Generate interpretation: 'Attributes do not match' error
anaRodrigues
MemberPosts:33
Contributor II
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.
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.
0
Answers
Dortmund, Germany
Yes, the attributes have the same type (both are Real).
Thanks,
Ana