Commit Graph

54 Commits

Author SHA1 Message Date
Thomas J. Fan 82df48934e
MNT Applies black formatting to most of the code base (#18948) 2021-06-17 14:21:09 -04:00
Jérémie du Boisberranger 57d3668f2a
MNT Avoid catastrophic cancellation in mean_variance_axis (#19766) 2021-03-31 09:20:22 +02:00
Maria Telenczuk f1111be2fa
incremental mean and var for weighted sparse X (#18569)
Co-authored-by: Alexandre Gramfort <alexandre.gramfort@m4x.org>
Co-authored-by: Olivier Grisel <olivier.grisel@ensta.org>
2020-10-23 22:16:27 +02:00
Alexandre Gramfort 20b27858b9
FIX raise proper error in incr_mean_variance_axis with axis=1 and mismatch shape input arrays (#18655) 2020-10-21 14:11:39 +02:00
Thomas J Fan e770715c43
MNT Completes position arg deprecation (#17272)
* ENH Adds public functions with position args warning

* BUG Circlar import

* STY

* BUG Add keywords

* BUG Fixes warnings

* ENH Adds other public functions

* CLN Suggestion

* CLN Suggestion

* REV Revert pairwise

* ENH Positoinal args for compute_class_weight
2020-05-19 09:07:25 +02:00
Guillaume Lemaitre b8c402722b
BUG avoid nan variance with sparse input in StandardScaler (#16466) 2020-02-27 00:00:22 +11:00
Thomas J Fan c1696ed18a
TST Sets random state in test_csr_row_norms (#16509) 2020-02-21 16:11:44 +01:00
Jérémie du Boisberranger 769794249c
[MRG] new K-means implementation for improved performances (#11950) 2020-02-20 13:39:16 +01:00
Roman Feldbauer 9b39c4c4d2
TST Fix unreachable code in tests (#16110) 2020-02-16 14:41:26 +01:00
Alexandre Batisse 418e0d55d3
[MRG] CLN Remove the use of assert_raises in utils/ (#16337)
* CLN use pytest raises ctxt manager for class weights & estimators [skip ci]

* CLN use pytest raises ctxt manager for multiclass test

* CLN use pytest warns in estimator checks test

* CLN use pytest raises ctxt manager for random test

* CLN use pytest raises ctxt manager for sparsefuncs test

* CLN use pytest raises ctxt manager for testing utils tests

* CLN use pytest raises ctxt manager for utils tests

* CLN use pytest raises ctxt manager for validation tests

* revert test_estimator_check.py to upstream master
2020-02-02 14:44:19 +01:00
Nicolas Hug b92455a6b2 MAINT Deprecate all of utils.testing except all_estimators (#15367) 2019-10-28 17:28:56 +01:00
Nicolas Hug f3eb9d2206 [MRG] MNT removing assert_equal, etc -- continued (#14232) 2019-07-02 14:53:51 +10:00
Adrin Jalali 19c068a2ec MNT towards removing assert_equal, etc (#14222) 2019-07-01 09:13:32 -04:00
rlms dc5e4d8896 FIX downcast large matrix indices where possible in sparsefuncs._minor_reduce (fix #13737) (#13741) 2019-05-23 22:55:09 +10:00
Hanmin Qin 575064b8e1
MNT More clean up after we remove python < 3.5 (#13078) 2019-02-02 22:01:13 +08:00
Hanmin Qin afe0a9b70d MNT skip test falling on master in legacy platforms (#12382) 2018-10-16 07:57:12 +11:00
Joel Nothman 98357ec4d4 FIX (0.21) make count_nonzero dtype invariant wrt axis (#12341) 2018-10-14 21:12:02 +08:00
Guillaume Lemaitre 5718466bb5 [MRG] ENH: Ignore NaNs in StandardScaler and scale (#11206) 2018-06-21 15:48:38 +02:00
Guillaume Lemaitre b9f3758dfd MAINT add function to ignore nan in min/max sparse functions (#11196) 2018-06-07 06:57:30 +10:00
Roman Yurchak d551227d97 ENH Add 64 bit indices support in csr_row_norms and inplace L2/L1 csr norm (#9663) 2017-09-05 18:35:44 +10:00
Thierry Guillemot 78a674875e Correct the deprecation of the random_integers numpy function. (#6712) 2016-04-26 16:08:48 +02:00
Yen 28758cc5eb [MRG+2] Use fused types in sparse mean variance functions (#6593)
* Use fused types in mean variance functions

* Add test for mean-variance functions using fused trpes

* Add whats_new
2016-04-16 11:16:05 -04:00
YenChenLin e34bbc1002 Use fused type in inplace normalize 2016-03-26 10:38:17 +08:00
YenChenLin 42d49c8c14 Make assign_rows_csr support Cython fused types 2016-03-25 00:25:36 +08:00
seales 0485ada58b General spelling fixes 2015-12-16 09:46:42 -08:00
giorgiop 6a5a2f7960 partial_fit for scalers 2015-10-13 11:37:00 +02:00
Lars Buitinck cec3bf98f2 FIX assign_rows_csr: should not zero the entire out array
Fixes #4985.
2015-07-31 10:59:23 +02:00
Raghav R V cd2ee7e454 MAINT docstring --> comments to prevent nose from using doc in verbose mode 2015-03-21 11:16:49 +05:30
MechCoder 2daa0289ef Renamed csc_row_median to csc_median_axis_0 2014-10-21 09:21:30 +02:00
MechCoder 91450c6960 FIX: Wrap csc_row_median around the _get_median imputer function 2014-10-21 09:21:30 +02:00
MechCoder 6f41bd1c88 ENH: Patches Nearest Centroid for metric=manhattan for sparse and dense data 2014-10-21 09:21:30 +02:00
Thomas Unterthiner 52adb5cae7 ENH Add 'axis' argument to sparsefuncs.mean_variance_axis 2014-09-03 13:47:48 +02:00
Joel Nothman 0e31553de5 ENH Sparse multilabel target support
All multilabel metric calculation is now performed efficiently over
sparse CSR matrices.
2014-07-30 22:13:11 +10:00
Lars Buitinck c9789c7edf COSMIT clean up tests with pyflakes
test_ridge.py contained two duplicate functions.
2014-07-22 16:10:16 +02:00
Thomas Unterthiner 325e998fcc Extend utils.sparsefuncs: inplace scale and axis min/max 2014-05-18 20:03:37 +10:00
Manoj-Kumar-S 5d47cfcbcb Improved formatting of unsupported sparse matrices in swapping 2014-05-10 11:37:53 +05:30
Manoj-Kumar-S fc634f95f2 Made the following changes
1. Minor changes to docs
2. Replaced swap with inplace_swap
2014-04-30 23:27:26 +05:30
Manoj-Kumar-S 924db9b07e Made the following changes
a] Replaced numpy slicing with concatanate
b] Added swap_sparse_column
2014-04-24 20:58:11 +05:30
Manoj-Kumar-S 5119d68982 ENH: Swap rows in sparsefuncs 2014-04-24 17:19:09 +05:30
cgohlke 627c564fae TST: Fix ValueError: Buffer dtype mismatch, expected 'npy_intp' but got 'long' on win-amd64 2014-03-30 11:53:02 +02:00
Manoj-Kumar-S 22c412d357 TST: Added tests for non-CSR/CSR format 2014-03-27 15:03:31 +05:30
Manoj-Kumar-S bbb8f1d1bf Moved sparsefuncs to sparsefuncs_fast 2014-03-27 13:20:45 +05:30
Manoj-Kumar-S 8cf30f1089 ENH: Optimise sparsefuncs
a] Removed numpy overhead in finding mean
b] Removed enumerate
c] Added tests to test_sparsefuncs
d] Replaced multiple looping with a single loop
2014-03-27 11:56:20 +05:30
Lars Buitinck b37791bab9 ENH cut MiniBatchKMeans memory usage in half for large n_clusters
Also, changed logging message to reflect the (public) class that is logging,
not the (private) function.

New Cython code lives in utils.sparsefuncs, because it may prove more
generally useful.
2013-03-13 15:23:50 +01:00
Mathieu Blondel e47039d71a Fix warning. 2012-05-21 12:58:32 +09:00
Mathieu Blondel f6729848c4 Unused import. 2012-05-21 12:55:34 +09:00
Alexis Mignon 4842ebb16b Unified access to 'mean_variance_axis0' for CSC and CSR matrices 2012-05-16 07:58:02 +02:00
Alexis Mignon c2c35da51c Made code pep compliant 2012-05-16 07:57:55 +02:00
Alexis Mignon 95ba957556 Corrected missing import of csc_mean_variance_axis0 2012-05-16 07:57:53 +02:00
Andreas Mueller 3a6e490e05 COSMIT pep8 2011-12-23 19:12:52 +01:00