Filter Pandas DataFrame by time index

Question

I have a pandas DataFrame from 6:36 AM to 5:31 PM. I want to remove all observations where the time is less than 8:00:00 AM. Here is my attempt:

df = df[df.index < '2013-10-16 08:00:00']

This does nothing, please help.

possible duplicate: stackoverflow.com/questions/16341367/… — Ryan Saxe
– Ryan Saxe, Commented Nov 27, 2013 at 3:05
that filters by column value, not by timeindex though. Thanks for the link — user2113095
– user2113095, Commented Nov 27, 2013 at 3:08
This isn't a duplicate the other post doesn't filter by datetimeindex — user2113095
– user2113095, Commented Nov 27, 2013 at 3:11

Kevin Stone · Accepted Answer · 2013-11-27 04:01:40Z

90

You want df.loc[df.index < '2013-10-16 08:00:00'] since you're selecting by label (index) and not by value.

answered Nov 27, 2013 at 4:01

Kevin Stone

9,01143 silver badges30 bronze badges

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

3 Comments

Oh I had no idea you could use .loc with boolean tests on the index! Great.

what if it has a multiindex?

don't have a usecase handy, but I'd suspect it's: df.loc[(df.index['a'] < something & df.index['b'] < another_thing)]

rachwa · Accepted Answer · 2022-05-31 19:09:59Z

1

You can use query for a more concise option:

df.query("index < '2013-10-16 08:00:00'")

answered May 31, 2022 at 19:09

rachwa

2,3901 gold badge21 silver badges20 bronze badges