Alternatives to group_map

Question

I am curious what would be the base R code and data.table code that do the same work as group_map in dplyr?

For example, how to