133

I have the following:

> date1
Timestamp('2014-01-23 00:00:00', tz=None)

> date2
datetime.date(2014, 3, 26)

and I read on this answer that I could use pandas.to_datetime() to convert from Timestamps to datetime objects, but it doesn't seem to work:

> pd.to_datetime(date1)   
Timestamp('2014-01-23 00:00:00', tz=None)

Why? How can I convert between these two formats?

4 Answers 4

170

You can use the to_pydatetime method to be more explicit:

In [11]: ts = pd.Timestamp('2014-01-23 00:00:00', tz=None)

In [12]: ts.to_pydatetime()
Out[12]: datetime.datetime(2014, 1, 23, 0, 0)

It's also available on a DatetimeIndex:

In [13]: rng = pd.date_range('1/10/2011', periods=3, freq='D')

In [14]: rng.to_pydatetime()
Out[14]:
array([datetime.datetime(2011, 1, 10, 0, 0),
       datetime.datetime(2011, 1, 11, 0, 0),
       datetime.datetime(2011, 1, 12, 0, 0)], dtype=object)
Sign up to request clarification or add additional context in comments.

3 Comments

Thanks. Do you know that difference is between to_datetime() and to_pydatetime() and why passing a Timestamp object to the class method pd.to_datetime doesn't work?
@user815423426 it works on the global names function (pd.to_datetime(ts))... difference is that to_pydatetime returns a datetime :)
this answer doesn't seem to be be valid anymore.
39

Pandas Timestamp to datetime.datetime:

pd.Timestamp('2014-01-23 00:00:00', tz=None).to_pydatetime()

datetime.datetime to Timestamp

pd.Timestamp(datetime(2014, 1, 23))

Comments

12
>>> pd.Timestamp('2014-01-23 00:00:00', tz=None).to_datetime()
datetime.datetime(2014, 1, 23, 0, 0)
>>> pd.Timestamp(datetime.date(2014, 3, 26))
Timestamp('2014-03-26 00:00:00')

4 Comments

Thanks - I guess one should run to_datetime as an object method, but not as a class method, but why?
@user815423426 it seems contrary to its name pd.to_datetime always returns Timestamp
doesn't work on pandas 0.24. says 'Timestamp' object has no attribute 'to_datetime'
@Phoenix, the method you are looking for is to_pydatetime().
-1

To answer the question of going from an existing python datetime to a pandas Timestamp do the following:

    import time, calendar, pandas as pd
    from datetime import datetime
    
    def to_posix_ts(d: datetime, utc:bool=True) -> float:
        tt=d.timetuple()
        return (calendar.timegm(tt) if utc else time.mktime(tt)) + round(d.microsecond/1000000, 0)
    
    def pd_timestamp_from_datetime(d: datetime) -> pd.Timestamp:
        return pd.to_datetime(to_posix_ts(d), unit='s')
    
    dt = pd_timestamp_from_datetime(datetime.now())
    print('({}) {}'.format(type(dt), dt))

Output:

(<class 'pandas._libs.tslibs.timestamps.Timestamp'>) 2020-09-05 23:38:55

I was hoping for a more elegant way to do this but the to_posix_ts is already in my standard tool chain so I'm moving on.

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.