Converting date and time string to appropriate format for arithmetic calculations

Question

I am using Python and I have dates in the table of the type '2013-06-24 17:04:28' stored as strings and I want to do things such as subtract and average them.

I dont care much about the time.

What is the appropriate format to convert the string to and how do I do it?

Thanks in advance for any help.

Max.

Possible duplicate of http://stackoverflow.com/questions/466345/converting-string-into-datetime — Lajos Arpad, Aug 06 '13 at 14:15
@LajosArpad it's not because it's pandas-specific, however it may be a dupe of another question :) — Andy Hayden, Aug 06 '13 at 14:46

Andy Hayden · Accepted Answer · 2013-08-06T14:24:54.543

4

pandas has it's own Timestamp object. You can convert strings, or entire columns (Series), to Timestamps using the to_datetime function:

pd.to_datetime(df['date_col'])

See "converting to Timestamps" section of the docs.

Note: you can convert to Timestamps when reading in your DataFrame e.g. using read_csv.

It works with your date format:

In [11]: pd.to_datetime('2013-06-24 17:04:28')
Out[11]: Timestamp('2013-06-24 17:04:28', tz=None)

edited Aug 06 '13 at 14:24

answered Aug 06 '13 at 14:18

Andy Hayden

359,921
101
625
535

So if I now 'df['date_col'].groupby(df['som_categorical_var']) - can I .mean() on the group object over dates? – user7289 Aug 06 '13 at 14:26
maybe `df['date_col'].groupby(['som_categorical_var', 'date_col']).mean()`, maybe? Ask as different question if it's not working, since it's rather different from this one – Andy Hayden Aug 06 '13 at 14:29
@AndyHayden How do you get to all the SO questions so *fast*? – Phillip Cloud Aug 06 '13 at 14:55
@AndyHayden Have you actually played that?? Or maybe you play that on a regular basis :) – Phillip Cloud Aug 06 '13 at 22:19

score 0 · Answer 2 · answered Aug 06 '13 at 14:16

0

time.strptime("2013-06-24 17:04:28", "%Y-%m-%d %H:%M:%S")

then you can use python's time module

answered Aug 06 '13 at 14:16

jongusmoe

676
4
16

Converting date and time string to appropriate format for arithmetic calculations

2 Answers2