You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Now i am using LightGBM via treesnip and got an error about categorical column parameter. I found that the package figure out categrocal columns itself using categorical_columns() but I think this does not match with the LGBM options.
prepare_df_lgbm <- function(x, y = NULL) {
categorical_cols <- categorical_columns(x)
x <- categorical_features_to_int(x, categorical_cols)
x <- as.matrix(x)
return(x)
}
Can we change the package allows to pick what is the categorical columns? This is needed because usually we have all numerical variables after the recipe in tidymodels (for example step_integer() in recipe), so I guess this categorical_columns() never run in this tidymodels frame work.
The text was updated successfully, but these errors were encountered:
Hi, thank you for making this awesome package!
Now i am using LightGBM via treesnip and got an error about categorical column parameter. I found that the package figure out categrocal columns itself using
categorical_columns()
but I think this does not match with the LGBM options.Can we change the package allows to pick what is the categorical columns? This is needed because usually we have all numerical variables after the recipe in tidymodels (for example step_integer() in recipe), so I guess this
categorical_columns()
never run in this tidymodels frame work.The text was updated successfully, but these errors were encountered: