Speed up pandas groupby.apply

Question

I have a function group_fn that needs to be applied to each group of a pandas dataframe:

df.groupby(group_col_name, sort=False).apply(group_fn)
<