The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

save predicted labels

fromfrom Member Posts: 2 Contributor I
edited November 2018 in Help
hi guys,
Im using a standard 'cross validation with SVM' scheme and want to get the predicted labels computed by the trained SVM on the test data set. How do I do that?
Thanks! P.

Answers

  • IngoRMIngoRM Employee-RapidMiner, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder
    Hi,

    simply replace the X-Validation operator by the operator X-Prediction. If you want both, predicted labels and estimated performance in a single run you would have to append the predictions after model application to a data set (via Remember, Recall, and Append or directly writing in a database for example).

    Cheers,
    Ingo
  • fromfrom Member Posts: 2 Contributor I
    thanks! that helped..
  • wesselwessel Member Posts: 537 Maven
    How come the split validation operator can not return the labelled test set?

    It only returns the full "data set". It does not return the part used for training and the part used for testing.
  • TobiasMalbrechtTobiasMalbrecht Moderator, Employee-RapidMiner, Member Posts: 295 RM Product Management
    Hi,
    wessel wrote:

    How come the split validation operator can not return the labelled test set?

    It only returns the full "data set". It does not return the part used for training and the part used for testing.
    that is because the main purpose of that operator is validation, i.e. computation of the performance, not the model application - normally, you would want to get the performance as main output and not the example set.

    Kind regards,
    Tobias
Sign In or Register to comment.