CODEWITHSUNDEEP
pythondatetimepandastime-series
HaggarTheHorrible
HaggarTheHorrible

Reputation: 7403

Pandas: How to extract rows which are just within a time duration?

I have a data-frame like this.

                       value  estimated  \
dttm_timezone                                                             
2011-12-31 20:10:00  10.7891          0   
2011-12-31 20:15:00  11.2060          0   
2011-12-31 20:20:00  19.9975          0   
2011-12-31 20:25:00  15.9975          0   
2011-12-31 20:30:00  10.9975          0   
2011-12-31 20:35:00  13.9975          0   
2011-12-31 20:40:00  15.9975          0   
2011-12-31 20:45:00  11.7891          0   
2011-12-31 20:50:00  10.9975          0   
2011-12-31 20:55:00  10.3933          0

By using the dttm_timezone column information, I would like to extract all the rows which are just within a day or a week or a month.

I have data of 1 year, so if I select day as the duration I should extract 365 days data separately, if I select month then I should extract a 12 months data separately.

How can I achieve this?

Upvotes: 1

Views: 324

Answers (1)

piRSquared
piRSquared

Reputation: 294218

Let's use 

import pandas as pd
import numpy as np

tidx = pd.date_range('2010-01-01', '2014-12-31', freq='H', name='dtime')

np.random.seed([3,1415])
df = pd.DataFrame(np.random.rand(len(tidx)), tidx, ['value'])

You can limit to '2010' like this:

df['2010']

Or

df[df.index.year == 2010]

You can limit to a specific month by:

df['2010-04']

or all Aprils:

df[df.index.month == 4]

You can limit to a specific day:

df['2010-04-28']

all 1:00 pm's:

df[df.index.hour == 13]

range of dates:

df['2011':'2013']

or

df['2011-01-01':'2013-06-30']

There is ton of ways to do this:

df.loc[(df.index.month == 11) & (df.index.hour == 22)]

link ---> The list can go on and on. Please read the docs <--- link

Upvotes: 3

Related Questions