force format when parsing dates in pandas from an Excel file

Question

I am trying to parse dates from an Excel file following a given format %d/%m/%y . I am not able to set the format, I am always getting %m/%d/%Y .

The source format is %d/%m/%y so that the first date should be first of June 2016. Any ideas on how to do it?

import pandas as pd

url = 'https://www.dropbox.com/s/8gqmq3jx27unsta/example_dates.xlsx?dl=1'

file = pd.ExcelFile(url, parse_date=True, 
       date_parser = (lambda x: pd.to_datetime(x, format ='%d/%m/%y')))

df = file.parse(0)

df
    date    variable
0   2016-01-06  1
1   2016-06-07  2
2   2016-12-10  3
3   2016-12-29  4

Something like this might work but it doesn't:

df = file.parse(0, converters={'date' : lambda x: pd.to_datetime(x, dayfirst=True)}

date    variable
0   2016-01-06  1
1   2016-06-07  2
2   2016-12-10  3
3   2016-12-29  4

Answer 1

Actually pandas have a format to display datetime object.

So it will display in that format till you change that

meanwhile you can do this:

df['date1'] = df['date'].dt.strftime('%d/%m/%y')

output:

        date    variable    date1
0   2016-01-06  1   06/01/16
1   2016-06-07  2   07/06/16
2   2016-12-10  3   10/12/16
3   2016-12-29  4   29/12/16

Answer 2

You can use converters :

>>> file.parse(0, parse_dates=True, index_col=0, date_parser=lambda x: pd.to_datetime(x).strftime("%d/%m/%Y"))

     variable
date    
2016-06-01  1
2016-07-06  2
2016-10-12  3
2016-12-29  4

Answer 3

You can use pandas.read_excel in combination with datetime.datetime.strptime .

import pandas as pd
from datetime import datetime

URL = 'https://www.dropbox.com/s/8gqmq3jx27unsta/example_dates.xlsx?dl=1'

file = (
    pd
    .read_excel(
        io=URL, 
        parse_date=True,
        date_parser=lambda x: datetime.strptime(x, '%d/%m/%Y'),
    )
)