-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathfeature_extraction.html
1762 lines (1522 loc) · 143 KB
/
feature_extraction.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="6.2. Feature extraction" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/feature_extraction.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="The sklearn.feature_extraction module can be used to extract features in a format supported by machine learning algorithms from datasets consisting of formats such as text and image. Loading featur..." />
<meta property="og:image" content="https://scikit-learn/stable/_images/sphx_glr_plot_coin_ward_segmentation_001.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="The sklearn.feature_extraction module can be used to extract features in a format supported by machine learning algorithms from datasets consisting of formats such as text and image. Loading featur..." />
<title>6.2. Feature extraction — scikit-learn 1.6.1 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=a746c00c" />
<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=e3ca86de" />
<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=d67e4bb0" />
<!-- 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=d6a008b6"></script>
<script src="../_static/doctools.js?v=9a2dae69"></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=d6bdf5f8"></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/feature_extraction';</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 = '1.6.1';
DOCUMENTATION_OPTIONS.show_version_warning_banner =
true;
</script>
<script src="../_static/scripts/dropdown.js?v=e2048168"></script>
<script src="../_static/scripts/version-switcher.js?v=a6dd8357"></script>
<script src="../_static/scripts/sg_plotly_resize.js?v=eeb41cab"></script>
<link rel="canonical" href="https://scikit-learn.org/stable/modules/feature_extraction.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="6.3. Preprocessing data" href="preprocessing.html" />
<link rel="prev" title="6.1. Pipelines and composite estimators" href="compose.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
<meta name="docsearch:version" content="1.6" />
</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-external" href="https://scikit-learn.org/dev/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-external" href="https://scikit-learn.org/dev/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 current active"><a class="current reference internal" href="#">6.2. Feature extraction</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing.html">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">6.2. </span>Feature extraction</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="feature-extraction">
<span id="id1"></span><h1><span class="section-number">6.2. </span>Feature extraction<a class="headerlink" href="#feature-extraction" title="Link to this heading">#</a></h1>
<p>The <a class="reference internal" href="../api/sklearn.feature_extraction.html#module-sklearn.feature_extraction" title="sklearn.feature_extraction"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.feature_extraction</span></code></a> module can be used to extract
features in a format supported by machine learning algorithms from datasets
consisting of formats such as text and image.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Feature extraction is very different from <a class="reference internal" href="feature_selection.html#feature-selection"><span class="std std-ref">Feature selection</span></a>:
the former consists in transforming arbitrary data, such as text or
images, into numerical features usable for machine learning. The latter
is a machine learning technique applied on these features.</p>
</div>
<section id="loading-features-from-dicts">
<span id="dict-feature-extraction"></span><h2><span class="section-number">6.2.1. </span>Loading features from dicts<a class="headerlink" href="#loading-features-from-dicts" title="Link to this heading">#</a></h2>
<p>The class <a class="reference internal" href="generated/sklearn.feature_extraction.DictVectorizer.html#sklearn.feature_extraction.DictVectorizer" title="sklearn.feature_extraction.DictVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">DictVectorizer</span></code></a> can be used to convert feature
arrays represented as lists of standard Python <code class="docutils literal notranslate"><span class="pre">dict</span></code> objects to the
NumPy/SciPy representation used by scikit-learn estimators.</p>
<p>While not particularly fast to process, Python’s <code class="docutils literal notranslate"><span class="pre">dict</span></code> has the
advantages of being convenient to use, being sparse (absent features
need not be stored) and storing feature names in addition to values.</p>
<p><a class="reference internal" href="generated/sklearn.feature_extraction.DictVectorizer.html#sklearn.feature_extraction.DictVectorizer" title="sklearn.feature_extraction.DictVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">DictVectorizer</span></code></a> implements what is called one-of-K or “one-hot”
coding for categorical (aka nominal, discrete) features. Categorical
features are “attribute-value” pairs where the value is restricted
to a list of discrete possibilities without ordering (e.g. topic
identifiers, types of objects, tags, names…).</p>
<p>In the following, “city” is a categorical attribute while “temperature”
is a traditional numerical feature:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">measurements</span> <span class="o">=</span> <span class="p">[</span>
<span class="gp">... </span> <span class="p">{</span><span class="s1">'city'</span><span class="p">:</span> <span class="s1">'Dubai'</span><span class="p">,</span> <span class="s1">'temperature'</span><span class="p">:</span> <span class="mf">33.</span><span class="p">},</span>
<span class="gp">... </span> <span class="p">{</span><span class="s1">'city'</span><span class="p">:</span> <span class="s1">'London'</span><span class="p">,</span> <span class="s1">'temperature'</span><span class="p">:</span> <span class="mf">12.</span><span class="p">},</span>
<span class="gp">... </span> <span class="p">{</span><span class="s1">'city'</span><span class="p">:</span> <span class="s1">'San Francisco'</span><span class="p">,</span> <span class="s1">'temperature'</span><span class="p">:</span> <span class="mf">18.</span><span class="p">},</span>
<span class="gp">... </span><span class="p">]</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.feature_extraction</span> <span class="kn">import</span> <span class="n">DictVectorizer</span>
<span class="gp">>>> </span><span class="n">vec</span> <span class="o">=</span> <span class="n">DictVectorizer</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">measurements</span><span class="p">)</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="go">array([[ 1., 0., 0., 33.],</span>
<span class="go"> [ 0., 1., 0., 12.],</span>
<span class="go"> [ 0., 0., 1., 18.]])</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">get_feature_names_out</span><span class="p">()</span>
<span class="go">array(['city=Dubai', 'city=London', 'city=San Francisco', 'temperature'], ...)</span>
</pre></div>
</div>
<p><a class="reference internal" href="generated/sklearn.feature_extraction.DictVectorizer.html#sklearn.feature_extraction.DictVectorizer" title="sklearn.feature_extraction.DictVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">DictVectorizer</span></code></a> accepts multiple string values for one
feature, like, e.g., multiple categories for a movie.</p>
<p>Assume a database classifies each movie using some categories (not mandatories)
and its year of release.</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">movie_entry</span> <span class="o">=</span> <span class="p">[{</span><span class="s1">'category'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'thriller'</span><span class="p">,</span> <span class="s1">'drama'</span><span class="p">],</span> <span class="s1">'year'</span><span class="p">:</span> <span class="mi">2003</span><span class="p">},</span>
<span class="gp">... </span> <span class="p">{</span><span class="s1">'category'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'animation'</span><span class="p">,</span> <span class="s1">'family'</span><span class="p">],</span> <span class="s1">'year'</span><span class="p">:</span> <span class="mi">2011</span><span class="p">},</span>
<span class="gp">... </span> <span class="p">{</span><span class="s1">'year'</span><span class="p">:</span> <span class="mi">1974</span><span class="p">}]</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">movie_entry</span><span class="p">)</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="go">array([[0.000e+00, 1.000e+00, 0.000e+00, 1.000e+00, 2.003e+03],</span>
<span class="go"> [1.000e+00, 0.000e+00, 1.000e+00, 0.000e+00, 2.011e+03],</span>
<span class="go"> [0.000e+00, 0.000e+00, 0.000e+00, 0.000e+00, 1.974e+03]])</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">get_feature_names_out</span><span class="p">()</span>
<span class="go">array(['category=animation', 'category=drama', 'category=family',</span>
<span class="go"> 'category=thriller', 'year'], ...)</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">transform</span><span class="p">({</span><span class="s1">'category'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'thriller'</span><span class="p">],</span>
<span class="gp">... </span> <span class="s1">'unseen_feature'</span><span class="p">:</span> <span class="s1">'3'</span><span class="p">})</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="go">array([[0., 0., 0., 1., 0.]])</span>
</pre></div>
</div>
<p><a class="reference internal" href="generated/sklearn.feature_extraction.DictVectorizer.html#sklearn.feature_extraction.DictVectorizer" title="sklearn.feature_extraction.DictVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">DictVectorizer</span></code></a> is also a useful representation transformation
for training sequence classifiers in Natural Language Processing models
that typically work by extracting feature windows around a particular
word of interest.</p>
<p>For example, suppose that we have a first algorithm that extracts Part of
Speech (PoS) tags that we want to use as complementary tags for training
a sequence classifier (e.g. a chunker). The following dict could be
such a window of features extracted around the word ‘sat’ in the sentence
‘The cat sat on the mat.’:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">pos_window</span> <span class="o">=</span> <span class="p">[</span>
<span class="gp">... </span> <span class="p">{</span>
<span class="gp">... </span> <span class="s1">'word-2'</span><span class="p">:</span> <span class="s1">'the'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'pos-2'</span><span class="p">:</span> <span class="s1">'DT'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'word-1'</span><span class="p">:</span> <span class="s1">'cat'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'pos-1'</span><span class="p">:</span> <span class="s1">'NN'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'word+1'</span><span class="p">:</span> <span class="s1">'on'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'pos+1'</span><span class="p">:</span> <span class="s1">'PP'</span><span class="p">,</span>
<span class="gp">... </span> <span class="p">},</span>
<span class="gp">... </span> <span class="c1"># in a real application one would extract many such dictionaries</span>
<span class="gp">... </span><span class="p">]</span>
</pre></div>
</div>
<p>This description can be vectorized into a sparse two-dimensional matrix
suitable for feeding into a classifier (maybe after being piped into a
<a class="reference internal" href="generated/sklearn.feature_extraction.text.TfidfTransformer.html#sklearn.feature_extraction.text.TfidfTransformer" title="sklearn.feature_extraction.text.TfidfTransformer"><code class="xref py py-class docutils literal notranslate"><span class="pre">TfidfTransformer</span></code></a> for normalization):</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">vec</span> <span class="o">=</span> <span class="n">DictVectorizer</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">pos_vectorized</span> <span class="o">=</span> <span class="n">vec</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">pos_window</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">pos_vectorized</span>
<span class="go"><Compressed Sparse...dtype 'float64'</span>
<span class="go"> with 6 stored elements and shape (1, 6)></span>
<span class="gp">>>> </span><span class="n">pos_vectorized</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="go">array([[1., 1., 1., 1., 1., 1.]])</span>
<span class="gp">>>> </span><span class="n">vec</span><span class="o">.</span><span class="n">get_feature_names_out</span><span class="p">()</span>
<span class="go">array(['pos+1=PP', 'pos-1=NN', 'pos-2=DT', 'word+1=on', 'word-1=cat',</span>
<span class="go"> 'word-2=the'], ...)</span>
</pre></div>
</div>
<p>As you can imagine, if one extracts such a context around each individual
word of a corpus of documents the resulting matrix will be very wide
(many one-hot-features) with most of them being valued to zero most
of the time. So as to make the resulting data structure able to fit in
memory the <code class="docutils literal notranslate"><span class="pre">DictVectorizer</span></code> class uses a <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code> matrix by
default instead of a <code class="docutils literal notranslate"><span class="pre">numpy.ndarray</span></code>.</p>
</section>
<section id="feature-hashing">
<span id="id2"></span><h2><span class="section-number">6.2.2. </span>Feature hashing<a class="headerlink" href="#feature-hashing" title="Link to this heading">#</a></h2>
<p>The class <a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a> is a high-speed, low-memory vectorizer that
uses a technique known as
<a class="reference external" href="https://en.wikipedia.org/wiki/Feature_hashing">feature hashing</a>,
or the “hashing trick”.
Instead of building a hash table of the features encountered in training,
as the vectorizers do, instances of <a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a>
apply a hash function to the features
to determine their column index in sample matrices directly.
The result is increased speed and reduced memory usage,
at the expense of inspectability;
the hasher does not remember what the input features looked like
and has no <code class="docutils literal notranslate"><span class="pre">inverse_transform</span></code> method.</p>
<p>Since the hash function might cause collisions between (unrelated) features,
a signed hash function is used and the sign of the hash value
determines the sign of the value stored in the output matrix for a feature.
This way, collisions are likely to cancel out rather than accumulate error,
and the expected mean of any output feature’s value is zero. This mechanism
is enabled by default with <code class="docutils literal notranslate"><span class="pre">alternate_sign=True</span></code> and is particularly useful
for small hash table sizes (<code class="docutils literal notranslate"><span class="pre">n_features</span> <span class="pre"><</span> <span class="pre">10000</span></code>). For large hash table
sizes, it can be disabled, to allow the output to be passed to estimators like
<a class="reference internal" href="generated/sklearn.naive_bayes.MultinomialNB.html#sklearn.naive_bayes.MultinomialNB" title="sklearn.naive_bayes.MultinomialNB"><code class="xref py py-class docutils literal notranslate"><span class="pre">MultinomialNB</span></code></a> or
<a class="reference internal" href="generated/sklearn.feature_selection.chi2.html#sklearn.feature_selection.chi2" title="sklearn.feature_selection.chi2"><code class="xref py py-class docutils literal notranslate"><span class="pre">chi2</span></code></a>
feature selectors that expect non-negative inputs.</p>
<p><a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a> accepts either mappings
(like Python’s <code class="docutils literal notranslate"><span class="pre">dict</span></code> and its variants in the <code class="docutils literal notranslate"><span class="pre">collections</span></code> module),
<code class="docutils literal notranslate"><span class="pre">(feature,</span> <span class="pre">value)</span></code> pairs, or strings,
depending on the constructor parameter <code class="docutils literal notranslate"><span class="pre">input_type</span></code>.
Mapping are treated as lists of <code class="docutils literal notranslate"><span class="pre">(feature,</span> <span class="pre">value)</span></code> pairs,
while single strings have an implicit value of 1,
so <code class="docutils literal notranslate"><span class="pre">['feat1',</span> <span class="pre">'feat2',</span> <span class="pre">'feat3']</span></code> is interpreted as
<code class="docutils literal notranslate"><span class="pre">[('feat1',</span> <span class="pre">1),</span> <span class="pre">('feat2',</span> <span class="pre">1),</span> <span class="pre">('feat3',</span> <span class="pre">1)]</span></code>.
If a single feature occurs multiple times in a sample,
the associated values will be summed
(so <code class="docutils literal notranslate"><span class="pre">('feat',</span> <span class="pre">2)</span></code> and <code class="docutils literal notranslate"><span class="pre">('feat',</span> <span class="pre">3.5)</span></code> become <code class="docutils literal notranslate"><span class="pre">('feat',</span> <span class="pre">5.5)</span></code>).
The output from <a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a> is always a <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code> matrix
in the CSR format.</p>
<p>Feature hashing can be employed in document classification,
but unlike <a class="reference internal" href="generated/sklearn.feature_extraction.text.CountVectorizer.html#sklearn.feature_extraction.text.CountVectorizer" title="sklearn.feature_extraction.text.CountVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">CountVectorizer</span></code></a>,
<a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a> does not do word
splitting or any other preprocessing except Unicode-to-UTF-8 encoding;
see <a class="reference internal" href="#hashing-vectorizer"><span class="std std-ref">Vectorizing a large text corpus with the hashing trick</span></a>, below, for a combined tokenizer/hasher.</p>
<p>As an example, consider a word-level natural language processing task
that needs features extracted from <code class="docutils literal notranslate"><span class="pre">(token,</span> <span class="pre">part_of_speech)</span></code> pairs.
One could use a Python generator function to extract features:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">token_features</span><span class="p">(</span><span class="n">token</span><span class="p">,</span> <span class="n">part_of_speech</span><span class="p">):</span>
<span class="k">if</span> <span class="n">token</span><span class="o">.</span><span class="n">isdigit</span><span class="p">():</span>
<span class="k">yield</span> <span class="s2">"numeric"</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">yield</span> <span class="s2">"token=</span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">token</span><span class="o">.</span><span class="n">lower</span><span class="p">())</span>
<span class="k">yield</span> <span class="s2">"token,pos=</span><span class="si">{}</span><span class="s2">,</span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">token</span><span class="p">,</span> <span class="n">part_of_speech</span><span class="p">)</span>
<span class="k">if</span> <span class="n">token</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">isupper</span><span class="p">():</span>
<span class="k">yield</span> <span class="s2">"uppercase_initial"</span>
<span class="k">if</span> <span class="n">token</span><span class="o">.</span><span class="n">isupper</span><span class="p">():</span>
<span class="k">yield</span> <span class="s2">"all_uppercase"</span>
<span class="k">yield</span> <span class="s2">"pos=</span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">part_of_speech</span><span class="p">)</span>
</pre></div>
</div>
<p>Then, the <code class="docutils literal notranslate"><span class="pre">raw_X</span></code> to be fed to <code class="docutils literal notranslate"><span class="pre">FeatureHasher.transform</span></code>
can be constructed using:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">raw_X</span> <span class="o">=</span> <span class="p">(</span><span class="n">token_features</span><span class="p">(</span><span class="n">tok</span><span class="p">,</span> <span class="n">pos_tagger</span><span class="p">(</span><span class="n">tok</span><span class="p">))</span> <span class="k">for</span> <span class="n">tok</span> <span class="ow">in</span> <span class="n">corpus</span><span class="p">)</span>
</pre></div>
</div>
<p>and fed to a hasher with:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">hasher</span> <span class="o">=</span> <span class="n">FeatureHasher</span><span class="p">(</span><span class="n">input_type</span><span class="o">=</span><span class="s1">'string'</span><span class="p">)</span>
<span class="n">X</span> <span class="o">=</span> <span class="n">hasher</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">raw_X</span><span class="p">)</span>
</pre></div>
</div>
<p>to get a <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code> matrix <code class="docutils literal notranslate"><span class="pre">X</span></code>.</p>
<p>Note the use of a generator comprehension,
which introduces laziness into the feature extraction:
tokens are only processed on demand from the hasher.</p>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="implementation-details">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Implementation details<a class="headerlink" href="#implementation-details" 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"><a class="reference internal" href="generated/sklearn.feature_extraction.FeatureHasher.html#sklearn.feature_extraction.FeatureHasher" title="sklearn.feature_extraction.FeatureHasher"><code class="xref py py-class docutils literal notranslate"><span class="pre">FeatureHasher</span></code></a> uses the signed 32-bit variant of MurmurHash3.
As a result (and because of limitations in <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code>),
the maximum number of features supported is currently <span class="math notranslate nohighlight">\(2^{31} - 1\)</span>.</p>
<p class="sd-card-text">The original formulation of the hashing trick by Weinberger et al.
used two separate hash functions <span class="math notranslate nohighlight">\(h\)</span> and <span class="math notranslate nohighlight">\(\xi\)</span>
to determine the column index and sign of a feature, respectively.
The present implementation works under the assumption
that the sign bit of MurmurHash3 is independent of its other bits.</p>
<p class="sd-card-text">Since a simple modulo is used to transform the hash function to a column index,
it is advisable to use a power of two as the <code class="docutils literal notranslate"><span class="pre">n_features</span></code> parameter;
otherwise the features will not be mapped evenly to the columns.</p>
<p class="rubric">References</p>
<ul class="simple">
<li><p class="sd-card-text"><a class="reference external" href="https://github.com/aappleby/smhasher">MurmurHash3</a>.</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p>Kilian Weinberger, Anirban Dasgupta, John Langford, Alex Smola and
Josh Attenberg (2009). <a class="reference external" href="https://alex.smola.org/papers/2009/Weinbergeretal09.pdf">Feature hashing for large scale multitask learning</a>. Proc. ICML.</p></li>
</ul>
</section>
<section id="text-feature-extraction">
<span id="id4"></span><h2><span class="section-number">6.2.3. </span>Text feature extraction<a class="headerlink" href="#text-feature-extraction" title="Link to this heading">#</a></h2>
<section id="the-bag-of-words-representation">
<h3><span class="section-number">6.2.3.1. </span>The Bag of Words representation<a class="headerlink" href="#the-bag-of-words-representation" title="Link to this heading">#</a></h3>
<p>Text Analysis is a major application field for machine learning
algorithms. However the raw data, a sequence of symbols cannot be fed
directly to the algorithms themselves as most of them expect numerical
feature vectors with a fixed size rather than the raw text documents
with variable length.</p>
<p>In order to address this, scikit-learn provides utilities for the most
common ways to extract numerical features from text content, namely:</p>
<ul class="simple">
<li><p><strong>tokenizing</strong> strings and giving an integer id for each possible token,
for instance by using white-spaces and punctuation as token separators.</p></li>
<li><p><strong>counting</strong> the occurrences of tokens in each document.</p></li>
<li><p><strong>normalizing</strong> and weighting with diminishing importance tokens that
occur in the majority of samples / documents.</p></li>
</ul>
<p>In this scheme, features and samples are defined as follows:</p>
<ul class="simple">
<li><p>each <strong>individual token occurrence frequency</strong> (normalized or not)
is treated as a <strong>feature</strong>.</p></li>
<li><p>the vector of all the token frequencies for a given <strong>document</strong> is
considered a multivariate <strong>sample</strong>.</p></li>
</ul>
<p>A corpus of documents can thus be represented by a matrix with one row
per document and one column per token (e.g. word) occurring in the corpus.</p>
<p>We call <strong>vectorization</strong> the general process of turning a collection
of text documents into numerical feature vectors. This specific strategy
(tokenization, counting and normalization) is called the <strong>Bag of Words</strong>
or “Bag of n-grams” representation. Documents are described by word
occurrences while completely ignoring the relative position information
of the words in the document.</p>
</section>
<section id="sparsity">
<h3><span class="section-number">6.2.3.2. </span>Sparsity<a class="headerlink" href="#sparsity" title="Link to this heading">#</a></h3>
<p>As most documents will typically use a very small subset of the words used in
the corpus, the resulting matrix will have many feature values that are
zeros (typically more than 99% of them).</p>
<p>For instance a collection of 10,000 short text documents (such as emails)
will use a vocabulary with a size in the order of 100,000 unique words in
total while each document will use 100 to 1000 unique words individually.</p>
<p>In order to be able to store such a matrix in memory but also to speed
up algebraic operations matrix / vector, implementations will typically
use a sparse representation such as the implementations available in the
<code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code> package.</p>
</section>
<section id="common-vectorizer-usage">
<h3><span class="section-number">6.2.3.3. </span>Common Vectorizer usage<a class="headerlink" href="#common-vectorizer-usage" title="Link to this heading">#</a></h3>
<p><a class="reference internal" href="generated/sklearn.feature_extraction.text.CountVectorizer.html#sklearn.feature_extraction.text.CountVectorizer" title="sklearn.feature_extraction.text.CountVectorizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">CountVectorizer</span></code></a> implements both tokenization and occurrence
counting in a single class:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.feature_extraction.text</span> <span class="kn">import</span> <span class="n">CountVectorizer</span>
</pre></div>
</div>
<p>This model has many parameters, however the default values are quite
reasonable (please see the <a class="reference internal" href="../api/sklearn.feature_extraction.html#feature-extraction-ref-from-text"><span class="std std-ref">reference documentation</span></a> for the details):</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">vectorizer</span> <span class="o">=</span> <span class="n">CountVectorizer</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">vectorizer</span>
<span class="go">CountVectorizer()</span>
</pre></div>
</div>
<p>Let’s use it to tokenize and count the word occurrences of a minimalistic
corpus of text documents:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">corpus</span> <span class="o">=</span> <span class="p">[</span>
<span class="gp">... </span> <span class="s1">'This is the first document.'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'This is the second second document.'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'And the third one.'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s1">'Is this the first document?'</span><span class="p">,</span>
<span class="gp">... </span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">corpus</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X</span>
<span class="go"><Compressed Sparse...dtype 'int64'</span>
<span class="go"> with 19 stored elements and shape (4, 9)></span>
</pre></div>
</div>
<p>The default configuration tokenizes the string by extracting words of
at least 2 letters. The specific function that does this step can be
requested explicitly:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">analyze</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">build_analyzer</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">analyze</span><span class="p">(</span><span class="s2">"This is a text document to analyze."</span><span class="p">)</span> <span class="o">==</span> <span class="p">(</span>
<span class="gp">... </span> <span class="p">[</span><span class="s1">'this'</span><span class="p">,</span> <span class="s1">'is'</span><span class="p">,</span> <span class="s1">'text'</span><span class="p">,</span> <span class="s1">'document'</span><span class="p">,</span> <span class="s1">'to'</span><span class="p">,</span> <span class="s1">'analyze'</span><span class="p">])</span>
<span class="go">True</span>
</pre></div>
</div>
<p>Each term found by the analyzer during the fit is assigned a unique
integer index corresponding to a column in the resulting matrix. This
interpretation of the columns can be retrieved as follows:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">vectorizer</span><span class="o">.</span><span class="n">get_feature_names_out</span><span class="p">()</span>
<span class="go">array(['and', 'document', 'first', 'is', 'one', 'second', 'the',</span>
<span class="go"> 'third', 'this'], ...)</span>
<span class="gp">>>> </span><span class="n">X</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="go">array([[0, 1, 1, 1, 0, 0, 1, 0, 1],</span>
<span class="go"> [0, 1, 0, 1, 0, 2, 1, 0, 1],</span>
<span class="go"> [1, 0, 0, 0, 1, 0, 1, 1, 0],</span>
<span class="go"> [0, 1, 1, 1, 0, 0, 1, 0, 1]]...)</span>
</pre></div>
</div>
<p>The converse mapping from feature name to column index is stored in the
<code class="docutils literal notranslate"><span class="pre">vocabulary_</span></code> attribute of the vectorizer:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">vectorizer</span><span class="o">.</span><span class="n">vocabulary_</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'document'</span><span class="p">)</span>
<span class="go">1</span>
</pre></div>