Unable to do calculations with categorical dtypes

[ahmedsoe]

df = pd.DataFrame({'a':[1,2,3,2,1,3,4,5,6,7],
                   'b':[1,2,3,2,1,3,4,5,6,7]})
df['a'] = df['a'].astype('category')
df['a'] * df['b']

Problem description

On pandas v0.20.3, this is possible:

Expected Output

Out[23]: 
0     1
1     4
2     9
3     4
4     1
5     9
6    16
7    25
8    36
9    49
dtype: int64

On v0.23.0, the output is:

TypeError: Series cannot perform the operation *

[jreback]

this is a deliberate choice as numeric operations on categories do not make sense. if you are using categorizals as a memory savings device then wait for #20899