Free cookie consent management tool by TermsFeed Policy Generator

Ignore:
Timestamp:
12/25/11 20:52:38 (12 years ago)
Author:
gkronber
Message:

#1685:

  • fixed bug in simplification view for discriminating function classification solutions (incorrect threshold calculation was used)
  • fixed bug in linear discriminant analysis algorithm (an additional constant term was added to the model incorrectly)
File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/sources/HeuristicLab.Algorithms.DataAnalysis/3.4/Linear/LinearDiscriminantAnalysis.cs

    r6649 r7233  
    100100      }
    101101
    102       ConstantTreeNode cNode = (ConstantTreeNode)new Constant().CreateTreeNode();
    103       cNode.Value = w[w.Length - 1];
    104       addition.AddSubtree(cNode);
    105 
    106 
    107102      var model = LinearDiscriminantAnalysis.CreateDiscriminantFunctionModel(tree, new SymbolicDataAnalysisExpressionTreeInterpreter(), problemData, rows);
    108103      SymbolicDiscriminantFunctionClassificationSolution solution = new SymbolicDiscriminantFunctionClassificationSolution(model, (IClassificationProblemData)problemData.Clone());
Note: See TracChangeset for help on using the changeset viewer.