Select prediction attribute from classification output

phivuphivu Member Posts: 34 Guru
edited November 2018 in Help

Hi RapidMiner,

 

I used Apply Model to do a classification, I tried to use "Select Attributes" to extract the special attributes from the output (e.g. prediction(label), confidence(yes), confidence(no)) but could not find these attributes in "select attributes" list (although i checked "include special attributes").

 

My ultimate goal is to gather all the predictions from many binary classifiers to build the output for a multi-class classifier.

 

Thank you!

Best Answers

  • phivuphivu Member Posts: 34 Guru
    Solution Accepted

    Thanks Martin, it worked, but when i put a Rename operator after the Select Attributes and manually type in the attribute name in the Rename, this time it said "Attribute not found", i also tried the "rename additional attributes".

  • phivuphivu Member Posts: 34 Guru
    Solution Accepted

    Hi,

     

    I just want to update the solution for my question. I tried to run the process again, although it still showed a red color at the Rename operator input port, but when I run the process, I got what I want. The error checking process and the GUI synchronization/update are quite lagging.

     

    Thanks.

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist

    Hi Phivucs,

     

    sometimes the meta-data propagation simply does not catch the newly created attributes. you can simply add them to the list of select subset by hand (typing it in the upper right field and press +). It works afterwards.

     

    ~Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.