5db2adf93c
MAINT Removed deprecated stuff.
2015-10-21 10:24:23 +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
51f01d6bc8
FIX cross_validation and pairwise_kernels treat GP kernels as special case
2015-10-19 18:07:11 +02:00
a4e4cf9875
FIX: remove shuffling in LabelKFold
2015-10-19 16:27:04 +02:00
1f4bc26a02
FIX: Ensure correct LabelKFold folds when shuffle=True
2015-09-25 13:26:34 +02:00
882c346abd
DOC Make cv documentation consistent across our codebase
2015-09-10 09:15:13 +05:30
410e7f2c2a
ENH: remove unnecessary assignments
2015-09-08 07:52:12 +02:00
c769a65e15
COSMIT: variable names, documentation, etc
2015-09-07 10:08:17 +02:00
cf7bc7b035
ENH: rename to LabelKFold
...
COSMIT: doc, pep8, etc
Refactor code
2015-09-07 09:56:19 +02:00
6eb4028519
Added subject independent KFold
...
Changed SubjectIndependentKFold to DisjointGroupKFold
cosmetic changes test (fix seed correctly, use assert_equal for
meaningful error messages)
Changed name to DisjointLabelKFold
Added example of use
FIX: whitespace related doctest failure
FIX: Python 2.6 requires the field numbers in print
FIX: change docstring to comment in test function
DOC: moved docstring from function to class
FIX: added call to parent class
FIX: error in calling the parent
DOC: fixed doctest
FIX: doctest
Cosmetic changes (minor refactoring)
Optimised code (use np.bincount)
Cosmetic: use samples instead of weight for clarity
Minor fix: removed shuffle parameter
Cosmetic
Use mergesort instead of quicksort for reproducibility.
Changed variable name 'y' to 'label'.
Added test for degenerate case where n_folds > n_labels.
Documented the requirement n_labels > n_folds.
DOC: improved description + added see also sections.
Fixed dtype of temporary arrays.
Improved test: check that one label is not in both test and training.
Added documentation for DisjoinLabelKFold.
2015-09-07 09:53:58 +02:00
e2e5fbd8e1
Merge pull request #5201 from christophebourguignat/master
...
DOC Clarify random_state parameter in KFold() doc
2015-09-02 07:16:11 +02:00
b10e78edbf
[DOC] Precision on random_state in KFold() doc
...
Without paying attention, we could think KFold(30, random_state=123)
and KFold(30, random_state=124) generated two different versions
2015-09-01 20:12:07 +02:00
364825ae67
Merge pull request #5194 from christophebourguignat/master
...
[MRG+1] DOC Updated documentation for cv parameter
2015-09-01 09:51:58 +02:00
53c3530ba3
DOC Updated documentation for cv parameter (issue #4533 )
...
Added « optional »
2015-08-31 23:18:15 +02:00
1991c43923
DOC Updated documentation for cv parameter (issue #4533 )
2015-08-31 23:09:10 +02:00
21a966ae8a
ENH: added LabelShuffleSplit cv iterator
2015-08-30 12:00:18 +02:00
ea42c55b7b
PEP8 in cross_validation module + tests
2015-08-27 15:34:41 +02:00
26d3323242
Remove redundant p variable
2015-08-26 16:15:02 +05:30
e35c90ceeb
Use Inverted locations to reorder the predictions
2015-08-26 14:27:44 +05:30
48adf8be7d
Use a single vstack for concatenating all blocks in prediction matrix
2015-08-26 08:07:15 +05:30
aa2d6bfb5f
Add check for sparse prediction in cross_val_predict
2015-08-26 07:37:12 +05:30
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
c5b395fb3e
FIX / TST make cross_val_predict work on lists, test pass-through and compatibility with various input types.
2015-05-19 15:04:45 -04:00
8fff05c3e8
Merge pull request #4438 from mbatchkarov/stratified_train_test_split
...
[MRG + 2] Stratified `train_test_split`
2015-05-09 17:31:14 -04:00
52244dc6df
incorporated amueller's feedback
2015-04-24 10:08:19 +01:00
00409d746b
DOC use standart X, y notation
2015-04-15 11:37:27 +02:00
6cd1cae830
using StratifiedShuffleSplit instead of StratifiedKFold
...
this resolves API incompatibility in 803dcc10a2
2015-03-24 09:07:03 +00:00
803dcc10a2
fixed issue when train_size < test_size for stratified
2015-03-23 19:44:51 +00:00
34c85ed995
[WIP] preliminary implementation of stratified train_test_split
...
Broken when train_size < test_size
2015-03-23 19:19:40 +00:00
bf5fd93f47
MAINT Remove the deprecated n_iterations from StratifiedShuffleSplit
2015-03-19 05:13:09 +05:30
88759ba187
MAINT remove the deprecated n_iterations param
2015-03-19 04:14:24 +05:30
e1af92b640
remove cross_validation.Bootstrap
2015-03-18 14:51:37 -04:00
6beacc3fb6
remove deprecated stuff from 0.17
2015-03-18 14:49:04 -04:00
67dcd99d82
TST fix tests with numpy 1.6.1
...
np.bincount raises an Exception with empty input arrays for numpy versions < 1.6.2
Add utils.fixes.bincount to tackle this issue and use it instead of np.bincount
2015-02-01 15:49:49 +01:00
17726d2481
Merge pull request #4141 from amueller/bootstrap_int_one_fix
...
[FIX] Bootstrap with test_size=1 fix
2015-01-24 15:24:18 +01:00
5c5fbf5ab0
COSMIT pep8 cleanups
2015-01-21 21:54:11 +01:00
edb73cc693
FIX check for integer 1 before checking for floating 1 (isinstance(1, numbers.real) == True)
2015-01-21 15:53:03 -05:00
a6723edca5
ENH: cross-validate over predefined splits
2015-01-21 13:04:50 +01:00
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
91c03a433a
Added helper function
2015-01-06 19:21:08 +05:30
5b9653334f
Sparse inputs for fit_params
2015-01-05 16:14:36 +05:30
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
7a38062c24
FIX backward compatibility.
...
force_arrays was never in a release, allow_nd was a legal option to ``train_test_split`` that raises an error in master.
2014-12-19 12:15:03 -05:00
1e8bdb859e
FIX bug in fit_params assumed to be an array
...
Fixes gh-3957.
2014-12-15 18:07:05 +01:00
1d6566bd59
Merge pull request #3690 from queqichao/fix_bug_in_cross_validation_when_using_sparse_matrix_for_fit_params
...
+ Fix the bug of calling len when using sparse matrix for fit_params in ...
2014-11-24 23:37:33 +01:00
2f275de41d
ENH Add cross_val_predict function
...
This function returns the cross-validated prediction for each element in
the input data (i.e., the prediction that is made when that object is in
the test set).
2014-11-03 11:51:10 +11:00
58be184f5f
Enable grid search with classifiers that may throw an error on individual fits.
2014-10-02 10:03:17 +10:00
f1f9760ff6
+ Fix the bug of calling len when using sparse matrix for fit_params in cross_validation.py.
...
+ Add test case for the above fix.
2014-09-22 23:09:25 -04:00
e8942057cc
FIX Implemented correct handling of multilabel y in cross_val_score
2014-07-29 11:49:58 +02:00
Arnaud Rachez