pd.TargetVariable, rows).ToList();
|
---|
167 |
|
---|
168 |
|
---|
169 | foreach (var v in pd.AllowedInputVariables) {
|
---|
170 | dataset.ReplaceVariable(v, medianValues[v]);
|
---|
171 |
|
---|
172 | var estimatedValues = model.GetEstimatedValues(dataset, rows);
|
---|
173 | if (estimationLimits != null)
|
---|
174 | estimatedValues = estimatedValues.LimitToRange(estimationLimits.Lower, estimationLimits.Upper);
|
---|
175 |
|
---|
176 | OnlineCalculatorError error;
|
---|
177 | var r = OnlinePearsonsRCalculator.Calculate(targetValues, estimatedValues, out error);
|
---|
178 | var newQuality = error == OnlineCalculatorError.None ? r * r : double.NaN;
|
---|
179 | var originalQuality = solution.TrainingRSquared;
|
---|
180 | impacts[v] = originalQuality - newQuality;
|
---|
181 |
|
---|
182 | dataset.ReplaceVariable(v, originalValues[v]);
|
---|
183 | }
|
---|
184 | return impacts;
|
---|
185 | }
|
---|
186 |
|
---|
187 | private static Dictionary<string, Tuple<IEnumerable<IRun>, Dictionary<string, double>>> CalculateVariableImpactsFromRunResults(RunCollection runs,
|
---|
188 | string qualityResultName, bool maximization, string impactsResultName, bool useBestRunsPerTarget = false) {
|
---|
189 | var targets = runs.GroupBy(x => ((IRegressionProblemData)x.Parameters["ProblemData"]).TargetVariable).ToList();
|
---|
190 | var targetImpacts = new Dictionary<string, Tuple<IEnumerable<IRun>, Dictionary<string, double>>>();
|
---|
191 | if (useBestRunsPerTarget) {
|
---|
192 | var bestRunsPerTarget = maximization
|
---|
193 | ? targets.Select(x => x.OrderBy(y => ((DoubleValue)y.Results[qualityResultName]).Value).Last())
|
---|
194 | : targets.Select(x => x.OrderBy(y => ((DoubleValue)y.Results[qualityResultName]).Value).First());
|
---|
195 |
|
---|
196 | foreach (var run in bestRunsPerTarget) {
|
---|
197 | var pd = (IRegressionProblemData)run.Parameters["ProblemData"];
|
---|
198 | var target = pd.TargetVariable;
|
---|
199 | var impacts = (DoubleMatrix)run.Results[impactsResultName];
|
---|
200 | targetImpacts[target] = new Tuple<IEnumerable<IRun>, Dictionary<string, double>>(new[] { run }, impacts.RowNames.Select((x, i) => new { Name = x, Index = i }).ToDictionary(x => x.Name, x => impacts[x.Index, 0]));
|
---|
201 | }
|
---|
202 | } else {
|
---|
203 | foreach (var target in targets) {
|
---|
204 | var averageImpacts = CalculateAverageImpacts(new RunCollection(target), impactsResultName);
|
---|
205 | targetImpacts[target.Key] = new Tuple<IEnumerable<IRun>, Dictionary<string, double>>(target, averageImpacts);
|
---|
206 | }
|
---|
207 | }
|
---|
208 | return targetImpacts;
|
---|
209 | }
|
---|
210 |
|
---|
211 | private static VariableInteractionNetwork CreateNetwork(Dictionary<string, Tuple<IEnumerable<IRun>, Dictionary<string, double>>> targetImpacts) {
|
---|
212 | var nodes = new Dictionary<string, IVertex>();
|
---|
213 | var vn = new VariableInteractionNetwork();
|
---|
214 | foreach (var ti in targetImpacts) {
|
---|
215 | var target = ti.Key;
|
---|
216 | var variableImpacts = ti.Value.Item2;
|
---|
217 | var targetRuns = ti.Value.Item1;
|
---|
218 | IVertex targetNode;
|
---|
219 |
|
---|
220 | var variables = variableImpacts.Keys.ToList();
|
---|
221 | if (variables.Count == 0) continue;
|
---|
222 |
|
---|
223 | if (!nodes.TryGetValue(target, out targetNode)) {
|
---|
224 | targetNode = new VariableNetworkNode { Label = target };
|
---|
225 | vn.AddVertex(targetNode);
|
---|
226 | nodes[target] = targetNode;
|
---|
227 | }
|
---|
228 |
|
---|
229 | IVertex variableNode;
|
---|
230 | if (variables.Count > 1) {
|
---|
231 | var variableList = new List<string>(variables) { target };
|
---|
232 | var junctionLabel = Concatenate(variableList);
|
---|
233 | IVertex junctionNode;
|
---|
234 | if (!nodes.TryGetValue(junctionLabel, out junctionNode)) {
|
---|
235 | var solutionsEnsemble = CreateEnsembleSolution(targetRuns);
|
---|
236 | junctionNode = new JunctionNetworkNode { Label = string.Empty, Data = solutionsEnsemble };
|
---|
237 | vn.AddVertex(junctionNode);
|
---|
238 | nodes[junctionLabel] = junctionNode;
|
---|
239 | junctionNode.Label = string.Format("Target quality: {0:0.000}", solutionsEnsemble.TrainingRSquared);
|
---|
240 | }
|
---|
241 | IArc arc;
|
---|
242 | foreach (var v in variables) {
|
---|
243 | var impact = variableImpacts[v];
|
---|
244 | if (!nodes.TryGetValue(v, out variableNode)) {
|
---|
245 | variableNode = new VariableNetworkNode { Label = v };
|
---|
246 | vn.AddVertex(variableNode);
|
---|
247 | nodes[v] = variableNode;
|
---|
248 | }
|
---|
249 | arc = new Arc(variableNode, junctionNode) { Weight = impact, Label = string.Format("Impact: {0:0.000}", impact) };
|
---|
250 | vn.AddArc(arc);
|
---|
251 | }
|
---|
252 | var trainingR2 = ((IRegressionSolution)((JunctionNetworkNode)junctionNode).Data).TrainingRSquared;
|
---|
253 | arc = new Arc(junctionNode, targetNode) { Weight = junctionNode.InArcs.Sum(x => x.Weight), Label = string.Format("Quality: {0:0.000}", trainingR2) };
|
---|
254 | vn.AddArc(arc);
|
---|
255 | } else {
|
---|
256 | foreach (var v in variables) {
|
---|
257 | var impact = variableImpacts[v];
|
---|
258 | if (!nodes.TryGetValue(v, out variableNode)) {
|
---|
259 | variableNode = new VariableNetworkNode { Label = v };
|
---|
260 | vn.AddVertex(variableNode);
|
---|
261 | nodes[v] = variableNode;
|
---|
262 | }
|
---|
263 | var arc = new Arc(variableNode, targetNode) { Weight = impact, Label = string.Format("Impact: {0:0.000}", impact) };
|
---|
264 | vn.AddArc(arc);
|
---|
265 | }
|
---|
266 | }
|
---|
267 | }
|
---|
268 | return vn;
|
---|
269 | }
|
---|
270 |
|
---|
271 | private static double CalculateAverageQuality(RunCollection runs) {
|
---|
272 | var pd = (IRegressionProblemData)runs.First().Parameters["ProblemData"];
|
---|
273 | var target = pd.TargetVariable;
|
---|
274 | var inputs = pd.AllowedInputVariables;
|
---|
275 |
|
---|
276 | if (!runs.All(x => {
|
---|
277 | var problemData = (IRegressionProblemData)x.Parameters["ProblemData"];
|
---|
278 | return target == problemData.TargetVariable && inputs.SequenceEqual(problemData.AllowedInputVariables);
|
---|
279 | })) {
|
---|
280 | throw new ArgumentException("All runs must have the same target and inputs.");
|
---|
281 | }
|
---|
282 | return runs.Average(x => ((DoubleValue)x.Results["Best training solution quality"]).Value);
|
---|
283 | }
|
---|
284 |
|
---|
285 | private static Dictionary<string, double> CalculateAverageImpacts(RunCollection runs, string resultName) {
|
---|
286 | var pd = (IRegressionProblemData)runs.First().Parameters["ProblemData"];
|
---|
287 | var target = pd.TargetVariable;
|
---|
288 | var inputs = pd.AllowedInputVariables.ToList();
|
---|
289 |
|
---|
290 | var impacts = inputs.ToDictionary(x => x, x => 0d);
|
---|
291 |
|
---|
292 | // check if all the runs have the same target and same inputs
|
---|
293 | if (!runs.All(x => {
|
---|
294 | var problemData = (IRegressionProblemData)x.Parameters["ProblemData"];
|
---|
295 | return target == problemData.TargetVariable && inputs.SequenceEqual(problemData.AllowedInputVariables);
|
---|
296 | })) {
|
---|
297 | throw new ArgumentException("All runs must have the same target and inputs.");
|
---|
298 | }
|
---|
299 |
|
---|
300 | foreach (var run in runs) {
|
---|
301 | var impactsMatrix = (DoubleMatrix)run.Results[resultName];
|
---|
302 |
|
---|
303 | int i = 0;
|
---|
304 | foreach (var v in impactsMatrix.RowNames) {
|
---|
305 | impacts[v] += impactsMatrix[i, 0];
|
---|
306 | ++i;
|
---|
307 | }
|
---|
308 | }
|
---|
309 |
|
---|
310 | foreach (var v in inputs) {
|
---|
311 | impacts[v] /= runs.Count;
|
---|
312 | }
|
---|
313 |
|
---|
314 | return impacts;
|
---|
315 | }
|
---|
316 |
|
---|
317 | private static string Concatenate(IEnumerable<string> strings) {
|
---|
318 | var sb = new StringBuilder();
|
---|
319 | foreach (var s in strings) {
|
---|
320 | sb.Append(s);
|
---|
321 | }
|
---|
322 | return sb.ToString();
|
---|
323 | }
|
---|
324 |
|
---|
325 | private void ConfigureNodeShapes() {
|
---|
326 | graphChart.ClearShapes();
|
---|
327 | var font = new Font(FontFamily.GenericSansSerif, 12);
|
---|
328 | graphChart.AddShape(typeof(VariableNetworkNode), new LabeledPrimitive(new Ellipse(graphChart.Chart, new PointD(0, 0), new PointD(30, 30), Pens.Black, Brushes.White), "", font));
|
---|
329 | graphChart.AddShape(typeof(JunctionNetworkNode), new LabeledPrimitive(new Rectangle(graphChart.Chart, new PointD(0, 0), new PointD(15, 15), Pens.Black, Brushes.DarkGray), "", font));
|
---|
330 | }
|
---|
331 |
|
---|
332 | #region events
|
---|
333 | protected override void OnContentChanged() {
|
---|
334 | base.OnContentChanged();
|
---|
335 | var run = Content.First();
|
---|
336 | var pd = (IRegressionProblemData)run.Parameters["ProblemData"];
|
---|
337 | var variables = new HashSet<string>(new List<string>(pd.Dataset.DoubleVariables));
|
---|
338 | impactResultNameComboBox.Items.Clear();
|
---|
339 | foreach (var result in run.Results.Where(x => x.Value is DoubleMatrix)) {
|
---|
340 | var m = (DoubleMatrix)result.Value;
|
---|
341 | if (m.RowNames.All(x => variables.Contains(x)))
|
---|
342 | impactResultNameComboBox.Items.Add(result.Key);
|
---|
343 | }
|
---|
344 | qualityResultNameComboBox.Items.Clear();
|
---|
345 | foreach (var result in run.Results.Where(x => x.Value is DoubleValue)) {
|
---|
346 | qualityResultNameComboBox.Items.Add(result.Key);
|
---|
347 | }
|
---|
348 | if (impactResultNameComboBox.Items.Count > 0) {
|
---|
349 | impactResultNameComboBox.Text = (string)impactResultNameComboBox.Items[0];
|
---|
350 | }
|
---|
351 | if (qualityResultNameComboBox.Items.Count > 0) {
|
---|
352 | qualityResultNameComboBox.Text = (string)qualityResultNameComboBox.Items[0];
|
---|
353 | }
|
---|
354 | if (impactResultNameComboBox.Items.Count > 0 && qualityResultNameComboBox.Items.Count > 0)
|
---|
355 | NetworkConfigurationChanged(this, EventArgs.Empty);
|
---|
356 | }
|
---|
357 |
|
---|
358 | private void TextBoxValidating(object sender, CancelEventArgs e) {
|
---|
359 | double v;
|
---|
360 | string errorMsg = "Could not parse the entered value. Please input a real number.";
|
---|
361 | var tb = (TextBox)sender;
|
---|
362 | if (!double.TryParse(tb.Text, out v)) {
|
---|
363 | e.Cancel = true;
|
---|
364 | tb.Select(0, tb.Text.Length);
|
---|
365 |
|
---|
366 | // Set the ErrorProvider error with the text to display.
|
---|
367 | this.errorProvider.SetError(tb, errorMsg);
|
---|
368 | errorProvider.BlinkStyle = ErrorBlinkStyle.NeverBlink;
|
---|
369 | errorProvider.SetIconPadding(tb, -20);
|
---|
370 | }
|
---|
371 | }
|
---|
372 |
|
---|
373 | private void ImpactThresholdTextBoxValidated(object sender, EventArgs e) {
|
---|
374 | var tb = (TextBox)sender;
|
---|
375 | errorProvider.SetError(tb, string.Empty);
|
---|
376 | var network = ApplyThreshold(variableInteractionNetwork, double.Parse(tb.Text));
|
---|
377 | graphChart.Graph = network;
|
---|
378 | }
|
---|
379 |
|
---|
380 | private static VariableInteractionNetwork ApplyThreshold(VariableInteractionNetwork originalNetwork, double threshold) {
|
---|
381 | var arcs = originalNetwork.Arcs.Where(x => x.Weight >= threshold).ToList();
|
---|
382 | if (!arcs.Any()) return originalNetwork;
|
---|
383 | var filteredNetwork = new VariableInteractionNetwork();
|
---|
384 | var cloner = new Cloner();
|
---|
385 | var vertices = arcs.SelectMany(x => new[] { x.Source, x.Target }).Select(cloner.Clone).Distinct(); // arcs are not cloned
|
---|
386 | filteredNetwork.AddVertices(vertices);
|
---|
387 | filteredNetwork.AddArcs(arcs.Select(x => (IArc)x.Clone(cloner)));
|
---|
388 |
|
---|
389 | var unusedJunctions = filteredNetwork.Vertices.Where(x => x.InDegree == 0 && x is JunctionNetworkNode).ToList();
|
---|
390 | filteredNetwork.RemoveVertices(unusedJunctions);
|
---|
391 | var orphanedNodes = filteredNetwork.Vertices.Where(x => x.Degree == 0).ToList();
|
---|
392 | filteredNetwork.RemoveVertices(orphanedNodes);
|
---|
393 | return filteredNetwork;
|
---|
394 | }
|
---|
395 |
|
---|
396 | private void LayoutConfigurationBoxValidated(object sender, EventArgs e) {
|
---|
397 | var tb = (TextBox)sender;
|
---|
398 | errorProvider.SetError(tb, string.Empty);
|
---|
399 | LayoutConfigurationChanged(sender, e);
|
---|
400 | }
|
---|
401 |
|
---|
402 | private void NetworkConfigurationChanged(object sender, EventArgs e) {
|
---|
403 | var useBest = impactAggregationComboBox.SelectedIndex <= 0;
|
---|
404 | var threshold = double.Parse(impactThresholdTextBox.Text);
|
---|
405 | var qualityResultName = qualityResultNameComboBox.Text;
|
---|
406 | var impactsResultName = impactResultNameComboBox.Text;
|
---|
407 | if (string.IsNullOrEmpty(qualityResultName) || string.IsNullOrEmpty(impactsResultName))
|
---|
408 | return;
|
---|
409 | var maximization = maximizationCheckBox.Checked;
|
---|
410 | var impacts = CalculateVariableImpactsFromRunResults(Content, qualityResultName, maximization, impactsResultName, useBest);
|
---|
411 | variableInteractionNetwork = CreateNetwork(impacts);
|
---|
412 | var network = ApplyThreshold(variableInteractionNetwork, threshold);
|
---|
413 | graphChart.Graph = network;
|
---|
414 | }
|
---|
415 |
|
---|
416 | private void LayoutConfigurationChanged(object sender, EventArgs e) {
|
---|
417 | ConstrainedForceDirectedLayout.EdgeRouting routingMode;
|
---|
418 | switch (edgeRoutingComboBox.SelectedIndex) {
|
---|
419 | case 0:
|
---|
420 | routingMode = ConstrainedForceDirectedLayout.EdgeRouting.None;
|
---|
421 | break;
|
---|
422 | case 1:
|
---|
423 | routingMode = ConstrainedForceDirectedLayout.EdgeRouting.Polyline;
|
---|
424 | break;
|
---|
425 | case 2:
|
---|
426 | routingMode = ConstrainedForceDirectedLayout.EdgeRouting.Orthogonal;
|
---|
427 | break;
|
---|
428 | default:
|
---|
429 | throw new ArgumentException("Invalid edge routing mode.");
|
---|
430 | }
|
---|
431 | var idealEdgeLength = double.Parse(idealEdgeLengthTextBox.Text);
|
---|
432 | if (routingMode == graphChart.RoutingMode && idealEdgeLength.IsAlmost(graphChart.IdealEdgeLength)) return;
|
---|
433 | graphChart.RoutingMode = routingMode;
|
---|
434 | graphChart.PerformEdgeRouting = routingMode != ConstrainedForceDirectedLayout.EdgeRouting.None;
|
---|
435 | graphChart.IdealEdgeLength = idealEdgeLength;
|
---|
436 | graphChart.Draw();
|
---|
437 | }
|
---|
438 |
|
---|
439 | private void onlineImpactCalculationButton_Click(object sender, EventArgs args) {
|
---|
440 | var button = (Button)sender;
|
---|
441 | var worker = new BackgroundWorker();
|
---|
442 | worker.DoWork += (o, e) => {
|
---|
443 | button.Enabled = false;
|
---|
444 | var impacts = CalculateVariableImpactsOnline(Content, false);
|
---|
445 | variableInteractionNetwork = CreateNetwork(impacts);
|
---|
446 | var threshold = double.Parse(impactThresholdTextBox.Text);
|
---|
447 | graphChart.Graph = ApplyThreshold(variableInteractionNetwork, threshold);
|
---|
448 | };
|
---|
449 | worker.RunWorkerCompleted += (o, e) => button.Enabled = true;
|
---|
450 | worker.RunWorkerAsync();
|
---|
451 | }
|
---|
452 | #endregion
|
---|
453 | }
|
---|
454 | }
|
---|