8ff339eb5b
Update test_utils.py
...
Comment instead of docstring to follow sklearn convention.
Regression test made easier by adding small value to main diagonal of s.p.d. matrix
2015-10-23 08:57:31 +02:00
6cb51b2a28
Initialize ARPACK eigsh
...
`v0 = random_state.rand(M.shape[0])` leads to an initial residual vector in ARPACK which is all positive. However, this is not the absolute or squared residual, but a true difference. Thus, it is better to initialize with `v0=random_state.uniform(-1, 1, M.shape[0])` to have an equally distributed sign. This is the way that ARPACK initializes the residuals.
The effect of the previous initialization is that eigsh frequently does not converge to the correct eigenvalues, e.g. negative eigenvalues for s.p.d. matrix, which leads to an incorrect null-space.
- initialized all occurences of sklearn.utils.arpack.eigsh the same way it would be initialzed by ARPACK
- regression test to test behavior of new initialization
2015-10-23 08:57:30 +02:00
1c7168d398
Fix fit_transform, stability issue and scale issue in PLS
2015-10-22 14:25:59 +02:00
744d161d37
Merge pull request #5431 from hlin117/nan-targets
...
[MRG + 2] Add check to regression models to raise error when targets are NaN
2015-10-22 09:36:12 +02:00
f725485427
#5322 : Small change to estimator_check.py
2015-10-21 11:59:21 -05:00
1f48814799
#5322 : Using multioutput_estimator_convert_y_2d in check_supervised_y_no_nan
2015-10-21 11:54:12 -05:00
3c988d5d1c
Merge pull request #5512 from giorgiop/randomized-svd-versionadded
...
DOC versionadded randomized_svd
2015-10-21 18:45:24 +02:00
f45f260ff7
DOC versionadded randomized_svd
2015-10-21 18:28:37 +02:00
290e0ea6f2
#5322 : Changed check_supervised_y_no_nan to not seed global
2015-10-21 11:26:20 -05:00
c45075123a
Added documentation about y nan handling in check_X_y
2015-10-21 11:11:08 -04:00
0cb93b0996
Merge pull request #5141 from giorgiop/power-iter-randomized-svd
...
[MRG + 2] ENH: optimizing power iterations phase for randomized_svd
2015-10-21 13:01:45 +02:00
b18f2951e4
randomized_svd: power iter, normalization, benchmark
2015-10-21 11:21:40 +02:00
62633f701b
Addressed comments on PR #5451
2015-10-21 10:24:23 +02:00
5db2adf93c
MAINT Removed deprecated stuff.
2015-10-21 10:24:23 +02:00
52a6acc15c
#5322 : Addressing @MechCoder's changes
2015-10-21 00:04:42 -05:00
3bfbe79fa0
#5322 : Making check_supervised_y_no_nan deterministic
2015-10-20 14:26:21 -05:00
eea5ba3d7f
#5322 : Made error message in check_supervised_y_no_nan more helpful
2015-10-20 10:36:48 -05:00
c657d77c3c
#5322 : Regression test for test_check_estimator
2015-10-20 10:36:48 -05:00
011dcd2c04
#5322 : Fixed python3 build
2015-10-20 10:36:48 -05:00
ef4faf69ca
#5322 : check_supervised_y_no_nan raises error when error message is incorrect
2015-10-20 10:36:48 -05:00
7f820269b8
#5322 : Fixing test in test_estimator_checks.test_check_estimator
2015-10-20 10:36:47 -05:00
a5860a1df2
#5322 : Fixed build for tree failures, fixing estimator_checks.check_estimator
2015-10-20 10:36:47 -05:00
b1a1aa8395
#5322 : Resolving errors for SVR and NuSVR
2015-10-20 10:36:47 -05:00
7745319e33
#5322 : Added check for DecisionTreeRegressor
2015-10-20 10:34:56 -05:00
92068ef7de
#5322 : Wrote check_supervised_y_no_nan in estimator_checks
2015-10-20 10:31:24 -05:00
857cb09254
FIX/ENH Make the moved class resilient to the deprecation patching
2015-10-20 01:24:35 +02:00
61ae8d56c4
MAINT move deprecated into deprecation.py
2015-10-19 22:35:36 +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
03718f80fa
FIX Remove BinaryGaussianProcessClassifierLaplace from packages __all__ list
2015-10-19 18:07:13 +02:00
d42aa19d02
TST GaussianProcessRegressor added to MULTI_OUTPUT estimators
2015-10-19 18:07:13 +02:00
de3889d3ad
TST Fixing some GP-releated issues in estimator checks
2015-10-19 18:07:13 +02:00
008adf11ad
Merge pull request #5375 from rvraghav93/set_precision
...
[MRG + 2] FIX precision to float64 across the codebase
2015-10-19 11:02:04 +02:00
0d26d54df5
Merge pull request #5423 from MechCoder/example_fix
...
[MRG + 1] Fix broken examples using RandomTreeEmbeddings
2015-10-17 09:24:04 +02:00
c31de116f9
Fix broken examples using RandomTreeEmbeddings
...
RandomTreeEmbeddings overwrites the fit_transform method of
_LearntSelectorMixin
2015-10-16 12:44:04 -04:00
a87011262a
FIX precision to float64 across the codebase
2015-10-16 11:20:29 +02:00
79fa25ade3
MAINT don't print things in testing.
2015-10-15 15:23:00 -04:00
6fe94b57f2
Merge pull request #4478 from amueller/fix_randomized_svd_transpose
...
[MRG] Fix randomized_svd transpose heuristic.
2015-10-15 13:39:46 +02:00
85223b9bcc
a common test to check if classifiers fail when fed regression targets
2015-10-14 14:46:13 -04:00
838570cc8a
Fixes #4455 .
2015-10-13 17:44:36 -04:00
6a5a2f7960
partial_fit for scalers
2015-10-13 11:37:00 +02:00
8d273a1e49
Merge pull request #5234 from vighneshbirodkar/mcd_fix
...
[MRG + 1]Deprecating 1D inputs in fast_mcd
2015-10-12 13:33:54 +02:00
652b950074
Merge pull request #4242 from MechCoder/select_from_model
...
[MRG+1] Implemented SelectFromModel meta-transformer
2015-10-10 23:15:06 -04:00
c805fbc79b
Refactor tests
2015-10-10 22:05:40 -04:00
f0121b7baa
removed reshaping in fast_mcd and replaced by check_array
...
Added fast_mcd 1D test
fixed syntax error cause py34 was failing
raise value error in check_array when min_samples>=2
formatting
added checl_array to MCD class and a test for the class
pep8 formatting
fixed string formatting and modified tests
2015-10-09 18:41:59 -04:00
5a0db1717a
1. Added parameter prefit to pass in a fitted estimator.
...
2. Use assert_warns instead of catch_warnings
3. Remove depracation warnings in common tests.
2015-10-09 01:57:36 -04:00
f60ba72d7b
Increase length of array indexing type in ArrayDataset
...
int (32-bit on all platforms) will loop around once the number of
elements exceed 2^31-1. While ints are used elsewhere for row index,
this is particularly significant here, since this int is used as an
index into a dense matrix. i.e. as long as #rows * #cols is greater
than 2^31, this will fail. This is probably the root cause of #2393 .
2015-10-08 14:56:05 -07:00
735dd1fdca
Fix warnings during tests
2015-10-08 14:52:21 -04:00
2ee718cc75
Add narrative docs and fix examples
2015-10-05 11:25:06 -04:00
8a28ea8169
fix test failures
2015-10-05 11:19:57 -04:00
c438f78996
Implemented SelectFromModel meta-transformer
2015-10-05 11:19:57 -04:00
Peter Fischer