python exercise [03_1_encoding]: mlr3 xgboost support for categorical data? #39
Description
Maybe also update the R version of the exercise, if mlr3 xgboost can handle categorical data.
The issue should maybe also be opened in MLR3? But besides that, the text states one hot encoding or correct categorical splits. But shouldn't we try to show what is possible in the language, in this case Python can do more, in other cases such as parallelization, R has more features?
Originally posted by @mfeurer in #26 (comment)