pandas dataframe column value to unix-like timestamp value

Asked Apr 26 '17 at 08:42

Active Apr 26 '17 at 08:44

Viewed 98 times

my question is simple. How do I convert pandas column value to unix-like timestamp?

data['time'] = pd.to_datetime(data['time']) 
print(data['time']) 

# output 
0   2016-05-04 18:38:32.076
1   2016-05-04 18:39:15.556
Name: time, dtype: datetime64[ns]  


my desired result: 
0    1238475858
1    1238475859

edited Apr 26 '17 at 08:44

Serenity

32,301
18
107
109

asked Apr 26 '17 at 08:42

Chris Joo

1

possible dupe: https://stackoverflow.com/questions/19553464/python-pandas-series-of-datetimes-to-seconds-since-the-epoch – EdChum Apr 26 '17 at 08:43
1

Do you think `df['time'] = df['time'].values.astype(np.int64)` ? – jezrael Apr 26 '17 at 08:44
Small correction to @jezrael's comment: `df['unix_ts'] = df['time'].astype(np.int64)//10**9` – MaxU - stop genocide of UA Apr 26 '17 at 08:58
@MaxU - Yes, question is if need `ns` or `s` ;) – jezrael Apr 26 '17 at 08:59
2

@jezrael, [number of __seconds__ that have elapsed since 00:00:00 Coordinated Universal Time (UTC), Thursday, 1 January 1970](https://en.wikipedia.org/wiki/Unix_time) ;-) – MaxU - stop genocide of UA Apr 26 '17 at 09:12
1

ok, then you are right. – jezrael Apr 26 '17 at 09:14
You can you 'dt' attribute to access datetime methods, such as timestamp(). So to answer your question is should be doable with data['time'].dt.timestamp() – Victor G. Apr 26 '17 at 09:51

pandas dataframe column value to unix-like timestamp value

0 Answers0