Free cookie consent management tool by TermsFeed Policy Generator

Ignore:
Timestamp:
12/07/16 16:25:07 (8 years ago)
Author:
mkommend
Message:

#2670: Minor cahnge in TargetResponseGradientView to includes variables the are used for prediction but not allowed as input variables in the problem data.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/sources/HeuristicLab.Problems.DataAnalysis.Views/3.4/Regression/RegressionSolutionTargetResponseGradientView.cs

    r14307 r14464  
    112112      limitView.Content.Upper = axisMax;
    113113
    114       // create dataset
    115       var allowedInputVariables = Content.ProblemData.AllowedInputVariables;
     114      // create dataset of problemData input variables and model input variables
     115      // necessary workaround to have the variables in the occuring order
     116      var inputvariables =
     117        new HashSet<string>(Content.ProblemData.AllowedInputVariables.Union(Content.Model.VariablesUsedForPrediction));
     118      var allowedInputVariables =
     119        Content.ProblemData.Dataset.VariableNames.Where(v => inputvariables.Contains(v)).ToList();
     120
     121
    116122      var variableValues = allowedInputVariables.Select(x => new List<double> { problemData.Dataset.GetDoubleValues(x, problemData.TrainingIndices).Median() });
    117123      if (sharedFixedVariables != null)
Note: See TracChangeset for help on using the changeset viewer.