-
Notifications
You must be signed in to change notification settings - Fork 34
/
library.html
1019 lines (822 loc) · 94.7 KB
/
library.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 Library API — PyTorch main documentation</title>
<link rel="canonical" href="https://pytorch.org/docs/stable/library.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="_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="_static/cpp_theme.css" type="text/css" />
<link rel="stylesheet" href="_static/collapsible-lists/css/tree_view.css" type="text/css" />
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<!-- Google Tag Manager -->
<script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
})(window,document,'script','dataLayer','');</script>
<!-- End Google Tag Manager -->
<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">
<link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" 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 class="main-menu-item">
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
Learn
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/get-started">
<span class=dropdown-title>Get Started</span>
<p>Run PyTorch locally or get started quickly with one of the supported cloud platforms</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/tutorials">
<span class="dropdown-title">Tutorials</span>
<p>Whats new in PyTorch tutorials</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/tutorials/beginner/basics/intro.html">
<span class="dropdown-title">Learn the Basics</span>
<p>Familiarize yourself with PyTorch concepts and modules</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/tutorials/recipes/recipes_index.html">
<span class="dropdown-title">PyTorch Recipes</span>
<p>Bite-size, ready-to-deploy PyTorch code examples</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/tutorials/beginner/introyt.html">
<span class="dropdown-title">Intro to PyTorch - YouTube Series</span>
<p>Master PyTorch basics with our engaging YouTube tutorial series</p>
</a>
</div>
</div>
</li>
<li>
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
Ecosystem
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/ecosystem">
<span class="dropdown-title">Tools</span>
<p>Learn about the tools and frameworks in the PyTorch Ecosystem</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/#community-module">
<span class=dropdown-title>Community</span>
<p>Join the PyTorch developer community to contribute, learn, and get your questions answered</p>
</a>
<a class="nav-dropdown-item" href="https://discuss.pytorch.org/" target="_blank">
<span class=dropdown-title>Forums</span>
<p>A place to discuss PyTorch code, issues, install, research</p>
</a>
<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/ecosystem/contributor-awards-2023">
<span class="dropdown-title">Contributor Awards - 2023</span>
<p>Award winners announced at this year's PyTorch Conference</p>
</a>
</div>
</div>
</li>
<li>
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
Edge
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/edge">
<span class="dropdown-title">About PyTorch Edge</span>
<p>Build innovative and privacy-aware AI experiences for edge devices</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/executorch-overview">
<span class="dropdown-title">ExecuTorch</span>
<p>End-to-end solution for enabling on-device inference capabilities across mobile and edge devices</p>
</a>
</div>
</div>
</li>
<li class="main-menu-item">
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
Docs
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/docs/stable/index.html">
<span class="dropdown-title">PyTorch</span>
<p>Explore the documentation for comprehensive guidance on how to use PyTorch</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/pytorch-domains">
<span class="dropdown-title">PyTorch Domains</span>
<p>Read the PyTorch Domains documentation to learn more about domain-specific libraries</p>
</a>
</div>
</div>
</li>
<li>
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
Blogs & News
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/blog/">
<span class="dropdown-title">PyTorch Blog</span>
<p>Catch up on the latest technical news and happenings</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/community-blog">
<span class="dropdown-title">Community Blog</span>
<p>Stories from the PyTorch ecosystem</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/videos">
<span class="dropdown-title">Videos</span>
<p>Learn about the latest PyTorch tutorials, new, and more </p>
<a class="nav-dropdown-item" href="https://pytorch.org/community-stories">
<span class="dropdown-title">Community Stories</span>
<p>Learn how our community solves real, everyday machine learning problems with PyTorch</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/events">
<span class="dropdown-title">Events</span>
<p>Find events, webinars, and podcasts</p>
</a>
</div>
</li>
<li>
<div id="resourcesDropdownButton" data-toggle="resources-dropdown" class="resources-dropdown">
<a class="with-down-arrow">
About
</a>
<div class="resources-dropdown-menu">
<a class="nav-dropdown-item" href="https://pytorch.org/foundation">
<span class="dropdown-title">PyTorch Foundation</span>
<p>Learn more about the PyTorch Foundation</p>
</a>
<a class="nav-dropdown-item" href="https://pytorch.org/governing-board">
<span class="dropdown-title">Governing Board</span>
<p></p>
</a>
</div>
</div>
</li>
<li class="main-menu-item">
<div class="no-dropdown">
<a href="https://pytorch.org/join" data-cta="join">
Become a Member
</a>
</div>
</li>
<li>
<div class="main-menu-item">
<a href="https://github.com/pytorch/pytorch" class="github-icon">
</a>
</div>
</li>
<!--- TODO: This block adds the search icon to the nav bar. We will enable it later.
<li>
<div class="main-menu-item">
<a href="https://github.com/pytorch/pytorch" class="search-icon">
</a>
</div>
</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">
main
</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>
<ul>
<li class="toctree-l1"><a class="reference internal" href="installing.html">Installing C++ Distributions of PyTorch</a></li>
<li class="toctree-l1"><a class="reference internal" href="frontend.html">The C++ Frontend</a></li>
<li class="toctree-l1"><a class="reference internal" href="api/library_root.html">Library API</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Notes</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="notes/faq.html">FAQ</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/inference_mode.html">Inference Mode</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/maybe_owned.html">MaybeOwned<Tensor></a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/tensor_basics.html">Tensor Basics</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/tensor_creation.html">Tensor Creation API</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/tensor_cuda_stream.html">Tensor CUDA Stream API</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/tensor_indexing.html">Tensor Indexing API</a></li>
<li class="toctree-l1"><a class="reference internal" href="notes/versioning.html">Library Versioning</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 Library API</li>
<li class="pytorch-breadcrumbs-aside">
<a href="_sources/library.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">
<!-- Google Tag Manager (noscript) -->
<noscript><iframe src="https://www.googletagmanager.com/ns.html?id="
height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>
<!-- End Google Tag Manager (noscript) -->
<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-library-api">
<h1>Torch Library API<a class="headerlink" href="#torch-library-api" title="Permalink to this heading">¶</a></h1>
<p>The PyTorch C++ API provides capabilities for extending PyTorch’s core library
of operators with user defined operators and data types. Extensions implemented
using the Torch Library API are made available for use in both the PyTorch eager
API as well as in TorchScript.</p>
<p>For a tutorial style introduction to the library API, check out the
<a class="reference external" href="https://pytorch.org/tutorials/advanced/torch_script_custom_ops.html">Extending TorchScript with Custom C++ Operators</a>
tutorial.</p>
<div class="section" id="macros">
<h2>Macros<a class="headerlink" href="#macros" title="Permalink to this heading">¶</a></h2>
<dl class="cpp macro">
<dt class="sig sig-object cpp" id="c.TORCH_LIBRARY">
<span class="target" id="library_8h_1a0bd5fb09d25dfb58e750d712fc5afb84"></span><span class="sig-name descname"><span class="n"><span class="pre">TORCH_LIBRARY</span></span></span><span class="sig-paren">(</span><span class="n"><span class="pre">ns</span></span>, <span class="n"><span class="pre">m</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Macro for defining a function that will be run at static initialization time to define a library of operators in the namespace <code class="docutils literal notranslate"><span class="pre">ns</span></code> (must be a valid C++ identifier, no quotes). </p>
<p>Use this macro when you want to define a new set of custom operators that do not already exist in PyTorch.</p>
<p>Example usage:</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="n">TORCH_LIBRARY</span><span class="p">(</span><span class="n">myops</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="c1">// m is a torch::Library; methods on it will define</span>
<span class="w"> </span><span class="c1">// operators in the myops namespace</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="s">"add"</span><span class="p">,</span><span class="w"> </span><span class="n">add_impl</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<p>The <code class="docutils literal notranslate"><span class="pre">m</span></code> argument is bound to a <a class="reference internal" href="#classtorch_1_1_library"><span class="std std-ref">torch::Library</span></a> that is used to register operators. There may only be one <a class="reference internal" href="#library_8h_1a0bd5fb09d25dfb58e750d712fc5afb84"><span class="std std-ref">TORCH_LIBRARY()</span></a> for any given namespace. </p>
</dd></dl>
<dl class="cpp macro">
<dt class="sig sig-object cpp" id="c.TORCH_LIBRARY_IMPL">
<span class="target" id="library_8h_1a800b28569c0f7d0d9732bf8b6342d9f3"></span><span class="sig-name descname"><span class="n"><span class="pre">TORCH_LIBRARY_IMPL</span></span></span><span class="sig-paren">(</span><span class="n"><span class="pre">ns</span></span>, <span class="n"><span class="pre">k</span></span>, <span class="n"><span class="pre">m</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Macro for defining a function that will be run at static initialization time to define operator overrides for dispatch key <code class="docutils literal notranslate"><span class="pre">k</span></code> (must be an unqualified enum member of c10::DispatchKey) in namespace <code class="docutils literal notranslate"><span class="pre">ns</span></code> (must be a valid C++ identifer, no quotes). </p>
<p>Use this macro when you want to implement a preexisting set of custom operators on a new dispatch key (e.g., you want to provide CUDA implementations of already existing operators). One common usage pattern is to use <a class="reference internal" href="#library_8h_1a0bd5fb09d25dfb58e750d712fc5afb84"><span class="std std-ref">TORCH_LIBRARY()</span></a> to define schema for all new operators you want to define, and then use several <a class="reference internal" href="#library_8h_1a800b28569c0f7d0d9732bf8b6342d9f3"><span class="std std-ref">TORCH_LIBRARY_IMPL()</span></a> blocks to provide implementations of the operator for CPU, CUDA and Autograd.</p>
<p>In some cases, you need to define something that applies to all namespaces, not just one namespace (usually a fallback). In that case, use the reserved namespace _, e.g.,</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="n">TORCH_LIBRARY_IMPL</span><span class="p">(</span><span class="n">_</span><span class="p">,</span><span class="w"> </span><span class="n">XLA</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">fallback</span><span class="p">(</span><span class="n">xla_fallback</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<p>Example usage:</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="n">TORCH_LIBRARY_IMPL</span><span class="p">(</span><span class="n">myops</span><span class="p">,</span><span class="w"> </span><span class="n">CPU</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="c1">// m is a torch::Library; methods on it will define</span>
<span class="w"> </span><span class="c1">// CPU implementations of operators in the myops namespace.</span>
<span class="w"> </span><span class="c1">// It is NOT valid to call torch::Library::def()</span>
<span class="w"> </span><span class="c1">// in this context.</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">impl</span><span class="p">(</span><span class="s">"add"</span><span class="p">,</span><span class="w"> </span><span class="n">add_cpu_impl</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<p>If <code class="docutils literal notranslate"><span class="pre">add_cpu_impl</span></code> is an overloaded function, use a <code class="docutils literal notranslate"><span class="pre">static_cast</span></code> to specify which overload you want (by providing the full type). </p>
</dd></dl>
</div>
<div class="section" id="classes">
<h2>Classes<a class="headerlink" href="#classes" title="Permalink to this heading">¶</a></h2>
<dl class="cpp class">
<dt class="sig sig-object cpp" id="_CPPv4N5torch7LibraryE">
<span id="_CPPv3N5torch7LibraryE"></span><span id="_CPPv2N5torch7LibraryE"></span><span id="torch::Library"></span><span class="target" id="classtorch_1_1_library"></span><span class="k"><span class="pre">class</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Library</span></span></span><br /></dt>
<dd><p>This object provides the API for defining operators and providing implementations at dispatch keys. </p>
<p>Typically, a <a class="reference internal" href="#classtorch_1_1_library"><span class="std std-ref">torch::Library</span></a> is not allocated directly; instead it is created by the <a class="reference internal" href="#library_8h_1a0bd5fb09d25dfb58e750d712fc5afb84"><span class="std std-ref">TORCH_LIBRARY()</span></a> or <a class="reference internal" href="#library_8h_1a800b28569c0f7d0d9732bf8b6342d9f3"><span class="std std-ref">TORCH_LIBRARY_IMPL()</span></a> macros.</p>
<p>Most methods on <a class="reference internal" href="#classtorch_1_1_library"><span class="std std-ref">torch::Library</span></a> return a reference to itself, supporting method chaining.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Examples:</span>
<span class="n">TORCH_LIBRARY</span><span class="p">(</span><span class="n">torchvision</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="c1">// m is a torch::Library</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="s">"roi_align"</span><span class="p">,</span><span class="w"> </span><span class="p">...);</span>
<span class="w"> </span><span class="p">...</span>
<span class="p">}</span>
<span class="n">TORCH_LIBRARY_IMPL</span><span class="p">(</span><span class="n">aten</span><span class="p">,</span><span class="w"> </span><span class="n">XLA</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="c1">// m is a torch::Library</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">impl</span><span class="p">(</span><span class="s">"add"</span><span class="p">,</span><span class="w"> </span><span class="p">...);</span>
<span class="w"> </span><span class="p">...</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<div class="breathe-sectiondef docutils container">
<p class="breathe-sectiondef-title rubric" id="breathe-section-title-public-functions">Public Functions</p>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0ENR5torch7Library3defER7LibraryRR6SchemaRKNSt6vectorIN2at3TagEEE17_RegisterOrVerify">
<span id="_CPPv3I0ENR5torch7Library3defERR6SchemaRKNSt6vectorIN2at3TagEEE17_RegisterOrVerify"></span><span id="_CPPv2I0ENR5torch7Library3defERR6SchemaRKNSt6vectorIN2at3TagEEE17_RegisterOrVerify"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Schema</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_library_1a983e5fef5b234ad18cf389ed820e33c0"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">def</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I0ENR5torch7Library3defER7LibraryRR6SchemaRKNSt6vectorIN2at3TagEEE17_RegisterOrVerify" title="torch::Library::def::Schema"><span class="n"><span class="pre">Schema</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_schema</span></span>, <span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">vector</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">at</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">Tag</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">tags</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="p"><span class="pre">{</span></span><span class="p"><span class="pre">}</span></span>, <a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerifyE" title="torch::_RegisterOrVerify"><span class="n"><span class="pre">_RegisterOrVerify</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">rv</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerifyE" title="torch::_RegisterOrVerify"><span class="n"><span class="pre">_RegisterOrVerify</span></span></a><span class="p"><span class="pre">::</span></span><a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerify8REGISTERE" title="torch::_RegisterOrVerify::REGISTER"><span class="n"><span class="pre">REGISTER</span></span></a><span class="sig-paren">)</span><span class="w"> </span><span class="p"><span class="pre">&</span></span><br /></dt>
<dd><p>Declare an operator with a schema, but don’t provide any implementations for it. </p>
<p>You’re expected to then provide implementations using the <a class="reference internal" href="#classtorch_1_1_library_1a59330802ab9deaff247d32a22092bdfb"><span class="std std-ref">impl()</span></a> method. All template arguments are inferred.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Example:</span>
<span class="n">TORCH_LIBRARY</span><span class="p">(</span><span class="n">myops</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="s">"add(Tensor self, Tensor other) -> Tensor"</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>raw_schema</strong> – The schema of the operator to be defined. Typically, this is a <code class="docutils literal notranslate"><span class="pre">const</span> <span class="pre">char*</span></code> string literal, but any type accepted by <a class="reference internal" href="#group__torch-schema-overloads_1ga4c64ba9b665cbc07c31c94b469827a6c"><span class="std std-ref">torch::schema()</span></a> is accepted here.</p>
</dd>
</dl>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4N5torch7Library17set_python_moduleEPKcPKc">
<span id="_CPPv3N5torch7Library17set_python_moduleEPKcPKc"></span><span id="_CPPv2N5torch7Library17set_python_moduleEPKcPKc"></span><span id="torch::Library::set_python_module__cCP.cCP"></span><span class="target" id="classtorch_1_1_library_1a24d61ded9256fff1d15d79c22166eb6a"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">set_python_module</span></span></span><span class="sig-paren">(</span><span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">pymodule</span></span>, <span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">context</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="s"><span class="pre">""</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Declares that for all operators that are subsequently def’ed, their fake impls may be found in the given Python module (pymodule). </p>
<p>This registers some help text that is used if the fake impl cannot be found.</p>
<p>Args:<ul class="simple">
<li><p>pymodule: the python module</p></li>
<li><p>context: We may include this in the error message. </p></li>
</ul>
</p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4N5torch7Library20impl_abstract_pystubEPKcPKc">
<span id="_CPPv3N5torch7Library20impl_abstract_pystubEPKcPKc"></span><span id="_CPPv2N5torch7Library20impl_abstract_pystubEPKcPKc"></span><span id="torch::Library::impl_abstract_pystub__cCP.cCP"></span><span class="target" id="classtorch_1_1_library_1a06ecb7b3d0d82e0f0a5b7857dc70903b"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">impl_abstract_pystub</span></span></span><span class="sig-paren">(</span><span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">pymodule</span></span>, <span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">context</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="s"><span class="pre">""</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Deprecated; use set_python_module instead. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I00ENR5torch7Library3defER7LibraryRR12NameOrSchemaRR4FuncRKNSt6vectorIN2at3TagEEE">
<span id="_CPPv3I00ENR5torch7Library3defERR12NameOrSchemaRR4FuncRKNSt6vectorIN2at3TagEEE"></span><span id="_CPPv2I00ENR5torch7Library3defERR12NameOrSchemaRR4FuncRKNSt6vectorIN2at3TagEEE"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">NameOrSchema</span></span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_library_1ac399b254981eb4fa1d7f9cd335dfb32b"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">def</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I00ENR5torch7Library3defER7LibraryRR12NameOrSchemaRR4FuncRKNSt6vectorIN2at3TagEEE" title="torch::Library::def::NameOrSchema"><span class="n"><span class="pre">NameOrSchema</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_name_or_schema</span></span>, <a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I00ENR5torch7Library3defER7LibraryRR12NameOrSchemaRR4FuncRKNSt6vectorIN2at3TagEEE" title="torch::Library::def::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_f</span></span>, <span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">vector</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">at</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">Tag</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">tags</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="p"><span class="pre">{</span></span><span class="p"><span class="pre">}</span></span><span class="sig-paren">)</span><span class="w"> </span><span class="p"><span class="pre">&</span></span><br /></dt>
<dd><p>Define an operator for a schema and then register an implementation for it. </p>
<p>This is typically what you would use if you aren’t planning on making use of the dispatcher to structure your operator implementation. It’s roughly equivalent to calling <a class="reference internal" href="#classtorch_1_1_library_1a983e5fef5b234ad18cf389ed820e33c0"><span class="std std-ref">def()</span></a> and then <a class="reference internal" href="#classtorch_1_1_library_1a59330802ab9deaff247d32a22092bdfb"><span class="std std-ref">impl()</span></a>, but if you omit the schema of the operator, we will infer it from the type of your C++ function. All template arguments are inferred.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Example:</span>
<span class="n">TORCH_LIBRARY</span><span class="p">(</span><span class="n">myops</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="s">"add"</span><span class="p">,</span><span class="w"> </span><span class="n">add_fn</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>raw_name_or_schema</strong> – The schema of the operator to be defined, or just the name of the operator if the schema is to be inferred from <code class="docutils literal notranslate"><span class="pre">raw_f</span></code>. Typically a <code class="docutils literal notranslate"><span class="pre">const</span> <span class="pre">char*</span></code> literal. </p></li>
<li><p><strong>raw_f</strong> – The C++ function that implements this operator. Any valid constructor of <a class="reference internal" href="#classtorch_1_1_cpp_function"><span class="std std-ref">torch::CppFunction</span></a> is accepted here; typically you provide a function pointer or lambda.</p></li>
</ul>
</dd>
</dl>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I00ENR5torch7Library4implER7Library4NameRR4Func17_RegisterOrVerify">
<span id="_CPPv3I00ENR5torch7Library4implE4NameRR4Func17_RegisterOrVerify"></span><span id="_CPPv2I00ENR5torch7Library4implE4NameRR4Func17_RegisterOrVerify"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Name</span></span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_library_1a59330802ab9deaff247d32a22092bdfb"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">impl</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I00ENR5torch7Library4implER7Library4NameRR4Func17_RegisterOrVerify" title="torch::Library::impl::Name"><span class="n"><span class="pre">Name</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">name</span></span>, <a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I00ENR5torch7Library4implER7Library4NameRR4Func17_RegisterOrVerify" title="torch::Library::impl::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_f</span></span>, <a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerifyE" title="torch::_RegisterOrVerify"><span class="n"><span class="pre">_RegisterOrVerify</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">rv</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerifyE" title="torch::_RegisterOrVerify"><span class="n"><span class="pre">_RegisterOrVerify</span></span></a><span class="p"><span class="pre">::</span></span><a class="reference internal" href="api/enum_namespacetorch_1a86b58f71d08ebe107f81fdb87d3cf161.html#_CPPv4N5torch17_RegisterOrVerify8REGISTERE" title="torch::_RegisterOrVerify::REGISTER"><span class="n"><span class="pre">REGISTER</span></span></a><span class="sig-paren">)</span><span class="w"> </span><span class="p"><span class="pre">&</span></span><br /></dt>
<dd><p>Register an implementation for an operator. </p>
<p>You may register multiple implementations for a single operator at different dispatch keys (see <a class="reference internal" href="#group__torch-dispatch-overloads_1ga91120ca3079a024775f01ea37e7fcbd8"><span class="std std-ref">torch::dispatch()</span></a>). Implementations must have a corresponding declaration (from <a class="reference internal" href="#classtorch_1_1_library_1a983e5fef5b234ad18cf389ed820e33c0"><span class="std std-ref">def()</span></a>), otherwise they are invalid. If you plan to register multiple implementations, DO NOT provide a function implementation when you <a class="reference internal" href="#classtorch_1_1_library_1a983e5fef5b234ad18cf389ed820e33c0"><span class="std std-ref">def()</span></a> the operator.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Example:</span>
<span class="n">TORCH_LIBRARY_IMPL</span><span class="p">(</span><span class="n">myops</span><span class="p">,</span><span class="w"> </span><span class="n">CUDA</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">impl</span><span class="p">(</span><span class="s">"add"</span><span class="p">,</span><span class="w"> </span><span class="n">add_cuda</span><span class="p">);</span>
<span class="p">}</span>
</pre></div>
</div>
</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>name</strong> – The name of the operator to implement. Do NOT provide schema here. </p></li>
<li><p><strong>raw_f</strong> – The C++ function that implements this operator. Any valid constructor of <a class="reference internal" href="#classtorch_1_1_cpp_function"><span class="std std-ref">torch::CppFunction</span></a> is accepted here; typically you provide a function pointer or lambda.</p></li>
</ul>
</dd>
</dl>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0ENR5torch7Library8fallbackER7LibraryRR4Func">
<span id="_CPPv3I0ENR5torch7Library8fallbackERR4Func"></span><span id="_CPPv2I0ENR5torch7Library8fallbackERR4Func"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_library_1a662da4bc1e19e98ff41d3e483d0865bc"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4N5torch7LibraryE" title="torch::Library"><span class="n"><span class="pre">Library</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="sig-name descname"><span class="n"><span class="pre">fallback</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_library.html#_CPPv4I0ENR5torch7Library8fallbackER7LibraryRR4Func" title="torch::Library::fallback::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_f</span></span><span class="sig-paren">)</span><span class="w"> </span><span class="p"><span class="pre">&</span></span><br /></dt>
<dd><p>Register a fallback implementation for all operators which will be used if there is not a specific implementation for an operator available. </p>
<p>There MUST be a DispatchKey associated with a fallback; e.g., only call this from <a class="reference internal" href="#library_8h_1a800b28569c0f7d0d9732bf8b6342d9f3"><span class="std std-ref">TORCH_LIBRARY_IMPL()</span></a> with namespace <code class="docutils literal notranslate"><span class="pre">_</span></code>.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Example:</span>
<span class="n">TORCH_LIBRARY_IMPL</span><span class="p">(</span><span class="n">_</span><span class="p">,</span><span class="w"> </span><span class="n">AutogradXLA</span><span class="p">,</span><span class="w"> </span><span class="n">m</span><span class="p">)</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="c1">// If there is not a kernel explicitly registered</span>
<span class="w"> </span><span class="c1">// for AutogradXLA, fallthrough to the next</span>
<span class="w"> </span><span class="c1">// available kernel</span>
<span class="w"> </span><span class="n">m</span><span class="p">.</span><span class="n">fallback</span><span class="p">(</span><span class="n">torch</span><span class="o">::</span><span class="n">CppFunction</span><span class="o">::</span><span class="n">makeFallthrough</span><span class="p">());</span>
<span class="p">}</span>
<span class="c1">// See aten/src/ATen/core/dispatch/backend_fallback_test.cpp</span>
<span class="c1">// for a full example of boxed fallback</span>
</pre></div>
</div>
</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>raw_f</strong> – The function that implements the fallback. Unboxed functions typically do not work as fallback functions, as fallback functions must work for every operator (even though they have varying type signatures). Typical arguments are <a class="reference internal" href="#classtorch_1_1_cpp_function_1a69be67b7ac572937f1786a81905ef2c7"><span class="std std-ref">CppFunction::makeFallthrough()</span></a> or <a class="reference internal" href="#classtorch_1_1_cpp_function_1a6591d6e48059314a6b0789b68c9abe2e"><span class="std std-ref">CppFunction::makeFromBoxedFunction()</span></a></p>
</dd>
</dl>
</dd></dl>
</div>
</dd></dl>
<dl class="cpp class">
<dt class="sig sig-object cpp" id="_CPPv4N5torch11CppFunctionE">
<span id="_CPPv3N5torch11CppFunctionE"></span><span id="_CPPv2N5torch11CppFunctionE"></span><span id="torch::CppFunction"></span><span class="target" id="classtorch_1_1_cpp_function"></span><span class="k"><span class="pre">class</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">CppFunction</span></span></span><br /></dt>
<dd><p>Represents a C++ function that implements an operator. </p>
<p>Most users won’t interact directly with this class, except via error messages: the constructors this function define the set of permissible “function”-like things you can bind via the interface.</p>
<p>This class erases the type of the passed in function, but durably records the type via an inferred schema for the function. </p>
<div class="breathe-sectiondef docutils container">
<p class="breathe-sectiondef-title rubric" id="breathe-section-title-public-functions">Public Functions</p>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0EN5torch11CppFunction11CppFunctionEP4FuncNSt11enable_if_tIN3c104guts16is_function_typeI4FuncE5valueENSt9nullptr_tEEE">
<span id="_CPPv3I0EN5torch11CppFunction11CppFunctionEP4FuncNSt11enable_if_tIN3c104guts16is_function_typeI4FuncE5valueENSt9nullptr_tEEE"></span><span id="_CPPv2I0EN5torch11CppFunction11CppFunctionEP4FuncNSt11enable_if_tIN3c104guts16is_function_typeI4FuncE5valueENSt9nullptr_tEEE"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a2d8fc800e4cbe1634abfe61a464ed748"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="k"><span class="pre">explicit</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">CppFunction</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionEP4FuncNSt11enable_if_tIN3c104guts16is_function_typeI4FuncE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">f</span></span>, <span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">enable_if_t</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">guts</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">is_function_type</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionEP4FuncNSt11enable_if_tIN3c104guts16is_function_typeI4FuncE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::Func"><span class="n"><span class="pre">Func</span></span></a><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">nullptr_t</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">nullptr</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>This overload accepts function pointers, e.g., <code class="docutils literal notranslate"><span class="pre">CppFunction(&add_impl)</span></code> </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0EN5torch11CppFunction11CppFunctionE7FuncPtrNSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEE">
<span id="_CPPv3I0EN5torch11CppFunction11CppFunctionE7FuncPtrNSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEE"></span><span id="_CPPv2I0EN5torch11CppFunction11CppFunctionE7FuncPtrNSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEE"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">FuncPtr</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a2e4b990ec1d900a80e5b5558593d9f74"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="k"><span class="pre">explicit</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">CppFunction</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionE7FuncPtrNSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">f</span></span>, <span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">enable_if_t</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">is_compile_time_function_pointer</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionE7FuncPtrNSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">nullptr_t</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">nullptr</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>This overload accepts compile time function pointers, e.g., <code class="docutils literal notranslate"><span class="pre">CppFunction(TORCH_FN(add_impl))</span></code> </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0EN5torch11CppFunction11CppFunctionERR6LambdaNSt11enable_if_tIN3c104guts10is_functorINSt7decay_tI6LambdaEEE5valueENSt9nullptr_tEEE">
<span id="_CPPv3I0EN5torch11CppFunction11CppFunctionERR6LambdaNSt11enable_if_tIN3c104guts10is_functorINSt7decay_tI6LambdaEEE5valueENSt9nullptr_tEEE"></span><span id="_CPPv2I0EN5torch11CppFunction11CppFunctionERR6LambdaNSt11enable_if_tIN3c104guts10is_functorINSt7decay_tI6LambdaEEE5valueENSt9nullptr_tEEE"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Lambda</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a93095d8f4285805a6984476ef5ccae56"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="k"><span class="pre">explicit</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">CppFunction</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionERR6LambdaNSt11enable_if_tIN3c104guts10is_functorINSt7decay_tI6LambdaEEE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::Lambda"><span class="n"><span class="pre">Lambda</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">f</span></span>, <span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">enable_if_t</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">guts</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">is_functor</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">decay_t</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction11CppFunctionERR6LambdaNSt11enable_if_tIN3c104guts10is_functorINSt7decay_tI6LambdaEEE5valueENSt9nullptr_tEEE" title="torch::CppFunction::CppFunction::Lambda"><span class="n"><span class="pre">Lambda</span></span></a><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">nullptr_t</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">nullptr</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>This overload accepts lambdas, e.g., <code class="docutils literal notranslate"><a class="reference internal" href="#classtorch_1_1_cpp_function"><span class="std std-ref"><span class="pre">CppFunction</span></span></a><span class="pre">([](const</span> <span class="pre">Tensor&</span> <span class="pre">self)</span> <span class="pre">{</span> <span class="pre">...</span> </code></p>
<p>}) </p>
</dd></dl>
</div>
<div class="breathe-sectiondef docutils container">
<p class="breathe-sectiondef-title rubric" id="breathe-section-title-public-static-functions">Public Static Functions</p>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4N5torch11CppFunction15makeFallthroughEv">
<span id="_CPPv3N5torch11CppFunction15makeFallthroughEv"></span><span id="_CPPv2N5torch11CppFunction15makeFallthroughEv"></span><span id="torch::CppFunction::makeFallthrough"></span><span class="target" id="classtorch_1_1_cpp_function_1a69be67b7ac572937f1786a81905ef2c7"></span><span class="k"><span class="pre">static</span></span><span class="w"> </span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4N5torch11CppFunctionE" title="torch::CppFunction"><span class="n"><span class="pre">CppFunction</span></span></a><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">makeFallthrough</span></span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><br /></dt>
<dd><p>This creates a fallthrough function. </p>
<p>Fallthrough functions immediately redispatch to the next available dispatch key, but are implemented more efficiently than a hand written function done in the same way. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I_PN3c1011BoxedKernel19BoxedKernelFunctionEEN5torch11CppFunction21makeFromBoxedFunctionE11CppFunctionv">
<span id="_CPPv3I_PN3c1011BoxedKernel19BoxedKernelFunctionEEN5torch11CppFunction21makeFromBoxedFunctionEv"></span><span id="_CPPv2I_PN3c1011BoxedKernel19BoxedKernelFunctionEEN5torch11CppFunction21makeFromBoxedFunctionEv"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">BoxedKernel</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">BoxedKernelFunction</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="sig-name descname"><span class="n"><span class="pre">func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a6591d6e48059314a6b0789b68c9abe2e"></span><span class="k"><span class="pre">static</span></span><span class="w"> </span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4N5torch11CppFunctionE" title="torch::CppFunction"><span class="n"><span class="pre">CppFunction</span></span></a><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">makeFromBoxedFunction</span></span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><br /></dt>
<dd><p>Create a function from a boxed kernel function with signature <code class="docutils literal notranslate"><span class="pre">void(const</span> <span class="pre">OperatorHandle&,</span> <span class="pre">Stack*)</span></code>; i.e., they receive a stack of arguments in a boxed calling convention, rather than in the native C++ calling convention. </p>
<p>Boxed functions are typically only used to register backend fallbacks via <a class="reference internal" href="#classtorch_1_1_library_1a662da4bc1e19e98ff41d3e483d0865bc"><span class="std std-ref">torch::Library::fallback()</span></a>. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0EN5torch11CppFunction20makeFromBoxedFunctorE11CppFunctionNSt10unique_ptrI13KernelFunctorEE">
<span id="_CPPv3I0EN5torch11CppFunction20makeFromBoxedFunctorENSt10unique_ptrI13KernelFunctorEE"></span><span id="_CPPv2I0EN5torch11CppFunction20makeFromBoxedFunctorENSt10unique_ptrI13KernelFunctorEE"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">class</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">KernelFunctor</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a5a49c17224c86b300a918ef75d70f51b"></span><span class="k"><span class="pre">static</span></span><span class="w"> </span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4N5torch11CppFunctionE" title="torch::CppFunction"><span class="n"><span class="pre">CppFunction</span></span></a><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">makeFromBoxedFunctor</span></span></span><span class="sig-paren">(</span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">unique_ptr</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0EN5torch11CppFunction20makeFromBoxedFunctorE11CppFunctionNSt10unique_ptrI13KernelFunctorEE" title="torch::CppFunction::makeFromBoxedFunctor::KernelFunctor"><span class="n"><span class="pre">KernelFunctor</span></span></a><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="n sig-param"><span class="pre">kernelFunctor</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Create a function from a boxed kernel functor which defines <code class="docutils literal notranslate"><span class="pre">operator()(const</span> <span class="pre">OperatorHandle&,</span> <span class="pre">DispatchKeySet,</span> <span class="pre">Stack*)</span></code> (receiving arguments from boxed calling convention) and inherits from <code class="docutils literal notranslate"><span class="pre">c10::OperatorKernel</span></code>. </p>
<p>Unlike makeFromBoxedFunction, functions registered in this way can also carry additional state which is managed by the functor; this is useful if you’re writing an adapter to some other implementation, e.g., a Python callable, which is dynamically associated with the registered kernel. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0_NSt11enable_if_tIN3c104guts16is_function_typeI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunctionP7FuncPtr">
<span id="_CPPv3I0_NSt11enable_if_tIN3c104guts16is_function_typeI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionEP7FuncPtr"></span><span id="_CPPv2I0_NSt11enable_if_tIN3c104guts16is_function_typeI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionEP7FuncPtr"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">FuncPtr</span></span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">enable_if_t</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">guts</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">is_function_type</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0_NSt11enable_if_tIN3c104guts16is_function_typeI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunctionP7FuncPtr" title="torch::CppFunction::makeFromUnboxedFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">nullptr_t</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">nullptr</span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1aba8dfeff023d6bd13d0fa3f72cb7e648"></span><span class="k"><span class="pre">static</span></span><span class="w"> </span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4N5torch11CppFunctionE" title="torch::CppFunction"><span class="n"><span class="pre">CppFunction</span></span></a><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">makeFromUnboxedFunction</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0_NSt11enable_if_tIN3c104guts16is_function_typeI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunctionP7FuncPtr" title="torch::CppFunction::makeFromUnboxedFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">f</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Create a function from an unboxed kernel function. </p>
<p>This is typically used to register common operators. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0_NSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunction7FuncPtr">
<span id="_CPPv3I0_NSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE7FuncPtr"></span><span id="_CPPv2I0_NSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE7FuncPtr"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">FuncPtr</span></span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">enable_if_t</span></span><span class="p"><span class="pre"><</span></span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">is_compile_time_function_pointer</span></span><span class="p"><span class="pre"><</span></span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0_NSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunction7FuncPtr" title="torch::CppFunction::makeFromUnboxedFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="p"><span class="pre">></span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">value</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="n"><span class="pre">std</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">nullptr_t</span></span><span class="p"><span class="pre">></span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">nullptr</span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="classtorch_1_1_cpp_function_1a8432c854c1e080e0a904a1e4e7e93d11"></span><span class="k"><span class="pre">static</span></span><span class="w"> </span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4N5torch11CppFunctionE" title="torch::CppFunction"><span class="n"><span class="pre">CppFunction</span></span></a><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">makeFromUnboxedFunction</span></span></span><span class="sig-paren">(</span><a class="reference internal" href="api/classtorch_1_1_cpp_function.html#_CPPv4I0_NSt11enable_if_tIN3c1032is_compile_time_function_pointerI7FuncPtrE5valueENSt9nullptr_tEEEEN5torch11CppFunction23makeFromUnboxedFunctionE11CppFunction7FuncPtr" title="torch::CppFunction::makeFromUnboxedFunction::FuncPtr"><span class="n"><span class="pre">FuncPtr</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">f</span></span><span class="sig-paren">)</span><br /></dt>
<dd><p>Create a function from a compile time unboxed kernel function pointer. </p>
<p>This is typically used to register common operators. Compile time function pointers can be used to allow the compiler to optimize (e.g. inline) calls to it. </p>
</dd></dl>
</div>
</dd></dl>
</div>
<div class="section" id="functions">
<h2>Functions<a class="headerlink" href="#functions" title="Permalink to this heading">¶</a></h2>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0E8dispatch11CppFunctionN3c1011DispatchKeyERR4Func">
<span id="_CPPv3I0E8dispatchN3c1011DispatchKeyERR4Func"></span><span id="_CPPv2I0E8dispatchN3c1011DispatchKeyERR4Func"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="group__torch-dispatch-overloads_1ga91120ca3079a024775f01ea37e7fcbd8"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="n"><span class="pre">CppFunction</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">dispatch</span></span></span><span class="sig-paren">(</span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">DispatchKey</span></span><span class="w"> </span><span class="n sig-param"><span class="pre">k</span></span>, <a class="reference internal" href="#_CPPv4I0E8dispatch11CppFunctionN3c1011DispatchKeyERR4Func" title="dispatch::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_f</span></span><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv4I0E8dispatch11CppFunctionN3c1011DispatchKeyERR4Func" title="Permalink to this definition">¶</a><br /></dt>
<dd><p>Create a <a class="reference internal" href="#classtorch_1_1_cpp_function"><span class="std std-ref">torch::CppFunction</span></a> which is associated with a specific dispatch key. </p>
<p>torch::CppFunctions that are tagged with a c10::DispatchKey don’t get invoked unless the dispatcher determines that this particular c10::DispatchKey is the one that should be dispatched to.</p>
<p>This function is generally not used directly, instead, prefer using <a class="reference internal" href="#library_8h_1a800b28569c0f7d0d9732bf8b6342d9f3"><span class="std std-ref">TORCH_LIBRARY_IMPL()</span></a>, which will implicitly set the c10::DispatchKey for all registration calls inside of its body. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv4I0E8dispatch11CppFunctionN3c1010DeviceTypeERR4Func">
<span id="_CPPv3I0E8dispatchN3c1010DeviceTypeERR4Func"></span><span id="_CPPv2I0E8dispatchN3c1010DeviceTypeERR4Func"></span><span class="k"><span class="pre">template</span></span><span class="p"><span class="pre"><</span></span><span class="k"><span class="pre">typename</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">Func</span></span></span><span class="p"><span class="pre">></span></span><br /><span class="target" id="group__torch-dispatch-overloads_1ga2037ceb48b4bcf8a09f3f2def4e1b545"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="n"><span class="pre">CppFunction</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">dispatch</span></span></span><span class="sig-paren">(</span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><a class="reference internal" href="api/enum_namespacec10_1a815bc73d9ef8591e4a92a70311b71697.html#_CPPv4N3c1010DeviceTypeE" title="c10::DeviceType"><span class="n"><span class="pre">DeviceType</span></span></a><span class="w"> </span><span class="n sig-param"><span class="pre">type</span></span>, <a class="reference internal" href="#_CPPv4I0E8dispatch11CppFunctionN3c1010DeviceTypeERR4Func" title="dispatch::Func"><span class="n"><span class="pre">Func</span></span></a><span class="w"> </span><span class="p"><span class="pre">&</span></span><span class="p"><span class="pre">&</span></span><span class="n sig-param"><span class="pre">raw_f</span></span><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv4I0E8dispatch11CppFunctionN3c1010DeviceTypeERR4Func" title="Permalink to this definition">¶</a><br /></dt>
<dd><p>Convenience overload of <a class="reference internal" href="#group__torch-dispatch-overloads_1ga91120ca3079a024775f01ea37e7fcbd8"><span class="std std-ref">dispatch()</span></a> which accepts <a class="reference internal" href="api/enum_namespacec10_1a815bc73d9ef8591e4a92a70311b71697.html#namespacec10_1a815bc73d9ef8591e4a92a70311b71697"><span class="std std-ref">c10::DeviceType</span></a>. </p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv46schemaPKcN3c1017AliasAnalysisKindEb">
<span id="_CPPv36schemaPKcN3c1017AliasAnalysisKindEb"></span><span id="_CPPv26schemaPKcN3c1017AliasAnalysisKindEb"></span><span id="schema__cCP.c10::AliasAnalysisKind.b"></span><span class="target" id="group__torch-schema-overloads_1ga4c64ba9b665cbc07c31c94b469827a6c"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">FunctionSchema</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">schema</span></span></span><span class="sig-paren">(</span><span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">str</span></span>, <span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">AliasAnalysisKind</span></span><span class="w"> </span><span class="n sig-param"><span class="pre">k</span></span>, <span class="kt"><span class="pre">bool</span></span><span class="w"> </span><span class="n sig-param"><span class="pre">allow_typevars</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">false</span></span><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv46schemaPKcN3c1017AliasAnalysisKindEb" title="Permalink to this definition">¶</a><br /></dt>
<dd><p>Construct a c10::FunctionSchema from a string, with an explicitly specified c10::AliasAnalysisKind. </p>
<p>Ordinarily, schemas are simply passed in as strings, but if you need to specify a custom alias analysis, you can replace the string with a call to this function.</p>
<p><div class="highlight-cpp notranslate"><div class="highlight"><pre><span></span><span class="c1">// Default alias analysis (FROM_SCHEMA)</span>
<span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="s">"def3(Tensor self) -> Tensor"</span><span class="p">);</span>
<span class="c1">// Pure function alias analysis</span>
<span class="n">m</span><span class="p">.</span><span class="n">def</span><span class="p">(</span><span class="n">torch</span><span class="o">::</span><span class="n">schema</span><span class="p">(</span><span class="s">"def3(Tensor self) -> Tensor"</span><span class="p">,</span>
<span class="n">c10</span><span class="o">::</span><span class="n">AliasAnalysisKind</span><span class="o">::</span><span class="n">PURE_FUNCTION</span><span class="p">));</span>
</pre></div>
</div>
</p>
</dd></dl>
<dl class="cpp function">
<dt class="sig sig-object cpp" id="_CPPv46schemaPKcb">
<span id="_CPPv36schemaPKcb"></span><span id="_CPPv26schemaPKcb"></span><span id="schema__cCP.b"></span><span class="target" id="group__torch-schema-overloads_1ga5f4f77472f06677ffe52db6602d1dfd0"></span><span class="k"><span class="pre">inline</span></span><span class="w"> </span><span class="n"><span class="pre">c10</span></span><span class="p"><span class="pre">::</span></span><span class="n"><span class="pre">FunctionSchema</span></span><span class="w"> </span><span class="sig-name descname"><span class="n"><span class="pre">schema</span></span></span><span class="sig-paren">(</span><span class="k"><span class="pre">const</span></span><span class="w"> </span><span class="kt"><span class="pre">char</span></span><span class="w"> </span><span class="p"><span class="pre">*</span></span><span class="n sig-param"><span class="pre">s</span></span>, <span class="kt"><span class="pre">bool</span></span><span class="w"> </span><span class="n sig-param"><span class="pre">allow_typevars</span></span><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="k"><span class="pre">false</span></span><span class="sig-paren">)</span><a class="headerlink" href="#_CPPv46schemaPKcb" title="Permalink to this definition">¶</a><br /></dt>
<dd><p>Function schemas can be directly constructed from string literals. </p>
</dd></dl>
</div>
</div>
</article>
</div>
<footer>
<hr>
<div role="contentinfo">
<p>
© Copyright 2024, PyTorch Contributors.
</p>
</div>
<div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</div>
</footer>
</div>
</div>
<div class="pytorch-content-right" id="pytorch-content-right">
<div class="pytorch-right-menu" id="pytorch-right-menu">
<div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
<ul>
<li><a class="reference internal" href="#">Torch Library API</a><ul>
<li><a class="reference internal" href="#macros">Macros</a></li>
<li><a class="reference internal" href="#classes">Classes</a></li>
<li><a class="reference internal" href="#functions">Functions</a></li>
</ul>
</li>
</ul>
</div>
</div>
</div>
</section>
</div>
<script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
<script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
<script src="_static/jquery.js"></script>
<script src="_static/underscore.js"></script>
<script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
<script src="_static/doctools.js"></script>
<script src="_static/sphinx_highlight.js"></script>
<script src="_static/collapsible-lists/js/CollapsibleLists.compressed.js"></script>
<script src="_static/collapsible-lists/js/apply-collapsible-lists.js"></script>
<script type="text/javascript" src="_static/js/vendor/popper.min.js"></script>
<script type="text/javascript" src="_static/js/vendor/bootstrap.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
<script type="text/javascript" src="_static/js/theme.js"></script>
<script type="text/javascript">
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
<!-- Begin Footer -->
<div class="container-fluid docs-tutorials-resources" id="docs-tutorials-resources">
<div class="container">
<div class="row">
<div class="col-md-4 text-center">
<h2>Docs</h2>
<p>Access comprehensive developer documentation for PyTorch</p>
<a class="with-right-arrow" href="https://pytorch.org/docs/stable/index.html">View Docs</a>
</div>
<div class="col-md-4 text-center">
<h2>Tutorials</h2>
<p>Get in-depth tutorials for beginners and advanced developers</p>
<a class="with-right-arrow" href="https://pytorch.org/tutorials">View Tutorials</a>
</div>
<div class="col-md-4 text-center">
<h2>Resources</h2>
<p>Find development resources and get your questions answered</p>
<a class="with-right-arrow" href="https://pytorch.org/resources">View Resources</a>
</div>
</div>
</div>
</div>
<footer class="site-footer">
<div class="container footer-container">
<div class="footer-logo-wrapper">
<a href="https://pytorch.org/" class="footer-logo"></a>
</div>
<div class="footer-links-wrapper">
<div class="footer-links-col">
<ul>
<li class="list-title"><a href="https://pytorch.org/">PyTorch</a></li>
<li><a href="https://pytorch.org/get-started">Get Started</a></li>
<li><a href="https://pytorch.org/features">Features</a></li>
<li><a href="https://pytorch.org/ecosystem">Ecosystem</a></li>
<li><a href="https://pytorch.org/blog/">Blog</a></li>
<li><a href="https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md">Contributing</a></li>
</ul>
</div>
<div class="footer-links-col">
<ul>
<li class="list-title"><a href="https://pytorch.org/resources">Resources</a></li>
<li><a href="https://pytorch.org/tutorials">Tutorials</a></li>
<li><a href="https://pytorch.org/docs/stable/index.html">Docs</a></li>
<li><a href="https://discuss.pytorch.org" target="_blank">Discuss</a></li>
<li><a href="https://github.com/pytorch/pytorch/issues" target="_blank">Github Issues</a></li>
<li><a href="https://pytorch.org/assets/brand-guidelines/PyTorch-Brand-Guidelines.pdf" target="_blank">Brand Guidelines</a></li>
</ul>
</div>
<div class="footer-links-col">
<ul>
<li class="list-title">Stay up to date</li>
<li><a href="https://www.facebook.com/pytorch" target="_blank">Facebook</a></li>
<li><a href="https://twitter.com/pytorch" target="_blank">Twitter</a></li>
<li><a href="https://www.youtube.com/pytorch" target="_blank">YouTube</a></li>
<li><a href="https://www.linkedin.com/company/pytorch" target="_blank">LinkedIn</a></li>
</ul>
</div>
<div class="footer-links-col">
<ul>
<li class="list-title">PyTorch Podcasts</li>
<li><a href="https://open.spotify.com/show/6UzHKeiy368jKfQMKKvJY5" target="_blank">Spotify</a></li>
<li><a href="https://podcasts.apple.com/us/podcast/pytorch-developer-podcast/id1566080008" target="_blank">Apple</a></li>
<li><a href="https://www.google.com/podcasts?feed=aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9PQjVGa0lsOA%3D%3D" target="_blank">Google</a></li>
<li><a href="https://music.amazon.com/podcasts/7a4e6f0e-26c2-49e9-a478-41bd244197d0/PyTorch-Developer-Podcast?" target="_blank">Amazon</a></li>
</ul>
</div>
</div>
<div class="privacy-policy">
<ul>
<li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/terms/" target="_blank">Terms</a></li>
<li class="privacy-policy-links">|</li>
<li class="privacy-policy-links"><a href="https://www.linuxfoundation.org/privacy-policy/" target="_blank">Privacy</a></li>
</ul>
</div>
<div class="copyright">
<p>© Copyright The Linux Foundation. The PyTorch Foundation is a project of The Linux Foundation.
For web site terms of use, trademark policy and other policies applicable to The PyTorch Foundation please see
<a href="https://www.linuxfoundation.org/policies/">www.linuxfoundation.org/policies/</a>. The PyTorch Foundation supports the PyTorch open source
project, which has been established as PyTorch Project a Series of LF Projects, LLC. For policies applicable to the PyTorch Project a Series of LF Projects, LLC,
please see <a href="https://www.lfprojects.org/policies/">www.lfprojects.org/policies/</a>.</p>
</div>
</div>
</footer>
<div class="cookie-banner-wrapper">
<div class="container">
<p class="gdpr-notice">To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: <a href="https://www.facebook.com/policies/cookies/">Cookies Policy</a>.</p>
<img class="close-button" src="_static/images/pytorch-x.svg">
</div>
</div>
<!-- End Footer -->
<!-- Begin Mobile Menu -->
<div class="mobile-main-menu">
<div class="container-fluid">
<div class="container">
<div class="mobile-main-menu-header-container">
<a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
<a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
</div>
</div>
</div>
<div class="mobile-main-menu-links-container">
<div class="main-menu">
<ul>
<li class="resources-mobile-menu-title">
<a>Learn</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/get-started">Get Started</a>
</li>
<li>
<a href="https://pytorch.org/tutorials">Tutorials</a>
</li>
<li>
<a href="https://pytorch.org/tutorials/beginner/basics/intro.html">Learn the Basics</a>
</li>
<li>
<a href="https://pytorch.org/tutorials/recipes/recipes_index.html">PyTorch Recipes</a>
</li>
<li>
<a href="https://pytorch.org/tutorials/beginner/introyt.html">Introduction to PyTorch - YouTube Series</a>
</li>
</ul>
<li class="resources-mobile-menu-title">
<a>Ecosystem</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/ecosystem">Tools</a>
</li>
<li>
<a href="https://pytorch.org/#community-module">Community</a>
</li>
<li>
<a href="https://discuss.pytorch.org/">Forums</a>
</li>
<li>
<a href="https://pytorch.org/resources">Developer Resources</a>
</li>
<li>
<a href="https://pytorch.org/ecosystem/contributor-awards-2023">Contributor Awards - 2023</a>
</li>
</ul>
<li class="resources-mobile-menu-title">
<a>Edge</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/edge">About PyTorch Edge</a>
</li>
<li>
<a href="https://pytorch.org/executorch-overview">ExecuTorch</a>
</li>
</ul>
<li class="resources-mobile-menu-title">
<a>Docs</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/docs/stable/index.html">PyTorch</a>
</li>
<li>
<a href="https://pytorch.org/pytorch-domains">PyTorch Domains</a>
</li>
</ul>
<li class="resources-mobile-menu-title">
<a>Blog & News</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/blog/">PyTorch Blog</a>
</li>
<li>
<a href="https://pytorch.org/community-blog">Community Blog</a>
</li>
<li>
<a href="https://pytorch.org/videos">Videos</a>
</li>
<li>
<a href="https://pytorch.org/community-stories">Community Stories</a>
</li>
<li>
<a href="https://pytorch.org/events">Events</a>
</li>
</ul>
<li class="resources-mobile-menu-title">
<a>About</a>
</li>
<ul class="resources-mobile-menu-items">
<li>
<a href="https://pytorch.org/foundation">PyTorch Foundation</a>
</li>
<li>
<a href="https://pytorch.org/governing-board">Governing Board</a>
</li>
</ul>
</ul>
</div>
</div>
</div>
<!-- End Mobile Menu -->
<script type="text/javascript" src="_static/js/vendor/anchor.min.js"></script>