13a83155eb
FIX sparse matrix indexing in BernoulliRBM
...
Fixes #2545 , reported by @jfelectron.
2013-10-27 17:51:33 +01:00
e24e391dcf
FIX don't raise file-level SkipTests
2013-10-20 14:00:58 -07:00
7e7dc78c6b
don't import tests
2013-10-20 13:54:08 -07:00
e42f211435
ENH make sure all "SkipTest" calls have an error message.
2013-10-20 13:43:34 -07:00
311172e31a
TST fix still broken test_fast_dot
...
On Travis, SciPy seems to raise AttributeError instead of ValueError.
2013-10-20 20:45:57 +02:00
c14a619de6
TST fix broken fast_dot test
...
Due to a misspelling ("get_blas" vs. "get_blas_funcs") and a bare except
clause, the tests weren't actually being run.
2013-10-20 19:17:26 +02:00
34a9a1a5ca
Remove deprecated ransac implementation
2013-10-20 15:57:57 +02:00
c74b1f8c1e
Move RANSAC implementation to linea_model subpackage
2013-10-20 15:57:55 +02:00
f6b2608615
Add empty line between parameter and return value description
2013-10-20 15:57:55 +02:00
44d98b32cd
Fix estimator parameter description
2013-10-20 15:57:55 +02:00
223edc3a51
Add some reference papers
2013-10-20 15:57:55 +02:00
fabf192b66
Move algorithm description to notes section
2013-10-20 15:57:55 +02:00
5b46b598ec
Skip iteration for empty inlier sample set
2013-10-20 15:57:55 +02:00
5ce892fda2
Add missing information predict method of estimator
2013-10-20 15:57:55 +02:00
755d90cff7
Remove unused variable
2013-10-20 15:57:55 +02:00
5c5953f6f1
Set numpy random seed for all test functions
2013-10-20 15:57:55 +02:00
99d2624f64
Move ransac.py to _ransac.py to avoid nosetest namespace conflict
2013-10-20 15:57:55 +02:00
cec4ed78fc
Add test for stop_score parameter
2013-10-20 15:57:55 +02:00
cddbf5562f
Add test for stop_n_inliers parameter
2013-10-20 15:57:54 +02:00
1f3e7fbe24
Add n_trials return value to doc string
2013-10-20 15:57:54 +02:00
d7fdb14ca6
Fix n_trials return value and add additional test
2013-10-20 15:57:54 +02:00
354fe5636c
Return number of trials in RANSAC
2013-10-20 15:57:54 +02:00
623e8fe5ff
Add unit tests for RANSAC max_trials option
2013-10-20 15:57:54 +02:00
87ad0da318
Add unit tests for RANSAC is_model_valid and is_data_valid
2013-10-20 15:57:54 +02:00
d0d10749c7
Add simple RANSAC unit test
2013-10-20 15:57:54 +02:00
26818ff895
Add parameter description to RANSAC and modify API
2013-10-20 15:57:54 +02:00
a971649537
Add step-by-step description of RANSAC algorithm
2013-10-20 15:57:54 +02:00
63137a4479
Fix bug in residual determination
2013-10-20 15:57:54 +02:00
d2c5f98dd4
Add ransac function to __init__.py of utils package
2013-10-20 15:57:53 +02:00
daca76320a
Add implementation of RANSAC algorithm
2013-10-20 15:57:53 +02:00
551b6c2f06
ENH honor Y_norm_squared when X=Y in euclidean_distances
...
Also a micro-optimization to row_norms.
2013-10-13 18:54:21 +02:00
24af95a211
ENH refactor squared-norms computation to extmath
...
k-means can now use the memory-efficient dense implementation from
metrics.pairwise, while pairwise can use the fast sparse implementation
from k-means.
2013-10-12 20:27:51 +02:00
edf747e6a2
DOC docstring for extmath.norm
2013-10-12 20:17:33 +02:00
ef108b0b21
MAINT add authors to validation.py and pairwise.py
2013-10-12 20:09:41 +02:00
c1df89edf2
... add the fix to import check
2013-10-08 23:01:48 +02:00
add591cbf7
FIX scipy API
2013-10-08 22:54:14 +02:00
aa2d0452e4
Merge pull request #2426 from larsmans/sgd-improvements
...
[MRG] SGD Cython improvements
2013-09-21 08:39:07 -07:00
6454b5f0f7
DOC: add docstring to _logaddexp
2013-09-11 22:00:31 +02:00
f8bf44a251
FIX: logaddexp(-inf, -inf) == -inf and not NaN
2013-09-11 21:31:46 +02:00
d0c981fa1d
MAINT deprecate indices=False for cross validation generators
2013-09-10 10:58:24 +02:00
b8e1a6c357
ENH SGD Cython improvements
...
* nogil declarations where possible (but no with nogil blocks, since the
performance benefit of those is not clear)
* use BLAS where appropriate
* replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's
the type used by scipy.sparse
* explicit checks that dataset dims don't exceed INT_MAX
* replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 18:24:36 +02:00
ceffed1d6e
MAINT remove useless deprecation in sklearn.utils
...
This module is off-limits to users anyway.
2013-09-04 18:43:56 +02:00
2fb55da670
BUG duplicate finity check in input validation
2013-09-04 18:36:40 +02:00
b0de2e3f5b
fix visual indentation
2013-08-31 12:28:28 +05:30
fcce3032c0
fix visual indentation
2013-08-31 12:27:16 +05:30
7fac2038dc
Small fixes
2013-08-28 09:20:43 +02:00
1cd73c645d
Added gen_batches() to utils
...
This generator is similar to gen_even_slices(), but takes a slice length
as input instead of a number of slices.
2013-08-27 10:29:40 +02:00
e0011e4bd6
[BUG] safe_asarray() converts sparse matrices dtype
...
When called with a dtype kw argument, utils.safe_asarray() would
not change the dtype of a sparse matrix. This has been fixed.
Unit tests have been created for safe_asarray().
2013-08-27 10:28:27 +02:00
2eddb1a29f
BUG fix whitespace in error messages
...
Partial fix for #2380 .
Signed-off-by: Lars Buitinck <L.J.Buitinck@uva.nl>
2013-08-26 17:02:52 +02:00
6dcfe690a7
MAINT: remove utf-8 headers
...
Without encoding headers, Python imports will fail if UTF8 characters
creep in. Thus, they will not go unnoticed.
2013-08-26 14:55:51 +02:00
Lars Buitinck