2e4aafd19c
MAINT remove my email address
...
I'm getting too many individual support requests. People should
use the mailing list.
2016-03-04 10:43:41 +01:00
a6604c3c1d
inverse_transformation for NMF decomposition
2016-02-11 23:30:10 +05:30
625471bb29
DOC: few minor typos spotted
2015-12-15 20:56:34 -05:00
fb123ed24b
More doc fixes. Latex builds again.
2015-11-20 16:30:45 -05:00
f2e35411fa
Added more versions of 0.17
2015-11-04 00:02:52 +01:00
2249daaea8
Version added for all new classes
2015-11-03 23:54:42 +01:00
1eb9bed151
COSMIT nicer output from NMF with verbose > 0
...
Used to print "('violation:', 1.0)". Now "violation: 1.0".
2015-10-21 13:14:56 +02:00
e3afc0e8c9
MAINT move custom error/warning classes into sklearn.exceptions
...
ENH NonBLASDotWarning -> EfficiencyWarning; Improve error message
DOC Add exceptions module to modules/classes.rst
MAINT Move ConvergenceWarning, UndefinedMetricWarning et al into exceptions
MAINT Remove ChangedBehaviorWarning from base
DOC/FIX Improve DataConversionWarning's docstring
2015-10-19 22:35:35 +02:00
46edeec111
COSMIT some refactoring in NMF
...
* de-indent loop
* move regularization calculations out of loop
2015-10-02 00:07:01 +02:00
e7ecc4f976
BUG shuffle components, not samples in NMF CD
2015-09-30 21:45:03 +02:00
ceeef70037
ENH refactor NMF and add CD solver
2015-09-21 11:42:07 +02:00
a6c6e732c5
Merge pull request #3659 from chyikwei/onlineldavb
...
ENH Latent Dirichlet Allocation (LDA) with online variational Bayes
2015-08-09 19:13:01 +02:00
71feb53bf6
FIX seed the initialization of NMF
2015-07-23 08:46:14 +02:00
a1d6bb1e2f
DOC Fix doc for NMF `init` default
...
Docs had claimed that the `init` argument defaulted to "nndsvdar" when `n_components < n_features`, but it actually defaults to "nndsvd".
Resolves issue #4929 .
2015-07-11 10:47:29 -05:00
0650d5502e
DOC adding backlinks to docstrings
2015-06-03 00:24:04 -04:00
6b16e9de61
move check_non_negative to utils/validation.py
2015-05-15 23:01:12 -04:00
b3fbccca38
FIX various mismatch between docstring and signature params
...
DOC max_iterations -> max_iter. Make it consistent with kmeans
MAINT Replace the deprecated dx parameter with d in the docstrings
MAINT Deprecation warning for max_iterations parameter.
2015-01-16 11:54:57 -05:00
abd31d2cad
MAINT Make uniform the error raised for not fitted condition
2015-01-12 00:33:40 +05:30
b3bdb08964
DOC fix formatting of attributes etc. in docstrings
2014-07-28 19:04:59 +10:00
bbb140aaf0
ENH: Added n_iter_ parameters across all iterative solvers
2014-07-22 15:09:39 +02:00
c235c3b566
ENH add allowed_sparse named argument for @ogrisel
2014-07-20 15:31:28 +02:00
6e2a83b4e1
remove check_arrays stuff and old input validation
2014-07-20 13:31:45 +02:00
e6e9fb6490
Make fit_transform and fit().transform() equivalent in nmf
2014-07-18 17:22:02 +02:00
c7db70ac9f
removed unused variables and imports, add used imports where missing.
2014-06-08 21:44:01 +02:00
fb795ebef0
ENH factor out squared norm helper
2014-04-08 23:02:54 +02:00
0abe21eaac
MAINT remove some deprecated stuff
2014-03-12 13:33:42 +01:00
abe3d05b4c
COSMIT micro-optimize norm computation in NMF
...
... for clarity, although it should also be slightly faster, and
corrected a docstring while I was there.
2013-10-30 23:31:13 +01:00
02d972644a
ENH micro-optimize NMF inner loop
...
Also fixed the random seed on the example.
2013-10-20 19:17:26 +02:00
a2a4257bc6
ENH optimize NMF inner loop
...
The call to np.maximum was taking 20% of the time.
Multiplying by Hn > 0 is about twice as fast.
This reverts commit 6ef08cfac2
2013-10-01 18:09:58 +02:00
6ef08cfac2
FIX failing test in NMF due to negative zeros
2013-09-02 14:03:06 +02:00
841fb88727
ENH prettier output from NMF example
...
Better stopword and noise filtering, increased n_samples to 2000
since the recent optimizations to tf-idf and NMF allow this.
Also some micro-optimizations to NMF itself.
2013-09-01 19:04:56 +02:00
c64163f822
wip
2013-09-01 16:47:04 +02:00
41b3ddd2d6
ENH speed up NMF (about 30% off topic extraction runtime)
2013-08-30 03:19:28 +02:00
7aac367666
Use scipy.nnls.optimize for dense data
2013-07-27 16:12:19 +02:00
ae74c9df86
Use same parameters when solving the transform
2013-07-27 16:12:19 +02:00
206d22793d
Use projected gradient solver in transform to support sparse matrices
2013-07-27 16:12:19 +02:00
77695ee27a
COSMIT pep8
2013-07-11 11:50:23 +02:00
8861833ec9
DOC spellfixes
2013-06-29 16:18:23 +03:00
2ffb484adb
COSMIT pep8
2013-05-04 10:43:39 +02:00
dacfd8bd5d
DOC: Replaced all BSD style licenses with "BSD 3 clause"
...
Replaced all BSD style licenses with "BSD 3 clause"
Not checked yet!
Removed duplicate "3 clause, 3 clause"
Removed trailing period if exists
Fixed some missed licences, still about 50 to do, but those can be automated
Think I got the last of them.
Apparently me and sed have different ideas of regex.
Found a few more
2013-04-30 08:34:46 +02:00
9cf9e9f335
Fixx doctest.
2013-04-10 00:32:16 +09:00
8a473b6e12
BUG: fix replacement for _neg.
2013-04-09 23:37:21 +09:00
d3bc5ac1ba
Demystify magic values in NNLS implementation.
2013-04-09 23:25:50 +09:00
5ece0e12c6
Cosmit: more idiomatic way of clipping to zero.
...
The test also runs 300 milliseconds faster on my machine (1.3s instead
of 1.6).
2013-04-09 23:07:04 +09:00
bb7da14726
COSMIT typo
2013-03-10 23:05:21 +01:00
01a745eb96
DOC FIX column span alignment problem in NMF ^^
2013-03-10 19:24:58 +01:00
21ba76d4a4
DOC some sphinx / rst fixes
2013-03-10 18:45:23 +01:00
a3b55f4eae
ENH fix test_estimators_overwrite_params to also test regressors and transformers. Then fix all the regressors and transformers ... meh!
2013-02-11 17:13:25 +01:00
1967a0b323
P3K use six to have a python 2 & 3 compatible code base
2013-02-09 18:08:44 +01:00
9f1c346c66
FIX: variable naming inconsistency in NMF
2013-02-09 11:41:28 +00:00
Lars Buitinck