9bc4de8dae
some fixes for sphinx and in examples
2015-03-02 12:50:53 -05:00
feceab6f62
FIX test for consistent handling on empty input data
2015-02-26 10:45:06 +01:00
cd5166e82f
make check_array convert object to float.
...
fix dtype check, add test. unfriend all multi-output estimators on facebook.
try to fix what is happening to y (by doing nothing to y)
make test work...
Make everything accept object y or say "invalid label"
fix multioutput linear models
add test for sensible error message.
2015-02-15 12:13:41 -05:00
cd931fa6ba
Merge pull request #4136 from amueller/test_dtypes_all
...
[MRG+1] More robust input validation, more testing.
2015-02-13 17:13:59 -05:00
8df25493b9
TST Check for equality in alphas_ when uniform targets are used.
2015-02-12 18:56:10 +05:30
3ac2da61bd
FIX Use resolution instead of min from np.finfo(float)
2015-02-12 02:10:05 +05:30
7926262ae5
TST Add enet test for the case of uniform targets
2015-02-11 23:08:45 +05:30
5fb923aacc
FIX Return np.zeros(n_alphas) when the targets are uniform
2015-02-11 22:58:15 +05:30
e3e0827243
FIX check (and enforce) that estimators can accept different dtypes.
2015-02-10 15:37:15 -05:00
54d68d247f
remove inplace multiplication in ridge
2015-02-02 18:37:25 -08:00
965bf51636
use safe in1d, use compute_sample_weight in ridge
2015-02-01 14:28:01 -08:00
e3f94351f2
TST Silence some tests.
2015-01-29 17:24:41 -05:00
59648f337b
Simplify rescaling.
2015-01-25 00:16:45 +09:00
6acbb8ce73
Handle sample_weight upfront.
2015-01-25 00:16:45 +09:00
ff1ec8d963
REFACTOR Using lstsq rather than manual pinv in _solve_cholesky_kernel
2015-01-18 11:01:44 +09:00
f3df83f068
FIX We must not use overwrite_a=True in _solve_cholesky_kernel as K might be reused
2015-01-18 11:01:43 +09:00
c48c9711b3
MISC Let sample_weight default to None
2015-01-18 11:01:43 +09:00
eaaf84d730
REFACTOR Backported changes in _solve_cholesky_kernel to ridge.py
2015-01-18 11:01:43 +09:00
d3eb311304
DOC Added documentation for kernel_ridge module
2015-01-18 11:01:42 +09:00
a413f875f1
DOC minor fixes do docstrings, don't document deprecated parameters.
2015-01-16 11:55:02 -05:00
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
c150608282
FIX Remove unused score_overrides_loss param from check_scoring
2015-01-16 11:54:14 -05:00
c970fb51e7
COSMIT spelling
2015-01-15 15:09:35 -05:00
0bca2ffdfa
FIX Refactor _is_fitted method into check_is_fitted.
...
DOC Make the docstring of NotFittedError class to conform to pep8.
DOC Make the error message of check_is_fitted more newbie friendly.
2015-01-12 00:33:40 +05:30
abd31d2cad
MAINT Make uniform the error raised for not fitted condition
2015-01-12 00:33:40 +05:30
88555987cc
PEP8 Fix E101, E111 errors and W191, W293, W293 and W391 warnings.
2015-01-08 04:28:24 +05:30
3a33776003
FIX : avoid unecessary computation sparse dual gap
2015-01-04 22:18:19 +01:00
4e270a5157
remove deprecation tests.
2014-12-29 14:00:06 -05:00
46c1c42131
remove deprecated ``n_bootstraps`` parameter in Bootstrap cv object, verbose in FactorAnalysis and SelectorMixin. Also adjust some deprecation strings.
2014-12-29 13:53:42 -05:00
f79ac7c187
MAINT: Remove deprecation warnings in enet_path and lasso_path
2014-12-27 00:47:32 +01:00
26901932d7
FIX: Bug in sparse coordinate solver in lazy centering
2014-12-26 22:53:52 +01:00
caf27db052
Small typo fix in the comments: "whther" --> "whether"
2014-12-20 21:28:28 -05:00
69d393af6f
Merge pull request #3931 from dsullivan7/cwwarn
...
[MRG+1] raising warning for class_weight in fit method
2014-12-12 14:57:37 +01:00
6cb191cfd8
changing warning test to use assert_raises_message
2014-12-12 13:27:49 +01:00
ae614649d7
Merge pull request #3954 from amueller/fix_sgd_learningrate_docs
...
DOC explain t0 in the SGD docstring (currently somewhat confusing)
2014-12-11 15:17:32 +01:00
bf1691b69a
'none' is an acceptable value for penalty
2014-12-10 18:33:45 -05:00
9392445767
DOC explain t0 in the docstring (currently somewhat confusing)
2014-12-10 18:32:28 -05:00
5a019b7202
adding simplefilter for warning
2014-12-10 16:07:46 +01:00
081a554698
FIX P/R/F metrics and scorers are now for binary problems only by default
...
Scorers for different average parameters have been added.
2014-12-09 13:56:51 +11:00
b99824fb46
changing warn message, making it Deprecation Warning, and removing negative index for py2.7
2014-12-05 09:39:39 +01:00
afe890b2a8
adding warning if passing class_weight through fit
2014-12-04 13:17:34 +01:00
8dd7f16bef
adding support for class_weight in fit method
2014-12-03 17:15:49 +01:00
7c50e6c8eb
various docstring fixes for web docs
2014-11-30 17:59:42 -08:00
e7bab895b7
Merge pull request #3823 from ragv/omp-3644
...
[MRG+1] Fixes #3644
2014-11-26 11:22:25 +01:00
e4f62b1f17
Added NRT for return_path propagation check
2014-11-25 23:24:54 +05:30
4a8c7a789c
Fixes #3644
2014-11-25 23:24:54 +05:30
37f11953d0
TST update numerical overflow (non-regression) tests
2014-11-25 17:55:33 +01:00
d0eb09b4b5
FIX clip SGD gradient on linear models for stability
...
The l2 weight decay rescaling is also kept positive (or null)
in case of strong regularization.
2014-11-25 13:33:36 +01:00
f0fe4afd1e
Merge pull request #2949 from FlorianWilhelm/theilsen
...
[MRG+1] TheilSen robust linear regression
2014-11-20 15:34:01 +01:00
429d06493a
Merge pull request #3853 from ogrisel/backport-lsqr-fix
...
[MRG] Fix ZeroDivisionError in LinearRegression on sparse data
2014-11-15 17:33:46 +01:00
Andreas Mueller