Options

For Loop implementation

sectynsectyn Member Posts: 25 Maven
I'm looking to implement something like a 'for loop' in RapidMiner. The input to the 'for loop' will be a dataset of say 5 attributes. For each run of the 'for loop', I need to add one new attribute. So if the 'for loop' runs 5 times, I end up with a dataset of 10 attributes. With the Loop operator the dataset that goes into the loop is the initial one with 5 attributes. How do I append attributes for each run and get that new dataset with the additional attribute as the input for the next run of my Loop operator? Or should I be using some other operator for this?

Best Answers

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,517 RM Data Scientist
    Solution Accepted
    Hi,
    you likely want to use the normal Loop with the "reuse results" option.

    Best,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • Options
    kaymankayman Member Posts: 662 Unicorn
    Solution Accepted
    You can do as follows : 

    Use the loop operator -> Number of iterations = 5 and ensure you select 'reuse results'
    In the loop operator add 'generate attribute' and use the iteration value (1 to 5) or use a macro to generate something more fancy.

    Attached some example to get you started
Sign In or Register to comment.