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
c66689c4f4
FIX Don't compare arrays to strings!!!!
2015-10-15 16:07:17 -04:00
21fab6e8ec
FIX Move validation from helper to main function
2015-10-15 18:21:51 +02:00
4f4c3598e6
FIX dtypes to conform to the stricter type cast rules of numpy 1.10
...
FIX set copy to (copy & whiten).
FIX/DOC Use float outputs for doctest
2015-10-15 18:21:51 +02:00
af56154a77
Merge pull request #5238 from rvraghav93/cvdoc
...
[MRG+1] DOC Make cv documentation consistent across our codebase
2015-09-13 13:29:27 -04:00
6dd6f8ffe8
minor fixes to the doc build
2015-09-11 17:32:51 -04:00
882c346abd
DOC Make cv documentation consistent across our codebase
2015-09-10 09:15:13 +05:30
53324a1677
docstrings updated with considerations regarding small alphas when positive=True.
2015-08-28 13:12:52 +09:00
db1f67bd51
pep-8 fix
2015-08-26 01:53:29 +09:00
f157606a32
positive option passed on to LarsCV estimator
2015-08-25 15:14:41 +09:00
7e16f9d12b
fit_intercept docstring comment removed from lars_path method
2015-08-25 14:34:59 +09:00
535536e537
merge with upstream/master
2015-08-25 14:07:56 +09:00
6c1c6eca36
agramfort code review input
2015-08-25 12:45:19 +09:00
c369af4e81
doctest fix.
2015-08-18 15:05:21 +09:00
7f1bf0de45
positivity constrained can now also be used for the CV estimators.
2015-08-18 14:30:11 +09:00
b2cea17053
cleanup
2015-08-18 13:14:41 +09:00
e3f46cbff4
docstrings added + fit_intercept reintroduced.
2015-08-18 13:10:51 +09:00
d6908a971e
todos for pull request
2015-08-17 13:59:03 +09:00
ceb066d29a
cleanup
2015-08-17 13:45:02 +09:00
e50aaba7dc
"nonnegative" -> "positive" + cleanup
2015-08-17 13:25:14 +09:00
8e3b22ccdc
lars_path method complemented with nonnegative option.
2015-08-17 13:25:14 +09:00
99eff75e8d
fix TheilSenRegressor input validation, LassoLarsIC input validation
2015-07-31 10:34:39 -04:00
62325cbb58
MAINT merge _check_cv into check_cv as indices argument is removed in 0.17
2015-06-08 16:37:03 +05:30
0650d5502e
DOC adding backlinks to docstrings
2015-06-03 00:24:04 -04:00
dd9641f707
FIX #4597 : LassoLarsCV on with readonly folds
2015-05-07 15:43:37 +01:00
09e6e12f7b
fix LarsCV and LassoLarsCV fails for numpy 1.8.0
2015-03-19 20:47:54 -07:00
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
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
88555987cc
PEP8 Fix E101, E111 errors and W191, W293, W293 and W391 warnings.
2015-01-08 04:28:24 +05:30
7c50e6c8eb
various docstring fixes for web docs
2014-11-30 17:59:42 -08:00
90e5b2a817
DOC: Document criterion_ attribute in LassoLarsIC
2014-10-07 17:13:25 +02:00
78fbd2576f
ENH comment on drop for good not being triggered in the tests
2014-09-11 13:58:24 -07:00
d9067fa295
FIX #3370 : better lars alpha path inequality checks for 32 bit support
2014-09-11 13:58:23 -07:00
cb1798ad1f
FIX more tolerant early stopping for least angle
2014-09-04 19:17:51 +02:00
a228e8a49b
FIX division by zero warning in LassoLarsIC
...
Fixes #3576 . First time I named a variable after myself.
2014-09-04 15:52:47 +02:00
ff948965fa
FIX finfo.eps should be used instead of tiny for equality checks
2014-09-04 10:06:47 +02:00
dbfb25e51e
FIX heisenfailure in test_lasso_lars_path_length
...
Tentative fix for #3370 : random failure under Python 32 bit and
Windows.
2014-09-03 16:17:46 +02:00
b3bdb08964
DOC fix formatting of attributes etc. in docstrings
2014-07-28 19:04:59 +10:00
c508965201
MAINT: Fixed a few docstrings and cosmits
2014-07-22 15:09:39 +02:00
d72ca5974f
Added n_iter attribute to OMP (and CV)
2014-07-22 15:09:39 +02:00
c838c382b9
FIX: Preserve public API by introducing return_n_iter param
2014-07-22 15:09:39 +02:00
bbb140aaf0
ENH: Added n_iter_ parameters across all iterative solvers
2014-07-22 15:09:39 +02:00
6e2a83b4e1
remove check_arrays stuff and old input validation
2014-07-20 13:31:45 +02:00
cc36ba1b15
Cosmetic np.abs -> abs
2014-07-14 18:21:00 +02:00
0295fdae71
FIX: check with tolerance on lars_path
...
Hopefully this will fix issue #3370 but I haven't checked.
2014-07-14 18:21:00 +02:00
9c80d11efc
TST: relax warning class checked in lars
2014-04-02 13:47:41 +02:00
092b50e44b
BUG: avoid NaNs in arrays passed to scipy.linalg
...
Initializing with "np.empty" creates NaNs every once in a while. Scipy
doesn't like it as of scipy 0.10: we cannot turn of "check_finite"
in this version.
These NaNs are harmless because they are in the upper diagonal, while the
scipy solve_triangular uses only the lower.
2014-03-05 09:59:27 +01:00
1b67fe33e5
MAINT: scipy 0.10 and 0.11
...
linalg.solve_triangular does not have a check_finite keyword argument
2014-03-03 23:51:38 +01:00
dc27992c1a
MAINT: remove our solve_triangular
...
No longer needed, since we depend on scipy >= 0.9
2014-03-03 20:25:33 +01:00
b92b7b4856
handle the case where length of idx in least_angle greater than one
2014-01-31 11:13:28 -05:00
Raghav R V