I'm using reticulate to convert a python saved .pkl into an R dataframe. I see the categorical python type is kept in the R dataframe. I'm guessing R doesn't know how to deal with python categorical variables. How do I convert those into factor variables in R?
#Looking at the class of some of the variables in the R dataframe, I get
class(df$variable)
$variable
[1] "pandas.core.arrays.categorical.Categorical"
"pandas.core.arrays.base.ExtensionArray"
[3] "pandas.core.base.PandasObject"
"pandas.core.base.StringMixin"
[5] "pandas.core.accessor.DirNamesMixin" "python.builtin.object"
#I get the right levels, but this doesn't seem to be R-friendly
> table(levels(factor(as.character(df$varibale))))
[level 1, level 2, level 3, level 4, level 5, ..., NaN, NaN, NaN, NaN,
NaN]\nLength: 464121\nCategories (6, object): [level 1 < level 2 < level 3
< level 4 < level 5 < level 6]