pandas dropping rows based on datetime condition

Question

I am dropping rows based on a datetime condition, I have it working with the following line

df.drop(df[df.index.date == datetime(2017,9,14).date()].index, inplace=True)

However when I actually run the code I am not passing a datetime(2017,9,14).date() for comparison I am passing a datetime.date(2017,9,14) . So the code would look something like this...

df.drop(df[df.index.date == datetime.date(2017,9,14)].index, inplace=True)

but that obviously throws a error :

"descriptor 'date' requires a 'datetime.datetime' object but received a 'int'"

what would be the best way to fix this problem to be able to compare dates.

jezrael · Accepted Answer · 2019-10-16 05:41:18Z

2

You can simplify code - select all rows if not datetime.date(2017,9,14) - so it remove rows with datetime.date(2017,9,14):

rng = pd.date_range('2017-09-13', periods=10, freq='10H')
df = pd.DataFrame({'a': range(10)}, index=rng)  
print (df)
                     a
2017-09-13 00:00:00  0
2017-09-13 10:00:00  1
2017-09-13 20:00:00  2
2017-09-14 06:00:00  3
2017-09-14 16:00:00  4
2017-09-15 02:00:00  5
2017-09-15 12:00:00  6
2017-09-15 22:00:00  7
2017-09-16 08:00:00  8
2017-09-16 18:00:00  9

import datetime
df1 = df[df.index.date != datetime.date(2017,9,14)]
print (df1)
                     a
2017-09-13 00:00:00  0
2017-09-13 10:00:00  1
2017-09-13 20:00:00  2
2017-09-15 02:00:00  5
2017-09-15 12:00:00  6
2017-09-15 22:00:00  7
2017-09-16 08:00:00  8
2017-09-16 18:00:00  9

edited Oct 16, 2019 at 5:41

answered Oct 16, 2019 at 5:31

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

4 Comments

blonc Over a year ago

throws the same error : descriptor 'date' requires a 'datetime.datetime' object but received a 'int'. When looking farther into it df.index.date is a NP array and datetime.date is a object. I think this is the issue?

jezrael Over a year ago

@blonc - What is print (df.index.dtype) ?

blonc Over a year ago

it is a datetime64[ns]

blonc Over a year ago

appreciate the link to other question. thanks for help, problem solved.

Fra · Accepted Answer · 2019-10-16 05:47:11Z

0

The error is thrown by datetime.date(2017,9,14), it has nothing to do with Pandas.

Try:

from datetime import datetime
datetime.date(2017,9,14)

Throws:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-2-8658ce936e92> in <module>
----> 1 datetime.date(2017,9,14)

TypeError: descriptor 'date' requires a 'datetime.datetime' object but received a 'int'

but:

import datetime
datetime.date(2017,9,14)

works and it also works in your code.

answered Oct 16, 2019 at 5:47

Fra

5,2189 gold badges36 silver badges51 bronze badges

Collectives™ on Stack Overflow

pandas dropping rows based on datetime condition

2 Answers 2

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related