Options

Logarithmically scaled steps in Create Learning Curve

yzanyzan Member Posts: 66 Unicorn
edited June 2019 in Help
Currently "Create Learning Curve" operator supports only linear steps. But commonly we are more interested into details at the lower count of samples than at the higher count. Also calculations are fast with low sample count. Hence wouldn't it be nice if "Create Learning Curve" operator supported logarithmically scaled steps?

PS: I know I can get the logarithmic scale with "Loop Parameters" supplemented with other operators. But what do I like on "Create Learning Curve" is that it is hard to use it wrongly.

Answers

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,525 RM Data Scientist
    Hi,

    you are completly right regarding that operator. Personally i need to say i don't use it regulary. Usually i use Optimize Paramters, Log and a sample operator. That way i can also create different learning curves at a time by changing parameters of the learner etc.

    Might be worth a building block.

    Cheers,

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