How to select index by datetime range with Pandas

Question

I have a Pandas datafame indexed by datetime and I would like to select indexes between 2 dates: start and end.

type(start)
<class 'pandas.core.indexes.datetimes.DatetimeIndex'>

type(end)
<class 'pandas.core.indexes.datetimes.DatetimeIndex'>

EDIT

Reproducible code:

from datetime import timedelta
idx = pd.DatetimeIndex(['2021-10-16 15:00:00', '2021-10-16 16:00:00',
           '2021-10-16 17:00:00', '2021-10-16 18:00:00',
           '2021-10-16 19:00:00', '2021-10-16 20:00:00',
           '2021-10-16 21:00:00', '2021-10-16 22:00:00',
           '2021-10-16 23:00:00', '2021-10-17 00:00:00',
           '2021-10-17 01:00:00', '2021-10-17 02:00:00',
           '2021-10-17 03:00:00', '2021-10-17 04:00:00',
           '2021-10-17 05:00:00', '2021-10-17 06:00:00',
           '2021-10-17 07:00:00', '2021-10-17 08:00:00',
           '2021-10-17 09:00:00', '2021-10-17 10:00:00',
           '2021-10-17 11:00:00', '2021-10-17 12:00:00',
           '2021-10-17 13:00:00', '2021-10-17 14:00:00',
           '2021-10-17 15:00:00', '2021-10-17 16:00:00'],
          dtype='datetime64[ns]', freq='H')
df = pd.DataFrame([range(25)], index=idx)
end = df.tail(1).index
start = end - timedelta(hours=5)
df.loc[(df.index >= start) & (df.index <= end)]

ValueError: Lengths must match

How can I do that ?

Does this answer your question? Select DataFrame rows between two dates — ajsp
– ajsp, Commented Oct 17, 2021 at 15:18
Unfortunatly not, because when I try prices.index > start it throws an error saying ValueError: Lengths must match — Florent
– Florent, Commented Oct 17, 2021 at 15:19
if your search date is a string, then you could convert the column to string. But for obvious reasons, it's quicker to change the searchable object into a datetime object. — ajsp
– ajsp, Commented Oct 17, 2021 at 15:27

Fergus McClean · Accepted Answer · 2021-10-17 15:33:39Z

2

Your start and end variables need to be datetimes, not datetime indexes.

Try end = df.index[-1]

answered Oct 17, 2021 at 15:33

Fergus McClean

1863 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

ajsp · Accepted Answer · 2021-10-17 15:35:35Z

2

You can use an ordinary query, but your dates need to be datetime objects.

dt1 = datetime.datetime(some date)
dt2 ...

rng = df[(df.index >= dt1 ) & (df.index <= dt2)]

This will give you the frame of the desired range of dates.

See Fergus answer, you were searching with an index and not a date.

edited Oct 17, 2021 at 15:35

answered Oct 17, 2021 at 15:14

ajsp

2,69025 silver badges35 bronze badges

3 Comments

Florent Over a year ago

When I try prices.index > start it throws an error saying ValueError: Lengths must match

ajsp Over a year ago

@Florent, you need to post both the date you are searching and the method, please post some reproducible code.

Florent Over a year ago

I eddited the question with reproducible code. Thanks

Collectives™ on Stack Overflow

How to select index by datetime range with Pandas

2 Answers 2

Comments

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related