baohan
/
scikit-learn
mirror of https://github.com/scikit-learn/scikit-learn.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
21,216 Commits 31 Branches 142 Tags 892 MiB
Commit Graph

21216 Commits

Author SHA1 Message Date
Scott Andrews f241f5f14e Fixed issue #5830, capitalization in advanced installation documenation 2016-09-12 18:31:04 -04:00
Joel Nothman c726085e20 DOC mention true positives etc in confusion matrix docs (#7392) 2016-09-12 12:22:41 -04:00
Raghav RV b18b1611cd [MRG+1] TST Stronger test for _check_is_permutation (#7395) 
* TST Stronger test for _check_is_permutation

* TST Ensure additional duplicate indices are caught
 2016-09-12 11:44:16 -04:00
Raghav RV c3b21153e8 MNT Remove unused constants from public API(?) (#6517) 2016-09-12 10:17:59 -04:00
Olivier Grisel 5a018a3d72 Merge pull request #7393 from ogrisel/mlp-multilabel 
[MRG] Fix for MLP multilabel predict_proba and removal of decision_function
 2016-09-12 13:09:27 +02:00
Olivier Grisel 1e8ca7df87 ENH more informative validation error message 2016-09-12 11:38:14 +02:00
Jiyuan Qian e1b5d601a2 MLP bug fixes with random state and loss, removed decision_function, and added test 2016-09-12 11:37:39 +02:00
Olivier Grisel b966b44585 DOC fix class.rst for Label / Group CV tools 2016-09-12 09:06:44 +02:00
Raghav RV 9a12555e6d [MRG+1] ENH/MNT Rename labels --> groups in CV tools (#6660) 2016-09-11 19:14:41 +02:00
Manoj Kumar 8da5092e83 [MRG] Fail imputer early when number of features are not the same in fit and transform (#7374) 2016-09-11 23:22:13 +10:00
Olivier Grisel 12a0125a3c DOC fix for svd_solver in PCA docstring 2016-09-11 12:02:20 +02:00
jblackburne ba7cf38f99 Removed isclose from _binary_clf_curve. (#7353) 
Added a bugfix report to whats_new.rst. Modified a unit test to check for the kind of problems caused by isclose.
 2016-09-11 12:48:32 +10:00
Olivier Grisel a86fd7f5e7 Add Dirichlet process prior to BayesianGaussianMixture (#7386) 2016-09-10 15:41:52 +02:00
Ruifeng Zheng 105d03314c [MRG+1] Add hyperlink and example for kappa (#7214) 2016-09-10 15:34:35 +02:00
Andreas Mueller b4872fe30d [MRG] add warning when importing old or new pickle. (#7248) 2016-09-10 14:17:42 +02:00
Andreas Mueller 31a4691d76 [MRG+1] fix sampling in stratified shuffle split (#6472) 
Fix sampling in stratified shuffle split, break tests that test sampling.
 2016-09-09 19:00:49 +02:00
Joel Nothman 49d126fd8f [MRG + 1] FIX use high precision cumsum and check it is stable enough (#7331) 
* FIX use high precision cumsum and check it is stable enough
 2016-09-09 06:17:12 +02:00
Olivier Grisel 2bf96df8da Merge pull request #7304 from jnothman/binary_prf_mine 
[MRG + 1] Complete deprecation of default average in P/R/F metrics
 2016-09-08 23:48:55 +02:00
Gael Varoquaux 4b2304f05a Merge pull request #7261 from betatim/negative-scorers 
[MRG + 2] Rename scorers like `mse` to `neg_mse`
 2016-09-08 20:40:17 +02:00
Tim Head 393540492a Test that warning is raised when using deprecated scorer 2016-09-08 17:34:54 +02:00
Joel Nothman 62ad5fed90 FIX Revert unintentionally committed code 2016-09-08 18:47:02 +10:00
Andreas Mueller 2c4b43c289 change installation description to not discourage people from installing scipy or numpy with pip. (#7313) 2016-09-08 10:09:56 +02:00
Joel Nothman 22e43054fc TST fix test to match altered deprecation msg 2016-09-08 10:29:06 +10:00
Andreas Mueller 6972d6c15c DOC minor doc fixes for sphinx. (#7357) 2016-09-08 10:07:40 +10:00
Andreas Mueller 680ab517df [MRG + 2] remove complicated equality checks in clone as __init__ shouldn't touch anything. (#5540) 
* BF: issue 5522 (cloning objects with pandas.Dataframe attributes)

* super conservative fix, pending GP and VBGMM fixes.

* TST improved test for df param

* add deprecation warning, add whatsnew entry

* fixed place of deprecation warning, added test for deprecation warning.

* pep8

* fix whitespace error
 2016-09-08 10:07:11 +10:00
b0noI a03db89eba [MRG+1] Support of the collections.Sequence type has been added to the _check_param_grid method from model_selection. (#7323) 
* Support of the collections.Sequence type has been added to the _check_param_grid method from model_selection.

* test_grid_search_when_param_grid_includes_range test was refactored (parts that are not nessesary have been removed).

* test_grid_search_bad_param_grid now checks that value is not string. This is important since string is a Sequence.

* _check_param_grid now checks is the type is not string together with the check for other types.

* whats_new.rst has been updated to include information about bug fix for bug #7322.

* Description of the fix for the bug #7322 has been updated.

* Fix for indented in model_selection._search.py.
 2016-09-07 17:06:44 -04:00
Nelle Varoquaux 35bb1c6ce2 Merge pull request #7324 from raghavrv/rename_results_ 
ENH/MNT results_ --> cv_results; test_mean_score --> mean_test_score et al.
 2016-09-07 09:27:30 -07:00
Raghav R V 4dbab2680e results_ --> cv_results_; "test_mean_score" --> "mean_test_score" et al. 2016-09-06 18:46:09 +02:00
Kyle Jackson 6c12514f25 [MRG] DOC removed redundant word from KNN docs (#7345) 2016-09-06 12:17:13 -04:00
Loïc Estève 53058619b5 Fix flake8_diff.sh when Travis runs in a fork 
The grep was not matching anything so the script exited with a 1 exit code.
 2016-09-06 16:11:31 +02:00
Tim Head 6673f75828 Update examples to new scorer names 2016-09-06 11:25:11 +02:00
Olivier Grisel 5daee9627d DOC fix iterated_power='auto' in doctests 2016-09-06 11:11:07 +02:00
Andreas Mueller 4a72792db4 change default iterated_power to auto. 2016-09-06 11:03:16 +02:00
Andreas Mueller 927f00ca9a fix warning and behavior in randomized_svd wrt power iterations 2016-09-06 11:03:16 +02:00
Olivier Grisel b9d16b26a8 Merge pull request #7342 from themrmax/pydotplus-decision-tree-plotting 
change decision tree example to use pydotplus
 2016-09-06 10:05:16 +02:00
Tim Head 2ba347841a Keep abstractmethod decorator 2016-09-06 08:38:51 +02:00
Nelson Liu 287a3298e1 [MRG] Add link to AUTHORS.rst in readme (#7320) 2016-09-06 10:28:27 +10:00
Joel Nothman d20170884e DOC move changelog heading to correct place 2016-09-06 10:26:42 +10:00
Brandon Carter 3dcfb564a0 [DOC] add release dates for each version on whats_new page (#7225) 2016-09-06 10:23:20 +10:00
saihttam a627a079c6 [MRG+1] Fix for test_neighbors_metrics() check kd-tree algorithm when possible (#7339) 
* Fix for test_neighbors_metrics() check kd-tree algorithm when possible

* Improving check in test_neighbors_metrics by using a dictionary for the results and checking for existence of key for kd_tree

* Fix flake8 problem due to line too long

* remove enumerate, since not needed when using results dict
 2016-09-05 22:49:14 +10:00
Joel Nothman ce3c84fb56 DOC fix code formatting 2016-09-05 22:08:44 +10:00
Olivier Grisel f916449ec0 Merge pull request #7317 from amueller/common_test_names 
[MRG+1] make more explicit which checks are run
 2016-09-05 13:11:31 +02:00
themrmax fed401c831 change decision tree example to use pydotplus for compatibility with Python 3 2016-09-05 14:45:57 +10:00
Tim Head f4419a9f2a Small documentation fix 2016-09-04 17:21:54 +02:00
Manoj Kumar 49fb295561 [MRG+2] Fix repr on isotropic kernels when a 1-D length scale is given (#7259) 2016-09-03 22:02:07 +10:00
Joel Nothman 2b11962b05 Move comment on relationship between labels and pos_label to docstring 2016-09-01 22:10:47 +10:00
Mridul Seth 0f2a00f890 Remove dtype conversion warning (#7265) 2016-09-01 10:57:21 +02:00
Olivier Grisel ccad485dd0 Merge pull request #7316 from totalgood/doc-clf-reg-patch 
[MRG + 1] fix regressor var name
 2016-09-01 08:59:58 +02:00
Tim Head 88a395abb3 Link to mean_squared_error function 
Makes the distinction between the metric and the scorer a bit
easier for the reader.
 2016-09-01 08:16:24 +02:00
Andreas Mueller 7fc41768b1 pep8 2016-08-31 22:37:23 -04:00