Mathieu Blondel
|
91fdb160d3
|
Fix problem when output is a vector.
|
2011-01-20 21:12:44 +09:00 |
Mathieu Blondel
|
678458eaca
|
Add dense_output option to safe_sparse_dot.
|
2011-01-20 20:43:33 +09:00 |
Fabian Pedregosa
|
ae5561f7ea
|
FIX: scipy 0.9 compatibility fixes
fixes for linalg.qr and sparse.linalg.eigen.
Also: import scipy.linalg lazily whenever possible.
|
2010-12-16 22:05:07 +01:00 |
Olivier Grisel
|
7b68216491
|
s/_sparsedot/safe_sparse_dot/g
|
2010-12-06 02:16:54 +01:00 |
Olivier Grisel
|
f3003230f5
|
switching back to scipy.linalg.qr with econ=True to avoid half-installed numpy issues with wrong lapack bindings
|
2010-12-01 11:45:19 +01:00 |
Olivier Grisel
|
cdeee0d448
|
OPTIM: fast_svd now has a auto transpose mode that switch to the fastest impl
|
2010-12-01 03:22:15 +01:00 |
Olivier Grisel
|
e61c7df1e5
|
OPTIM: do not allocate a (n_samples, n_samples) temporary array with scipy.linalg.qr when (n_samples, k + p)) is all what is needed
|
2010-12-01 01:42:51 +01:00 |
Olivier Grisel
|
078bee9e5d
|
more tests for the iterated power refinement of the Martinsson randomized SVD
|
2010-11-29 02:14:04 +01:00 |
Olivier Grisel
|
531f25f96c
|
more documentation on fast SVD + missing reference
|
2010-11-28 18:55:06 +01:00 |
Alexandre Passos
|
29480da44d
|
Merging the rng changes
|
2010-11-27 19:32:43 -02:00 |
Alexandre Passos
|
3976608734
|
Adding the power iteration parameter to fast_svd (to make it better in high-rank very-big very-sparse matrices according to the Martinsson et al survey
|
2010-11-27 19:18:50 -02:00 |
Olivier Grisel
|
6a5d4c28a5
|
make fast_svd deteriministc by default while allowing to pass rng seeds
|
2010-11-27 22:06:05 +01:00 |
Olivier Grisel
|
8fa07c3971
|
drop the warning since useful even if approximate as demoed in the faces example
|
2010-11-27 21:54:34 +01:00 |
Olivier Grisel
|
2d0ce097cc
|
FIX: actually truncate the SVD to make it faster + add some test
|
2010-11-27 21:16:57 +01:00 |
Olivier Grisel
|
4b310a6807
|
cosmit
|
2010-11-22 12:09:33 +01:00 |
Gael Varoquaux
|
8cf803d463
|
MISC: Cosmit+Doc in fast truncated PCA
|
2010-11-22 11:41:04 +01:00 |
Alexandre Passos
|
b2b1e9c7e0
|
Adding random projections SVD to scikits.learn.pca as an option
|
2010-11-22 11:24:27 +01:00 |
Fabian Pedregosa
|
ba8e530746
|
Correct typo in docstring.
|
2010-10-14 16:21:31 +02:00 |
Gael varoquaux
|
d1a9d8e508
|
BUG: Fix bug with numpy >= 1.5 introduced by my previous (stupid) commit.
|
2010-08-30 22:49:20 +02:00 |
Gael varoquaux
|
1d645ea9b3
|
ENH: For numpy >= 1.5, use np.linalg.slogdet as a fast_logdet
|
2010-08-30 22:34:26 +02:00 |
Fabian Pedregosa
|
b3b1fee67c
|
Refactoring: squash glm module in a single file.
No behaviour or API changes were made.
|
2010-06-07 09:20:32 +02:00 |
ogrisel
|
8ac9d24fda
|
fix name conflict issue with stdlib math module
|
2010-05-25 14:39:11 +02:00 |