Changeset 17080 for trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees

Timestamp:

07/05/19 11:08:57 (5 years ago)

Author:

gkronber

Message:

#2847: more changes for renaming the M5 plugin

Location:

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4

Files:

• DecisionTreeRegression.cs (modified) (6 diffs)
• HeuristicLab.Algorithms.DataAnalysis.DecisionTrees-3.4.csproj (modified) (3 diffs)
• Interfaces/IDecisionTreeModel.cs (moved) (moved from trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/Interfaces/IM5Model.cs)
• LeafTypes/Leaf.cs (moved) (moved from trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/LeafTypes/M5Leaf.cs)
• LeafTypes/LeafBase.cs (modified) (1 diff)
• LeafTypes/LinearLeaf.cs (modified) (1 diff)
• LeafTypes/RegularizedLeaf.cs (moved) (moved from trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/LeafTypes/M5regLeaf.cs)
• MetaModels/RegressionNodeTreeModel.cs (modified) (3 diffs)
• MetaModels/RegressionRuleModel.cs (modified) (2 diffs)
• MetaModels/RegressionRuleSetModel.cs (modified) (1 diff)
• Pruning/ComplexityPruning.cs (modified) (1 diff)
• Splitting/Splitter.cs (moved) (moved from trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/Splitting/M5Splitter.cs) (1 diff)
• Splitting/SplitterBase.cs (modified) (1 diff)

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/DecisionTreeRegression.cs

@@ -38 +38 @@
   [StorableType("FC8D8E5A-D16D-41BB-91CF-B2B35D17ADD7")]
   [Creatable(CreatableAttribute.Categories.DataAnalysisRegression, Priority = 95)]
-  [Item("M5RegressionTree", "A M5 regression tree / rule set")]
-  public sealed class M5Regression : FixedDataAnalysisAlgorithm<IRegressionProblem> {
+  [Item("Decision tree regression", "A regression tree / rule set learner")]
+  public sealed class DecisionTreeRegression : FixedDataAnalysisAlgorithm<IRegressionProblem> {
     public override bool SupportsPause {
       get { return true; }
@@ -135 +135 @@
     #region Constructors and Cloning
     [StorableConstructor]
-    private M5Regression(StorableConstructorFlag _) : base(_) { }
-    private M5Regression(M5Regression original, Cloner cloner) : base(original, cloner) {
+    private DecisionTreeRegression(StorableConstructorFlag _) : base(_) { }
+    private DecisionTreeRegression(DecisionTreeRegression original, Cloner cloner) : base(original, cloner) {
       stateScope = cloner.Clone(stateScope);
     }
-    public M5Regression() {
+    public DecisionTreeRegression() {
       var modelSet = new ItemSet<ILeafModel>(ApplicationManager.Manager.GetInstances<ILeafModel>());
       var pruningSet = new ItemSet<IPruning>(ApplicationManager.Manager.GetInstances<IPruning>());
@@ -145 +145 @@
       Parameters.Add(new FixedValueParameter<BoolValue>(GenerateRulesParameterName, "Whether a set of rules or a decision tree shall be created (default=false)", new BoolValue(false)));
       Parameters.Add(new FixedValueParameter<PercentValue>(HoldoutSizeParameterName, "How much of the training set shall be reserved for pruning (default=20%).", new PercentValue(0.2)));
-      Parameters.Add(new ConstrainedValueParameter<ISplitter>(SplitterParameterName, "The type of split function used to create node splits (default='M5Splitter').", splitterSet, splitterSet.OfType<M5Splitter>().First()));
+      Parameters.Add(new ConstrainedValueParameter<ISplitter>(SplitterParameterName, "The type of split function used to create node splits (default='Splitter').", splitterSet, splitterSet.OfType<M5Splitter>().First()));
       Parameters.Add(new FixedValueParameter<IntValue>(MinimalNodeSizeParameterName, "The minimal number of samples in a leaf node (default=1).", new IntValue(1)));
       Parameters.Add(new ConstrainedValueParameter<ILeafModel>(LeafModelParameterName, "The type of model used for the nodes (default='LinearLeaf').", modelSet, modelSet.OfType<LinearLeaf>().First()));
@@ -155 +155 @@
     }
     public override IDeepCloneable Clone(Cloner cloner) {
-      return new M5Regression(this, cloner);
+      return new DecisionTreeRegression(this, cloner);
     }
     #endregion
@@ -204 +204 @@
       var doubleVars = new HashSet<string>(problemData.Dataset.DoubleVariables);
       var vars = problemData.AllowedInputVariables.Concat(new[] {problemData.TargetVariable}).ToArray();
-      if (vars.Any(v => !doubleVars.Contains(v))) throw new NotSupportedException("M5 regression supports only double valued input or output features.");
+      if (vars.Any(v => !doubleVars.Contains(v))) throw new NotSupportedException("Decision tree regression supports only double valued input or output features.");
       var doubles = vars.Select(v => problemData.Dataset.GetDoubleValues(v, problemData.TrainingIndices).ToArray()).ToArray();
       if (doubles.Any(v => v.Any(x => double.IsNaN(x) || double.IsInfinity(x))))
-        throw new NotSupportedException("M5 regression does not support NaN or infinity values in the input dataset.");
+        throw new NotSupportedException("Decision tree regression does not support NaN or infinity values in the input dataset.");
       var trainingData = new Dataset(vars, doubles);
       var pd = new RegressionProblemData(trainingData, problemData.AllowedInputVariables, problemData.TargetVariable);
@@ -283 +283 @@
       var ruleSet = solution.Model as RegressionRuleSetModel;
       if (ruleSet != null) {
-        results.Add(RegressionTreeAnalyzer.CreateRulesResult(ruleSet, problemData, "M5Rules", true));
+        results.Add(RegressionTreeAnalyzer.CreateRulesResult(ruleSet, problemData, "Rules", true));
         frequencies = RegressionTreeAnalyzer.GetRuleVariableFrequences(ruleSet);
         results.Add(RegressionTreeAnalyzer.CreateCoverageDiagram(ruleSet, problemData));

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees-3.4.csproj

@@ -68 +68 @@
   <ItemGroup>
     <Compile Include="DecisionTreeRegression.cs" />
+    <Compile Include="Interfaces\IDecisionTreeModel.cs" />
     <Compile Include="Interfaces\ILeafModel.cs" />
-    <Compile Include="Interfaces\IM5Model.cs" />
     <Compile Include="Interfaces\IPruning.cs" />
     <Compile Include="Interfaces\ISplitter.cs" />
@@ -79 +79 @@
     <Compile Include="LeafTypes\ConstantLeaf.cs" />
     <Compile Include="LeafTypes\GaussianProcessLeaf.cs" />
+    <Compile Include="LeafTypes\Leaf.cs" />
     <Compile Include="LeafTypes\LeafBase.cs" />
     <Compile Include="LeafTypes\LinearLeaf.cs" />
-    <Compile Include="LeafTypes\M5Leaf.cs" />
-    <Compile Include="LeafTypes\M5regLeaf.cs" />
+    <Compile Include="LeafTypes\RegularizedLeaf.cs" />
     <Compile Include="MetaModels\RegressionNodeModel.cs" />
     <Compile Include="MetaModels\RegressionNodeTreeModel.cs" />
@@ -93 +93 @@
     <Compile Include="Splitting\CorrelationImpurityCalculator.cs" />
     <Compile Include="Splitting\CorrelationSplitter.cs" />
-    <Compile Include="Splitting\M5Splitter.cs" />
     <Compile Include="Splitting\NeumaierSum.cs" />
     <Compile Include="Splitting\OrderImpurityCalculator.cs" />
+    <Compile Include="Splitting\Splitter.cs" />
     <Compile Include="Splitting\SplitterBase.cs" />
     <Compile Include="Splitting\UnivariateOnlineLR.cs" />

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/LeafTypes/LeafBase.cs

@@ -72 +72 @@
 
     public void Build(RegressionNodeTreeModel tree, IReadOnlyList<int> trainingRows, IScope stateScope, CancellationToken cancellationToken) {
-      var parameters = (RegressionTreeParameters)stateScope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var parameters = (RegressionTreeParameters)stateScope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       var state = (LeafBuildingState)stateScope.Variables[LeafBuildingStateVariableName].Value;
 

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/LeafTypes/LinearLeaf.cs

@@ -30 +30 @@
 namespace HeuristicLab.Algorithms.DataAnalysis {
   [StorableType("347CA25D-FB37-4C4F-9B61-9D79288B2B28")]
-  [Item("LinearLeaf", "A leaf type that uses linear models as leaf models. This is the standard for M5' regression")]
+  [Item("LinearLeaf", "A leaf type that uses linear models as leaf models. This is the standard for decision tree regression")]
   public class LinearLeaf : LeafBase {
     #region Constructors & Cloning

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/MetaModels/RegressionNodeTreeModel.cs

@@ -71 +71 @@
     #region IM5Model
     public void Build(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, ResultCollection results, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       //start with one node
       if (Root == null)
@@ -87 +87 @@
 
     public void Update(IReadOnlyList<int> rows, IScope statescope, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       regressionTreeParams.LeafModel.Build(this, rows, statescope, cancellationToken);
     }
 
     public static void Initialize(IScope stateScope) {
-      var param = (RegressionTreeParameters)stateScope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var param = (RegressionTreeParameters)stateScope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       stateScope.Variables.Add(new Variable(RootVariableName, RegressionNodeModel.CreateNode(param.TargetVariable, param)));
     }
@@ -98 +98 @@
 
     public void BuildModel(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, ResultCollection results, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       //start with one node
       Root = RegressionNodeModel.CreateNode(regressionTreeParams.TargetVariable, regressionTreeParams);

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/MetaModels/RegressionRuleModel.cs

@@ -83 +83 @@
 
     public void Build(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, ResultCollection results, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       variables = regressionTreeParams.AllowedInputVariables.ToList();
 
@@ -109 +109 @@
 
     public void Update(IReadOnlyList<int> rows, IScope statescope, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       int np;
       RuleModel = regressionTreeParams.LeafModel.BuildModel(rows, regressionTreeParams, cancellationToken, out np);

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/MetaModels/RegressionRuleSetModel.cs

@@ -77 +77 @@
     #region IM5Model
     public void Build(IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope stateScope, ResultCollection results, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)stateScope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)stateScope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       var ruleSetState = (RuleSetState)stateScope.Variables[RuleSetStateVariableName].Value;
 

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/Pruning/ComplexityPruning.cs

@@ -86 +86 @@
 
     public void Prune(RegressionNodeTreeModel treeModel, IReadOnlyList<int> trainingRows, IReadOnlyList<int> pruningRows, IScope statescope, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)statescope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       var state = (PruningState)statescope.Variables[PruningStateVariableName].Value;
 

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/Splitting/Splitter.cs

@@ -30 +30 @@
 namespace HeuristicLab.Algorithms.DataAnalysis {
   [StorableType("502B1429-7A28-45C1-A60A-93E72CB3AF4A")]
-  [Item("M5Splitter", "A split selector that uses the ratio between Variances^(1/Order) to determine good splits.")]
+  [Item("Splitter", "A split selector that uses the ratio between Variances^(1/Order) to determine good splits.")]
   public sealed class M5Splitter : SplitterBase {
     public const string OrderParameterName = "Order";

trunk/HeuristicLab.Algorithms.DataAnalysis.DecisionTrees/3.4/Splitting/SplitterBase.cs

@@ -48 +48 @@
 
     public void Split(RegressionNodeTreeModel tree, IReadOnlyList<int> trainingRows, IScope stateScope, CancellationToken cancellationToken) {
-      var regressionTreeParams = (RegressionTreeParameters)stateScope.Variables[M5Regression.RegressionTreeParameterVariableName].Value;
+      var regressionTreeParams = (RegressionTreeParameters)stateScope.Variables[DecisionTreeRegression.RegressionTreeParameterVariableName].Value;
       var splittingState = (SplittingState)stateScope.Variables[SplittingStateVariableName].Value;
       var variables = regressionTreeParams.AllowedInputVariables.ToArray();