search

curso-r / treesnip

Parsnip backends for `tree`, `lightGBM` and `Catboost`
https://curso-r.github.io/treesnip
GNU General Public License v3.0
85 stars 13 forks source link

Error in catboost #59

Open vitaminasnow opened 2 years ago

vitaminasnow commented 2 years ago

I'm getting the following error when trying to tune a model

1 "preprocessor 1/1, model 1/5: Error in catboost::catboost.train(learn_pool = d, params = list(iterations = 500, : tools/enum_parser/enum_serialization_runtime/enum…
2 "preprocessor 1/1, model 2/5: Error in catboost::catboost.train(learn_pool = d, params = list(iterations = 500, : tools/enum_parser/enum_serialization_runtime/enum…
3 "preprocessor 1/1, model 3/5: Error in catboost::catboost.train(learn_pool = d, params = list(iterations = 500, : tools/enum_parser/enum_serialization_runtime/enum…
4 "preprocessor 1/1, model 4/5: Error in catboost::catboost.train(learn_pool = d, params = list(iterations = 500, : tools/enum_parser/enum_serialization_runtime/enum…
5 "preprocessor 1/1, model 5/5: Error in catboost::catboost.train(learn_pool = d, params = list(iterations = 500, : tools/enum_parser/enum_serialization_runtime/enum…

The same workflow works when using xgboost.

nipnipj commented 2 years ago

Using just set_engine("catboost") instead of set_engine("catboost", loss_function = "Logloss") seems to work.