-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathpreprocessing.html
2014 lines (1772 loc) · 198 KB
/
preprocessing.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>
<html lang="en" data-content_root="../" >
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<meta property="og:title" content="6.3. Preprocessing data" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/preprocessing.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="The sklearn.preprocessing package provides several common utility functions and transformer classes to change raw feature vectors into a representation that is more suitable for the downstream esti..." />
<meta property="og:image" content="https://scikit-learn/stable/_images/sphx_glr_plot_map_data_to_normal_001.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="The sklearn.preprocessing package provides several common utility functions and transformer classes to change raw feature vectors into a representation that is more suitable for the downstream esti..." />
<title>6.3. Preprocessing data — scikit-learn 1.7.dev0 documentation</title>
<script data-cfasync="false">
document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
</script>
<!--
this give us a css class that will be invisible only if js is disabled
-->
<noscript>
<style>
.pst-js-only { display: none !important; }
</style>
</noscript>
<!-- Loaded before other Sphinx assets -->
<link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
<link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
<link rel="stylesheet" type="text/css" href="../_static/copybutton.css?v=76b2166b" />
<link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
<link rel="stylesheet" type="text/css" href="https://fonts.googleapis.com/css?family=Vibur" />
<link rel="stylesheet" type="text/css" href="../_static/jupyterlite_sphinx.css?v=2c9f8f05" />
<link rel="stylesheet" type="text/css" href="../_static/sg_gallery.css?v=d2d258e8" />
<link rel="stylesheet" type="text/css" href="../_static/sg_gallery-binder.css?v=f4aeca0c" />
<link rel="stylesheet" type="text/css" href="../_static/sg_gallery-dataframe.css?v=2082cf3c" />
<link rel="stylesheet" type="text/css" href="../_static/sg_gallery-rendered-html.css?v=1277b6f3" />
<link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
<link rel="stylesheet" type="text/css" href="../_static/styles/colors.css?v=cc94ab7d" />
<link rel="stylesheet" type="text/css" href="../_static/styles/custom.css?v=8f525996" />
<!-- So that users can add custom icons -->
<script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
<script src="../_static/documentation_options.js?v=473747f4"></script>
<script src="../_static/doctools.js?v=9bcbadda"></script>
<script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
<script src="../_static/clipboard.min.js?v=a7894cd8"></script>
<script src="../_static/copybutton.js?v=97f0b27d"></script>
<script src="../_static/jupyterlite_sphinx.js?v=96e329c5"></script>
<script src="../_static/design-tabs.js?v=f930bc37"></script>
<script data-domain="scikit-learn.org" defer="defer" src="https://views.scientific-python.org/js/script.js"></script>
<script async="async" src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml.js"></script>
<script>DOCUMENTATION_OPTIONS.pagename = 'modules/preprocessing';</script>
<script>
DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
DOCUMENTATION_OPTIONS.theme_switcher_json_url = 'https://scikit-learn.org/dev/_static/versions.json';
DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.7.dev0';
DOCUMENTATION_OPTIONS.show_version_warning_banner =
true;
</script>
<script src="../_static/scripts/dropdown.js?v=d6825577"></script>
<script src="../_static/scripts/version-switcher.js?v=a6dd8357"></script>
<script src="../_static/scripts/sg_plotly_resize.js?v=2167d4db"></script>
<link rel="canonical" href="https://scikit-learn.org/stable/modules/preprocessing.html" />
<link rel="icon" href="../_static/favicon.ico"/>
<link rel="author" title="About these documents" href="../about.html" />
<link rel="search" title="Search" href="../search.html" />
<link rel="next" title="6.4. Imputation of missing values" href="impute.html" />
<link rel="prev" title="6.2. Feature extraction" href="feature_extraction.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
<meta name="docsearch:version" content="1.7" />
</head>
<body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
<div id="pst-scroll-pixel-helper"></div>
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
<dialog id="pst-search-dialog">
<form class="bd-search d-flex align-items-center"
action="../search.html"
method="get">
<i class="fa-solid fa-magnifying-glass"></i>
<input type="search"
class="form-control"
name="q"
placeholder="Search the docs ..."
aria-label="Search the docs ..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form>
</dialog>
<div class="pst-async-banner-revealer d-none">
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>
<header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
<div class="bd-header__inner bd-page-width">
<button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
<span class="fa-solid fa-bars"></span>
</button>
<div class=" navbar-header-items__start">
<div class="navbar-item">
<a class="navbar-brand logo" href="../index.html">
<img src="../_static/scikit-learn-logo-small.png" class="logo__image only-light" alt="scikit-learn homepage"/>
<img src="../_static/scikit-learn-logo-small.png" class="logo__image only-dark pst-js-only" alt="scikit-learn homepage"/>
</a></div>
</div>
<div class=" navbar-header-items">
<div class="me-auto navbar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="../install.html">
Install
</a>
</li>
<li class="nav-item current active">
<a class="nav-link nav-internal" href="../user_guide.html">
User Guide
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../api/index.html">
API
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../auto_examples/index.html">
Examples
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-external" href="https://blog.scikit-learn.org/">
Community
</a>
</li>
<li class="nav-item dropdown">
<button class="btn dropdown-toggle nav-item" type="button"
data-bs-toggle="dropdown" aria-expanded="false"
aria-controls="pst-nav-more-links">
More
</button>
<ul id="pst-nav-more-links" class="dropdown-menu">
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../getting_started.html">
Getting Started
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../whats_new.html">
Release History
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../glossary.html">
Glossary
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../developers/index.html">
Development
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../faq.html">
FAQ
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../support.html">
Support
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../related_projects.html">
Related Projects
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../roadmap.html">
Roadmap
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../governance.html">
Governance
</a>
</li>
<li class=" ">
<a class="nav-link dropdown-item nav-internal" href="../about.html">
About us
</a>
</li>
</ul>
</li>
</ul>
</nav></div>
</div>
<div class="navbar-header-items__end">
<div class="navbar-item navbar-persistent--container">
<button class="btn btn-sm pst-navbar-icon search-button search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
</div>
<div class="navbar-item">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
</button></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/scikit-learn/scikit-learn" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
<div class="navbar-item">
<div class="version-switcher__container dropdown pst-js-only">
<button id="pst-version-switcher-button-2"
type="button"
class="version-switcher__button btn btn-sm dropdown-toggle"
data-bs-toggle="dropdown"
aria-haspopup="listbox"
aria-controls="pst-version-switcher-list-2"
aria-label="Version switcher list"
>
Choose version <!-- this text may get changed later by javascript -->
<span class="caret"></span>
</button>
<div id="pst-version-switcher-list-2"
class="version-switcher__menu dropdown-menu list-group-flush py-0"
role="listbox" aria-labelledby="pst-version-switcher-button-2">
<!-- dropdown will be populated by javascript on page load -->
</div>
</div></div>
</div>
</div>
<div class="navbar-persistent--mobile">
<button class="btn btn-sm pst-navbar-icon search-button search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
</div>
<button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
<span class="fa-solid fa-outdent"></span>
</button>
</div>
</header>
<div class="bd-container">
<div class="bd-container__inner bd-page-width">
<dialog id="pst-primary-sidebar-modal"></dialog>
<div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
<div class="sidebar-header-items sidebar-primary__section">
<div class="sidebar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="../install.html">
Install
</a>
</li>
<li class="nav-item current active">
<a class="nav-link nav-internal" href="../user_guide.html">
User Guide
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../api/index.html">
API
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../auto_examples/index.html">
Examples
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-external" href="https://blog.scikit-learn.org/">
Community
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../getting_started.html">
Getting Started
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../whats_new.html">
Release History
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../glossary.html">
Glossary
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../developers/index.html">
Development
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../faq.html">
FAQ
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../support.html">
Support
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../related_projects.html">
Related Projects
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../roadmap.html">
Roadmap
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../governance.html">
Governance
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="../about.html">
About us
</a>
</li>
</ul>
</nav></div>
</div>
<div class="sidebar-header-items__end">
<div class="navbar-item">
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light" title="Light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark" title="Dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto" title="System Settings"></i>
</button></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/scikit-learn/scikit-learn" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
<div class="navbar-item">
<div class="version-switcher__container dropdown pst-js-only">
<button id="pst-version-switcher-button-3"
type="button"
class="version-switcher__button btn btn-sm dropdown-toggle"
data-bs-toggle="dropdown"
aria-haspopup="listbox"
aria-controls="pst-version-switcher-list-3"
aria-label="Version switcher list"
>
Choose version <!-- this text may get changed later by javascript -->
<span class="caret"></span>
</button>
<div id="pst-version-switcher-list-3"
class="version-switcher__menu dropdown-menu list-group-flush py-0"
role="listbox" aria-labelledby="pst-version-switcher-button-3">
<!-- dropdown will be populated by javascript on page load -->
</div>
</div></div>
</div>
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item">
<nav class="bd-docs-nav bd-links"
aria-label="Section Navigation">
<p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
<div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
<li class="toctree-l1 has-children"><a class="reference internal" href="../supervised_learning.html">1. Supervised learning</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="linear_model.html">1.1. Linear Models</a></li>
<li class="toctree-l2"><a class="reference internal" href="lda_qda.html">1.2. Linear and Quadratic Discriminant Analysis</a></li>
<li class="toctree-l2"><a class="reference internal" href="kernel_ridge.html">1.3. Kernel ridge regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="svm.html">1.4. Support Vector Machines</a></li>
<li class="toctree-l2"><a class="reference internal" href="sgd.html">1.5. Stochastic Gradient Descent</a></li>
<li class="toctree-l2"><a class="reference internal" href="neighbors.html">1.6. Nearest Neighbors</a></li>
<li class="toctree-l2"><a class="reference internal" href="gaussian_process.html">1.7. Gaussian Processes</a></li>
<li class="toctree-l2"><a class="reference internal" href="cross_decomposition.html">1.8. Cross decomposition</a></li>
<li class="toctree-l2"><a class="reference internal" href="naive_bayes.html">1.9. Naive Bayes</a></li>
<li class="toctree-l2"><a class="reference internal" href="tree.html">1.10. Decision Trees</a></li>
<li class="toctree-l2"><a class="reference internal" href="ensemble.html">1.11. Ensembles: Gradient boosting, random forests, bagging, voting, stacking</a></li>
<li class="toctree-l2"><a class="reference internal" href="multiclass.html">1.12. Multiclass and multioutput algorithms</a></li>
<li class="toctree-l2"><a class="reference internal" href="feature_selection.html">1.13. Feature selection</a></li>
<li class="toctree-l2"><a class="reference internal" href="semi_supervised.html">1.14. Semi-supervised learning</a></li>
<li class="toctree-l2"><a class="reference internal" href="isotonic.html">1.15. Isotonic regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="calibration.html">1.16. Probability calibration</a></li>
<li class="toctree-l2"><a class="reference internal" href="neural_networks_supervised.html">1.17. Neural network models (supervised)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../unsupervised_learning.html">2. Unsupervised learning</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="mixture.html">2.1. Gaussian mixture models</a></li>
<li class="toctree-l2"><a class="reference internal" href="manifold.html">2.2. Manifold learning</a></li>
<li class="toctree-l2"><a class="reference internal" href="clustering.html">2.3. Clustering</a></li>
<li class="toctree-l2"><a class="reference internal" href="biclustering.html">2.4. Biclustering</a></li>
<li class="toctree-l2"><a class="reference internal" href="decomposition.html">2.5. Decomposing signals in components (matrix factorization problems)</a></li>
<li class="toctree-l2"><a class="reference internal" href="covariance.html">2.6. Covariance estimation</a></li>
<li class="toctree-l2"><a class="reference internal" href="outlier_detection.html">2.7. Novelty and Outlier Detection</a></li>
<li class="toctree-l2"><a class="reference internal" href="density.html">2.8. Density Estimation</a></li>
<li class="toctree-l2"><a class="reference internal" href="neural_networks_unsupervised.html">2.9. Neural network models (unsupervised)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../model_selection.html">3. Model selection and evaluation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="cross_validation.html">3.1. Cross-validation: evaluating estimator performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="grid_search.html">3.2. Tuning the hyper-parameters of an estimator</a></li>
<li class="toctree-l2"><a class="reference internal" href="classification_threshold.html">3.3. Tuning the decision threshold for class prediction</a></li>
<li class="toctree-l2"><a class="reference internal" href="model_evaluation.html">3.4. Metrics and scoring: quantifying the quality of predictions</a></li>
<li class="toctree-l2"><a class="reference internal" href="learning_curve.html">3.5. Validation curves: plotting scores to evaluate models</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../inspection.html">4. Inspection</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="partial_dependence.html">4.1. Partial Dependence and Individual Conditional Expectation plots</a></li>
<li class="toctree-l2"><a class="reference internal" href="permutation_importance.html">4.2. Permutation feature importance</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../visualizations.html">5. Visualizations</a></li>
<li class="toctree-l1 current active has-children"><a class="reference internal" href="../data_transforms.html">6. Dataset transformations</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
<li class="toctree-l2"><a class="reference internal" href="compose.html">6.1. Pipelines and composite estimators</a></li>
<li class="toctree-l2"><a class="reference internal" href="feature_extraction.html">6.2. Feature extraction</a></li>
<li class="toctree-l2 current active"><a class="current reference internal" href="#">6.3. Preprocessing data</a></li>
<li class="toctree-l2"><a class="reference internal" href="impute.html">6.4. Imputation of missing values</a></li>
<li class="toctree-l2"><a class="reference internal" href="unsupervised_reduction.html">6.5. Unsupervised dimensionality reduction</a></li>
<li class="toctree-l2"><a class="reference internal" href="random_projection.html">6.6. Random Projection</a></li>
<li class="toctree-l2"><a class="reference internal" href="kernel_approximation.html">6.7. Kernel Approximation</a></li>
<li class="toctree-l2"><a class="reference internal" href="metrics.html">6.8. Pairwise metrics, Affinities and Kernels</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing_targets.html">6.9. Transforming the prediction target (<code class="docutils literal notranslate"><span class="pre">y</span></code>)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../datasets.html">7. Dataset loading utilities</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="../datasets/toy_dataset.html">7.1. Toy datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/real_world.html">7.2. Real world datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/sample_generators.html">7.3. Generated datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/loading_other_datasets.html">7.4. Loading other datasets</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../computing.html">8. Computing with scikit-learn</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="../computing/scaling_strategies.html">8.1. Strategies to scale computationally: bigger data</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/computational_performance.html">8.2. Computational Performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/parallelism.html">8.3. Parallelism, resource management, and configuration</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../model_persistence.html">9. Model persistence</a></li>
<li class="toctree-l1"><a class="reference internal" href="../common_pitfalls.html">10. Common pitfalls and recommended practices</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../dispatching.html">11. Dispatching</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="array_api.html">11.1. Array API support (experimental)</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../machine_learning_map.html">12. Choosing the right estimator</a></li>
<li class="toctree-l1"><a class="reference internal" href="../presentations.html">13. External Resources, Videos and Talks</a></li>
</ul>
</div>
</nav></div>
</div>
<div class="sidebar-primary-items__end sidebar-primary__section">
</div>
</div>
<main id="main-content" class="bd-main" role="main">
<div class="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item">
<nav aria-label="Breadcrumb" class="d-print-none">
<ul class="bd-breadcrumbs">
<li class="breadcrumb-item breadcrumb-home">
<a href="../index.html" class="nav-link" aria-label="Home">
<i class="fa-solid fa-home"></i>
</a>
</li>
<li class="breadcrumb-item"><a href="../user_guide.html" class="nav-link">User Guide</a></li>
<li class="breadcrumb-item"><a href="../data_transforms.html" class="nav-link"><span class="section-number">6. </span>Dataset transformations</a></li>
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis"><span class="section-number">6.3. </span>Preprocessing data</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="preprocessing-data">
<span id="preprocessing"></span><h1><span class="section-number">6.3. </span>Preprocessing data<a class="headerlink" href="#preprocessing-data" title="Link to this heading">#</a></h1>
<p>The <code class="docutils literal notranslate"><span class="pre">sklearn.preprocessing</span></code> package provides several common
utility functions and transformer classes to change raw feature vectors
into a representation that is more suitable for the downstream estimators.</p>
<p>In general, many learning algorithms such as linear models benefit from standardization of the data set
(see <a class="reference internal" href="../auto_examples/preprocessing/plot_scaling_importance.html#sphx-glr-auto-examples-preprocessing-plot-scaling-importance-py"><span class="std std-ref">Importance of Feature Scaling</span></a>).
If some outliers are present in the set, robust scalers or other transformers can
be more appropriate. The behaviors of the different scalers, transformers, and
normalizers on a dataset containing marginal outliers are highlighted in
<a class="reference internal" href="../auto_examples/preprocessing/plot_all_scaling.html#sphx-glr-auto-examples-preprocessing-plot-all-scaling-py"><span class="std std-ref">Compare the effect of different scalers on data with outliers</span></a>.</p>
<section id="standardization-or-mean-removal-and-variance-scaling">
<span id="preprocessing-scaler"></span><h2><span class="section-number">6.3.1. </span>Standardization, or mean removal and variance scaling<a class="headerlink" href="#standardization-or-mean-removal-and-variance-scaling" title="Link to this heading">#</a></h2>
<p><strong>Standardization</strong> of datasets is a <strong>common requirement for many
machine learning estimators</strong> implemented in scikit-learn; they might behave
badly if the individual features do not more or less look like standard
normally distributed data: Gaussian with <strong>zero mean and unit variance</strong>.</p>
<p>In practice we often ignore the shape of the distribution and just
transform the data to center it by removing the mean value of each
feature, then scale it by dividing non-constant features by their
standard deviation.</p>
<p>For instance, many elements used in the objective function of
a learning algorithm (such as the RBF kernel of Support Vector
Machines or the l1 and l2 regularizers of linear models) may assume that
all features are centered around zero or have variance in the same
order. If a feature has a variance that is orders of magnitude larger
than others, it might dominate the objective function and make the
estimator unable to learn from other features correctly as expected.</p>
<p>The <a class="reference internal" href="../api/sklearn.preprocessing.html#module-sklearn.preprocessing" title="sklearn.preprocessing"><code class="xref py py-mod docutils literal notranslate"><span class="pre">preprocessing</span></code></a> module provides the
<a class="reference internal" href="generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">StandardScaler</span></code></a> utility class, which is a quick and
easy way to perform the following operation on an array-like
dataset:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn</span><span class="w"> </span><span class="kn">import</span> <span class="n">preprocessing</span>
<span class="gp">>>> </span><span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
<span class="gp">>>> </span><span class="n">X_train</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">2.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">]])</span>
<span class="gp">>>> </span><span class="n">scaler</span> <span class="o">=</span> <span class="n">preprocessing</span><span class="o">.</span><span class="n">StandardScaler</span><span class="p">()</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">scaler</span>
<span class="go">StandardScaler()</span>
<span class="gp">>>> </span><span class="n">scaler</span><span class="o">.</span><span class="n">mean_</span>
<span class="go">array([1. ..., 0. ..., 0.33...])</span>
<span class="gp">>>> </span><span class="n">scaler</span><span class="o">.</span><span class="n">scale_</span>
<span class="go">array([0.81..., 0.81..., 1.24...])</span>
<span class="gp">>>> </span><span class="n">X_scaled</span> <span class="o">=</span> <span class="n">scaler</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_scaled</span>
<span class="go">array([[ 0. ..., -1.22..., 1.33...],</span>
<span class="go"> [ 1.22..., 0. ..., -0.26...],</span>
<span class="go"> [-1.22..., 1.22..., -1.06...]])</span>
</pre></div>
</div>
<p>Scaled data has zero mean and unit variance:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">X_scaled</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="go">array([0., 0., 0.])</span>
<span class="gp">>>> </span><span class="n">X_scaled</span><span class="o">.</span><span class="n">std</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="go">array([1., 1., 1.])</span>
</pre></div>
</div>
<p>This class implements the <code class="docutils literal notranslate"><span class="pre">Transformer</span></code> API to compute the mean and
standard deviation on a training set so as to be able to later re-apply the
same transformation on the testing set. This class is hence suitable for
use in the early steps of a <a class="reference internal" href="generated/sklearn.pipeline.Pipeline.html#sklearn.pipeline.Pipeline" title="sklearn.pipeline.Pipeline"><code class="xref py py-class docutils literal notranslate"><span class="pre">Pipeline</span></code></a>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">make_classification</span>
<span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.linear_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">LogisticRegression</span>
<span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.model_selection</span><span class="w"> </span><span class="kn">import</span> <span class="n">train_test_split</span>
<span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">make_pipeline</span>
<span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.preprocessing</span><span class="w"> </span><span class="kn">import</span> <span class="n">StandardScaler</span>
<span class="gp">>>> </span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">make_classification</span><span class="p">(</span><span class="n">random_state</span><span class="o">=</span><span class="mi">42</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">42</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">pipe</span> <span class="o">=</span> <span class="n">make_pipeline</span><span class="p">(</span><span class="n">StandardScaler</span><span class="p">(),</span> <span class="n">LogisticRegression</span><span class="p">())</span>
<span class="gp">>>> </span><span class="n">pipe</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span> <span class="c1"># apply scaling on training data</span>
<span class="go">Pipeline(steps=[('standardscaler', StandardScaler()),</span>
<span class="go"> ('logisticregression', LogisticRegression())])</span>
<span class="gp">>>> </span><span class="n">pipe</span><span class="o">.</span><span class="n">score</span><span class="p">(</span><span class="n">X_test</span><span class="p">,</span> <span class="n">y_test</span><span class="p">)</span> <span class="c1"># apply scaling on testing data, without leaking training data.</span>
<span class="go">0.96</span>
</pre></div>
</div>
<p>It is possible to disable either centering or scaling by either
passing <code class="docutils literal notranslate"><span class="pre">with_mean=False</span></code> or <code class="docutils literal notranslate"><span class="pre">with_std=False</span></code> to the constructor
of <a class="reference internal" href="generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">StandardScaler</span></code></a>.</p>
<section id="scaling-features-to-a-range">
<h3><span class="section-number">6.3.1.1. </span>Scaling features to a range<a class="headerlink" href="#scaling-features-to-a-range" title="Link to this heading">#</a></h3>
<p>An alternative standardization is scaling features to
lie between a given minimum and maximum value, often between zero and one,
or so that the maximum absolute value of each feature is scaled to unit size.
This can be achieved using <a class="reference internal" href="generated/sklearn.preprocessing.MinMaxScaler.html#sklearn.preprocessing.MinMaxScaler" title="sklearn.preprocessing.MinMaxScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MinMaxScaler</span></code></a> or <a class="reference internal" href="generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler" title="sklearn.preprocessing.MaxAbsScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxAbsScaler</span></code></a>,
respectively.</p>
<p>The motivation to use this scaling includes robustness to very small
standard deviations of features and preserving zero entries in sparse data.</p>
<p>Here is an example to scale a toy data matrix to the <code class="docutils literal notranslate"><span class="pre">[0,</span> <span class="pre">1]</span></code> range:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">X_train</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">2.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">]])</span>
<span class="gp">...</span>
<span class="gp">>>> </span><span class="n">min_max_scaler</span> <span class="o">=</span> <span class="n">preprocessing</span><span class="o">.</span><span class="n">MinMaxScaler</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">X_train_minmax</span> <span class="o">=</span> <span class="n">min_max_scaler</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_train_minmax</span>
<span class="go">array([[0.5 , 0. , 1. ],</span>
<span class="go"> [1. , 0.5 , 0.33333333],</span>
<span class="go"> [0. , 1. , 0. ]])</span>
</pre></div>
</div>
<p>The same instance of the transformer can then be applied to some new test data
unseen during the fit call: the same scaling and shifting operations will be
applied to be consistent with the transformation performed on the train data:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">X_test</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span><span class="o">-</span><span class="mf">3.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">,</span> <span class="mf">4.</span><span class="p">]])</span>
<span class="gp">>>> </span><span class="n">X_test_minmax</span> <span class="o">=</span> <span class="n">min_max_scaler</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_test_minmax</span>
<span class="go">array([[-1.5 , 0. , 1.66666667]])</span>
</pre></div>
</div>
<p>It is possible to introspect the scaler attributes to find about the exact
nature of the transformation learned on the training data:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">min_max_scaler</span><span class="o">.</span><span class="n">scale_</span>
<span class="go">array([0.5 , 0.5 , 0.33...])</span>
<span class="gp">>>> </span><span class="n">min_max_scaler</span><span class="o">.</span><span class="n">min_</span>
<span class="go">array([0. , 0.5 , 0.33...])</span>
</pre></div>
</div>
<p>If <a class="reference internal" href="generated/sklearn.preprocessing.MinMaxScaler.html#sklearn.preprocessing.MinMaxScaler" title="sklearn.preprocessing.MinMaxScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MinMaxScaler</span></code></a> is given an explicit <code class="docutils literal notranslate"><span class="pre">feature_range=(min,</span> <span class="pre">max)</span></code> the
full formula is:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X_std</span> <span class="o">=</span> <span class="p">(</span><span class="n">X</span> <span class="o">-</span> <span class="n">X</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span> <span class="o">/</span> <span class="p">(</span><span class="n">X</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> <span class="o">-</span> <span class="n">X</span><span class="o">.</span><span class="n">min</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">))</span>
<span class="n">X_scaled</span> <span class="o">=</span> <span class="n">X_std</span> <span class="o">*</span> <span class="p">(</span><span class="nb">max</span> <span class="o">-</span> <span class="nb">min</span><span class="p">)</span> <span class="o">+</span> <span class="nb">min</span>
</pre></div>
</div>
<p><a class="reference internal" href="generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler" title="sklearn.preprocessing.MaxAbsScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxAbsScaler</span></code></a> works in a very similar fashion, but scales in a way
that the training data lies within the range <code class="docutils literal notranslate"><span class="pre">[-1,</span> <span class="pre">1]</span></code> by dividing through
the largest maximum value in each feature. It is meant for data
that is already centered at zero or sparse data.</p>
<p>Here is how to use the toy data from the previous example with this scaler:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">X_train</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">2.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">0.</span><span class="p">],</span>
<span class="gp">... </span> <span class="p">[</span> <span class="mf">0.</span><span class="p">,</span> <span class="mf">1.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">]])</span>
<span class="gp">...</span>
<span class="gp">>>> </span><span class="n">max_abs_scaler</span> <span class="o">=</span> <span class="n">preprocessing</span><span class="o">.</span><span class="n">MaxAbsScaler</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">X_train_maxabs</span> <span class="o">=</span> <span class="n">max_abs_scaler</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_train_maxabs</span>
<span class="go">array([[ 0.5, -1. , 1. ],</span>
<span class="go"> [ 1. , 0. , 0. ],</span>
<span class="go"> [ 0. , 1. , -0.5]])</span>
<span class="gp">>>> </span><span class="n">X_test</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span> <span class="o">-</span><span class="mf">3.</span><span class="p">,</span> <span class="o">-</span><span class="mf">1.</span><span class="p">,</span> <span class="mf">4.</span><span class="p">]])</span>
<span class="gp">>>> </span><span class="n">X_test_maxabs</span> <span class="o">=</span> <span class="n">max_abs_scaler</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_test_maxabs</span>
<span class="go">array([[-1.5, -1. , 2. ]])</span>
<span class="gp">>>> </span><span class="n">max_abs_scaler</span><span class="o">.</span><span class="n">scale_</span>
<span class="go">array([2., 1., 2.])</span>
</pre></div>
</div>
</section>
<section id="scaling-sparse-data">
<h3><span class="section-number">6.3.1.2. </span>Scaling sparse data<a class="headerlink" href="#scaling-sparse-data" title="Link to this heading">#</a></h3>
<p>Centering sparse data would destroy the sparseness structure in the data, and
thus rarely is a sensible thing to do. However, it can make sense to scale
sparse inputs, especially if features are on different scales.</p>
<p><a class="reference internal" href="generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler" title="sklearn.preprocessing.MaxAbsScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MaxAbsScaler</span></code></a> was specifically designed for scaling
sparse data, and is the recommended way to go about this.
However, <a class="reference internal" href="generated/sklearn.preprocessing.StandardScaler.html#sklearn.preprocessing.StandardScaler" title="sklearn.preprocessing.StandardScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">StandardScaler</span></code></a> can accept <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code>
matrices as input, as long as <code class="docutils literal notranslate"><span class="pre">with_mean=False</span></code> is explicitly passed
to the constructor. Otherwise a <code class="docutils literal notranslate"><span class="pre">ValueError</span></code> will be raised as
silently centering would break the sparsity and would often crash the
execution by allocating excessive amounts of memory unintentionally.
<a class="reference internal" href="generated/sklearn.preprocessing.RobustScaler.html#sklearn.preprocessing.RobustScaler" title="sklearn.preprocessing.RobustScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">RobustScaler</span></code></a> cannot be fitted to sparse inputs, but you can use
the <code class="docutils literal notranslate"><span class="pre">transform</span></code> method on sparse inputs.</p>
<p>Note that the scalers accept both Compressed Sparse Rows and Compressed
Sparse Columns format (see <code class="docutils literal notranslate"><span class="pre">scipy.sparse.csr_matrix</span></code> and
<code class="docutils literal notranslate"><span class="pre">scipy.sparse.csc_matrix</span></code>). Any other sparse input will be <strong>converted to
the Compressed Sparse Rows representation</strong>. To avoid unnecessary memory
copies, it is recommended to choose the CSR or CSC representation upstream.</p>
<p>Finally, if the centered data is expected to be small enough, explicitly
converting the input to an array using the <code class="docutils literal notranslate"><span class="pre">toarray</span></code> method of sparse matrices
is another option.</p>
</section>
<section id="scaling-data-with-outliers">
<h3><span class="section-number">6.3.1.3. </span>Scaling data with outliers<a class="headerlink" href="#scaling-data-with-outliers" title="Link to this heading">#</a></h3>
<p>If your data contains many outliers, scaling using the mean and variance
of the data is likely to not work very well. In these cases, you can use
<a class="reference internal" href="generated/sklearn.preprocessing.RobustScaler.html#sklearn.preprocessing.RobustScaler" title="sklearn.preprocessing.RobustScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">RobustScaler</span></code></a> as a drop-in replacement instead. It uses
more robust estimates for the center and range of your data.</p>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="references">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">References<a class="headerlink" href="#references" title="Link to this dropdown">#</a></span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
<p class="sd-card-text">Further discussion on the importance of centering and scaling data is
available on this FAQ: <a class="reference external" href="http://www.faqs.org/faqs/ai-faq/neural-nets/part2/section-16.html">Should I normalize/standardize/rescale the data?</a></p>
</div>
</details><details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="scaling-vs-whitening">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Scaling vs Whitening<a class="headerlink" href="#scaling-vs-whitening" title="Link to this dropdown">#</a></span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
<p class="sd-card-text">It is sometimes not enough to center and scale the features
independently, since a downstream model can further make some assumption
on the linear independence of the features.</p>
<p class="sd-card-text">To address this issue you can use <a class="reference internal" href="generated/sklearn.decomposition.PCA.html#sklearn.decomposition.PCA" title="sklearn.decomposition.PCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">PCA</span></code></a> with
<code class="docutils literal notranslate"><span class="pre">whiten=True</span></code> to further remove the linear correlation across features.</p>
</div>
</details></section>
<section id="centering-kernel-matrices">
<span id="kernel-centering"></span><h3><span class="section-number">6.3.1.4. </span>Centering kernel matrices<a class="headerlink" href="#centering-kernel-matrices" title="Link to this heading">#</a></h3>
<p>If you have a kernel matrix of a kernel <span class="math notranslate nohighlight">\(K\)</span> that computes a dot product
in a feature space (possibly implicitly) defined by a function
<span class="math notranslate nohighlight">\(\phi(\cdot)\)</span>, a <a class="reference internal" href="generated/sklearn.preprocessing.KernelCenterer.html#sklearn.preprocessing.KernelCenterer" title="sklearn.preprocessing.KernelCenterer"><code class="xref py py-class docutils literal notranslate"><span class="pre">KernelCenterer</span></code></a> can transform the kernel matrix
so that it contains inner products in the feature space defined by <span class="math notranslate nohighlight">\(\phi\)</span>
followed by the removal of the mean in that space. In other words,
<a class="reference internal" href="generated/sklearn.preprocessing.KernelCenterer.html#sklearn.preprocessing.KernelCenterer" title="sklearn.preprocessing.KernelCenterer"><code class="xref py py-class docutils literal notranslate"><span class="pre">KernelCenterer</span></code></a> computes the centered Gram matrix associated to a
positive semidefinite kernel <span class="math notranslate nohighlight">\(K\)</span>.</p>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="mathematical-formulation">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Mathematical formulation<a class="headerlink" href="#mathematical-formulation" title="Link to this dropdown">#</a></span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
<p class="sd-card-text">We can have a look at the mathematical formulation now that we have the
intuition. Let <span class="math notranslate nohighlight">\(K\)</span> be a kernel matrix of shape <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_samples)</span></code>
computed from <span class="math notranslate nohighlight">\(X\)</span>, a data matrix of shape <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_features)</span></code>,
during the <code class="docutils literal notranslate"><span class="pre">fit</span></code> step. <span class="math notranslate nohighlight">\(K\)</span> is defined by</p>
<div class="math notranslate nohighlight">
\[K(X, X) = \phi(X) . \phi(X)^{T}\]</div>
<p class="sd-card-text"><span class="math notranslate nohighlight">\(\phi(X)\)</span> is a function mapping of <span class="math notranslate nohighlight">\(X\)</span> to a Hilbert space. A
centered kernel <span class="math notranslate nohighlight">\(\tilde{K}\)</span> is defined as:</p>
<div class="math notranslate nohighlight">
\[\tilde{K}(X, X) = \tilde{\phi}(X) . \tilde{\phi}(X)^{T}\]</div>
<p class="sd-card-text">where <span class="math notranslate nohighlight">\(\tilde{\phi}(X)\)</span> results from centering <span class="math notranslate nohighlight">\(\phi(X)\)</span> in the
Hilbert space.</p>
<p class="sd-card-text">Thus, one could compute <span class="math notranslate nohighlight">\(\tilde{K}\)</span> by mapping <span class="math notranslate nohighlight">\(X\)</span> using the
function <span class="math notranslate nohighlight">\(\phi(\cdot)\)</span> and center the data in this new space. However,
kernels are often used because they allow some algebra calculations that
avoid computing explicitly this mapping using <span class="math notranslate nohighlight">\(\phi(\cdot)\)</span>. Indeed, one
can implicitly center as shown in Appendix B in <a class="reference internal" href="#scholkopf1998" id="id1"><span>[Scholkopf1998]</span></a>:</p>
<div class="math notranslate nohighlight">
\[\tilde{K} = K - 1_{\text{n}_{samples}} K - K 1_{\text{n}_{samples}} + 1_{\text{n}_{samples}} K 1_{\text{n}_{samples}}\]</div>
<p class="sd-card-text"><span class="math notranslate nohighlight">\(1_{\text{n}_{samples}}\)</span> is a matrix of <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_samples)</span></code> where
all entries are equal to <span class="math notranslate nohighlight">\(\frac{1}{\text{n}_{samples}}\)</span>. In the
<code class="docutils literal notranslate"><span class="pre">transform</span></code> step, the kernel becomes <span class="math notranslate nohighlight">\(K_{test}(X, Y)\)</span> defined as:</p>
<div class="math notranslate nohighlight">
\[K_{test}(X, Y) = \phi(Y) . \phi(X)^{T}\]</div>
<p class="sd-card-text"><span class="math notranslate nohighlight">\(Y\)</span> is the test dataset of shape <code class="docutils literal notranslate"><span class="pre">(n_samples_test,</span> <span class="pre">n_features)</span></code> and thus
<span class="math notranslate nohighlight">\(K_{test}\)</span> is of shape <code class="docutils literal notranslate"><span class="pre">(n_samples_test,</span> <span class="pre">n_samples)</span></code>. In this case,
centering <span class="math notranslate nohighlight">\(K_{test}\)</span> is done as:</p>
<div class="math notranslate nohighlight">
\[\tilde{K}_{test}(X, Y) = K_{test} - 1'_{\text{n}_{samples}} K - K_{test} 1_{\text{n}_{samples}} + 1'_{\text{n}_{samples}} K 1_{\text{n}_{samples}}\]</div>
<p class="sd-card-text"><span class="math notranslate nohighlight">\(1'_{\text{n}_{samples}}\)</span> is a matrix of shape
<code class="docutils literal notranslate"><span class="pre">(n_samples_test,</span> <span class="pre">n_samples)</span></code> where all entries are equal to
<span class="math notranslate nohighlight">\(\frac{1}{\text{n}_{samples}}\)</span>.</p>
<p class="rubric">References</p>
<div role="list" class="citation-list">
<div class="citation" id="scholkopf1998" role="doc-biblioentry">
<span class="label"><span class="fn-bracket">[</span><a role="doc-backlink" href="#id1">Scholkopf1998</a><span class="fn-bracket">]</span></span>
<p class="sd-card-text">B. Schölkopf, A. Smola, and K.R. Müller,
<a class="reference external" href="https://www.mlpack.org/papers/kpca.pdf">“Nonlinear component analysis as a kernel eigenvalue problem.”</a>
Neural computation 10.5 (1998): 1299-1319.</p>
</div>
</div>
</div>
</details></section>
</section>
<section id="non-linear-transformation">
<span id="preprocessing-transformer"></span><h2><span class="section-number">6.3.2. </span>Non-linear transformation<a class="headerlink" href="#non-linear-transformation" title="Link to this heading">#</a></h2>
<p>Two types of transformations are available: quantile transforms and power
transforms. Both quantile and power transforms are based on monotonic
transformations of the features and thus preserve the rank of the values
along each feature.</p>
<p>Quantile transforms put all features into the same desired distribution based
on the formula <span class="math notranslate nohighlight">\(G^{-1}(F(X))\)</span> where <span class="math notranslate nohighlight">\(F\)</span> is the cumulative
distribution function of the feature and <span class="math notranslate nohighlight">\(G^{-1}\)</span> the
<a class="reference external" href="https://en.wikipedia.org/wiki/Quantile_function">quantile function</a> of the
desired output distribution <span class="math notranslate nohighlight">\(G\)</span>. This formula is using the two following
facts: (i) if <span class="math notranslate nohighlight">\(X\)</span> is a random variable with a continuous cumulative
distribution function <span class="math notranslate nohighlight">\(F\)</span> then <span class="math notranslate nohighlight">\(F(X)\)</span> is uniformly distributed on
<span class="math notranslate nohighlight">\([0,1]\)</span>; (ii) if <span class="math notranslate nohighlight">\(U\)</span> is a random variable with uniform distribution
on <span class="math notranslate nohighlight">\([0,1]\)</span> then <span class="math notranslate nohighlight">\(G^{-1}(U)\)</span> has distribution <span class="math notranslate nohighlight">\(G\)</span>. By performing
a rank transformation, a quantile transform smooths out unusual distributions
and is less influenced by outliers than scaling methods. It does, however,
distort correlations and distances within and across features.</p>
<p>Power transforms are a family of parametric transformations that aim to map
data from any distribution to as close to a Gaussian distribution.</p>
<section id="mapping-to-a-uniform-distribution">
<h3><span class="section-number">6.3.2.1. </span>Mapping to a Uniform distribution<a class="headerlink" href="#mapping-to-a-uniform-distribution" title="Link to this heading">#</a></h3>
<p><a class="reference internal" href="generated/sklearn.preprocessing.QuantileTransformer.html#sklearn.preprocessing.QuantileTransformer" title="sklearn.preprocessing.QuantileTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">QuantileTransformer</span></code></a> provides a non-parametric
transformation to map the data to a uniform distribution
with values between 0 and 1:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">load_iris</span>
<span class="gp">>>> </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.model_selection</span><span class="w"> </span><span class="kn">import</span> <span class="n">train_test_split</span>
<span class="gp">>>> </span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_iris</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">quantile_transformer</span> <span class="o">=</span> <span class="n">preprocessing</span><span class="o">.</span><span class="n">QuantileTransformer</span><span class="p">(</span><span class="n">random_state</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_train_trans</span> <span class="o">=</span> <span class="n">quantile_transformer</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">X_train</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_test_trans</span> <span class="o">=</span> <span class="n">quantile_transformer</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">np</span><span class="o">.</span><span class="n">percentile</span><span class="p">(</span><span class="n">X_train</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">75</span><span class="p">,</span> <span class="mi">100</span><span class="p">])</span>
<span class="go">array([ 4.3, 5.1, 5.8, 6.5, 7.9])</span>
</pre></div>
</div>
<p>This feature corresponds to the sepal length in cm. Once the quantile
transformation is applied, those landmarks approach closely the percentiles
previously defined:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">np</span><span class="o">.</span><span class="n">percentile</span><span class="p">(</span><span class="n">X_train_trans</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">75</span><span class="p">,</span> <span class="mi">100</span><span class="p">])</span>
<span class="gp">... </span>
<span class="go">array([ 0.00... , 0.24..., 0.49..., 0.73..., 0.99... ])</span>
</pre></div>
</div>
<p>This can be confirmed on an independent testing set with similar remarks:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">np</span><span class="o">.</span><span class="n">percentile</span><span class="p">(</span><span class="n">X_test</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">75</span><span class="p">,</span> <span class="mi">100</span><span class="p">])</span>
<span class="gp">... </span>
<span class="go">array([ 4.4 , 5.125, 5.75 , 6.175, 7.3 ])</span>
<span class="gp">>>> </span><span class="n">np</span><span class="o">.</span><span class="n">percentile</span><span class="p">(</span><span class="n">X_test_trans</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">75</span><span class="p">,</span> <span class="mi">100</span><span class="p">])</span>
<span class="gp">... </span>
<span class="go">array([ 0.01..., 0.25..., 0.46..., 0.60... , 0.94...])</span>