a47c3b9a07
ENH use expit in ``_BinaryGaussianProcessClassifierLaplace`` ( #9011 )
...
SciPy already provides a numerically stable ``special.expit``
function, which does not overflow on the example mentioned in #8641 .
2017-06-07 10:20:54 +10:00
623ee42260
Fix link to cross_validation ( #9025 )
...
Previously read "search [...] for the best Cross-validation: evaluating estimator performance score", because the title of 'cross_validation' changed.
2017-06-06 22:19:55 +02:00
38c7e93b1e
[MRG+1] LDA verbose should print iteration number in all cases ( #8944 )
...
* Printing iter and max_iter in online_lda if verbose
* Fixed typo to print iter and max_iter
* TST test verbose feature of LatentDirichletAllocation
* Py2 compatibility
2017-06-06 18:06:43 +02:00
bdb00d17b9
Fix typos in test_covariance.py and extmath.py ( #8997 )
2017-06-06 16:54:51 +02:00
7a23b2493f
Update to Sphinx-Gallery v0.1.11 ( #8222 )
2017-06-06 16:35:24 +02:00
1c41368bac
[MRG+1] Uncontroversial fixes from estimator tags branch ( #8086 )
...
* some bug fixes.
* minor fixes to whatsnew
* typo in whatsnew
* add test for n_components = 1 transform in dict learning
* feature extraction doc fix
* fix broken test
* revert aggressive input validation changes
* in SelectFromModel, don't store threshold_ in transform. If we called "fit", use estimates from last "fit".
* move score from EllipticEnvelope to OutlierDetectionMixin
* revert changes to Tfidf documentation
* remove dummy input validation from whatsnew
* fix text feature tests
* rewrite from_model threshold again...
* remove stray condition
* fix self.estimator -> estimator, slightly more interesting test
* typo in comment
* Fix issues in SparseEncoder, add tests.
more explicit explanation of SparseEncoder change, add issue numbers to whatsnew
* minor fixes in whats_new.rst
* slightly more consistency with tuples for shapes
* not longer typo
2017-06-06 16:34:47 +02:00
e2d011d953
Division.
2017-06-06 23:13:42 +09:00
fa5926afae
[ #8972 ] Add links to FAQ ( #9002 )
...
* Add links to FAQ
* Update faq.rst
* Fix typo
2017-06-06 15:18:57 +02:00
f4b230cc65
BF: fixing error on plot_image_denoising.py with Python 2.7 ( #9003 )
2017-06-06 14:36:13 +02:00
7772e8102c
DOC add reference in probabilities to SVC doc ( #8990 )
2017-06-06 18:42:47 +10:00
Hanmin Qin
41b84f84cc
[MRG+1] Change the restriction of n_repeats in RepeatedKFold and RepeatedStratifiedKFold ( #8909 )
...
* change the restriction of n_repeats
* change test case
2017-06-05 16:51:17 +02:00
25a73db451
[MRG+1] Documentation: Add xcessiv to related projects ( #8985 )
...
* Documentation: Add xcessiv to related projects
For your consideration, this adds Xcessiv to the list of related projects under experimentation frameworks. Xcessiv lets you fully define your data, cross-validation technique, estimators, and metrics and is based heavily on the scikit-learn API.
* Modify second sentence
2017-06-05 15:25:03 +02:00
dd7a3dd0eb
Fix typos ( #8980 )
2017-06-03 23:54:04 +02:00
5fc30fd6f8
DOC recommend editable install using pip in contributing ( #8974 )
2017-06-03 18:47:20 +10:00
bd0fc236e0
[MRG+1] MAINT drop SciPy < 0.13 ( #8854 )
...
Remove sklearn.utils.fixes functions that are not needed for scipy >= 0.13 and keep deprecated wrappers in other modules.
2017-06-02 23:36:33 +02:00
a150d93078
Updated error message for max_leaf_nodes ( #8975 )
...
Solves #8964
2017-06-02 09:29:59 -07:00
7ce7134a57
[MRG+2] ENH Loop over candidates as outer loop in search ( #8322 )
...
This encourages concurrent fits to be over *different datasets* so that
fits over the same data subset are more likely to run in serial and
hence generate cache hits where memoisation is used.
2017-06-01 16:29:34 +02:00
Hanmin Qin
098fd31ce2
[MRG+1] DOC use a table in load_linnerud docstring ( #8970 )
2017-06-01 14:32:59 +02:00
57415c4610
[MRG] Fix gradient descent in TSNE ( #8768 )
...
Skip flaky test_n_iter_without_progress on 32bit
2017-06-01 13:14:35 +02:00
f198f9945e
DOC fixed typo in online_lda.py ( #8969 )
2017-06-01 19:53:22 +10:00
Hanmin Qin
453446affa
[MRG] DOC add function load_wine to the document ( #8965 )
...
* add wine dataset to the document
* Update classes.rst
* Update index.rst
2017-05-31 19:36:38 +02:00
cb63a6d0bc
[MRG+1] Add set cardinality formulation for mutual_info_score ( #8821 )
...
* Remove unnecessary braces.
* Replace R, C with |U|, |V|.
* Remove the probability formulation from docstring.
2017-05-30 10:26:12 -07:00
1cc346cece
BUILD Give clear error message when there are merge conflicts ( #8945 )
2017-05-29 13:42:47 +02:00
a2db2a0298
ENH mlp convergence warning text ( #8947 )
2017-05-29 09:18:28 +10:00
65f7c4e9c5
DOC update whats_new with LDA n_topics deprecation
2017-05-28 21:59:51 +10:00
3efefab4c8
[MRG+1] Rename n_topics to n_components in LDA ( #8938 )
2017-05-27 18:46:26 +10:00
26978b5459
Revert "[MRG+1] Added n_components parameter to LatentDirichletAllocation to replace … ( #8922 )" ( #8937 )
...
This reverts commit 890e652898
2017-05-26 00:02:51 -07:00
890e652898
[MRG+1] Added n_components parameter to LatentDirichletAllocation to replace … ( #8922 )
...
[MRG+2] Added n_components parameter to LatentDirichletAllocation to replace …
2017-05-25 23:07:10 -07:00
9131f89e6c
Update docs with new line_profiler signature and extension method
...
Update memory_profiler docs as well
Append new extensions instead of setting them directly
2017-05-25 08:25:17 +10:00
964aedbebc
[MRG] documentation fix in neighbors module ( #8929 )
2017-05-25 07:04:02 +10:00
1ada91c341
documentation fix ( #8928 )
2017-05-24 18:22:14 +10:00
83d65b84fe
Update numpy refs to 1.8.2 ( #8926 )
2017-05-24 11:53:37 +10:00
db25f57781
[MRG+1] Update version dependencies to numpy 1.8.2, scipy 1.13.3 ( #8855 )
2017-05-24 08:24:04 +10:00
bbf43483c1
Fixed markdown in doc/README.md ( #8925 )
2017-05-23 20:28:52 +02:00
97590191fe
DOC add plot_stock_market.py to expected failing examples ( #8921 )
...
as a stop gap solution until we figure out a longer-term work-around.
Also remove nibabel from conf.py.
2017-05-23 20:27:23 +02:00
1b5bff3ad0
[MRG] DOC add documentation on adding non-regression tests ( #8850 )
2017-05-23 15:23:03 +02:00
92cb166647
[MRG] DOC Homogeneous deprecation in docstring ( #8907 )
...
Update developers doc to use "deprecated" sphinx directive.
2017-05-23 08:42:08 +02:00
fcdbaca5a9
Fix typo in PCA test ( #8917 )
2017-05-23 12:10:35 +10:00
Hanmin Qin
937b47fddc
Update model_evaluation.rst ( #8915 )
2017-05-22 14:56:46 -07:00
a0c6e6ca32
DOC: styling of deprecated boxes ( #8914 )
2017-05-22 14:53:24 -07:00
Hanmin Qin
e19bb7ce5a
Fix n_splits in KFold instantiation in model_selection tests ( #8910 )
2017-05-21 01:35:02 -04:00
29d200249b
[MRG+1] De-italicize non-variables in formulas ( #8901 )
...
* De-italicize non-variables in formulas
* Fix some TeX syntax errors
2017-05-20 18:28:34 +10:00
6252f99c16
[MRG+1] Fixing Math domain error on NMF due to numpy.dot ( #8765 )
...
* Fixing overflow error on NMF due to numpy.dot
2017-05-20 09:47:43 +02:00
4397e7ec6a
[MRG+1] removed precomputed support in nearestcentroid ( #8515 )
...
* removed test_precomputed
* Revert "removed test_precomputed"
This reverts commit 71ace3a39c
2017-05-18 23:22:10 -04:00
15b7d47a2f
relax skewness assumption ( #7573 )
...
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
last corrections
whats_new
merge
complying whats_new
removed unnecessary _assert_X
increased coverage
relax skewness assumption
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
relax skewedness assumption
last corrections
whats_new
merge
complying whats_new
removed unnecessary _assert_X
increased coverage
remove cythonize.dat and merge
simplify tests
simplify tests
2017-05-18 14:17:10 -04:00
55c9443ca4
Add information what +1 and -1 means ( #8891 )
2017-05-17 17:29:17 +02:00
4c33ea2ee0
[MRG+1] EHN/DOC Make error msg about Memory more explicit ( #8865 )
2017-05-17 22:51:16 +10:00
33f50ad948
minor fixes to whats_new.rst for 0.19-dev ( #8886 )
2017-05-16 21:24:52 +10:00
0301e06b19
DOC: add missing default value in LinearRegression docstring ( #8873 )
...
* DOC fix typos in linear_model.rst
* Another typo
* DOC: add missing default value in LinearRegression docstring and fix typo
2017-05-15 13:53:05 +02:00
3a9b3f5e6a
DOC Fix user guide link in MultiTaskElasticNetCV docstring ( #8880 )
2017-05-15 07:38:22 +10:00
Sergei Lebedev
Hanmin Qin