Parallel task based on attribute values

mfulgerimfulgeri Member Posts: 3 Contributor I
edited December 2018 in Help

Hi guys,

I'm searching for an elegan solution of the following problem: to train multiple instances of a predictor, separated by the value of an attribute.

Example: say I have a "Country" attribute in may data, and the nomial values of Country are "US" and "UK". 

I could design my process in this way (may in 2 separated files..) and everything would works fine. 

2017-12-08_17-45-58.jpg

Or I could create a loop on the attributes that would do the same job: 

2017-12-08_17-49-16.jpg

and my question is: is there any component or modelling tecnique to implement a logic that is splitting the data based on an attribute (country, in this example) and generating different instances of the model?

2017-12-08_17-51-57.jpg

Many thanks for your help!

 

Thanks, 

Matteo 

Answers

  • Thomas_OttThomas_Ott RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,761 Unicorn

    Yes, you can use Extract Macros (extract the country values), use a Filter inside the loop set to the Macro Value, and then churn out a model for each country by appending the extracted country value to the model file.

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

    Hi,

     

    or use my favourite Group into Collection (from Operator toolbox) + Loop Collection combo :)

     

    Best,

    Martin

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