Pandas-dev

Download

pandas-dev@python.org

April 2015

6 participants
2 discussions

Upcoming Index repr changes
by Joris Van den Bossche May 22, 2015

May 22, 2015

Hi all, We have a PR pending to unify the string representation of the different Index objects: https://github.com/pydata/pandas/pull/9901 What are the most important changes: - We propose to reduce the default number of values shown from 100 to 10 (an option controllable as pd.options.display.max_seq_items). - The datetime-like indices (DatetimeIndex, TimedeltaIndex, PeriodIndex) were always somewhat different and get a new repr that is now more consistent with how it is for other Index types like Int64Index. This is the biggest change. So for eg Int64Index not much changes (only 'name' is now also shown, and the number of shown values has changed), but for DatetimeIndex the change is larger. *But we would like to get some feedback on this!* Do you like the changes? For DatetimeIndex? For the number of shown values? Would you want different behaviour for repr() and str()? Some examples of the changes with the current state of the PR are shown below: Previous Behavior In [1]: pd.get_option('max_seq_items') Out[1]: 100 In [2]: pd.Index(range(4), name='foo') Out[2]: Int64Index([0, 1, 2, 3], dtype='int64') In [3]: pd.Index(range(104), name='foo') Out[3]: Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, ...], dtype='int64') In [4]: pd.date_range('20130101', periods=4, name='foo', tz='US/Eastern') Out[4]: <class 'pandas.tseries.index.DatetimeIndex'> [2013-01-01 00:00:00-05:00, ..., 2013-01-04 00:00:00-05:00] Length: 4, Freq: D, Timezone: US/Eastern In [5]: pd.date_range('20130101', periods=104, name='foo', tz='US/Eastern') Out[5]: <class 'pandas.tseries.index.DatetimeIndex'> [2013-01-01 00:00:00-05:00, ..., 2013-04-14 00:00:00-04:00] Length: 104, Freq: D, Timezone: US/Eastern New Behavior In [1]: pd.get_option('max_seq_items') Out[1]: 10 In [9]: pd.Index(range(4), name='foo') Out[9]: Int64Index([0, 1, 2, 3], dtype='int64', name=u'foo') In [10]: pd.Index(range(104), name='foo') Out[10]: Int64Index([0, 1, ..., 102, 103], dtype='int64', name=u'foo', length=104) In [11]: pd.date_range('20130101', periods=4, name='foo', tz='US/Eastern') Out[11]: DatetimeIndex(['2013-01-01 00:00:00-05:00', '2013-01-02 00:00:00-05:00', '2013-01-03 00:00:00-05:00', '2013-01-04 00:00:00-05:00'], dtype='datetime64[ns]', name=u'foo', freq='D', tz='US/Eastern') In [12]: pd.date_range('20130101', periods=104 ,name='foo', tz='US/Eastern') Out[12]: DatetimeIndex(['2013-01-01 00:00:00-05:00', '2013-01-02 00:00:00-05:00', ..., '2013-04-13 00:00:00-04:00', '2013-04-14 00:00:00-04:00'], dtype='datetime64[ns]', name=u'foo', length=104, freq='D', tz='US/Eastern')

4 5

sinhrks (Masaaki Horikoshi)
by Jeff Reback April 18, 2015

April 18, 2015

All, @sinhrks (Masaaki Horikoshi) has accepted the invitation to be a *pandas core dev!* Thanks for all of the effort you have put in. Let keep it up to get even more people into pandas contributions. This project is quite popular, from just seeing the reaction at PyCon and is used in a lot of different fields / ways. Jeff

3 2