Problem with RVM learner

ehteshamehtesham Member Posts: 1 Contributor I
edited November 2018 in Help
Friends,
Does Rapidminer supports multiclass RVM,  because i am getting strange error while classification task.
The training dataset has 500 datapoints with 1051 attributes. The first attribute is label. It is a 49 class problem.

When i run it, after some time it gives me error saying  "the setup does not seem to contain any obvious error but you should check the log messages of start debug mode".
in debug mode it shows two errors. 'ArrayIndexOutOfBoundException' occured in 1st application of RVM Learner.

Please guide me.

Answers

  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    Hi,
    if you click on Operator Info in the context menu of the RVM Operator, you will be shown the abilities of the learner. The RVM Learner only supports binominal classes and regression. No multiclass is supported. This is due to the statistical method: Like SVMs, the principially cannot difference between more than 2 classes. If you want to classify more than 2 classes, you will need a wrapper approche which is implemented in the Binary2Multiclass Operator.
    Some SVM implementations do this internally, but the operator provides this functionality generally.

    Greetings,
      Sebastian
Sign In or Register to comment.