"Sampling to an known distribution of labes"

pickmay
pickmay New Altair Community Member
edited November 5 in Community Q&A
Hi all.

I have a large data set, and I want to to produce a small sample in which there will be the same amount of examples from each label. is there any Rapid function that does that?

thanks
Yishai

Answers

  • TobiasMalbrecht
    TobiasMalbrecht New Altair Community Member
    Hi,

    that is already on our todo list, but unfortunately we have not managed to implement such a sample operator yet, since there is plenty of other things to do at the moment. One thing RM does already have is an operator to distribute weights among the examples giving every class the same sum of weights. The operator is called [tt]EqualLabelWeighting[/tt].

    Regards,
    Tobias