Options

Attaching performance operator for testing data.

nn_herenn_here Member Posts: 31 Contributor I
Hi,
I am having an issue where in i need to find out the performance of a testing dataset.In my current scenario i am using multi label modelling to predict multiple temperature attributes. So the predicted values are coming out.But when i attach performance operator to the apply model of the testing dataset, it says 'Input example set doesn't have a label attribute'. When i used set role operator for the same,it displayed  'Input example set doesnt have a predicted label attribute'.In our earlier cases ,i had used crossvalidation operator which had performance operator inside testing part. For multilabel modelling operator iam unaware how to proceed.  Please find the processes used. DecisionTree.rmp, without attaching performance, DecisionTreePerf.rmp after attaching performance operator and throwing error. Kindly help.
Thanks and regards,
nn_here

 

Answers

  • Options
    nn_herenn_here Member Posts: 31 Contributor I
    Hi,
    To add to the above doubt,in the same process if i use linear regression or neural network,getting error like 'Insufficient capability-the operator LRdoes not have sufficient capabilities for the given dataset. Polynomial attributes not supported' .All the attributes are of datatype Real and one of Datetime .In the process i am converting datetime to nominal and back to nominal to date. Please let me know if that would have caused the issue.If so,what can i do to rectify the same.Kindly help.

    Thanks and regards,
    nn_here.
Sign In or Register to comment.