python - Datetime milliseconds to seconds in Pandas

Question

Have a datetime column in pandas dataframe with values like these:

time
2018-04-11 22:18:30.122
2018-04-11 23:00:21.399

I'm wondering how can I round these values, get rid of milliseconds and represent only date, hour, minutes and 00 as seconds like this:

time
2018-04-11 22:18:00
2018-04-11 23:00:00

Answer 1

Use floor with T for minutes for set 0 seconds:

#if necessary
#df['time'] = pd.to_datetime(df['time'])
df['time'] = df['time'].dt.floor('T')
#alternative solution
#df['time'] = df['time'].dt.floor('Min')

print (df)
                 time
0 2018-04-11 22:18:00
1 2018-04-11 23:00:00

I want round values time after 30sec is changed to next one:

df['time'] = df['time'].dt.round('T')
print (df)
                 time
0 2018-04-11 22:19:00
1 2018-04-11 23:00:00