forked from pytorch/pytorch.github.io
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathnn.html
1376 lines (1146 loc) · 128 KB
/
nn.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>torch.nn — PyTorch master documentation</title>
<link rel="canonical" href="https://pytorch.org/docs/stable/nn.html"/>
<link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
<!-- <link rel="stylesheet" href="_static/pygments.css" type="text/css" /> -->
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/katex.min.css" type="text/css" />
<link rel="stylesheet" href="_static/css/jit.css" type="text/css" />
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/katex.min.css" type="text/css" />
<link rel="stylesheet" href="_static/katex-math.css" type="text/css" />
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Parameter" href="generated/torch.nn.parameter.Parameter.html" />
<link rel="prev" title="torch.promote_types" href="generated/torch.promote_types.html" />
<script src="_static/js/modernizr.min.js"></script>
<!-- Preload the theme fonts -->
<link rel="preload" href="_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<!-- Preload the katex fonts -->
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
</head>
<div class="container-fluid header-holder tutorials-header" id="header-holder">
<div class="container">
<div class="header-container">
<a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
<div class="main-menu">
<ul>
<li>
<a href="https://pytorch.org/get-started">Get Started</a>
</li>
<li>
<div class="ecosystem-dropdown">
<a id="dropdownMenuButton" data-toggle="ecosystem-dropdown">
Ecosystem
</a>
<div class="ecosystem-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/hub"">
<span class=dropdown-title>Models (Beta)</span>
<p>Discover, publish, and reuse pre-trained models</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/ecosystem">
<span class=dropdown-title>Tools & Libraries</span>
<p>Explore the ecosystem of tools and libraries</p>
</a>
</div>
</div>
</li>
<li>
<a href="https://pytorch.org/mobile">Mobile</a>
</li>
<li>
<a href="https://pytorch.org/blog/">Blog</a>
</li>
<li>
<a href="https://pytorch.org/tutorials">Tutorials</a>
</li>
<li class="active">
<a href="https://pytorch.org/docs/stable/index.html">Docs</a>
</li>
<li>
<div class="resources-dropdown">
<a id="resourcesDropdownButton" data-toggle="resources-dropdown">
Resources
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/resources"">
<span class=dropdown-title>Developer Resources</span>
<p>Find resources and get questions answered</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/features">
<span class=dropdown-title>About</span>
<p>Learn about PyTorch’s features and capabilities</p>
</a>
</div>
</div>
</li>
<li>
<a href="https://github.com/pytorch/pytorch">Github</a>
</li>
</ul>
</div>
<a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
</div>
</div>
</div>
<body class="pytorch-body">
<div class="table-of-contents-link-wrapper">
<span>Table of Contents</span>
<a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
</div>
<nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
<div class="pytorch-side-scroll">
<div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<div class="pytorch-left-menu-search">
<div class="version">
<a href='http://pytorch.org/docs/versions.html'>1.6.0a0+f2f8027 ▼</a>
</div>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="search.html" method="get">
<input type="text" name="q" placeholder="Search Docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<div>
<a style="color:#F05732" href="https://pytorch.org/docs/stable/nn.html">
You are viewing unstable developer preview docs.
Click here to view docs for latest stable release.
</a>
</div>
<p class="caption"><span class="caption-text">Notes</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="notes/amp_examples.html">Automatic Mixed Precision examples</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/autograd.html">Autograd mechanics</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/broadcasting.html">Broadcasting semantics</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/cpu_threading_torchscript_inference.html">CPU threading and TorchScript inference</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/cuda.html">CUDA semantics</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/ddp.html">Distributed Data Parallel</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/extending.html">Extending PyTorch</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/faq.html">Frequently Asked Questions</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/large_scale_deployments.html">Features for large-scale deployments</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/multiprocessing.html">Multiprocessing best practices</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/randomness.html">Reproducibility</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/serialization.html">Serialization semantics</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/windows.html">Windows FAQ</a></li>
</ul>
<p class="caption"><span class="caption-text">Language Bindings</span></p>
<ul>
<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/cppdocs/">C++ API</a></li>
<li class="toctree-l1"><a class="reference internal" href="packages.html">Javadoc</a></li>
</ul>
<p class="caption"><span class="caption-text">Python API</span></p>
<ul class="current">
<li class="toctree-l1"><a class="reference internal" href="torch.html">torch</a></li>
<li class="toctree-l1 current"><a class="current reference internal" href="#">torch.nn</a></li>
<li class="toctree-l1"><a class="reference internal" href="nn.functional.html">torch.nn.functional</a></li>
<li class="toctree-l1"><a class="reference internal" href="tensors.html">torch.Tensor</a></li>
<li class="toctree-l1"><a class="reference internal" href="tensor_attributes.html">Tensor Attributes</a></li>
<li class="toctree-l1"><a class="reference internal" href="tensor_view.html">Tensor Views</a></li>
<li class="toctree-l1"><a class="reference internal" href="autograd.html">torch.autograd</a></li>
<li class="toctree-l1"><a class="reference internal" href="cuda.html">torch.cuda</a></li>
<li class="toctree-l1"><a class="reference internal" href="amp.html">torch.cuda.amp</a></li>
<li class="toctree-l1"><a class="reference internal" href="distributed.html">torch.distributed</a></li>
<li class="toctree-l1"><a class="reference internal" href="distributions.html">torch.distributions</a></li>
<li class="toctree-l1"><a class="reference internal" href="hub.html">torch.hub</a></li>
<li class="toctree-l1"><a class="reference internal" href="jit.html">torch.jit</a></li>
<li class="toctree-l1"><a class="reference internal" href="nn.init.html">torch.nn.init</a></li>
<li class="toctree-l1"><a class="reference internal" href="onnx.html">torch.onnx</a></li>
<li class="toctree-l1"><a class="reference internal" href="optim.html">torch.optim</a></li>
<li class="toctree-l1"><a class="reference internal" href="quantization.html">Quantization</a></li>
<li class="toctree-l1"><a class="reference internal" href="rpc.html">Distributed RPC Framework</a></li>
<li class="toctree-l1"><a class="reference internal" href="random.html">torch.random</a></li>
<li class="toctree-l1"><a class="reference internal" href="sparse.html">torch.sparse</a></li>
<li class="toctree-l1"><a class="reference internal" href="storage.html">torch.Storage</a></li>
<li class="toctree-l1"><a class="reference internal" href="bottleneck.html">torch.utils.bottleneck</a></li>
<li class="toctree-l1"><a class="reference internal" href="checkpoint.html">torch.utils.checkpoint</a></li>
<li class="toctree-l1"><a class="reference internal" href="cpp_extension.html">torch.utils.cpp_extension</a></li>
<li class="toctree-l1"><a class="reference internal" href="data.html">torch.utils.data</a></li>
<li class="toctree-l1"><a class="reference internal" href="dlpack.html">torch.utils.dlpack</a></li>
<li class="toctree-l1"><a class="reference internal" href="model_zoo.html">torch.utils.model_zoo</a></li>
<li class="toctree-l1"><a class="reference internal" href="tensorboard.html">torch.utils.tensorboard</a></li>
<li class="toctree-l1"><a class="reference internal" href="type_info.html">Type Info</a></li>
<li class="toctree-l1"><a class="reference internal" href="named_tensor.html">Named Tensors</a></li>
<li class="toctree-l1"><a class="reference internal" href="name_inference.html">Named Tensors operator coverage</a></li>
<li class="toctree-l1"><a class="reference internal" href="__config__.html">torch.__config__</a></li>
</ul>
<p class="caption"><span class="caption-text">Libraries</span></p>
<ul>
<li class="toctree-l1"><a class="reference external" href="http://pytorch.org/xla/">PyTorch on XLA Devices</a></li>
<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/elastic/">PyTorch Elastic (torchelastic)</a></li>
<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/audio">torchaudio</a></li>
<li class="toctree-l1"><a class="reference external" href="https://pytorch.org/text">torchtext</a></li>
<li class="toctree-l1"><a class="reference internal" href="torchvision/index.html">torchvision</a></li>
</ul>
<p class="caption"><span class="caption-text">Community</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="community/contribution_guide.html">PyTorch Contribution Guide</a></li>
<li class="toctree-l1"><a class="reference internal" href="community/governance.html">PyTorch Governance</a></li>
<li class="toctree-l1"><a class="reference internal" href="community/persons_of_interest.html">PyTorch Governance | Persons of Interest</a></li>
</ul>
</div>
</div>
</nav>
<div class="pytorch-container">
<div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
<div class="pytorch-breadcrumbs-wrapper">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="pytorch-breadcrumbs">
<li>
<a href="index.html">
Docs
</a> >
</li>
<li>torch.nn</li>
<li class="pytorch-breadcrumbs-aside">
<a href="_sources/nn.rst.txt" rel="nofollow"><img src="_static/images/view-page-source-icon.svg"></a>
</li>
</ul>
</div>
</div>
<div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
Shortcuts
</div>
</div>
<section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
<div class="pytorch-content-left">
<div class="rst-content">
<div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
<article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
<div class="section" id="torch-nn">
<h1>torch.nn<a class="headerlink" href="#torch-nn" title="Permalink to this headline">¶</a></h1>
<p>These are the basic building block for graphs</p>
<div class="contents local topic" id="id1">
<p class="topic-title">torch.nn</p>
<ul class="simple">
<li><p><a class="reference internal" href="#containers" id="id2">Containers</a></p></li>
<li><p><a class="reference internal" href="#convolution-layers" id="id3">Convolution Layers</a></p></li>
<li><p><a class="reference internal" href="#pooling-layers" id="id4">Pooling layers</a></p></li>
<li><p><a class="reference internal" href="#padding-layers" id="id5">Padding Layers</a></p></li>
<li><p><a class="reference internal" href="#non-linear-activations-weighted-sum-nonlinearity" id="id6">Non-linear Activations (weighted sum, nonlinearity)</a></p></li>
<li><p><a class="reference internal" href="#non-linear-activations-other" id="id7">Non-linear Activations (other)</a></p></li>
<li><p><a class="reference internal" href="#normalization-layers" id="id8">Normalization Layers</a></p></li>
<li><p><a class="reference internal" href="#recurrent-layers" id="id9">Recurrent Layers</a></p></li>
<li><p><a class="reference internal" href="#transformer-layers" id="id10">Transformer Layers</a></p></li>
<li><p><a class="reference internal" href="#linear-layers" id="id11">Linear Layers</a></p></li>
<li><p><a class="reference internal" href="#dropout-layers" id="id12">Dropout Layers</a></p></li>
<li><p><a class="reference internal" href="#sparse-layers" id="id13">Sparse Layers</a></p></li>
<li><p><a class="reference internal" href="#distance-functions" id="id14">Distance Functions</a></p></li>
<li><p><a class="reference internal" href="#loss-functions" id="id15">Loss Functions</a></p></li>
<li><p><a class="reference internal" href="#vision-layers" id="id16">Vision Layers</a></p></li>
<li><p><a class="reference internal" href="#dataparallel-layers-multi-gpu-distributed" id="id17">DataParallel Layers (multi-GPU, distributed)</a></p></li>
<li><p><a class="reference internal" href="#utilities" id="id18">Utilities</a></p></li>
<li><p><a class="reference internal" href="#quantized-functions" id="id19">Quantized Functions</a></p></li>
</ul>
</div>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.parameter.Parameter.html#torch.nn.parameter.Parameter" title="torch.nn.parameter.Parameter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Parameter</span></code></a></p></td>
<td><p>A kind of Tensor that is to be considered a module parameter.</p></td>
</tr>
</tbody>
</table>
<div class="section" id="containers">
<h2><a class="toc-backref" href="#id1">Containers</a><a class="headerlink" href="#containers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Module.html#torch.nn.Module" title="torch.nn.Module"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Module</span></code></a></p></td>
<td><p>Base class for all neural network modules.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Sequential.html#torch.nn.Sequential" title="torch.nn.Sequential"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Sequential</span></code></a></p></td>
<td><p>A sequential container.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ModuleList.html#torch.nn.ModuleList" title="torch.nn.ModuleList"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ModuleList</span></code></a></p></td>
<td><p>Holds submodules in a list.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ModuleDict.html#torch.nn.ModuleDict" title="torch.nn.ModuleDict"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ModuleDict</span></code></a></p></td>
<td><p>Holds submodules in a dictionary.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ParameterList.html#torch.nn.ParameterList" title="torch.nn.ParameterList"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ParameterList</span></code></a></p></td>
<td><p>Holds parameters in a list.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ParameterDict.html#torch.nn.ParameterDict" title="torch.nn.ParameterDict"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ParameterDict</span></code></a></p></td>
<td><p>Holds parameters in a dictionary.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="convolution-layers">
<h2><a class="toc-backref" href="#id1">Convolution Layers</a><a class="headerlink" href="#convolution-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Conv1d.html#torch.nn.Conv1d" title="torch.nn.Conv1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Conv1d</span></code></a></p></td>
<td><p>Applies a 1D convolution over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Conv2d.html#torch.nn.Conv2d" title="torch.nn.Conv2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Conv2d</span></code></a></p></td>
<td><p>Applies a 2D convolution over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Conv3d.html#torch.nn.Conv3d" title="torch.nn.Conv3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Conv3d</span></code></a></p></td>
<td><p>Applies a 3D convolution over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ConvTranspose1d.html#torch.nn.ConvTranspose1d" title="torch.nn.ConvTranspose1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConvTranspose1d</span></code></a></p></td>
<td><p>Applies a 1D transposed convolution operator over an input image composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ConvTranspose2d.html#torch.nn.ConvTranspose2d" title="torch.nn.ConvTranspose2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConvTranspose2d</span></code></a></p></td>
<td><p>Applies a 2D transposed convolution operator over an input image composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ConvTranspose3d.html#torch.nn.ConvTranspose3d" title="torch.nn.ConvTranspose3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConvTranspose3d</span></code></a></p></td>
<td><p>Applies a 3D transposed convolution operator over an input image composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Unfold.html#torch.nn.Unfold" title="torch.nn.Unfold"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Unfold</span></code></a></p></td>
<td><p>Extracts sliding local blocks from a batched input tensor.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Fold.html#torch.nn.Fold" title="torch.nn.Fold"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Fold</span></code></a></p></td>
<td><p>Combines an array of sliding local blocks into a large containing tensor.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="pooling-layers">
<h2><a class="toc-backref" href="#id1">Pooling layers</a><a class="headerlink" href="#pooling-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.MaxPool1d.html#torch.nn.MaxPool1d" title="torch.nn.MaxPool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxPool1d</span></code></a></p></td>
<td><p>Applies a 1D max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MaxPool2d.html#torch.nn.MaxPool2d" title="torch.nn.MaxPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxPool2d</span></code></a></p></td>
<td><p>Applies a 2D max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.MaxPool3d.html#torch.nn.MaxPool3d" title="torch.nn.MaxPool3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxPool3d</span></code></a></p></td>
<td><p>Applies a 3D max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MaxUnpool1d.html#torch.nn.MaxUnpool1d" title="torch.nn.MaxUnpool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxUnpool1d</span></code></a></p></td>
<td><p>Computes a partial inverse of <code class="xref py py-class docutils literal notranslate"><span class="pre">MaxPool1d</span></code>.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.MaxUnpool2d.html#torch.nn.MaxUnpool2d" title="torch.nn.MaxUnpool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxUnpool2d</span></code></a></p></td>
<td><p>Computes a partial inverse of <code class="xref py py-class docutils literal notranslate"><span class="pre">MaxPool2d</span></code>.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MaxUnpool3d.html#torch.nn.MaxUnpool3d" title="torch.nn.MaxUnpool3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MaxUnpool3d</span></code></a></p></td>
<td><p>Computes a partial inverse of <code class="xref py py-class docutils literal notranslate"><span class="pre">MaxPool3d</span></code>.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AvgPool1d.html#torch.nn.AvgPool1d" title="torch.nn.AvgPool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AvgPool1d</span></code></a></p></td>
<td><p>Applies a 1D average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.AvgPool2d.html#torch.nn.AvgPool2d" title="torch.nn.AvgPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AvgPool2d</span></code></a></p></td>
<td><p>Applies a 2D average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AvgPool3d.html#torch.nn.AvgPool3d" title="torch.nn.AvgPool3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AvgPool3d</span></code></a></p></td>
<td><p>Applies a 3D average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.FractionalMaxPool2d.html#torch.nn.FractionalMaxPool2d" title="torch.nn.FractionalMaxPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.FractionalMaxPool2d</span></code></a></p></td>
<td><p>Applies a 2D fractional max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.LPPool1d.html#torch.nn.LPPool1d" title="torch.nn.LPPool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LPPool1d</span></code></a></p></td>
<td><p>Applies a 1D power-average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.LPPool2d.html#torch.nn.LPPool2d" title="torch.nn.LPPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LPPool2d</span></code></a></p></td>
<td><p>Applies a 2D power-average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveMaxPool1d.html#torch.nn.AdaptiveMaxPool1d" title="torch.nn.AdaptiveMaxPool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveMaxPool1d</span></code></a></p></td>
<td><p>Applies a 1D adaptive max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveMaxPool2d.html#torch.nn.AdaptiveMaxPool2d" title="torch.nn.AdaptiveMaxPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveMaxPool2d</span></code></a></p></td>
<td><p>Applies a 2D adaptive max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveMaxPool3d.html#torch.nn.AdaptiveMaxPool3d" title="torch.nn.AdaptiveMaxPool3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveMaxPool3d</span></code></a></p></td>
<td><p>Applies a 3D adaptive max pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveAvgPool1d.html#torch.nn.AdaptiveAvgPool1d" title="torch.nn.AdaptiveAvgPool1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveAvgPool1d</span></code></a></p></td>
<td><p>Applies a 1D adaptive average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveAvgPool2d.html#torch.nn.AdaptiveAvgPool2d" title="torch.nn.AdaptiveAvgPool2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveAvgPool2d</span></code></a></p></td>
<td><p>Applies a 2D adaptive average pooling over an input signal composed of several input planes.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveAvgPool3d.html#torch.nn.AdaptiveAvgPool3d" title="torch.nn.AdaptiveAvgPool3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveAvgPool3d</span></code></a></p></td>
<td><p>Applies a 3D adaptive average pooling over an input signal composed of several input planes.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="padding-layers">
<h2><a class="toc-backref" href="#id1">Padding Layers</a><a class="headerlink" href="#padding-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ReflectionPad1d.html#torch.nn.ReflectionPad1d" title="torch.nn.ReflectionPad1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReflectionPad1d</span></code></a></p></td>
<td><p>Pads the input tensor using the reflection of the input boundary.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ReflectionPad2d.html#torch.nn.ReflectionPad2d" title="torch.nn.ReflectionPad2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReflectionPad2d</span></code></a></p></td>
<td><p>Pads the input tensor using the reflection of the input boundary.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ReplicationPad1d.html#torch.nn.ReplicationPad1d" title="torch.nn.ReplicationPad1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReplicationPad1d</span></code></a></p></td>
<td><p>Pads the input tensor using replication of the input boundary.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ReplicationPad2d.html#torch.nn.ReplicationPad2d" title="torch.nn.ReplicationPad2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReplicationPad2d</span></code></a></p></td>
<td><p>Pads the input tensor using replication of the input boundary.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ReplicationPad3d.html#torch.nn.ReplicationPad3d" title="torch.nn.ReplicationPad3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReplicationPad3d</span></code></a></p></td>
<td><p>Pads the input tensor using replication of the input boundary.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ZeroPad2d.html#torch.nn.ZeroPad2d" title="torch.nn.ZeroPad2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ZeroPad2d</span></code></a></p></td>
<td><p>Pads the input tensor boundaries with zero.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ConstantPad1d.html#torch.nn.ConstantPad1d" title="torch.nn.ConstantPad1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConstantPad1d</span></code></a></p></td>
<td><p>Pads the input tensor boundaries with a constant value.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ConstantPad2d.html#torch.nn.ConstantPad2d" title="torch.nn.ConstantPad2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConstantPad2d</span></code></a></p></td>
<td><p>Pads the input tensor boundaries with a constant value.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ConstantPad3d.html#torch.nn.ConstantPad3d" title="torch.nn.ConstantPad3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ConstantPad3d</span></code></a></p></td>
<td><p>Pads the input tensor boundaries with a constant value.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="non-linear-activations-weighted-sum-nonlinearity">
<h2><a class="toc-backref" href="#id1">Non-linear Activations (weighted sum, nonlinearity)</a><a class="headerlink" href="#non-linear-activations-weighted-sum-nonlinearity" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ELU.html#torch.nn.ELU" title="torch.nn.ELU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ELU</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Hardshrink.html#torch.nn.Hardshrink" title="torch.nn.Hardshrink"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Hardshrink</span></code></a></p></td>
<td><p>Applies the hard shrinkage function element-wise:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Hardtanh.html#torch.nn.Hardtanh" title="torch.nn.Hardtanh"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Hardtanh</span></code></a></p></td>
<td><p>Applies the HardTanh function element-wise</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.LeakyReLU.html#torch.nn.LeakyReLU" title="torch.nn.LeakyReLU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LeakyReLU</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.LogSigmoid.html#torch.nn.LogSigmoid" title="torch.nn.LogSigmoid"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LogSigmoid</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MultiheadAttention.html#torch.nn.MultiheadAttention" title="torch.nn.MultiheadAttention"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MultiheadAttention</span></code></a></p></td>
<td><p>Allows the model to jointly attend to information from different representation subspaces.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.PReLU.html#torch.nn.PReLU" title="torch.nn.PReLU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.PReLU</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.ReLU.html#torch.nn.ReLU" title="torch.nn.ReLU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReLU</span></code></a></p></td>
<td><p>Applies the rectified linear unit function element-wise:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.ReLU6.html#torch.nn.ReLU6" title="torch.nn.ReLU6"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.ReLU6</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.RReLU.html#torch.nn.RReLU" title="torch.nn.RReLU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.RReLU</span></code></a></p></td>
<td><p>Applies the randomized leaky rectified liner unit function, element-wise, as described in the paper:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.SELU.html#torch.nn.SELU" title="torch.nn.SELU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.SELU</span></code></a></p></td>
<td><p>Applied element-wise, as:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.CELU.html#torch.nn.CELU" title="torch.nn.CELU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.CELU</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.GELU.html#torch.nn.GELU" title="torch.nn.GELU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.GELU</span></code></a></p></td>
<td><p>Applies the Gaussian Error Linear Units function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Sigmoid.html#torch.nn.Sigmoid" title="torch.nn.Sigmoid"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Sigmoid</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Softplus.html#torch.nn.Softplus" title="torch.nn.Softplus"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softplus</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Softshrink.html#torch.nn.Softshrink" title="torch.nn.Softshrink"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softshrink</span></code></a></p></td>
<td><p>Applies the soft shrinkage function elementwise:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Softsign.html#torch.nn.Softsign" title="torch.nn.Softsign"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softsign</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Tanh.html#torch.nn.Tanh" title="torch.nn.Tanh"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Tanh</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Tanhshrink.html#torch.nn.Tanhshrink" title="torch.nn.Tanhshrink"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Tanhshrink</span></code></a></p></td>
<td><p>Applies the element-wise function:</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Threshold.html#torch.nn.Threshold" title="torch.nn.Threshold"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Threshold</span></code></a></p></td>
<td><p>Thresholds each element of the input Tensor.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="non-linear-activations-other">
<h2><a class="toc-backref" href="#id1">Non-linear Activations (other)</a><a class="headerlink" href="#non-linear-activations-other" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Softmin.html#torch.nn.Softmin" title="torch.nn.Softmin"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softmin</span></code></a></p></td>
<td><p>Applies the Softmin function to an n-dimensional input Tensor rescaling them so that the elements of the n-dimensional output Tensor lie in the range <cite>[0, 1]</cite> and sum to 1.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Softmax.html#torch.nn.Softmax" title="torch.nn.Softmax"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softmax</span></code></a></p></td>
<td><p>Applies the Softmax function to an n-dimensional input Tensor rescaling them so that the elements of the n-dimensional output Tensor lie in the range [0,1] and sum to 1.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Softmax2d.html#torch.nn.Softmax2d" title="torch.nn.Softmax2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Softmax2d</span></code></a></p></td>
<td><p>Applies SoftMax over features to each spatial location.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.LogSoftmax.html#torch.nn.LogSoftmax" title="torch.nn.LogSoftmax"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LogSoftmax</span></code></a></p></td>
<td><p>Applies the <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>log</mi><mo></mo><mo stretchy="false">(</mo><mtext>Softmax</mtext><mo stretchy="false">(</mo><mi>x</mi><mo stretchy="false">)</mo><mo stretchy="false">)</mo></mrow><annotation encoding="application/x-tex">\log(\text{Softmax}(x))</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mop">lo<span style="margin-right:0.01389em;">g</span></span><span class="mopen">(</span><span class="mord text"><span class="mord">Softmax</span></span><span class="mopen">(</span><span class="mord mathdefault">x</span><span class="mclose">)</span><span class="mclose">)</span></span></span></span>
</span> function to an n-dimensional input Tensor.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.AdaptiveLogSoftmaxWithLoss.html#torch.nn.AdaptiveLogSoftmaxWithLoss" title="torch.nn.AdaptiveLogSoftmaxWithLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AdaptiveLogSoftmaxWithLoss</span></code></a></p></td>
<td><p>Efficient softmax approximation as described in <a class="reference external" href="https://arxiv.org/abs/1609.04309">Efficient softmax approximation for GPUs by Edouard Grave, Armand Joulin, Moustapha Cissé, David Grangier, and Hervé Jégou</a>.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="normalization-layers">
<h2><a class="toc-backref" href="#id1">Normalization Layers</a><a class="headerlink" href="#normalization-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.BatchNorm1d.html#torch.nn.BatchNorm1d" title="torch.nn.BatchNorm1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.BatchNorm1d</span></code></a></p></td>
<td><p>Applies Batch Normalization over a 2D or 3D input (a mini-batch of 1D inputs with optional additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift</a> .</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.BatchNorm2d.html#torch.nn.BatchNorm2d" title="torch.nn.BatchNorm2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.BatchNorm2d</span></code></a></p></td>
<td><p>Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift</a> .</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.BatchNorm3d.html#torch.nn.BatchNorm3d" title="torch.nn.BatchNorm3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.BatchNorm3d</span></code></a></p></td>
<td><p>Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift</a> .</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.GroupNorm.html#torch.nn.GroupNorm" title="torch.nn.GroupNorm"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.GroupNorm</span></code></a></p></td>
<td><p>Applies Group Normalization over a mini-batch of inputs as described in the paper <a class="reference external" href="https://arxiv.org/abs/1803.08494">Group Normalization</a></p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.SyncBatchNorm.html#torch.nn.SyncBatchNorm" title="torch.nn.SyncBatchNorm"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.SyncBatchNorm</span></code></a></p></td>
<td><p>Applies Batch Normalization over a N-Dimensional input (a mini-batch of [N-2]D inputs with additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1502.03167">Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift</a> .</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.InstanceNorm1d.html#torch.nn.InstanceNorm1d" title="torch.nn.InstanceNorm1d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.InstanceNorm1d</span></code></a></p></td>
<td><p>Applies Instance Normalization over a 3D input (a mini-batch of 1D inputs with optional additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1607.08022">Instance Normalization: The Missing Ingredient for Fast Stylization</a>.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.InstanceNorm2d.html#torch.nn.InstanceNorm2d" title="torch.nn.InstanceNorm2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.InstanceNorm2d</span></code></a></p></td>
<td><p>Applies Instance Normalization over a 4D input (a mini-batch of 2D inputs with additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1607.08022">Instance Normalization: The Missing Ingredient for Fast Stylization</a>.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.InstanceNorm3d.html#torch.nn.InstanceNorm3d" title="torch.nn.InstanceNorm3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.InstanceNorm3d</span></code></a></p></td>
<td><p>Applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper <a class="reference external" href="https://arxiv.org/abs/1607.08022">Instance Normalization: The Missing Ingredient for Fast Stylization</a>.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.LayerNorm.html#torch.nn.LayerNorm" title="torch.nn.LayerNorm"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LayerNorm</span></code></a></p></td>
<td><p>Applies Layer Normalization over a mini-batch of inputs as described in the paper <a class="reference external" href="https://arxiv.org/abs/1607.06450">Layer Normalization</a></p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.LocalResponseNorm.html#torch.nn.LocalResponseNorm" title="torch.nn.LocalResponseNorm"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LocalResponseNorm</span></code></a></p></td>
<td><p>Applies local response normalization over an input signal composed of several input planes, where channels occupy the second dimension.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="recurrent-layers">
<h2><a class="toc-backref" href="#id1">Recurrent Layers</a><a class="headerlink" href="#recurrent-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.RNNBase.html#torch.nn.RNNBase" title="torch.nn.RNNBase"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.RNNBase</span></code></a></p></td>
<td><p></p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.RNN.html#torch.nn.RNN" title="torch.nn.RNN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.RNN</span></code></a></p></td>
<td><p>Applies a multi-layer Elman RNN with <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>tanh</mi><mo></mo></mrow><annotation encoding="application/x-tex">\tanh</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.69444em;vertical-align:0em;"></span><span class="mop">tanh</span></span></span></span>
</span> or <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mtext>ReLU</mtext></mrow><annotation encoding="application/x-tex">\text{ReLU}</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.68333em;vertical-align:0em;"></span><span class="mord text"><span class="mord">ReLU</span></span></span></span></span>
</span> non-linearity to an input sequence.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.LSTM.html#torch.nn.LSTM" title="torch.nn.LSTM"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LSTM</span></code></a></p></td>
<td><p>Applies a multi-layer long short-term memory (LSTM) RNN to an input sequence.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.GRU.html#torch.nn.GRU" title="torch.nn.GRU"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.GRU</span></code></a></p></td>
<td><p>Applies a multi-layer gated recurrent unit (GRU) RNN to an input sequence.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.RNNCell.html#torch.nn.RNNCell" title="torch.nn.RNNCell"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.RNNCell</span></code></a></p></td>
<td><p>An Elman RNN cell with tanh or ReLU non-linearity.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.LSTMCell.html#torch.nn.LSTMCell" title="torch.nn.LSTMCell"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.LSTMCell</span></code></a></p></td>
<td><p>A long short-term memory (LSTM) cell.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.GRUCell.html#torch.nn.GRUCell" title="torch.nn.GRUCell"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.GRUCell</span></code></a></p></td>
<td><p>A gated recurrent unit (GRU) cell</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="transformer-layers">
<h2><a class="toc-backref" href="#id1">Transformer Layers</a><a class="headerlink" href="#transformer-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Transformer.html#torch.nn.Transformer" title="torch.nn.Transformer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Transformer</span></code></a></p></td>
<td><p>A transformer model.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.TransformerEncoder.html#torch.nn.TransformerEncoder" title="torch.nn.TransformerEncoder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.TransformerEncoder</span></code></a></p></td>
<td><p>TransformerEncoder is a stack of N encoder layers</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.TransformerDecoder.html#torch.nn.TransformerDecoder" title="torch.nn.TransformerDecoder"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.TransformerDecoder</span></code></a></p></td>
<td><p>TransformerDecoder is a stack of N decoder layers</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.TransformerEncoderLayer.html#torch.nn.TransformerEncoderLayer" title="torch.nn.TransformerEncoderLayer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.TransformerEncoderLayer</span></code></a></p></td>
<td><p>TransformerEncoderLayer is made up of self-attn and feedforward network.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.TransformerDecoderLayer.html#torch.nn.TransformerDecoderLayer" title="torch.nn.TransformerDecoderLayer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.TransformerDecoderLayer</span></code></a></p></td>
<td><p>TransformerDecoderLayer is made up of self-attn, multi-head-attn and feedforward network.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="linear-layers">
<h2><a class="toc-backref" href="#id1">Linear Layers</a><a class="headerlink" href="#linear-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Identity.html#torch.nn.Identity" title="torch.nn.Identity"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Identity</span></code></a></p></td>
<td><p>A placeholder identity operator that is argument-insensitive.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Linear.html#torch.nn.Linear" title="torch.nn.Linear"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Linear</span></code></a></p></td>
<td><p>Applies a linear transformation to the incoming data: <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi><mo>=</mo><mi>x</mi><msup><mi>A</mi><mi>T</mi></msup><mo>+</mo><mi>b</mi></mrow><annotation encoding="application/x-tex">y = xA^T + b</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span><span class="mrel">=</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span></span><span class="base"><span class="strut" style="height:0.924661em;vertical-align:-0.08333em;"></span><span class="mord mathdefault">x</span><span class="mord"><span class="mord mathdefault">A</span><span class="msupsub"><span class="vlist-t"><span class="vlist-r"><span class="vlist" style="height:0.8413309999999999em;"><span style="top:-3.063em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathdefault mtight" style="margin-right:0.13889em;">T</span></span></span></span></span></span></span></span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">+</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:0.69444em;vertical-align:0em;"></span><span class="mord mathdefault">b</span></span></span></span>
</span></p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Bilinear.html#torch.nn.Bilinear" title="torch.nn.Bilinear"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Bilinear</span></code></a></p></td>
<td><p>Applies a bilinear transformation to the incoming data: <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi><mo>=</mo><msubsup><mi>x</mi><mn>1</mn><mi>T</mi></msubsup><mi>A</mi><msub><mi>x</mi><mn>2</mn></msub><mo>+</mo><mi>b</mi></mrow><annotation encoding="application/x-tex">y = x_1^T A x_2 + b</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span><span class="mrel">=</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span></span><span class="base"><span class="strut" style="height:1.0894389999999998em;vertical-align:-0.24810799999999997em;"></span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.8413309999999999em;"><span style="top:-2.4518920000000004em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">1</span></span></span><span style="top:-3.063em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mathdefault mtight" style="margin-right:0.13889em;">T</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.24810799999999997em;"><span></span></span></span></span></span></span><span class="mord mathdefault">A</span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">2</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">+</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:0.69444em;vertical-align:0em;"></span><span class="mord mathdefault">b</span></span></span></span>
</span></p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="dropout-layers">
<h2><a class="toc-backref" href="#id1">Dropout Layers</a><a class="headerlink" href="#dropout-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Dropout.html#torch.nn.Dropout" title="torch.nn.Dropout"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Dropout</span></code></a></p></td>
<td><p>During training, randomly zeroes some of the elements of the input tensor with probability <code class="xref py py-attr docutils literal notranslate"><span class="pre">p</span></code> using samples from a Bernoulli distribution.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Dropout2d.html#torch.nn.Dropout2d" title="torch.nn.Dropout2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Dropout2d</span></code></a></p></td>
<td><p>Randomly zero out entire channels (a channel is a 2D feature map, e.g., the <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>j</mi></mrow><annotation encoding="application/x-tex">j</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.85396em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.05724em;">j</span></span></span></span>
</span>-th channel of the <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>i</mi></mrow><annotation encoding="application/x-tex">i</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.65952em;vertical-align:0em;"></span><span class="mord mathdefault">i</span></span></span></span>
</span>-th sample in the batched input is a 2D tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mtext>input</mtext><mo stretchy="false">[</mo><mi>i</mi><mo separator="true">,</mo><mi>j</mi><mo stretchy="false">]</mo></mrow><annotation encoding="application/x-tex">\text{input}[i, j]</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mord text"><span class="mord">input</span></span><span class="mopen">[</span><span class="mord mathdefault">i</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.05724em;">j</span><span class="mclose">]</span></span></span></span>
</span>).</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Dropout3d.html#torch.nn.Dropout3d" title="torch.nn.Dropout3d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Dropout3d</span></code></a></p></td>
<td><p>Randomly zero out entire channels (a channel is a 3D feature map, e.g., the <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>j</mi></mrow><annotation encoding="application/x-tex">j</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.85396em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.05724em;">j</span></span></span></span>
</span>-th channel of the <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>i</mi></mrow><annotation encoding="application/x-tex">i</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.65952em;vertical-align:0em;"></span><span class="mord mathdefault">i</span></span></span></span>
</span>-th sample in the batched input is a 3D tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mtext>input</mtext><mo stretchy="false">[</mo><mi>i</mi><mo separator="true">,</mo><mi>j</mi><mo stretchy="false">]</mo></mrow><annotation encoding="application/x-tex">\text{input}[i, j]</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mord text"><span class="mord">input</span></span><span class="mopen">[</span><span class="mord mathdefault">i</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.05724em;">j</span><span class="mclose">]</span></span></span></span>
</span>).</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.AlphaDropout.html#torch.nn.AlphaDropout" title="torch.nn.AlphaDropout"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.AlphaDropout</span></code></a></p></td>
<td><p>Applies Alpha Dropout over the input.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="sparse-layers">
<h2><a class="toc-backref" href="#id1">Sparse Layers</a><a class="headerlink" href="#sparse-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.Embedding.html#torch.nn.Embedding" title="torch.nn.Embedding"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Embedding</span></code></a></p></td>
<td><p>A simple lookup table that stores embeddings of a fixed dictionary and size.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.EmbeddingBag.html#torch.nn.EmbeddingBag" title="torch.nn.EmbeddingBag"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.EmbeddingBag</span></code></a></p></td>
<td><p>Computes sums or means of ‘bags’ of embeddings, without instantiating the intermediate embeddings.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="distance-functions">
<h2><a class="toc-backref" href="#id1">Distance Functions</a><a class="headerlink" href="#distance-functions" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.CosineSimilarity.html#torch.nn.CosineSimilarity" title="torch.nn.CosineSimilarity"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.CosineSimilarity</span></code></a></p></td>
<td><p>Returns cosine similarity between <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>x</mi><mn>1</mn></msub></mrow><annotation encoding="application/x-tex">x_1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">1</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span> and <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>x</mi><mn>2</mn></msub></mrow><annotation encoding="application/x-tex">x_2</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">2</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span>, computed along dim.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.PairwiseDistance.html#torch.nn.PairwiseDistance" title="torch.nn.PairwiseDistance"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.PairwiseDistance</span></code></a></p></td>
<td><p>Computes the batchwise pairwise distance between vectors <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>v</mi><mn>1</mn></msub></mrow><annotation encoding="application/x-tex">v_1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault" style="margin-right:0.03588em;">v</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:-0.03588em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">1</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span>, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>v</mi><mn>2</mn></msub></mrow><annotation encoding="application/x-tex">v_2</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault" style="margin-right:0.03588em;">v</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:-0.03588em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">2</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span> using the p-norm:</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="loss-functions">
<h2><a class="toc-backref" href="#id1">Loss Functions</a><a class="headerlink" href="#loss-functions" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.L1Loss.html#torch.nn.L1Loss" title="torch.nn.L1Loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.L1Loss</span></code></a></p></td>
<td><p>Creates a criterion that measures the mean absolute error (MAE) between each element in the input <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> and target <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span>.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MSELoss.html#torch.nn.MSELoss" title="torch.nn.MSELoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MSELoss</span></code></a></p></td>
<td><p>Creates a criterion that measures the mean squared error (squared L2 norm) between each element in the input <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> and target <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span>.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.CrossEntropyLoss.html#torch.nn.CrossEntropyLoss" title="torch.nn.CrossEntropyLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.CrossEntropyLoss</span></code></a></p></td>
<td><p>This criterion combines <a class="reference internal" href="generated/torch.nn.LogSoftmax.html#torch.nn.LogSoftmax" title="torch.nn.LogSoftmax"><code class="xref py py-func docutils literal notranslate"><span class="pre">nn.LogSoftmax()</span></code></a> and <a class="reference internal" href="generated/torch.nn.NLLLoss.html#torch.nn.NLLLoss" title="torch.nn.NLLLoss"><code class="xref py py-func docutils literal notranslate"><span class="pre">nn.NLLLoss()</span></code></a> in one single class.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.CTCLoss.html#torch.nn.CTCLoss" title="torch.nn.CTCLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.CTCLoss</span></code></a></p></td>
<td><p>The Connectionist Temporal Classification loss.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.NLLLoss.html#torch.nn.NLLLoss" title="torch.nn.NLLLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.NLLLoss</span></code></a></p></td>
<td><p>The negative log likelihood loss.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.PoissonNLLLoss.html#torch.nn.PoissonNLLLoss" title="torch.nn.PoissonNLLLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.PoissonNLLLoss</span></code></a></p></td>
<td><p>Negative log likelihood loss with Poisson distribution of target.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.KLDivLoss.html#torch.nn.KLDivLoss" title="torch.nn.KLDivLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.KLDivLoss</span></code></a></p></td>
<td><p>The <a href="#id20"><span class="problematic" id="id21">`Kullback-Leibler divergence`_</span></a> Loss</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.BCELoss.html#torch.nn.BCELoss" title="torch.nn.BCELoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.BCELoss</span></code></a></p></td>
<td><p>Creates a criterion that measures the Binary Cross Entropy between the target and the output:</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.BCEWithLogitsLoss.html#torch.nn.BCEWithLogitsLoss" title="torch.nn.BCEWithLogitsLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.BCEWithLogitsLoss</span></code></a></p></td>
<td><p>This loss combines a <cite>Sigmoid</cite> layer and the <cite>BCELoss</cite> in one single class.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MarginRankingLoss.html#torch.nn.MarginRankingLoss" title="torch.nn.MarginRankingLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MarginRankingLoss</span></code></a></p></td>
<td><p>Creates a criterion that measures the loss given inputs <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi><mn>1</mn></mrow><annotation encoding="application/x-tex">x1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord mathdefault">x</span><span class="mord">1</span></span></span></span>
</span>, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi><mn>2</mn></mrow><annotation encoding="application/x-tex">x2</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord mathdefault">x</span><span class="mord">2</span></span></span></span>
</span>, two 1D mini-batch <cite>Tensors</cite>, and a label 1D mini-batch tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> (containing 1 or -1).</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.HingeEmbeddingLoss.html#torch.nn.HingeEmbeddingLoss" title="torch.nn.HingeEmbeddingLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.HingeEmbeddingLoss</span></code></a></p></td>
<td><p>Measures the loss given an input tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> and a labels tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> (containing 1 or -1).</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.MultiLabelMarginLoss.html#torch.nn.MultiLabelMarginLoss" title="torch.nn.MultiLabelMarginLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MultiLabelMarginLoss</span></code></a></p></td>
<td><p>Creates a criterion that optimizes a multi-class multi-classification hinge loss (margin-based loss) between input <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> (a 2D mini-batch <cite>Tensor</cite>) and output <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> (which is a 2D <cite>Tensor</cite> of target class indices).</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.SmoothL1Loss.html#torch.nn.SmoothL1Loss" title="torch.nn.SmoothL1Loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.SmoothL1Loss</span></code></a></p></td>
<td><p>Creates a criterion that uses a squared term if the absolute element-wise error falls below 1 and an L1 term otherwise.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.SoftMarginLoss.html#torch.nn.SoftMarginLoss" title="torch.nn.SoftMarginLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.SoftMarginLoss</span></code></a></p></td>
<td><p>Creates a criterion that optimizes a two-class classification logistic loss between input tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> and target tensor <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> (containing 1 or -1).</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.MultiLabelSoftMarginLoss.html#torch.nn.MultiLabelSoftMarginLoss" title="torch.nn.MultiLabelSoftMarginLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MultiLabelSoftMarginLoss</span></code></a></p></td>
<td><p>Creates a criterion that optimizes a multi-label one-versus-all loss based on max-entropy, between input <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> and target <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> of size <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mo stretchy="false">(</mo><mi>N</mi><mo separator="true">,</mo><mi>C</mi><mo stretchy="false">)</mo></mrow><annotation encoding="application/x-tex">(N, C)</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mopen">(</span><span class="mord mathdefault" style="margin-right:0.10903em;">N</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.07153em;">C</span><span class="mclose">)</span></span></span></span>
</span>.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.CosineEmbeddingLoss.html#torch.nn.CosineEmbeddingLoss" title="torch.nn.CosineEmbeddingLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.CosineEmbeddingLoss</span></code></a></p></td>
<td><p>Creates a criterion that measures the loss given input tensors <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>x</mi><mn>1</mn></msub></mrow><annotation encoding="application/x-tex">x_1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">1</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span>, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>x</mi><mn>2</mn></msub></mrow><annotation encoding="application/x-tex">x_2</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.58056em;vertical-align:-0.15em;"></span><span class="mord"><span class="mord mathdefault">x</span><span class="msupsub"><span class="vlist-t vlist-t2"><span class="vlist-r"><span class="vlist" style="height:0.30110799999999993em;"><span style="top:-2.5500000000000003em;margin-left:0em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">2</span></span></span></span><span class="vlist-s"></span></span><span class="vlist-r"><span class="vlist" style="height:0.15em;"><span></span></span></span></span></span></span></span></span></span>
</span> and a <cite>Tensor</cite> label <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> with values 1 or -1.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.MultiMarginLoss.html#torch.nn.MultiMarginLoss" title="torch.nn.MultiMarginLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.MultiMarginLoss</span></code></a></p></td>
<td><p>Creates a criterion that optimizes a multi-class classification hinge loss (margin-based loss) between input <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.43056em;vertical-align:0em;"></span><span class="mord mathdefault">x</span></span></span></span>
</span> (a 2D mini-batch <cite>Tensor</cite>) and output <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>y</mi></mrow><annotation encoding="application/x-tex">y</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.625em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span></span></span></span>
</span> (which is a 1D tensor of target class indices, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mn>0</mn><mo>≤</mo><mi>y</mi><mo>≤</mo><mtext>x.size</mtext><mo stretchy="false">(</mo><mn>1</mn><mo stretchy="false">)</mo><mo>−</mo><mn>1</mn></mrow><annotation encoding="application/x-tex">0 \leq y \leq \text{x.size}(1)-1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.78041em;vertical-align:-0.13597em;"></span><span class="mord">0</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span><span class="mrel">≤</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span></span><span class="base"><span class="strut" style="height:0.8304100000000001em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.03588em;">y</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span><span class="mrel">≤</span><span class="mspace" style="margin-right:0.2777777777777778em;"></span></span><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mord text"><span class="mord">x.size</span></span><span class="mopen">(</span><span class="mord">1</span><span class="mclose">)</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">−</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">1</span></span></span></span>
</span>):</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.TripletMarginLoss.html#torch.nn.TripletMarginLoss" title="torch.nn.TripletMarginLoss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.TripletMarginLoss</span></code></a></p></td>
<td><p>Creates a criterion that measures the triplet loss given an input tensors <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi><mn>1</mn></mrow><annotation encoding="application/x-tex">x1</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord mathdefault">x</span><span class="mord">1</span></span></span></span>
</span>, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi><mn>2</mn></mrow><annotation encoding="application/x-tex">x2</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord mathdefault">x</span><span class="mord">2</span></span></span></span>
</span>, <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi><mn>3</mn></mrow><annotation encoding="application/x-tex">x3</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord mathdefault">x</span><span class="mord">3</span></span></span></span>
</span> and a margin with a value greater than <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mn>0</mn></mrow><annotation encoding="application/x-tex">0</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">0</span></span></span></span>
</span>.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="vision-layers">
<h2><a class="toc-backref" href="#id1">Vision Layers</a><a class="headerlink" href="#vision-layers" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.PixelShuffle.html#torch.nn.PixelShuffle" title="torch.nn.PixelShuffle"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.PixelShuffle</span></code></a></p></td>
<td><p>Rearranges elements in a tensor of shape <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mo stretchy="false">(</mo><mo>∗</mo><mo separator="true">,</mo><mi>C</mi><mo>×</mo><msup><mi>r</mi><mn>2</mn></msup><mo separator="true">,</mo><mi>H</mi><mo separator="true">,</mo><mi>W</mi><mo stretchy="false">)</mo></mrow><annotation encoding="application/x-tex">(*, C \times r^2, H, W)</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mopen">(</span><span class="mord">∗</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.07153em;">C</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">×</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:1.064108em;vertical-align:-0.25em;"></span><span class="mord"><span class="mord mathdefault" style="margin-right:0.02778em;">r</span><span class="msupsub"><span class="vlist-t"><span class="vlist-r"><span class="vlist" style="height:0.8141079999999999em;"><span style="top:-3.063em;margin-right:0.05em;"><span class="pstrut" style="height:2.7em;"></span><span class="sizing reset-size6 size3 mtight"><span class="mord mtight">2</span></span></span></span></span></span></span></span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.08125em;">H</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.13889em;">W</span><span class="mclose">)</span></span></span></span>
</span> to a tensor of shape <span class="math"><span class="katex"><span class="katex-mathml"><math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mo stretchy="false">(</mo><mo>∗</mo><mo separator="true">,</mo><mi>C</mi><mo separator="true">,</mo><mi>H</mi><mo>×</mo><mi>r</mi><mo separator="true">,</mo><mi>W</mi><mo>×</mo><mi>r</mi><mo stretchy="false">)</mo></mrow><annotation encoding="application/x-tex">(*, C, H \times r, W \times r)</annotation></semantics></math></span><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mopen">(</span><span class="mord">∗</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.07153em;">C</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.08125em;">H</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">×</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:0.8777699999999999em;vertical-align:-0.19444em;"></span><span class="mord mathdefault" style="margin-right:0.02778em;">r</span><span class="mpunct">,</span><span class="mspace" style="margin-right:0.16666666666666666em;"></span><span class="mord mathdefault" style="margin-right:0.13889em;">W</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span><span class="mbin">×</span><span class="mspace" style="margin-right:0.2222222222222222em;"></span></span><span class="base"><span class="strut" style="height:1em;vertical-align:-0.25em;"></span><span class="mord mathdefault" style="margin-right:0.02778em;">r</span><span class="mclose">)</span></span></span></span>
</span>.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.Upsample.html#torch.nn.Upsample" title="torch.nn.Upsample"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.Upsample</span></code></a></p></td>
<td><p>Upsamples a given multi-channel 1D (temporal), 2D (spatial) or 3D (volumetric) data.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.UpsamplingNearest2d.html#torch.nn.UpsamplingNearest2d" title="torch.nn.UpsamplingNearest2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.UpsamplingNearest2d</span></code></a></p></td>
<td><p>Applies a 2D nearest neighbor upsampling to an input signal composed of several input channels.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.UpsamplingBilinear2d.html#torch.nn.UpsamplingBilinear2d" title="torch.nn.UpsamplingBilinear2d"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.UpsamplingBilinear2d</span></code></a></p></td>
<td><p>Applies a 2D bilinear upsampling to an input signal composed of several input channels.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="dataparallel-layers-multi-gpu-distributed">
<h2><a class="toc-backref" href="#id1">DataParallel Layers (multi-GPU, distributed)</a><a class="headerlink" href="#dataparallel-layers-multi-gpu-distributed" title="Permalink to this headline">¶</a></h2>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.DataParallel.html#torch.nn.DataParallel" title="torch.nn.DataParallel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.DataParallel</span></code></a></p></td>
<td><p>Implements data parallelism at the module level.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.parallel.DistributedDataParallel.html#torch.nn.parallel.DistributedDataParallel" title="torch.nn.parallel.DistributedDataParallel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">nn.parallel.DistributedDataParallel</span></code></a></p></td>
<td><p>Implements distributed data parallelism that is based on <code class="docutils literal notranslate"><span class="pre">torch.distributed</span></code> package at the module level.</p></td>
</tr>
</tbody>
</table>
</div>
<div class="section" id="utilities">
<h2><a class="toc-backref" href="#id1">Utilities</a><a class="headerlink" href="#utilities" title="Permalink to this headline">¶</a></h2>
<p>From the <code class="docutils literal notranslate"><span class="pre">torch.nn.utils</span></code> module</p>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.clip_grad_norm_.html#torch.nn.utils.clip_grad_norm_" title="torch.nn.utils.clip_grad_norm_"><code class="xref py py-obj docutils literal notranslate"><span class="pre">clip_grad_norm_</span></code></a></p></td>
<td><p>Clips gradient norm of an iterable of parameters.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.clip_grad_value_.html#torch.nn.utils.clip_grad_value_" title="torch.nn.utils.clip_grad_value_"><code class="xref py py-obj docutils literal notranslate"><span class="pre">clip_grad_value_</span></code></a></p></td>
<td><p>Clips gradient of an iterable of parameters at specified value.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.parameters_to_vector.html#torch.nn.utils.parameters_to_vector" title="torch.nn.utils.parameters_to_vector"><code class="xref py py-obj docutils literal notranslate"><span class="pre">parameters_to_vector</span></code></a></p></td>
<td><p>Convert parameters to one vector</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.vector_to_parameters.html#torch.nn.utils.vector_to_parameters" title="torch.nn.utils.vector_to_parameters"><code class="xref py py-obj docutils literal notranslate"><span class="pre">vector_to_parameters</span></code></a></p></td>
<td><p>Convert one vector to the parameters</p></td>
</tr>
</tbody>
</table>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.BasePruningMethod.html#torch.nn.utils.prune.BasePruningMethod" title="torch.nn.utils.prune.BasePruningMethod"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.BasePruningMethod</span></code></a></p></td>
<td><p>Abstract base class for creation of new pruning techniques.</p></td>
</tr>
</tbody>
</table>
<table class="longtable docutils colwidths-auto align-default">
<tbody>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.PruningContainer.html#torch.nn.utils.prune.PruningContainer" title="torch.nn.utils.prune.PruningContainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.PruningContainer</span></code></a></p></td>
<td><p>Container holding a sequence of pruning methods for iterative pruning.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.Identity.html#torch.nn.utils.prune.Identity" title="torch.nn.utils.prune.Identity"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.Identity</span></code></a></p></td>
<td><p>Utility pruning method that does not prune any units but generates the pruning parametrization with a mask of ones.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.RandomUnstructured.html#torch.nn.utils.prune.RandomUnstructured" title="torch.nn.utils.prune.RandomUnstructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.RandomUnstructured</span></code></a></p></td>
<td><p>Prune (currently unpruned) units in a tensor at random.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.L1Unstructured.html#torch.nn.utils.prune.L1Unstructured" title="torch.nn.utils.prune.L1Unstructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.L1Unstructured</span></code></a></p></td>
<td><p>Prune (currently unpruned) units in a tensor by zeroing out the ones with the lowest L1-norm.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.RandomStructured.html#torch.nn.utils.prune.RandomStructured" title="torch.nn.utils.prune.RandomStructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.RandomStructured</span></code></a></p></td>
<td><p>Prune entire (currently unpruned) channels in a tensor at random.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.LnStructured.html#torch.nn.utils.prune.LnStructured" title="torch.nn.utils.prune.LnStructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.LnStructured</span></code></a></p></td>
<td><p>Prune entire (currently unpruned) channels in a tensor based on their Ln-norm.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.CustomFromMask.html#torch.nn.utils.prune.CustomFromMask" title="torch.nn.utils.prune.CustomFromMask"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.CustomFromMask</span></code></a></p></td>
<td><p></p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.identity.html#torch.nn.utils.prune.identity" title="torch.nn.utils.prune.identity"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.identity</span></code></a></p></td>
<td><p>Applies pruning reparametrization to the tensor corresponding to the parameter called <code class="docutils literal notranslate"><span class="pre">name</span></code> in <code class="docutils literal notranslate"><span class="pre">module</span></code> without actually pruning any units.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.random_unstructured.html#torch.nn.utils.prune.random_unstructured" title="torch.nn.utils.prune.random_unstructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.random_unstructured</span></code></a></p></td>
<td><p>Prunes tensor corresponding to parameter called <code class="docutils literal notranslate"><span class="pre">name</span></code> in <code class="docutils literal notranslate"><span class="pre">module</span></code> by removing the specified <code class="docutils literal notranslate"><span class="pre">amount</span></code> of (currently unpruned) units selected at random.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.l1_unstructured.html#torch.nn.utils.prune.l1_unstructured" title="torch.nn.utils.prune.l1_unstructured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.l1_unstructured</span></code></a></p></td>
<td><p>Prunes tensor corresponding to parameter called <code class="docutils literal notranslate"><span class="pre">name</span></code> in <code class="docutils literal notranslate"><span class="pre">module</span></code> by removing the specified <cite>amount</cite> of (currently unpruned) units with the lowest L1-norm.</p></td>
</tr>
<tr class="row-odd"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.random_structured.html#torch.nn.utils.prune.random_structured" title="torch.nn.utils.prune.random_structured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.random_structured</span></code></a></p></td>
<td><p>Prunes tensor corresponding to parameter called <code class="docutils literal notranslate"><span class="pre">name</span></code> in <code class="docutils literal notranslate"><span class="pre">module</span></code> by removing the specified <code class="docutils literal notranslate"><span class="pre">amount</span></code> of (currently unpruned) channels along the specified <code class="docutils literal notranslate"><span class="pre">dim</span></code> selected at random.</p></td>
</tr>
<tr class="row-even"><td><p><a class="reference internal" href="generated/torch.nn.utils.prune.ln_structured.html#torch.nn.utils.prune.ln_structured" title="torch.nn.utils.prune.ln_structured"><code class="xref py py-obj docutils literal notranslate"><span class="pre">prune.ln_structured</span></code></a></p></td>