-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathfaq.html
More file actions
639 lines (485 loc) · 43 KB
/
faq.html
File metadata and controls
639 lines (485 loc) · 43 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
<!DOCTYPE html>
<html lang="en" data-content_root="" >
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
<title>What the FAQ — InterpretML documentation</title>
<script data-cfasync="false">
document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
</script>
<!-- Loaded before other Sphinx assets -->
<link href="_static/styles/theme.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/styles/bootstrap.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/styles/pydata-sphinx-theme.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link href="_static/vendor/fontawesome/6.5.2/css/all.min.css?digest=dfe6caa3a7d634c4db9b" rel="stylesheet" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-solid-900.woff2" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-brands-400.woff2" />
<link rel="preload" as="font" type="font/woff2" crossorigin href="_static/vendor/fontawesome/6.5.2/webfonts/fa-regular-400.woff2" />
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" href="_static/styles/sphinx-book-theme.css?digest=14f4ca6b54d191a8c7657f6c759bf11a5fb86285" type="text/css" />
<link rel="stylesheet" type="text/css" href="_static/togglebutton.css" />
<link rel="stylesheet" type="text/css" href="_static/copybutton.css" />
<link rel="stylesheet" type="text/css" href="_static/mystnb.4510f1fc1dee50b3e5859aac5469c37c29e427902b24a333a5f9fcb2f0b3ac41.css" />
<link rel="stylesheet" type="text/css" href="_static/sphinx-thebe.css" />
<link rel="stylesheet" type="text/css" href="_static/design-style.4045f2051d55cab465a707391d5b2007.min.css" />
<!-- Pre-loaded scripts that we'll load fully later -->
<link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=dfe6caa3a7d634c4db9b" />
<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=dfe6caa3a7d634c4db9b" />
<script src="_static/vendor/fontawesome/6.5.2/js/all.min.js?digest=dfe6caa3a7d634c4db9b"></script>
<script data-url_root="./" id="documentation_options" src="_static/documentation_options.js"></script>
<script src="_static/jquery.js"></script>
<script src="_static/underscore.js"></script>
<script src="_static/_sphinx_javascript_frameworks_compat.js"></script>
<script src="_static/doctools.js"></script>
<script src="_static/clipboard.min.js"></script>
<script src="_static/copybutton.js"></script>
<script src="_static/scripts/sphinx-book-theme.js?digest=5a5c038af52cf7bc1a1ec88eea08e6366ee68824"></script>
<script>let toggleHintShow = 'Click to show';</script>
<script>let toggleHintHide = 'Click to hide';</script>
<script>let toggleOpenOnPrint = 'true';</script>
<script src="_static/togglebutton.js"></script>
<script>var togglebuttonSelector = '.toggle, .admonition.dropdown';</script>
<script src="_static/design-tabs.js"></script>
<script>const THEBE_JS_URL = "https://unpkg.com/thebe@0.8.2/lib/index.js"
const thebe_selector = ".thebe,.cell"
const thebe_selector_input = "pre"
const thebe_selector_output = ".output, .cell_output"
</script>
<script async="async" src="_static/sphinx-thebe.js"></script>
<script>DOCUMENTATION_OPTIONS.pagename = 'faq';</script>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Examples" href="examples.html" />
<link rel="prev" title="Logging and Debugging" href="debugging-guide.html" />
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<meta name="docsearch:language" content="en"/>
</head>
<body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
<div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
<div id="pst-scroll-pixel-helper"></div>
<button type="button" class="btn rounded-pill" id="pst-back-to-top">
<i class="fa-solid fa-arrow-up"></i>Back to top</button>
<input type="checkbox"
class="sidebar-toggle"
id="pst-primary-sidebar-checkbox"/>
<label class="overlay overlay-primary" for="pst-primary-sidebar-checkbox"></label>
<input type="checkbox"
class="sidebar-toggle"
id="pst-secondary-sidebar-checkbox"/>
<label class="overlay overlay-secondary" for="pst-secondary-sidebar-checkbox"></label>
<div class="search-button__wrapper">
<div class="search-button__overlay"></div>
<div class="search-button__search-container">
<form class="bd-search d-flex align-items-center"
action="search.html"
method="get">
<i class="fa-solid fa-magnifying-glass"></i>
<input type="search"
class="form-control"
name="q"
id="search-input"
placeholder="Search this book..."
aria-label="Search this book..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form></div>
</div>
<div class="pst-async-banner-revealer d-none">
<aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>
<header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
</header>
<div class="bd-container">
<div class="bd-container__inner bd-page-width">
<div class="bd-sidebar-primary bd-sidebar">
<div class="sidebar-header-items sidebar-primary__section">
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item">
<a class="navbar-brand logo" href="index.html">
<img src="_static/logo.png" class="logo__image only-light" alt="InterpretML documentation - Home"/>
<script>document.write(`<img src="_static/logo.png" class="logo__image only-dark" alt="InterpretML documentation - Home"/>`);</script>
</a></div>
<div class="sidebar-primary-item"><nav class="bd-links" id="bd-docs-nav" aria-label="Main">
<div class="bd-toc-item navbar-nav active">
<ul class="nav bd-sidenav bd-sidenav__home-link">
<li class="toctree-l1">
<a class="reference internal" href="index.html">
Getting Started
</a>
</li>
</ul>
<ul class="current nav bd-sidenav">
<li class="toctree-l1 has-children"><a class="reference internal" href="interpret.html">API Reference</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-glassbox.html">interpret.glassbox</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/ExplainableBoostingClassifier.html">ExplainableBoostingClassifier</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/ExplainableBoostingRegressor.html">ExplainableBoostingRegressor</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/APLRRegressor.html">APLRRegressor</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/APLRClassifier.html">APLRClassifier</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/LogisticRegression.html">LogisticRegression</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/LinearRegression.html">LinearRegression</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/ClassificationTree.html">ClassificationTree</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/RegressionTree.html">RegressionTree</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/merge_ebms.html">merge_ebms</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-blackbox.html">interpret.blackbox</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/ShapKernel.html">ShapKernel</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/LimeTabular.html">LimeTabular</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/PartialDependence.html">PartialDependence</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/MorrisSensitivity.html">MorrisSensitivity</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-greybox.html">interpret.greybox</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/ShapTree.html">ShapTree</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-perf.html">interpret.perf</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/RegressionPerf.html">RegressionPerf</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/ROC.html">ROC</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/PR.html">PR</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-data.html">interpret.data</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/ClassHistogram.html">ClassHistogram</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/Marginal.html">Marginal</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-privacy.html">interpret.privacy</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/DPExplainableBoostingClassifier.html">DPExplainableBoostingClassifier</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/DPExplainableBoostingRegressor.html">DPExplainableBoostingRegressor</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-utils.html">interpret.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/link_func.html">link_func</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/inv_link.html">inv_link</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/measure_interactions.html">measure_interactions</a></li>
<li class="toctree-l3"><a class="reference internal" href="python/api/purify.html">purify</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="python/api/interpret-develop.html">interpret.develop</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="python/api/debug_mode.html">debug_mode</a></li>
</ul>
</details></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/show.html">show</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/show_link.html">show_link</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/set_show_addr.html">set_show_addr</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/get_show_addr.html">get_show_addr</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/preserve.html">preserve</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/shutdown_show_server.html">shutdown_show_server</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/init_show_server.html">init_show_server</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/status_show_server.html">status_show_server</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/set_visualize_provider.html">set_visualize_provider</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/api/get_visualize_provider.html">get_visualize_provider</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="algorithms.html">Algorithms</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2 has-children"><a class="reference internal" href="glassbox.html">Glassbox Models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="ebm.html">Explainable Boosting Machine</a></li>
<li class="toctree-l3"><a class="reference internal" href="aplr.html">Automatic Piecewise Linear Regression (APLR)</a></li>
<li class="toctree-l3"><a class="reference internal" href="lr.html">Linear Model</a></li>
<li class="toctree-l3"><a class="reference internal" href="dt.html">Decision Tree</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="blackbox.html">Blackbox Explainers</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="shap.html">Shapley Additive Explanations</a></li>
<li class="toctree-l3"><a class="reference internal" href="lime.html">Local Interpretable Model-agnostic Explanations</a></li>
<li class="toctree-l3"><a class="reference internal" href="pdp.html">Partial Dependence Plot</a></li>
<li class="toctree-l3"><a class="reference internal" href="msa.html">Morris Sensitivity Analysis</a></li>
</ul>
</details></li>
<li class="toctree-l2 has-children"><a class="reference internal" href="privacy.html">Differentially Private Models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l3"><a class="reference internal" href="dpebm.html">Differentially Private EBMs</a></li>
</ul>
</details></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="hyperparameters.html">Hyperparameters</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="ebm-internals.html">EBM internals</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="ebm-internals-regression.html">EBM Internals - Regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="ebm-internals-classification.html">EBM Internals - Binary classification</a></li>
<li class="toctree-l2"><a class="reference internal" href="ebm-internals-multiclass.html">EBM Internals - Multiclass</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="framework.html">Visualizations</a></li>
<li class="toctree-l1"><a class="reference internal" href="deployment-guide.html">Deployments</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="development.html">Development</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="installation-guide.html">Installation</a></li>
<li class="toctree-l2"><a class="reference internal" href="debugging-guide.html">Logging and Debugging</a></li>
</ul>
</details></li>
<li class="toctree-l1 current active"><a class="current reference internal" href="#">What the FAQ</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="examples.html">Examples</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="python/examples/interpretable-classification.html">Interpretable Classification</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/interpretable-regression.html">Interpretable Regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/interpretable-regression-synthetic.html">Smoothing EBMs</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/explain-blackbox-classifiers.html">Explain Blackbox Classifiers</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/explain-blackbox-regressors.html">Explain Blackbox Regressors</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/merge-ebms.html">Merging EBMs</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/custom-interactions.html">Custom Interactions</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/differential-privacy.html">Differential Privacy</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/group-importances.html">Group Importances</a></li>
<li class="toctree-l2"><a class="reference internal" href="python/examples/prototype-selection.html">Prototype Selection</a></li>
</ul>
</details></li>
</ul>
</div>
</nav></div>
</div>
<div class="sidebar-primary-items__end sidebar-primary__section">
</div>
<div id="rtd-footer-container"></div>
</div>
<main id="main-content" class="bd-main" role="main">
<div class="sbt-scroll-pixel-helper"></div>
<div class="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item"><label class="sidebar-toggle primary-toggle btn btn-sm" for="__primary" title="Toggle primary sidebar" data-bs-placement="bottom" data-bs-toggle="tooltip">
<span class="fa-solid fa-bars"></span>
</label></div>
</div>
<div class="header-article-items__end">
<div class="header-article-item">
<div class="article-header-buttons">
<div class="dropdown dropdown-download-buttons">
<button class="btn dropdown-toggle" type="button" data-bs-toggle="dropdown" aria-expanded="false" aria-label="Download this page">
<i class="fas fa-download"></i>
</button>
<ul class="dropdown-menu">
<li><a href="_sources/faq.ipynb" target="_blank"
class="btn btn-sm btn-download-source-button dropdown-item"
title="Download source file"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-file"></i>
</span>
<span class="btn__text-container">.ipynb</span>
</a>
</li>
<li>
<button onclick="window.print()"
class="btn btn-sm btn-download-pdf-button dropdown-item"
title="Print to PDF"
data-bs-placement="left" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-file-pdf"></i>
</span>
<span class="btn__text-container">.pdf</span>
</button>
</li>
</ul>
</div>
<button onclick="toggleFullScreen()"
class="btn btn-sm btn-fullscreen-button"
title="Fullscreen mode"
data-bs-placement="bottom" data-bs-toggle="tooltip"
>
<span class="btn__icon-container">
<i class="fas fa-expand"></i>
</span>
</button>
<script>
document.write(`
<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button" title="light/dark" aria-label="light/dark" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="theme-switch fa-solid fa-sun fa-lg" data-mode="light"></i>
<i class="theme-switch fa-solid fa-moon fa-lg" data-mode="dark"></i>
<i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"></i>
</button>
`);
</script>
<script>
document.write(`
<button class="btn btn-sm pst-navbar-icon search-button search-button__button" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
`);
</script>
</div></div>
</div>
</div>
</div>
<div id="jb-print-docs-body" class="onlyprint">
<h1>What the FAQ</h1>
<!-- Table of contents -->
<div id="print-main-content">
<div id="jb-print-toc">
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section class="tex2jax_ignore mathjax_ignore" id="what-the-faq">
<h1>What the FAQ<a class="headerlink" href="#what-the-faq" title="Permalink to this heading">#</a></h1>
<p>You have questions, too many of them. We have answers.</p>
<h2>Why doesn't anything happen when I run show(...)?</h2><p>Interpret’s visualizations are designed to work best in Jupyter notebook-like environments (like Jupyter notebook, VS Code, Colab, …). If you’re running show() from a command line script, you may not be able to render visualizations directly – check the printed console output for a link to open in a browser.</p>
<p>By default, interpret hosts visualizations on a local web server using Plotly Dash. In some restricted environments, where applications are not allowed to host a local web server, we embed visualizations directly into the notebook. If Interpret did not automatically detect and switch the rendering mode, you can manually embed visualizations in restricted environments with the following code:</p>
<div class="cell docutils container">
<div class="cell_input docutils container">
<div class="highlight-ipython3 notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span><span class="w"> </span><span class="nn">interpret.provider</span><span class="w"> </span><span class="kn">import</span> <span class="n">InlineProvider</span>
<span class="kn">from</span><span class="w"> </span><span class="nn">interpret</span><span class="w"> </span><span class="kn">import</span> <span class="n">set_visualize_provider</span>
<span class="n">set_visualize_provider</span><span class="p">(</span><span class="n">InlineProvider</span><span class="p">())</span>
</pre></div>
</div>
</div>
</div>
<p>The <code class="docutils literal notranslate"><span class="pre">InlineProvider</span></code> is compatible with a broader range of environments than the default rendering logic in Interpret. However, <em>warning</em>: you may experience slower performance when calling <code class="docutils literal notranslate"><span class="pre">show()</span></code>, and the full interpret dashboard is currently unsupported with <code class="docutils literal notranslate"><span class="pre">InlineProvider</span></code>.</p>
<h2>How do I generate the full interpret dashboard instead of the small single-explanation dropdown?</h2><p>Make sure you are passing in a <em>list</em> of explanations to show. Note that you can also pass in a single explanation wrapped in a list. Ex:</p>
<div class="highlight-python3 notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span><span class="w"> </span><span class="nn">interpret</span><span class="w"> </span><span class="kn">import</span> <span class="n">show</span>
<span class="n">show</span><span class="p">(</span><span class="n">ebm_local</span><span class="p">)</span> <span class="c1"># Returns small dropdown</span>
<span class="n">show</span><span class="p">([</span> <span class="n">ebm_local</span> <span class="p">])</span> <span class="c1"># Produces interpret dashboard</span>
</pre></div>
</div>
<h2>How can I extract the underlying data used to visualize explanations?</h2><p>Every explanation object in interpret supports a <code class="docutils literal notranslate"><span class="pre">.data()</span></code> method, which returns a JSON-compatible dictionary of the underlying data used to produce the visualizations. Most explanations contain many visualizations – for example, <code class="docutils literal notranslate"><span class="pre">explain_local()</span></code> calls will produce visualizations for each individual instance passed to the function. <code class="docutils literal notranslate"><span class="pre">data</span></code> takes in a single parameter which indexes the explanation object and returns the data used for that individual explanation (ex: explanation.data(0) returns the data used for the first visualization in the object). To return data used for <em>all</em> visualizations, use the <code class="docutils literal notranslate"><span class="pre">-1</span></code> key (explanation.data(-1)).</p>
<h2>How can I save an explanation graph to disk?</h2>
<p>Every explanation graph is a Plotly object, and can be saved to disk using the camera icon in the UI:</p>
<p><img alt="Save plotly graph" src="_images/save-plotly-graph.png" /></p>
<p>or with the <a class="reference external" href="https://plotly.com/python/static-image-export/">Plotly Static Image Export</a> tools. You can access individual Plotly figures with:</p>
<div class="highlight-python3 notranslate"><div class="highlight"><pre><span></span><span class="n">ebm_explanation</span> <span class="o">=</span> <span class="n">ebm</span><span class="o">.</span><span class="n">explain_global</span><span class="p">()</span>
<span class="n">plotly_fig</span> <span class="o">=</span> <span class="n">ebm_explanation</span><span class="o">.</span><span class="n">visualize</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
</pre></div>
</div>
<p>For example, to save every graph in a global explanation to the “images” directory on disk:</p>
<div class="highlight-python3 notranslate"><div class="highlight"><pre><span></span><span class="n">ebm_global</span> <span class="o">=</span> <span class="n">ebm</span><span class="o">.</span><span class="n">explain_global</span><span class="p">()</span>
<span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">ebm</span><span class="o">.</span><span class="n">feature_groups_</span><span class="p">):</span>
<span class="n">plotly_fig</span> <span class="o">=</span> <span class="n">ebm_global</span><span class="o">.</span><span class="n">visualize</span><span class="p">(</span><span class="n">index</span><span class="p">)</span>
<span class="n">plotly_fig</span><span class="o">.</span><span class="n">write_image</span><span class="p">(</span><span class="sa">f</span><span class="s2">"images/fig_</span><span class="si">{</span><span class="n">index</span><span class="si">}</span><span class="s2">.png"</span><span class="p">)</span>
</pre></div>
</div>
<h2>What does the 'density' at the bottom of each graph mean?</h2>
<p>The density is a histogram describing the data distribution for that feature (estimated using any data passed into the <code class="docutils literal notranslate"><span class="pre">explain_*</span></code> methods). It is often useful to understand how much data is in each region of the feature space when visualizing an explanation – models can perform very differently with large and small samples.</p>
<h2>Does interpret support explainability for image and text data?</h2><p>Not yet, but keep an eye out for future releases!</p>
<h2>How do I install just a single explainer without any other dependencies?</h2>
<p>This is possible by installing directly from the <code class="docutils literal notranslate"><span class="pre">interpret-core</span></code> base package for advanced users. For example, if you want to install EBM without any other explainers, run the following codeblock below:</p>
<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>pip<span class="w"> </span>install<span class="w"> </span>interpret-core<span class="o">[</span>required,ebm<span class="o">]</span>
</pre></div>
</div>
<p>The various installation options provided by <code class="docutils literal notranslate"><span class="pre">interpret-core</span></code> are specified in the interpret-core <a class="reference external" href="https://github.com/interpretml/interpret/blob/d1ee9ae602d7086ed4074092de4041f98efe0d68/python/interpret-core/setup.py#L54">setup.py</a> file. The <code class="docutils literal notranslate"><span class="pre">required</span></code> tag is generally recommended for all installs.</p>
<h2>Why aren't the feature names showing in graphs?</h2>
<p>If you are passing a numpy array to an explainer, make sure that the <code class="docutils literal notranslate"><span class="pre">feature_names</span></code> property is set (either by passing at initialization, or manually setting before calling an explain function). This should work automatically when using pandas dataframes.</p>
<h2>EBM</h2><h2>Should I be parameter tuning EBMs (and if so, what parameters should I tune)?</h2>
<p>In general, the default parameters for EBMs should perform reasonably well on most problems. We recommend training a model with defaults and looking through the learned functions to catch abnormal behavior before parameter tuning – oftentimes, these graphs help indicate which parameters to tune. Here’s some general recommendations:</p>
<ul class="simple">
<li><p>To produce the best models, we generally recommend setting <code class="docutils literal notranslate"><span class="pre">outer_bags</span></code> and <code class="docutils literal notranslate"><span class="pre">inner_bags</span></code> to 25 or more each. This will significantly slow down the training time of the algorithm, and may be infeasible on larger datasets, but tends to produce smoother graphs with marginally higher accuracy.</p></li>
<li><p>If you believe the models might be overfitting (ex: large difference between train and test error, or high degrees of instability in graphs), consider the following options:</p>
<ul>
<li><p>Reduce <code class="docutils literal notranslate"><span class="pre">max_bins</span></code> for smaller datasets, to clump more data together.</p></li>
<li><p>Make the early stopping more aggressive by decreasing <code class="docutils literal notranslate"><span class="pre">early_stopping_rounds</span></code> and increasing <code class="docutils literal notranslate"><span class="pre">early_stopping_tolerance</span></code>.</p></li>
</ul>
</li>
<li><p>Conversely, for underfit models which appear too conservative, you can do the opposite of the previous suggestions – increase <code class="docutils literal notranslate"><span class="pre">max_bins</span></code> and make early stopping less aggressive. It might also be helpful to increase the total allowed <code class="docutils literal notranslate"><span class="pre">max_rounds</span></code>.</p></li>
<li><p>If many of the included interaction terms are significant (e.g. learning large values or ranking highly on the overall importance list), there’s a chance the 10 interaction terms included by default is insufficient for your dataset. Consider increasing this number.</p></li>
<li><p>For general tuning, we recommend sweeping <code class="docutils literal notranslate"><span class="pre">max_bins</span></code> with values between 32 and 1024, and <code class="docutils literal notranslate"><span class="pre">max_leaves</span></code> from 2 to 5. These improvements are normally marginal, but may help significantly on some datasets.</p></li>
</ul>
<h2>What does the error bar on an EBM graph mean, and how are they calculated?</h2>
<p>The error bars are rough estimates of the uncertainty of the model in each region of the feature space. A large error bar means that the learned function may have changed substantially with minor changes to the training data, and indicates that the interpretation of the model in that region should be treated with more skepticisim.</p>
<p>The size of the error bar is typically determined by two factors – the amount of training data in that region of the feature space, and the inherent uncertainty of the learned model. For example, consider this graph of the learned “Age” feature from the Adult income dataset:</p>
<p><img alt="Age graph" src="_images/age-graph-adult.png" /></p>
<p>On the right side of the graph (Ages 70+), the model predictions become unstable, and the error bars become much larger to indicate this uncertainty. From the density graph at the bottom, we can see that this is likely due to the small number of samples in this area.</p>
<p>The error estimates are derived through <a class="reference external" href="https://en.wikipedia.org/wiki/Bootstrap_aggregating">bagging</a>. By default, EBM trains 8 different mini-EBMs on random 85% subsamples of the training dataset. The number of mini-EBMs trained is controlled by the <code class="docutils literal notranslate"><span class="pre">outer_bags</span></code> parameter, and the proportion of data sampled is 1 - <code class="docutils literal notranslate"><span class="pre">validation_size</span></code>. The outputs of these models are averaged together to produce the final EBM, and the standard deviation of estimates for each region of the graph is published as the error bar. You can programmatically access the error bar sizes from the <code class="docutils literal notranslate"><span class="pre">standard_deviations_</span></code> property on any EBM.</p>
<h2>Do you have p-values for significance of EBM terms?</h2>
<p>Not yet, though this is an area of active research. If you’re interested in discussing this or helping us figure it out, reach out!</p>
<h2>What's the difference between EBMs in classification and regression?</h2>
<p>The differences are largely analogous to the differences between <a class="reference external" href="https://en.wikipedia.org/wiki/Linear_regression">linear regression</a> and <a class="reference external" href="https://en.wikipedia.org/wiki/Logistic_regression">logistic regression</a>. Like logistic regression, EBMs need to be adapted to the classification setting through the use of the <a class="reference external" href="https://en.wikipedia.org/wiki/Logistic_function">logistic link function</a>.</p>
<p>This link is used because probabilities are not additive – two features cannot each contribute +0.6 probability to a prediction, as the resulting final probability of 1.2 is undefined. The logistic link allows the models to train in the logit space – where the contribution of each feature can be additive – and convert back to a bounded probability at prediction time. Therefore, when interpreting the graphs of an <code class="docutils literal notranslate"><span class="pre">ExplainableBoostingClassifier</span></code>, it’s important to remember that the y-axis values are in <a class="reference external" href="https://en.wikipedia.org/wiki/Logit">logits</a>. The interpretation of the shape is generally the same – positive values push the prediction towards a positive label for that class. However, because these graphs are in logarithm space, differences of +1 or +2 are quite significant.</p>
<p>In the regression setting, the interpretation is simpler: the y-axis values are directly in the units of the target. For example, if you are training on housing prices, the y-axis for each graph will represent exactly how many dollars that feature contributes to the final price of a model – no additional transformations needed!</p>
<h2>How can I edit EBM models to remove unwanted learned effects?</h2>
<p>The final EBM models are stored as numpy arrays in the <code class="docutils literal notranslate"><span class="pre">terms_scores_</span></code> property. Indexing this array returns the function learned for a specific term in the EBM – for example, <code class="docutils literal notranslate"><span class="pre">ebm.terms_scores_[0]</span></code> will return the array for the first term. Editing this array directly will change all predictions made by the model for that corresponding region of the graph. A nicer API for this with more granular controls will be introduced shortly!</p>
<h2>Can we enforce monotonicity for individual EBM terms?</h2>
<p>It is possible to enforce monotonicity for individual terms in an EBM via two methods:</p>
<ul class="simple">
<li><p>By setting <code class="docutils literal notranslate"><span class="pre">monotone_constraints</span></code> in the <code class="docutils literal notranslate"><span class="pre">ExplainableBoostingClassifier</span></code> or <code class="docutils literal notranslate"><span class="pre">ExplainableBoostingRegressor</span></code> constructor. This parameter is a list of integers, where each integer corresponds to the monotonicity constraint for the corresponding feature. A value of -1 enforces decreasing monotonicity, 0 enforces no monotonicity, and 1 enforces increasing monotonicity. For example, <code class="docutils literal notranslate"><span class="pre">monotone_constraints=[1,</span> <span class="pre">-1,</span> <span class="pre">0]</span></code> would enforce increasing monotonicity for the first feature, decreasing monotonicity for the second feature, and no monotonicity for the third feature.</p></li>
<li><p>Through postprocessing a graph. We generally recommend using <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.isotonic.IsotonicRegression.html#sklearn.isotonic.IsotonicRegression">isotonic regression</a> on each graph output to force positive or negative monotonicity. This can be done by calling the <code class="docutils literal notranslate"><span class="pre">monotonize</span></code> method on the EBM object.
Postprocessing is the recommended method, as it prevents the model from compensating for the monotonicity constraints by learning non-monotonic effects in other highly-correlated features.</p></li>
</ul>
<h2>How can we serialize EBMs and use them in production?</h2>
<p>For full functionality, we recommend using <a class="reference external" href="https://docs.python.org/3/library/pickle.html">pickle</a> to serialize and deserialize EBM objects. Explanations can be serialized as JSON through the <code class="docutils literal notranslate"><span class="pre">data</span></code> method.</p>
<p>Thanks to Github user <a class="reference external" href="https://github.com/MainRo">@MainRo</a>, there is now an <code class="docutils literal notranslate"><span class="pre">ebm2onnx</span></code> package available which enables high speed inference on EBM objects through ONNX compatible runtimes. Check out the package here: <a class="github reference external" href="https://github.com/SoftAtHome/ebm2onnx/">SoftAtHome/ebm2onnx</a> and install from PyPi via <code class="docutils literal notranslate"><span class="pre">pip</span> <span class="pre">install</span> <span class="pre">ebm2onnx</span></code></p>
<h2>Support</h2>
<h2>I need help, how do I get in touch?</h2>
<p>For most questions we recommend raising a <a class="reference external" href="https://github.com/interpretml/interpret/issues">GitHub issue</a>. Solving issues on GitHub means that other users like yourself can benefit from the same solutions. For anything else (or questions that need to be kept private), feel free to send us an email at <a class="reference external" href="mailto:interpret%40microsoft.com">interpret<span>@</span>microsoft<span>.</span>com</a> .</p>
<h2>I have a great idea, how can I contribute?</h2>
<p>If you have code you’d like to commit, make sure you read the <a class="reference external" href="https://github.com/interpretml/interpret/blob/master/CONTRIBUTING.md">contributing guidelines</a>, and send us a pull request.</p>
<p>If you’d like to request a new feature or discuss any ideas, just raise a <a class="reference external" href="https://github.com/interpretml/interpret/issues">GitHub issue</a>.</p>
</section>
<script type="text/x-thebe-config">
{
requestKernel: true,
binderOptions: {
repo: "binder-examples/jupyter-stacks-datascience",
ref: "master",
},
codeMirrorConfig: {
theme: "abcdef",
mode: "python"
},
kernelOptions: {
name: "python3",
path: "./."
},
predefinedOutput: true
}
</script>
<script>kernelName = 'python3'</script>
</article>
<footer class="prev-next-footer d-print-none">
<div class="prev-next-area">
<a class="left-prev"
href="debugging-guide.html"
title="previous page">
<i class="fa-solid fa-angle-left"></i>
<div class="prev-next-info">
<p class="prev-next-subtitle">previous</p>
<p class="prev-next-title">Logging and Debugging</p>
</div>
</a>
<a class="right-next"
href="examples.html"
title="next page">
<div class="prev-next-info">
<p class="prev-next-subtitle">next</p>
<p class="prev-next-title">Examples</p>
</div>
<i class="fa-solid fa-angle-right"></i>
</a>
</div>
</footer>
</div>
</div>
<footer class="bd-footer-content">
<div class="bd-footer-content__inner container">
<div class="footer-item">
<p class="component-author">
By The InterpretML Contributors
</p>
</div>
<div class="footer-item">
<p class="copyright">
© Copyright 2023.
<br/>
</p>
</div>
<div class="footer-item">
</div>
<div class="footer-item">
</div>
</div>
</footer>
</main>
</div>
</div>
<!-- Scripts loaded after <body> so the DOM is not blocked -->
<script src="_static/scripts/bootstrap.js?digest=dfe6caa3a7d634c4db9b"></script>
<script src="_static/scripts/pydata-sphinx-theme.js?digest=dfe6caa3a7d634c4db9b"></script>
<footer class="bd-footer">
</footer>
</body>
</html>