forked from pandas-dev/pandas
-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathv0.18.0.txt
501 lines (308 loc) · 16 KB
/
v0.18.0.txt
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
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
.. _whatsnew_0180:
v0.18.0 (January ??, 2016)
--------------------------
This is a major release from 0.17.1 and includes a small number of API changes, several new features,
enhancements, and performance improvements along with a large number of bug fixes. We recommend that all
users upgrade to this version.
.. warning::
pandas >= 0.18.0 will no longer support compatibility with Python version 2.6 (:issue:`7718`)
.. warning::
pandas >= 0.18.0 will no longer support compatibility with Python version 3.3 (:issue:`11273`)
Highlights include:
- Window functions are now methods on ``.groupby`` like objects, see :ref:`here <whatsnew_0180.enhancements.moments>`.
Check the :ref:`API Changes <whatsnew_0180.api>` and :ref:`deprecations <whatsnew_0180.deprecations>` before updating.
.. contents:: What's new in v0.18.0
:local:
:backlinks: none
.. _whatsnew_0180.enhancements:
New features
~~~~~~~~~~~~
.. _whatsnew_0180.enhancements.moments:
Window functions are now methods
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Window functions have been refactored to be methods on ``Series/DataFrame`` objects, rather than top-level functions, which are now deprecated. This allows these window-type functions, to have a similar API to that of ``.groupby``. See the full documentation :ref:`here <stats.moments>` (:issue:`11603`)
.. ipython:: python
np.random.seed(1234)
df = DataFrame({'A' : range(10), 'B' : np.random.randn(10)})
df
Previous Behavior:
.. code-block:: python
In [8]: pd.rolling_mean(df,window=3)
FutureWarning: pd.rolling_mean is deprecated for DataFrame and will be removed in a future version, replace with
DataFrame.rolling(window=3,center=False).mean()
Out[8]:
A B
0 NaN NaN
1 NaN NaN
2 1 0.237722
3 2 -0.023640
4 3 0.133155
5 4 -0.048693
6 5 0.342054
7 6 0.370076
8 7 0.079587
9 8 -0.954504
New Behavior:
.. ipython:: python
r = df.rolling(window=3)
These show a descriptive repr
.. ipython:: python
r
with tab-completion of available methods and properties.
.. code-block:: python
In [9]: r.
r.A r.agg r.apply r.count r.exclusions r.max r.median r.name r.skew r.sum
r.B r.aggregate r.corr r.cov r.kurt r.mean r.min r.quantile r.std r.var
The methods operate on the ``Rolling`` object itself
.. ipython:: python
r.mean()
They provide getitem accessors
.. ipython:: python
r['A'].mean()
And multiple aggregations
.. ipython:: python
r.agg({'A' : ['mean','std'],
'B' : ['mean','std']})
.. _whatsnew_0180.enhancements.other:
Other enhancements
^^^^^^^^^^^^^^^^^^
- Handle truncated floats in SAS xport files (:issue:`11713`)
- Added option to hide index in ``Series.to_string`` (:issue:`11729`)
- ``read_excel`` now supports s3 urls of the format ``s3://bucketname/filename`` (:issue:`11447`)
- A simple version of ``Panel.round()`` is now implemented (:issue:`11763`)
- For Python 3.x, ``round(DataFrame)``, ``round(Series)``, ``round(Panel)`` will work (:issue:`11763`)
- ``DataFrame`` has gained a ``_repr_latex_`` method in order to allow for automatic conversion to latex in a ipython/jupyter notebook using nbconvert. Options ``display.latex.escape`` and ``display.latex.longtable`` have been added to the configuration and are used automatically by the ``to_latex`` method. (:issue:`11778`)
- ``sys.getsizeof(obj)`` returns the memory usage of a pandas object, including the
values it contains (:issue:`11597`)
- ``Series`` gained an ``is_unique`` attribute (:issue:`11946`)
- ``DataFrame.quantile`` and ``Series.quantile`` now accept ``interpolation`` keyword (:issue:`10174`).
.. _whatsnew_0180.enhancements.rounding:
Datetimelike rounding
^^^^^^^^^^^^^^^^^^^^^
``DatetimeIndex``, ``Timestamp``, ``TimedeltaIndex``, ``Timedelta`` have gained the ``.round()`` method for datetimelike rounding. (:issue:`4314`)
Naive datetimes
.. ipython:: python
dr = pd.date_range('20130101 09:12:56.1234', periods=3)
dr
dr.round('s')
# Timestamp scalar
dr[0]
dr[0].round('10s')
Tz-aware are rounded in local times
.. ipython:: python
dr = dr.tz_localize('US/Eastern')
dr
dr.round('s')
Timedeltas
.. ipython:: python
t = timedelta_range('1 days 2 hr 13 min 45 us',periods=3,freq='d')
t
t.round('10min')
# Timedelta scalar
t[0]
t[0].round('2h')
In addition, ``.round()`` will be available thru the ``.dt`` accessor of ``Series``.
.. ipython:: python
s = Series(dr)
s
s.dt.round('D')
.. _whatsnew_0180.api:
.. _whatsnew_0180.api_breaking:
Backwards incompatible API changes
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- the leading whitespaces have been removed from the output of ``.to_string(index=False)`` method (:issue:`11833`)
- the ``out`` parameter has been removed from the ``Series.round()`` method. (:issue:`11763`)
- ``DataFrame.round()`` leaves non-numeric columns unchanged in its return, rather than raises. (:issue:`11885`)
- ``DataFrame.head(0)`` and ``DataFrame.tail(0)`` return empty frames, rather than ``self``. (:issue:`11937`)
- ``Series.head(0)`` and ``Series.tail(0)`` return empty series, rather than ``self``. (:issue:`11937`)
NaT and Timedelta operations
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
``NaT`` and ``Timedelta`` have expanded arithmetic operations, which are extended to ``Series``
arithmetic where applicable. Operations defined for ``datetime64[ns]`` or ``timedelta64[ns]``
are now also defined for ``NaT`` (:issue:`11564`).
``NaT`` now supports arithmetic operations with integers and floats.
.. ipython:: python
pd.NaT * 1
pd.NaT * 1.5
pd.NaT / 2
pd.NaT * np.nan
``NaT`` defines more arithmetic operations with ``datetime64[ns]`` and ``timedelta64[ns]``.
.. ipython:: python
pd.NaT / pd.NaT
pd.Timedelta('1s') / pd.NaT
``NaT`` may represent either a ``datetime64[ns]`` null or a ``timedelta64[ns]`` null.
Given the ambiguity, it is treated as a ``timedelta64[ns]``, which allows more operations
to succeed.
.. ipython:: python
pd.NaT + pd.NaT
# same as
pd.Timedelta('1s') + pd.Timedelta('1s')
as opposed to
.. code-block:: python
In [3]: pd.Timestamp('19900315') + pd.Timestamp('19900315')
TypeError: unsupported operand type(s) for +: 'Timestamp' and 'Timestamp'
However, when wrapped in a ``Series`` whose ``dtype`` is ``datetime64[ns]`` or ``timedelta64[ns]``,
the ``dtype`` information is respected.
.. code-block:: python
In [1]: pd.Series([pd.NaT], dtype='<M8[ns]') + pd.Series([pd.NaT], dtype='<M8[ns]')
TypeError: can only operate on a datetimes for subtraction,
but the operator [__add__] was passed
.. ipython:: python
pd.Series([pd.NaT], dtype='<m8[ns]') + pd.Series([pd.NaT], dtype='<m8[ns]')
``Timedelta`` division by ``floats`` now works.
.. ipython:: python
pd.Timedelta('1s') / 2.0
Subtraction by ``Timedelta`` in a ``Series`` by a ``Timestamp`` works (:issue:`11925`)
.. ipython:: python
ser = pd.Series(pd.timedelta_range('1 day', periods=3))
ser
pd.Timestamp('2012-01-01') - ser
Bug in QuarterBegin with n=0
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
In previous versions, the behavior of the QuarterBegin offset was inconsistent
depending on the date when the ``n`` parameter was 0. (:issue:`11406`)
The general semantics of anchored offsets for ``n=0`` is to not move the date
when it is an anchor point (e.g., a quarter start date), and otherwise roll
forward to the next anchor point.
.. ipython:: python
d = pd.Timestamp('2014-02-01')
d
d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
d + pd.offsets.QuarterBegin(n=0, startingMonth=1)
For the ``QuarterBegin`` offset in previous versions, the date would be rolled
*backwards* if date was in the same month as the quarter start date.
.. code-block:: python
In [3]: d = pd.Timestamp('2014-02-15')
In [4]: d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
Out[4]: Timestamp('2014-02-01 00:00:00')
This behavior has been corrected in version 0.18.0, which is consistent with
other anchored offsets like ``MonthBegin`` and ``YearBegin``.
.. ipython:: python
d = pd.Timestamp('2014-02-15')
d + pd.offsets.QuarterBegin(n=0, startingMonth=2)
Other API Changes
^^^^^^^^^^^^^^^^^
- ``DataFrame.between_time`` and ``Series.between_time`` now only parse a fixed set of time strings. Parsing of date strings is no longer supported and raises a ``ValueError``. (:issue:`11818`)
.. ipython:: python
s = pd.Series(range(10), pd.date_range('2015-01-01', freq='H', periods=10))
s.between_time("7:00am", "9:00am")
This will now raise.
.. code-block:: python
In [2]: s.between_time('20150101 07:00:00','20150101 09:00:00')
ValueError: Cannot convert arg ['20150101 07:00:00'] to a time.
- ``.memory_usage`` now includes values in the index, as does memory_usage in ``.info`` (:issue:`11597`)
Changes to eval
^^^^^^^^^^^^^^^
In prior versions, new columns assignments in an ``eval`` expression resulted
in an inplace change to the ``DataFrame``. (:issue:`9297`)
.. ipython:: python
df = pd.DataFrame({'a': np.linspace(0, 10, 5), 'b': range(5)})
df.eval('c = a + b')
df
In version 0.18.0, a new ``inplace`` keyword was added to choose whether the
assignment should be done inplace or return a copy.
.. ipython:: python
df
df.eval('d = c - b', inplace=False)
df
df.eval('d = c - b', inplace=True)
df
.. warning::
For backwards compatability, ``inplace`` defaults to ``True`` if not specified.
This will change in a future version of pandas. If your code depends on an
inplace assignment you should update to explicitly set ``inplace=True``
The ``inplace`` keyword parameter was also added the ``query`` method.
.. ipython:: python
df.query('a > 5')
df.query('a > 5', inplace=True)
df
.. warning::
Note that the default value for ``inplace`` in a ``query``
is ``False``, which is consistent with prior versions.
``eval`` has also been updated to allow multi-line expressions for multiple
assignments. These expressions will be evaluated one at a time in order. Only
assignments are valid for multi-line expressions.
.. ipython:: python
df
df.eval("""
e = d + a
f = e - 22
g = f / 2.0""", inplace=True)
df
.. _whatsnew_0180.deprecations:
Deprecations
^^^^^^^^^^^^
.. _whatsnew_0180.window_deprecations:
- The functions ``pd.rolling_*``, ``pd.expanding_*``, and ``pd.ewm*`` are deprecated and replaced by the corresponding method call. Note that
the new suggested syntax includes all of the arguments (even if default) (:issue:`11603`)
.. code-block:: python
In [1]: s = Series(range(3))
In [2]: pd.rolling_mean(s,window=2,min_periods=1)
FutureWarning: pd.rolling_mean is deprecated for Series and
will be removed in a future version, replace with
Series.rolling(min_periods=1,window=2,center=False).mean()
Out[2]:
0 0.0
1 0.5
2 1.5
dtype: float64
In [3]: pd.rolling_cov(s, s, window=2)
FutureWarning: pd.rolling_cov is deprecated for Series and
will be removed in a future version, replace with
Series.rolling(window=2).cov(other=<Series>)
Out[3]:
0 NaN
1 0.5
2 0.5
dtype: float64
- The the ``freq`` and ``how`` arguments to the ``.rolling``, ``.expanding``, and ``.ewm`` (new) functions are deprecated, and will be removed in a future version. You can simply resample the input prior to creating a window function. (:issue:`11603`).
For example, instead of ``s.rolling(window=5,freq='D').max()`` to get the max value on a rolling 5 Day window, one could use ``s.resample('D',how='max').rolling(window=5).max()``, which first resamples the data to daily data, then provides a rolling 5 day window.
- ``pd.tseries.frequencies.get_offset_name`` function is deprecated. Use offset's ``.freqstr`` property as alternative (:issue:`11192`)
- ``pandas.stats.fama_macbeth`` routines are deprecated and will be removed in a future version (:issue:`6077`)
- ``pandas.stats.ols``, ``pandas.stats.plm`` and ``pandas.stats.var`` routines are deprecated and will be removed in a future version (:issue:`6077`)
.. _whatsnew_0180.prior_deprecations:
Removal of prior version deprecations/changes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
- Removal of ``rolling_corr_pairwise`` in favor of ``.rolling().corr(pairwise=True)`` (:issue:`4950`)
- Removal of ``expanding_corr_pairwise`` in favor of ``.expanding().corr(pairwise=True)`` (:issue:`4950`)
.. _whatsnew_0180.performance:
Performance Improvements
~~~~~~~~~~~~~~~~~~~~~~~~
- Improved performance of ``andrews_curves`` (:issue:`11534`)
- Improved huge ``DatetimeIndex``, ``PeriodIndex`` and ``TimedeltaIndex``'s ops performance including ``NaT`` (:issue:`10277`)
.. _whatsnew_0180.bug_fixes:
Bug Fixes
~~~~~~~~~
- Bug in ``GroupBy.size`` when data-frame is empty. (:issue:`11699`)
- Bug in ``Period.end_time`` when a multiple of time period is requested (:issue:`11738`)
- Regression in ``.clip`` with tz-aware datetimes (:issue:`11838`)
- Bug in ``date_range`` when the boundaries fell on the frequency (:issue:`11804`)
- Bug in consistency of passing nested dicts to ``.groupby(...).agg(...)`` (:issue:`9052`)
- Bug in vectorized ``DateOffset`` when ``n`` parameter is ``0`` (:issue:`11370`)
- Bug in ``Timedelta.round`` with negative values (:issue:`11690`)
- Bug in ``.loc`` against ``CategoricalIndex`` may result in normal ``Index`` (:issue:`11586`)
- Bug in ``DataFrame.info`` when duplicated column names exist (:issue:`11761`)
- Bug in ``.copy`` of datetime tz-aware objects (:issue:`11794`)
- Bug in ``Series.apply`` and ``Series.map`` where ``timedelta64`` was not boxed (:issue:`11349`)
- Bug in subclasses of ``DataFrame`` where ``AttributeError`` did not propagate (:issue:`11808`)
- Bug groupby on tz-aware data where selection not returning ``Timestamp`` (:issue:`11616`)
- Bug in ``pd.read_clipboard`` and ``pd.to_clipboard`` functions not supporting Unicode; upgrade included ``pyperclip`` to v1.5.15 (:issue:`9263`)
- Bug in ``DataFrame.query`` containing an assignment (:issue:`8664`)
- Bug in timezone info lost when broadcasting scalar datetime to ``DataFrame`` (:issue:`11682`)
- Bug in ``Index`` creation from ``Timestamp`` with mixed tz coerces to UTC (:issue:`11488`)
- Bug in ``to_numeric`` where it does not raise if input is more than one dimension (:issue:`11776`)
- Bug in parsing timezone offset strings with non-zero minutes (:issue:`11708`)
- Bug in ``df.plot`` using incorrect colors for bar plots under matplotlib 1.5+ (:issue:`11614`)
- Bug in the ``groupby`` ``plot`` method when using keyword arguments (:issue:`11805`).
- Bug in ``DataFrame.duplicated`` and ``drop_duplicates`` causing spurious matches when setting ``keep=False`` (:issue:`11864`)
- Bug in ``.loc`` result with duplicated key may have ``Index`` with incorrect dtype (:issue:`11497`)
- Bug in ``pd.rolling_median`` where memory allocation failed even with sufficient memory (:issue:`11696`)
- Bug in ``.style.bar`` may not rendered properly using specific browser (:issue:`11678`)
- Bug in rich comparison of ``Timedelta`` with a ``numpy.array`` of ``Timedelta`` that caused an infinite recursion (:issue:`11835`)
- Bug in ``df.replace`` while replacing value in mixed dtype ``Dataframe`` (:issue:`11698`)
- Bug in ``Index`` prevents copying name of passed ``Index``, when a new name is not provided (:issue:`11193`)
- Bug in ``read_excel`` failing to read any non-empty sheets when empty sheets exist and ``sheetname=None`` (:issue:`11711`)
- Bug in ``read_excel`` failing to raise ``NotImplemented`` error when keywords ``parse_dates`` and ``date_parser`` are provided (:issue:`11544`)
- Bug in ``read_sql`` with pymysql connections failing to return chunked data (:issue:`11522`)
- Bug in ``.to_csv`` ignoring formatting parameters ``decimal``, ``na_rep``, ``float_format`` for float indexes (:issue:`11553`)
- Bug in ``DataFrame`` when masking an empty ``DataFrame`` (:issue:`11859`)