Free cookie consent management tool by TermsFeed Policy Generator

Ignore:
Timestamp:
09/28/09 14:36:15 (15 years ago)
Author:
gkronber
Message:

Fixed #768 (CEDMA engines should use a limited number of trainingsamples to generate more models in less time)

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/sources/HeuristicLab.Modeling/3.2/DefaultRegressionOperators.cs

    r2388 r2396  
    3232      op.Name = "ProblemInjector";
    3333      SequentialProcessor seq = new SequentialProcessor();
     34      ProblemInjector probInjector = new ProblemInjector();
     35      probInjector.AddVariable(new Variable("MaxNumberOfTrainingSamples", new IntData(2000)));
    3436
    35       seq.AddSubOperator(new ProblemInjector());
     37      seq.AddSubOperator(probInjector);
    3638      DatasetShuffler shuffler = new DatasetShuffler();
    3739      shuffler.GetVariableInfo("ShuffleStart").ActualName = "TrainingSamplesStart";
Note: See TracChangeset for help on using the changeset viewer.