Changeset 15338 for branches/EfficientGlobalOptimization/HeuristicLab.Algorithms.EGO/Operators/ModelBuilder.cs

Timestamp:

08/22/17 15:01:46 (7 years ago)

Author:

bwerth

Message:

#2745 fixed bug concerning new Start and StartAsync methods; passed CancellationToken to sub algorithms

File:

• branches/EfficientGlobalOptimization/HeuristicLab.Algorithms.EGO/Operators/ModelBuilder.cs (modified) (7 diffs)

branches/EfficientGlobalOptimization/HeuristicLab.Algorithms.EGO/Operators/ModelBuilder.cs

@@ -22 +22 @@
 using System;
 using System.Linq;
+using System.Threading;
 using HeuristicLab.Algorithms.DataAnalysis;
 using HeuristicLab.Common;
@@ -31 +32 @@
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 using HeuristicLab.Problems.DataAnalysis;
+using HeuristicLab.Problems.SurrogateProblem;
 
 namespace HeuristicLab.Algorithms.EGO {
@@ -38 +40 @@
   [Item("ModelBuilder", "Builds a model from a dataset and a given RegressionAlgorithm")]
   [StorableClass]
-  public class ModelBuilder : InstrumentedOperator, IStochasticOperator {
+  public class ModelBuilder : InstrumentedOperator, IStochasticOperator, ICancellableOperator {
     public override bool CanChangeName => true;
+    public CancellationToken Cancellation { get; set; }
 
+    #region Parameter properties
     public ILookupParameter<IDataAnalysisAlgorithm<IRegressionProblem>> RegressionAlgorithmParameter => (ILookupParameter<IDataAnalysisAlgorithm<IRegressionProblem>>)Parameters["RegressionAlgorithm"];
     public ILookupParameter<IRegressionSolution> ModelParameter => (ILookupParameter<IRegressionSolution>)Parameters["Model"];
@@ -47 +51 @@
     public ILookupParameter<IntValue> MaxModelSizeParameter => (ILookupParameter<IntValue>)Parameters["Maximal Model Size"];
     public ILookupParameter<DoubleMatrix> InfillBoundsParameter => (ILookupParameter<DoubleMatrix>)Parameters["InfillBounds"];
+    #endregion
 
     [StorableConstructor]
@@ -93 +98 @@
     }
 
-    private static IRegressionSolution BuildModel(IRandom random, IDataAnalysisAlgorithm<IRegressionProblem> regressionAlgorithm, IDataset dataset, IRegressionSolution oldSolution) {
+    private IRegressionSolution BuildModel(IRandom random, IDataAnalysisAlgorithm<IRegressionProblem> regressionAlgorithm, IDataset dataset, IRegressionSolution oldSolution) {
       //var dataset = EgoUtilities.GetDataSet(dataSamples, RemoveDuplicates);
       var problemdata = new RegressionProblemData(dataset, dataset.VariableNames.Where(x => !x.Equals("output")), "output");
@@ -108 +113 @@
 
       while (solution == null && i++ < 100) {
-        var results = EgoUtilities.SyncRunSubAlgorithm(regressionAlgorithm, random.Next(int.MaxValue));
+        var results = EgoUtilities.SyncRunSubAlgorithm(regressionAlgorithm, random.Next(int.MaxValue), Cancellation);
         solution = results.Select(x => x.Value).OfType<IRegressionSolution>().SingleOrDefault();
       }
 
-      //try creating a model with old hyperparameters and new dataset;
-      var gp = regressionAlgorithm as GaussianProcessRegression;
-      var oldmodel = oldSolution as GaussianProcessRegressionSolution;
-      if (gp != null && oldmodel != null) {
-        var mean = (IMeanFunction)oldmodel.Model.MeanFunction.Clone();
-        var cov = (ICovarianceFunction)oldmodel.Model.CovarianceFunction.Clone();
-        try {
-          var model = new GaussianProcessModel(problemdata.Dataset, problemdata.TargetVariable,
-            problemdata.AllowedInputVariables, problemdata.TrainingIndices, new[] { 0.0 }, mean, cov);
-          model.FixParameters();
-          var sol = new GaussianProcessRegressionSolution(model, problemdata);
-          if (solution == null || solution.TrainingMeanSquaredError > sol.TrainingMeanSquaredError) {
-            solution = sol;
-          }
-        }
-        catch (ArgumentException) { }
-      }
+      if (regressionAlgorithm is GaussianProcessRegression && oldSolution != null)
+        solution = SanitizeGaussianProcess(oldSolution as GaussianProcessRegressionSolution, solution as GaussianProcessRegressionSolution, Cancellation);
+
+      if (regressionAlgorithm is M5RegressionTree && oldSolution != null)
+        solution = SanitizeM5Regression(oldSolution.Model as M5Model, solution, random, Cancellation);
+
 
       regressionAlgorithm.Runs.Clear();
@@ -134 +128 @@
 
     }
+
+    private static IRegressionSolution SanitizeM5Regression(M5Model oldmodel, IRegressionSolution newSolution, IRandom random, CancellationToken cancellation) {
+      var problemdata = newSolution.ProblemData;
+      oldmodel.UpdateLeafModels(problemdata, problemdata.AllIndices, random, cancellation);
+      var oldSolution = oldmodel.CreateRegressionSolution(problemdata);
+      var magicDecision = newSolution.TrainingRSquared < oldSolution.TrainingRSquared - 0.05;
+      return magicDecision ? newSolution : oldmodel.CreateRegressionSolution(problemdata);
+    }
+
+    //try creating a model with old hyperparameters and new dataset;
+    private static IRegressionSolution SanitizeGaussianProcess(GaussianProcessRegressionSolution oldmodel, GaussianProcessRegressionSolution newSolution, CancellationToken cancellation) {
+      var problemdata = newSolution.ProblemData;
+      var mean = (IMeanFunction)oldmodel.Model.MeanFunction.Clone();
+      var cov = (ICovarianceFunction)oldmodel.Model.CovarianceFunction.Clone();
+      try {
+        var model = new GaussianProcessModel(problemdata.Dataset, problemdata.TargetVariable, problemdata.AllowedInputVariables, problemdata.TrainingIndices, new[] { 0.0 }, mean, cov);
+        cancellation.ThrowIfCancellationRequested();
+        model.FixParameters();
+        var sol = new GaussianProcessRegressionSolution(model, problemdata);
+        if (newSolution.TrainingMeanSquaredError > sol.TrainingMeanSquaredError) {
+          newSolution = sol;
+        }
+      } catch (ArgumentException) { }
+      return newSolution;
+    }
+
   }
 }