-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathv0.23.html
1120 lines (1086 loc) · 167 KB
/
v0.23.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="generator" content="Docutils 0.19: https://docutils.sourceforge.io/" />
<meta property="og:title" content="Version 0.23.2" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/whats_new/v0.23.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="Changed models: The following estimators and functions, when fit with the same data and parameters, may produce different models from the previous version. This often occurs due to changes in the m..." />
<meta property="og:image" content="https://scikit-learn.org/stable/_static/scikit-learn-logo-small.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="Changed models: The following estimators and functions, when fit with the same data and parameters, may produce different models from the previous version. This often occurs due to changes in the m..." />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Version 0.23.2 — scikit-learn 1.2.2 documentation</title>
<link rel="canonical" href="http://scikit-learn.org/stable/whats_new/v0.23.html" />
<link rel="shortcut icon" href="../_static/favicon.ico"/>
<link rel="stylesheet" href="../_static/css/vendor/bootstrap.min.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../_static/plot_directive.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<script id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
<script src="../_static/js/vendor/jquery-3.6.3.slim.min.js"></script>
</head>
<body>
<nav id="navbar" class="sk-docs-navbar navbar navbar-expand-md navbar-light bg-light py-0">
<div class="container-fluid sk-docs-container px-0">
<a class="navbar-brand py-0" href="../index.html">
<img
class="sk-brand-img"
src="../_static/scikit-learn-logo-small.png"
alt="logo"/>
</a>
<button
id="sk-navbar-toggler"
class="navbar-toggler"
type="button"
data-toggle="collapse"
data-target="#navbarSupportedContent"
aria-controls="navbarSupportedContent"
aria-expanded="false"
aria-label="Toggle navigation"
>
<span class="navbar-toggler-icon"></span>
</button>
<div class="sk-navbar-collapse collapse navbar-collapse" id="navbarSupportedContent">
<ul class="navbar-nav mr-auto">
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../install.html">Install</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../user_guide.html">User Guide</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../modules/classes.html">API</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../auto_examples/index.html">Examples</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" target="_blank" rel="noopener noreferrer" href="https://blog.scikit-learn.org/">Community</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../getting_started.html" >Getting Started</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../tutorial/index.html" >Tutorial</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="v1.2.html" >What's new</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../glossary.html" >Glossary</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://scikit-learn.org/dev/developers/index.html" target="_blank" rel="noopener noreferrer">Development</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../faq.html" >FAQ</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../support.html" >Support</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../related_projects.html" >Related packages</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../roadmap.html" >Roadmap</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../governance.html" >Governance</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../about.html" >About us</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</li>
<li class="nav-item dropdown nav-more-item-dropdown">
<a class="sk-nav-link nav-link dropdown-toggle" href="#" id="navbarDropdown" role="button" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">More</a>
<div class="dropdown-menu" aria-labelledby="navbarDropdown">
<a class="sk-nav-dropdown-item dropdown-item" href="../getting_started.html" >Getting Started</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../tutorial/index.html" >Tutorial</a>
<a class="sk-nav-dropdown-item dropdown-item" href="v1.2.html" >What's new</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../glossary.html" >Glossary</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://scikit-learn.org/dev/developers/index.html" target="_blank" rel="noopener noreferrer">Development</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../faq.html" >FAQ</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../support.html" >Support</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../related_projects.html" >Related packages</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../roadmap.html" >Roadmap</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../governance.html" >Governance</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../about.html" >About us</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</div>
</li>
</ul>
<div id="searchbox" role="search">
<div class="searchformwrapper">
<form class="search" action="../search.html" method="get">
<input class="sk-search-text-input" type="text" name="q" aria-labelledby="searchlabel" />
<input class="sk-search-text-btn" type="submit" value="Go" />
</form>
</div>
</div>
</div>
</div>
</nav>
<div class="d-flex" id="sk-doc-wrapper">
<input type="checkbox" name="sk-toggle-checkbox" id="sk-toggle-checkbox">
<label id="sk-sidemenu-toggle" class="sk-btn-toggle-toc btn sk-btn-primary" for="sk-toggle-checkbox">Toggle Menu</label>
<div id="sk-sidebar-wrapper" class="border-right">
<div class="sk-sidebar-toc-wrapper">
<div class="btn-group w-100 mb-2" role="group" aria-label="rellinks">
<a href="v0.24.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Version 0.24.2">Prev</a><a href="../whats_new.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Release History">Up</a>
<a href="v0.22.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Version 0.22.2.post1">Next</a>
</div>
<div class="alert alert-danger p-1 mb-2" role="alert">
<p class="text-center mb-0">
<strong>scikit-learn 1.2.2</strong><br/>
<a href="http://scikit-learn.org/dev/versions.html">Other versions</a>
</p>
</div>
<div class="alert alert-warning p-1 mb-2" role="alert">
<p class="text-center mb-0">
Please <a class="font-weight-bold" href="../about.html#citing-scikit-learn"><string>cite us</string></a> if you use the software.
</p>
</div>
<div class="sk-sidebar-toc">
<ul>
<li><a class="reference internal" href="#">Version 0.23.2</a><ul>
<li><a class="reference internal" href="#changed-models">Changed models</a></li>
<li><a class="reference internal" href="#changelog">Changelog</a><ul>
<li><a class="reference internal" href="#sklearn-cluster"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#sklearn-decomposition"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#sklearn-ensemble"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.ensemble</span></code></a></li>
<li><a class="reference internal" href="#sklearn-feature-extraction"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_extraction</span></code></a></li>
<li><a class="reference internal" href="#sklearn-linear-model"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.linear_model</span></code></a></li>
<li><a class="reference internal" href="#sklearn-manifold"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.manifold</span></code></a></li>
<li><a class="reference internal" href="#sklearn-metrics"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.metrics</span></code></a></li>
<li><a class="reference internal" href="#sklearn-pipeline"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.pipeline</span></code></a></li>
<li><a class="reference internal" href="#sklearn-utils"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.utils</span></code></a></li>
</ul>
</li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-23-1">Version 0.23.1</a><ul>
<li><a class="reference internal" href="#id1">Changelog</a><ul>
<li><a class="reference internal" href="#id2"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#miscellaneous">Miscellaneous</a></li>
</ul>
</li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-23-0">Version 0.23.0</a><ul>
<li><a class="reference internal" href="#legend-for-changelogs">Legend for changelogs</a></li>
<li><a class="reference internal" href="#enforcing-keyword-only-arguments">Enforcing keyword-only arguments</a></li>
<li><a class="reference internal" href="#id3">Changed models</a></li>
<li><a class="reference internal" href="#id4">Changelog</a><ul>
<li><a class="reference internal" href="#id5"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#sklearn-compose"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#sklearn-datasets"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.datasets</span></code></a></li>
<li><a class="reference internal" href="#id6"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#id7"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.ensemble</span></code></a></li>
<li><a class="reference internal" href="#id8"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_extraction</span></code></a></li>
<li><a class="reference internal" href="#sklearn-feature-selection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_selection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-gaussian-process"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.gaussian_process</span></code></a></li>
<li><a class="reference internal" href="#sklearn-impute"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.impute</span></code></a></li>
<li><a class="reference internal" href="#sklearn-inspection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.inspection</span></code></a></li>
<li><a class="reference internal" href="#id9"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.linear_model</span></code></a></li>
<li><a class="reference internal" href="#id10"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.metrics</span></code></a></li>
<li><a class="reference internal" href="#sklearn-model-selection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.model_selection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-multioutput"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.multioutput</span></code></a></li>
<li><a class="reference internal" href="#sklearn-naive-bayes"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.naive_bayes</span></code></a></li>
<li><a class="reference internal" href="#sklearn-neural-network"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.neural_network</span></code></a></li>
<li><a class="reference internal" href="#id11"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.inspection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-preprocessing"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.preprocessing</span></code></a></li>
<li><a class="reference internal" href="#sklearn-semi-supervised"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.semi_supervised</span></code></a></li>
<li><a class="reference internal" href="#sklearn-svm"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.svm</span></code></a></li>
<li><a class="reference internal" href="#sklearn-tree"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.tree</span></code></a></li>
<li><a class="reference internal" href="#id12"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.utils</span></code></a></li>
<li><a class="reference internal" href="#id13">Miscellaneous</a></li>
</ul>
</li>
<li><a class="reference internal" href="#code-and-documentation-contributors">Code and Documentation Contributors</a></li>
</ul>
</li>
</ul>
</div>
</div>
</div>
<div id="sk-page-content-wrapper">
<div class="sk-page-content container-fluid body px-md-3" role="main">
<section id="version-0-23-2">
<span id="changes-0-23-2"></span><h1>Version 0.23.2<a class="headerlink" href="#version-0-23-2" title="Permalink to this heading">¶</a></h1>
<section id="changed-models">
<h2>Changed models<a class="headerlink" href="#changed-models" title="Permalink to this heading">¶</a></h2>
<p>The following estimators and functions, when fit with the same data and
parameters, may produce different models from the previous version. This often
occurs due to changes in the modelling logic (bug fixes or enhancements), or in
random sampling procedures.</p>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="docutils literal notranslate"><span class="pre">inertia_</span></code> attribute of <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.cluster.MiniBatchKMeans.html#sklearn.cluster.MiniBatchKMeans" title="sklearn.cluster.MiniBatchKMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.MiniBatchKMeans</span></code></a>.</p></li>
</ul>
<p>Details are listed in the changelog below.</p>
<p>(While we are trying to better inform users by providing this information, we
cannot assure that this list is complete.)</p>
</section>
<section id="changelog">
<h2>Changelog<a class="headerlink" href="#changelog" title="Permalink to this heading">¶</a></h2>
<section id="sklearn-cluster">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#sklearn-cluster" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> where rounding errors could
prevent convergence to be declared when <code class="docutils literal notranslate"><span class="pre">tol=0</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17959">#17959</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jérémie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.cluster.MiniBatchKMeans.html#sklearn.cluster.MiniBatchKMeans" title="sklearn.cluster.MiniBatchKMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.MiniBatchKMeans</span></code></a> where the reported inertia was incorrectly
weighted by the sample weights. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17848">#17848</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jérémie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.MeanShift.html#sklearn.cluster.MeanShift" title="sklearn.cluster.MeanShift"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.MeanShift</span></code></a> with <code class="docutils literal notranslate"><span class="pre">bin_seeding=True</span></code>. When
the estimated bandwidth is 0, the behavior is equivalent to
<code class="docutils literal notranslate"><span class="pre">bin_seeding=False</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17742">#17742</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.AffinityPropagation.html#sklearn.cluster.AffinityPropagation" title="sklearn.cluster.AffinityPropagation"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.AffinityPropagation</span></code></a>, that
gives incorrect clusters when the array dtype is float32.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17995">#17995</a> by <a class="reference external" href="https://github.com/Wikilicious">Thomaz Santana</a> and
<a class="reference external" href="https://github.com/amy12xx">Amanda Dsouza</a>.</p></li>
</ul>
</section>
<section id="sklearn-decomposition">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.decomposition" title="sklearn.decomposition"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#sklearn-decomposition" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in
<a class="reference internal" href="../modules/generated/sklearn.decomposition.MiniBatchDictionaryLearning.html#sklearn.decomposition.MiniBatchDictionaryLearning.partial_fit" title="sklearn.decomposition.MiniBatchDictionaryLearning.partial_fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">decomposition.MiniBatchDictionaryLearning.partial_fit</span></code></a> which should
update the dictionary by iterating only once over a mini-batch.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17433">#17433</a> by <a class="reference external" href="https://github.com/cmarmo">Chiara Marmo</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Avoid overflows on Windows in
<a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA.partial_fit" title="sklearn.decomposition.IncrementalPCA.partial_fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">decomposition.IncrementalPCA.partial_fit</span></code></a> for large <code class="docutils literal notranslate"><span class="pre">batch_size</span></code> and
<code class="docutils literal notranslate"><span class="pre">n_samples</span></code> values.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17985">#17985</a> by <a class="reference external" href="https://github.com/aldee153">Alan Butler</a> and
<a class="reference external" href="https://github.com/amy12xx">Amanda Dsouza</a>.</p></li>
</ul>
</section>
<section id="sklearn-ensemble">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.ensemble" title="sklearn.ensemble"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.ensemble</span></code></a><a class="headerlink" href="#sklearn-ensemble" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed bug in <code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.MultinomialDeviance</span></code> where the
average of logloss was incorrectly calculated as sum of logloss.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17694">#17694</a> by <a class="reference external" href="https://github.com/rempfler">Markus Rempfler</a> and
<a class="reference external" href="https://github.com/t-kusanagi2">Tsutomu Kusanagi</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixes <a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingClassifier.html#sklearn.ensemble.StackingClassifier" title="sklearn.ensemble.StackingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingRegressor.html#sklearn.ensemble.StackingRegressor" title="sklearn.ensemble.StackingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingRegressor</span></code></a> compatibility with estimators that
do not define <code class="docutils literal notranslate"><span class="pre">n_features_in_</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17357">#17357</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-feature-extraction">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.feature_extraction" title="sklearn.feature_extraction"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_extraction</span></code></a><a class="headerlink" href="#sklearn-feature-extraction" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixes bug in <a class="reference internal" href="../modules/generated/sklearn.feature_extraction.text.CountVectorizer.html#sklearn.feature_extraction.text.CountVectorizer" title="sklearn.feature_extraction.text.CountVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">feature_extraction.text.CountVectorizer</span></code></a> where
sample order invariance was broken when <code class="docutils literal notranslate"><span class="pre">max_features</span></code> was set and features
had the same count. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/18016">#18016</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>, <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>, and
<a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</p></li>
</ul>
</section>
<section id="sklearn-linear-model">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.linear_model" title="sklearn.linear_model"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.linear_model</span></code></a><a class="headerlink" href="#sklearn-linear-model" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.lars_path.html#sklearn.linear_model.lars_path" title="sklearn.linear_model.lars_path"><code class="xref py py-func docutils literal notranslate"><span class="pre">linear_model.lars_path</span></code></a> does not overwrite <code class="docutils literal notranslate"><span class="pre">X</span></code> when
<code class="docutils literal notranslate"><span class="pre">X_copy=True</span></code> and <code class="docutils literal notranslate"><span class="pre">Gram='auto'</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17914">#17914</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-manifold">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.manifold" title="sklearn.manifold"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.manifold</span></code></a><a class="headerlink" href="#sklearn-manifold" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug where <a class="reference internal" href="../modules/generated/sklearn.metrics.pairwise_distances.html#sklearn.metrics.pairwise_distances" title="sklearn.metrics.pairwise_distances"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.pairwise_distances</span></code></a> would raise an
error if <code class="docutils literal notranslate"><span class="pre">metric='seuclidean'</span></code> and <code class="docutils literal notranslate"><span class="pre">X</span></code> is not type <code class="docutils literal notranslate"><span class="pre">np.float64</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15730">#15730</a> by <a class="reference external" href="https://github.com/ForrestCKoch">Forrest Koch</a>.</p></li>
</ul>
</section>
<section id="sklearn-metrics">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.metrics" title="sklearn.metrics"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.metrics</span></code></a><a class="headerlink" href="#sklearn-metrics" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.mean_squared_error</span></code></a> where the
average of multiple RMSE values was incorrectly calculated as the root of the
average of multiple MSE values.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17309">#17309</a> by <a class="reference external" href="https://github.com/swierh">Swier Heeres</a>.</p></li>
</ul>
</section>
<section id="sklearn-pipeline">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.pipeline" title="sklearn.pipeline"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.pipeline</span></code></a><a class="headerlink" href="#sklearn-pipeline" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.pipeline.FeatureUnion.html#sklearn.pipeline.FeatureUnion" title="sklearn.pipeline.FeatureUnion"><code class="xref py py-class docutils literal notranslate"><span class="pre">pipeline.FeatureUnion</span></code></a> raises a deprecation warning when
<code class="docutils literal notranslate"><span class="pre">None</span></code> is included in <code class="docutils literal notranslate"><span class="pre">transformer_list</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17360">#17360</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-utils">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.utils" title="sklearn.utils"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.utils</span></code></a><a class="headerlink" href="#sklearn-utils" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fix <a class="reference internal" href="../modules/generated/sklearn.utils.estimator_checks.check_estimator.html#sklearn.utils.estimator_checks.check_estimator" title="sklearn.utils.estimator_checks.check_estimator"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.estimator_checks.check_estimator</span></code></a> so that all test
cases support the <code class="docutils literal notranslate"><span class="pre">binary_only</span></code> estimator tag.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17812">#17812</a> by <a class="reference external" href="https://github.com/brcharron">Bruno Charron</a>.</p></li>
</ul>
</section>
</section>
</section>
<section id="version-0-23-1">
<span id="changes-0-23-1"></span><h1>Version 0.23.1<a class="headerlink" href="#version-0-23-1" title="Permalink to this heading">¶</a></h1>
<p><strong>May 18 2020</strong></p>
<section id="id1">
<h2>Changelog<a class="headerlink" href="#id1" title="Permalink to this heading">¶</a></h2>
<section id="id2">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#id2" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> efficiency has been improved for very
small datasets. In particular it cannot spawn idle threads any more.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17210">#17210</a> and <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17235">#17235</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> where the sample weights
provided by the user were modified in place. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17204">#17204</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
</ul>
</section>
<section id="miscellaneous">
<h3>Miscellaneous<a class="headerlink" href="#miscellaneous" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in the <code class="docutils literal notranslate"><span class="pre">repr</span></code> of third-party estimators that use a
<code class="docutils literal notranslate"><span class="pre">**kwargs</span></code> parameter in their constructor, when <code class="docutils literal notranslate"><span class="pre">changed_only</span></code> is True
which is now the default. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17205">#17205</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
</ul>
</section>
</section>
</section>
<section id="version-0-23-0">
<span id="changes-0-23"></span><h1>Version 0.23.0<a class="headerlink" href="#version-0-23-0" title="Permalink to this heading">¶</a></h1>
<p><strong>May 12 2020</strong></p>
<p>For a short description of the main highlights of the release, please
refer to
<a class="reference internal" href="../auto_examples/release_highlights/plot_release_highlights_0_23_0.html#sphx-glr-auto-examples-release-highlights-plot-release-highlights-0-23-0-py"><span class="std std-ref">Release Highlights for scikit-learn 0.23</span></a>.</p>
<section id="legend-for-changelogs">
<h2>Legend for changelogs<a class="headerlink" href="#legend-for-changelogs" title="Permalink to this heading">¶</a></h2>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> : something big that you couldn’t do before.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> : something that you couldn’t do before.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> : an existing feature now may not require as much computation or
memory.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> : a miscellaneous minor improvement.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> : something that previously didn’t work as documentated – or according
to reasonable expectations – should now work.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> : you will need to change your code to have the same effect in the
future; or a feature will be removed in the future.</p></li>
</ul>
</section>
<section id="enforcing-keyword-only-arguments">
<h2>Enforcing keyword-only arguments<a class="headerlink" href="#enforcing-keyword-only-arguments" title="Permalink to this heading">¶</a></h2>
<p>In an effort to promote clear and non-ambiguous use of the library, most
constructor and function parameters are now expected to be passed as keyword
arguments (i.e. using the <code class="docutils literal notranslate"><span class="pre">param=value</span></code> syntax) instead of positional. To
ease the transition, a <code class="docutils literal notranslate"><span class="pre">FutureWarning</span></code> is raised if a keyword-only parameter
is used as positional. In version 1.0 (renaming of 0.25), these parameters
will be strictly keyword-only, and a <code class="docutils literal notranslate"><span class="pre">TypeError</span></code> will be raised.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/15005">#15005</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>, <a class="reference external" href="https://github.com/adrinjalali">Adrin Jalali</a>, <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>, and
<a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>. See <a class="reference external" href="https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep009/proposal.html">SLEP009</a>
for more details.</p>
</section>
<section id="id3">
<h2>Changed models<a class="headerlink" href="#id3" title="Permalink to this heading">¶</a></h2>
<p>The following estimators and functions, when fit with the same data and
parameters, may produce different models from the previous version. This often
occurs due to changes in the modelling logic (bug fixes or enhancements), or in
random sampling procedures.</p>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingClassifier.html#sklearn.ensemble.BaggingClassifier" title="sklearn.ensemble.BaggingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingClassifier</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingRegressor.html#sklearn.ensemble.BaggingRegressor" title="sklearn.ensemble.BaggingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingRegressor</span></code></a>,
and <a class="reference internal" href="../modules/generated/sklearn.ensemble.IsolationForest.html#sklearn.ensemble.IsolationForest" title="sklearn.ensemble.IsolationForest"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.IsolationForest</span></code></a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> with <code class="docutils literal notranslate"><span class="pre">algorithm="elkan"</span></code> and
<code class="docutils literal notranslate"><span class="pre">algorithm="full"</span></code>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.Birch.html#sklearn.cluster.Birch" title="sklearn.cluster.Birch"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.Birch</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="docutils literal notranslate"><span class="pre">compose.ColumnTransformer.get_feature_names</span></code></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.ColumnTransformer.html#sklearn.compose.ColumnTransformer.fit" title="sklearn.compose.ColumnTransformer.fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">compose.ColumnTransformer.fit</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.datasets.make_multilabel_classification.html#sklearn.datasets.make_multilabel_classification" title="sklearn.datasets.make_multilabel_classification"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.make_multilabel_classification</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.PCA.html#sklearn.decomposition.PCA" title="sklearn.decomposition.PCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.PCA</span></code></a> with <code class="docutils literal notranslate"><span class="pre">n_components='mle'</span></code></p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.NMF.html#sklearn.decomposition.NMF" title="sklearn.decomposition.NMF"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.NMF</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.decomposition.non_negative_factorization.html#sklearn.decomposition.non_negative_factorization" title="sklearn.decomposition.non_negative_factorization"><code class="xref py py-func docutils literal notranslate"><span class="pre">decomposition.non_negative_factorization</span></code></a> with float32 dtype input.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.KernelPCA.html#sklearn.decomposition.KernelPCA.inverse_transform" title="sklearn.decomposition.KernelPCA.inverse_transform"><code class="xref py py-func docutils literal notranslate"><span class="pre">decomposition.KernelPCA.inverse_transform</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> <a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="docutils literal notranslate"><span class="pre">estimator_samples_</span></code> in <a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingClassifier.html#sklearn.ensemble.BaggingClassifier" title="sklearn.ensemble.BaggingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingClassifier</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingRegressor.html#sklearn.ensemble.BaggingRegressor" title="sklearn.ensemble.BaggingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingRegressor</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.ensemble.IsolationForest.html#sklearn.ensemble.IsolationForest" title="sklearn.ensemble.IsolationForest"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.IsolationForest</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingClassifier.html#sklearn.ensemble.StackingClassifier" title="sklearn.ensemble.StackingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingRegressor.html#sklearn.ensemble.StackingRegressor" title="sklearn.ensemble.StackingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingRegressor</span></code></a> with <code class="docutils literal notranslate"><span class="pre">sample_weight</span></code></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.gaussian_process.GaussianProcessRegressor.html#sklearn.gaussian_process.GaussianProcessRegressor" title="sklearn.gaussian_process.GaussianProcessRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">gaussian_process.GaussianProcessRegressor</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.RANSACRegressor.html#sklearn.linear_model.RANSACRegressor" title="sklearn.linear_model.RANSACRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RANSACRegressor</span></code></a> with <code class="docutils literal notranslate"><span class="pre">sample_weight</span></code>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeClassifierCV.html#sklearn.linear_model.RidgeClassifierCV" title="sklearn.linear_model.RidgeClassifierCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeClassifierCV</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.mean_squared_error</span></code></a> with <code class="docutils literal notranslate"><span class="pre">squared</span></code> and
<code class="docutils literal notranslate"><span class="pre">multioutput='raw_values'</span></code>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.metrics.mutual_info_score.html#sklearn.metrics.mutual_info_score" title="sklearn.metrics.mutual_info_score"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.mutual_info_score</span></code></a> with negative scores.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.metrics.confusion_matrix.html#sklearn.metrics.confusion_matrix" title="sklearn.metrics.confusion_matrix"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.confusion_matrix</span></code></a> with zero length <code class="docutils literal notranslate"><span class="pre">y_true</span></code> and <code class="docutils literal notranslate"><span class="pre">y_pred</span></code></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.neural_network.MLPClassifier.html#sklearn.neural_network.MLPClassifier" title="sklearn.neural_network.MLPClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">neural_network.MLPClassifier</span></code></a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.StandardScaler</span></code></a> with <code class="docutils literal notranslate"><span class="pre">partial_fit</span></code> and sparse
input.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.preprocessing.Normalizer.html#sklearn.preprocessing.Normalizer" title="sklearn.preprocessing.Normalizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.Normalizer</span></code></a> with norm=’max’</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Any model using the <code class="xref py py-func docutils literal notranslate"><span class="pre">svm.libsvm</span></code> or the <code class="xref py py-func docutils literal notranslate"><span class="pre">svm.liblinear</span></code> solver,
including <a class="reference internal" href="../modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC" title="sklearn.svm.LinearSVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.LinearSVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.LinearSVR.html#sklearn.svm.LinearSVR" title="sklearn.svm.LinearSVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.LinearSVR</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.svm.NuSVC.html#sklearn.svm.NuSVC" title="sklearn.svm.NuSVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.NuSVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.NuSVR.html#sklearn.svm.NuSVR" title="sklearn.svm.NuSVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.NuSVR</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM" title="sklearn.svm.OneClassSVM"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.OneClassSVM</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.svm.SVC.html#sklearn.svm.SVC" title="sklearn.svm.SVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.SVR.html#sklearn.svm.SVR" title="sklearn.svm.SVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVR</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.linear_model.LogisticRegression.html#sklearn.linear_model.LogisticRegression" title="sklearn.linear_model.LogisticRegression"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.LogisticRegression</span></code></a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.DecisionTreeClassifier</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.tree.ExtraTreeClassifier.html#sklearn.tree.ExtraTreeClassifier" title="sklearn.tree.ExtraTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.ExtraTreeClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.GradientBoostingClassifier.html#sklearn.ensemble.GradientBoostingClassifier" title="sklearn.ensemble.GradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.GradientBoostingClassifier</span></code></a> as well as <code class="docutils literal notranslate"><span class="pre">predict</span></code> method of
<a class="reference internal" href="../modules/generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.DecisionTreeRegressor</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.tree.ExtraTreeRegressor.html#sklearn.tree.ExtraTreeRegressor" title="sklearn.tree.ExtraTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.ExtraTreeRegressor</span></code></a>, and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.GradientBoostingRegressor</span></code></a> and read-only float32 input in
<code class="docutils literal notranslate"><span class="pre">predict</span></code>, <code class="docutils literal notranslate"><span class="pre">decision_path</span></code> and <code class="docutils literal notranslate"><span class="pre">predict_proba</span></code>.</p></li>
</ul>
<p>Details are listed in the changelog below.</p>
<p>(While we are trying to better inform users by providing this information, we
cannot assure that this list is complete.)</p>
</section>
<section id="id4">
<h2>Changelog<a class="headerlink" href="#id4" title="Permalink to this heading">¶</a></h2>
<section id="id5">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#id5" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.Birch.html#sklearn.cluster.Birch" title="sklearn.cluster.Birch"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.Birch</span></code></a> implementation of the predict method
avoids high memory footprint by calculating the distances matrix using
a chunked scheme.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16149">#16149</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a> and
<a class="reference external" href="https://github.com/alexshacked">Alex Shacked</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <span class="raw-html"><span class="badge badge-success">Major Feature</span></span> The critical parts of <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a>
have a more optimized implementation. Parallelism is now over the data
instead of over initializations allowing better scalability. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11950">#11950</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> now supports sparse data when
<code class="docutils literal notranslate"><span class="pre">solver</span> <span class="pre">=</span> <span class="pre">"elkan"</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11950">#11950</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.AgglomerativeClustering.html#sklearn.cluster.AgglomerativeClustering" title="sklearn.cluster.AgglomerativeClustering"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.AgglomerativeClustering</span></code></a> has a faster and more
memory efficient implementation of single linkage clustering.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11514">#11514</a> by <a class="reference external" href="https://github.com/lmcinnes">Leland McInnes</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> with <code class="docutils literal notranslate"><span class="pre">algorithm="elkan"</span></code> now converges with
<code class="docutils literal notranslate"><span class="pre">tol=0</span></code> as with the default <code class="docutils literal notranslate"><span class="pre">algorithm="full"</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16075">#16075</a> by
<a class="reference external" href="https://github.com/kno10">Erich Schubert</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.Birch.html#sklearn.cluster.Birch" title="sklearn.cluster.Birch"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.Birch</span></code></a> where the <code class="docutils literal notranslate"><span class="pre">n_clusters</span></code> parameter
could not have a <code class="docutils literal notranslate"><span class="pre">np.int64</span></code> type. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16484">#16484</a>
by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.AgglomerativeCluClustering</span></code> add specific error when
distance matrix is not square and <code class="docutils literal notranslate"><span class="pre">affinity=precomputed</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16257">#16257</a> by <a class="reference external" href="https://github.com/simonamaggio">Simona Maggio</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> parameter of <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.cluster.SpectralCoclustering.html#sklearn.cluster.SpectralCoclustering" title="sklearn.cluster.SpectralCoclustering"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.SpectralCoclustering</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.cluster.SpectralBiclustering.html#sklearn.cluster.SpectralBiclustering" title="sklearn.cluster.SpectralBiclustering"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.SpectralBiclustering</span></code></a> is deprecated. They now use OpenMP
based parallelism. For more details on how to control the number of threads,
please refer to our <a class="reference internal" href="../computing/parallelism.html#parallelism"><span class="std std-ref">Parallelism</span></a> notes. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11950">#11950</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The <code class="docutils literal notranslate"><span class="pre">precompute_distances</span></code> parameter of <a class="reference internal" href="../modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans" title="sklearn.cluster.KMeans"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.KMeans</span></code></a> is
deprecated. It has no effect. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11950">#11950</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The <code class="docutils literal notranslate"><span class="pre">random_state</span></code> parameter has been added to
<a class="reference internal" href="../modules/generated/sklearn.cluster.AffinityPropagation.html#sklearn.cluster.AffinityPropagation" title="sklearn.cluster.AffinityPropagation"><code class="xref py py-class docutils literal notranslate"><span class="pre">cluster.AffinityPropagation</span></code></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16801">#16801</a> by <a class="reference external" href="https://github.com/rcwoolston">@rcwoolston</a>
and <a class="reference external" href="https://github.com/cmarmo">Chiara Marmo</a>.</p></li>
</ul>
</section>
<section id="sklearn-compose">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.compose" title="sklearn.compose"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#sklearn-compose" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.ColumnTransformer.html#sklearn.compose.ColumnTransformer" title="sklearn.compose.ColumnTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">compose.ColumnTransformer</span></code></a> is now faster when working
with dataframes and strings are used to specific subsets of data for
transformers. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16431">#16431</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.ColumnTransformer.html#sklearn.compose.ColumnTransformer" title="sklearn.compose.ColumnTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">compose.ColumnTransformer</span></code></a> method <code class="docutils literal notranslate"><span class="pre">get_feature_names</span></code>
now supports <code class="docutils literal notranslate"><span class="pre">'passthrough'</span></code> columns, with the feature name being either
the column name for a dataframe, or <code class="docutils literal notranslate"><span class="pre">'xi'</span></code> for column index <code class="docutils literal notranslate"><span class="pre">i</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14048">#14048</a> by <a class="reference external" href="https://github.com/lrjball">Lewis Ball</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.ColumnTransformer.html#sklearn.compose.ColumnTransformer" title="sklearn.compose.ColumnTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">compose.ColumnTransformer</span></code></a> method <code class="docutils literal notranslate"><span class="pre">get_feature_names</span></code> now
returns correct results when one of the transformer steps applies on an
empty list of columns <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15963">#15963</a> by <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.ColumnTransformer.html#sklearn.compose.ColumnTransformer.fit" title="sklearn.compose.ColumnTransformer.fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">compose.ColumnTransformer.fit</span></code></a> will error when selecting
a column name that is not unique in the dataframe. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16431">#16431</a> by
<a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-datasets">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.datasets" title="sklearn.datasets"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.datasets</span></code></a><a class="headerlink" href="#sklearn-datasets" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.datasets.fetch_openml.html#sklearn.datasets.fetch_openml" title="sklearn.datasets.fetch_openml"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.fetch_openml</span></code></a> has reduced memory usage because
it no longer stores the full dataset text stream in memory. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16084">#16084</a> by
<a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.datasets.fetch_california_housing.html#sklearn.datasets.fetch_california_housing" title="sklearn.datasets.fetch_california_housing"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.fetch_california_housing</span></code></a> now supports
heterogeneous data using pandas by setting <code class="docutils literal notranslate"><span class="pre">as_frame=True</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15950">#15950</a>
by <a class="reference external" href="https://github.com/gitsteph">Stephanie Andrews</a> and
<a class="reference external" href="https://github.com/reshamas">Reshama Shaikh</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> embedded dataset loaders <code class="xref py py-func docutils literal notranslate"><span class="pre">load_breast_cancer</span></code>,
<code class="xref py py-func docutils literal notranslate"><span class="pre">load_diabetes</span></code>, <code class="xref py py-func docutils literal notranslate"><span class="pre">load_digits</span></code>, <code class="xref py py-func docutils literal notranslate"><span class="pre">load_iris</span></code>,
<code class="xref py py-func docutils literal notranslate"><span class="pre">load_linnerud</span></code> and <code class="xref py py-func docutils literal notranslate"><span class="pre">load_wine</span></code> now support loading as a pandas
<code class="docutils literal notranslate"><span class="pre">DataFrame</span></code> by setting <code class="docutils literal notranslate"><span class="pre">as_frame=True</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15980">#15980</a> by <a class="reference external" href="https://github.com/wconnell">@wconnell</a> and
<a class="reference external" href="https://github.com/reshamas">Reshama Shaikh</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> Added <code class="docutils literal notranslate"><span class="pre">return_centers</span></code> parameter in
<a class="reference internal" href="../modules/generated/sklearn.datasets.make_blobs.html#sklearn.datasets.make_blobs" title="sklearn.datasets.make_blobs"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.make_blobs</span></code></a>, which can be used to return
centers for each cluster.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15709">#15709</a> by <a class="reference external" href="https://github.com/shivamgargsya">@shivamgargsya</a> and
<a class="reference external" href="https://github.com/venkyyuvy">Venkatachalam N</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> Functions <a class="reference internal" href="../modules/generated/sklearn.datasets.make_circles.html#sklearn.datasets.make_circles" title="sklearn.datasets.make_circles"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.make_circles</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.datasets.make_moons.html#sklearn.datasets.make_moons" title="sklearn.datasets.make_moons"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.make_moons</span></code></a> now accept two-element tuple.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15707">#15707</a> by <a class="reference external" href="https://github.com/mjmikulski">Maciej J Mikulski</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.datasets.make_multilabel_classification.html#sklearn.datasets.make_multilabel_classification" title="sklearn.datasets.make_multilabel_classification"><code class="xref py py-func docutils literal notranslate"><span class="pre">datasets.make_multilabel_classification</span></code></a> now generates
<code class="docutils literal notranslate"><span class="pre">ValueError</span></code> for arguments <code class="docutils literal notranslate"><span class="pre">n_classes</span> <span class="pre"><</span> <span class="pre">1</span></code> OR <code class="docutils literal notranslate"><span class="pre">length</span> <span class="pre"><</span> <span class="pre">1</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16006">#16006</a> by <a class="reference external" href="https://github.com/rushabh-v">Rushabh Vasani</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The <code class="docutils literal notranslate"><span class="pre">StreamHandler</span></code> was removed from <code class="docutils literal notranslate"><span class="pre">sklearn.logger</span></code> to avoid
double logging of messages in common cases where a handler is attached
to the root logger, and to follow the Python logging documentation
recommendation for libraries to leave the log message handling to
users and application code. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16451">#16451</a> by <a class="reference external" href="https://github.com/cdeil">Christoph Deil</a>.</p></li>
</ul>
</section>
<section id="id6">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.decomposition" title="sklearn.decomposition"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#id6" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.NMF.html#sklearn.decomposition.NMF" title="sklearn.decomposition.NMF"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.NMF</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.decomposition.non_negative_factorization.html#sklearn.decomposition.non_negative_factorization" title="sklearn.decomposition.non_negative_factorization"><code class="xref py py-func docutils literal notranslate"><span class="pre">decomposition.non_negative_factorization</span></code></a> now preserves float32 dtype.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16280">#16280</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <code class="xref py py-func docutils literal notranslate"><span class="pre">TruncatedSVD.transform</span></code> is now faster on given sparse
<code class="docutils literal notranslate"><span class="pre">csc</span></code> matrices. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16837">#16837</a> by <a class="reference external" href="https://github.com/wornbb">@wornbb</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.PCA.html#sklearn.decomposition.PCA" title="sklearn.decomposition.PCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.PCA</span></code></a> with a float <code class="docutils literal notranslate"><span class="pre">n_components</span></code> parameter, will
exclusively choose the components that explain the variance greater than
<code class="docutils literal notranslate"><span class="pre">n_components</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15669">#15669</a> by <a class="reference external" href="https://github.com/krishnachaitanya9">Krishna Chaitanya</a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.PCA.html#sklearn.decomposition.PCA" title="sklearn.decomposition.PCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.PCA</span></code></a> with <code class="docutils literal notranslate"><span class="pre">n_components='mle'</span></code> now correctly
handles small eigenvalues, and does not infer 0 as the correct number of
components. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16224">#16224</a> by <a class="reference external" href="https://github.com/lschwetlick">Lisa Schwetlick</a>, and
<a class="reference external" href="https://github.com/gelavizh1">Gelavizh Ahmadi</a> and <a class="reference external" href="https://github.com/marijavlajic">Marija Vlajic Wheeler</a> and <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16841">#16841</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.KernelPCA.html#sklearn.decomposition.KernelPCA" title="sklearn.decomposition.KernelPCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.KernelPCA</span></code></a> method <code class="docutils literal notranslate"><span class="pre">inverse_transform</span></code> now
applies the correct inverse transform to the transformed data. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16655">#16655</a>
by <a class="reference external" href="https://github.com/lrjball">Lewis Ball</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed bug that was causing <a class="reference internal" href="../modules/generated/sklearn.decomposition.KernelPCA.html#sklearn.decomposition.KernelPCA" title="sklearn.decomposition.KernelPCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.KernelPCA</span></code></a> to sometimes
raise <code class="docutils literal notranslate"><span class="pre">invalid</span> <span class="pre">value</span> <span class="pre">encountered</span> <span class="pre">in</span> <span class="pre">multiply</span></code> during <code class="docutils literal notranslate"><span class="pre">fit</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16718">#16718</a> by <a class="reference external" href="https://github.com/gui-miotto">Gui Miotto</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> Added <code class="docutils literal notranslate"><span class="pre">n_components_</span></code> attribute to <a class="reference internal" href="../modules/generated/sklearn.decomposition.SparsePCA.html#sklearn.decomposition.SparsePCA" title="sklearn.decomposition.SparsePCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.SparsePCA</span></code></a>
and <a class="reference internal" href="../modules/generated/sklearn.decomposition.MiniBatchSparsePCA.html#sklearn.decomposition.MiniBatchSparsePCA" title="sklearn.decomposition.MiniBatchSparsePCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.MiniBatchSparsePCA</span></code></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16981">#16981</a> by
<a class="reference external" href="https://github.com/Reksbril">Mateusz Górski</a>.</p></li>
</ul>
</section>
<section id="id7">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.ensemble" title="sklearn.ensemble"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.ensemble</span></code></a><a class="headerlink" href="#id7" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a> now support
<a class="reference internal" href="../glossary.html#term-sample_weight"><span class="xref std std-term">sample_weight</span></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14696">#14696</a> by <a class="reference external" href="https://github.com/adrinjalali">Adrin Jalali</a> and <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> Early stopping in
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a> is now determined with a
new <code class="docutils literal notranslate"><span class="pre">early_stopping</span></code> parameter instead of <code class="docutils literal notranslate"><span class="pre">n_iter_no_change</span></code>. Default value
is ‘auto’, which enables early stopping if there are at least 10,000
samples in the training set. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14516">#14516</a> by <a class="reference external" href="https://github.com/johannfaouzi">Johann Faouzi</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a> now support monotonic
constraints, useful when features are supposed to have a positive/negative
effect on the target. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15582">#15582</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Added boolean <code class="docutils literal notranslate"><span class="pre">verbose</span></code> flag to classes:
<a class="reference internal" href="../modules/generated/sklearn.ensemble.VotingClassifier.html#sklearn.ensemble.VotingClassifier" title="sklearn.ensemble.VotingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.VotingClassifier</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.ensemble.VotingRegressor.html#sklearn.ensemble.VotingRegressor" title="sklearn.ensemble.VotingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.VotingRegressor</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16069">#16069</a> by <a class="reference external" href="https://github.com/spbail">Sam Bail</a>,
<a class="reference external" href="https://github.com/hannahbrucemacdonald">Hanna Bruce MacDonald</a>,
<a class="reference external" href="https://github.com/reshamas">Reshama Shaikh</a>, and
<a class="reference external" href="https://github.com/cmarmo">Chiara Marmo</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a> that would not respect the
<code class="docutils literal notranslate"><span class="pre">max_leaf_nodes</span></code> parameter if the criteria was reached at the same time as
the <code class="docutils literal notranslate"><span class="pre">max_depth</span></code> criteria. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16183">#16183</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Changed the convention for <code class="docutils literal notranslate"><span class="pre">max_depth</span></code> parameter of
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a>. The depth now corresponds to
the number of edges to go from the root to the deepest leaf.
Stumps (trees with one split) are now allowed.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16182">#16182</a> by <a class="reference external" href="https://github.com/santhoshbala18">Santhosh B</a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingClassifier.html#sklearn.ensemble.BaggingClassifier" title="sklearn.ensemble.BaggingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingClassifier</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.ensemble.BaggingRegressor.html#sklearn.ensemble.BaggingRegressor" title="sklearn.ensemble.BaggingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.BaggingRegressor</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.ensemble.IsolationForest.html#sklearn.ensemble.IsolationForest" title="sklearn.ensemble.IsolationForest"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.IsolationForest</span></code></a>
where the attribute <code class="docutils literal notranslate"><span class="pre">estimators_samples_</span></code> did not generate the proper indices
used during <code class="docutils literal notranslate"><span class="pre">fit</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16437">#16437</a> by <a class="reference external" href="https://github.com/chofchof">Jin-Hwan CHO</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingClassifier.html#sklearn.ensemble.StackingClassifier" title="sklearn.ensemble.StackingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.StackingRegressor.html#sklearn.ensemble.StackingRegressor" title="sklearn.ensemble.StackingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.StackingRegressor</span></code></a> where the <code class="docutils literal notranslate"><span class="pre">sample_weight</span></code>
argument was not being passed to <code class="docutils literal notranslate"><span class="pre">cross_val_predict</span></code> when
evaluating the base estimators on cross-validation folds
to obtain the input to the meta estimator.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16539">#16539</a> by <a class="reference external" href="https://github.com/wderose">Bill DeRose</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> Added additional option <code class="docutils literal notranslate"><span class="pre">loss="poisson"</span></code> to
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a>, which adds Poisson deviance
with log-link useful for modeling count data.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16692">#16692</a> by <a class="reference external" href="https://github.com/lorentzenchr">Christian Lorentzen</a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug where <a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingRegressor</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.HistGradientBoostingClassifier.html#sklearn.ensemble.HistGradientBoostingClassifier" title="sklearn.ensemble.HistGradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.HistGradientBoostingClassifier</span></code></a> would fail with multiple
calls to fit when <code class="docutils literal notranslate"><span class="pre">warm_start=True</span></code>, <code class="docutils literal notranslate"><span class="pre">early_stopping=True</span></code>, and there is no
validation set. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16663">#16663</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="id8">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.feature_extraction" title="sklearn.feature_extraction"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_extraction</span></code></a><a class="headerlink" href="#id8" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.feature_extraction.text.CountVectorizer.html#sklearn.feature_extraction.text.CountVectorizer" title="sklearn.feature_extraction.text.CountVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">feature_extraction.text.CountVectorizer</span></code></a> now sorts
features after pruning them by document frequency. This improves performances
for datasets with large vocabularies combined with <code class="docutils literal notranslate"><span class="pre">min_df</span></code> or <code class="docutils literal notranslate"><span class="pre">max_df</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15834">#15834</a> by <a class="reference external" href="https://github.com/smola">Santiago M. Mola</a>.</p></li>
</ul>
</section>
<section id="sklearn-feature-selection">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.feature_selection" title="sklearn.feature_selection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_selection</span></code></a><a class="headerlink" href="#sklearn-feature-selection" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> Added support for multioutput data in
<a class="reference internal" href="../modules/generated/sklearn.feature_selection.RFE.html#sklearn.feature_selection.RFE" title="sklearn.feature_selection.RFE"><code class="xref py py-class docutils literal notranslate"><span class="pre">feature_selection.RFE</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.feature_selection.RFECV.html#sklearn.feature_selection.RFECV" title="sklearn.feature_selection.RFECV"><code class="xref py py-class docutils literal notranslate"><span class="pre">feature_selection.RFECV</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16103">#16103</a> by <a class="reference external" href="https://github.com/divyaprabha123">Divyaprabha M</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Adds <a class="reference internal" href="../modules/generated/sklearn.feature_selection.SelectorMixin.html#sklearn.feature_selection.SelectorMixin" title="sklearn.feature_selection.SelectorMixin"><code class="xref py py-class docutils literal notranslate"><span class="pre">feature_selection.SelectorMixin</span></code></a> back to public API.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16132">#16132</a> by <a class="reference external" href="https://github.com/trimeta">@trimeta</a>.</p></li>
</ul>
</section>
<section id="sklearn-gaussian-process">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.gaussian_process" title="sklearn.gaussian_process"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.gaussian_process</span></code></a><a class="headerlink" href="#sklearn-gaussian-process" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.gaussian_process.kernels.Matern.html#sklearn.gaussian_process.kernels.Matern" title="sklearn.gaussian_process.kernels.Matern"><code class="xref py py-func docutils literal notranslate"><span class="pre">gaussian_process.kernels.Matern</span></code></a> returns the RBF kernel when <code class="docutils literal notranslate"><span class="pre">nu=np.inf</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15503">#15503</a> by <a class="reference external" href="https://github.com/sam-dixon">Sam Dixon</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed bug in <a class="reference internal" href="../modules/generated/sklearn.gaussian_process.GaussianProcessRegressor.html#sklearn.gaussian_process.GaussianProcessRegressor" title="sklearn.gaussian_process.GaussianProcessRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">gaussian_process.GaussianProcessRegressor</span></code></a> that
caused predicted standard deviations to only be between 0 and 1 when
WhiteKernel is not used. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15782">#15782</a>
by <a class="reference external" href="https://github.com/plgreenLIRU">@plgreenLIRU</a>.</p></li>
</ul>
</section>
<section id="sklearn-impute">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.impute" title="sklearn.impute"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.impute</span></code></a><a class="headerlink" href="#sklearn-impute" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.impute.IterativeImputer.html#sklearn.impute.IterativeImputer" title="sklearn.impute.IterativeImputer"><code class="xref py py-class docutils literal notranslate"><span class="pre">impute.IterativeImputer</span></code></a> accepts both scalar and array-like inputs for
<code class="docutils literal notranslate"><span class="pre">max_value</span></code> and <code class="docutils literal notranslate"><span class="pre">min_value</span></code>. Array-like inputs allow a different max and min to be specified
for each feature. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16403">#16403</a> by <a class="reference external" href="https://github.com/narendramukherjee">Narendra Mukherjee</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.impute.SimpleImputer.html#sklearn.impute.SimpleImputer" title="sklearn.impute.SimpleImputer"><code class="xref py py-class docutils literal notranslate"><span class="pre">impute.SimpleImputer</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.impute.KNNImputer.html#sklearn.impute.KNNImputer" title="sklearn.impute.KNNImputer"><code class="xref py py-class docutils literal notranslate"><span class="pre">impute.KNNImputer</span></code></a>, and
<a class="reference internal" href="../modules/generated/sklearn.impute.IterativeImputer.html#sklearn.impute.IterativeImputer" title="sklearn.impute.IterativeImputer"><code class="xref py py-class docutils literal notranslate"><span class="pre">impute.IterativeImputer</span></code></a> accepts pandas’ nullable integer dtype with
missing values. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16508">#16508</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-inspection">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.inspection" title="sklearn.inspection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.inspection</span></code></a><a class="headerlink" href="#sklearn-inspection" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.inspection.partial_dependence.html#sklearn.inspection.partial_dependence" title="sklearn.inspection.partial_dependence"><code class="xref py py-func docutils literal notranslate"><span class="pre">inspection.partial_dependence</span></code></a> and
<code class="xref py py-func docutils literal notranslate"><span class="pre">inspection.plot_partial_dependence</span></code> now support the fast ‘recursion’
method for <a class="reference internal" href="../modules/generated/sklearn.ensemble.RandomForestRegressor.html#sklearn.ensemble.RandomForestRegressor" title="sklearn.ensemble.RandomForestRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.RandomForestRegressor</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.DecisionTreeRegressor</span></code></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15864">#15864</a> by
<a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
</ul>
</section>
<section id="id9">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.linear_model" title="sklearn.linear_model"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.linear_model</span></code></a><a class="headerlink" href="#id9" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> Added generalized linear models (GLM) with non normal error
distributions, including <a class="reference internal" href="../modules/generated/sklearn.linear_model.PoissonRegressor.html#sklearn.linear_model.PoissonRegressor" title="sklearn.linear_model.PoissonRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.PoissonRegressor</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.GammaRegressor.html#sklearn.linear_model.GammaRegressor" title="sklearn.linear_model.GammaRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.GammaRegressor</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.linear_model.TweedieRegressor.html#sklearn.linear_model.TweedieRegressor" title="sklearn.linear_model.TweedieRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.TweedieRegressor</span></code></a>
which use Poisson, Gamma and Tweedie distributions respectively.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14300">#14300</a> by <a class="reference external" href="https://github.com/lorentzenchr">Christian Lorentzen</a>, <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>,
and <a class="reference external" href="https://twitter.com/ogrisel">Olivier Grisel</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> Support of <code class="docutils literal notranslate"><span class="pre">sample_weight</span></code> in
<a class="reference internal" href="../modules/generated/sklearn.linear_model.ElasticNet.html#sklearn.linear_model.ElasticNet" title="sklearn.linear_model.ElasticNet"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.ElasticNet</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.linear_model.Lasso.html#sklearn.linear_model.Lasso" title="sklearn.linear_model.Lasso"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.Lasso</span></code></a> for dense
feature matrix <code class="docutils literal notranslate"><span class="pre">X</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15436">#15436</a> by <a class="reference external" href="https://github.com/lorentzenchr">Christian Lorentzen</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeCV.html#sklearn.linear_model.RidgeCV" title="sklearn.linear_model.RidgeCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeCV</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeClassifierCV.html#sklearn.linear_model.RidgeClassifierCV" title="sklearn.linear_model.RidgeClassifierCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeClassifierCV</span></code></a> now does not allocate a
potentially large array to store dual coefficients for all hyperparameters
during its <code class="docutils literal notranslate"><span class="pre">fit</span></code>, nor an array to store all error or LOO predictions unless
<code class="docutils literal notranslate"><span class="pre">store_cv_values</span></code> is <code class="docutils literal notranslate"><span class="pre">True</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15652">#15652</a> by <a class="reference external" href="https://github.com/jeromedockes">Jérôme Dockès</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.LassoLars.html#sklearn.linear_model.LassoLars" title="sklearn.linear_model.LassoLars"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.LassoLars</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.linear_model.Lars.html#sklearn.linear_model.Lars" title="sklearn.linear_model.Lars"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.Lars</span></code></a> now support a <code class="docutils literal notranslate"><span class="pre">jitter</span></code> parameter that adds
random noise to the target. This might help with stability in some edge
cases. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15179">#15179</a> by <a class="reference external" href="https://github.com/angelaambroz">@angelaambroz</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug where if a <code class="docutils literal notranslate"><span class="pre">sample_weight</span></code> parameter was passed to the fit
method of <a class="reference internal" href="../modules/generated/sklearn.linear_model.RANSACRegressor.html#sklearn.linear_model.RANSACRegressor" title="sklearn.linear_model.RANSACRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RANSACRegressor</span></code></a>, it would not be passed to
the wrapped <code class="docutils literal notranslate"><span class="pre">base_estimator</span></code> during the fitting of the final model.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15773">#15773</a> by <a class="reference external" href="https://github.com/J-A16">Jeremy Alexandre</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Add <code class="docutils literal notranslate"><span class="pre">best_score_</span></code> attribute to <a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeCV.html#sklearn.linear_model.RidgeCV" title="sklearn.linear_model.RidgeCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeCV</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeClassifierCV.html#sklearn.linear_model.RidgeClassifierCV" title="sklearn.linear_model.RidgeClassifierCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeClassifierCV</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15655">#15655</a> by <a class="reference external" href="https://github.com/jeromedockes">Jérôme Dockès</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.linear_model.RidgeClassifierCV.html#sklearn.linear_model.RidgeClassifierCV" title="sklearn.linear_model.RidgeClassifierCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.RidgeClassifierCV</span></code></a> to pass a
specific scoring strategy. Before the internal estimator outputs score
instead of predictions.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14848">#14848</a> by <a class="reference external" href="https://github.com/venkyyuvy">Venkatachalam N</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.LogisticRegression.html#sklearn.linear_model.LogisticRegression" title="sklearn.linear_model.LogisticRegression"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.LogisticRegression</span></code></a> will now avoid an unnecessary
iteration when <code class="docutils literal notranslate"><span class="pre">solver='newton-cg'</span></code> by checking for inferior or equal instead
of strictly inferior for maximum of <code class="docutils literal notranslate"><span class="pre">absgrad</span></code> and <code class="docutils literal notranslate"><span class="pre">tol</span></code> in <code class="docutils literal notranslate"><span class="pre">utils.optimize._newton_cg</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16266">#16266</a> by <a class="reference external" href="https://github.com/rushabh-v">Rushabh Vasani</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Deprecated public attributes <code class="docutils literal notranslate"><span class="pre">standard_coef_</span></code>, <code class="docutils literal notranslate"><span class="pre">standard_intercept_</span></code>,
<code class="docutils literal notranslate"><span class="pre">average_coef_</span></code>, and <code class="docutils literal notranslate"><span class="pre">average_intercept_</span></code> in
<a class="reference internal" href="../modules/generated/sklearn.linear_model.SGDClassifier.html#sklearn.linear_model.SGDClassifier" title="sklearn.linear_model.SGDClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.SGDClassifier</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.SGDRegressor.html#sklearn.linear_model.SGDRegressor" title="sklearn.linear_model.SGDRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.SGDRegressor</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.PassiveAggressiveClassifier.html#sklearn.linear_model.PassiveAggressiveClassifier" title="sklearn.linear_model.PassiveAggressiveClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.PassiveAggressiveClassifier</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.PassiveAggressiveRegressor.html#sklearn.linear_model.PassiveAggressiveRegressor" title="sklearn.linear_model.PassiveAggressiveRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.PassiveAggressiveRegressor</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16261">#16261</a> by <a class="reference external" href="https://github.com/chbrandt">Carlos Brandt</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.linear_model.ARDRegression.html#sklearn.linear_model.ARDRegression" title="sklearn.linear_model.ARDRegression"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.ARDRegression</span></code></a> is more stable and
much faster when <code class="docutils literal notranslate"><span class="pre">n_samples</span> <span class="pre">></span> <span class="pre">n_features</span></code>. It can now scale to hundreds of
thousands of samples. The stability fix might imply changes in the number
of non-zero coefficients and in the predicted output. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16849">#16849</a> by
<a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.linear_model.ElasticNetCV.html#sklearn.linear_model.ElasticNetCV" title="sklearn.linear_model.ElasticNetCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.ElasticNetCV</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskElasticNetCV.html#sklearn.linear_model.MultiTaskElasticNetCV" title="sklearn.linear_model.MultiTaskElasticNetCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskElasticNetCV</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.linear_model.LassoCV.html#sklearn.linear_model.LassoCV" title="sklearn.linear_model.LassoCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.LassoCV</span></code></a>
and <a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskLassoCV.html#sklearn.linear_model.MultiTaskLassoCV" title="sklearn.linear_model.MultiTaskLassoCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskLassoCV</span></code></a> where fitting would fail when
using joblib loky backend. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14264">#14264</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jérémie du Boisberranger</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> Speed up <a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskLasso.html#sklearn.linear_model.MultiTaskLasso" title="sklearn.linear_model.MultiTaskLasso"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskLasso</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskLassoCV.html#sklearn.linear_model.MultiTaskLassoCV" title="sklearn.linear_model.MultiTaskLassoCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskLassoCV</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskElasticNet.html#sklearn.linear_model.MultiTaskElasticNet" title="sklearn.linear_model.MultiTaskElasticNet"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskElasticNet</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.linear_model.MultiTaskElasticNetCV.html#sklearn.linear_model.MultiTaskElasticNetCV" title="sklearn.linear_model.MultiTaskElasticNetCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.MultiTaskElasticNetCV</span></code></a> by avoiding slower
BLAS Level 2 calls on small arrays
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17021">#17021</a> by <a class="reference external" href="https://github.com/agramfort">Alex Gramfort</a> and
<a class="reference external" href="https://github.com/mathurinm">Mathurin Massias</a>.</p></li>
</ul>
</section>
<section id="id10">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.metrics" title="sklearn.metrics"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.metrics</span></code></a><a class="headerlink" href="#id10" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.pairwise.pairwise_distances_chunked</span></code> now allows
its <code class="docutils literal notranslate"><span class="pre">reduce_func</span></code> to not have a return value, enabling in-place operations.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16397">#16397</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.mean_squared_error</span></code></a> to not ignore
argument <code class="docutils literal notranslate"><span class="pre">squared</span></code> when argument <code class="docutils literal notranslate"><span class="pre">multioutput='raw_values'</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16323">#16323</a> by <a class="reference external" href="https://github.com/rushabh-v">Rushabh Vasani</a></p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.metrics.mutual_info_score.html#sklearn.metrics.mutual_info_score" title="sklearn.metrics.mutual_info_score"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.mutual_info_score</span></code></a> where negative
scores could be returned. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16362">#16362</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.metrics.confusion_matrix.html#sklearn.metrics.confusion_matrix" title="sklearn.metrics.confusion_matrix"><code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.confusion_matrix</span></code></a> that would raise
an error when <code class="docutils literal notranslate"><span class="pre">y_true</span></code> and <code class="docutils literal notranslate"><span class="pre">y_pred</span></code> were length zero and <code class="docutils literal notranslate"><span class="pre">labels</span></code> was
not <code class="docutils literal notranslate"><span class="pre">None</span></code>. In addition, we raise an error when an empty list is given to
the <code class="docutils literal notranslate"><span class="pre">labels</span></code> parameter.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16442">#16442</a> by <a class="reference external" href="https://github.com/parsons-kyle-89">Kyle Parsons</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Changed the formatting of values in
<a class="reference internal" href="../modules/generated/sklearn.metrics.ConfusionMatrixDisplay.html#sklearn.metrics.ConfusionMatrixDisplay.plot" title="sklearn.metrics.ConfusionMatrixDisplay.plot"><code class="xref py py-meth docutils literal notranslate"><span class="pre">metrics.ConfusionMatrixDisplay.plot</span></code></a> and
<code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.plot_confusion_matrix</span></code> to pick the shorter format (either ‘2g’
or ‘d’). <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16159">#16159</a> by <a class="reference external" href="https://github.com/Rick-Mackenbach">Rick Mackenbach</a> and
<a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> From version 0.25, <code class="xref py py-func docutils literal notranslate"><span class="pre">metrics.pairwise.pairwise_distances</span></code> will no
longer automatically compute the <code class="docutils literal notranslate"><span class="pre">VI</span></code> parameter for Mahalanobis distance
and the <code class="docutils literal notranslate"><span class="pre">V</span></code> parameter for seuclidean distance if <code class="docutils literal notranslate"><span class="pre">Y</span></code> is passed. The user
will be expected to compute this parameter on the training data of their
choice and pass it to <code class="docutils literal notranslate"><span class="pre">pairwise_distances</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16993">#16993</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</p></li>
</ul>
</section>
<section id="sklearn-model-selection">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.model_selection" title="sklearn.model_selection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.model_selection</span></code></a><a class="headerlink" href="#sklearn-model-selection" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.model_selection.GridSearchCV.html#sklearn.model_selection.GridSearchCV" title="sklearn.model_selection.GridSearchCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">model_selection.GridSearchCV</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.model_selection.RandomizedSearchCV.html#sklearn.model_selection.RandomizedSearchCV" title="sklearn.model_selection.RandomizedSearchCV"><code class="xref py py-class docutils literal notranslate"><span class="pre">model_selection.RandomizedSearchCV</span></code></a> yields stack trace information
in fit failed warning messages in addition to previously emitted
type and details.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15622">#15622</a> by <a class="reference external" href="https://github.com/GregoryMorse">Gregory Morse</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.model_selection.cross_val_predict.html#sklearn.model_selection.cross_val_predict" title="sklearn.model_selection.cross_val_predict"><code class="xref py py-func docutils literal notranslate"><span class="pre">model_selection.cross_val_predict</span></code></a> supports
<code class="docutils literal notranslate"><span class="pre">method="predict_proba"</span></code> when <code class="docutils literal notranslate"><span class="pre">y=None</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15918">#15918</a> by
<a class="reference external" href="https://github.com/lkubin">Luca Kubin</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="xref py py-func docutils literal notranslate"><span class="pre">model_selection.fit_grid_point</span></code> is deprecated in 0.23 and will
be removed in 0.25. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16401">#16401</a> by
<a class="reference external" href="https://github.com/ariepratama">Arie Pratama Sutiono</a></p></li>
</ul>
</section>
<section id="sklearn-multioutput">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.multioutput" title="sklearn.multioutput"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.multioutput</span></code></a><a class="headerlink" href="#sklearn-multioutput" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.multioutput.MultiOutputRegressor.html#sklearn.multioutput.MultiOutputRegressor.fit" title="sklearn.multioutput.MultiOutputRegressor.fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">multioutput.MultiOutputRegressor.fit</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.multioutput.MultiOutputClassifier.html#sklearn.multioutput.MultiOutputClassifier.fit" title="sklearn.multioutput.MultiOutputClassifier.fit"><code class="xref py py-func docutils literal notranslate"><span class="pre">multioutput.MultiOutputClassifier.fit</span></code></a> now can accept <code class="docutils literal notranslate"><span class="pre">fit_params</span></code>
to pass to the <code class="docutils literal notranslate"><span class="pre">estimator.fit</span></code> method of each step. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/15953">#15953</a>
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15959">#15959</a> by <a class="reference external" href="https://github.com/huangk10">Ke Huang</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.multioutput.RegressorChain.html#sklearn.multioutput.RegressorChain" title="sklearn.multioutput.RegressorChain"><code class="xref py py-class docutils literal notranslate"><span class="pre">multioutput.RegressorChain</span></code></a> now supports <code class="docutils literal notranslate"><span class="pre">fit_params</span></code>
for <code class="docutils literal notranslate"><span class="pre">base_estimator</span></code> during <code class="docutils literal notranslate"><span class="pre">fit</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16111">#16111</a> by <a class="reference external" href="https://github.com/venkyyuvy">Venkatachalam N</a>.</p></li>
</ul>
</section>
<section id="sklearn-naive-bayes">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.naive_bayes" title="sklearn.naive_bayes"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.naive_bayes</span></code></a><a class="headerlink" href="#sklearn-naive-bayes" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> A correctly formatted error message is shown in
<a class="reference internal" href="../modules/generated/sklearn.naive_bayes.CategoricalNB.html#sklearn.naive_bayes.CategoricalNB" title="sklearn.naive_bayes.CategoricalNB"><code class="xref py py-class docutils literal notranslate"><span class="pre">naive_bayes.CategoricalNB</span></code></a> when the number of features in the input
differs between <code class="docutils literal notranslate"><span class="pre">predict</span></code> and <code class="docutils literal notranslate"><span class="pre">fit</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16090">#16090</a> by <a class="reference external" href="https://github.com/madhuracj">Madhura Jayaratne</a>.</p></li>
</ul>
</section>
<section id="sklearn-neural-network">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.neural_network" title="sklearn.neural_network"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.neural_network</span></code></a><a class="headerlink" href="#sklearn-neural-network" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.neural_network.MLPClassifier.html#sklearn.neural_network.MLPClassifier" title="sklearn.neural_network.MLPClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">neural_network.MLPClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.neural_network.MLPRegressor.html#sklearn.neural_network.MLPRegressor" title="sklearn.neural_network.MLPRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">neural_network.MLPRegressor</span></code></a> has reduced memory footprint when using
stochastic solvers, <code class="docutils literal notranslate"><span class="pre">'sgd'</span></code> or <code class="docutils literal notranslate"><span class="pre">'adam'</span></code>, and <code class="docutils literal notranslate"><span class="pre">shuffle=True</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14075">#14075</a> by
<a class="reference external" href="https://github.com/meyer89">@meyer89</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Increases the numerical stability of the logistic loss function in
<a class="reference internal" href="../modules/generated/sklearn.neural_network.MLPClassifier.html#sklearn.neural_network.MLPClassifier" title="sklearn.neural_network.MLPClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">neural_network.MLPClassifier</span></code></a> by clipping the probabilities.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16117">#16117</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="id11">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.inspection" title="sklearn.inspection"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.inspection</span></code></a><a class="headerlink" href="#id11" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.inspection.PartialDependenceDisplay.html#sklearn.inspection.PartialDependenceDisplay" title="sklearn.inspection.PartialDependenceDisplay"><code class="xref py py-class docutils literal notranslate"><span class="pre">inspection.PartialDependenceDisplay</span></code></a> now exposes the
deciles lines as attributes so they can be hidden or customized. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15785">#15785</a>
by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a></p></li>
</ul>
</section>
<section id="sklearn-preprocessing">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.preprocessing" title="sklearn.preprocessing"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.preprocessing</span></code></a><a class="headerlink" href="#sklearn-preprocessing" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Feature</span></span> argument <code class="docutils literal notranslate"><span class="pre">drop</span></code> of <a class="reference internal" href="../modules/generated/sklearn.preprocessing.OneHotEncoder.html#sklearn.preprocessing.OneHotEncoder" title="sklearn.preprocessing.OneHotEncoder"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.OneHotEncoder</span></code></a>
will now accept value ‘if_binary’ and will drop the first category of
each feature with two categories. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16245">#16245</a>
by <a class="reference external" href="https://github.com/rushabh-v">Rushabh Vasani</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.preprocessing.OneHotEncoder.html#sklearn.preprocessing.OneHotEncoder" title="sklearn.preprocessing.OneHotEncoder"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.OneHotEncoder</span></code></a>’s <code class="docutils literal notranslate"><span class="pre">drop_idx_</span></code> ndarray
can now contain <code class="docutils literal notranslate"><span class="pre">None</span></code>, where <code class="docutils literal notranslate"><span class="pre">drop_idx_[i]</span> <span class="pre">=</span> <span class="pre">None</span></code> means that no category
is dropped for index <code class="docutils literal notranslate"><span class="pre">i</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16585">#16585</a> by <a class="reference external" href="https://github.com/cmarmo">Chiara Marmo</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler" title="sklearn.preprocessing.MaxAbsScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.MaxAbsScaler</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.MinMaxScaler.html#sklearn.preprocessing.MinMaxScaler" title="sklearn.preprocessing.MinMaxScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.MinMaxScaler</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.StandardScaler</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.PowerTransformer.html#sklearn.preprocessing.PowerTransformer" title="sklearn.preprocessing.PowerTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.PowerTransformer</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.QuantileTransformer.html#sklearn.preprocessing.QuantileTransformer" title="sklearn.preprocessing.QuantileTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.QuantileTransformer</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.RobustScaler.html#sklearn.preprocessing.RobustScaler" title="sklearn.preprocessing.RobustScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.RobustScaler</span></code></a> now supports pandas’ nullable integer
dtype with missing values. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16508">#16508</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.preprocessing.OneHotEncoder.html#sklearn.preprocessing.OneHotEncoder" title="sklearn.preprocessing.OneHotEncoder"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.OneHotEncoder</span></code></a> is now faster at
transforming. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15762">#15762</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fix a bug in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.StandardScaler</span></code></a> which was incorrectly
computing statistics when calling <code class="docutils literal notranslate"><span class="pre">partial_fit</span></code> on sparse inputs.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16466">#16466</a> by <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fix a bug in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.Normalizer.html#sklearn.preprocessing.Normalizer" title="sklearn.preprocessing.Normalizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.Normalizer</span></code></a> with norm=’max’,
which was not taking the absolute value of the maximum values before
normalizing the vectors. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16632">#16632</a> by
<a class="reference external" href="https://github.com/Maupin1991">Maura Pintor</a> and <a class="reference external" href="https://github.com/bbiggio">Battista Biggio</a>.</p></li>
</ul>
</section>
<section id="sklearn-semi-supervised">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.semi_supervised" title="sklearn.semi_supervised"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.semi_supervised</span></code></a><a class="headerlink" href="#sklearn-semi-supervised" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.semi_supervised.LabelSpreading.html#sklearn.semi_supervised.LabelSpreading" title="sklearn.semi_supervised.LabelSpreading"><code class="xref py py-class docutils literal notranslate"><span class="pre">semi_supervised.LabelSpreading</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.semi_supervised.LabelPropagation.html#sklearn.semi_supervised.LabelPropagation" title="sklearn.semi_supervised.LabelPropagation"><code class="xref py py-class docutils literal notranslate"><span class="pre">semi_supervised.LabelPropagation</span></code></a> avoids divide by zero warnings
when normalizing <code class="docutils literal notranslate"><span class="pre">label_distributions_</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15946">#15946</a> by <a class="reference external" href="https://github.com/ngshya">@ngshya</a>.</p></li>
</ul>
</section>
<section id="sklearn-svm">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.svm" title="sklearn.svm"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.svm</span></code></a><a class="headerlink" href="#sklearn-svm" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <span class="raw-html"><span class="badge badge-info">Efficiency</span></span> Improved <code class="docutils literal notranslate"><span class="pre">libsvm</span></code> and <code class="docutils literal notranslate"><span class="pre">liblinear</span></code> random number
generators used to randomly select coordinates in the coordinate descent
algorithms. Platform-dependent C <code class="docutils literal notranslate"><span class="pre">rand()</span></code> was used, which is only able to
generate numbers up to <code class="docutils literal notranslate"><span class="pre">32767</span></code> on windows platform (see this <a class="reference external" href="https://codeforces.com/blog/entry/61587">blog
post</a>) and also has poor
randomization power as suggested by <a class="reference external" href="https://channel9.msdn.com/Events/GoingNative/2013/rand-Considered-Harmful">this presentation</a>.
It was replaced with C++11 <code class="docutils literal notranslate"><span class="pre">mt19937</span></code>, a Mersenne Twister that correctly
generates 31bits/63bits random numbers on all platforms. In addition, the
crude “modulo” postprocessor used to get a random number in a bounded
interval was replaced by the tweaked Lemire method as suggested by <a class="reference external" href="http://www.pcg-random.org/posts/bounded-rands.html">this blog
post</a>.
Any model using the <code class="xref py py-func docutils literal notranslate"><span class="pre">svm.libsvm</span></code> or the <code class="xref py py-func docutils literal notranslate"><span class="pre">svm.liblinear</span></code> solver,
including <a class="reference internal" href="../modules/generated/sklearn.svm.LinearSVC.html#sklearn.svm.LinearSVC" title="sklearn.svm.LinearSVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.LinearSVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.LinearSVR.html#sklearn.svm.LinearSVR" title="sklearn.svm.LinearSVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.LinearSVR</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.svm.NuSVC.html#sklearn.svm.NuSVC" title="sklearn.svm.NuSVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.NuSVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.NuSVR.html#sklearn.svm.NuSVR" title="sklearn.svm.NuSVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.NuSVR</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM" title="sklearn.svm.OneClassSVM"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.OneClassSVM</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.svm.SVC.html#sklearn.svm.SVC" title="sklearn.svm.SVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVC</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.svm.SVR.html#sklearn.svm.SVR" title="sklearn.svm.SVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVR</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.linear_model.LogisticRegression.html#sklearn.linear_model.LogisticRegression" title="sklearn.linear_model.LogisticRegression"><code class="xref py py-class docutils literal notranslate"><span class="pre">linear_model.LogisticRegression</span></code></a>,
is affected. In particular users can expect a better convergence when the
number of samples (LibSVM) or the number of features (LibLinear) is large.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/13511">#13511</a> by <a class="reference external" href="https://github.com/smarie">Sylvain Marié</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fix use of custom kernel not taking float entries such as string
kernels in <a class="reference internal" href="../modules/generated/sklearn.svm.SVC.html#sklearn.svm.SVC" title="sklearn.svm.SVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVC</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.svm.SVR.html#sklearn.svm.SVR" title="sklearn.svm.SVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVR</span></code></a>. Note that custom kennels
are now expected to validate their input where they previously received
valid numeric arrays.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/11296">#11296</a> by <a class="reference external" href="http://alexandre.gramfort.net">Alexandre Gramfort</a> and <a class="reference external" href="https://github.com/georgipeev">Georgi Peev</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> <a class="reference internal" href="../modules/generated/sklearn.svm.SVR.html#sklearn.svm.SVR" title="sklearn.svm.SVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.SVR</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.svm.OneClassSVM.html#sklearn.svm.OneClassSVM" title="sklearn.svm.OneClassSVM"><code class="xref py py-class docutils literal notranslate"><span class="pre">svm.OneClassSVM</span></code></a> attributes, <code class="docutils literal notranslate"><span class="pre">probA_</span></code> and
<code class="docutils literal notranslate"><span class="pre">probB_</span></code>, are now deprecated as they were not useful. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15558">#15558</a> by
<a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="sklearn-tree">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.tree" title="sklearn.tree"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.tree</span></code></a><a class="headerlink" href="#sklearn-tree" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <a class="reference internal" href="../modules/generated/sklearn.tree.plot_tree.html#sklearn.tree.plot_tree" title="sklearn.tree.plot_tree"><code class="xref py py-func docutils literal notranslate"><span class="pre">tree.plot_tree</span></code></a> <code class="docutils literal notranslate"><span class="pre">rotate</span></code> parameter was unused and has been
deprecated.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15806">#15806</a> by <a class="reference external" href="https://github.com/cmarmo">Chiara Marmo</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> Fix support of read-only float32 array input in <code class="docutils literal notranslate"><span class="pre">predict</span></code>,
<code class="docutils literal notranslate"><span class="pre">decision_path</span></code> and <code class="docutils literal notranslate"><span class="pre">predict_proba</span></code> methods of
<a class="reference internal" href="../modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.DecisionTreeClassifier</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.tree.ExtraTreeClassifier.html#sklearn.tree.ExtraTreeClassifier" title="sklearn.tree.ExtraTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.ExtraTreeClassifier</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.GradientBoostingClassifier.html#sklearn.ensemble.GradientBoostingClassifier" title="sklearn.ensemble.GradientBoostingClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.GradientBoostingClassifier</span></code></a> as well as <code class="docutils literal notranslate"><span class="pre">predict</span></code> method of
<a class="reference internal" href="../modules/generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.DecisionTreeRegressor</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.tree.ExtraTreeRegressor.html#sklearn.tree.ExtraTreeRegressor" title="sklearn.tree.ExtraTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">tree.ExtraTreeRegressor</span></code></a>, and
<a class="reference internal" href="../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">ensemble.GradientBoostingRegressor</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16331">#16331</a> by <a class="reference external" href="https://github.com/batalex">Alexandre Batisse</a>.</p></li>
</ul>
</section>
<section id="id12">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.utils" title="sklearn.utils"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.utils</span></code></a><a class="headerlink" href="#id12" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> Estimators can now be displayed with a rich html
representation. This can be enabled in Jupyter notebooks by setting
<code class="docutils literal notranslate"><span class="pre">display='diagram'</span></code> in <a class="reference internal" href="../modules/generated/sklearn.set_config.html#sklearn.set_config" title="sklearn.set_config"><code class="xref py py-func docutils literal notranslate"><span class="pre">set_config</span></code></a>. The raw html can be
returned by using <a class="reference internal" href="../modules/generated/sklearn.utils.estimator_html_repr.html#sklearn.utils.estimator_html_repr" title="sklearn.utils.estimator_html_repr"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.estimator_html_repr</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14180">#14180</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> improve error message in <a class="reference internal" href="../modules/generated/sklearn.utils.validation.column_or_1d.html#sklearn.utils.validation.column_or_1d" title="sklearn.utils.validation.column_or_1d"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.validation.column_or_1d</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15926">#15926</a> by <a class="reference external" href="https://github.com/lesteve">Loïc Estève</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> add warning in <a class="reference internal" href="../modules/generated/sklearn.utils.check_array.html#sklearn.utils.check_array" title="sklearn.utils.check_array"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.check_array</span></code></a> for
pandas sparse DataFrame.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16021">#16021</a> by <a class="reference external" href="https://github.com/rushabh-v">Rushabh Vasani</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <a class="reference internal" href="../modules/generated/sklearn.utils.check_array.html#sklearn.utils.check_array" title="sklearn.utils.check_array"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.check_array</span></code></a> now constructs a sparse
matrix from a pandas DataFrame that contains only <code class="docutils literal notranslate"><span class="pre">SparseArray</span></code> columns.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16728">#16728</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <code class="xref py py-func docutils literal notranslate"><span class="pre">utils.validation.check_array</span></code> supports pandas’
nullable integer dtype with missing values when <code class="docutils literal notranslate"><span class="pre">force_all_finite</span></code> is set to
<code class="docutils literal notranslate"><span class="pre">False</span></code> or <code class="docutils literal notranslate"><span class="pre">'allow-nan'</span></code> in which case the data is converted to floating
point values where <code class="docutils literal notranslate"><span class="pre">pd.NA</span></code> values are replaced by <code class="docutils literal notranslate"><span class="pre">np.nan</span></code>. As a consequence,
all <a class="reference internal" href="../modules/classes.html#module-sklearn.preprocessing" title="sklearn.preprocessing"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.preprocessing</span></code></a> transformers that accept numeric inputs with
missing values represented as <code class="docutils literal notranslate"><span class="pre">np.nan</span></code> now also accepts being directly fed
pandas dataframes with <code class="docutils literal notranslate"><span class="pre">pd.Int*</span> <span class="pre">or</span> <span class="pre">`pd.Uint*</span></code> typed columns that use <code class="docutils literal notranslate"><span class="pre">pd.NA</span></code>
as a missing value marker. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16508">#16508</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Passing classes to <a class="reference internal" href="../modules/generated/sklearn.utils.estimator_checks.check_estimator.html#sklearn.utils.estimator_checks.check_estimator" title="sklearn.utils.estimator_checks.check_estimator"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.estimator_checks.check_estimator</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.utils.estimator_checks.parametrize_with_checks.html#sklearn.utils.estimator_checks.parametrize_with_checks" title="sklearn.utils.estimator_checks.parametrize_with_checks"><code class="xref py py-func docutils literal notranslate"><span class="pre">utils.estimator_checks.parametrize_with_checks</span></code></a> is now deprecated,
and support for classes will be removed in 0.24. Pass instances instead.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17032">#17032</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The private utility <code class="docutils literal notranslate"><span class="pre">_safe_tags</span></code> in <code class="docutils literal notranslate"><span class="pre">utils.estimator_checks</span></code> was
removed, hence all tags should be obtained through <code class="docutils literal notranslate"><span class="pre">estimator._get_tags()</span></code>.
Note that Mixins like <code class="docutils literal notranslate"><span class="pre">RegressorMixin</span></code> must come <em>before</em> base classes
in the MRO for <code class="docutils literal notranslate"><span class="pre">_get_tags()</span></code> to work properly.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16950">#16950</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-danger">Fix</span></span> <code class="xref py py-func docutils literal notranslate"><span class="pre">utils.all_estimators</span></code> now only returns public estimators.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/15380">#15380</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
</ul>
</section>
<section id="id13">
<h3>Miscellaneous<a class="headerlink" href="#id13" title="Permalink to this heading">¶</a></h3>
<ul class="simple">
<li><p><span class="raw-html"><span class="badge badge-success">Major Feature</span></span> Adds a HTML representation of estimators to be shown in
a jupyter notebook or lab. This visualization is acitivated by setting the
<code class="docutils literal notranslate"><span class="pre">display</span></code> option in <a class="reference internal" href="../modules/generated/sklearn.set_config.html#sklearn.set_config" title="sklearn.set_config"><code class="xref py py-func docutils literal notranslate"><span class="pre">sklearn.set_config</span></code></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/14180">#14180</a> by
<a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-info">Enhancement</span></span> <code class="docutils literal notranslate"><span class="pre">scikit-learn</span></code> now works with <code class="docutils literal notranslate"><span class="pre">mypy</span></code> without errors.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16726">#16726</a> by <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Most estimators now expose a <code class="docutils literal notranslate"><span class="pre">n_features_in_</span></code> attribute. This
attribute is equal to the number of features passed to the <code class="docutils literal notranslate"><span class="pre">fit</span></code> method.
See <a class="reference external" href="https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep010/proposal.html">SLEP010</a>
for details. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16112">#16112</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> Estimators now have a <code class="docutils literal notranslate"><span class="pre">requires_y</span></code> tags which is False by default
except for estimators that inherit from <code class="docutils literal notranslate"><span class="pre">~sklearn.base.RegressorMixin</span></code> or
<code class="docutils literal notranslate"><span class="pre">~sklearn.base.ClassifierMixin</span></code>. This tag is used to ensure that a proper
error message is raised when y was expected but None was passed.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/16622">#16622</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
<li><p><span class="raw-html"><span class="badge badge-warning">API Change</span></span> The default setting <code class="docutils literal notranslate"><span class="pre">print_changed_only</span></code> has been changed from False
to True. This means that the <code class="docutils literal notranslate"><span class="pre">repr</span></code> of estimators is now more concise and
only shows the parameters whose default value has been changed when
printing an estimator. You can restore the previous behaviour by using
<code class="docutils literal notranslate"><span class="pre">sklearn.set_config(print_changed_only=False)</span></code>. Also, note that it is
always possible to quickly inspect the parameters of any estimator using
<code class="docutils literal notranslate"><span class="pre">est.get_params(deep=False)</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/pull/17061">#17061</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</p></li>
</ul>
</section>
</section>
<section id="code-and-documentation-contributors">
<h2>Code and Documentation Contributors<a class="headerlink" href="#code-and-documentation-contributors" title="Permalink to this heading">¶</a></h2>
<p>Thanks to everyone who has contributed to the maintenance and improvement of the
project since version 0.22, including:</p>
<p>Abbie Popa, Adrin Jalali, Aleksandra Kocot, Alexandre Batisse, Alexandre
Gramfort, Alex Henrie, Alex Itkes, Alex Liang, alexshacked, Alonso Silva
Allende, Ana Casado, Andreas Mueller, Angela Ambroz, Ankit810, Arie Pratama
Sutiono, Arunav Konwar, Baptiste Maingret, Benjamin Beier Liu, bernie gray,
Bharathi Srinivasan, Bharat Raghunathan, Bibhash Chandra Mitra, Brian Wignall,
brigi, Brigitta Sipőcz, Carlos H Brandt, CastaChick, castor, cgsavard, Chiara
Marmo, Chris Gregory, Christian Kastner, Christian Lorentzen, Corrie
Bartelheimer, Daniël van Gelder, Daphne, David Breuer, david-cortes, dbauer9,
Divyaprabha M, Edward Qian, Ekaterina Borovikova, ELNS, Emily Taylor, Erich
Schubert, Eric Leung, Evgeni Chasnovski, Fabiana, Facundo Ferrín, Fan,
Franziska Boenisch, Gael Varoquaux, Gaurav Sharma, Geoffrey Bolmier, Georgi
Peev, gholdman1, Gonthier Nicolas, Gregory Morse, Gregory R. Lee, Guillaume
Lemaitre, Gui Miotto, Hailey Nguyen, Hanmin Qin, Hao Chun Chang, HaoYin, Hélion
du Mas des Bourboux, Himanshu Garg, Hirofumi Suzuki, huangk10, Hugo van
Kemenade, Hye Sung Jung, indecisiveuser, inderjeet, J-A16, Jérémie du
Boisberranger, Jin-Hwan CHO, JJmistry, Joel Nothman, Johann Faouzi, Jon Haitz