Ticket #2857: PDP.patch

Controls/PartialDependencePlot.cs

@@ -36 +36 @@
 namespace HeuristicLab.Problems.DataAnalysis.Views {
   public partial class PartialDependencePlot : UserControl, IPartialDependencePlot {
     private ModifiableDataset sharedFixedVariables; // used for syncronising variable values between charts
+    private ModifiableDataset sharedIndependentVariables;
     private ModifiableDataset internalDataset; // holds the x values for each point drawn
 
     private CancellationTokenSource cancelCurrentRecalculateSource;
@@ -236 +237 @@
         cancelCurrentRecalculateSource.Cancel();
     }
 
-    public void Configure(IEnumerable<IRegressionSolution> solutions, ModifiableDataset sharedFixedVariables, string freeVariable, int drawingSteps, bool initializeAxisRanges = true) {
+    public void Configure(
+      IEnumerable<IRegressionSolution> solutions,
+      ModifiableDataset sharedFixedVariables,
+      ModifiableDataset sharedIndependentVariables,
+      string freeVariable, int drawingSteps, bool initializeAxisRanges = true) {
       if (!SolutionsCompatible(solutions))
         throw new ArgumentException("Solutions are not compatible with the problem data.");
       this.freeVariable = freeVariable;
@@ -249 +254 @@
       // this change is reflected in the internal dataset (where the value becomes a whole column)
       if (this.sharedFixedVariables != null)
         this.sharedFixedVariables.ItemChanged -= sharedFixedVariables_ItemChanged;
+
       this.sharedFixedVariables = sharedFixedVariables;
       this.sharedFixedVariables.ItemChanged += sharedFixedVariables_ItemChanged;
+      this.sharedIndependentVariables = sharedIndependentVariables;
 
       RecalculateTrainingLimits(initializeAxisRanges);
       RecalculateInternalDataset();
@@ -285 +292 @@
       OrderAndColorSeries();
     }
 
+    private async void SharedIndependentVariables_ItemChanged(object sender, EventArgs<int, int> e) {
+      await RecalculateAsync(true, false);
+    }
+
     public async Task RecalculateAsync(bool updateOnFinish = true, bool resetYAxis = true) {
       if (IsDisposed
         || sharedFixedVariables == null || !solutions.Any() || string.IsNullOrEmpty(freeVariable)
@@ -316 +327 @@
         calculationPendingLabel.Visible = false;
         if (updateOnFinish)
           Update();
-      }
-      catch (OperationCanceledException) { }
-      catch (AggregateException ae) {
+      } catch (OperationCanceledException) { } catch (AggregateException ae) {
         if (!ae.InnerExceptions.Any(e => e is OperationCanceledException))
           throw;
       }
@@ -411 +420 @@
       };
       series.LegendText = series.Name;
 
-      var confidenceBoundSolution = solution as IConfidenceRegressionSolution;
+      //var confidenceBoundSolution = solution as IConfidenceRegressionSolution;
       Series confidenceIntervalSeries = null;
-      if (confidenceBoundSolution != null) {
-        confidenceIntervalSeries = new Series {
-          ChartType = SeriesChartType.Range,
-          YValuesPerPoint = 2,
-          Name = "95% Conf. Interval " + series.Name,
-          IsVisibleInLegend = false
-        };
-      }
+      //if (confidenceBoundSolution != null) {
+      confidenceIntervalSeries = new Series {
+        ChartType = SeriesChartType.Range,
+        YValuesPerPoint = 2,
+        Name = "95% Conf. Interval " + series.Name,
+        IsVisibleInLegend = false
+      };
+      //}
       return Tuple.Create(series, confidenceIntervalSeries);
     }
 
@@ -467 +476 @@
     private Task<DoubleLimit> UpdateSeriesDataAsync(IRegressionSolution solution, CancellationToken cancellationToken) {
       return Task.Run(() => {
         var xvalues = internalDataset.GetDoubleValues(FreeVariable).ToList();
-        var yvalues = solution.Model.GetEstimatedValues(internalDataset, Enumerable.Range(0, internalDataset.Rows)).ToList();
+        var yvalues = GetSamples(solution).ToList();
 
         double min = double.MaxValue, max = double.MinValue;
 
         var series = seriesCache[solution];
         for (int i = 0; i < xvalues.Count; i++) {
-          series.Points[i].SetValueXY(xvalues[i], yvalues[i]);
-          if (yvalues[i] < min) min = yvalues[i];
-          if (yvalues[i] > max) max = yvalues[i];
+          var avg = yvalues[i].Average();
+          series.Points[i].SetValueXY(xvalues[i], avg);
+          if (avg < min) min = avg;
+          if (avg > max) max = avg;
         }
 
         cancellationToken.ThrowIfCancellationRequested();
 
-        var confidenceBoundSolution = solution as IConfidenceRegressionSolution;
-        if (confidenceBoundSolution != null) {
-          var confidenceIntervalSeries = ciSeriesCache[solution];
-          var variances = confidenceBoundSolution.Model.GetEstimatedVariances(internalDataset, Enumerable.Range(0, internalDataset.Rows)).ToList();
-          for (int i = 0; i < xvalues.Count; i++) {
-            var lower = yvalues[i] - 1.96 * Math.Sqrt(variances[i]);
-            var upper = yvalues[i] + 1.96 * Math.Sqrt(variances[i]);
-            confidenceIntervalSeries.Points[i].SetValueXY(xvalues[i], lower, upper);
-            if (lower < min) min = lower;
-            if (upper > max) max = upper;
-          }
+        // var confidenceBoundSolution = solution as IConfidenceRegressionSolution;
+        // if (confidenceBoundSolution != null) {
+        //   var confidenceIntervalSeries = ciSeriesCache[solution];
+        //   var variances = GetEstimatedVariance(confidenceBoundSolution).ToList();
+        //   for (int i = 0; i < xvalues.Count; i++) {
+        //     var lower = yvalues[i] - 1.96 * Math.Sqrt(variances[i]);
+        //     var upper = yvalues[i] + 1.96 * Math.Sqrt(variances[i]);
+        //     confidenceIntervalSeries.Points[i].SetValueXY(xvalues[i], lower, upper);
+        //     if (lower < min) min = lower;
+        //     if (upper > max) max = upper;
+        //   }
+        // }
+        var confidenceIntervalSeries = ciSeriesCache[solution];
+        for (int i = 0; i < xvalues.Count; i++) {
+          var avg = yvalues[i].Average();
+          var variance = yvalues[i].Variance();
+          var lower = avg - 1.96 * Math.Sqrt(variance);
+          var upper = avg + 1.96 * Math.Sqrt(variance);
+          confidenceIntervalSeries.Points[i].SetValueXY(xvalues[i], lower, upper);
+          if (lower < min) min = lower;
+          if (upper > max) max = upper;
         }
 
         cancellationToken.ThrowIfCancellationRequested();
@@ -498 +518 @@
       }, cancellationToken);
     }
 
+    private IEnumerable<double[]> GetSamples(IRegressionSolution solution) {
+      var independentVarNames = sharedIndependentVariables.DoubleVariables.ToArray();
+
+      var predictions = new List<double[]>();
+      for (int rIdx = 0; rIdx < sharedIndependentVariables.Rows; rIdx++) {
+        foreach (var varName in independentVarNames) {
+          var varVal = sharedIndependentVariables.GetDoubleValue(varName, rIdx);
+          if (internalDataset.VariableNames.Contains(varName))
+            internalDataset.ReplaceVariable(varName, Enumerable.Repeat(varVal, internalDataset.Rows).ToList());
+          else
+            internalDataset.AddVariable(varName, Enumerable.Repeat(varVal, internalDataset.Rows).ToList());
+
+        }
+        var pred = solution.Model.GetEstimatedValues(internalDataset, Enumerable.Range(0, internalDataset.Rows));
+        predictions.Add(pred.ToArray());
+      }
+      for (int i = 0; i < internalDataset.Rows; i++) {
+        yield return predictions.Select(p => p[i]).ToArray();
+      }
+    }
+
     private void ResizeAllSeriesData() {
       if (internalDataset == null)
         return;

Regression/RegressionSolutionGradientView.cs

@@ -93 +93 @@
       var newTrackbars = CreateConfiguration();
 
       sharedFixedVariables = new ModifiableDataset(variableNames, newTrackbars.Select(tb => new List<double>(1) { (double)tb.Value }));
-      _partialDependencePlot.Configure(new[] { Content }, sharedFixedVariables, variableNames.First(), DrawingSteps);
+      _partialDependencePlot.Configure(new[] { Content }, sharedFixedVariables, new ModifiableDataset(), variableNames.First(), DrawingSteps);
       await _partialDependencePlot.RecalculateAsync();
 
       // Add to table and observable lists

Regression/RegressionSolutionPartialDependencePlotView.cs

@@ -39 +39 @@
     private readonly Dictionary<string, DensityChart> densityCharts;
     private readonly Dictionary<string, Panel> groupingPanels;
     private ModifiableDataset sharedFixedVariables;
+    private ModifiableDataset sharedIndependentVariables;
 
     private const int Points = 200;
     private int MaxColumns = 4;
@@ -45 +46 @@
 
     private IEnumerable<string> VisibleVariables {
       get {
-        foreach (ListViewItem item in variableListView.CheckedItems)
-          yield return item.Text;
+        for (int i = 0; i < checkedVariableList.Items.Count; i++) {
+          var item = checkedVariableList.Items[i];
+          if (checkedVariableList.GetItemCheckState(i) == CheckState.Checked)
+            yield return (string)item;
+        }
       }
     }
     private IEnumerable<IPartialDependencePlot> VisiblePartialDependencePlots {
@@ -133 +137 @@
         sharedFixedVariables.ItemChanged -= SharedFixedVariables_ItemChanged;
 
       sharedFixedVariables = new ModifiableDataset(doubleVariables.Concat(factorVariables), doubleVariableValues.Concat(factorVariableValues));
+      sharedIndependentVariables = new ModifiableDataset(new string[] { "$$$RowIdx" }, new IList[] { problemData.TrainingIndices.ToList() });
 
 
       // create controls
@@ -140 +145 @@
       densityCharts.Clear();
       groupingPanels.Clear();
       foreach (var variableName in doubleVariables) {
-        var plot = CreatePartialDependencePlot(variableName, sharedFixedVariables);
+        var plot = CreatePartialDependencePlot(variableName, sharedFixedVariables, sharedIndependentVariables);
         partialDependencePlots.Add(variableName, plot);
 
         var densityChart = new DensityChart() {
@@ -218 +223 @@
         panel.Controls.Add(plot);
         groupingPanels.Add(variableName, panel);
       }
+
       // update variable list
-      variableListView.ItemChecked -= variableListView_ItemChecked;
-      variableListView.Items.Clear();
+      checkedVariableList.ItemCheck -= checkedVariableList_ItemCheck;
+      checkedVariableList.Items.Clear();
       foreach (var variable in allowedInputVariables)
-        variableListView.Items.Add(key: variable, text: variable, imageIndex: 0);
+        checkedVariableList.Items.Add(variable, isChecked: Content.Model.VariablesUsedForPrediction.Contains(variable));
 
-      foreach (var variable in Content.Model.VariablesUsedForPrediction)
-        variableListView.Items[variable].Checked = true;
-      variableListView.ItemChecked += variableListView_ItemChecked;
+      checkedVariableList.ItemCheck += checkedVariableList_ItemCheck;
 
       sharedFixedVariables.ItemChanged += SharedFixedVariables_ItemChanged;
 
@@ -294 +298 @@
         }
       }
     }
-    private PartialDependencePlot CreatePartialDependencePlot(string variableName, ModifiableDataset sharedFixedVariables) {
+    private PartialDependencePlot CreatePartialDependencePlot(string variableName, ModifiableDataset sharedFixedVariables, ModifiableDataset sharedIndependentVariables) {
       var plot = new PartialDependencePlot {
         Dock = DockStyle.Fill,
         Margin = Padding.Empty,
@@ -314 +318 @@
         if (recalculations.All(t => t.IsCompleted))
           SetupYAxis();
       };
-      plot.Configure(new[] { Content }, sharedFixedVariables, variableName, Points);
+      plot.Configure(new[] { Content }, sharedFixedVariables, sharedIndependentVariables, variableName, Points);
       plot.SolutionAdded += partialDependencePlot_SolutionAdded;
       plot.SolutionRemoved += partialDependencePlot_SolutionRemoved;
       return plot;
@@ -372 +376 @@
       var tl = partialDependencePlotTableLayout;
       tl.Controls.Clear();
       int row = 0, column = 0;
-      double yValue = Content.Model.GetEstimatedValues(sharedFixedVariables, new[] { 0 }).Single();
+      double yValue = Content.Model.GetEstimatedValues(sharedFixedVariables, new[] { 0 }).Single(); // TODO mean estimation
       string title = Content.ProblemData.TargetVariable + ": " + yValue.ToString("G5", CultureInfo.CurrentCulture);
 
       foreach (var v in VisibleVariables) {
@@ -427 +431 @@
       }
     }
 
-    private async void variableListView_ItemChecked(object sender, ItemCheckedEventArgs e) {
-      var item = e.Item;
-      var variable = item.Text;
+
+    private async void checkedVariableList_ItemCheck(object sender, ItemCheckEventArgs e) {
+      // workaround to set the checkedState of the item to the new value so that remaining code works
+      // see https://stackoverflow.com/questions/3666682/which-checkedlistbox-event-triggers-after-a-item-is-checked
+      // we handle three states here
+
+      CheckedListBox clb = (CheckedListBox)sender;
+      // Switch off event handler
+      clb.ItemCheck -= checkedVariableList_ItemCheck;
+      CheckState newCheckState;
+      if (e.CurrentValue == CheckState.Unchecked) newCheckState = CheckState.Checked;
+      else if (e.CurrentValue == CheckState.Checked) newCheckState = CheckState.Indeterminate;
+      else newCheckState = CheckState.Unchecked;
+      e.NewValue = newCheckState;
+      clb.SetItemCheckState(e.Index, newCheckState);
+      // Switch on event handler
+      clb.ItemCheck += checkedVariableList_ItemCheck;
+
+      var itemChecked = newCheckState == CheckState.Checked;
+      var item = checkedVariableList.Items[e.Index];
+      var variable = (string)item;
       var plot = partialDependencePlots[variable];
       var chartsPanel = groupingPanels[variable];
       var tl = partialDependencePlotTableLayout;
 
-      tl.SuspendLayout();
-      if (item.Checked) {
-        tl.Controls.Add(chartsPanel);
-        await plot.RecalculateAsync(false, false);
+
+      if (newCheckState == CheckState.Indeterminate) {
+        tl.Controls.Remove(chartsPanel);
+        if (!sharedIndependentVariables.DoubleVariables.Contains(variable)) {
+          sharedIndependentVariables.AddVariable(variable, Content.ProblemData.Dataset.GetDoubleValues(variable, Content.ProblemData.TrainingIndices));
+        }
       } else {
-        tl.Controls.Remove(chartsPanel);
+        if (sharedIndependentVariables.DoubleVariables.Contains(variable))
+          sharedIndependentVariables.RemoveVariable(variable);
       }
-
-      if (tl.Controls.Count > 0) {
-        SetupYAxis();
-        ReOrderControls();
-        SetStyles();
-      }
-      tl.ResumeLayout();
-      tl.Refresh();
-      densityComboBox_SelectedIndexChanged(this, EventArgs.Empty);
+      RecalculateAndRelayoutCharts();
     }
 
     private void automaticYAxisCheckBox_CheckedChanged(object sender, EventArgs e) {

Regression/RegressionSolutionPartialDependencePlotView.Designer.cs

@@ -24 +24 @@
     /// </summary>
     private void InitializeComponent() {
       this.components = new System.ComponentModel.Container();
-      this.variableListView = new System.Windows.Forms.ListView();
       this.partialDependencePlotTableLayout = new System.Windows.Forms.TableLayoutPanel();
       this.yAxisConfigGroupBox = new System.Windows.Forms.GroupBox();
       this.limitView = new HeuristicLab.Problems.DataAnalysis.Views.DoubleLimitView();
       this.automaticYAxisCheckBox = new System.Windows.Forms.CheckBox();
       this.densityGroupBox = new System.Windows.Forms.GroupBox();
+      this.columnsNumericUpDown = new System.Windows.Forms.NumericUpDown();
       this.columnsLabel = new System.Windows.Forms.Label();
       this.densityComboBox = new System.Windows.Forms.ComboBox();
       this.label1 = new System.Windows.Forms.Label();
       this.configSplitContainer = new System.Windows.Forms.SplitContainer();
       this.variableGroupBox = new System.Windows.Forms.GroupBox();
+      this.checkedVariableList = new System.Windows.Forms.CheckedListBox();
       this.scrollPanel = new System.Windows.Forms.Panel();
       this.errorProvider = new System.Windows.Forms.ErrorProvider(this.components);
-      this.columnsNumericUpDown = new System.Windows.Forms.NumericUpDown();
       this.yAxisConfigGroupBox.SuspendLayout();
       this.densityGroupBox.SuspendLayout();
+      ((System.ComponentModel.ISupportInitialize)(this.columnsNumericUpDown)).BeginInit();
       ((System.ComponentModel.ISupportInitialize)(this.configSplitContainer)).BeginInit();
       this.configSplitContainer.Panel1.SuspendLayout();
       this.configSplitContainer.Panel2.SuspendLayout();
@@ -47 +48 @@
       this.variableGroupBox.SuspendLayout();
       this.scrollPanel.SuspendLayout();
       ((System.ComponentModel.ISupportInitialize)(this.errorProvider)).BeginInit();
-      ((System.ComponentModel.ISupportInitialize)(this.columnsNumericUpDown)).BeginInit();
       this.SuspendLayout();
       // 
-      // variableListView
-      // 
-      this.variableListView.CheckBoxes = true;
-      this.variableListView.Dock = System.Windows.Forms.DockStyle.Fill;
-      this.variableListView.Location = new System.Drawing.Point(3, 16);
-      this.variableListView.Name = "variableListView";
-      this.variableListView.Size = new System.Drawing.Size(163, 478);
-      this.variableListView.TabIndex = 0;
-      this.variableListView.UseCompatibleStateImageBehavior = false;
-      this.variableListView.View = System.Windows.Forms.View.List;
-      this.variableListView.ItemChecked += new System.Windows.Forms.ItemCheckedEventHandler(this.variableListView_ItemChecked);
-      // 
       // partialDependencePlotTableLayout
       // 
       this.partialDependencePlotTableLayout.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Left) 
@@ -126 +114 @@
       this.densityGroupBox.TabStop = false;
       this.densityGroupBox.Text = "Settings";
       // 
+      // columnsNumericUpDown
+      // 
+      this.columnsNumericUpDown.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Left) 
+            | System.Windows.Forms.AnchorStyles.Right)));
+      this.columnsNumericUpDown.Location = new System.Drawing.Point(66, 46);
+      this.columnsNumericUpDown.Minimum = new decimal(new int[] {
+            1,
+            0,
+            0,
+            0});
+      this.columnsNumericUpDown.Name = "columnsNumericUpDown";
+      this.columnsNumericUpDown.Size = new System.Drawing.Size(94, 20);
+      this.columnsNumericUpDown.TabIndex = 1;
+      this.columnsNumericUpDown.Value = new decimal(new int[] {
+            4,
+            0,
+            0,
+            0});
+      this.columnsNumericUpDown.ValueChanged += new System.EventHandler(this.columnsNumericUpDown_ValueChanged);
+      // 
       // columnsLabel
       // 
       this.columnsLabel.AutoSize = true;
@@ -184 +192 @@
       // 
       // variableGroupBox
       // 
-      this.variableGroupBox.Controls.Add(this.variableListView);
+      this.variableGroupBox.Controls.Add(this.checkedVariableList);
       this.variableGroupBox.Dock = System.Windows.Forms.DockStyle.Fill;
       this.variableGroupBox.Location = new System.Drawing.Point(0, 151);
       this.variableGroupBox.Name = "variableGroupBox";
@@ -193 +201 @@
       this.variableGroupBox.TabStop = false;
       this.variableGroupBox.Text = "Variables";
       // 
+      // checkedVariableList
+      // 
+      this.checkedVariableList.CheckOnClick = true;
+      this.checkedVariableList.Dock = System.Windows.Forms.DockStyle.Fill;
+      this.checkedVariableList.FormattingEnabled = true;
+      this.checkedVariableList.Location = new System.Drawing.Point(3, 16);
+      this.checkedVariableList.Name = "checkedVariableList";
+      this.checkedVariableList.Size = new System.Drawing.Size(163, 478);
+      this.checkedVariableList.TabIndex = 0;
+      // 
       // scrollPanel
       // 
       this.scrollPanel.Controls.Add(this.partialDependencePlotTableLayout);
@@ -207 +225 @@
       this.errorProvider.BlinkStyle = System.Windows.Forms.ErrorBlinkStyle.NeverBlink;
       this.errorProvider.ContainerControl = this;
       // 
-      // columnsNumericUpDown
-      // 
-      this.columnsNumericUpDown.Anchor = ((System.Windows.Forms.AnchorStyles)(((System.Windows.Forms.AnchorStyles.Top | System.Windows.Forms.AnchorStyles.Left) 
-            | System.Windows.Forms.AnchorStyles.Right)));
-      this.columnsNumericUpDown.Location = new System.Drawing.Point(66, 46);
-      this.columnsNumericUpDown.Minimum = new decimal(new int[] {
-            1,
-            0,
-            0,
-            0});
-      this.columnsNumericUpDown.Name = "columnsNumericUpDown";
-      this.columnsNumericUpDown.Size = new System.Drawing.Size(94, 20);
-      this.columnsNumericUpDown.TabIndex = 1;
-      this.columnsNumericUpDown.Value = new decimal(new int[] {
-            4,
-            0,
-            0,
-            0});
-      this.columnsNumericUpDown.ValueChanged += new System.EventHandler(this.columnsNumericUpDown_ValueChanged);
-      // 
       // RegressionSolutionPartialDependencePlotView
       // 
       this.AutoScaleDimensions = new System.Drawing.SizeF(6F, 13F);
@@ -238 +236 @@
       this.yAxisConfigGroupBox.PerformLayout();
       this.densityGroupBox.ResumeLayout(false);
       this.densityGroupBox.PerformLayout();
+      ((System.ComponentModel.ISupportInitialize)(this.columnsNumericUpDown)).EndInit();
       this.configSplitContainer.Panel1.ResumeLayout(false);
       this.configSplitContainer.Panel2.ResumeLayout(false);
       ((System.ComponentModel.ISupportInitialize)(this.configSplitContainer)).EndInit();
@@ -246 +245 @@
       this.scrollPanel.ResumeLayout(false);
       this.scrollPanel.PerformLayout();
       ((System.ComponentModel.ISupportInitialize)(this.errorProvider)).EndInit();
-      ((System.ComponentModel.ISupportInitialize)(this.columnsNumericUpDown)).EndInit();
       this.ResumeLayout(false);
 
     }
 
     #endregion
-
-    private System.Windows.Forms.ListView variableListView;
     private System.Windows.Forms.TableLayoutPanel partialDependencePlotTableLayout;
     private System.Windows.Forms.GroupBox yAxisConfigGroupBox;
     private System.Windows.Forms.CheckBox automaticYAxisCheckBox;
@@ -267 +263 @@
     private System.Windows.Forms.ErrorProvider errorProvider;
     private System.Windows.Forms.Label label1;
     private System.Windows.Forms.NumericUpDown columnsNumericUpDown;
+    private System.Windows.Forms.CheckedListBox checkedVariableList;
   }
 }