Changeset 15110 for trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content

Timestamp:

07/03/17 10:57:53 (7 years ago)

Author:

pfleck

Message:

#2709: merged branch to trunk

Location:

trunk/sources/HeuristicLab.DataPreprocessing

Files:

• . (modified) (1 prop)
• 3.4 (modified) (1 prop)
• 3.4/Content/CorrelationMatrixContent.cs (modified) (1 diff)
• 3.4/Content/DataCompletenessChartContent.cs (modified) (1 diff)
• 3.4/Content/DataGridContent.cs (modified) (5 diffs)
• 3.4/Content/FilterContent.cs (modified) (1 diff)
• 3.4/Content/HistogramContent.cs (modified) (3 diffs)
• 3.4/Content/IViewChartShortcut.cs (deleted)
• 3.4/Content/LineChartContent.cs (modified) (2 diffs)
• 3.4/Content/ManipulationContent.cs (modified) (2 diffs)
• 3.4/Content/MultiScatterPlotContent.cs (copied) (copied from branches/DataPreprocessing Enhancements/HeuristicLab.DataPreprocessing/3.4/Content/MultiScatterPlotContent.cs)
• 3.4/Content/PreprocessingChartContent.cs (modified) (5 diffs)
• 3.4/Content/PreprocessingDataTable.cs (deleted)
• 3.4/Content/ScatterPlotContent.cs (modified) (1 diff)
• 3.4/Content/SingleScatterPlotContent.cs (copied) (copied from branches/DataPreprocessing Enhancements/HeuristicLab.DataPreprocessing/3.4/Content/SingleScatterPlotContent.cs)
• 3.4/Content/StatisticsContent.cs (modified) (2 diffs)
• 3.4/Content/TransformationContent.cs (modified) (2 diffs)

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/CorrelationMatrixContent.cs

@@ -27 +27 @@
 namespace HeuristicLab.DataPreprocessing {
   [Item("Feature Correlation Matrix", "Represents the feature correlation matrix.")]
-  public class CorrelationMatrixContent : Item, IViewChartShortcut {
+  public class CorrelationMatrixContent : Item, IViewShortcut {
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.Gradient; }
     }
 
-    private PreprocessingContext Context { get; set; }
-    private ITransactionalPreprocessingData PreprocessingData {
+    public PreprocessingContext Context { get; private set; }
+    public ITransactionalPreprocessingData PreprocessingData {
       get { return Context.Data; }
     }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/DataCompletenessChartContent.cs

@@ -1 @@
-using System.Drawing;
+#region License Information
+/* HeuristicLab
+ * Copyright (C) 2002-2016 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
+ *
+ * This file is part of HeuristicLab.
+ *
+ * HeuristicLab is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * HeuristicLab is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
+ */
+#endregion
+
+using System.Drawing;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
 
 namespace HeuristicLab.DataPreprocessing {
-  [Item("DataCompletenessChart", "Represents a datacompleteness chart.")]
+  [Item("Data Completeness Chart", "Represents a datacompleteness chart.")]
 
-  public class DataCompletenessChartContent : Item, IViewChartShortcut {
+  public class DataCompletenessChartContent : Item, IViewShortcut {
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.EditBrightnessContrast; }
     }
 
-    //public DataGridLogic DataGridLogic { get; private set; }
     public SearchLogic SearchLogic { get; private set; }
 

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/DataGridContent.cs

@@ -30 +30 @@
 namespace HeuristicLab.DataPreprocessing {
 
-  [Item("DataGrid", "Represents a data grid.")]
+  [Item("Data Grid", "Represents a data grid.")]
   public class DataGridContent : Item, IStringConvertibleMatrix, IViewShortcut {
-
-    public ITransactionalPreprocessingData PreProcessingData { get; private set; }
-
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.Table; }
     }
+
+    public ITransactionalPreprocessingData PreProcessingData { get; private set; }
 
     public ManipulationLogic ManipulationLogic { get; private set; }
@@ -43 +42 @@
 
     public int Rows {
-      get {
-        return PreProcessingData.Rows;
-      }
-      set {
-        //does nothing
-      }
+      get { return PreProcessingData.Rows; }
+      set { }
     }
 
     public int Columns {
-      get {
-        return PreProcessingData.Columns;
-      }
-      set {
-        //does nothing
-      }
+      get { return PreProcessingData.Columns; }
+      set { }
     }
 
     public IEnumerable<string> ColumnNames {
-      get {
-        return PreProcessingData.VariableNames;
-      }
-      set {
-
-      }
+      get { return PreProcessingData.VariableNames; }
+      set { }
     }
 
     public IEnumerable<string> RowNames {
-      get {
-        return Enumerable.Range(1, Rows).Select(n => n.ToString());
-      }
-      set {
-        throw new NotSupportedException();
-      }
+      get { return Enumerable.Range(1, Rows).Select(n => n.ToString()); }
+      set { throw new NotSupportedException(); }
     }
 
     public bool SortableView {
-      get {
-        return true;
-      }
-      set {
-        throw new NotSupportedException();
-      }
+      get { return true; }
+      set { throw new NotSupportedException(); }
     }
 
@@ -91 +70 @@
     }
 
-
     public IDictionary<int, IList<int>> Selection {
       get { return PreProcessingData.Selection; }
       set { PreProcessingData.Selection = value; }
     }
-
 
     public DataGridContent(ITransactionalPreprocessingData preProcessingData, ManipulationLogic theManipulationLogic, FilterLogic theFilterLogic) {
@@ -137 +114 @@
     }
 
-
     #region unused stuff/not implemented but necessary due to IStringConvertibleMatrix
 #pragma warning disable 0067
@@ -154 +130 @@
     public event EventHandler<EventArgs<int, int>> ItemChanged;
     public event EventHandler Reset;
-
 #pragma warning restore 0067
     #endregion
-
   }
 }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/FilterContent.cs

@@ -28 +28 @@
   [Item("Filter", "Represents the filter grid.")]
   public class FilterContent : Item, IViewShortcut {
-
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.Filter; }
     }
 
-    private ICheckedItemCollection<IFilter> filters = new CheckedItemCollection<IFilter>();
-
     public FilterLogic FilterLogic { get; private set; }
 
-    public ICheckedItemCollection<IFilter> Filters {
-      get {
-        return this.filters;
-      }
-      set {
-        this.filters = value;
-      }
-    }
+    public ICheckedItemCollection<IFilter> Filters { get; private set; }
 
-    private bool isAndCombination = true;
-    public bool IsAndCombination {
-      get {
-        return this.isAndCombination;
-      }
-      set {
-        this.isAndCombination = value;
-      }
-    }
+    public bool IsAndCombination { get; set; }
 
     public FilterContent(FilterLogic filterLogic) {
+      Filters = new CheckedItemCollection<IFilter>();
+      IsAndCombination = true;
       FilterLogic = filterLogic;
     }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/HistogramContent.cs

@@ -20 +20 @@
 #endregion
 
-using System.Collections.Generic;
+using System;
 using System.Drawing;
 using System.Linq;
+using HeuristicLab.Analysis;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
@@ -29 +30 @@
   [Item("Histogram", "Represents the histogram grid.")]
   public class HistogramContent : PreprocessingChartContent {
-
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.Statistics; }
     }
-    private const int MAX_DISTINCT_VALUES_FOR_CLASSIFCATION = 20;
 
-    private int classifierVariableIndex = 0;
+    public string GroupingVariableName { get; set; }
 
-    public int ClassifierVariableIndex {
-      get { return this.classifierVariableIndex; }
-      set { this.classifierVariableIndex = value; }
-    }
-    public bool IsDetailedChartViewEnabled { get; set; }
+    public int Bins { get; set; }
+    public bool ExactBins { get; set; }
 
+    public LegendOrder Order { get; set; }
 
     public HistogramContent(IFilteredPreprocessingData preprocessingData)
       : base(preprocessingData) {
-      AllInOneMode = false;
+      Bins = 10;
+      ExactBins = false;
     }
 
@@ -56 +54 @@
     }
 
-    public IEnumerable<string> GetVariableNamesForHistogramClassification() {
-      List<string> doubleVariableNames = new List<string>();
+    public static DataTable CreateHistogram(IFilteredPreprocessingData preprocessingData, string variableName, string groupingVariableName, DataTableVisualProperties.DataTableHistogramAggregation aggregation, LegendOrder legendOrder = LegendOrder.Appearance) {
+      var dataTable = new DataTable {
+        VisualProperties = { Title = variableName, HistogramAggregation = aggregation },
+      };
 
-      //only return variable names from type double
-      for (int i = 0; i < PreprocessingData.Columns; ++i) {
-        if (PreprocessingData.VariableHasType<double>(i)) {
-          double distinctValueCount = PreprocessingData.GetValues<double>(i).GroupBy(x => x).Count();
-          bool distinctValuesOk = distinctValueCount <= MAX_DISTINCT_VALUES_FOR_CLASSIFCATION;
-          if (distinctValuesOk)
-            doubleVariableNames.Add(PreprocessingData.GetVariableName(i));
-        }
+      if (string.IsNullOrEmpty(groupingVariableName)) {
+        var row = PreprocessingChartContent.CreateDataRow(preprocessingData, variableName, DataRowVisualProperties.DataRowChartType.Histogram);
+        row.VisualProperties.IsVisibleInLegend = false;
+        dataTable.Rows.Add(row);
+        return dataTable;
       }
-      return doubleVariableNames;
+
+      int variableIndex = preprocessingData.GetColumnIndex(variableName);
+      var variableValues = preprocessingData.GetValues<double>(variableIndex);
+      int groupVariableIndex = preprocessingData.GetColumnIndex(groupingVariableName);
+      var groupingValues = Enumerable.Empty<string>();
+
+      if (preprocessingData.VariableHasType<double>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<double>(groupVariableIndex).Select(x => x.ToString());
+      } else if (preprocessingData.VariableHasType<string>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<string>(groupVariableIndex);
+      } else if (preprocessingData.VariableHasType<DateTime>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<DateTime>(groupVariableIndex).Select(x => x.ToString());
+      }
+
+      var groups = groupingValues.Zip(variableValues, Tuple.Create).GroupBy(t => t.Item1, t => t.Item2);
+
+      if (legendOrder == LegendOrder.Alphabetically)
+        groups = groups.OrderBy(x => x.Key, new NaturalStringComparer());
+
+      foreach (var group in groups) {
+        var classRow = new DataRow {
+          Name = group.Key,
+          VisualProperties = {
+              ChartType = DataRowVisualProperties.DataRowChartType.Histogram,
+              IsVisibleInLegend = !string.IsNullOrEmpty(groupingVariableName)
+            }
+        };
+        classRow.Values.AddRange(group);
+        dataTable.Rows.Add(classRow);
+      }
+      return dataTable;
     }
-
   }
 }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/LineChartContent.cs

@@ -26 +26 @@
 namespace HeuristicLab.DataPreprocessing {
 
-  [Item("LineChart", "Represents the line chart grid.")]
+  [Item("Line Chart", "Represents the line chart grid.")]
   public class LineChartContent : PreprocessingChartContent {
+    public bool AllInOneMode { get; set; }
 
     public static new Image StaticItemImage {
@@ -35 +36 @@
     public LineChartContent(IFilteredPreprocessingData preprocessingData)
       : base(preprocessingData) {
+      AllInOneMode = true;
     }
 
     public LineChartContent(LineChartContent content, Cloner cloner)
       : base(content, cloner) {
+      this.AllInOneMode = content.AllInOneMode;
     }
     public override IDeepCloneable Clone(Cloner cloner) {

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/ManipulationContent.cs

@@ -28 +28 @@
   [Item("Manipulation", "Represents the available manipulations on a data set.")]
   public class ManipulationContent : Item, IViewShortcut {
-
-    private ManipulationLogic manipulationLogic;
-    private SearchLogic searchLogic;
-    private FilterLogic filterLogic;
-
-    public ManipulationLogic ManipulationLogic { get { return manipulationLogic; } }
-    public SearchLogic SearchLogic { get { return searchLogic; } }
-    public FilterLogic FilterLogic { get { return filterLogic; } }
-
     public static new Image StaticItemImage {
       get { return HeuristicLab.Common.Resources.VSImageLibrary.Method; }
     }
 
-    public ManipulationContent(ManipulationLogic theManipulationLogic, SearchLogic theSearchLogic, FilterLogic theFilterLogic) {
-      manipulationLogic = theManipulationLogic;
-      searchLogic = theSearchLogic;
-      filterLogic = theFilterLogic;
+    public ManipulationLogic ManipulationLogic { get; private set; }
+    public SearchLogic SearchLogic { get; private set; }
+    public FilterLogic FilterLogic { get; private set; }
+
+    public ManipulationContent(ManipulationLogic manipulationLogic, SearchLogic searchLogic, FilterLogic filterLogic) {
+      ManipulationLogic = manipulationLogic;
+      SearchLogic = searchLogic;
+      FilterLogic = filterLogic;
     }
 
@@ -52 +47 @@
       return new ManipulationContent(this, cloner);
     }
-
-
   }
 }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/PreprocessingChartContent.cs

@@ -23 +23 @@
 using System.Collections.Generic;
 using System.Drawing;
+using System.Linq;
 using HeuristicLab.Analysis;
 using HeuristicLab.Common;
+using HeuristicLab.Common.Resources;
 using HeuristicLab.Core;
 using HeuristicLab.Data;
@@ -30 +32 @@
 namespace HeuristicLab.DataPreprocessing {
   [Item("PreprocessingChart", "Represents a preprocessing chart.")]
-  public class PreprocessingChartContent : Item, IViewChartShortcut {
-    public static new Image StaticItemImage {
-      get { return HeuristicLab.Common.Resources.VSImageLibrary.PieChart; }
+  public class PreprocessingChartContent : Item, IViewShortcut {
+    public enum LegendOrder {
+      Appearance,
+      Alphabetically
     }
 
-    private bool allInOneMode = true;
-    public bool AllInOneMode {
-      get { return this.allInOneMode; }
-      set { this.allInOneMode = value; }
+    public static new Image StaticItemImage {
+      get { return VSImageLibrary.PieChart; }
     }
 
     private ICheckedItemList<StringValue> variableItemList = null;
     public ICheckedItemList<StringValue> VariableItemList {
-      get { return this.variableItemList; }
-      set { this.variableItemList = value; }
+      get {
+        if (variableItemList == null)
+          variableItemList = CreateVariableItemList(PreprocessingData);
+        return this.variableItemList;
+      }
     }
 
     public IFilteredPreprocessingData PreprocessingData { get; private set; }
+    public event DataPreprocessingChangedEventHandler Changed {
+      add { PreprocessingData.Changed += value; }
+      remove { PreprocessingData.Changed -= value; }
+    }
 
     public PreprocessingChartContent(IFilteredPreprocessingData preprocessingData) {
@@ -55 +63 @@
     public PreprocessingChartContent(PreprocessingChartContent content, Cloner cloner)
       : base(content, cloner) {
-      this.allInOneMode = content.allInOneMode;
       this.PreprocessingData = content.PreprocessingData;
       this.variableItemList = cloner.Clone<ICheckedItemList<StringValue>>(variableItemList);
@@ -63 +70 @@
     }
 
+    public DataRow CreateDataRow(string variableName, DataRowVisualProperties.DataRowChartType chartType) {
+      return CreateDataRow(PreprocessingData, variableName, chartType);
+    }
 
-    public DataRow CreateDataRow(string variableName, DataRowVisualProperties.DataRowChartType chartType) {
-      IList<double> values = PreprocessingData.GetValues<double>(PreprocessingData.GetColumnIndex(variableName));
+    public static DataRow CreateDataRow(IFilteredPreprocessingData preprocessingData, string variableName, DataRowVisualProperties.DataRowChartType chartType) {
+      IList<double> values = preprocessingData.GetValues<double>(preprocessingData.GetColumnIndex(variableName));
       DataRow row = new DataRow(variableName, "", values);
       row.VisualProperties.ChartType = chartType;
@@ -71 +81 @@
     }
 
-    public List<DataRow> CreateAllDataRows(DataRowVisualProperties.DataRowChartType chartType) {
-      List<DataRow> dataRows = new List<DataRow>();
-      foreach (var name in PreprocessingData.GetDoubleVariableNames())
-        dataRows.Add(CreateDataRow(name, chartType));
-      return dataRows;
-    }
-
-    public DataRow CreateSelectedDataRow(string variableName, DataRowVisualProperties.DataRowChartType chartType) {
-
-      IDictionary<int, IList<int>> selection = PreprocessingData.Selection;
-      int variableIndex = PreprocessingData.GetColumnIndex(variableName);
-
-      if (selection.Keys.Contains(variableIndex)) {
-        List<int> selectedIndices = new List<int>(selection[variableIndex]);
-        //need selection with more than 1 value
-        if (selectedIndices.Count < 2)
-          return null;
-
-        selectedIndices.Sort();
-        int start = selectedIndices[0];
-        int end = selectedIndices[selectedIndices.Count - 1];
-
-        DataRow rowSelect = CreateDataRowRange(variableName, start, end, chartType);
-        return rowSelect;
-      } else
-        return null;
-    }
-
-    public DataRow CreateDataRowRange(string variableName, int start, int end, DataRowVisualProperties.DataRowChartType chartType) {
-      IList<double> values = PreprocessingData.GetValues<double>(PreprocessingData.GetColumnIndex(variableName));
-      IList<double> valuesRange = new List<double>();
-      for (int i = 0; i < values.Count; i++) {
-        if (i >= start && i <= end)
-          valuesRange.Add(values[i]);
-        else
-          valuesRange.Add(Double.NaN);
-      }
-
-      DataRow row = new DataRow(variableName, "", valuesRange);
-      row.VisualProperties.ChartType = chartType;
-      return row;
-    }
-
-    public List<DataRow> CreateAllSelectedDataRows(DataRowVisualProperties.DataRowChartType chartType) {
-      List<DataRow> dataRows = new List<DataRow>();
-      foreach (var name in PreprocessingData.GetDoubleVariableNames()) {
-        DataRow row = CreateSelectedDataRow(name, chartType);
-        if (row != null)
-          dataRows.Add(row);
-      }
-      return dataRows;
-    }
-
-
-    public ICheckedItemList<StringValue> CreateVariableItemList(IList<string> checkedItems = null) {
-      if (checkedItems == null) checkedItems = new string[0];
+    private static ICheckedItemList<StringValue> CreateVariableItemList(IPreprocessingData preprocessingData) {
       ICheckedItemList<StringValue> itemList = new CheckedItemList<StringValue>();
-      foreach (string name in PreprocessingData.GetDoubleVariableNames()) {
+      foreach (string name in preprocessingData.GetDoubleVariableNames()) {
         var n = new StringValue(name);
-        itemList.Add(n, checkedItems.Contains(name));
+        bool isInputTarget = preprocessingData.InputVariables.Contains(name) || preprocessingData.TargetVariable == name;
+        itemList.Add(n, isInputTarget);
       }
       return new ReadOnlyCheckedItemList<StringValue>(itemList);
     }
 
-    public event DataPreprocessingChangedEventHandler Changed {
-      add { PreprocessingData.Changed += value; }
-      remove { PreprocessingData.Changed -= value; }
+    private const int MAX_DISTINCT_VALUES_FOR_GROUPING = 20;
+    public static IEnumerable<string> GetVariableNamesForGrouping(IPreprocessingData preprocessingData) {
+      var variableNames = new List<string>();
+
+      for (int i = 0; i < preprocessingData.Columns; ++i) {
+        int distinctValues = Int32.MaxValue;
+        if (preprocessingData.VariableHasType<double>(i))
+          distinctValues = preprocessingData.GetValues<double>(i).GroupBy(x => x).Count();
+        else if (preprocessingData.VariableHasType<string>(i))
+          distinctValues = preprocessingData.GetValues<string>(i).GroupBy(x => x).Count();
+        else if (preprocessingData.VariableHasType<DateTime>(i))
+          distinctValues = preprocessingData.GetValues<DateTime>(i).GroupBy(x => x).Count();
+
+        if (distinctValues <= MAX_DISTINCT_VALUES_FOR_GROUPING)
+          variableNames.Add(preprocessingData.GetVariableName(i));
+      }
+      return variableNames;
     }
-
   }
 }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/ScatterPlotContent.cs

@@ -20 +20 @@
 #endregion
 
+using System;
 using System.Collections.Generic;
-using System.Drawing;
 using System.Linq;
 using HeuristicLab.Analysis;
 using HeuristicLab.Common;
-using HeuristicLab.Core;
+using HeuristicLab.Visualization.ChartControlsExtensions;
 
 namespace HeuristicLab.DataPreprocessing {
 
-  [Item("ScatterPlot", "Represents a scatter plot.")]
-  public class ScatterPlotContent : PreprocessingChartContent {
+  public abstract class ScatterPlotContent : PreprocessingChartContent {
+    public string GroupingVariable { get; set; }
 
-    public string SelectedXVariable { get; set; }
-    public string SelectedYVariable { get; set; }
-    public string SelectedColorVariable { get; set; }
-
-    public ScatterPlotContent(IFilteredPreprocessingData preprocessingData)
+    protected ScatterPlotContent(IFilteredPreprocessingData preprocessingData)
       : base(preprocessingData) {
     }
 
-    public ScatterPlotContent(ScatterPlotContent content, Cloner cloner)
+    protected ScatterPlotContent(ScatterPlotContent content, Cloner cloner)
       : base(content, cloner) {
-      this.SelectedXVariable = content.SelectedXVariable;
-      this.SelectedYVariable = content.SelectedYVariable;
-      this.SelectedColorVariable = content.SelectedColorVariable;
     }
 
-    public static new Image StaticItemImage {
-      get { return HeuristicLab.Common.Resources.VSImageLibrary.Performance; }
-    }
-
-    public override IDeepCloneable Clone(Cloner cloner) {
-      return new ScatterPlotContent(this, cloner);
-    }
-
-    public ScatterPlot CreateScatterPlot(string variableNameX, string variableNameY, string variableNameColor = "-") {
+    public static ScatterPlot CreateScatterPlot(IFilteredPreprocessingData preprocessingData, string variableNameX, string variableNameY, string variableNameGroup = "-", LegendOrder legendOrder = LegendOrder.Appearance) {
       ScatterPlot scatterPlot = new ScatterPlot();
 
-      IList<double> xValues = PreprocessingData.GetValues<double>(PreprocessingData.GetColumnIndex(variableNameX));
-      IList<double> yValues = PreprocessingData.GetValues<double>(PreprocessingData.GetColumnIndex(variableNameY));
-      if (variableNameColor == null || variableNameColor == "-") {
-        List<Point2D<double>> points = new List<Point2D<double>>();
+      IList<double> xValues = preprocessingData.GetValues<double>(preprocessingData.GetColumnIndex(variableNameX));
+      IList<double> yValues = preprocessingData.GetValues<double>(preprocessingData.GetColumnIndex(variableNameY));
 
-        for (int i = 0; i < xValues.Count; i++) {
-          Point2D<double> point = new Point2D<double>(xValues[i], yValues[i]);
-          points.Add(point);
-        }
+      var points = xValues.Zip(yValues, (x, y) => new Point2D<double>(x, y)).ToList();
+      var validPoints = points.Where(p => !double.IsNaN(p.X) && !double.IsNaN(p.Y) && !double.IsInfinity(p.X) && !double.IsInfinity(p.Y)).ToList();
+      if (validPoints.Any()) {
+        try {
+          double axisMin, axisMax, axisInterval;
+          ChartUtil.CalculateOptimalAxisInterval(validPoints.Min(p => p.X), validPoints.Max(p => p.X), out axisMin, out axisMax, out axisInterval);
+          scatterPlot.VisualProperties.XAxisMinimumAuto = false;
+          scatterPlot.VisualProperties.XAxisMaximumAuto = false;
+          scatterPlot.VisualProperties.XAxisMinimumFixedValue = axisMin;
+          scatterPlot.VisualProperties.XAxisMaximumFixedValue = axisMax;
+        } catch (ArgumentOutOfRangeException) { } // error during CalculateOptimalAxisInterval 
+        try {
+          double axisMin, axisMax, axisInterval;
+          ChartUtil.CalculateOptimalAxisInterval(validPoints.Min(p => p.Y), validPoints.Max(p => p.Y), out axisMin, out axisMax, out axisInterval);
+          scatterPlot.VisualProperties.YAxisMinimumAuto = false;
+          scatterPlot.VisualProperties.YAxisMaximumAuto = false;
+          scatterPlot.VisualProperties.YAxisMinimumFixedValue = axisMin;
+          scatterPlot.VisualProperties.YAxisMaximumFixedValue = axisMax;
+        } catch (ArgumentOutOfRangeException) { } // error during CalculateOptimalAxisInterval 
+      }
 
-        ScatterPlotDataRow scdr = new ScatterPlotDataRow(variableNameX + " - " + variableNameY, "", points);
+
+      //No Grouping
+      if (string.IsNullOrEmpty(variableNameGroup) || variableNameGroup == "-") {
+        ScatterPlotDataRow scdr = new ScatterPlotDataRow(variableNameX + " - " + variableNameY, "", validPoints);
+        scdr.VisualProperties.IsVisibleInLegend = false;
         scatterPlot.Rows.Add(scdr);
+        return scatterPlot;
+      }
 
-      } else {
-        var colorValues = PreprocessingData.GetValues<double>(PreprocessingData.GetColumnIndex(variableNameColor));
-        var data = xValues.Zip(yValues, (x, y) => new { x, y }).Zip(colorValues, (v, c) => new { v.x, v.y, c }).ToList();
-        var gradients = ColorGradient.Colors;
-        int curGradient = 0;
-        int numColors = colorValues.Distinct().Count();
-        foreach (var colorValue in colorValues.Distinct()) {
-          var values = data.Where(x => x.c == colorValue);
-          var row = new ScatterPlotDataRow(
-            variableNameX + " - " + variableNameY + " (" + colorValue + ")",
-            "",
-            values.Select(v => new Point2D<double>(v.x, v.y)),
-            new ScatterPlotDataRowVisualProperties() { Color = gradients[curGradient] });
-          curGradient += gradients.Count / numColors;
-          scatterPlot.Rows.Add(row);
-        }
+      //Grouping
+      int groupVariableIndex = preprocessingData.GetColumnIndex(variableNameGroup);
+      var groupingValues = Enumerable.Empty<string>();
+
+      if (preprocessingData.VariableHasType<double>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<double>(groupVariableIndex).Select(x => x.ToString());
+      } else if (preprocessingData.VariableHasType<string>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<string>(groupVariableIndex);
+      } else if (preprocessingData.VariableHasType<DateTime>(groupVariableIndex)) {
+        groupingValues = preprocessingData.GetValues<DateTime>(groupVariableIndex).Select(x => x.ToString());
+      }
+      var groups = groupingValues.Zip(validPoints, Tuple.Create).GroupBy(t => t.Item1, t => t.Item2);
+
+      if (legendOrder == LegendOrder.Alphabetically)
+        groups = groups.OrderBy(x => x.Key, new NaturalStringComparer());
+
+      foreach (var group in groups) {
+        var scdr = new ScatterPlotDataRow {
+          Name = group.Key,
+          VisualProperties = {
+            IsVisibleInLegend = true,
+            PointSize = 6
+          }
+        };
+        scdr.Points.AddRange(group);
+        scatterPlot.Rows.Add(scdr);
       }
       return scatterPlot;

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/StatisticsContent.cs

@@ -27 +27 @@
   [Item("Statistics", "Represents the statistics grid.")]
   public class StatisticsContent : Item, IViewShortcut {
+    public static new Image StaticItemImage {
+      get { return HeuristicLab.Common.Resources.VSImageLibrary.Object; }
+    }
 
-    private readonly StatisticsLogic statisticsLogic;
-    public StatisticsContent(StatisticsLogic theStatisticsLogic) {
-      statisticsLogic = theStatisticsLogic;
+    public ITransactionalPreprocessingData PreprocessingData { get; private set; }
+    public StatisticsLogic StatisticsLogic { get; private set; }
+
+    public StatisticsContent(ITransactionalPreprocessingData preProcessingData, StatisticsLogic statisticsLogic) {
+      PreprocessingData = preProcessingData;
+      StatisticsLogic = statisticsLogic;
     }
 
     public StatisticsContent(StatisticsContent content, Cloner cloner)
       : base(content, cloner) {
-
-    }
-
-    public StatisticsLogic StatisticsLogic {
-      get { return statisticsLogic; }
-    }
-
-    public static new Image StaticItemImage {
-      get { return HeuristicLab.Common.Resources.VSImageLibrary.Object; }
     }
 
@@ -51 +48 @@
 
     public event DataPreprocessingChangedEventHandler Changed {
-      add { statisticsLogic.Changed += value; }
-      remove { statisticsLogic.Changed -= value; }
+      add { StatisticsLogic.Changed += value; }
+      remove { StatisticsLogic.Changed -= value; }
     }
-
   }
 }

trunk/sources/HeuristicLab.DataPreprocessing/3.4/Content/TransformationContent.cs

@@ -28 +28 @@
   [Item("Transformation", "Represents the transformation grid.")]
   public class TransformationContent : Item, IViewShortcut {
+    public static new Image StaticItemImage {
+      get { return HeuristicLab.Common.Resources.VSImageLibrary.Method; }
+    }
 
     public IPreprocessingData Data { get; private set; }
@@ -33 +36 @@
 
     public ICheckedItemList<ITransformation> CheckedTransformationList { get; private set; }
-
-    public static new Image StaticItemImage {
-      get { return HeuristicLab.Common.Resources.VSImageLibrary.Method; }
-    }
 
     public TransformationContent(IPreprocessingData data, FilterLogic filterLogic) {