在 Python/pandas 中格式化每月日期

Question

I want to modify the Monthly_idxs so that it outputs the monthly data ranges starting from the beginning minute of the month -01 00:00:00+00:00 instead of the current output. I want to also include the month of the initial index of which is October but the output starts the initial Monthly_idxs from November. How would I be able to get the Expected Output below?

import pandas as pd 

# Creates 1 minute data range between date_range(a, b)
l = (pd.DataFrame(columns=['NULL'],
                  index=pd.date_range('2015-10-08T13:40:00Z', '2016-01-04T21:00:00Z',
                                      freq='1T'))
       .index.strftime('%Y-%m-%dT%H:%M:%SZ')
       .tolist()
)

#Month Indexes
Monthly_idxs = pd.date_range(l[0], l[-1], freq='MS')

Output:

['2015-11-01 13:40:00+00:00', '2015-12-01 13:40:00+00:00',
               '2016-01-01 13:40:00+00:00']

Expected Output:

['2015-10-01 00:00:00+00:00', '2015-11-01 00:00:00+00:00','2015-12-01 00:00:00+00:00'
               '2016-01-01 00:00:00+00:00']

Answer 1

We can write Monthly_idxs using round and DateOffset to get the expected result:

from pandas.tseries.offsets import DateOffset

Monthly_idxs = pd.date_range(pd.Timestamp(min(l)).round('1d') - DateOffset(months=1), pd.Timestamp(max(l)).round('1d'), freq='MS').strftime("%Y-%m-%d %H:%M:%S%z").tolist()

Output:

['2015-10-01 00:00:00+0000',
 '2015-11-01 00:00:00+0000',
 '2015-12-01 00:00:00+0000',
 '2016-01-01 00:00:00+0000']

Thanks to @MrFuppes for the DateOffset idea.

Answer 2

Your list conversion occurs too soon. You can use resample on your dataframe and then use format to get the string list of your resampled index:

df = pd.DataFrame(columns=['NULL'],
                  index=pd.date_range('2015-10-08T13:40:00Z', '2016-01-04T21:00:00Z',
                                      freq='1T'))

Month_begin = df.resample('MS').asfreq()
Monthly_idxs = Month_begin.index.format()
print(Monthly_idxs)

Output:

['2015-10-01 00:00:00+00:00', '2015-11-01 00:00:00+00:00', '2015-12-01 00:00:00+00:00', '2016-01-01 00:00:00+00:00']