Selecting rows in a pandas dataframe with two unique identifiers and storing these as new dataframes

Question

I have an extremely large, unsorted pandas dataframe (over two million rows) with multiple columns, two columns of which identify which category these rows belong to. Where