Calculating Python Pandas group by percentage

Question

I want to find out the percentage of occurrence particular group(group by 2 columns) with the no of occurrence of group by of single column. Like in below i want to calculate the occurence of group by(Cardno,Sitedesc) divided by the group by(Cardno). Thanks in advance

Data

         Cardno  Sitedesc
0        883484        30
1       1017011        59
2        304531        48
3        304531        59
4        304531        32
5        687253        46
6        351841        59
7        976365        58
8        983485        46
9        326465        30
10       326465        40
11       983485        58
12       983485        46
13       983485        48
14       847535        31
15       687250        47
16       687250        46

Sorry, I dont understand why `50`. Is possible use only 5-6 rows data [minimal, complete, and verifiable example](http://stackoverflow.com/help/mcve) with expected output? — jezrael, Jan 28 '19 at 12:41
output:- P(687250)=50 As the 687250 is occurring twice and both has the different Sitedesc. so for same Cardno we have 2 different site desc then probability of occurrence of particular Sitedesc for a particular card no will be 50 . That's why probability is 50 — Amarjeet Singh, Jan 28 '19 at 12:41
Yes I changed the data. Now we have different data for 326465 ,304531,687250 cardno — Amarjeet Singh, Jan 28 '19 at 12:44
Basically, its the count of groupby(Cardno,Sitedesc) divided by the count of groupby(Cardno). — Amarjeet Singh, Jan 28 '19 at 12:47
Possible duplicate of [Pandas percentage of total with groupby](https://stackoverflow.com/questions/23377108/pandas-percentage-of-total-with-groupby) — Venkatesh Garnepudi, Jan 28 '19 at 12:51

score 0 · Accepted Answer · answered Jan 28 '19 at 12:49

Use:

s = df.groupby(['Cardno','Sitedesc']).size()
out = s.div(s.sum(level=0), level=0)
print (out)

Cardno   Sitedesc
304531   48          1.0
326465   30          1.0
351841   59          1.0
395354   59          1.0
687253   46          1.0
687463   59          1.0
847535   31          1.0
883484   30          1.0
976365   58          1.0
983485   46          1.0
1017011  59          1.0
1038308  58          1.0
dtype: float64

Calculating Python Pandas group by percentage

1 Answers1