Python Forum

Hi guys,
I have one table with a lot of columns: Two of them are Start Station and End Station

Start Station
0 Columbus Dr & Randolph St
1 Kingsbury St & Erie St
2 Canal St & Madison St
3 Spaulding Ave & Armitage Ave
4 Clark St & Randolph St


End Station \
Federal St & Polk St
Orleans St & Merchandise Mart Plaza
Paulina Ave & North Ave
California Ave & Milwaukee Ave
Financial Pl & Congress Pkwy

Now I want to display the most frequent combination of start station and end station trip

This is my code:

import pandas as pd
df = pd.read_csv('chicago.csv')
# I exctract the column with start stations:
x= df.iloc[:, 3]
# I exctract the column with end stations:
y =df.iloc[:, 4]
# Put this two columns together and 
df2 = (x+' & '+y)
#display the most frequent combination of start station and end station trip
df1 = df.groupby(df2).count().sorted(df2)
print (df1)

However, the code doesn't work.
This is what I get

 'DataFrame' object has no attribute 'sorted'

Actually I excpected smth like this:
Columbus Dr & Randolph St & Federal St & Polk St 8

How can I fix it?
Thanks
J

There is no reason to use Pandas on a simple task like this. You can just read the csv file normally.

with open('chicago.csv', 'r') as fp:
    for rec in fp:
        split_rec=rec.strip().split(",")
        ## you now have a list of columns and can use a 
        ## dictionary to count the 2 combined columns
        print(split_rec)

        ## ****** I am not going to do your homework for you *****