I sometimes use pandas to analyze access logs and utilize binning feature, but usually I forget how to make range objects with pd.*_range methods. So I memorize popular methods to me.

I usually use date_range (bdate_range).

In [28]: pd.date_range('2017-08-23', periods=3)
Out[28]: DatetimeIndex(['2017-08-23', '2017-08-24', '2017-08-25'], dtype='datetime64[ns]', freq='D')

In [29]: pd.date_range('2017-08-23', periods=3, freq=4*pd.offsets.BDay())
Out[29]: DatetimeIndex(['2017-08-23', '2017-08-29', '2017-09-04'], dtype='datetime64[ns]', freq='4B')

Here is a reference for frequency.

• DateOffset objects
• Anchored Offsets

DateOffset is useful.

In [38]: pd.to_datetime('2017-01-01 03:45:05 +09:00') + pd.offsets.DateOffset(days=5, seconds=1)
Out[38]: Timestamp('2017-01-05 18:45:06')