https://github.com/scikit-learn-contrib/metric-learn

sort by:
Revision Author Date Message Commit Date
96a8abc Version bump to 0.3, updating docs 13 July 2016, 21:44:46 UTC
ce5f238 Converting constraints to a nice object A big refactor, but I think it makes things cleaner. 13 July 2016, 21:21:36 UTC
e31fb50 More example code cleanup 13 July 2016, 19:21:13 UTC
7fdf150 Style fixes 13 July 2016, 17:53:11 UTC
e3066f3 Fixing sandwich example 13 July 2016, 17:53:01 UTC
038e8dd Style fixes, docstring tweaks, constraint renaming 21 June 2016, 19:51:35 UTC
d48e4e7 Simple Covariance metric-learner (#21) * Simple Covariance metric-learner * Updated docstring 21 June 2016, 19:27:07 UTC
3c57c64 Fit constraints (#19) * ITML fit_constraints * SDML fit_constraints * LSML fit_constraints * RCA fit_constraints * Renamed fit->fit_constrains in test * Created new supervised classes for methods with semi-supervised constraints * Comment * Removed duplicate code * Super compatible with Python 2 * Code standards * Fix overriding of params 20 June 2016, 18:14:24 UTC
c1372af Remove downloads stats for now (#20) 02 June 2016, 19:02:39 UTC
5475a26 Attempting to simplify travis config 28 May 2016, 01:11:46 UTC
8576941 Bumping docs version [ci skip] 16 May 2016, 18:45:13 UTC
75ac175 Bumping version 16 May 2016, 18:40:04 UTC
f7b422f Minor PEP8 tweaks, fixing return value 16 May 2016, 18:32:53 UTC
b9402a9 Merge pull request #18 from svecon/master Added get_params, set_params and params moved to dictionary 16 May 2016, 18:21:52 UTC
947a66b set_params rewritten to be compatible with python 2 16 May 2016, 13:56:37 UTC
3b2029c Moved params to dictionary for all methods 16 May 2016, 13:10:31 UTC
195884c Comment typo 16 May 2016, 13:01:17 UTC
405f8e1 Added comments for get_params, set_params 16 May 2016, 12:59:37 UTC
3d23367 Reversed setting of params 12 May 2016, 13:04:00 UTC
167a172 LFDA.fit returns self5 12 May 2016, 12:56:51 UTC
1026f26 LFDA params moved to dictionary 12 May 2016, 12:43:56 UTC
a00c585 NCA params moved to dictionary 12 May 2016, 12:22:51 UTC
543aca4 Fixed indentation 12 May 2016, 12:14:25 UTC
3e0be2e get_params, set_params on base_matric 12 May 2016, 11:40:51 UTC
b14d96c Bumping version for Py3k support [ci skip] 07 November 2015, 22:52:19 UTC
2bf0266 Choosing a better seed + adding note 07 November 2015, 22:48:41 UTC
eae4cd5 Fix for Python 3 07 November 2015, 22:46:40 UTC
63372d0 Adding Py3k support + Travis tester 07 November 2015, 22:36:25 UTC
2eda8a3 removing TODO section see issue #13 for the current list 14 October 2015, 19:43:10 UTC
b48e433 Merge pull request #12 from terrytangyuan/patch-1 Added Download Stats Badge 12 October 2015, 16:15:37 UTC
3eef45e Added Download Stats Badge 12 October 2015, 01:17:51 UTC
699dc6d Merge pull request #11 from terrytangyuan/patch-1 Updated version on PyPI 07 October 2015, 18:02:03 UTC
8254591 Updated version on PyPI I've just uploaded the latest version on PyPI so that users can use the newly added algorithms. https://pypi.python.org/pypi?name=metric-learn&version=0.1.1&:action=display 07 October 2015, 02:34:50 UTC
605b4ad Adding a link to the docs [ci skip] 23 September 2015, 00:18:39 UTC
d8726d3 Re-made docs hierarchy using sphinx-apidoc This is a bit of a massive change, sorry! I moved all of the method descriptions to docstrings in the code, then made it so those get pulled into the generated docs. I also merged the intro with the index, so the landing page should be pretty nice to look at. Finally, I started working on converting the docstrings to numpydoc format 23 September 2015, 00:03:28 UTC
7707e6e Merge pull request #9 from terrytangyuan/dev Initial doc page 22 September 2015, 18:51:45 UTC
630b1ba Updated references for doc 22 September 2015, 15:33:06 UTC
3710b64 Added doc for itml 22 September 2015, 15:11:10 UTC
0c37099 Added intro to rca 22 September 2015, 14:55:26 UTC
93f9aea Restructured doc to remove some warnings 22 September 2015, 14:41:45 UTC
b41b676 Updated intro and lmnn doc 22 September 2015, 14:30:18 UTC
e47a39a adding RCA method 22 September 2015, 05:25:29 UTC
11945d1 typo fix [ci skip] 22 September 2015, 05:24:12 UTC
f885031 Added PyPI version badge 22 September 2015, 03:39:04 UTC
a53a6be Changed doc structure 22 September 2015, 03:27:52 UTC
77798c2 Added lmnn doc page 22 September 2015, 03:03:51 UTC
036a514 Added intro page 22 September 2015, 02:44:50 UTC
a402857 Adding LFDA method Iris results seem to match the figure in the original paper. 22 September 2015, 02:42:09 UTC
3d695b1 Added intial doc version 22 September 2015, 02:32:55 UTC
46a0711 Renaming for clarity + use einsum trick 21 September 2015, 19:58:31 UTC
669456f memory improvements for LMNN This should alleviate some of the memory pressure when running on large datasets. 21 September 2015, 17:35:52 UTC
648684b fixing readme typo [ci skip] 17 September 2015, 16:45:26 UTC
c7c8316 Merge pull request #6 from terrytangyuan/patch-1 Added Install from PyPI 17 September 2015, 16:37:16 UTC
764bd69 Added Install from PyPI 17 September 2015, 14:31:23 UTC
dfd9892 fixing travis link 16 September 2015, 21:33:03 UTC
cdb0cc6 updating to reflect the new name + URL 16 September 2015, 21:26:56 UTC
316f8af converting to RST format, renaming to metric-learn 16 September 2015, 21:25:06 UTC
2c21ea7 Merge pull request #5 from terrytangyuan/dev Some minor additions 16 September 2015, 21:13:34 UTC
92ea557 Added one more line before title 16 September 2015, 21:10:49 UTC
6ec24e6 Reformat keywords in setup.py 16 September 2015, 20:57:26 UTC
fbb24f2 Changed format of LICENSE to .txt 16 September 2015, 20:36:21 UTC
239a006 Fixed tests 16 September 2015, 20:32:07 UTC
d3f830e Changed .md to .rst 16 September 2015, 20:25:25 UTC
4506281 More info in setup.py 16 September 2015, 20:22:19 UTC
b501468 Merge pull request #4 from terrytangyuan/dev License, badges, utf-8 coding 16 September 2015, 20:14:13 UTC
6f71652 Added utf-8 unicoding 16 September 2015, 19:59:51 UTC
40b3d72 Travis and License badge and LICENSE document 16 September 2015, 19:43:34 UTC
f7c15ba Use new Travis infra 14 July 2015, 23:33:47 UTC
964a7dc Adding TravisCI support 09 July 2015, 04:45:08 UTC
36b852c Fixing error with older scipy array.dot(sparse) was creating an object array, but sparse.dot(array) correctly makes an array. 01 July 2015, 22:43:11 UTC
1667364 Use logdet for numerical stability 04 April 2015, 16:36:44 UTC
194da3a Adding NCA to methods list, adding TODO 02 April 2015, 17:54:16 UTC
42176c5 Precompute label masks for speed 26 March 2015, 21:46:50 UTC
1b8c099 Adding Neighborhood Components Analysis (NCA) 26 March 2015, 21:06:04 UTC
9f68554 Remove version constraints on numpy/scipy 26 March 2015, 21:05:29 UTC
30d1502 use a fixed seed for SDML test 14 October 2014, 18:02:45 UTC
0958b33 adding SDML.fit docstring 08 September 2014, 22:14:33 UTC
15642e8 seed random, test sparse SDML input 08 September 2014, 22:11:13 UTC
6ce16a0 updating SDML interface 08 September 2014, 21:21:35 UTC
11b36a4 converting LSML 08 September 2014, 21:18:04 UTC
c6853f2 push prior to fit 08 September 2014, 21:14:32 UTC
a304b6a updating tests, returning self 08 September 2014, 21:12:36 UTC
63de27f updating ITML interface 08 September 2014, 21:10:05 UTC
0ec1290 starting migration to sklearn convention 08 September 2014, 21:02:06 UTC
f82bc3d updating README 08 September 2014, 20:39:20 UTC
4fda224 Moving demo to examples/ 08 September 2014, 20:36:00 UTC
e64aa89 Moving tests out, using setuptools 08 September 2014, 20:33:02 UTC
05d6fd5 adding setup.py 08 September 2014, 20:18:49 UTC
7112188 style fixes 08 September 2014, 15:14:32 UTC
90d35e6 fix some LMNN issues 08 September 2014, 15:09:58 UTC
6d4209f use a simpler choice function avoids dependency on numpy 1.7+ 05 November 2013, 20:48:43 UTC
5c2cefe avoiding einsum because it depends on numpy 1.6+ also, it's potentially more opaque to read 05 November 2013, 20:42:47 UTC
82fb591 adding workaround for older numpy versions 05 November 2013, 17:01:05 UTC
d015531 adding usage notes 04 November 2013, 00:39:44 UTC
ac518c7 formatting 02 November 2013, 08:35:52 UTC
b787e5a adding note about shogun 02 November 2013, 08:33:44 UTC
d5c1966 initial commit 02 November 2013, 08:29:23 UTC
back to top