Pandas to_csv - Not save index values repeatedly?

Question

Pandas 0.20.3 - Excuse the poor title. I don't know the best way to describe it.

DataFrame:

In [363]: df
Out[363]: 
           IP   Jroid   ST
0   127.0.0.3  Joid12  stq
1   127.0.0.2  Jroid2  stt
2   127.0.0.1  Jroid1  sth
3   127.0.0.1  Jroid1  stl
4   127.0.0.1  Jroid3  stj
5   127.0.0.1  Jroid2   ll
6   127.0.0.2  Jroid1  stq
7   127.0.0.1  Jroid3  stt
8   127.0.0.1  Jroid1  sth
9   127.0.0.2  Jroid1  stl
10  127.0.0.3  Jroid1  stq
11  127.0.0.3  Jroid3  stt

DataFrame grouped perfectly for my needs:

In [365]: df.groupby(['IP','Jroid','ST']).ST.agg('count')
Out[365]: 
IP         Jroid   ST 
127.0.0.1  Jroid1  sth    2
                   stl    1
           Jroid2  ll     1
           Jroid3  stj    1
                   stt    1
127.0.0.2  Jroid1  stl    1
                   stq    1
           Jroid2  stt    1
127.0.0.3  Joid12  stq    1
           Jroid1  stq    1
           Jroid3  stt    1
Name: ST, dtype: int64

Now

df.groupby(['IP','Jroid','ST']).ST.agg('count').to_csv('/tmp/hi.csv')

renders as such:

127.0.0.1,Jroid1,sth,2
127.0.0.1,Jroid1,stl,1
127.0.0.1,Jroid2,ll,1
127.0.0.1,Jroid3,stj,1
127.0.0.1,Jroid3,stt,1
127.0.0.2,Jroid1,stl,1
127.0.0.2,Jroid1,stq,1
127.0.0.2,Jroid2,stt,1
127.0.0.3,Joid12,stq,1
127.0.0.3,Jroid1,stq,1
127.0.0.3,Jroid3,stt,1

How can I can have "to_csv" save the exact way it is represented by "df.groupby(['IP','Jroid','ST']).ST.agg('count')"? Is it possible?

Creating a new df doesn't seem to help:

print_df = df.groupby(['IP','Jroid','ST']).ST.agg('count')
print_df.to_csv('/tmp/hi.csv')

Does the same thing though.

Answer 1

As DeepSpace stated it won't be a valid CSV format

You can do it within .xlsx

df.to_excel('df.xlsx', engine='xlsxwriter')