-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathv0.20.html
1831 lines (1802 loc) · 278 KB
/
v0.20.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 PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Version 0.20.4 — scikit-learn 0.20.4 documentation</title>
<!-- htmltitle is before nature.css - we use this hack to load bootstrap first -->
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<link rel="stylesheet" href="../_static/css/bootstrap.min.css" media="screen" />
<link rel="stylesheet" href="../_static/css/bootstrap-responsive.css"/>
<link rel="stylesheet" href="../_static/nature.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/gallery.css" type="text/css" />
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../',
VERSION: '0.20.4',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: '.txt'
};
</script>
<script type="text/javascript" src="../_static/jquery.js"></script>
<script type="text/javascript" src="../_static/underscore.js"></script>
<script type="text/javascript" src="../_static/doctools.js"></script>
<script type="text/javascript" src="../_static/js/copybutton.js"></script>
<script type="text/javascript" src="../_static/js/extra.js"></script>
<script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS_SVG"></script>
<link rel="shortcut icon" href="../_static/favicon.ico"/>
<link rel="author" title="About these documents" href="../about.html" />
<link rel="search" title="Search" href="../search.html" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<script src="../_static/js/bootstrap.min.js" type="text/javascript"></script>
<script>
VERSION_SUBDIR = (function(groups) {
return groups ? groups[1] : null;
})(location.href.match(/^https?:\/\/scikit-learn.org\/([^\/]+)/));
</script>
<link rel="canonical" href="https://scikit-learn.org/stable/whats_new/v0.20.html" />
<script type="text/javascript">
$("div.buttonNext, div.buttonPrevious").hover(
function () {
$(this).css('background-color', '#FF9C34');
},
function () {
$(this).css('background-color', '#A7D6E2');
}
);
function showMenu() {
var topNav = document.getElementById("scikit-navbar");
if (topNav.className === "navbar") {
topNav.className += " responsive";
} else {
topNav.className = "navbar";
}
};
</script>
</head>
<body>
<div class="header-wrapper">
<div class="header">
<p class="logo"><a href="../index.html">
<img src="../_static/scikit-learn-logo-small.png" alt="Logo"/>
</a>
</p><div class="navbar" id="scikit-navbar">
<ul>
<li><a href="../index.html">Home</a></li>
<li><a href="../install.html">Installation</a></li>
<li class="btn-li"><div class="btn-group">
<a href="../documentation.html">Documentation</a>
<a class="btn dropdown-toggle" data-toggle="dropdown">
<span class="caret"></span>
</a>
<ul class="dropdown-menu">
<li class="link-title">Scikit-learn <script>document.write(DOCUMENTATION_OPTIONS.VERSION + (VERSION_SUBDIR ? " (" + VERSION_SUBDIR + ")" : ""));</script></li>
<li><a href="../tutorial/index.html">Tutorials</a></li>
<li><a href="../user_guide.html">User guide</a></li>
<li><a href="../modules/classes.html">API</a></li>
<li><a href="../glossary.html">Glossary</a></li>
<li><a href="../faq.html">FAQ</a></li>
<li><a href="../developers/index.html">Development</a></li>
<li><a href="../roadmap.html">Roadmap</a></li>
<li class="divider"></li>
<script>if (VERSION_SUBDIR != "stable") document.write('<li><a href="http://scikit-learn.org/stable/documentation.html">Stable version</a></li>')</script>
<script>if (VERSION_SUBDIR != "dev") document.write('<li><a href="http://scikit-learn.org/dev/documentation.html">Development version</a></li>')</script>
<li><a href="http://scikit-learn.org/dev/versions.html">All available versions</a></li>
<li><a href="../_downloads/scikit-learn-docs.pdf">PDF documentation</a></li>
</ul>
</div>
</li>
<li><a href="../auto_examples/index.html">Examples</a></li>
</ul>
<a href="javascript:void(0);" onclick="showMenu()">
<div class="nav-icon">
<div class="hamburger-line"></div>
<div class="hamburger-line"></div>
<div class="hamburger-line"></div>
</div>
</a>
<div class="search_form">
<div class="gcse-search" id="cse" style="width: 100%;"></div>
</div>
</div> <!-- end navbar --></div>
</div>
<!-- GitHub "fork me" ribbon -->
<a href="https://github.com/scikit-learn/scikit-learn">
<img class="fork-me"
style="position: absolute; top: 0; right: 0; border: 0;"
src="../_static/img/forkme.png"
alt="Fork me on GitHub" />
</a>
<div class="content-wrapper">
<div class="sphinxsidebar">
<div class="sphinxsidebarwrapper">
<p class="doc-version"><b>scikit-learn v0.20.4</b><br/>
<a href="http://scikit-learn.org/dev/versions.html">Other versions</a></p>
<p class="citing">Please <b><a href="../about.html#citing-scikit-learn" style="font-size: 110%;">cite us </a></b>if you use the software.</p>
<ul>
<li><a class="reference internal" href="#">Version 0.20.4</a><ul>
<li><a class="reference internal" href="#changelog">Changelog</a><ul>
<li><a class="reference internal" href="#sklearn-cluster"><code class="docutils literal"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#sklearn-compose"><code class="docutils literal"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#sklearn-decomposition"><code class="docutils literal"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#sklearn-model-selection"><code class="docutils literal"><span class="pre">sklearn.model_selection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-neighbors"><code class="docutils literal"><span class="pre">sklearn.neighbors</span></code></a></li>
</ul>
</li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-20-3">Version 0.20.3</a><ul>
<li><a class="reference internal" href="#id5">Changelog</a><ul>
<li><a class="reference internal" href="#id6"><code class="docutils literal"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#id7"><code class="docutils literal"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#sklearn-covariance"><code class="docutils literal"><span class="pre">sklearn.covariance</span></code></a></li>
<li><a class="reference internal" href="#id8"><code class="docutils literal"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#sklearn-datasets"><code class="docutils literal"><span class="pre">sklearn.datasets</span></code></a></li>
<li><a class="reference internal" href="#sklearn-feature-extraction"><code class="docutils literal"><span class="pre">sklearn.feature_extraction</span></code></a></li>
<li><a class="reference internal" href="#sklearn-impute"><code class="docutils literal"><span class="pre">sklearn.impute</span></code></a></li>
<li><a class="reference internal" href="#sklearn-linear-model"><code class="docutils literal"><span class="pre">sklearn.linear_model</span></code></a></li>
<li><a class="reference internal" href="#sklearn-preprocessing"><code class="docutils literal"><span class="pre">sklearn.preprocessing</span></code></a></li>
<li><a class="reference internal" href="#sklearn-svm"><code class="docutils literal"><span class="pre">sklearn.svm</span></code></a></li>
</ul>
</li>
<li><a class="reference internal" href="#code-and-documentation-contributors">Code and Documentation Contributors</a></li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-20-2">Version 0.20.2</a><ul>
<li><a class="reference internal" href="#changed-models">Changed models</a></li>
<li><a class="reference internal" href="#id9">Changelog</a><ul>
<li><a class="reference internal" href="#id10"><code class="docutils literal"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#sklearn-metrics"><code class="docutils literal"><span class="pre">sklearn.metrics</span></code></a></li>
<li><a class="reference internal" href="#id11"><code class="docutils literal"><span class="pre">sklearn.neighbors</span></code></a></li>
<li><a class="reference internal" href="#sklearn-utils"><code class="docutils literal"><span class="pre">sklearn.utils</span></code></a></li>
</ul>
</li>
<li><a class="reference internal" href="#id12">Code and Documentation Contributors</a></li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-20-1">Version 0.20.1</a><ul>
<li><a class="reference internal" href="#id13">Changed models</a></li>
<li><a class="reference internal" href="#id14">Changelog</a><ul>
<li><a class="reference internal" href="#id15"><code class="docutils literal"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#id16"><code class="docutils literal"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#id17"><code class="docutils literal"><span class="pre">sklearn.datasets</span></code></a></li>
<li><a class="reference internal" href="#id18"><code class="docutils literal"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#sklearn-ensemble"><code class="docutils literal"><span class="pre">sklearn.ensemble</span></code></a></li>
<li><a class="reference internal" href="#id19"><code class="docutils literal"><span class="pre">sklearn.feature_extraction</span></code></a></li>
<li><a class="reference internal" href="#id20"><code class="docutils literal"><span class="pre">sklearn.linear_model</span></code></a></li>
<li><a class="reference internal" href="#id21"><code class="docutils literal"><span class="pre">sklearn.metrics</span></code></a></li>
<li><a class="reference internal" href="#sklearn-mixture"><code class="docutils literal"><span class="pre">sklearn.mixture</span></code></a></li>
<li><a class="reference internal" href="#id22"><code class="docutils literal"><span class="pre">sklearn.neighbors</span></code></a></li>
<li><a class="reference internal" href="#id23"><code class="docutils literal"><span class="pre">sklearn.preprocessing</span></code></a></li>
<li><a class="reference internal" href="#id24"><code class="docutils literal"><span class="pre">sklearn.utils</span></code></a></li>
<li><a class="reference internal" href="#miscellaneous">Miscellaneous</a></li>
</ul>
</li>
<li><a class="reference internal" href="#id25">Code and Documentation Contributors</a></li>
</ul>
</li>
<li><a class="reference internal" href="#version-0-20-0">Version 0.20.0</a><ul>
<li><a class="reference internal" href="#highlights">Highlights</a></li>
<li><a class="reference internal" href="#id26">Changed models</a></li>
<li><a class="reference internal" href="#known-major-bugs">Known Major Bugs</a></li>
<li><a class="reference internal" href="#id27">Changelog</a><ul>
<li><a class="reference internal" href="#id28"><code class="docutils literal"><span class="pre">sklearn.cluster</span></code></a></li>
<li><a class="reference internal" href="#id29"><code class="docutils literal"><span class="pre">sklearn.compose</span></code></a></li>
<li><a class="reference internal" href="#id30"><code class="docutils literal"><span class="pre">sklearn.covariance</span></code></a></li>
<li><a class="reference internal" href="#id31"><code class="docutils literal"><span class="pre">sklearn.datasets</span></code></a></li>
<li><a class="reference internal" href="#id32"><code class="docutils literal"><span class="pre">sklearn.decomposition</span></code></a></li>
<li><a class="reference internal" href="#sklearn-discriminant-analysis"><code class="docutils literal"><span class="pre">sklearn.discriminant_analysis</span></code></a></li>
<li><a class="reference internal" href="#sklearn-dummy"><code class="docutils literal"><span class="pre">sklearn.dummy</span></code></a></li>
<li><a class="reference internal" href="#id33"><code class="docutils literal"><span class="pre">sklearn.ensemble</span></code></a></li>
<li><a class="reference internal" href="#id34"><code class="docutils literal"><span class="pre">sklearn.feature_extraction</span></code></a></li>
<li><a class="reference internal" href="#sklearn-feature-selection"><code class="docutils literal"><span class="pre">sklearn.feature_selection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-gaussian-process"><code class="docutils literal"><span class="pre">sklearn.gaussian_process</span></code></a></li>
<li><a class="reference internal" href="#id35"><code class="docutils literal"><span class="pre">sklearn.impute</span></code></a></li>
<li><a class="reference internal" href="#sklearn-isotonic"><code class="docutils literal"><span class="pre">sklearn.isotonic</span></code></a></li>
<li><a class="reference internal" href="#id36"><code class="docutils literal"><span class="pre">sklearn.linear_model</span></code></a></li>
<li><a class="reference internal" href="#sklearn-manifold"><code class="docutils literal"><span class="pre">sklearn.manifold</span></code></a></li>
<li><a class="reference internal" href="#id37"><code class="docutils literal"><span class="pre">sklearn.metrics</span></code></a></li>
<li><a class="reference internal" href="#id38"><code class="docutils literal"><span class="pre">sklearn.mixture</span></code></a></li>
<li><a class="reference internal" href="#id39"><code class="docutils literal"><span class="pre">sklearn.model_selection</span></code></a></li>
<li><a class="reference internal" href="#sklearn-multioutput"><code class="docutils literal"><span class="pre">sklearn.multioutput</span></code></a></li>
<li><a class="reference internal" href="#sklearn-naive-bayes"><code class="docutils literal"><span class="pre">sklearn.naive_bayes</span></code></a></li>
<li><a class="reference internal" href="#id40"><code class="docutils literal"><span class="pre">sklearn.neighbors</span></code></a></li>
<li><a class="reference internal" href="#sklearn-neural-network"><code class="docutils literal"><span class="pre">sklearn.neural_network</span></code></a></li>
<li><a class="reference internal" href="#sklearn-pipeline"><code class="docutils literal"><span class="pre">sklearn.pipeline</span></code></a></li>
<li><a class="reference internal" href="#id41"><code class="docutils literal"><span class="pre">sklearn.preprocessing</span></code></a></li>
<li><a class="reference internal" href="#id42"><code class="docutils literal"><span class="pre">sklearn.svm</span></code></a></li>
<li><a class="reference internal" href="#sklearn-tree"><code class="docutils literal"><span class="pre">sklearn.tree</span></code></a></li>
<li><a class="reference internal" href="#id43"><code class="docutils literal"><span class="pre">sklearn.utils</span></code></a></li>
<li><a class="reference internal" href="#multiple-modules">Multiple modules</a></li>
<li><a class="reference internal" href="#id44">Miscellaneous</a></li>
</ul>
</li>
<li><a class="reference internal" href="#changes-to-estimator-checks">Changes to estimator checks</a></li>
<li><a class="reference internal" href="#id45">Code and Documentation Contributors</a></li>
</ul>
</li>
</ul>
</div>
</div>
<input type="checkbox" id="nav-trigger" class="nav-trigger" checked />
<label for="nav-trigger"></label>
<div class="content">
<div class="documentwrapper">
<div class="bodywrapper">
<div class="body" role="main">
<div class="section" id="version-0-20-4">
<span id="changes-0-20-4"></span><h1>Version 0.20.4<a class="headerlink" href="#version-0-20-4" title="Permalink to this headline">¶</a></h1>
<p><strong>July 30, 2019</strong></p>
<p>This is a bug-fix release with some bug fixes applied to version 0.20.3.</p>
<div class="section" id="changelog">
<h2>Changelog<a class="headerlink" href="#changelog" title="Permalink to this headline">¶</a></h2>
<p>The bundled version of joblib was upgraded from 0.13.0 to 0.13.2.</p>
<div class="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"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#sklearn-cluster" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-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"><span class="pre">cluster.KMeans</span></code></a> where KMeans++ initialisation
could rarely result in an IndexError. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11756">#11756</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#sklearn-compose" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed an issue in <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"><span class="pre">compose.ColumnTransformer</span></code></a> where using
DataFrames whose column order differs between :func:<code class="docutils literal"><span class="pre">fit</span></code> and
:func:<code class="docutils literal"><span class="pre">transform</span></code> could lead to silently passing incorrect columns to the
<code class="docutils literal"><span class="pre">remainder</span></code> transformer.
<a href="#id1"><span class="problematic" id="id2">:pr:`14237`</span></a> by <cite>Andreas Schuderer <schuderer></cite>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#sklearn-decomposition" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cross_decomposition.CCA.html#sklearn.cross_decomposition.CCA" title="sklearn.cross_decomposition.CCA"><code class="xref py py-class docutils literal"><span class="pre">cross_decomposition.CCA</span></code></a> improving numerical
stability when <cite>Y</cite> is close to zero. <a href="#id3"><span class="problematic" id="id4">:pr:`13903`</span></a> by <a href="#id46"><span class="problematic" id="id47">`Thomas Fan`_</span></a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.model_selection</span></code></a><a class="headerlink" href="#sklearn-model-selection" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug where <a class="reference internal" href="../modules/generated/sklearn.model_selection.StratifiedKFold.html#sklearn.model_selection.StratifiedKFold" title="sklearn.model_selection.StratifiedKFold"><code class="xref py py-class docutils literal"><span class="pre">model_selection.StratifiedKFold</span></code></a>
shuffles each class’s samples with the same <code class="docutils literal"><span class="pre">random_state</span></code>,
making <code class="docutils literal"><span class="pre">shuffle=True</span></code> ineffective.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13124">#13124</a> by <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
</ul>
</div>
<div class="section" id="sklearn-neighbors">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.neighbors" title="sklearn.neighbors"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.neighbors</span></code></a><a class="headerlink" href="#sklearn-neighbors" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.neighbors.KernelDensity.html#sklearn.neighbors.KernelDensity" title="sklearn.neighbors.KernelDensity"><code class="xref py py-class docutils literal"><span class="pre">neighbors.KernelDensity</span></code></a> which could not be
restored from a pickle if <code class="docutils literal"><span class="pre">sample_weight</span></code> had been used.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13772">#13772</a> by <a class="reference external" href="https://github.com/aditya1702">Aditya Vyas</a>.</li>
</ul>
<blockquote>
<div></div></blockquote>
</div>
</div>
</div>
<div class="section" id="version-0-20-3">
<span id="changes-0-20-3"></span><h1>Version 0.20.3<a class="headerlink" href="#version-0-20-3" title="Permalink to this headline">¶</a></h1>
<p><strong>March 1, 2019</strong></p>
<p>This is a bug-fix release with some minor documentation improvements and
enhancements to features released in 0.20.0.</p>
<div class="section" id="id5">
<h2>Changelog<a class="headerlink" href="#id5" title="Permalink to this headline">¶</a></h2>
<div class="section" id="id6">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#id6" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-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"><span class="pre">cluster.KMeans</span></code></a> where computation was single
threaded when <cite>n_jobs > 1</cite> or <cite>n_jobs = -1</cite>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12949">#12949</a> by <a class="reference external" href="https://github.com/nixphix">Prabakaran Kumaresshan</a>.</li>
</ul>
</div>
<div class="section" id="id7">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.compose" title="sklearn.compose"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#id7" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">compose.ColumnTransformer</span></code></a> to handle
negative indexes in the columns list of the transformers.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12946">#12946</a> by <a class="reference external" href="https://github.com/pierretallotte">Pierre Tallotte</a>.</li>
</ul>
</div>
<div class="section" id="sklearn-covariance">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.covariance" title="sklearn.covariance"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.covariance</span></code></a><a class="headerlink" href="#sklearn-covariance" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a regression in <a class="reference internal" href="../modules/generated/sklearn.covariance.graphical_lasso.html#sklearn.covariance.graphical_lasso" title="sklearn.covariance.graphical_lasso"><code class="xref py py-func docutils literal"><span class="pre">covariance.graphical_lasso</span></code></a> so that
the case <cite>n_features=2</cite> is handled correctly. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13276">#13276</a> by
<a class="reference external" href="https://github.com/bellet">Aurélien Bellet</a>.</li>
</ul>
</div>
<div class="section" id="id8">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.decomposition" title="sklearn.decomposition"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#id8" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.decomposition.sparse_encode.html#sklearn.decomposition.sparse_encode" title="sklearn.decomposition.sparse_encode"><code class="xref py py-func docutils literal"><span class="pre">decomposition.sparse_encode</span></code></a> where computation was single
threaded when <cite>n_jobs > 1</cite> or <cite>n_jobs = -1</cite>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13005">#13005</a> by <a class="reference external" href="https://github.com/nixphix">Prabakaran Kumaresshan</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.datasets</span></code></a><a class="headerlink" href="#sklearn-datasets" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-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"><span class="pre">sklearn.datasets.fetch_openml</span></code></a> now loads data by
streaming, avoiding high memory usage. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13312">#13312</a> by <a class="reference external" href="https://github.com/jorisvandenbossche">Joris Van den
Bossche</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.feature_extraction</span></code></a><a class="headerlink" href="#sklearn-feature-extraction" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a 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"><span class="pre">feature_extraction.text.CountVectorizer</span></code></a> which
would result in the sparse feature matrix having conflicting <cite>indptr</cite> and
<cite>indices</cite> precisions under very large vocabularies. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11295">#11295</a> by
<a class="reference external" href="https://github.com/gvacaliuc">Gabriel Vacaliuc</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.impute</span></code></a><a class="headerlink" href="#sklearn-impute" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> add support for non-numeric data in
<a class="reference internal" href="../modules/generated/sklearn.impute.MissingIndicator.html#sklearn.impute.MissingIndicator" title="sklearn.impute.MissingIndicator"><code class="xref py py-class docutils literal"><span class="pre">sklearn.impute.MissingIndicator</span></code></a> which was not supported while
<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"><span class="pre">sklearn.impute.SimpleImputer</span></code></a> was supporting this for some
imputation strategies.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13046">#13046</a> by <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.linear_model</span></code></a><a class="headerlink" href="#sklearn-linear-model" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">linear_model.MultiTaskElasticNet</span></code></a> and
<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"><span class="pre">linear_model.MultiTaskLasso</span></code></a> which were breaking when
<code class="docutils literal"><span class="pre">warm_start</span> <span class="pre">=</span> <span class="pre">True</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12360">#12360</a> by <a class="reference external" href="https://github.com/joaak">Aakanksha Joshi</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.preprocessing</span></code></a><a class="headerlink" href="#sklearn-preprocessing" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.KBinsDiscretizer.html#sklearn.preprocessing.KBinsDiscretizer" title="sklearn.preprocessing.KBinsDiscretizer"><code class="xref py py-class docutils literal"><span class="pre">preprocessing.KBinsDiscretizer</span></code></a> where
<code class="docutils literal"><span class="pre">strategy='kmeans'</span></code> fails with an error during transformation due to unsorted
bin edges. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13134">#13134</a> by <a class="reference external" href="https://github.com/SandroCasagrande">Sandro Casagrande</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">preprocessing.OneHotEncoder</span></code></a> where the
deprecation of <code class="docutils literal"><span class="pre">categorical_features</span></code> was handled incorrectly in
combination with <code class="docutils literal"><span class="pre">handle_unknown='ignore'</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12881">#12881</a> by <a class="reference external" href="https://github.com/jorisvandenbossche">Joris Van den Bossche</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Bins whose width are too small (i.e., <= 1e-8) are removed
with a warning in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.KBinsDiscretizer.html#sklearn.preprocessing.KBinsDiscretizer" title="sklearn.preprocessing.KBinsDiscretizer"><code class="xref py py-class docutils literal"><span class="pre">preprocessing.KBinsDiscretizer</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13165">#13165</a> by <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.svm</span></code></a><a class="headerlink" href="#sklearn-svm" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug 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"><span class="pre">svm.SVC</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"><span class="pre">svm.NuSVC</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"><span class="pre">svm.SVR</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"><span class="pre">svm.NuSVR</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"><span class="pre">svm.OneClassSVM</span></code></a> where the <code class="docutils literal"><span class="pre">scale</span></code> option
of parameter <code class="docutils literal"><span class="pre">gamma</span></code> is erroneously defined as
<code class="docutils literal"><span class="pre">1</span> <span class="pre">/</span> <span class="pre">(n_features</span> <span class="pre">*</span> <span class="pre">X.std())</span></code>. It’s now defined as
<code class="docutils literal"><span class="pre">1</span> <span class="pre">/</span> <span class="pre">(n_features</span> <span class="pre">*</span> <span class="pre">X.var())</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/13221">#13221</a> by <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
</ul>
</div>
</div>
<div class="section" id="code-and-documentation-contributors">
<h2>Code and Documentation Contributors<a class="headerlink" href="#code-and-documentation-contributors" title="Permalink to this headline">¶</a></h2>
<p>With thanks to:</p>
<p>Adrin Jalali, Agamemnon Krasoulis, Albert Thomas, Andreas Mueller, Aurélien
Bellet, bertrandhaut, Bharat Raghunathan, Dowon, Emmanuel Arias, Fibinse
Xavier, Finn O’Shea, Gabriel Vacaliuc, Gael Varoquaux, Guillaume Lemaitre,
Hanmin Qin, joaak, Joel Nothman, Joris Van den Bossche, Jérémie Méhault, kms15,
Kossori Aruku, Lakshya KD, maikia, Manuel López-Ibáñez, Marco Gorelli,
MarcoGorelli, mferrari3, Mickaël Schoentgen, Nicolas Hug, pavlos kallis, Pierre
Glaser, pierretallotte, Prabakaran Kumaresshan, Reshama Shaikh, Rohit Kapoor,
Roman Yurchak, SandroCasagrande, Tashay Green, Thomas Fan, Vishaal Kapoor,
Zhuyi Xue, Zijie (ZJ) Poh</p>
</div>
</div>
<div class="section" id="version-0-20-2">
<span id="changes-0-20-2"></span><h1>Version 0.20.2<a class="headerlink" href="#version-0-20-2" title="Permalink to this headline">¶</a></h1>
<p><strong>December 20, 2018</strong></p>
<p>This is a bug-fix release with some minor documentation improvements and
enhancements to features released in 0.20.0.</p>
<div class="section" id="changed-models">
<h2>Changed models<a class="headerlink" href="#changed-models" title="Permalink to this headline">¶</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><a class="reference internal" href="../modules/classes.html#module-sklearn.neighbors" title="sklearn.neighbors"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.neighbors</span></code></a> when <code class="docutils literal"><span class="pre">metric=='jaccard'</span></code> (bug fix)</li>
<li>use of <code class="docutils literal"><span class="pre">'seuclidean'</span></code> or <code class="docutils literal"><span class="pre">'mahalanobis'</span></code> metrics in some cases (bug fix)</li>
</ul>
</div>
<div class="section" id="id9">
<h2>Changelog<a class="headerlink" href="#id9" title="Permalink to this headline">¶</a></h2>
<div class="section" id="id10">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.compose" title="sklearn.compose"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#id10" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed an issue in <a class="reference internal" href="../modules/generated/sklearn.compose.make_column_transformer.html#sklearn.compose.make_column_transformer" title="sklearn.compose.make_column_transformer"><code class="xref py py-func docutils literal"><span class="pre">compose.make_column_transformer</span></code></a> which raises
unexpected error when columns is pandas Index or pandas Series.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12704">#12704</a> by <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.metrics</span></code></a><a class="headerlink" href="#sklearn-metrics" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">metrics.pairwise_distances</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.metrics.pairwise_distances_chunked.html#sklearn.metrics.pairwise_distances_chunked" title="sklearn.metrics.pairwise_distances_chunked"><code class="xref py py-func docutils literal"><span class="pre">metrics.pairwise_distances_chunked</span></code></a> where parameters <code class="docutils literal"><span class="pre">V</span></code> of
<code class="docutils literal"><span class="pre">"seuclidean"</span></code> and <code class="docutils literal"><span class="pre">VI</span></code> of <code class="docutils literal"><span class="pre">"mahalanobis"</span></code> metrics were computed after
the data was split into chunks instead of being pre-computed on whole data.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12701">#12701</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</li>
</ul>
</div>
<div class="section" id="id11">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.neighbors" title="sklearn.neighbors"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.neighbors</span></code></a><a class="headerlink" href="#id11" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed <a class="reference internal" href="../modules/generated/sklearn.neighbors.DistanceMetric.html#sklearn.neighbors.DistanceMetric" title="sklearn.neighbors.DistanceMetric"><code class="xref py py-class docutils literal"><span class="pre">sklearn.neighbors.DistanceMetric</span></code></a> jaccard distance
function to return 0 when two all-zero vectors are compared.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12685">#12685</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.utils</span></code></a><a class="headerlink" href="#sklearn-utils" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Calling <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"><span class="pre">utils.check_array</span></code></a> on <cite>pandas.Series</cite> with categorical
data, which raised an error in 0.20.0, now returns the expected output again.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12699">#12699</a> by <a class="reference external" href="https://github.com/jorisvandenbossche">Joris Van den Bossche</a>.</li>
</ul>
</div>
</div>
<div class="section" id="id12">
<h2>Code and Documentation Contributors<a class="headerlink" href="#id12" title="Permalink to this headline">¶</a></h2>
<p>With thanks to:</p>
<p>adanhawth, Adrin Jalali, Albert Thomas, Andreas Mueller, Dan Stine, Feda Curic,
Hanmin Qin, Jan S, jeremiedbb, Joel Nothman, Joris Van den Bossche,
josephsalmon, Katrin Leinweber, Loic Esteve, Muhammad Hassaan Rafique, Nicolas
Hug, Olivier Grisel, Paul Paczuski, Reshama Shaikh, Sam Waterbury, Shivam
Kotwalia, Thomas Fan</p>
</div>
</div>
<div class="section" id="version-0-20-1">
<span id="changes-0-20-1"></span><h1>Version 0.20.1<a class="headerlink" href="#version-0-20-1" title="Permalink to this headline">¶</a></h1>
<p><strong>November 21, 2018</strong></p>
<p>This is a bug-fix release with some minor documentation improvements and
enhancements to features released in 0.20.0. Note that we also include some
API changes in this release, so you might get some extra warnings after
updating from 0.20.0 to 0.20.1.</p>
<div class="section" id="id13">
<h2>Changed models<a class="headerlink" href="#id13" title="Permalink to this headline">¶</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><a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a> (bug fix)</li>
</ul>
</div>
<div class="section" id="id14">
<h2>Changelog<a class="headerlink" href="#id14" title="Permalink to this headline">¶</a></h2>
<div class="section" id="id15">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#id15" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-info">Efficiency</span></span> make <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"><span class="pre">cluster.MeanShift</span></code></a> no longer try to do nested
parallelism as the overhead would hurt performance significantly when
<code class="docutils literal"><span class="pre">n_jobs</span> <span class="pre">></span> <span class="pre">1</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12159">#12159</a> by <a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.DBSCAN.html#sklearn.cluster.DBSCAN" title="sklearn.cluster.DBSCAN"><code class="xref py py-class docutils literal"><span class="pre">cluster.DBSCAN</span></code></a> with precomputed sparse neighbors
graph, which would add explicitly zeros on the diagonal even when already
present. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12105">#12105</a> by <a class="reference external" href="https://github.com/TomDLT">Tom Dupre la Tour</a>.</li>
</ul>
</div>
<div class="section" id="id16">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.compose" title="sklearn.compose"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#id16" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed an issue in <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"><span class="pre">compose.ColumnTransformer</span></code></a> when stacking
columns with types not convertible to a numeric.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11912">#11912</a> by <a class="reference external" href="https://github.com/adrinjalali">Adrin Jalali</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</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"><span class="pre">compose.ColumnTransformer</span></code></a> now applies the <code class="docutils literal"><span class="pre">sparse_threshold</span></code>
even if all transformation results are sparse. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12304">#12304</a> by <a class="reference external" href="https://amueller.github.io/">Andreas
Müller</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> <a class="reference internal" href="../modules/generated/sklearn.compose.make_column_transformer.html#sklearn.compose.make_column_transformer" title="sklearn.compose.make_column_transformer"><code class="xref py py-func docutils literal"><span class="pre">compose.make_column_transformer</span></code></a> now expects
<code class="docutils literal"><span class="pre">(transformer,</span> <span class="pre">columns)</span></code> instead of <code class="docutils literal"><span class="pre">(columns,</span> <span class="pre">transformer)</span></code> to keep
consistent with <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"><span class="pre">compose.ColumnTransformer</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12339">#12339</a> by <a class="reference external" href="https://github.com/adrinjalali">Adrin Jalali</a>.</li>
</ul>
</div>
<div class="section" id="id17">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.datasets" title="sklearn.datasets"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.datasets</span></code></a><a class="headerlink" href="#id17" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</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"><span class="pre">datasets.fetch_openml</span></code></a> to correctly use the local cache.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12246">#12246</a> by <a class="reference external" href="https://github.com/janvanrijn">Jan N. van Rijn</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</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"><span class="pre">datasets.fetch_openml</span></code></a> to correctly handle ignore attributes and
row id attributes. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12330">#12330</a> by <a class="reference external" href="https://github.com/janvanrijn">Jan N. van Rijn</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed integer overflow in <a class="reference internal" href="../modules/generated/sklearn.datasets.make_classification.html#sklearn.datasets.make_classification" title="sklearn.datasets.make_classification"><code class="xref py py-func docutils literal"><span class="pre">datasets.make_classification</span></code></a>
for values of <code class="docutils literal"><span class="pre">n_informative</span></code> parameter larger than 64.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10811">#10811</a> by <a class="reference external" href="https://github.com/VarIr">Roman Feldbauer</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed olivetti faces dataset <code class="docutils literal"><span class="pre">DESCR</span></code> attribute to point to the right
location in <a class="reference internal" href="../modules/generated/sklearn.datasets.fetch_olivetti_faces.html#sklearn.datasets.fetch_olivetti_faces" title="sklearn.datasets.fetch_olivetti_faces"><code class="xref py py-func docutils literal"><span class="pre">datasets.fetch_olivetti_faces</span></code></a>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12441">#12441</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jérémie du Boisberranger</a></li>
<li><span class="raw-html"><span class="label label-danger">Fix</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"><span class="pre">datasets.fetch_openml</span></code></a> to retry downloading when reading
from local cache fails. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12517">#12517</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</li>
</ul>
</div>
<div class="section" id="id18">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.decomposition" title="sklearn.decomposition"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#id18" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a regression in <a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a> where
0.20.0 raised an error if the number of samples in the final batch for
fitting IncrementalPCA was smaller than n_components.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12234">#12234</a> by <a class="reference external" href="https://github.com/minggli">Ming Li</a>.</li>
</ul>
</div>
<div class="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"><span class="pre">sklearn.ensemble</span></code></a><a class="headerlink" href="#sklearn-ensemble" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug mostly affecting <a class="reference internal" href="../modules/generated/sklearn.ensemble.RandomForestClassifier.html#sklearn.ensemble.RandomForestClassifier" title="sklearn.ensemble.RandomForestClassifier"><code class="xref py py-class docutils literal"><span class="pre">ensemble.RandomForestClassifier</span></code></a>
where <code class="docutils literal"><span class="pre">class_weight='balanced_subsample'</span></code> failed with more than 32 classes.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12165">#12165</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug affecting <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"><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"><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"><span class="pre">ensemble.IsolationForest</span></code></a>,
where <code class="docutils literal"><span class="pre">max_features</span></code> was sometimes rounded down to zero.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12388">#12388</a> by <a class="reference external" href="https://github.com/Connossor">Connor Tann</a>.</li>
</ul>
</div>
<div class="section" id="id19">
<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"><span class="pre">sklearn.feature_extraction</span></code></a><a class="headerlink" href="#id19" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a regression in v0.20.0 where
<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-func docutils literal"><span class="pre">feature_extraction.text.CountVectorizer</span></code></a> and other text vectorizers
could error during stop words validation with custom preprocessors
or tokenizers. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12393">#12393</a> by <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>.</li>
</ul>
</div>
<div class="section" id="id20">
<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"><span class="pre">sklearn.linear_model</span></code></a><a class="headerlink" href="#id20" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> <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"><span class="pre">linear_model.SGDClassifier</span></code></a> and variants
with <code class="docutils literal"><span class="pre">early_stopping=True</span></code> would not use a consistent validation
split in the multiclass case and this would cause a crash when using
those estimators as part of parallel parameter search or cross-validation.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12122">#12122</a> by <a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug affecting <code class="xref py py-class docutils literal"><span class="pre">SGDClassifier</span></code> in the multiclass
case. Each one-versus-all step is run in a <a class="reference external" href="https://joblib.readthedocs.io/en/latest/generated/joblib.Parallel.html#joblib.Parallel" title="(in joblib v0.13.2)"><code class="xref py py-class docutils literal"><span class="pre">joblib.Parallel</span></code></a> call and
mutating a common parameter, causing a segmentation fault if called within a
backend using processes and not threads. We now use <code class="docutils literal"><span class="pre">require=sharedmem</span></code>
at the <a class="reference external" href="https://joblib.readthedocs.io/en/latest/generated/joblib.Parallel.html#joblib.Parallel" title="(in joblib v0.13.2)"><code class="xref py py-class docutils literal"><span class="pre">joblib.Parallel</span></code></a> instance creation. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12518">#12518</a> by
<a class="reference external" href="https://github.com/pierreglaser">Pierre Glaser</a> and <a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
</ul>
</div>
<div class="section" id="id21">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.metrics" title="sklearn.metrics"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.metrics</span></code></a><a class="headerlink" href="#id21" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <code class="xref py py-func docutils literal"><span class="pre">metrics.pairwise.pairwise_distances_argmin_min</span></code>
which returned the square root of the distance when the metric parameter was
set to “euclidean”. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12481">#12481</a> by
<a class="reference external" href="https://github.com/jeremiedbb">Jérémie du Boisberranger</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <code class="xref py py-func docutils literal"><span class="pre">metrics.pairwise.pairwise_distances_chunked</span></code>
which didn’t ensure the diagonal is zero for euclidean distances.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12612">#12612</a> by <a class="reference external" href="https://github.com/amueller">Andreas Müller</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> The <a class="reference internal" href="../modules/generated/sklearn.metrics.calinski_harabaz_score.html#sklearn.metrics.calinski_harabaz_score" title="sklearn.metrics.calinski_harabaz_score"><code class="xref py py-func docutils literal"><span class="pre">metrics.calinski_harabaz_score</span></code></a> has been renamed to
<code class="xref py py-func docutils literal"><span class="pre">metrics.calinski_harabasz_score</span></code> and will be removed in version 0.23.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12211">#12211</a> by <a class="reference external" href="https://github.com/LisaThomas9">Lisa Thomas</a>,
<a class="reference external" href="https://github.com/markhannel">Mark Hannel</a> and <a class="reference external" href="https://github.com/mferrari3">Melissa Ferrari</a>.</li>
</ul>
</div>
<div class="section" id="sklearn-mixture">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.mixture" title="sklearn.mixture"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.mixture</span></code></a><a class="headerlink" href="#sklearn-mixture" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Ensure that the <code class="docutils literal"><span class="pre">fit_predict</span></code> method of
<a class="reference internal" href="../modules/generated/sklearn.mixture.GaussianMixture.html#sklearn.mixture.GaussianMixture" title="sklearn.mixture.GaussianMixture"><code class="xref py py-class docutils literal"><span class="pre">mixture.GaussianMixture</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.mixture.BayesianGaussianMixture.html#sklearn.mixture.BayesianGaussianMixture" title="sklearn.mixture.BayesianGaussianMixture"><code class="xref py py-class docutils literal"><span class="pre">mixture.BayesianGaussianMixture</span></code></a>
always yield assignments consistent with <code class="docutils literal"><span class="pre">fit</span></code> followed by <code class="docutils literal"><span class="pre">predict</span></code> even
if the convergence criterion is too loose or not met. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12451">#12451</a>
by <a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
</ul>
</div>
<div class="section" id="id22">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.neighbors" title="sklearn.neighbors"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.neighbors</span></code></a><a class="headerlink" href="#id22" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> force the parallelism backend to <code class="code docutils literal"><span class="pre">threading</span></code> for
<a class="reference internal" href="../modules/generated/sklearn.neighbors.KDTree.html#sklearn.neighbors.KDTree" title="sklearn.neighbors.KDTree"><code class="xref py py-class docutils literal"><span class="pre">neighbors.KDTree</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.neighbors.BallTree.html#sklearn.neighbors.BallTree" title="sklearn.neighbors.BallTree"><code class="xref py py-class docutils literal"><span class="pre">neighbors.BallTree</span></code></a> in Python 2.7 to
avoid pickling errors caused by the serialization of their methods.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12171">#12171</a> by <a class="reference external" href="https://github.com/tomMoral">Thomas Moreau</a>.</li>
</ul>
</div>
<div class="section" id="id23">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.preprocessing" title="sklearn.preprocessing"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.preprocessing</span></code></a><a class="headerlink" href="#id23" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed bug in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.OrdinalEncoder.html#sklearn.preprocessing.OrdinalEncoder" title="sklearn.preprocessing.OrdinalEncoder"><code class="xref py py-class docutils literal"><span class="pre">preprocessing.OrdinalEncoder</span></code></a> when passing
manually specified categories. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12365">#12365</a> by <a class="reference external" href="https://github.com/jorisvandenbossche">Joris Van den Bossche</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed bug in <a class="reference internal" href="../modules/generated/sklearn.preprocessing.KBinsDiscretizer.html#sklearn.preprocessing.KBinsDiscretizer" title="sklearn.preprocessing.KBinsDiscretizer"><code class="xref py py-class docutils literal"><span class="pre">preprocessing.KBinsDiscretizer</span></code></a> where the
<code class="docutils literal"><span class="pre">transform</span></code> method mutates the <code class="docutils literal"><span class="pre">_encoder</span></code> attribute. The <code class="docutils literal"><span class="pre">transform</span></code>
method is now thread safe. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12514">#12514</a> by
<a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">preprocessing.PowerTransformer</span></code></a> where the
Yeo-Johnson transform was incorrect for lambda parameters outside of <cite>[0, 2]</cite>
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12522">#12522</a> by <a class="reference external" href="https://github.com/NicolasHug">Nicolas Hug</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">preprocessing.OneHotEncoder</span></code></a> where transform
failed when set to ignore unknown numpy strings of different lengths
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12471">#12471</a> by <a class="reference external" href="https://github.com/GMarzinotto">Gabriel Marzinotto</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> The default value of the <code class="code docutils literal"><span class="pre">method</span></code> argument in
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.power_transform.html#sklearn.preprocessing.power_transform" title="sklearn.preprocessing.power_transform"><code class="xref py py-func docutils literal"><span class="pre">preprocessing.power_transform</span></code></a> will be changed from <code class="code docutils literal"><span class="pre">box-cox</span></code>
to <code class="code docutils literal"><span class="pre">yeo-johnson</span></code> to match <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"><span class="pre">preprocessing.PowerTransformer</span></code></a>
in version 0.23. A FutureWarning is raised when the default value is used.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12317">#12317</a> by <a class="reference external" href="https://github.com/chang">Eric Chang</a>.</li>
</ul>
</div>
<div class="section" id="id24">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.utils" title="sklearn.utils"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.utils</span></code></a><a class="headerlink" href="#id24" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Use float64 for mean accumulator to avoid floating point
precision issues 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"><span class="pre">preprocessing.StandardScaler</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a> when using float32 datasets.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12338">#12338</a> by <a class="reference external" href="https://github.com/bauks">bauks</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Calling <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"><span class="pre">utils.check_array</span></code></a> on <cite>pandas.Series</cite>, which
raised an error in 0.20.0, now returns the expected output again.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12625">#12625</a> by <a class="reference external" href="https://amueller.github.io/">Andreas Müller</a></li>
</ul>
</div>
<div class="section" id="miscellaneous">
<h3>Miscellaneous<a class="headerlink" href="#miscellaneous" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> When using site joblib by setting the environment variable
<cite>SKLEARN_SITE_JOBLIB</cite>, added compatibility with joblib 0.11 in addition
to 0.12+. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12350">#12350</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a> and <a class="reference external" href="https://github.com/rth">Roman Yurchak</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Make sure to avoid raising <code class="docutils literal"><span class="pre">FutureWarning</span></code> when calling
<code class="docutils literal"><span class="pre">np.vstack</span></code> with numpy 1.16 and later (use list comprehensions
instead of generator expressions in many locations of the scikit-learn
code base). <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12467">#12467</a> by <a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> Removed all mentions of <code class="docutils literal"><span class="pre">sklearn.externals.joblib</span></code>, and deprecated
joblib methods exposed in <code class="docutils literal"><span class="pre">sklearn.utils</span></code>, except for
<a class="reference internal" href="../modules/generated/sklearn.utils.parallel_backend.html#sklearn.utils.parallel_backend" title="sklearn.utils.parallel_backend"><code class="xref py py-func docutils literal"><span class="pre">utils.parallel_backend</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.utils.register_parallel_backend.html#sklearn.utils.register_parallel_backend" title="sklearn.utils.register_parallel_backend"><code class="xref py py-func docutils literal"><span class="pre">utils.register_parallel_backend</span></code></a>,
which allow users to configure parallel computation in scikit-learn.
Other functionalities are part of <a class="reference external" href="https://joblib.readthedocs.io/">joblib</a>.
package and should be used directly, by installing it.
The goal of this change is to prepare for
unvendoring joblib in future version of scikit-learn.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/12345">#12345</a> by <a class="reference external" href="https://github.com/tomMoral">Thomas Moreau</a></li>
</ul>
</div>
</div>
<div class="section" id="id25">
<h2>Code and Documentation Contributors<a class="headerlink" href="#id25" title="Permalink to this headline">¶</a></h2>
<p>With thanks to:</p>
<p>^__^, Adrin Jalali, Andrea Navarrete, Andreas Mueller,
bauks, BenjaStudio, Cheuk Ting Ho, Connossor,
Corey Levinson, Dan Stine, daten-kieker, Denis Kataev,
Dillon Gardner, Dmitry Vukolov, Dougal J. Sutherland, Edward J Brown,
Eric Chang, Federico Caselli, Gabriel Marzinotto, Gael Varoquaux,
GauravAhlawat, Gustavo De Mari Pereira, Hanmin Qin, haroldfox,
JackLangerman, Jacopo Notarstefano, janvanrijn, jdethurens,
jeremiedbb, Joel Nothman, Joris Van den Bossche, Koen,
Kushal Chauhan, Lee Yi Jie Joel, Lily Xiong, mail-liam,
Mark Hannel, melsyt, Ming Li, Nicholas Smith,
Nicolas Hug, Nikolay Shebanov, Oleksandr Pavlyk, Olivier Grisel,
Peter Hausamann, Pierre Glaser, Pulkit Maloo, Quentin Batista,
Radostin Stoyanov, Ramil Nugmanov, Rebekah Kim, Reshama Shaikh,
Rohan Singh, Roman Feldbauer, Roman Yurchak, Roopam Sharma,
Sam Waterbury, Scott Lowe, Sebastian Raschka, Stephen Tierney,
SylvainLan, TakingItCasual, Thomas Fan, Thomas Moreau,
Tom Dupré la Tour, Tulio Casagrande, Utkarsh Upadhyay, Xing Han Lu,
Yaroslav Halchenko, Zach Miller</p>
</div>
</div>
<div class="section" id="version-0-20-0">
<span id="changes-0-20"></span><h1>Version 0.20.0<a class="headerlink" href="#version-0-20-0" title="Permalink to this headline">¶</a></h1>
<p><strong>September 25, 2018</strong></p>
<p>This release packs in a mountain of bug fixes, features and enhancements for
the Scikit-learn library, and improvements to the documentation and examples.
Thanks to our contributors!</p>
<p>This release is dedicated to the memory of Raghav Rajagopalan.</p>
<div class="admonition warning">
<p class="first admonition-title">Warning</p>
<p class="last">Version 0.20 is the last version of scikit-learn to support Python 2.7 and Python 3.4.
Scikit-learn 0.21 will require Python 3.5 or higher.</p>
</div>
<div class="section" id="highlights">
<h2>Highlights<a class="headerlink" href="#highlights" title="Permalink to this headline">¶</a></h2>
<p>We have tried to improve our support for common data-science use-cases
including missing values, categorical variables, heterogeneous data, and
features/targets with unusual distributions.
Missing values in features, represented by NaNs, are now accepted in
column-wise preprocessing such as scalers. Each feature is fitted disregarding
NaNs, and data containing NaNs can be transformed. The new <code class="xref py py-mod docutils literal"><span class="pre">impute</span></code>
module provides estimators for learning despite missing data.</p>
<p><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"><span class="pre">ColumnTransformer</span></code></a> handles the case where different features
or columns of a pandas.DataFrame need different preprocessing.
String or pandas Categorical columns can now be encoded with
<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"><span class="pre">OneHotEncoder</span></code></a> or
<a class="reference internal" href="../modules/generated/sklearn.preprocessing.OrdinalEncoder.html#sklearn.preprocessing.OrdinalEncoder" title="sklearn.preprocessing.OrdinalEncoder"><code class="xref py py-class docutils literal"><span class="pre">OrdinalEncoder</span></code></a>.</p>
<p><a class="reference internal" href="../modules/generated/sklearn.compose.TransformedTargetRegressor.html#sklearn.compose.TransformedTargetRegressor" title="sklearn.compose.TransformedTargetRegressor"><code class="xref py py-class docutils literal"><span class="pre">TransformedTargetRegressor</span></code></a> helps when the regression target
needs to be transformed to be modeled. <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"><span class="pre">PowerTransformer</span></code></a>
and <a class="reference internal" href="../modules/generated/sklearn.preprocessing.KBinsDiscretizer.html#sklearn.preprocessing.KBinsDiscretizer" title="sklearn.preprocessing.KBinsDiscretizer"><code class="xref py py-class docutils literal"><span class="pre">KBinsDiscretizer</span></code></a> join
<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"><span class="pre">QuantileTransformer</span></code></a> as non-linear transformations.</p>
<p>Beyond this, we have added <a class="reference internal" href="../glossary.html#term-sample-weight"><span class="xref std std-term">sample_weight</span></a> support to several estimators
(including <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"><span class="pre">KMeans</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.linear_model.BayesianRidge.html#sklearn.linear_model.BayesianRidge" title="sklearn.linear_model.BayesianRidge"><code class="xref py py-class docutils literal"><span class="pre">BayesianRidge</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.neighbors.KernelDensity.html#sklearn.neighbors.KernelDensity" title="sklearn.neighbors.KernelDensity"><code class="xref py py-class docutils literal"><span class="pre">KernelDensity</span></code></a>) and improved stopping criteria in others
(including <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"><span class="pre">MLPRegressor</span></code></a>,
<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"><span class="pre">GradientBoostingRegressor</span></code></a> and
<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"><span class="pre">SGDRegressor</span></code></a>).</p>
<p>This release is also the first to be accompanied by a <a class="reference internal" href="../glossary.html#glossary"><span class="std std-ref">Glossary of Common Terms and API Elements</span></a> developed
by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>. The glossary is a reference resource to help users and
contributors become familiar with the terminology and conventions used in
Scikit-learn.</p>
<p>Sorry if your contribution didn’t make it into the highlights. There’s a lot
here…</p>
</div>
<div class="section" id="id26">
<h2>Changed models<a class="headerlink" href="#id26" title="Permalink to this headline">¶</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><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"><span class="pre">cluster.MeanShift</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a> in Python 2 (bug fix)</li>
<li><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"><span class="pre">decomposition.SparsePCA</span></code></a> (bug fix)</li>
<li><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"><span class="pre">ensemble.GradientBoostingClassifier</span></code></a> (bug fix affecting feature importances)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.isotonic.IsotonicRegression.html#sklearn.isotonic.IsotonicRegression" title="sklearn.isotonic.IsotonicRegression"><code class="xref py py-class docutils literal"><span class="pre">isotonic.IsotonicRegression</span></code></a> (bug fix)</li>
<li><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"><span class="pre">linear_model.ARDRegression</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.linear_model.LogisticRegressionCV.html#sklearn.linear_model.LogisticRegressionCV" title="sklearn.linear_model.LogisticRegressionCV"><code class="xref py py-class docutils literal"><span class="pre">linear_model.LogisticRegressionCV</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.linear_model.OrthogonalMatchingPursuit.html#sklearn.linear_model.OrthogonalMatchingPursuit" title="sklearn.linear_model.OrthogonalMatchingPursuit"><code class="xref py py-class docutils literal"><span class="pre">linear_model.OrthogonalMatchingPursuit</span></code></a> (bug fix)</li>
<li><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"><span class="pre">linear_model.PassiveAggressiveClassifier</span></code></a> (bug fix)</li>
<li><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"><span class="pre">linear_model.PassiveAggressiveRegressor</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.linear_model.Perceptron.html#sklearn.linear_model.Perceptron" title="sklearn.linear_model.Perceptron"><code class="xref py py-class docutils literal"><span class="pre">linear_model.Perceptron</span></code></a> (bug fix)</li>
<li><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"><span class="pre">linear_model.SGDClassifier</span></code></a> (bug fix)</li>
<li><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"><span class="pre">linear_model.SGDRegressor</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.metrics.roc_auc_score.html#sklearn.metrics.roc_auc_score" title="sklearn.metrics.roc_auc_score"><code class="xref py py-class docutils literal"><span class="pre">metrics.roc_auc_score</span></code></a> (bug fix)</li>
<li><a class="reference internal" href="../modules/generated/sklearn.metrics.roc_curve.html#sklearn.metrics.roc_curve" title="sklearn.metrics.roc_curve"><code class="xref py py-class docutils literal"><span class="pre">metrics.roc_curve</span></code></a> (bug fix)</li>
<li><code class="xref py py-class docutils literal"><span class="pre">neural_network.BaseMultilayerPerceptron</span></code> (bug fix)</li>
<li><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"><span class="pre">neural_network.MLPClassifier</span></code></a> (bug fix)</li>
<li><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"><span class="pre">neural_network.MLPRegressor</span></code></a> (bug fix)</li>
<li>The v0.19.0 release notes failed to mention a backwards incompatibility with
<a class="reference internal" href="../modules/generated/sklearn.model_selection.StratifiedKFold.html#sklearn.model_selection.StratifiedKFold" title="sklearn.model_selection.StratifiedKFold"><code class="xref py py-class docutils literal"><span class="pre">model_selection.StratifiedKFold</span></code></a> when <code class="docutils literal"><span class="pre">shuffle=True</span></code> due to
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/7823">#7823</a>.</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>
</div>
<div class="section" id="known-major-bugs">
<h2>Known Major Bugs<a class="headerlink" href="#known-major-bugs" title="Permalink to this headline">¶</a></h2>
<ul class="simple">
<li><a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11924">#11924</a>: <a class="reference internal" href="../modules/generated/sklearn.linear_model.LogisticRegressionCV.html#sklearn.linear_model.LogisticRegressionCV" title="sklearn.linear_model.LogisticRegressionCV"><code class="xref py py-class docutils literal"><span class="pre">linear_model.LogisticRegressionCV</span></code></a> with
<cite>solver=’lbfgs’</cite> and <cite>multi_class=’multinomial’</cite> may be non-deterministic or
otherwise broken on macOS. This appears to be the case on Travis CI servers,
but has not been confirmed on personal MacBooks! This issue has been present
in previous releases.</li>
<li><a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9354">#9354</a>: <a class="reference internal" href="../modules/generated/sklearn.metrics.pairwise.euclidean_distances.html#sklearn.metrics.pairwise.euclidean_distances" title="sklearn.metrics.pairwise.euclidean_distances"><code class="xref py py-func docutils literal"><span class="pre">metrics.pairwise.euclidean_distances</span></code></a> (which is used
several times throughout the library) gives results with poor precision,
which particularly affects its use with 32-bit float inputs. This became
more problematic in versions 0.18 and 0.19 when some algorithms were changed
to avoid casting 32-bit data into 64-bit.</li>
</ul>
</div>
<div class="section" id="id27">
<h2>Changelog<a class="headerlink" href="#id27" title="Permalink to this headline">¶</a></h2>
<p>Support for Python 3.3 has been officially dropped.</p>
<div class="section" id="id28">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.cluster" title="sklearn.cluster"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.cluster</span></code></a><a class="headerlink" href="#id28" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-success">Major Feature</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"><span class="pre">cluster.AgglomerativeClustering</span></code></a> now supports Single
Linkage clustering via <code class="docutils literal"><span class="pre">linkage='single'</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9372">#9372</a> by <a class="reference external" href="https://github.com/lmcinnes">Leland
McInnes</a> and <a class="reference external" href="https://github.com/sastels">Steve Astels</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</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"><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"><span class="pre">cluster.MiniBatchKMeans</span></code></a> now support
sample weights via new parameter <code class="docutils literal"><span class="pre">sample_weight</span></code> in <code class="docutils literal"><span class="pre">fit</span></code> function.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10933">#10933</a> by <a class="reference external" href="https://github.com/jnhansen">Johannes Hansen</a>.</li>
<li><span class="raw-html"><span class="label label-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"><span class="pre">cluster.KMeans</span></code></a>, <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"><span class="pre">cluster.MiniBatchKMeans</span></code></a> and
<a class="reference internal" href="../modules/generated/sklearn.cluster.k_means.html#sklearn.cluster.k_means" title="sklearn.cluster.k_means"><code class="xref py py-func docutils literal"><span class="pre">cluster.k_means</span></code></a> passed with <code class="docutils literal"><span class="pre">algorithm='full'</span></code> now enforces
row-major ordering, improving runtime.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10471">#10471</a> by <a class="reference external" href="https://github.com/gxyd">Gaurav Dhingra</a>.</li>
<li><span class="raw-html"><span class="label label-info">Efficiency</span></span> <a class="reference internal" href="../modules/generated/sklearn.cluster.DBSCAN.html#sklearn.cluster.DBSCAN" title="sklearn.cluster.DBSCAN"><code class="xref py py-class docutils literal"><span class="pre">cluster.DBSCAN</span></code></a> now is parallelized according to <code class="docutils literal"><span class="pre">n_jobs</span></code>
regardless of <code class="docutils literal"><span class="pre">algorithm</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/8003">#8003</a> by <a class="reference external" href="https://github.com/recamshak">Joël Billaud</a>.</li>
<li><span class="raw-html"><span class="label label-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"><span class="pre">cluster.KMeans</span></code></a> now gives a warning if the number of
distinct clusters found is smaller than <code class="docutils literal"><span class="pre">n_clusters</span></code>. This may occur when
the number of distinct points in the data set is actually smaller than the
number of cluster one is looking for.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10059">#10059</a> by <a class="reference external" href="https://github.com/christianbraune79">Christian Braune</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug where the <code class="docutils literal"><span class="pre">fit</span></code> method of
<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"><span class="pre">cluster.AffinityPropagation</span></code></a> stored cluster
centers as 3d array instead of 2d array in case of non-convergence. For the
same class, fixed undefined and arbitrary behavior in case of training data
where all samples had equal similarity.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9612">#9612</a>. By <a class="reference external" href="https://github.com/jsamoocha">Jonatan Samoocha</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.spectral_clustering.html#sklearn.cluster.spectral_clustering" title="sklearn.cluster.spectral_clustering"><code class="xref py py-func docutils literal"><span class="pre">cluster.spectral_clustering</span></code></a> where the normalization of
the spectrum was using a division instead of a multiplication. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/8129">#8129</a>
by <a class="reference external" href="https://github.com/jmargeta">Jan Margeta</a>, <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>,
and <a class="reference external" href="https://github.com/devanshdalal">Devansh D.</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <code class="xref py py-func docutils literal"><span class="pre">cluster.k_means_elkan</span></code> where the returned
<code class="docutils literal"><span class="pre">iteration</span></code> was 1 less than the correct value. Also added the missing
<code class="docutils literal"><span class="pre">n_iter_</span></code> attribute in the docstring 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"><span class="pre">cluster.KMeans</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11353">#11353</a> by <a class="reference external" href="https://github.com/jeremiedbb">Jeremie du Boisberranger</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.cluster.mean_shift.html#sklearn.cluster.mean_shift" title="sklearn.cluster.mean_shift"><code class="xref py py-func docutils literal"><span class="pre">cluster.mean_shift</span></code></a> where the assigned labels
were not deterministic if there were multiple clusters with the same
intensities.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11901">#11901</a> by <a class="reference external" href="https://github.com/adrinjalali">Adrin Jalali</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> Deprecate <code class="docutils literal"><span class="pre">pooling_func</span></code> unused parameter in
<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"><span class="pre">cluster.AgglomerativeClustering</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9875">#9875</a> by <a class="reference external" href="https://github.com/thechargedneutron">Kumar Ashutosh</a>.</li>
</ul>
</div>
<div class="section" id="id29">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.compose" title="sklearn.compose"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.compose</span></code></a><a class="headerlink" href="#id29" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>New module.</li>
<li><span class="raw-html"><span class="label label-success">Major Feature</span></span> Added <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"><span class="pre">compose.ColumnTransformer</span></code></a>, which allows to
apply different transformers to different columns of arrays or pandas
DataFrames. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9012">#9012</a> by <a class="reference external" href="https://amueller.github.io/">Andreas Müller</a> and <a class="reference external" href="https://github.com/jorisvandenbossche">Joris Van den Bossche</a>,
and <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11315">#11315</a> by <a class="reference external" href="https://github.com/thomasjpfan">Thomas Fan</a>.</li>
<li><span class="raw-html"><span class="label label-success">Major Feature</span></span> Added the <a class="reference internal" href="../modules/generated/sklearn.compose.TransformedTargetRegressor.html#sklearn.compose.TransformedTargetRegressor" title="sklearn.compose.TransformedTargetRegressor"><code class="xref py py-class docutils literal"><span class="pre">compose.TransformedTargetRegressor</span></code></a> which
transforms the target y before fitting a regression model. The predictions
are mapped back to the original space via an inverse transform. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9041">#9041</a>
by <a class="reference external" href="https://amueller.github.io/">Andreas Müller</a> and <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</li>
</ul>
</div>
<div class="section" id="id30">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.covariance" title="sklearn.covariance"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.covariance</span></code></a><a class="headerlink" href="#id30" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-info">Efficiency</span></span> Runtime improvements to <a class="reference internal" href="../modules/generated/sklearn.covariance.GraphicalLasso.html#sklearn.covariance.GraphicalLasso" title="sklearn.covariance.GraphicalLasso"><code class="xref py py-class docutils literal"><span class="pre">covariance.GraphicalLasso</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9858">#9858</a> by <a class="reference external" href="https://github.com/stevendbrown">Steven Brown</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> The <a class="reference internal" href="../modules/generated/sklearn.covariance.graph_lasso.html#sklearn.covariance.graph_lasso" title="sklearn.covariance.graph_lasso"><code class="xref py py-func docutils literal"><span class="pre">covariance.graph_lasso</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.covariance.GraphLasso.html#sklearn.covariance.GraphLasso" title="sklearn.covariance.GraphLasso"><code class="xref py py-class docutils literal"><span class="pre">covariance.GraphLasso</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.covariance.GraphLassoCV.html#sklearn.covariance.GraphLassoCV" title="sklearn.covariance.GraphLassoCV"><code class="xref py py-class docutils literal"><span class="pre">covariance.GraphLassoCV</span></code></a> have been
renamed to <a class="reference internal" href="../modules/generated/sklearn.covariance.graphical_lasso.html#sklearn.covariance.graphical_lasso" title="sklearn.covariance.graphical_lasso"><code class="xref py py-func docutils literal"><span class="pre">covariance.graphical_lasso</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.covariance.GraphicalLasso.html#sklearn.covariance.GraphicalLasso" title="sklearn.covariance.GraphicalLasso"><code class="xref py py-class docutils literal"><span class="pre">covariance.GraphicalLasso</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.covariance.GraphicalLassoCV.html#sklearn.covariance.GraphicalLassoCV" title="sklearn.covariance.GraphicalLassoCV"><code class="xref py py-class docutils literal"><span class="pre">covariance.GraphicalLassoCV</span></code></a>
respectively and will be removed in version 0.22.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9993">#9993</a> by <a class="reference external" href="https://github.com/artiemq">Artiem Krinitsyn</a></li>
</ul>
</div>
<div class="section" id="id31">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.datasets" title="sklearn.datasets"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.datasets</span></code></a><a class="headerlink" href="#id31" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-success">Major Feature</span></span> Added <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"><span class="pre">datasets.fetch_openml</span></code></a> to fetch datasets from
<a class="reference external" href="https://openml.org">OpenML</a>. OpenML is a free, open data sharing platform
and will be used instead of mldata as it provides better service availability.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9908">#9908</a> by <a class="reference external" href="https://amueller.github.io/">Andreas Müller</a> and <a class="reference external" href="https://github.com/janvanrijn">Jan N. van Rijn</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> 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"><span class="pre">datasets.make_blobs</span></code></a>, one can now pass a list to the
<code class="docutils literal"><span class="pre">n_samples</span></code> parameter to indicate the number of samples to generate per
cluster. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/8617">#8617</a> by <a class="reference external" href="https://github.com/maskani-moh">Maskani Filali Mohamed</a> and
<a class="reference external" href="https://github.com/kkatrio">Konstantinos Katrioplas</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> Add <code class="docutils literal"><span class="pre">filename</span></code> attribute to <code class="xref py py-mod docutils literal"><span class="pre">datasets</span></code> that have a CSV file.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9101">#9101</a> by <a class="reference external" href="https://github.com/alex-33">alex-33</a>
and <a class="reference external" href="https://github.com/maskani-moh">Maskani Filali Mohamed</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <code class="docutils literal"><span class="pre">return_X_y</span></code> parameter has been added to several dataset loaders.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10774">#10774</a> by <a class="reference external" href="https://github.com/ccatalfo">Chris Catalfo</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.datasets.load_boston.html#sklearn.datasets.load_boston" title="sklearn.datasets.load_boston"><code class="xref py py-func docutils literal"><span class="pre">datasets.load_boston</span></code></a> which had a wrong data
point. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10795">#10795</a> by <a class="reference external" href="https://github.com/tarcusx">Takeshi Yoshizawa</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.datasets.load_iris.html#sklearn.datasets.load_iris" title="sklearn.datasets.load_iris"><code class="xref py py-func docutils literal"><span class="pre">datasets.load_iris</span></code></a> which had two wrong data points.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11082">#11082</a> by <a class="reference external" href="https://github.com/rotuna">Sadhana Srinivasan</a>
and <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.datasets.fetch_kddcup99.html#sklearn.datasets.fetch_kddcup99" title="sklearn.datasets.fetch_kddcup99"><code class="xref py py-func docutils literal"><span class="pre">datasets.fetch_kddcup99</span></code></a>, where data were not
properly shuffled. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9731">#9731</a> by <a class="reference external" href="https://ngoix.github.io/">Nicolas Goix</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">datasets.make_circles</span></code></a>, where no odd number of
data points could be generated. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10045">#10045</a> by <a class="reference external" href="https://github.com/christianbraune79">Christian Braune</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> Deprecated <a class="reference internal" href="../modules/generated/sklearn.datasets.fetch_mldata.html#sklearn.datasets.fetch_mldata" title="sklearn.datasets.fetch_mldata"><code class="xref py py-func docutils literal"><span class="pre">sklearn.datasets.fetch_mldata</span></code></a> to be removed in
version 0.22. mldata.org is no longer operational. Until removal it will
remain possible to load cached datasets. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11466">#11466</a> by <a class="reference external" href="https://joelnothman.com/">Joel Nothman</a>.</li>
</ul>
</div>
<div class="section" id="id32">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.decomposition" title="sklearn.decomposition"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.decomposition</span></code></a><a class="headerlink" href="#id32" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.decomposition.dict_learning.html#sklearn.decomposition.dict_learning" title="sklearn.decomposition.dict_learning"><code class="xref py py-func docutils literal"><span class="pre">decomposition.dict_learning</span></code></a> functions and models now
support positivity constraints. This applies to the dictionary and sparse
code. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/6374">#6374</a> by <a class="reference external" href="https://github.com/jakirkham">John Kirkham</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <span class="raw-html"><span class="label label-danger">Fix</span></span> <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"><span class="pre">decomposition.SparsePCA</span></code></a> now exposes
<code class="docutils literal"><span class="pre">normalize_components</span></code>. When set to True, the train and test data are
centered with the train mean repsectively during the fit phase and the
transform phase. This fixes the behavior of SparsePCA. When set to False,
which is the default, the previous abnormal behaviour still holds. The False
value is for backward compatibility and should not be used. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11585">#11585</a>
by <a class="reference external" href="https://github.com/FollowKenny">Ivan Panico</a>.</li>
<li><span class="raw-html"><span class="label label-info">Efficiency</span></span> Efficiency improvements in <a class="reference internal" href="../modules/generated/sklearn.decomposition.dict_learning.html#sklearn.decomposition.dict_learning" title="sklearn.decomposition.dict_learning"><code class="xref py py-func docutils literal"><span class="pre">decomposition.dict_learning</span></code></a>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11420">#11420</a> and others by <a class="reference external" href="https://github.com/jakirkham">John Kirkham</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fix for uninformative error in <a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a>:
now an error is raised if the number of components is larger than the
chosen batch size. The <code class="docutils literal"><span class="pre">n_components=None</span></code> case was adapted accordingly.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/6452">#6452</a>. By <a class="reference external" href="https://github.com/wallygauze">Wally Gauze</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug where the <code class="docutils literal"><span class="pre">partial_fit</span></code> method of
<a class="reference internal" href="../modules/generated/sklearn.decomposition.IncrementalPCA.html#sklearn.decomposition.IncrementalPCA" title="sklearn.decomposition.IncrementalPCA"><code class="xref py py-class docutils literal"><span class="pre">decomposition.IncrementalPCA</span></code></a> used integer division instead of float
division on Python 2.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9492">#9492</a> by <a class="reference external" href="https://github.com/jrbourbeau">James Bourbeau</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> In <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"><span class="pre">decomposition.PCA</span></code></a> selecting a n_components parameter greater
than the number of samples now raises an error. Similarly, the
<code class="docutils literal"><span class="pre">n_components=None</span></code> case now selects the minimum of <code class="docutils literal"><span class="pre">n_samples</span></code> and
<code class="docutils literal"><span class="pre">n_features</span></code>.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/8484">#8484</a> by <a class="reference external" href="https://github.com/wallygauze">Wally Gauze</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">decomposition.PCA</span></code></a> where users will get
unexpected error with large datasets when <code class="docutils literal"><span class="pre">n_components='mle'</span></code> on Python 3
versions.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9886">#9886</a> by <a class="reference external" href="https://github.com/qinhanmin2014">Hanmin Qin</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed an underflow in calculating KL-divergence for
<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"><span class="pre">decomposition.NMF</span></code></a> <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10142">#10142</a> by <a class="reference external" href="https://github.com/TomDLT">Tom Dupre la Tour</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <a class="reference internal" href="../modules/generated/sklearn.decomposition.SparseCoder.html#sklearn.decomposition.SparseCoder" title="sklearn.decomposition.SparseCoder"><code class="xref py py-class docutils literal"><span class="pre">decomposition.SparseCoder</span></code></a> when running OMP
sparse coding in parallel using read-only memory mapped datastructures.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/5956">#5956</a> by <a class="reference external" href="https://github.com/vighneshbirodkar">Vighnesh Birodkar</a> and
<a class="reference external" href="https://github.com/ogrisel">Olivier Grisel</a>.</li>
</ul>
</div>
<div class="section" id="sklearn-discriminant-analysis">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.discriminant_analysis" title="sklearn.discriminant_analysis"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.discriminant_analysis</span></code></a><a class="headerlink" href="#sklearn-discriminant-analysis" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-info">Efficiency</span></span> Memory usage improvement for <code class="xref py py-func docutils literal"><span class="pre">_class_means</span></code> and
<code class="xref py py-func docutils literal"><span class="pre">_class_cov</span></code> in <code class="xref py py-mod docutils literal"><span class="pre">discriminant_analysis</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/10898">#10898</a> by
<a class="reference external" href="https://github.com/bobchennan">Nanxin Chen</a>.</li>
</ul>
</div>
<div class="section" id="sklearn-dummy">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.dummy" title="sklearn.dummy"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.dummy</span></code></a><a class="headerlink" href="#sklearn-dummy" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.dummy.DummyRegressor.html#sklearn.dummy.DummyRegressor" title="sklearn.dummy.DummyRegressor"><code class="xref py py-class docutils literal"><span class="pre">dummy.DummyRegressor</span></code></a> now has a <code class="docutils literal"><span class="pre">return_std</span></code> option in its
<code class="docutils literal"><span class="pre">predict</span></code> method. The returned standard deviations will be zeros.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.dummy.DummyClassifier.html#sklearn.dummy.DummyClassifier" title="sklearn.dummy.DummyClassifier"><code class="xref py py-class docutils literal"><span class="pre">dummy.DummyClassifier</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.dummy.DummyRegressor.html#sklearn.dummy.DummyRegressor" title="sklearn.dummy.DummyRegressor"><code class="xref py py-class docutils literal"><span class="pre">dummy.DummyRegressor</span></code></a> now
only require X to be an object with finite length or shape. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9832">#9832</a> by
<a class="reference external" href="https://github.com/vrishank97">Vrishank Bhardwaj</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <a class="reference internal" href="../modules/generated/sklearn.dummy.DummyClassifier.html#sklearn.dummy.DummyClassifier" title="sklearn.dummy.DummyClassifier"><code class="xref py py-class docutils literal"><span class="pre">dummy.DummyClassifier</span></code></a> and <a class="reference internal" href="../modules/generated/sklearn.dummy.DummyRegressor.html#sklearn.dummy.DummyRegressor" title="sklearn.dummy.DummyRegressor"><code class="xref py py-class docutils literal"><span class="pre">dummy.DummyRegressor</span></code></a>
can now be scored without supplying test samples.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11951">#11951</a> by <a class="reference external" href="https://github.com/JarnoRFB">Rüdiger Busche</a>.</li>
</ul>
</div>
<div class="section" id="id33">
<h3><a class="reference internal" href="../modules/classes.html#module-sklearn.ensemble" title="sklearn.ensemble"><code class="xref py py-mod docutils literal"><span class="pre">sklearn.ensemble</span></code></a><a class="headerlink" href="#id33" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <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"><span class="pre">ensemble.BaggingRegressor</span></code></a> and
<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"><span class="pre">ensemble.BaggingClassifier</span></code></a> can now be fit with missing/non-finite
values in X and/or multi-output Y to support wrapping pipelines that perform
their own imputation. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9707">#9707</a> by <a class="reference external" href="https://github.com/jimmywan">Jimmy Wan</a>.</li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> <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"><span class="pre">ensemble.GradientBoostingClassifier</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"><span class="pre">ensemble.GradientBoostingRegressor</span></code></a> now support early stopping
via <code class="docutils literal"><span class="pre">n_iter_no_change</span></code>, <code class="docutils literal"><span class="pre">validation_fraction</span></code> and <code class="docutils literal"><span class="pre">tol</span></code>. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/7071">#7071</a>
by <a class="reference external" href="https://github.com/raghavrv">Raghav RV</a></li>
<li><span class="raw-html"><span class="label label-success">Feature</span></span> Added <code class="docutils literal"><span class="pre">named_estimators_</span></code> parameter in
<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"><span class="pre">ensemble.VotingClassifier</span></code></a> to access fitted estimators.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9157">#9157</a> by <a class="reference external" href="https://github.com/herilalaina">Herilalaina Rakotoarison</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug when fitting <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"><span class="pre">ensemble.GradientBoostingClassifier</span></code></a> or
<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"><span class="pre">ensemble.GradientBoostingRegressor</span></code></a> with <code class="docutils literal"><span class="pre">warm_start=True</span></code> which
previously raised a segmentation fault due to a non-conversion of CSC matrix
into CSR format expected by <code class="docutils literal"><span class="pre">decision_function</span></code>. Similarly, Fortran-ordered
arrays are converted to C-ordered arrays in the dense case. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9991">#9991</a> by
<a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> Fixed a bug in <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"><span class="pre">ensemble.GradientBoostingRegressor</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"><span class="pre">ensemble.GradientBoostingClassifier</span></code></a> to have
feature importances summed and then normalized, rather than normalizing on a
per-tree basis. The previous behavior over-weighted the Gini importance of
features that appear in later stages. This issue only affected feature
importances. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11176">#11176</a> by <a class="reference external" href="https://github.com/gforsyth">Gil Forsyth</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> The default value of the <code class="docutils literal"><span class="pre">n_estimators</span></code> parameter of
<a class="reference internal" href="../modules/generated/sklearn.ensemble.RandomForestClassifier.html#sklearn.ensemble.RandomForestClassifier" title="sklearn.ensemble.RandomForestClassifier"><code class="xref py py-class docutils literal"><span class="pre">ensemble.RandomForestClassifier</span></code></a>, <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"><span class="pre">ensemble.RandomForestRegressor</span></code></a>,
<a class="reference internal" href="../modules/generated/sklearn.ensemble.ExtraTreesClassifier.html#sklearn.ensemble.ExtraTreesClassifier" title="sklearn.ensemble.ExtraTreesClassifier"><code class="xref py py-class docutils literal"><span class="pre">ensemble.ExtraTreesClassifier</span></code></a>, <a class="reference internal" href="../modules/generated/sklearn.ensemble.ExtraTreesRegressor.html#sklearn.ensemble.ExtraTreesRegressor" title="sklearn.ensemble.ExtraTreesRegressor"><code class="xref py py-class docutils literal"><span class="pre">ensemble.ExtraTreesRegressor</span></code></a>,
and <a class="reference internal" href="../modules/generated/sklearn.ensemble.RandomTreesEmbedding.html#sklearn.ensemble.RandomTreesEmbedding" title="sklearn.ensemble.RandomTreesEmbedding"><code class="xref py py-class docutils literal"><span class="pre">ensemble.RandomTreesEmbedding</span></code></a> will change from 10 in version 0.20
to 100 in 0.22. A FutureWarning is raised when the default value is used.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/11542">#11542</a> by <a class="reference external" href="https://github.com/annaayzenshtat">Anna Ayzenshtat</a>.</li>
<li><span class="raw-html"><span class="label label-warning">API Change</span></span> Classes derived from <code class="xref py py-class docutils literal"><span class="pre">ensemble.BaseBagging</span></code>. The attribute
<code class="docutils literal"><span class="pre">estimators_samples_</span></code> will return a list of arrays containing the indices
selected for each bootstrap instead of a list of arrays containing the mask
of the samples selected for each bootstrap. Indices allows to repeat samples
while mask does not allow this functionality.
<a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9524">#9524</a> by <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</li>
<li><span class="raw-html"><span class="label label-danger">Fix</span></span> <code class="xref py py-class docutils literal"><span class="pre">ensemble.BaseBagging</span></code> where one could not deterministically
reproduce <code class="docutils literal"><span class="pre">fit</span></code> result using the object attributes when <code class="docutils literal"><span class="pre">random_state</span></code>
is set. <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/9723">#9723</a> by <a class="reference external" href="https://github.com/glemaitre">Guillaume Lemaitre</a>.</li>