baohan
/
scikit-learn
mirror of https://github.com/scikit-learn/scikit-learn.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
10,497 Commits 31 Branches 142 Tags 892 MiB
Commit Graph

108 Commits

Author SHA1 Message Date
Lars Buitinck 917f309559 ENH fit_transform on KMeans 
Naive implementation for now. Fixes Issue #1157 (partly, at least).
 2012-10-31 13:28:28 +01:00
Gael Varoquaux a69557e3b9 TST: increase coverage in spectral clustering 2012-10-27 22:10:51 +02:00
Gael Varoquaux 402a304941 BUG: fix test error 2012-10-27 17:10:44 +02:00
Gael Varoquaux d4a6958908 FIX: numerical stability in spectral 
This enables me to unskip a skipped test, hurray!
 2012-10-27 17:06:44 +02:00
Gael Varoquaux 315d42111f ENH/FIX add a lobpcg solver to spectral embedding 
This makes the solver a bit better behaved on some problems
 2012-10-27 17:06:44 +02:00
Mathieu Blondel 6eec2e651b Even more sklearn.utils.testing. 2012-10-26 00:09:46 +09:00
Gael Varoquaux 27baaf33d5 TST: cater for 0.9 not > 0.9 
Fixes #1228
 2012-10-12 11:26:59 +02:00
Andreas Mueller 689cd96b4e FIX removed isotropic spectral clustering test because of arpack problems. 2012-08-29 22:53:34 +02:00
Andreas Mueller fef96c785b FIX try to be a bit nicer to arpack - any one with a different setting care to try to make a more stable test? 2012-08-29 22:06:22 +02:00
Andreas Mueller a11c38d922 tst add check for isotropic data in spectral clustering 2012-08-29 21:56:56 +02:00
Gael Varoquaux ae377f6a39 BUG: fix my fix for affinity :( 2012-08-28 10:21:45 +03:00
Gael Varoquaux 1378012724 BUG: choose seed to get affinity test working 
Fragile test :)
 2012-08-28 10:19:21 +03:00
Andreas Mueller c59a7dd066 ENH updated whatsnew for bugfix, removed warning box, tightened test. 2012-08-27 11:41:52 +02:00
Andreas Mueller b797265ebf Merge branch 'fit_pairwise' 
Conflicts:
	sklearn/cluster/tests/test_spectral.py
 2012-08-27 10:39:24 +02:00
Andreas Mueller dcbe01f624 TST added tests for different spectral clustering affinities 2012-08-27 10:32:57 +02:00
Andreas Mueller 0df98bb3c0 ENH add test, revert affinity propagation to previous parametrization (was a bit over-eager there) 2012-08-27 09:35:14 +02:00
Andreas Mueller 949ddd10b1 ENH renamed ``neq_sqr_euclidean`` to ``euclidean`` so we it is easier to parse 2012-08-26 19:04:41 +02:00
Andreas Mueller f6aaa4d234 ENH affinity propagation now has an ``affinity`` parameter, instead of a ``precomputed`` parameter, to support other affinities in the future. 2012-08-26 17:20:42 +02:00
Andreas Mueller 3990e13d74 ENH catch expected warning in ward clustering 2012-08-18 21:53:52 +01:00
Andreas Mueller 82a762f69d COSMIT pep8 2012-08-10 11:06:42 +01:00
Andreas Mueller 194b0f10bc Merge branch 'master' into fit_pairwise 2012-08-05 16:49:48 +01:00
Andreas Mueller 5c3853cceb Merge branch 'master' into fit_pairwise 2012-08-05 14:44:05 +01:00
GaelVaroquaux 6c4ca28770 Typos 2012-07-30 13:40:56 +02:00
GaelVaroquaux c389d7d3d7 Cosmit: comment 
as requested by @ogrisel
 2012-07-30 11:33:34 +02:00
Gael Varoquaux 6d0c72724d TST: more tests for hierarchical 2012-07-26 09:15:57 +02:00
Gael Varoquaux 5ad4d0e90a TST: test TypeError in Ward 2012-07-26 09:15:57 +02:00
Gael Varoquaux bdb32d4714 TST: More testing in hierarchical 2012-07-26 09:15:57 +02:00
GaelVaroquaux 80220085a9 FIX: make ward_tree work on 1D data 2012-07-26 09:15:57 +02:00
GaelVaroquaux 99f58139e7 TEST: improve test coverage in hierarchical 2012-07-26 09:15:57 +02:00
GaelVaroquaux df1e71149c DOC: ward docstring and testing 2012-07-26 09:15:57 +02:00
GaelVaroquaux 7bc92afb71 BUG: fix ward tests 2012-07-26 09:15:57 +02:00
Lars Buitinck 4e1912a7c5 TST absolute imports in spectral clustering tests 2012-07-25 12:40:08 +02:00
Robert Layton 993b6b4ab7 Test that an unknown mode raises the appropritate error 2012-07-25 09:23:40 +02:00
Robert Layton 26dd857dd4 Spectral will raise an error if the mode is set to amg and pyamg is not available 2012-07-25 09:23:40 +02:00
Robert Layton bbc8038bd7 pep8 2012-07-25 09:23:40 +02:00
Robert Layton e8af7c6d33 Test for "amg" mode for spectral clustering added. 2012-07-25 09:23:40 +02:00
Andreas Mueller 37cadf03c0 Merge branch 'master' into fit_pairwise 
Conflicts:
	sklearn/svm/base.py
 2012-06-22 22:20:18 +02:00
Andreas Mueller 8f8f1c0724 Merge pull request #875 from AlexandreAbraham/ward_coo_bug 
Ward clustering fails if connectivity is not item assignable and there is more than one component
 2012-06-05 13:47:53 -07:00
Andreas Mueller ef82d75c5b ENH going for a universal property ``_pairwise`` instead of many functions. 2012-06-03 20:39:02 +02:00
Andreas Mueller 1d88edc86f ENH fixed affinity propagation test. Need more tests. 2012-06-03 20:33:06 +02:00
Andreas Mueller 86baa8a7d6 ENH "fit_pairwise" for spectral clustering. 2012-06-03 20:33:06 +02:00
GaelVaroquaux 9c5f9af79e FIX: unique in old numpy 2012-06-01 18:23:30 +02:00
Alexandre Abraham 4045b41e74 Fix test function name. 2012-06-01 17:46:44 +02:00
Alexandre Abraham 43a363e46f Add a non-regression test for the bug of connectivity fixing. 2012-05-31 10:42:01 +02:00
Vlad Niculae 6e2472af7b Remove a warning from kmeans tests 2012-05-30 14:36:38 +03:00
Gael Varoquaux 7054c7a0b7 COSMIT: avoid deprecation warnings 2012-05-30 07:46:23 +02:00
Andreas Mueller 805c0f3540 ENH renamed 'k' to n_clusters in SpectralClustering 2012-05-30 07:41:34 +02:00
Andreas Mueller 388c1af757 ENH rename 'k' in KMeans and MiniBatchKMeans 2012-05-30 07:41:34 +02:00
Andreas Mueller 8b24cbdb56 COSMIT pep8 2012-05-29 22:38:42 +02:00
Gael Varoquaux e04fe926e0 BUG: reassigning cluster centers with X sparse 
The code to reassign cluster centers didn't work with X sparse because a
sparse matrix is always 2D

closes #867
 2012-05-25 08:51:55 +02:00