count
The count very allows you quick count unique values of one or more variables.
df %>% add_count(size)
df.groupby("size").count()