When dealing with naive datetime objects in Python which belong to the datetime.datetime class, it's essential to remember that they represent local time. While this might seem straightforward, it can still be puzzling at times. Interestingly, calling the timestamp method on a naive datetime object returns a POSIX timestamp related to UTC (seconds since the epoch).
The behavior of a naive pandas.Timestamp may not be as intuitive when compared to the Python datetime object. If derived from a tz-naive string in a similar manner, a pandas.Timestamp will actually represent UTC rather than local time. This distinction becomes evident when you localize the datetime object to UTC:
from datetime import datetime, timezone
import pandas as pd
date = "2020-02-07T16:05:16.000000000"
t1 = datetime.strptime(date[:-3], '%Y-%m-%dT%H:%M:%S.%f')
t2 = pd.to_datetime(date)
print(t1.replace(tzinfo=timezone.utc).timestamp() - t2.timestamp())
# 0.0
Conversely, you have the option to make a pandas.Timestamp timezone-aware, like so:
t3 = pd.to_datetime(t1.astimezone())
# e.g. Timestamp('2020-02-07 16:05:16+0100', tz='Mitteleuropäische Zeit')
# confirming that both t1 and t3 now represent my local time:
print(t1.timestamp() - t3.timestamp())
# 0.0
In conclusion, if you are certain about the time zones represented by your timestamps, it is advisable to work with timezone-aware datetime objects. For instance, for handling timestamps in UTC:
import pytz # using pytz since pandas relies on it internally
t1 = datetime.strptime(date[:-3], '%Y-%m-%dT%H:%M:%S.%f').replace(tzinfo=pytz.UTC)
t2 = pd.to_datetime(date, utc=True)
print(t1 == t2)
# True
print(t1-t2)
# 0 days 00:00:00
print(t1.timestamp()-t2.timestamp())
# 0.0