How to change format of data to '%Y%m%d' in Pandas?

Question

I have a DF with first column showing as e.g. 2018-01-31 00:00:00.

I want to convert whole column (or during printing / saving to other variable) that date to 20180131 format. NOT looking to do that during saving to a CSV file.

Tried this but it did not work:

df['mydate'] = pd.to_datetime(df['mydate'], format='%Y%m%d')

Solution 1

pd.to_datetime is used to convert your series to datetime:

s = pd.Series(['2018-01-31 00:00:00'])
s = pd.to_datetime(s)

print(s)

0   2018-01-31
dtype: datetime64[ns]

pd.Series.dt.strftime converts your datetime series to a string in your desired format:

s = s.dt.strftime('%Y%m%d')

print(s)

0    20180131
dtype: object

Solution 2

pd.to_datetime will convert a string to a date. You want to covert a date to a string

df['mydate'].dt.strftime('%Y%m%d')

Note that it's possible your date is already a string, but in the wrong format in which case you might have to convert it to a date first:

pd.to_datetime(df['mydate'], format='%Y-%m-%d %H:%M:%S').dt.strftime('%Y%m%d')

Comments

[Joe]

What is .dt? What libraries you are importing?

[Dan]

No libraries, those are date functions built into pandas. Follow the link in the post.

[Joe]

This did not work for me: AttributeError: 'Timestamp' object has no attribute 'dt'

[Dan]

what is df['mydate'].dtype?

[Dan]

@Joe Then pd.to_datetime(df['mydate'], format='%Y-%m-%d %H:%M:S').dt.strftime('%Y%m%d') will work. Certainly for your example date.