Sign up to take part
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
Im noticing that the resulting coefficients for a OHE feature fit with a ridge regression does not have the real name of the category assigned to it. Instead it looks to me that it indexed the row that the category was in (they are distinct), but Im not at all sure about that. My question is why dd it not just give it the name of the category? Can I get that back somehow?
Maybe I can go into the preprocessing pipeline and use the OHE object and work it out that way?
There is one thing that's a little odd about this implementation that I should note. I'm making a custom preprocessing to be a sparse matrix.
Operating system used: Ubuntu
โ
I ended up grabbing the processor from the preprocessing pipeline and using the inverse_transform method. Problem solved
I ended up grabbing the processor from the preprocessing pipeline and using the inverse_transform method. Problem solved