Replace each element in pandas Series by how often it previously occured

Question

Assume there is a pandas Series containing values reflecting categories ('a','b','c' or 1,2,3):

pds = I.pd.Series(['a','a','b','c','c','c','a'])

I would like to generate a new Series, which indicates how often each element has already occured, i.e. the expected output would be:

pds_result = I.pd.Series([0,1,0,0,1,2,2])
                     #    ^ no 'a' prior to this position in pds
                     #      ^ one 'a' prior to this position in pds
                     #        ^ no 'b' prior to this position in pds
                     #                ^ two 'a' prior to this position in pds

How can this be achieved in a concise manner?