"[SOLVED] Loop Parameters operator - iteration get parameter values"

julobjulob Member Posts: 2 Contributor I
edited June 2019 in Help
Hello,

I decided to use loop parameters operator because I am looking for the best Neural Network settings. I am changing 3 parameters.

How can I get parameters values made by this operator? I need to know which combination of parameters values belongs to specified output set.

Or at least how the iteration is performed. Whether the looping is performed from the first or the last parameter.

Thank You for answers

Answers

  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hi,
    if the parameters you are looping are text fields, you can loop macro values instead, and insert the macros into the text fields (see process below).
    Or you use a Log operator to log the current parameter settings, then Log to Data to get them into an example set, and Extract Macro to get the actual values for further processing.
    Or you use Generate Macro directly to extract the parameter values.

    To know the current iteration number, you can create a macro construct to count the iterations, or, as in the first branch of the attached process, count the number of logged values.

    Everything is implemented as a demo case in the different branches of the attached process.

    Best,
     Marius
  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Update: instead of posting the process here, I uploaded it to myExperiement. To view it, please install the Community Extension of RapidMiner, then go to View->Show View->myExperiment Browser and search for "Macro Magic: get state of Loop Parameters".

    Best, Marius
  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Yes, see my "update" post about myExperiment above :)
  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    One more remark: the third branch, "Extract Macro with param() function", probably does not work in RapidMiner 5.2.8, but only in the current development version, which will be released as 5.2.9.
Sign In or Register to comment.