Opened 2 years ago

Closed 18 months ago

#2905 closed defect (done)

Partial dependence plots throws exception for support vector regression solutions

Reported by: abeham Owned by: gkronber
Priority: medium Milestone: HeuristicLab 3.3.16
Component: Algorithms.DataAnalysis Version: trunk
Keywords: Cc:


To reproduce: Open a new SVM, load a problem, e.g. Housing, open the partial dependence plot.

The error message is: "System.ArgumentException: The variable MEDV does not exist in the dataset."

   at HeuristicLab.Problems.DataAnalysis.Dataset.GetValues[T](String variableName) in HeuristicLab.Problems.DataAnalysis\3.4\Dataset.cs:line 243
   at HeuristicLab.Problems.DataAnalysis.Dataset.GetValues[T](String variableName, IEnumerable`1 rows) in HeuristicLab.Problems.DataAnalysis\3.4\Dataset.cs:line 236
   at HeuristicLab.Problems.DataAnalysis.Dataset.GetDoubleValues(String variableName, IEnumerable`1 rows) in HeuristicLab.Problems.DataAnalysis\3.4\Dataset.cs:line 206
   at HeuristicLab.Algorithms.DataAnalysis.SupportVectorMachineUtil.CreateSvmProblem(IDataset dataset, String targetVariable, IEnumerable`1 inputVariables, IEnumerable`1 rowIndices) in HeuristicLab.Algorithms.DataAnalysis\3.4\SupportVectorMachine\SupportVectorMachineUtil.cs:line 43
   at HeuristicLab.Algorithms.DataAnalysis.SupportVectorMachineModel.<GetEstimatedValuesHelper>d__8.MoveNext() in HeuristicLab.Algorithms.DataAnalysis\3.4\SupportVectorMachine\SupportVectorMachineModel.cs:line 157
   at System.Collections.Generic.List`1..ctor(IEnumerable`1 collection)
   at System.Linq.Enumerable.ToList[TSource](IEnumerable`1 source)
   at HeuristicLab.Problems.DataAnalysis.Views.PartialDependencePlot.<>c__DisplayClass1d.<UpdateSeriesDataAsync>b__1c() in HeuristicLab.Problems.DataAnalysis.Views\3.4\Controls\PartialDependencePlot.cs:line 470

Change History (7)

comment:1 Changed 2 years ago by gkronber

  • Owner set to gkronber
  • Status changed from new to accepted

comment:2 Changed 2 years ago by gkronber

The problem can be reproduced.

comment:3 Changed 2 years ago by gkronber

r15854: don't include the target variable in the svm_problem when a trained SVM is used for prediction

comment:4 Changed 2 years ago by gkronber

  • Owner changed from gkronber to abeham
  • Status changed from accepted to reviewing
  • Version set to trunk

comment:5 Changed 21 months ago by abeham

  • Owner changed from abeham to gkronber
  • Status changed from reviewing to readytorelease

I tested it and it works

comment:6 Changed 19 months ago by gkronber

r16160: merged r15854 from trunk to stable

comment:7 Changed 18 months ago by gkronber

  • Resolution set to done
  • Status changed from readytorelease to closed
Note: See TracTickets for help on using tickets.