Repository navigation
Expand file tree
/
Copy paththirdaibolt.html
More file actions
1408 lines (1339 loc) · 137 KB
/
Copy paththirdaibolt.html
File metadata and controls
1408 lines (1339 loc) · 137 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html class="writer-html5" lang="en" data-content_root="./">
<head>
<meta charset="utf-8" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>thirdai.bolt — thirdai documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=80d5e7a1" />
<link rel="stylesheet" type="text/css" href="_static/css/theme.css?v=e59714d7" />
<script src="_static/jquery.js?v=5d32c60e"></script>
<script src="_static/_sphinx_javascript_frameworks_compat.js?v=2cd50e6c"></script>
<script src="_static/documentation_options.js?v=5929fcd5"></script>
<script src="_static/doctools.js?v=9bcbadda"></script>
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
<script src="_static/js/theme.js"></script>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="thirdai.logging" href="thirdailog.html" />
<link rel="prev" title="thirdai.neural_db" href="thirdaineuraldb.html" />
</head>
<body class="wy-body-for-nav">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
<div class="wy-side-scroll">
<div class="wy-side-nav-search" >
<a href="index.html" class="icon icon-home">
thirdai
<img src="_static/thirdai_logo.png" class="logo" alt="Logo"/>
</a>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="search.html" method="get">
<input type="text" name="q" placeholder="Search docs" aria-label="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div><div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="Navigation menu">
<ul class="current">
<li class="toctree-l1"><a class="reference internal" href="installation.html">Installation</a></li>
<li class="toctree-l1"><a class="reference internal" href="thirdaibase.html">thirdai</a></li>
<li class="toctree-l1"><a class="reference internal" href="thirdaineuraldb.html">thirdai.neural_db</a></li>
<li class="toctree-l1 current"><a class="current reference internal" href="#">thirdai.bolt</a><ul>
<li class="toctree-l2"><a class="reference internal" href="#udt">UDT</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.__init__"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.__init__()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.train"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.train()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.cold_start"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.cold_start()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.train_batch"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.train_batch()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.evaluate"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.evaluate()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.predict"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.predict()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.predict_batch"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.predict_batch()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.explain"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.explain()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.save"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.save()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.checkpoint"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.checkpoint()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.load"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.load()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.embedding_representation"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.embedding_representation()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.get_entity_embedding"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.get_entity_embedding()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.class_name"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.class_name()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.index"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.index()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.index_batch"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.index_batch()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.reset_temporal_trackers"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.reset_temporal_trackers()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.index_nodes"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.index_nodes()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.UniversalDeepTransformer.clear_graph"><code class="docutils literal notranslate"><span class="pre">UniversalDeepTransformer.clear_graph()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.Validation"><code class="docutils literal notranslate"><span class="pre">Validation</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.Validation.__init__"><code class="docutils literal notranslate"><span class="pre">Validation.__init__()</span></code></a></li>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.Validation.filename"><code class="docutils literal notranslate"><span class="pre">Validation.filename</span></code></a></li>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.Validation.metrics"><code class="docutils literal notranslate"><span class="pre">Validation.metrics</span></code></a></li>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.Validation.sparse_validation"><code class="docutils literal notranslate"><span class="pre">Validation.sparse_validation</span></code></a></li>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.Validation.steps_per_validation"><code class="docutils literal notranslate"><span class="pre">Validation.steps_per_validation</span></code></a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="#module-thirdai.bolt.types">UDT Input Column Types</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.ColumnType"><code class="docutils literal notranslate"><span class="pre">ColumnType</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.categorical"><code class="docutils literal notranslate"><span class="pre">categorical</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.categorical.__init__"><code class="docutils literal notranslate"><span class="pre">categorical.__init__()</span></code></a></li>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.categorical.delimiter"><code class="docutils literal notranslate"><span class="pre">categorical.delimiter</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.date"><code class="docutils literal notranslate"><span class="pre">date</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.date.__init__"><code class="docutils literal notranslate"><span class="pre">date.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.metadata"><code class="docutils literal notranslate"><span class="pre">metadata</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.metadata.__init__"><code class="docutils literal notranslate"><span class="pre">metadata.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.neighbors"><code class="docutils literal notranslate"><span class="pre">neighbors</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.neighbors.__init__"><code class="docutils literal notranslate"><span class="pre">neighbors.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.node_id"><code class="docutils literal notranslate"><span class="pre">node_id</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.node_id.__init__"><code class="docutils literal notranslate"><span class="pre">node_id.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.numerical"><code class="docutils literal notranslate"><span class="pre">numerical</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.numerical.__init__"><code class="docutils literal notranslate"><span class="pre">numerical.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.sequence"><code class="docutils literal notranslate"><span class="pre">sequence</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.sequence.__init__"><code class="docutils literal notranslate"><span class="pre">sequence.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.text"><code class="docutils literal notranslate"><span class="pre">text</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.text.__init__"><code class="docutils literal notranslate"><span class="pre">text.__init__()</span></code></a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.types.token_tags"><code class="docutils literal notranslate"><span class="pre">token_tags</span></code></a><ul>
<li class="toctree-l4"><a class="reference internal" href="#thirdai.bolt.types.token_tags.__init__"><code class="docutils literal notranslate"><span class="pre">token_tags.__init__()</span></code></a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="#module-thirdai.bolt.temporal">UDT Temporal Options</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.temporal.TemporalConfig"><code class="docutils literal notranslate"><span class="pre">TemporalConfig</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.temporal.categorical"><code class="docutils literal notranslate"><span class="pre">categorical()</span></code></a></li>
<li class="toctree-l3"><a class="reference internal" href="#thirdai.bolt.temporal.numerical"><code class="docutils literal notranslate"><span class="pre">numerical()</span></code></a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="thirdailog.html">thirdai.logging</a></li>
<li class="toctree-l1"><a class="reference internal" href="thirdailicensing.html">thirdai.licensing</a></li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"><nav class="wy-nav-top" aria-label="Mobile navigation menu" >
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="index.html">thirdai</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="Page navigation">
<ul class="wy-breadcrumbs">
<li><a href="index.html" class="icon icon-home" aria-label="Home"></a></li>
<li class="breadcrumb-item active">thirdai.bolt</li>
<li class="wy-breadcrumbs-aside">
<a href="_sources/thirdaibolt.rst.txt" rel="nofollow"> View page source</a>
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<section id="thirdai-bolt">
<h1>thirdai.bolt<a class="headerlink" href="#thirdai-bolt" title="Link to this heading"></a></h1>
<section id="udt">
<h2>UDT<a class="headerlink" href="#udt" title="Link to this heading"></a></h2>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.__init__">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">__init__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">target</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_types</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.__init__" title="Link to this definition"></a></dt>
<dd></dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.train">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">train</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">learning_rate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">epochs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">validation</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="#thirdai.bolt.Validation" title="thirdai.bolt.Validation"><span class="pre">Validation</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_in_memory_batches</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">callbacks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Callback</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">[]</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">[]</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_interval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shuffle_reservoir_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">64000</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">comm</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.train" title="Link to this definition"></a></dt>
<dd><p>Trains a UniversalDeepTransformer (UDT) on a given dataset using a file on disk
or in a cloud storage bucket, such as s3 or google cloud storage (GCS). If the
file is on S3, it should be in the normal s3 form, i.e. s3://bucket/path/to/key.
For files in GCS, the path should have the form gcs://bucket/path/to/filename.
We currently support csv and parquet format files. If the file is parquet, it
should end in .parquet or .pqt. Otherwise, we will assume it is a csv file.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>filename</strong> (<em>str</em>) – Path to the dataset file. It Can be a path to a file on
disk or an S3 or GCS resource identifier. If the file is on s3 or GCS,
regular credentials files will be required for authentication.</p></li>
<li><p><strong>learning_rate</strong> (<em>float</em>) – Optional, uses default if not provided.</p></li>
<li><p><strong>epochs</strong> (<em>int</em>) – Optional, uses default if not provided.</p></li>
<li><p><strong>validation</strong> (<em>Optional</em><em>[</em><a class="reference internal" href="#thirdai.bolt.Validation" title="thirdai.bolt.Validation"><em>bolt.Validation</em></a><em>]</em>) – This is an optional parameter that
specifies a validation dataset, metrics, and interval to use during
training.</p></li>
<li><p><strong>batch_size</strong> (<em>Option</em><em>[</em><em>int</em><em>]</em>) – This is an optional parameter indicating which batch
size to use for training. If not specified, the batch size will be autotuned.</p></li>
<li><p><strong>max_in_memory_batches</strong> (<em>Option</em><em>[</em><em>int</em><em>]</em>) – The maximum number of batches to load in
memory at a given time. If this is specified then the dataset will be processed
in a streaming fashion.</p></li>
<li><p><strong>verbose</strong> (<em>bool</em>) – Optional, defaults to True. Controls if additional information
is printed during training.</p></li>
<li><p><strong>callbacks</strong> (<em>List</em><em>[</em><em>bolt.train.callbacks.Callback</em><em>]</em>) – List of callbacks to use during
training.</p></li>
<li><p><strong>metrics</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – List of metrics to compute during training. These are
logged if logging is enabled, and are accessible by any callbacks.</p></li>
<li><p><strong>logging_interval</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – How frequently to log training metrics,
represents the number of batches between logging metrics. If not specified
logging is done at the end of each epoch.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>The train method returns a dictionary providing the values of any metrics
computed during training. The format is: {“name of metric”: [list of values]}.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>(Dict[str, List[float]])</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span>
<span class="go"> filename="./train_file", epochs=5, learning_rate=0.01, max_in_memory_batches=12</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span>
<span class="go"> filename="s3://bucket/path/to/key"</span>
<span class="go"> )</span>
</pre></div>
</div>
<p class="rubric">Notes</p>
<ul class="simple">
<li><p>If temporal tracking relationships are provided, UDT can make better
predictions by taking temporal context into account. For example, UDT may
keep track of the last few movies that a user has watched to better
recommend the next movie. <cite>model.train()</cite> automatically updates UDT’s
temporal context.</p></li>
<li><p>If the prediction task is binary classification then the model will attempt
to find an optimal threshold for predictions that will be used if <cite>return_predicted_class=True</cite>
is passed to calls to evaluate, predict, and predict_batch. The optimal threshold
will be selected based on what threshold maximizes the first validation metric
on the validation data. If no validation data or metrics are passed in then
it will use the first 100 batches of the training data and the first training
metric. If there is also no training metrics then it will not choose a prediction
threshold.</p></li>
</ul>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.cold_start">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">cold_start</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">filename:</span> <span class="pre">str,</span> <span class="pre">strong_column_names:</span> <span class="pre">~typing.List[str],</span> <span class="pre">weak_column_names:</span> <span class="pre">~typing.List[str],</span> <span class="pre">variable_length:</span> <span class="pre">~data.transformations.VariableLengthConfig</span> <span class="pre">|</span> <span class="pre">None</span> <span class="pre">=</span> <span class="pre"><data.transformations.VariableLengthConfig</span> <span class="pre">object>,</span> <span class="pre">learning_rate:</span> <span class="pre">float</span> <span class="pre">=</span> <span class="pre">0.001,</span> <span class="pre">epochs:</span> <span class="pre">int</span> <span class="pre">=</span> <span class="pre">5,</span> <span class="pre">batch_size:</span> <span class="pre">int</span> <span class="pre">|</span> <span class="pre">None</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">metrics:</span> <span class="pre">~typing.List[str]</span> <span class="pre">=</span> <span class="pre">[],</span> <span class="pre">validation:</span> <span class="pre">~bolt.Validation</span> <span class="pre">|</span> <span class="pre">None</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">callbacks:</span> <span class="pre">~typing.List[~bolt.train.callbacks.Callback]</span> <span class="pre">=</span> <span class="pre">[],</span> <span class="pre">max_in_memory_batches:</span> <span class="pre">int</span> <span class="pre">|</span> <span class="pre">None</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">verbose:</span> <span class="pre">bool</span> <span class="pre">=</span> <span class="pre">True,</span> <span class="pre">logging_interval:</span> <span class="pre">int</span> <span class="pre">|</span> <span class="pre">None</span> <span class="pre">=</span> <span class="pre">None,</span> <span class="pre">comm=None,</span> <span class="pre">shuffle_reservoir_size:</span> <span class="pre">int</span> <span class="pre">=</span> <span class="pre">64000,</span> <span class="pre">**kwargs</span></em><span class="sig-paren">)</span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.cold_start" title="Link to this definition"></a></dt>
<dd><p>This method will perform cold start pretraining for UDT. This is a type of
pretraining for text classification models that is especially useful for query
to product recommendation models. It requires that the model takes in a single
text input and has a categorical/multi-categorical output.</p>
<p>The cold start pretraining typically takes in an unsupervised dataset of objects
where each object corresponds to one or more columns of textual metadata. This could
be something like a product catalog (with product ids as objects, and titles,
descriptions, and tags as metadata). The goal with cold start is to pre-train UDT
on unsupervised data so in the future it may be able to answer text search queries
and return the relevant objects. The dataset it takes in should be a csv file that
gives a class id column and some number of text columns, where for a given row
the text is related to the class also specified by that row.</p>
<p>You may cold_start the model and train with supervised data afterwards, typically
leading to faster convergence on the supervised data.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>filename</strong> (<em>str</em>) – Path to the dataset used for pretraining.</p></li>
<li><p><strong>strong_column_names</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – The strong column names indicate which
text columns are most closely related to the output class. In this
case closely related means that all of the words in the text are useful
in identifying the output class in that row. For example in the
case of a product catalog then a strong column could be the full title
of the product.</p></li>
<li><p><strong>weak_column_names</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – The weak column names indicate which text
columns are either more loosely related to the output class. In
this case loosely related means that parts of the text are useful in
identifying the output class, but there may also be parts of the
text that contain more generic words or phrases that don’t have as high
of a correlation. For example in a product catalog the description of
the product could be a weak column because while there is a correlation,
parts of the description may be fairly similar between products or be
too general to completly identify which products the correspond to.</p></li>
<li><p><strong>learning_rate</strong> (<em>float</em>) – Optional, uses default if not provided.</p></li>
<li><p><strong>epochs</strong> (<em>int</em>) – Optional, uses default if not provided.</p></li>
<li><p><strong>batch_size</strong> (<em>Option</em><em>[</em><em>int</em><em>]</em>) – This is an optional parameter indicating which batch
size to use for training. If not specified, the batch size will be autotuned.</p></li>
<li><p><strong>metrics</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – List of metrics to compute during training. These are
logged if logging is enabled, and are accessible by any callbacks.</p></li>
<li><p><strong>validation</strong> (<em>Optional</em><em>[</em><a class="reference internal" href="#thirdai.bolt.Validation" title="thirdai.bolt.Validation"><em>bolt.Validation</em></a><em>]</em>) – This is an optional parameter that
specifies a validation dataset, metrics, and interval to use during
training.</p></li>
<li><p><strong>callbacks</strong> (<em>List</em><em>[</em><em>bolt.train.callbacks.Callback</em><em>]</em>) – List of callbacks to use during
training.</p></li>
<li><p><strong>max_in_memory_batches</strong> (<em>Option</em><em>[</em><em>int</em><em>]</em>) – The maximum number of batches to load in
memory at a given time. If this is specified then the dataset will be processed
in a streaming fashion.</p></li>
<li><p><strong>verbose</strong> (<em>bool</em>) – Optional, defaults to True. Controls if additional information
is printed during training.</p></li>
<li><p><strong>logging_interval</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – How frequently to log training metrics,
represents the number of batches between logging metrics. If not specified
logging is done at the end of each epoch.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>The train method returns a dictionary providing the values of any metrics
computed during training. The format is: {“name of metric”: [list of values]}.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>(Dict[str, List[float]])</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "query": bolt.types.text(),</span>
<span class="go"> "product": bolt.types.categorical(n_classes=1000),</span>
<span class="go"> }</span>
<span class="go"> target="product",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">cold_start</span><span class="p">(</span>
<span class="go"> filename="product_catalog.csv",</span>
<span class="go"> strong_column_names=["title"],</span>
<span class="go"> weak_column_names=["description", "bullet_points"],</span>
<span class="go"> learning_rate=0.001,</span>
<span class="go"> epochs=5,</span>
<span class="go"> metrics=["f_measure(0.95)"]</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span>
<span class="go"> train_filename=supervised_query_product_data,</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">result</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict</span><span class="p">({</span><span class="s2">"QUERY"</span><span class="p">:</span> <span class="n">query</span><span class="p">})</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.train_batch">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">train_batch</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">batch</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">learning_rate</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">object</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.train_batch" title="Link to this definition"></a></dt>
<dd><p>Trains the model on the given training batch.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>batch</strong> (<em>List</em><em>[</em><em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em><em>]</em>) – The raw data comprising the training batch. This should
be in the form {“column_name”: “column_value”} for each column the model expects.</p></li>
<li><p><strong>learning_rate</strong> (<em>float</em>) – Optional, uses default if not provided.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>None</p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.evaluate">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">evaluate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">[]</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_sparse_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.evaluate" title="Link to this definition"></a></dt>
<dd><p>Evaluates the UniversalDeepTransformer (UDT) on the given dataset and returns a
numpy array of the activations. We currently support csv and parquet format
files. If the file is parquet, it should end in .parquet or .pqt. Otherwise,
we will assume it is a csv file.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>filename</strong> (<em>str</em>) – Path to the dataset file. Like train, this can be a path
to a local file or a path to a file that lives in an s3 or google cloud
storage (GCS) bucket.</p></li>
<li><p><strong>metrics</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – List of metrics to compute during evaluation.</p></li>
<li><p><strong>use_sparse_inference</strong> (<em>bool</em>) – Optional, defaults to False, determines if
sparse inference is used during evaluation.</p></li>
<li><p><strong>verbose</strong> (<em>bool</em>) – Optional, defaults to True. Controls if additional information
is printed during training.</p></li>
<li><p><strong>top_k</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – Optional, defaults to None. This parameter is only used
for query reformulation model to deterimine how many candidates to select
before computing evaluation metrics.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>Returns a list of values for the specified metrics, keyed by the metric names.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>(Dict[str, float])</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">metrics</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">evaluate</span><span class="p">(</span><span class="n">filename</span><span class="o">=</span><span class="s2">"./test_file"</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="p">[</span><span class="s2">"categorical_accuracy"</span><span class="p">])</span>
</pre></div>
</div>
<p class="rubric">Notes</p>
<ul class="simple">
<li><p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
<cite>model.evaluate()</cite> automatically updates UDT’s temporal context.</p></li>
</ul>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.predict">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">predict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sample</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sparse_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_predicted_class</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">top_k</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">object</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.predict" title="Link to this definition"></a></dt>
<dd><p>Performs inference on a single sample.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_sample</strong> (<em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em>) – The input sample as a dictionary
where the keys are column names and the values are the respective column values.</p></li>
<li><p><strong>use_sparse_inference</strong> (<em>bool</em>) – Whether or not to use sparse inference.</p></li>
<li><p><strong>return_predicted_class</strong> (<em>bool</em>) – If true then the model will return the id of the
predicted class instead of the activations of the output layer. This argument is only
applicable to classification models.</p></li>
<li><p><strong>top_k</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – If specified then the model will return the ids of the
top k predicted classes instead of the activations of the output layer. This argument
is only applicable to classification models.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>Returns a numpy array of the activations if the output is dense, or a tuple
of the active neurons and activations if the output is sparse. The shape of
each array will be (num_nonzeros_in_output, ). If return predicted class is specified
then the class id (an integer) will be returned. If top_k is specified then a list of
integer class ids will be returned. You can map neuron ids back to target class names
by calling the <cite>class_name()</cite> method. If the target column is a sequence, UDT will
perform inference recursively and return a sequence in the same format as the target
column.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>(np.ndarray, Tuple[np.ndarray, np.ndarray], List[int], or int)</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># Suppose we configure UDT as follows:</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "user_id": bolt.types.categorical(),</span>
<span class="go"> "timestamp": bolt.types.date(),</span>
<span class="go"> "special_event": bolt.types.categorical(),</span>
<span class="go"> "movie_title": bolt.types.categorical(n_classes=500)</span>
<span class="go"> },</span>
<span class="go"> temporal_tracking_relationships={</span>
<span class="go"> "user_id": ["movie_title"]</span>
<span class="go"> },</span>
<span class="go"> target="movie_title",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># Make a single prediction</span>
<span class="gp">>>> </span><span class="n">activations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"}</span>
<span class="go"> )</span>
</pre></div>
</div>
<p class="rubric">Notes</p>
<ul class="simple">
<li><p>The values of columns that are tracked temporally may be unknown during inference
(the column_known_during_inference attribute of the bolt.temporal objects are False
by default). These columns do not need to be passed into <cite>model.predict()</cite>.
For example, we did not pass the “movie_title” column to <cite>model.predict()</cite>.
All other columns must be passed in.</p></li>
<li><p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
Thus, UDT is at its best when its internal temporal context gets updated with
new true samples. <cite>model.predict()</cite> does not update UDT’s temporal context.
To do this without retraining the model, we need to use <cite>model.index()</cite> or
<cite>model.index_batch()</cite>. Read about <cite>model.index()</cite> and <cite>model.index_batch()</cite>
for details.</p></li>
</ul>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.predict_batch">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">predict_batch</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sparse_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_predicted_class</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">top_k</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">object</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.predict_batch" title="Link to this definition"></a></dt>
<dd><p>Performs inference on a batch of samples in parallel.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_samples</strong> (<em>List</em><em>[</em><em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em><em>]</em>) – A list of input sample as dictionaries
where the keys are column names as specified in data_types and the
values are the respective column values.</p></li>
<li><p><strong>use_sparse_inference</strong> (<em>bool</em><em>, </em><em>default=False</em>) – Whether or not to use sparse inference.</p></li>
<li><p><strong>return_predicted_class</strong> (<em>bool</em>) – If true then the model will return the id of the
predicted class instead of the activations of the output layer. This argument is only
applicable to classification models.</p></li>
<li><p><strong>top_k</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – If specified then the model will return the ids of the
top k predicted classes instead of the activations of the output layer. This argument
is only applicable to classification models.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>Returns a numpy array of the activations if the output is dense, or a tuple
of the active neurons and activations if the output is sparse. The shape of
each array will be (batch_size, num_nonzeros_in_output). If return predicted class is specified
then the class id (an integer) will be returned. If top_k is specified then a list of
integer class ids will be returned. You can map neuron ids back to target class names
by calling the <cite>class_name()</cite> method. If the target column is a sequence, UDT will
perform inference recursively and return a sequence in the same format as the target
column.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>(np.ndarray, Tuple[np.ndarray, np.ndarray], List[List[int]], or List[int])</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">activations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict_batch</span><span class="p">([</span>
<span class="go"> {"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"},</span>
<span class="go"> {"user_id": "A25978", "timestamp": "2022-12-25", "special_event": "christmas"},</span>
<span class="go"> {"user_id": "A25978", "timestamp": "2022-12-26", "special_event": "christmas"}"</span>
<span class="go"> ])</span>
</pre></div>
</div>
<p class="rubric">Notes</p>
<ul class="simple">
<li><p>The values of columns that are tracked temporally may be unknown during inference
(the column_known_during_inference attribute of the bolt.temporal objects are False
by default). These columns do not need to be passed into <cite>model.predict_batch()</cite>.
For example, we did not pass the “movie_title” column to <cite>model.predict_batch()</cite>.
All other columns must be passed in.</p></li>
<li><p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
Thus, UDT is at its best when its internal temporal context gets updated with
new true samples. <cite>model.predict_batch()</cite> does not update UDT’s temporal context.
To do this without retraining the model, we need to use <cite>model.index()</cite> or
<cite>model.index_batch()</cite>. Read about <cite>model.index()</cite> and <cite>model.index_batch()</cite>
for details.</p></li>
</ul>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.explain">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">explain</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_sample</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_class</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.explain" title="Link to this definition"></a></dt>
<dd><p>Identifies the columns that are most responsible for a predicted outcome
and provides a brief description of the column’s value.</p>
<p>If a target is provided, the model will identify the columns that need
to change for the model to predict the target class.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>input_sample</strong> (<em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em>) – The input sample as a dictionary
where the keys are column names as specified in data_types and the ”
values are the respective column values.</p></li>
<li><p><strong>target_class</strong> (<em>str</em>) – Optional. The desired target class. If provided, the
model will identify the columns that need to change for the model to
predict the target class.</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>A list of explanations from the input features along with weights representing the
significance of that feature.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>List[Tuple[str, float]]</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># Suppose we configure UDT as follows:</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "user_id": bolt.types.categorical(),</span>
<span class="go"> "timestamp": bolt.types.date(),</span>
<span class="go"> "special_event": bolt.types.categorical(),</span>
<span class="go"> "movie_title": bolt.types.categorical(n_classes=500)</span>
<span class="go"> },</span>
<span class="go"> temporal_tracking_relationships={</span>
<span class="go"> "user_id": "movie_title"</span>
<span class="go"> },</span>
<span class="go"> target="movie_title",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># Make a single prediction</span>
<span class="gp">>>> </span><span class="n">explanations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">explain</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-02-02", "special_event": "christmas"}, target_class=35</span>
<span class="go"> )</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.save">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.save" title="Link to this definition"></a></dt>
<dd><p>Serializes an instance of UniversalDeepTransformer (UDT) into a file on disk.
The serialized UDT includes its current temporal context. The <cite>save</cite> method just saves
the model parameters, the <cite>checkpoint</cite> method saves additional information such as
the optimizer state to use if training is resumed.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>filename</strong> (<em>str</em>) – The file on disk to serialize this instance of UDT into.</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="s2">"udt_savefile.bolt"</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">checkpoint</span><span class="p">(</span><span class="s2">"udt_savefile.bolt"</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.checkpoint">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">checkpoint</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.checkpoint" title="Link to this definition"></a></dt>
<dd><p>Serializes an instance of UniversalDeepTransformer (UDT) into a file on disk.
The serialized UDT includes its current temporal context. The <cite>save</cite> method just saves
the model parameters, the <cite>checkpoint</cite> method saves additional information such as
the optimizer state to use if training is resumed.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>filename</strong> (<em>str</em>) – The file on disk to serialize this instance of UDT into.</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="s2">"udt_savefile.bolt"</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">checkpoint</span><span class="p">(</span><span class="s2">"udt_savefile.bolt"</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.load">
<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">load</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">bolt.UniversalDeepTransformer</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.load" title="Link to this definition"></a></dt>
<dd><p>Loads a serialized instance of a UniversalDeepTransformer (UDT) model from a
file on disk.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>filename</strong> (<em>str</em>) – The file on disk from where to load an instance of UDT.</p>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>The loaded instance of UDT</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>UniversalDeepTransformer</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">"udt_savefile.bolt"</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.embedding_representation">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">embedding_representation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_sample</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">object</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.embedding_representation" title="Link to this definition"></a></dt>
<dd><p>Performs inference on a single sample and returns the penultimate layer of
UniversalDeepTransformer (UDT) so that it can be used as an embedding
representation for downstream applications.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>input_sample</strong> (<em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em>) – The input sample as a dictionary
where the keys are column names as specified in data_types and the
values are the respective column values.</p>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>Returns a numpy array of the penultimate layer’s activations.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>np.ndarray</p>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># Suppose we configure UDT as follows:</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "user_id": bolt.types.categorical(),</span>
<span class="go"> "timestamp": bolt.types.date(),</span>
<span class="go"> "special_event": bolt.types.categorical(),</span>
<span class="go"> "movie_title": bolt.types.categorical(n_classes=500)</span>
<span class="go"> },</span>
<span class="go"> temporal_tracking_relationships={</span>
<span class="go"> "user_id": ["movie_title"]</span>
<span class="go"> },</span>
<span class="go"> target="movie_title",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># Get an embedding representation</span>
<span class="gp">>>> </span><span class="n">embedding</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">embedding_representation</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"}</span>
<span class="go"> )</span>
</pre></div>
</div>
<p class="rubric">Notes</p>
<ul class="simple">
<li><p>The values of columns that are tracked temporally may be unknown during inference
(the column_known_during_inference attribute of the bolt.temporal objects are False
by default). These columns do not need to be passed into <cite>model.embedding_representation()</cite>.
For example, we did not pass the “movie_title” column to <cite>model.embedding_representation()</cite>.
All other columns must be passed in.</p></li>
<li><p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
Thus, UDT is at its best when its internal temporal context gets updated with
new true samples. <cite>model.predict()</cite> does not update UDT’s temporal context.
To do this without retraining the model, we need to use <cite>model.index()</cite> or
<cite>model.index_batch()</cite>. Read about <cite>model.index()</cite> and <cite>model.index_batch()</cite>
for details.</p></li>
</ul>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.get_entity_embedding">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">get_entity_embedding</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">label_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">object</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.get_entity_embedding" title="Link to this definition"></a></dt>
<dd><p>Returns an embedding representation for a given output entity, an entity being
the name of a class predicted as output.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>label_id</strong> (<em>Union</em><em>[</em><em>int</em><em>, </em><em>str</em><em>]</em>) – The the name of the entity to get an embedding for.</p></li>
<li><p><strong>target</strong> (<em>If type='int' for the</em>)</p></li>
<li><p><strong>to</strong> (<em>this function should take in an integer from 0</em>)</p></li>
<li><p><strong>string.</strong> (<em>n_classes - 1 instead</em><em> of </em><em>a</em>)</p></li>
</ul>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>A 1D numpy array of floats representing a dense embedding of that entity.</p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.class_name">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">class_name</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">arg0</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.class_name" title="Link to this definition"></a></dt>
<dd><p>Returns the target class name associated with an output neuron ID.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>neuron_id</strong> (<em>int</em>) – The index of the neuron in UDT’s output layer. This is
useful for mapping the activations returned by <cite>evaluate()</cite> and
<cite>predict()</cite> back to class names.</p>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>The class names that corresponds to the given neuron_id.</p>
</dd>
<dt class="field-odd">Return type<span class="colon">:</span></dt>
<dd class="field-odd"><p>str</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">activations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"}</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">top_recommendation</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">activations</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">class_name</span><span class="p">(</span><span class="n">top_recommendation</span><span class="p">)</span>
<span class="go">"Die Hard"</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.index">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">index</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_sample</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.index" title="Link to this definition"></a></dt>
<dd><p>Indexes a single true sample to keep UniversalDeepTransformer’s (UDT) temporal
context up to date.</p>
<p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
Thus, UDT is at its best when its internal temporal context gets updated with
new true samples. <cite>model.index()</cite> does exactly this.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>input_sample</strong> (<em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em>) – The input sample as a dictionary
where the keys are column names as specified in data_types and the ”
values are the respective column values.</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># Suppose we configure UDT to do movie recommendation as follows:</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "user_id": bolt.types.categorical(),</span>
<span class="go"> "timestamp": bolt.types.date(),</span>
<span class="go"> "special_event": bolt.types.categorical(),</span>
<span class="go"> "movie_title": bolt.types.categorical(n_classes=500)</span>
<span class="go"> },</span>
<span class="go"> temporal_tracking_relationships={</span>
<span class="go"> "user_id": ["movie_title"]</span>
<span class="go"> },</span>
<span class="go"> target="movie_title",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># We then deploy the model for inference. Inference is performed by calling model.predict()</span>
<span class="gp">>>> </span><span class="n">activations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"}</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># Suppose we later learn that user "A33225" ends up watching "Die Hard 3".</span>
<span class="gp">>>> </span><span class="c1"># We can call model.index() to keep UDT's temporal context up to date.</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">index</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas", "movie_title": "Die Hard 3"}</span>
<span class="go"> )</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.index_batch">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">index_batch</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.index_batch" title="Link to this definition"></a></dt>
<dd><p>Indexes a batch of true samples to keep UniversalDeepTransformer’s (UDT) temporal
context up to date.</p>
<p>If temporal tracking relationships are provided, UDT can make better predictions
by taking temporal context into account. For example, UDT may keep track of
the last few movies that a user has watched to better recommend the next movie.
Thus, UDT is at its best when its internal temporal context gets updated with
new true samples. <cite>model.index_batch()</cite> does exactly this with a batch of samples.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>input_samples</strong> (<em>List</em><em>[</em><em>Dict</em><em>[</em><em>str</em><em>, </em><em>str</em><em>]</em><em>]</em>) – The input sample as a dictionary
where the keys are column names as specified in data_types and the ”
values are the respective column values.</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># Suppose we configure UDT to do movie recommendation as follows:</span>
<span class="gp">>>> </span><span class="n">model</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types={</span>
<span class="go"> "user_id": bolt.types.categorical(),</span>
<span class="go"> "timestamp": bolt.types.date(),</span>
<span class="go"> "special_event": bolt.types.categorical(),</span>
<span class="go"> "movie_title": bolt.types.categorical(n_classes=500)</span>
<span class="go"> },</span>
<span class="go"> temporal_tracking_relationships={</span>
<span class="go"> "user_id": ["movie_title"]</span>
<span class="go"> },</span>
<span class="go"> target="movie_title",</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># We then deploy the model for inference. Inference is performed by calling model.predict()</span>
<span class="gp">>>> </span><span class="n">activations</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span>
<span class="go"> input_sample={"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas"}</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="c1"># Suppose we later learn what users actually watched.</span>
<span class="gp">>>> </span><span class="c1"># We can call model.index_batch() to keep UDT's temporal context up to date.</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">index_batch</span><span class="p">(</span>
<span class="go"> input_samples=[</span>
<span class="go"> {"user_id": "A33225", "timestamp": "2022-12-25", "special_event": "christmas", "movie_title": "Die Hard 3"},</span>
<span class="go"> {"user_id": "A39574", "timestamp": "2022-12-25", "special_event": "christmas", "movie_title": "Home Alone"},</span>
<span class="go"> {"user_id": "A39574", "timestamp": "2022-12-26", "special_event": "christmas", "movie_title": "Home Alone 2"},</span>
<span class="go"> ]</span>
<span class="go"> )</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.reset_temporal_trackers">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">reset_temporal_trackers</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.reset_temporal_trackers" title="Link to this definition"></a></dt>
<dd><p>Resets UniversalDeepTransformer’s (UDT) temporal context. When temporal
relationships are supplied, UDT assumes that we feed it data in chronological
order. Thus, if we break this assumption, we need to first reset the temporal
trackers. An example of when you would use this is when you want to repeat the
UDT training routine on the same dataset. Since you would be training on data
from the same time period as before, we need to first reset the temporal trackers
so that we don’t double count events.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>None</strong></p>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>None</p>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">reset_temporal_trackers</span><span class="p">()</span>
</pre></div>
</div>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.index_nodes">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">index_nodes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_source</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">dataset.DataSource</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.index_nodes" title="Link to this definition"></a></dt>
<dd><p>Updates the graph that the UDT model is performing graph node classification on. The file
should have the same node id, neighbors, and features columns as the model is configured to accept.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><p><strong>filename</strong> (<em>str</em>) – The filename to load the graph from.</p>
</dd>
<dt class="field-even">Returns<span class="colon">:</span></dt>
<dd class="field-even"><p>None</p>
</dd>
</dl>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.UniversalDeepTransformer.clear_graph">
<span class="sig-prename descclassname"><span class="pre">UniversalDeepTransformer.</span></span><span class="sig-name descname"><span class="pre">clear_graph</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.UniversalDeepTransformer.clear_graph" title="Link to this definition"></a></dt>
<dd><p>Clears all graph info that is being tracked by the model.</p>
<dl class="field-list simple">
<dt class="field-odd">Returns<span class="colon">:</span></dt>
<dd class="field-odd"><p>None</p>
</dd>
</dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="thirdai.bolt.Validation">
<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">thirdai.bolt.</span></span><span class="sig-name descname"><span class="pre">Validation</span></span><a class="headerlink" href="#thirdai.bolt.Validation" title="Link to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">pybind11_object</span></code></p>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.Validation.__init__">
<span class="sig-name descname"><span class="pre">__init__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">filename</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">interval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_sparse_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.Validation.__init__" title="Link to this definition"></a></dt>
<dd><p>Creates a validation object that stores the necessary information for the model
to perform validation during training.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>filename</strong> (<em>str</em>) – The name of the validation file.</p></li>
<li><p><strong>metrics</strong> (<em>List</em><em>[</em><em>str</em><em>]</em>) – The metrics to compute for validation.</p></li>
<li><p><strong>interval</strong> (<em>Optional</em><em>[</em><em>int</em><em>]</em>) – The interval, in number of batches, between computing
validation. For instance, <cite>interval=10</cite> means that validation metrics will
be computed every 10 batches. If it is not specified then validation will
be done after each epoch.</p></li>
<li><p><strong>use_sparse_inference</strong> (<em>bool</em>) – Optional, defaults to False. When True, sparse
inference will be used during validation.</p></li>
</ul>
</dd>
</dl>
<p class="rubric">Examples</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">validation</span> <span class="o">=</span> <span class="n">bolt</span><span class="o">.</span><span class="n">Validation</span><span class="p">(</span>
<span class="go"> filename="validation.csv", metrics=["categorical_accuracy"], interval=10</span>
<span class="go"> )</span>
<span class="gp">>>> </span><span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="s2">"train.csv"</span><span class="p">,</span> <span class="n">epochs</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">validation</span><span class="o">=</span><span class="n">validation</span><span class="p">)</span>
</pre></div>
</div>
</dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="thirdai.bolt.Validation.filename">
<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">filename</span></span><a class="headerlink" href="#thirdai.bolt.Validation.filename" title="Link to this definition"></a></dt>
<dd></dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="thirdai.bolt.Validation.metrics">
<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">metrics</span></span><a class="headerlink" href="#thirdai.bolt.Validation.metrics" title="Link to this definition"></a></dt>
<dd></dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="thirdai.bolt.Validation.sparse_validation">
<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">sparse_validation</span></span><a class="headerlink" href="#thirdai.bolt.Validation.sparse_validation" title="Link to this definition"></a></dt>
<dd></dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="thirdai.bolt.Validation.steps_per_validation">
<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">steps_per_validation</span></span><a class="headerlink" href="#thirdai.bolt.Validation.steps_per_validation" title="Link to this definition"></a></dt>
<dd></dd></dl>
</dd></dl>
</section>
<section id="module-thirdai.bolt.types">
<span id="udt-input-column-types"></span><h2>UDT Input Column Types<a class="headerlink" href="#module-thirdai.bolt.types" title="Link to this heading"></a></h2>
<dl class="py class">
<dt class="sig sig-object py" id="thirdai.bolt.types.ColumnType">
<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">thirdai.bolt.types.</span></span><span class="sig-name descname"><span class="pre">ColumnType</span></span><a class="headerlink" href="#thirdai.bolt.types.ColumnType" title="Link to this definition"></a></dt>
<dd><p>Bases: <code class="xref py py-class docutils literal notranslate"><span class="pre">pybind11_object</span></code></p>
<p>Base class for bolt types.</p>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="thirdai.bolt.types.categorical">
<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">thirdai.bolt.types.</span></span><span class="sig-name descname"><span class="pre">categorical</span></span><a class="headerlink" href="#thirdai.bolt.types.categorical" title="Link to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#thirdai.bolt.types.ColumnType" title="thirdai._thirdai.bolt.types.ColumnType"><code class="xref py py-class docutils literal notranslate"><span class="pre">ColumnType</span></code></a></p>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.types.categorical.__init__">
<span class="sig-name descname"><span class="pre">__init__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">n_classes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'str'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delimiter</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metadata</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="#thirdai.bolt.types.metadata" title="thirdai.bolt.types.metadata"><span class="pre">bolt.types.metadata</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.types.categorical.__init__" title="Link to this definition"></a></dt>
<dd><p>Categorical column type. Use this object if a column contains categorical
data (each unique value is treated as a class). Examples include user IDs,
movie titles, or age groups.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters<span class="colon">:</span></dt>
<dd class="field-odd"><ul class="simple">
<li><p><strong>n_classes</strong> (<em>Optiona</em><em>[</em><em>int</em><em>]</em>) – Optional, defaults to None. If the data type is for the
target column in a classification task then this must be specified to indicate
how many classes are present in the dataset.</p></li>
<li><p><strong>type</strong> (<em>str</em>) – Optional, defaults to “str”. How the categories are represented in the
dataset, i.e. are they stored as strings or integers. This is primarily relevant
for target columns since string categories will be mapped to integers to
correspond to output neurons, however if the categories are already represented
as integers then they can be used for labels directly.</p></li>
<li><p><strong>delimiter</strong> (<em>str</em>) – Optional. Defaults to None. A single character
(length-1 string) that separates multiple values in the same
column. This can only be used for the target column. If not
provided, UDT assumes that there is only one value in the column.</p></li>
<li><p><strong>metadata</strong> (<a class="reference internal" href="#thirdai.bolt.types.metadata" title="thirdai.bolt.types.metadata"><em>metadata</em></a>) – Optional. A metadata object to be used when there
is a separate metadata file corresponding to this categorical
column.</p></li>
</ul>
</dd>
</dl>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types: {</span>
<span class="go"> "user_id": bolt.types.categorical(</span>
<span class="go"> delimiter=' ',</span>
<span class="go"> metadata=bolt.types.metadata(filename="user_meta.csv", data_types={"age": bolt.types.numerical()}, key_column_name="user_id")</span>
<span class="go"> )</span>
<span class="go"> }</span>
<span class="go"> ...</span>
<span class="go"> )</span>
</pre></div>
</div>
</dd></dl>
<dl class="py property">
<dt class="sig sig-object py" id="thirdai.bolt.types.categorical.delimiter">
<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">delimiter</span></span><a class="headerlink" href="#thirdai.bolt.types.categorical.delimiter" title="Link to this definition"></a></dt>
<dd></dd></dl>
</dd></dl>
<dl class="py class">
<dt class="sig sig-object py" id="thirdai.bolt.types.date">
<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">thirdai.bolt.types.</span></span><span class="sig-name descname"><span class="pre">date</span></span><a class="headerlink" href="#thirdai.bolt.types.date" title="Link to this definition"></a></dt>
<dd><p>Bases: <a class="reference internal" href="#thirdai.bolt.types.ColumnType" title="thirdai._thirdai.bolt.types.ColumnType"><code class="xref py py-class docutils literal notranslate"><span class="pre">ColumnType</span></code></a></p>
<dl class="py method">
<dt class="sig sig-object py" id="thirdai.bolt.types.date.__init__">
<span class="sig-name descname"><span class="pre">__init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">→</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#thirdai.bolt.types.date.__init__" title="Link to this definition"></a></dt>
<dd><p>Date column type. Use this object if a column contains date strings.
Date strings must be in YYYY-MM-DD format.</p>
<p class="rubric">Example</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">bolt</span><span class="o">.</span><span class="n">UniversalDeepTransformer</span><span class="p">(</span>
<span class="go"> data_types: {</span>