Prohibit CatBoostPerSegmentModel and CatBoostMultiSegmentModel from using non-regressors
#587
d-a-bunin
started this conversation in
Improvements
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Description
Currently,
CatBoostPerSegmentModelandCatBoostMultiSegmentModeldoesn't filter-out non-regressor features. It could lead to using features that are unknown (contain NaNs) during forecasting. It differs from sklearn-models like e.g.LinearPerSegmentModelandLinearMultiSegmentModelwhich select only regressor features for training.Thoughts
DL models we also don't filter-out non-regressor features and there is inconsistency in handling. Probably the way of handling regressors should be reworked. E.g. transform that selects only regressor features could be created and applied at the end.
Related
No response
Beta Was this translation helpful? Give feedback.
All reactions