Convert datetime64 to seconds

Question

I subtracted two datetime64 columns and got a column of results such as '00:20:32' and '00:08:21.' How do I then convert those datetime64 values to seconds?

Please add more information regarding the question. A minimum reprodiclbe example and and expected output would be nice! — Celius Stingher, Nov 10 '19 at 21:02
Duplicate? https://stackoverflow.com/questions/14920903/time-difference-in-seconds-from-numpy-timedelta64 — Hallgeir Wilhelmsen, Nov 10 '19 at 21:05
Possible duplicate of [Time difference in seconds from numpy.timedelta64](https://stackoverflow.com/questions/14920903/time-difference-in-seconds-from-numpy-timedelta64) ... another, maybe better: [Convert timedelta64 column to seconds in Python Pandas DataFrame](https://stackoverflow.com/questions/26456825/convert-timedelta64ns-column-to-seconds-in-python-pandas-dataframe) — wwii, Nov 10 '19 at 21:12

score 2 · Answer 1 · edited Nov 11 '19 at 01:05

2

Can you paste the code to make to question more precise ?

Generally you can do that using NumPy timedelta64:

np.timedelta64( timedate1 - timedate2 ,'s')

edited Nov 11 '19 at 01:05

Beniamin H

2,048
1
11
17

answered Nov 10 '19 at 21:08

dexter103

58
3

score 1 · Accepted Answer · answered Nov 10 '19 at 21:05

Off the top of my head (and with the poor information in the question), if there's a column that contains hours, minutes and seconds, which I will call "time" I would create the following column in the dataframe:

df['full_seconds'] = df['time'].dt.seconds

Or use a simple calculus summing hours, minutes and seconds:

df['full_seconds'] = df['time'].dt.hour * 3600 + df['time'].dt.minute * 60 + df['time'].dt.second

Convert datetime64 to seconds

2 Answers2