image

🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

Generate interpretation: 'Attributes do not match' error

anaRodriguesanaRodrigues Member Posts: 33 Contributor II
edited July 27 in Help
Hi,

I'm trying to generate an interpretation for a few models i have stored in the repository, but the operator always gives this error.



I've tried a million things and can't figure out where the error comes from. As you can see the attribute is present in the example set:



But it's not present in the attributes the model is using:

Shouldn't the operator automatically select the relevant attributes just like the 'apply model' operator? Do I have to "manually" select the attributes in the example set before feeding it to the generate interpretation operator? That's fine in this case, but what happens when I have a random forest model, for instance?

Thank you,
Ana

**EDIT**
So apparently it works if I manually select only the attributes the model needs. But still, this solution is impossible for a random forest model or gradient boost.

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,053  RM Data Scientist
    Hi,
    can you make sure, that the attribute also has the same type as in training? Some models complain about the difference of Integers and Reals.

    Best,
    Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
  • anaRodriguesanaRodrigues Member Posts: 33 Contributor II
    Hi Martin,

    Yes, the attributes have the same type (both are Real).

    Thanks,
    Ana
Sign In or Register to comment.