How to do a conditional count after groupby on a Pandas Dataframe?

Question

I have the following dataframe:

   key1  key2
0    a   one
1    a   two
2    b   one
3    b   two
4    a   one
5    c