Parameters in SVM

hmhsinghmhsing Member Posts: 15  Maven

I am practicing on Rapidminer. When I redo the Credit Risk Modeling example in NEW PROCESS, although I input kernel gamma = 5.0 and C = 100.0 in Support Vector Machine (SVM), both will change to infinity after I run the process. The result is therefore different from what example shows. I couldn't find the reason, hope someone can help.  

Tagged:

Best Answer

  • mschmitzmschmitz Posts: 2,113  RM Data Scientist
    Solution Accepted

    Dear hmhsing,

     

    the Credit Risk template uses a Optimize Parameters around. This Optimize overwrites the concrete settings of the SVM. If you want to change it, you need to change the settings of optimize Parameters.

     

    ~Martin

    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.