1f5bcaeb39
FIX float16 overflow on accumulator operations in StandardScaler ( #13010 )
2019-01-26 18:15:18 +01:00
365c1b2071
MNT more informative warning in estimator_checks ( #13002 )
2019-01-17 21:59:40 +01:00
07b0a420a9
FIX builtin PrettyPrinter usage on estimators ( #12938 )
2019-01-14 17:29:57 +11:00
d300f406ae
MAINT Simplify super() calls ( #12812 )
2019-01-10 22:27:06 +01:00
e0697b01a7
MNT Used scipy.special.expit for the inverse of the logit function ( #12931 )
2019-01-08 18:35:19 +11:00
701144559f
MAINT Remove unused utils.fixes ( #12928 )
...
This continues the work done in https://github.com/scikit-learn/scikit-learn/pull/12639 on dropping the python 2 support by,
- ~~removing unnecessary `from __future__` imports~~
- removing unused `sklearn.utils.fixes` assuming we can agree in https://github.com/scikit-learn/scikit-learn/issues/12927 that `sklearn.utils.fixes` are private as was stated e.g. in https://github.com/scikit-learn/scikit-learn/issues/6616#issuecomment-245109979
2019-01-08 12:43:49 +11:00
acb8106472
MNT Use list and dict comprehension ( #12668 )
2019-01-08 09:32:34 +11:00
684d8a221d
MAINT Use set litterals when possible ( #12667 )
2019-01-06 19:21:45 +08:00
952ef6637a
MRG Drop legacy python / remove six dependencies ( #12639 )
2019-01-03 15:50:05 +02:00
a50c03f975
[MRG] Add pprint for estimators - continued ( #11705 )
...
* add pprint for estimators
* strip color from length, add color option
* Minor cleaning, fixes, factoring and docs
* Added some basic tests
* Fixed line length issue
* fixed flake8 and added visual test for review
* Fixed test
* Fixed Python 2 issues (inspect.signature import)
* Trying to fix flake8 again
* Added special repr for functions
* Added some other visual tests
* Changed _format_function in to _format_callable
because callable() returns True also for class objects (which we want to
reprensent with their name as well anyway)
* Consistent output in Python 2 and 3
* WIP
* Now using the builtin pprint module
* pep8
* Added changed_only param
* Fixed printing when string would fit in less than line width
* Fixed printing of steps parameter
* Fixed changed_only param for short estimators
* fixed pep8
* Added some more description in docstring
* changed_only is now an option from set_config()
* Put _pprint.py into sklearn/utils, added tests
* Added doctest NORMALIZE_WHITESPACE where needed
* Fixed tests
* fix test-doc
* fixing test that passed before....
* Fixed tests
* Added test for changed_only and long lines
* typo
* Added authors names
* Added license file
* Added ellipsis based on number of elements in sequence + added increasinly aggressive repr strategies
* Updated whatsnew
* dont use increaingly aggressive strategy
* Fixed tests
* Removed LICENSE file and put license text in _pprint.py
* fixed test_base
* Sorted parameters dictionary for consistent output in 3.5
* Actually using OrderedDict...
* Addressed comments
* Added test for NaN changed parameter
* Update whatsnew
* Added example to set_config()
* Removed example
* Added example in gallery
* Spelling
2018-12-19 23:48:21 -05:00
2bd87f6ed8
Remove python < 3.5 from CI ( #12746 )
2018-12-14 10:53:12 +01:00
5be8507dcb
MNT Deprecate assert_true and assert_false ( #12717 )
2018-12-13 23:36:53 +08:00
4f5dd77d19
BUG: fix check_array on pandas Series with custom dtype (eg categorical) ( #12706 )
...
Closes #12699 . Related to #12625
2018-12-03 21:25:52 +11:00
Hanmin Qin
ccf0d9286d
MNT Minor code refactoring in utils.testing ( #12688 )
2018-11-30 23:19:29 +08:00
fa98a72dcc
MNT Replaced all occurrences of assert_true and assert_false with assert ( #12588 )
2018-11-28 09:16:26 +08:00
104f684791
FIX check_array dtype check for pandas series ( #12625 )
2018-11-21 09:23:01 +11:00
d25da1be20
FIX make joblib utils private, and remove mentions of externals.joblib ( #12345 )
2018-11-20 10:53:52 +11:00
4e8194909c
FIX remove FutureWarning in _object_dtype_isnan and add test ( #12567 )
2018-11-12 21:28:05 +08:00
Hanmin Qin
43e3a02085
MNT Remove unused assert_true imports ( #12560 )
2018-11-11 11:08:37 +08:00
362cb3bcab
TST autoreplace assert_true(...==...) with plain assert ( #12547 )
2018-11-11 09:05:34 +08:00
1128094271
BLD we should ensure continued support for joblib 0.11 ( #12350 )
2018-11-06 22:10:23 +01:00
5d8dfc9677
FIX SkLearn `.score()` method generating error with Dask DataFrames ( #12462 )
2018-11-06 21:46:13 +11:00
6226cf59ad
TST skip test_backend_respected if joblib is forced into serial mode ( #12496 )
2018-11-02 23:10:57 +01:00
1c88b3c9c7
Fix IncrementalPCA when final batch is smaller than dimensions required for SVD ( #12379 )
2018-10-31 14:32:19 +01:00
f6b0c67290
TST Added estimator check for idempotence of fit() ( #12328 )
2018-10-29 17:31:12 +01:00
9ec5a15823
ALL Add resample and shuffle to __all__ ( #12456 )
2018-10-26 11:00:01 +02:00
5af272ac61
MNT Remove unused variables ( #12230 )
2018-10-22 12:48:40 +11:00
a1d0e96791
FIX Increase mean precision for large float32 arrays ( #12338 )
2018-10-16 09:08:39 +02:00
Hanmin Qin
afe0a9b70d
MNT skip test falling on master in legacy platforms ( #12382 )
2018-10-16 07:57:12 +11:00
98357ec4d4
FIX (0.21) make count_nonzero dtype invariant wrt axis ( #12341 )
2018-10-14 21:12:02 +08:00
0f94f2962b
MNT simple deprecations and removals for 0.21 ( #12238 )
...
Part of #11992 .
These were all the things that seemed pretty straight-forward. It's actually a bit bulky but should still be easy to review, hopefully.
2018-10-11 14:56:37 -04:00
c8a4132774
DOC check_array() and check_X_y() documentation update ( #12340 )
2018-10-10 21:16:44 +11:00
a1be3254c0
MNT Make check_X_y raise a better error when y is None ( #12283 )
2018-10-05 18:35:04 +08:00
1e052e9da9
Converting http to https (2)... ( #12292 )
2018-10-04 23:06:14 +02:00
74b56dbc57
MNT Converting http to https ( #12277 )
2018-10-04 22:37:57 +08:00
Hanmin Qin
f456a40828
MNT Change show_versions format to suit markdown ( #12255 )
2018-10-04 09:38:35 +08:00
e0e7387606
Remove unused private functions ( #12253 )
2018-10-03 21:08:17 -04:00
fb7be879a3
[MRG + 1] return_train_score deprecation ( #12241 )
2018-10-03 19:24:14 +02:00
da0cb32270
FIX Use take instead of choose in compute_sample_weight ( #12165 )
2018-09-27 21:51:49 +08:00
a11154e727
[MRG] improve check_non_negative for sparse matrices ( #12106 )
2018-09-27 14:13:50 +02:00
a0418215fe
TST FIX use match rather than message in pytest.raises ( #12001 )
...
Previously these assertions would pass without matching.
2018-09-05 07:36:15 +10:00
51b1b7cce3
Remove unnecessary copy for float64 in sparsefuncs_fast ( #11966 )
2018-09-02 10:17:03 +02:00
7ed61a24fe
ENH add multi_class='auto' for LogisticRegression, default from 0.22; default solver will be 'lbfgs' ( #11905 )
...
* Change default solver in LogisticRegression
* This is an API change, not a feature
* Decrease numerical precision in LogisticRegression doctest
* ENH add multi_class='auto' for LR, default from 0.22
* No warning when binary
2018-08-26 23:00:02 +03:00
b42a5152af
MAINT Fix test_logistic::test_dtype_match failure on 32 bit arch ( #11899 )
2018-08-25 11:54:30 +08:00
a85eeb28ee
MRG/REL fixes /skips for 32bit tests ( #11879 )
2018-08-22 22:48:32 +03:00
757663c881
MNT Clip percentile_idx in _weighted_percentile ( #11847 )
2018-08-19 08:54:28 +08:00
fc56da504d
Deprecate fetch_mldata ( #11466 )
...
* API Deprecate fetch_mldata and update examples
* Use pytest's filterwarnings
* Rm unused import
* Remove broken doctest
* Refer user to openml URL
* DOC whatsnew tweak
2018-08-18 19:27:48 +03:00
Hanmin Qin
5764baa4b5
MNT log1p -> log in cython files ( #11852 )
2018-08-18 19:51:18 +08:00
07e909ad66
MNT Use std math lib log1p instead of numpy log1p in cython inner loops ( #11848 )
2018-08-18 15:45:19 +08:00
477c921c0f
ENH extending BaseSearchCV with a custom search strategy ( #9599 )
2018-08-06 08:18:48 +10:00
Raf Baluyot
Hanmin Qin