Python: Indexing in Pandas with Datetime indices

Question

I've got problems regarding to indexing in Pandas and hope you can help me:

rng = pd.date_range('2015-12-31 21:00:00', periods=7, freq='H')
df = pd.DataFrame({ 'Val' : np.random.randn(len(rng)) }, index=rng) 
first_value_of_year = df['2016'].first('1H').index

returns the first value of the year as DatetimeIndex:

DatetimeIndex(['2016-01-01'], dtype='datetime64[ns]', freq='H')

When I call the dataframe with this index, everything seems to be working fine:

df.loc[first_value_of_year]

returns

+------------------------+-----------+
|                        |  Val      |
+------------------------+-----------+
| 2016-01-01             | 0.144044  |

So, everything is OK up to here! But if I want to get all values greater than this value, it doesn't work anymore:

df.loc[df.index >= first_value_of_year]

and gives ValueError (lenghts must match):

but if I take the same command with the timestamp itself as string it does what it should do

df.loc[df.index >= '2016-01-01 00:00:00']

returns

+------------------------+-----------+
|                        |  Val      |
+------------------------+-----------+
| 2016-01-01 01:00:00    | 1.454274  |
| 2016-01-01 02:00:00    | 0.761038  |
| 2016-01-01 03:00:00    | 0.121675  |

so, what am I missing here? How do I correctly access all values greater than the DatetimeIndex variable?

Thanks!

I could not understand if it works the way you want or not? Also, printing the type of variables inside the square brackets may give you a hint, e.g. print(type(df.index >= first_value_of_year)) — Erdal Dogan
– Erdal Dogan, Commented Aug 29, 2019 at 10:34

jezrael · Accepted Answer · 2019-08-29 10:36:26Z

3

I believe you need select first value of index to scalar by indexing - [0]:

df.loc[df.index >= first_value_of_year[0]]

answered Aug 29, 2019 at 10:36

jezrael

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

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

1 Comment

Exi Over a year ago

thank you so much... have spent over 2 hours on this :-)

Collectives™ on Stack Overflow

Python: Indexing in Pandas with Datetime indices

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related