-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathmanifold.html
1377 lines (1147 loc) · 85.4 KB
/
manifold.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="2.2. Manifold learning" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/manifold.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="Look for the bare necessities, The simple bare necessities, Forget about your worries and your strife, I mean the bare necessities, Old Mother Nature’s recipes, That bring the bare necessities of l..." />
<meta property="og:image" content="https://scikit-learn/stable/_images/sphx_glr_plot_compare_methods_001.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="Look for the bare necessities, The simple bare necessities, Forget about your worries and your strife, I mean the bare necessities, Old Mother Nature’s recipes, That bring the bare necessities of l..." />
<title>2.2. Manifold learning — 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/manifold';</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/manifold.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="2.3. Clustering" href="clustering.html" />
<link rel="prev" title="2.1. Gaussian mixture models" href="mixture.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 current active has-children"><a class="reference internal" href="../unsupervised_learning.html">2. Unsupervised learning</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="mixture.html">2.1. Gaussian mixture models</a></li>
<li class="toctree-l2 current active"><a class="current reference internal" href="#">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"><a class="reference internal" href="../metadata_routing.html">4. Metadata Routing</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../inspection.html">5. 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">5.1. Partial Dependence and Individual Conditional Expectation plots</a></li>
<li class="toctree-l2"><a class="reference internal" href="permutation_importance.html">5.2. Permutation feature importance</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../visualizations.html">6. Visualizations</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../data_transforms.html">7. Dataset transformations</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="compose.html">7.1. Pipelines and composite estimators</a></li>
<li class="toctree-l2"><a class="reference internal" href="feature_extraction.html">7.2. Feature extraction</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing.html">7.3. Preprocessing data</a></li>
<li class="toctree-l2"><a class="reference internal" href="impute.html">7.4. Imputation of missing values</a></li>
<li class="toctree-l2"><a class="reference internal" href="unsupervised_reduction.html">7.5. Unsupervised dimensionality reduction</a></li>
<li class="toctree-l2"><a class="reference internal" href="random_projection.html">7.6. Random Projection</a></li>
<li class="toctree-l2"><a class="reference internal" href="kernel_approximation.html">7.7. Kernel Approximation</a></li>
<li class="toctree-l2"><a class="reference internal" href="metrics.html">7.8. Pairwise metrics, Affinities and Kernels</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing_targets.html">7.9. 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">8. 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">8.1. Toy datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/real_world.html">8.2. Real world datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/sample_generators.html">8.3. Generated datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/loading_other_datasets.html">8.4. Loading other datasets</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../computing.html">9. 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">9.1. Strategies to scale computationally: bigger data</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/computational_performance.html">9.2. Computational Performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/parallelism.html">9.3. Parallelism, resource management, and configuration</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../model_persistence.html">10. Model persistence</a></li>
<li class="toctree-l1"><a class="reference internal" href="../common_pitfalls.html">11. Common pitfalls and recommended practices</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../dispatching.html">12. 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">12.1. Array API support (experimental)</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../machine_learning_map.html">13. Choosing the right estimator</a></li>
<li class="toctree-l1"><a class="reference internal" href="../presentations.html">14. 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="../unsupervised_learning.html" class="nav-link"><span class="section-number">2. </span>Unsupervised learning</a></li>
<li class="breadcrumb-item active" aria-current="page"><span class="ellipsis"><span class="section-number">2.2. </span>Manifold learning</span></li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="manifold-learning">
<span id="manifold"></span><h1><span class="section-number">2.2. </span>Manifold learning<a class="headerlink" href="#manifold-learning" title="Link to this heading">#</a></h1>
<div class="line-block">
<div class="line">Look for the bare necessities</div>
<div class="line">The simple bare necessities</div>
<div class="line">Forget about your worries and your strife</div>
<div class="line">I mean the bare necessities</div>
<div class="line">Old Mother Nature’s recipes</div>
<div class="line">That bring the bare necessities of life</div>
<div class="line"><br /></div>
<div class="line-block">
<div class="line">– Baloo’s song [The Jungle Book]</div>
</div>
</div>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_compare_methods.html"><img alt="../_images/sphx_glr_plot_compare_methods_001.png" src="../_images/sphx_glr_plot_compare_methods_001.png" style="width: 420.0px; height: 420.0px;" />
</a>
</figure>
<p class="centered">
<strong><a class="reference external" href="../auto_examples/manifold/plot_compare_methods.html"><img alt="manifold_img3" src="../_images/sphx_glr_plot_compare_methods_003.png" style="width: 180.0px; height: 180.0px;" /></a> <a class="reference external" href="../auto_examples/manifold/plot_compare_methods.html"><img alt="manifold_img4" src="../_images/sphx_glr_plot_compare_methods_004.png" style="width: 180.0px; height: 180.0px;" /></a> <a class="reference external" href="../auto_examples/manifold/plot_compare_methods.html"><img alt="manifold_img5" src="../_images/sphx_glr_plot_compare_methods_005.png" style="width: 180.0px; height: 180.0px;" /></a> <a class="reference external" href="../auto_examples/manifold/plot_compare_methods.html"><img alt="manifold_img6" src="../_images/sphx_glr_plot_compare_methods_006.png" style="width: 180.0px; height: 180.0px;" /></a></strong></p><p>Manifold learning is an approach to non-linear dimensionality reduction.
Algorithms for this task are based on the idea that the dimensionality of
many data sets is only artificially high.</p>
<section id="introduction">
<h2><span class="section-number">2.2.1. </span>Introduction<a class="headerlink" href="#introduction" title="Link to this heading">#</a></h2>
<p>High-dimensional datasets can be very difficult to visualize. While data
in two or three dimensions can be plotted to show the inherent
structure of the data, equivalent high-dimensional plots are much less
intuitive. To aid visualization of the structure of a dataset, the
dimension must be reduced in some way.</p>
<p>The simplest way to accomplish this dimensionality reduction is by taking
a random projection of the data. Though this allows some degree of
visualization of the data structure, the randomness of the choice leaves much
to be desired. In a random projection, it is likely that the more
interesting structure within the data will be lost.</p>
<p class="centered">
<strong><a class="reference external" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="digits_img" src="../_images/sphx_glr_plot_lle_digits_001.png" style="width: 300.0px; height: 300.0px;" /></a> <a class="reference external" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="projected_img" src="../_images/sphx_glr_plot_lle_digits_002.png" style="width: 320.0px; height: 240.0px;" /></a></strong></p><p>To address this concern, a number of supervised and unsupervised linear
dimensionality reduction frameworks have been designed, such as Principal
Component Analysis (PCA), Independent Component Analysis, Linear
Discriminant Analysis, and others. These algorithms define specific
rubrics to choose an “interesting” linear projection of the data.
These methods can be powerful, but often miss important non-linear
structure in the data.</p>
<p class="centered">
<strong><a class="reference external" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="PCA_img" src="../_images/sphx_glr_plot_lle_digits_003.png" style="width: 320.0px; height: 240.0px;" /></a> <a class="reference external" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="LDA_img" src="../_images/sphx_glr_plot_lle_digits_004.png" style="width: 320.0px; height: 240.0px;" /></a></strong></p><p>Manifold Learning can be thought of as an attempt to generalize linear
frameworks like PCA to be sensitive to non-linear structure in data. Though
supervised variants exist, the typical manifold learning problem is
unsupervised: it learns the high-dimensional structure of the data
from the data itself, without the use of predetermined classifications.</p>
<p class="rubric">Examples</p>
<ul class="simple">
<li><p>See <a class="reference internal" href="../auto_examples/manifold/plot_lle_digits.html#sphx-glr-auto-examples-manifold-plot-lle-digits-py"><span class="std std-ref">Manifold learning on handwritten digits: Locally Linear Embedding, Isomap…</span></a> for an example of
dimensionality reduction on handwritten digits.</p></li>
<li><p>See <a class="reference internal" href="../auto_examples/manifold/plot_compare_methods.html#sphx-glr-auto-examples-manifold-plot-compare-methods-py"><span class="std std-ref">Comparison of Manifold Learning methods</span></a> for an example of
dimensionality reduction on a toy “S-curve” dataset.</p></li>
<li><p>See <a class="reference internal" href="../auto_examples/applications/plot_stock_market.html#sphx-glr-auto-examples-applications-plot-stock-market-py"><span class="std std-ref">Visualizing the stock market structure</span></a> for an example of
using manifold learning to map the stock market structure based on historical stock
prices.</p></li>
</ul>
<p>The manifold learning implementations available in scikit-learn are
summarized below</p>
</section>
<section id="isomap">
<span id="id1"></span><h2><span class="section-number">2.2.2. </span>Isomap<a class="headerlink" href="#isomap" title="Link to this heading">#</a></h2>
<p>One of the earliest approaches to manifold learning is the Isomap
algorithm, short for Isometric Mapping. Isomap can be viewed as an
extension of Multi-dimensional Scaling (MDS) or Kernel PCA.
Isomap seeks a lower-dimensional embedding which maintains geodesic
distances between all points. Isomap can be performed with the object
<a class="reference internal" href="generated/sklearn.manifold.Isomap.html#sklearn.manifold.Isomap" title="sklearn.manifold.Isomap"><code class="xref py py-class docutils literal notranslate"><span class="pre">Isomap</span></code></a>.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="../_images/sphx_glr_plot_lle_digits_005.png" src="../_images/sphx_glr_plot_lle_digits_005.png" style="width: 320.0px; height: 240.0px;" />
</a>
</figure>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity" 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">The Isomap algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Nearest neighbor search.</strong> Isomap uses
<a class="reference internal" href="generated/sklearn.neighbors.BallTree.html#sklearn.neighbors.BallTree" title="sklearn.neighbors.BallTree"><code class="xref py py-class docutils literal notranslate"><span class="pre">BallTree</span></code></a> for efficient neighbor search.
The cost is approximately <span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)]\)</span>, for <span class="math notranslate nohighlight">\(k\)</span>
nearest neighbors of <span class="math notranslate nohighlight">\(N\)</span> points in <span class="math notranslate nohighlight">\(D\)</span> dimensions.</p></li>
<li><p class="sd-card-text"><strong>Shortest-path graph search.</strong> The most efficient known algorithms
for this are <em>Dijkstra’s Algorithm</em>, which is approximately
<span class="math notranslate nohighlight">\(O[N^2(k + \log(N))]\)</span>, or the <em>Floyd-Warshall algorithm</em>, which
is <span class="math notranslate nohighlight">\(O[N^3]\)</span>. The algorithm can be selected by the user with
the <code class="docutils literal notranslate"><span class="pre">path_method</span></code> keyword of <code class="docutils literal notranslate"><span class="pre">Isomap</span></code>. If unspecified, the code
attempts to choose the best algorithm for the input data.</p></li>
<li><p class="sd-card-text"><strong>Partial eigenvalue decomposition.</strong> The embedding is encoded in the
eigenvectors corresponding to the <span class="math notranslate nohighlight">\(d\)</span> largest eigenvalues of the
<span class="math notranslate nohighlight">\(N \times N\)</span> isomap kernel. For a dense solver, the cost is
approximately <span class="math notranslate nohighlight">\(O[d N^2]\)</span>. This cost can often be improved using
the <code class="docutils literal notranslate"><span class="pre">ARPACK</span></code> solver. The eigensolver can be specified by the user
with the <code class="docutils literal notranslate"><span class="pre">eigen_solver</span></code> keyword of <code class="docutils literal notranslate"><span class="pre">Isomap</span></code>. If unspecified, the
code attempts to choose the best algorithm for the input data.</p></li>
</ol>
<p class="sd-card-text">The overall complexity of Isomap is
<span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)] + O[N^2(k + \log(N))] + O[d N^2]\)</span>.</p>
<ul class="simple">
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(N\)</span> : number of training data points</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(D\)</span> : input dimension</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(k\)</span> : number of nearest neighbors</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(d\)</span> : output dimension</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p><a class="reference external" href="http://science.sciencemag.org/content/290/5500/2319.full">“A global geometric framework for nonlinear dimensionality reduction”</a>
Tenenbaum, J.B.; De Silva, V.; & Langford, J.C. Science 290 (5500)</p></li>
</ul>
</section>
<section id="locally-linear-embedding">
<span id="id2"></span><h2><span class="section-number">2.2.3. </span>Locally Linear Embedding<a class="headerlink" href="#locally-linear-embedding" title="Link to this heading">#</a></h2>
<p>Locally linear embedding (LLE) seeks a lower-dimensional projection of the data
which preserves distances within local neighborhoods. It can be thought
of as a series of local Principal Component Analyses which are globally
compared to find the best non-linear embedding.</p>
<p>Locally linear embedding can be performed with function
<a class="reference internal" href="generated/sklearn.manifold.locally_linear_embedding.html#sklearn.manifold.locally_linear_embedding" title="sklearn.manifold.locally_linear_embedding"><code class="xref py py-func docutils literal notranslate"><span class="pre">locally_linear_embedding</span></code></a> or its object-oriented counterpart
<a class="reference internal" href="generated/sklearn.manifold.LocallyLinearEmbedding.html#sklearn.manifold.LocallyLinearEmbedding" title="sklearn.manifold.LocallyLinearEmbedding"><code class="xref py py-class docutils literal notranslate"><span class="pre">LocallyLinearEmbedding</span></code></a>.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="../_images/sphx_glr_plot_lle_digits_006.png" src="../_images/sphx_glr_plot_lle_digits_006.png" style="width: 320.0px; height: 240.0px;" />
</a>
</figure>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity-2">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity-2" 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">The standard LLE algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Nearest Neighbors Search</strong>. See discussion under Isomap above.</p></li>
<li><p class="sd-card-text"><strong>Weight Matrix Construction</strong>. <span class="math notranslate nohighlight">\(O[D N k^3]\)</span>.
The construction of the LLE weight matrix involves the solution of a
<span class="math notranslate nohighlight">\(k \times k\)</span> linear equation for each of the <span class="math notranslate nohighlight">\(N\)</span> local
neighborhoods.</p></li>
<li><p class="sd-card-text"><strong>Partial Eigenvalue Decomposition</strong>. See discussion under Isomap above.</p></li>
</ol>
<p class="sd-card-text">The overall complexity of standard LLE is
<span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)] + O[D N k^3] + O[d N^2]\)</span>.</p>
<ul class="simple">
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(N\)</span> : number of training data points</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(D\)</span> : input dimension</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(k\)</span> : number of nearest neighbors</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(d\)</span> : output dimension</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p><a class="reference external" href="http://www.sciencemag.org/content/290/5500/2323.full">“Nonlinear dimensionality reduction by locally linear embedding”</a>
Roweis, S. & Saul, L. Science 290:2323 (2000)</p></li>
</ul>
</section>
<section id="modified-locally-linear-embedding">
<h2><span class="section-number">2.2.4. </span>Modified Locally Linear Embedding<a class="headerlink" href="#modified-locally-linear-embedding" title="Link to this heading">#</a></h2>
<p>One well-known issue with LLE is the regularization problem. When the number
of neighbors is greater than the number of input dimensions, the matrix
defining each local neighborhood is rank-deficient. To address this, standard
LLE applies an arbitrary regularization parameter <span class="math notranslate nohighlight">\(r\)</span>, which is chosen
relative to the trace of the local weight matrix. Though it can be shown
formally that as <span class="math notranslate nohighlight">\(r \to 0\)</span>, the solution converges to the desired
embedding, there is no guarantee that the optimal solution will be found
for <span class="math notranslate nohighlight">\(r > 0\)</span>. This problem manifests itself in embeddings which distort
the underlying geometry of the manifold.</p>
<p>One method to address the regularization problem is to use multiple weight
vectors in each neighborhood. This is the essence of <em>modified locally
linear embedding</em> (MLLE). MLLE can be performed with function
<a class="reference internal" href="generated/sklearn.manifold.locally_linear_embedding.html#sklearn.manifold.locally_linear_embedding" title="sklearn.manifold.locally_linear_embedding"><code class="xref py py-func docutils literal notranslate"><span class="pre">locally_linear_embedding</span></code></a> or its object-oriented counterpart
<a class="reference internal" href="generated/sklearn.manifold.LocallyLinearEmbedding.html#sklearn.manifold.LocallyLinearEmbedding" title="sklearn.manifold.LocallyLinearEmbedding"><code class="xref py py-class docutils literal notranslate"><span class="pre">LocallyLinearEmbedding</span></code></a>, with the keyword <code class="docutils literal notranslate"><span class="pre">method</span> <span class="pre">=</span> <span class="pre">'modified'</span></code>.
It requires <code class="docutils literal notranslate"><span class="pre">n_neighbors</span> <span class="pre">></span> <span class="pre">n_components</span></code>.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="../_images/sphx_glr_plot_lle_digits_007.png" src="../_images/sphx_glr_plot_lle_digits_007.png" style="width: 320.0px; height: 240.0px;" />
</a>
</figure>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity-3">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity-3" 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">The MLLE algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Nearest Neighbors Search</strong>. Same as standard LLE</p></li>
<li><p class="sd-card-text"><strong>Weight Matrix Construction</strong>. Approximately
<span class="math notranslate nohighlight">\(O[D N k^3] + O[N (k-D) k^2]\)</span>. The first term is exactly equivalent
to that of standard LLE. The second term has to do with constructing the
weight matrix from multiple weights. In practice, the added cost of
constructing the MLLE weight matrix is relatively small compared to the
cost of stages 1 and 3.</p></li>
<li><p class="sd-card-text"><strong>Partial Eigenvalue Decomposition</strong>. Same as standard LLE</p></li>
</ol>
<p class="sd-card-text">The overall complexity of MLLE is
<span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)] + O[D N k^3] + O[N (k-D) k^2] + O[d N^2]\)</span>.</p>
<ul class="simple">
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(N\)</span> : number of training data points</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(D\)</span> : input dimension</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(k\)</span> : number of nearest neighbors</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(d\)</span> : output dimension</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p><a class="reference external" href="https://citeseerx.ist.psu.edu/doc_view/pid/0b060fdbd92cbcc66b383bcaa9ba5e5e624d7ee3">“MLLE: Modified Locally Linear Embedding Using Multiple Weights”</a>
Zhang, Z. & Wang, J.</p></li>
</ul>
</section>
<section id="hessian-eigenmapping">
<h2><span class="section-number">2.2.5. </span>Hessian Eigenmapping<a class="headerlink" href="#hessian-eigenmapping" title="Link to this heading">#</a></h2>
<p>Hessian Eigenmapping (also known as Hessian-based LLE: HLLE) is another method
of solving the regularization problem of LLE. It revolves around a
hessian-based quadratic form at each neighborhood which is used to recover
the locally linear structure. Though other implementations note its poor
scaling with data size, <code class="docutils literal notranslate"><span class="pre">sklearn</span></code> implements some algorithmic
improvements which make its cost comparable to that of other LLE variants
for small output dimension. HLLE can be performed with function
<a class="reference internal" href="generated/sklearn.manifold.locally_linear_embedding.html#sklearn.manifold.locally_linear_embedding" title="sklearn.manifold.locally_linear_embedding"><code class="xref py py-func docutils literal notranslate"><span class="pre">locally_linear_embedding</span></code></a> or its object-oriented counterpart
<a class="reference internal" href="generated/sklearn.manifold.LocallyLinearEmbedding.html#sklearn.manifold.LocallyLinearEmbedding" title="sklearn.manifold.LocallyLinearEmbedding"><code class="xref py py-class docutils literal notranslate"><span class="pre">LocallyLinearEmbedding</span></code></a>, with the keyword <code class="docutils literal notranslate"><span class="pre">method</span> <span class="pre">=</span> <span class="pre">'hessian'</span></code>.
It requires <code class="docutils literal notranslate"><span class="pre">n_neighbors</span> <span class="pre">></span> <span class="pre">n_components</span> <span class="pre">*</span> <span class="pre">(n_components</span> <span class="pre">+</span> <span class="pre">3)</span> <span class="pre">/</span> <span class="pre">2</span></code>.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="../_images/sphx_glr_plot_lle_digits_008.png" src="../_images/sphx_glr_plot_lle_digits_008.png" style="width: 320.0px; height: 240.0px;" />
</a>
</figure>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity-4">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity-4" 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">The HLLE algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Nearest Neighbors Search</strong>. Same as standard LLE</p></li>
<li><p class="sd-card-text"><strong>Weight Matrix Construction</strong>. Approximately
<span class="math notranslate nohighlight">\(O[D N k^3] + O[N d^6]\)</span>. The first term reflects a similar
cost to that of standard LLE. The second term comes from a QR
decomposition of the local hessian estimator.</p></li>
<li><p class="sd-card-text"><strong>Partial Eigenvalue Decomposition</strong>. Same as standard LLE.</p></li>
</ol>
<p class="sd-card-text">The overall complexity of standard HLLE is
<span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)] + O[D N k^3] + O[N d^6] + O[d N^2]\)</span>.</p>
<ul class="simple">
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(N\)</span> : number of training data points</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(D\)</span> : input dimension</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(k\)</span> : number of nearest neighbors</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(d\)</span> : output dimension</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p><a class="reference external" href="http://www.pnas.org/content/100/10/5591">“Hessian Eigenmaps: Locally linear embedding techniques for
high-dimensional data”</a>
Donoho, D. & Grimes, C. Proc Natl Acad Sci USA. 100:5591 (2003)</p></li>
</ul>
</section>
<section id="spectral-embedding">
<span id="id3"></span><h2><span class="section-number">2.2.6. </span>Spectral Embedding<a class="headerlink" href="#spectral-embedding" title="Link to this heading">#</a></h2>
<p>Spectral Embedding is an approach to calculating a non-linear embedding.
Scikit-learn implements Laplacian Eigenmaps, which finds a low dimensional
representation of the data using a spectral decomposition of the graph
Laplacian. The graph generated can be considered as a discrete approximation of
the low dimensional manifold in the high dimensional space. Minimization of a
cost function based on the graph ensures that points close to each other on
the manifold are mapped close to each other in the low dimensional space,
preserving local distances. Spectral embedding can be performed with the
function <a class="reference internal" href="generated/sklearn.manifold.spectral_embedding.html#sklearn.manifold.spectral_embedding" title="sklearn.manifold.spectral_embedding"><code class="xref py py-func docutils literal notranslate"><span class="pre">spectral_embedding</span></code></a> or its object-oriented counterpart
<a class="reference internal" href="generated/sklearn.manifold.SpectralEmbedding.html#sklearn.manifold.SpectralEmbedding" title="sklearn.manifold.SpectralEmbedding"><code class="xref py py-class docutils literal notranslate"><span class="pre">SpectralEmbedding</span></code></a>.</p>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity-5">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity-5" 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">The Spectral Embedding (Laplacian Eigenmaps) algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Weighted Graph Construction</strong>. Transform the raw input data into
graph representation using affinity (adjacency) matrix representation.</p></li>
<li><p class="sd-card-text"><strong>Graph Laplacian Construction</strong>. unnormalized Graph Laplacian
is constructed as <span class="math notranslate nohighlight">\(L = D - A\)</span> for and normalized one as
<span class="math notranslate nohighlight">\(L = D^{-\frac{1}{2}} (D - A) D^{-\frac{1}{2}}\)</span>.</p></li>
<li><p class="sd-card-text"><strong>Partial Eigenvalue Decomposition</strong>. Eigenvalue decomposition is
done on graph Laplacian.</p></li>
</ol>
<p class="sd-card-text">The overall complexity of spectral embedding is
<span class="math notranslate nohighlight">\(O[D \log(k) N \log(N)] + O[D N k^3] + O[d N^2]\)</span>.</p>
<ul class="simple">
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(N\)</span> : number of training data points</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(D\)</span> : input dimension</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(k\)</span> : number of nearest neighbors</p></li>
<li><p class="sd-card-text"><span class="math notranslate nohighlight">\(d\)</span> : output dimension</p></li>
</ul>
</div>
</details><p class="rubric">References</p>
<ul class="simple">
<li><p><a class="reference external" href="https://web.cse.ohio-state.edu/~mbelkin/papers/LEM_NC_03.pdf">“Laplacian Eigenmaps for Dimensionality Reduction
and Data Representation”</a>
M. Belkin, P. Niyogi, Neural Computation, June 2003; 15 (6):1373-1396</p></li>
</ul>
</section>
<section id="local-tangent-space-alignment">
<h2><span class="section-number">2.2.7. </span>Local Tangent Space Alignment<a class="headerlink" href="#local-tangent-space-alignment" title="Link to this heading">#</a></h2>
<p>Though not technically a variant of LLE, Local tangent space alignment (LTSA)
is algorithmically similar enough to LLE that it can be put in this category.
Rather than focusing on preserving neighborhood distances as in LLE, LTSA
seeks to characterize the local geometry at each neighborhood via its
tangent space, and performs a global optimization to align these local
tangent spaces to learn the embedding. LTSA can be performed with function
<a class="reference internal" href="generated/sklearn.manifold.locally_linear_embedding.html#sklearn.manifold.locally_linear_embedding" title="sklearn.manifold.locally_linear_embedding"><code class="xref py py-func docutils literal notranslate"><span class="pre">locally_linear_embedding</span></code></a> or its object-oriented counterpart
<a class="reference internal" href="generated/sklearn.manifold.LocallyLinearEmbedding.html#sklearn.manifold.LocallyLinearEmbedding" title="sklearn.manifold.LocallyLinearEmbedding"><code class="xref py py-class docutils literal notranslate"><span class="pre">LocallyLinearEmbedding</span></code></a>, with the keyword <code class="docutils literal notranslate"><span class="pre">method</span> <span class="pre">=</span> <span class="pre">'ltsa'</span></code>.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/manifold/plot_lle_digits.html"><img alt="../_images/sphx_glr_plot_lle_digits_009.png" src="../_images/sphx_glr_plot_lle_digits_009.png" style="width: 320.0px; height: 240.0px;" />
</a>
</figure>
<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3" id="complexity-6">
<summary class="sd-summary-title sd-card-header">
<span class="sd-summary-text">Complexity<a class="headerlink" href="#complexity-6" 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">The LTSA algorithm comprises three stages:</p>
<ol class="arabic simple">
<li><p class="sd-card-text"><strong>Nearest Neighbors Search</strong>. Same as standard LLE</p></li>
<li><p class="sd-card-text"><strong>Weight Matrix Construction</strong>. Approximately
<span class="math notranslate nohighlight">\(O[D N k^3] + O[k^2 d]\)</span>. The first term reflects a similar