CODEWITHSUNDEEP
pythonpandas
Heisenberg
Heisenberg

Reputation: 5279

How to groupby in different keys

I have dflike following

A B C
a a d
a b d
a b e
b c e
b c f

When I try

df.groupby(A).size()

A
a 3
b 2

df.groupby(B).size()

B
a 1
b 2
c 2

my desired result is aggregated one

  A B
a 3 1
b 2 2
c 0 2

Are there any way to achieve this result ?

If someone has opinion,please let me know.

Thanks

Upvotes: 2

Views: 41

Answers (1)

BENY
BENY

Reputation: 323226

melt + crosstab

s = df[['A','B']].melt()
out = pd.crosstab(s['value'],s['variable'])
out
Out[18]: 
variable  A  B
value         
a         3  1
b         2  2
c         0  2

Or

df[['A','B']].apply(pd.Series.value_counts)
Out[19]: 
     A  B
a  3.0  1
b  2.0  2
c  NaN  2

Upvotes: 1

Related Questions