-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathtutorial.html
More file actions
657 lines (466 loc) · 45.9 KB
/
tutorial.html
File metadata and controls
657 lines (466 loc) · 45.9 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
<!DOCTYPE html>
<html lang="en" data-content_root="./" >
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<title>Tutorial — DeepQMC 1.2.0 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?v=a746c00c" />
<link rel="stylesheet" type="text/css" href="https://cdn.jsdelivr.net/npm/katex@0.16.10/dist/katex.min.css" />
<link rel="stylesheet" type="text/css" href="_static/katex-math.css?v=91adb8b6" />
<!-- 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 src="_static/documentation_options.js?v=6efca38a"></script>
<script src="_static/doctools.js?v=9bcbadda"></script>
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
<script src="_static/katex.min.js?v=be8ff15f"></script>
<script src="_static/auto-render.min.js?v=ad136472"></script>
<script src="_static/katex_autorenderer.js?v=bebc588a"></script>
<script>DOCUMENTATION_OPTIONS.pagename = 'tutorial';</script>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="API" href="api.html" />
<link rel="prev" title="Command-line interface" href="cli.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 the docs ..."
aria-label="Search the docs ..."
autocomplete="off"
autocorrect="off"
autocapitalize="off"
spellcheck="false"/>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form></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">
<div class="bd-header__inner bd-page-width">
<button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
<span class="fa-solid fa-bars"></span>
</button>
<div class="col-lg-3 navbar-header-items__start">
<div class="navbar-item">
<a class="navbar-brand logo" href="index.html">
<p class="title logo__title">DeepQMC 1.2.0 documentation</p>
</a></div>
</div>
<div class="col-lg-9 navbar-header-items">
<div class="me-auto navbar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="installation.html">
Installation
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="cli.html">
Command-line interface
</a>
</li>
<li class="nav-item current active">
<a class="nav-link nav-internal" href="#">
Tutorial
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="api.html">
API
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="refs.html">
References
</a>
</li>
</ul>
</nav></div>
</div>
<div class="navbar-header-items__end">
<div class="navbar-item navbar-persistent--container">
<script>
document.write(`
<button class="btn search-button-field search-button__button" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
`);
</script>
</div>
<div class="navbar-item">
<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></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/deepqmc/deepqmc" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="navbar-persistent--mobile">
<script>
document.write(`
<button class="btn search-button-field search-button__button" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
<i class="fa-solid fa-magnifying-glass"></i>
<span class="search-button__default-text">Search</span>
<span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
</button>
`);
</script>
</div>
</div>
</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 class="sidebar-header-items__center">
<div class="navbar-item">
<nav>
<ul class="bd-navbar-elements navbar-nav">
<li class="nav-item ">
<a class="nav-link nav-internal" href="installation.html">
Installation
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="cli.html">
Command-line interface
</a>
</li>
<li class="nav-item current active">
<a class="nav-link nav-internal" href="#">
Tutorial
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="api.html">
API
</a>
</li>
<li class="nav-item ">
<a class="nav-link nav-internal" href="refs.html">
References
</a>
</li>
</ul>
</nav></div>
</div>
<div class="sidebar-header-items__end">
<div class="navbar-item">
<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></div>
<div class="navbar-item"><ul class="navbar-icon-links"
aria-label="Icon Links">
<li class="nav-item">
<a href="https://github.com/deepqmc/deepqmc" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-github fa-lg" aria-hidden="true"></i>
<span class="sr-only">GitHub</span></a>
</li>
</ul></div>
</div>
</div>
<div class="sidebar-primary-items__start sidebar-primary__section">
<div class="sidebar-primary-item">
<div
id="pst-page-navigation-heading-2"
class="page-toc tocsection onthispage">
<i class="fa-solid fa-list"></i> On this page
</div>
<nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
<ul class="visible nav section-nav flex-column">
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#create-a-molecule">Create a molecule</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#create-the-molecular-hamiltonian">Create the molecular Hamiltonian</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#create-a-wave-function-ansatz">Create a wave function ansatz</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#instantiate-a-sampler">Instantiate a sampler</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#optimize-the-ansatz">Optimize the ansatz</a><ul class="nav section-nav flex-column">
<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#optimizing-electronic-excited-states">Optimizing electronic excited states</a></li>
</ul>
</li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#logging">Logging</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#evaluate-the-energy">Evaluate the energy</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#pseudopotentials">Pseudopotentials</a></li>
</ul>
</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="bd-content">
<div class="bd-article-container">
<div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
<div class="header-article-items__start">
<div class="header-article-item">
<nav aria-label="Breadcrumb" class="d-print-none">
<ul class="bd-breadcrumbs">
<li class="breadcrumb-item breadcrumb-home">
<a href="index.html" class="nav-link" aria-label="Home">
<i class="fa-solid fa-home"></i>
</a>
</li>
<li class="breadcrumb-item active" aria-current="page">Tutorial</li>
</ul>
</nav>
</div>
</div>
</div>
</div>
<div id="searchbox"></div>
<article class="bd-article">
<section id="tutorial">
<span id="id1"></span><h1>Tutorial<a class="headerlink" href="#tutorial" title="Link to this heading">#</a></h1>
<p>This section exemplifies the use of the API of DeepQMC. For running calculations it is recommended to use the high-level command line API, which can be fully configured through hydra (see <a class="reference internal" href="cli.html#cli"><span class="std std-ref">cli</span></a>). For further information and a more detailed descriptions of the functions presented here consult the <a class="reference internal" href="api.html#api"><span class="std std-ref">api</span></a> documentation and the accompanying software paper <a class="reference internal" href="refs.html#schaetzle23" id="id2"><span>[Schaetzle23]</span></a>.</p>
<section id="create-a-molecule">
<h2>Create a molecule<a class="headerlink" href="#create-a-molecule" title="Link to this heading">#</a></h2>
<p>A molecule is represented by the <code class="xref py py-class docutils literal notranslate"><span class="pre">Molecule</span></code> class in DeepQMC. The easiest way to get started is to work with one of the predefined molecules:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">deepqmc.molecule</span> <span class="kn">import</span> <span class="n">Molecule</span>
<span class="n">mol</span> <span class="o">=</span> <span class="n">Molecule</span><span class="o">.</span><span class="n">from_name</span><span class="p">(</span><span class="s1">'LiH'</span><span class="p">)</span>
</pre></div>
</div>
<p>To get all available predefined molecules use:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">Molecule</span><span class="o">.</span><span class="n">all_names</span>
<span class="go">{'B', 'B2', 'Be', ..., bicyclobutane'}</span>
</pre></div>
</div>
<p>A <code class="xref py py-class docutils literal notranslate"><span class="pre">Molecule</span></code> can be also created from scratch by specifying the nuclear coordinates and charges, as well as the total charge and spin multiplicity:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">mol</span> <span class="o">=</span> <span class="n">Molecule</span><span class="p">(</span> <span class="c1"># LiH</span>
<span class="n">coords</span><span class="o">=</span><span class="p">[[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span> <span class="p">[</span><span class="mf">3.015</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]],</span>
<span class="n">charges</span><span class="o">=</span><span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span>
<span class="n">charge</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
<span class="n">spin</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
<span class="n">unit</span><span class="o">=</span><span class="s1">'bohr'</span><span class="p">,</span>
<span class="p">)</span>
</pre></div>
</div>
</section>
<section id="create-the-molecular-hamiltonian">
<h2>Create the molecular Hamiltonian<a class="headerlink" href="#create-the-molecular-hamiltonian" title="Link to this heading">#</a></h2>
<p>From the molecule the <a class="reference internal" href="api.html#deepqmc.hamil.MolecularHamiltonian" title="deepqmc.hamil.MolecularHamiltonian"><code class="xref py py-class docutils literal notranslate"><span class="pre">MolecularHamiltonian</span></code></a> is constructed:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">deepqmc.hamil</span> <span class="kn">import</span> <span class="n">MolecularHamiltonian</span>
<span class="n">H</span> <span class="o">=</span> <span class="n">MolecularHamiltonian</span><span class="p">(</span><span class="n">mol</span><span class="o">=</span><span class="n">mol</span><span class="p">)</span>
</pre></div>
</div>
<p>The Hamiltonian provides the local energy function for the evaluation of the energy expectation value, as well as an educated guess for initial electron configurations to start the sampling.</p>
</section>
<section id="create-a-wave-function-ansatz">
<h2>Create a wave function ansatz<a class="headerlink" href="#create-a-wave-function-ansatz" title="Link to this heading">#</a></h2>
<p>The neural network wave function ansatz is available in the <code class="xref py py-mod docutils literal notranslate"><span class="pre">deepqmc.wf</span></code> submodule. A convenient way of initializing a wave function instance is to use a <code class="xref py py-mod docutils literal notranslate"><span class="pre">hydra</span></code> config file. DeepQMC comes with config files for predefined wave functions (at <code class="docutils literal notranslate"><span class="pre">../deepqmc/src/deepqmc/conf/ansatz</span></code>), however custom configurations may be used. Being a <a class="reference external" href="https://dm-haiku.readthedocs.io/en/latest/api.html#module-1" title="(in Haiku)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">haiku</span></code></a> module the ansatz has to be initialized inside a <a class="reference external" href="https://dm-haiku.readthedocs.io/en/latest/api.html#haiku.transform" title="(in Haiku)"><code class="xref py py-func docutils literal notranslate"><span class="pre">haiku.transform()</span></code></a>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span>
<span class="kn">import</span> <span class="nn">haiku</span> <span class="k">as</span> <span class="nn">hk</span>
<span class="kn">from</span> <span class="nn">hydra</span> <span class="kn">import</span> <span class="n">compose</span><span class="p">,</span> <span class="n">initialize_config_dir</span>
<span class="kn">from</span> <span class="nn">hydra.utils</span> <span class="kn">import</span> <span class="n">instantiate</span>
<span class="kn">import</span> <span class="nn">deepqmc</span>
<span class="kn">from</span> <span class="nn">deepqmc.app</span> <span class="kn">import</span> <span class="n">instantiate_ansatz</span>
<span class="n">deepqmc_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">dirname</span><span class="p">(</span><span class="n">deepqmc</span><span class="o">.</span><span class="vm">__file__</span><span class="p">)</span>
<span class="n">config_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">deepqmc_dir</span><span class="p">,</span> <span class="s1">'conf/ansatz'</span><span class="p">)</span>
<span class="k">with</span> <span class="n">initialize_config_dir</span><span class="p">(</span><span class="n">version_base</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">config_dir</span><span class="o">=</span><span class="n">config_dir</span><span class="p">):</span>
<span class="n">cfg</span> <span class="o">=</span> <span class="n">compose</span><span class="p">(</span><span class="n">config_name</span><span class="o">=</span><span class="s1">'default'</span><span class="p">)</span>
<span class="n">_ansatz</span> <span class="o">=</span> <span class="n">instantiate</span><span class="p">(</span><span class="n">cfg</span><span class="p">,</span> <span class="n">_recursive_</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">_convert_</span><span class="o">=</span><span class="s1">'all'</span><span class="p">)</span>
<span class="n">ansatz</span> <span class="o">=</span> <span class="n">instantiate_ansatz</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">_ansatz</span><span class="p">)</span>
</pre></div>
</div>
<p>The hyperparameters and their physical meaning are described in the <a class="reference internal" href="api.html#api"><span class="std std-ref">api</span></a> reference. The resulting <code class="docutils literal notranslate"><span class="pre">ansatz</span></code> object has two methods: <code class="docutils literal notranslate"><span class="pre">ansatz.init</span></code> can be used to initialize the ansatz parameters, while <code class="docutils literal notranslate"><span class="pre">ansatz.apply</span></code> evaluates the wave function.</p>
</section>
<section id="instantiate-a-sampler">
<h2>Instantiate a sampler<a class="headerlink" href="#instantiate-a-sampler" title="Link to this heading">#</a></h2>
<p>The variational Monte Carlo method requires sampling the probability density associated with the square of the wave function. A <code class="xref py py-class docutils literal notranslate"><span class="pre">Sampler</span></code> can be instantiated from a <a class="reference internal" href="api.html#deepqmc.hamil.MolecularHamiltonian" title="deepqmc.hamil.MolecularHamiltonian"><code class="xref py py-class docutils literal notranslate"><span class="pre">MolecularHamiltonian</span></code></a> and a wave function. The instantiation is handled within the <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a> wrapper of the training loop. <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a> therefore accepts a sampler factory, that is a function that constructs a <code class="xref py py-class docutils literal notranslate"><span class="pre">Sampler</span></code>.:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">deepqmc.sampling</span> <span class="kn">import</span> <span class="n">initialize_sampling</span><span class="p">,</span> <span class="n">MetropolisSampler</span><span class="p">,</span> <span class="n">DecorrSampler</span><span class="p">,</span> <span class="n">combine_samplers</span>
<span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
<span class="n">elec_sampler</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">combine_samplers</span><span class="p">,</span> <span class="n">samplers</span><span class="o">=</span><span class="p">[</span><span class="n">DecorrSampler</span><span class="p">(</span><span class="n">length</span><span class="o">=</span><span class="mi">20</span><span class="p">),</span> <span class="n">MetropolisSampler</span><span class="p">])</span>
<span class="n">sampler_factory</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">initialize_sampling</span><span class="p">,</span> <span class="n">elec_sampler</span><span class="o">=</span><span class="n">elec_sampler</span><span class="p">)</span>
</pre></div>
</div>
<p>The above example combines the <a class="reference internal" href="api.html#deepqmc.sampling.DecorrSampler" title="deepqmc.sampling.DecorrSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecorrSampler</span></code></a> and the <a class="reference internal" href="api.html#deepqmc.sampling.MetropolisSampler" title="deepqmc.sampling.MetropolisSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">MetropolisSampler</span></code></a> to create a Metropolis-Hastings sampler that performs 20 decorrelating steps each time before returning the next set of samples.</p>
</section>
<section id="optimize-the-ansatz">
<h2>Optimize the ansatz<a class="headerlink" href="#optimize-the-ansatz" title="Link to this heading">#</a></h2>
<p>The high-level <a class="reference internal" href="api.html#module-deepqmc.train" title="deepqmc.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a> function is used to train the deep neural networks in the ansatz. The train function takes a <a class="reference internal" href="api.html#deepqmc.hamil.MolecularHamiltonian" title="deepqmc.hamil.MolecularHamiltonian"><code class="xref py py-class docutils literal notranslate"><span class="pre">MolecularHamiltonian</span></code></a>, a <code class="xref py py-class docutils literal notranslate"><span class="pre">WaveFunction</span></code>, a sampler factory, and an <a class="reference internal" href="api.html#deepqmc.optimizer.Optimizer" title="deepqmc.optimizer.Optimizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">Optimizer</span></code></a>. The recommended KFAC optimizer can be instantiated using <code class="xref py py-mod docutils literal notranslate"><span class="pre">hydra</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">config_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">deepqmc_dir</span><span class="p">,</span> <span class="s1">'conf/task/opt'</span><span class="p">)</span>
<span class="k">with</span> <span class="n">initialize_config_dir</span><span class="p">(</span><span class="n">version_base</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">config_dir</span><span class="o">=</span><span class="n">config_dir</span><span class="p">):</span>
<span class="n">cfg</span> <span class="o">=</span> <span class="n">compose</span><span class="p">(</span><span class="n">config_name</span><span class="o">=</span><span class="s1">'kfac'</span><span class="p">)</span>
<span class="n">kfac</span> <span class="o">=</span> <span class="n">instantiate</span><span class="p">(</span><span class="n">cfg</span><span class="p">,</span> <span class="n">_recursive_</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">_convert_</span><span class="o">=</span><span class="s1">'all'</span><span class="p">)</span>
</pre></div>
</div>
<p>The object <code class="docutils literal notranslate"><span class="pre">kfac</span></code> can now be passed as the <code class="docutils literal notranslate"><span class="pre">opt</span></code> argument to <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a>, along with the requested number of training steps (<code class="docutils literal notranslate"><span class="pre">steps</span></code>), the number of electron position samples used in a training batch (<code class="docutils literal notranslate"><span class="pre">electron_batch_size</span></code>), and a seed (<code class="docutils literal notranslate"><span class="pre">seed</span></code>):</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">deepqmc.train</span> <span class="kn">import</span> <span class="n">train</span>
<span class="gp">>>> </span><span class="n">train</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">ansatz</span><span class="p">,</span> <span class="n">kfac</span><span class="p">,</span> <span class="n">sampler_factory</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">electron_batch_size</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">42</span><span class="p">)</span>
<span class="go">training: 0%|▋ | 102/10000 [01:00<23:01, 7.16it/s, E=-8.042(10)]</span>
</pre></div>
</div>
<p>If the argument <code class="docutils literal notranslate"><span class="pre">pretrain_steps</span></code> is set, the ansatz is pretrained with respect to a Hartree-Fock or CASSCF baseline obtained with <a class="reference external" href="https://pyscf.org/pyscf_api_docs/pyscf.html#module-pyscf" title="(in PySCF v2.7)"><code class="xref py py-mod docutils literal notranslate"><span class="pre">pyscf</span></code></a>. For more details as well as further training hyperparameters consult the <a class="reference internal" href="api.html#api"><span class="std std-ref">api</span></a> reference.</p>
<section id="optimizing-electronic-excited-states">
<h3>Optimizing electronic excited states<a class="headerlink" href="#optimizing-electronic-excited-states" title="Link to this heading">#</a></h3>
<p>DeepQMC can simultaneously optimize the lowest <code class="docutils literal notranslate"><span class="pre">n</span></code> electronic states of a molecule. By default, <code class="docutils literal notranslate"><span class="pre">n</span></code> is set to one, but it can be increased via the <code class="docutils literal notranslate"><span class="pre">electronic_states</span></code> argument to <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a>. In order to efficiently pretrain for more than one electronic state, a CASSCF pretraining target should be used, with an active space that contains at least <code class="docutils literal notranslate"><span class="pre">n</span></code> states.</p>
<p>For a detailed description of the excited states methodology, see <a class="reference internal" href="refs.html#szabo24" id="id3"><span>[Szabo24]</span></a>.</p>
</section>
</section>
<section id="logging">
<span id="id4"></span><h2>Logging<a class="headerlink" href="#logging" title="Link to this heading">#</a></h2>
<p>The terminal output shows only how far has the training progressed and the current estimate of the energy. More detailed monitoring of the training is available via <a class="reference external" href="https://www.tensorflow.org/tensorboard">Tensorboard</a>. When <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a> is called with an optional <code class="docutils literal notranslate"><span class="pre">workdir</span></code> argument, the training run creates a Tensorboard event file:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">train</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">ansatz</span><span class="p">,</span> <span class="n">kfac</span><span class="p">,</span> <span class="n">sampler_factory</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">electron_batch_size</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">42</span><span class="p">,</span> <span class="n">workdir</span><span class="o">=</span><span class="s1">'runs/01'</span><span class="p">)</span>
</pre></div>
</div>
<div class="highlight-none notranslate"><div class="highlight"><pre><span></span>$ tensorboard --logdir runs/
TensorFlow installation not found - running with reduced feature set.
Serving TensorBoard on localhost; to expose to the network, use a proxy or pass --bind_all
TensorBoard 2.11.0 at http://localhost:6006/ (Press CTRL+C to quit)
</pre></div>
</div>
<p>This launches a Tensorboard server which can be accessed via a web browser at the printed URL.</p>
<p>Furthermore, several other quantities are dumped during the training to the <code class="docutils literal notranslate"><span class="pre">workdir</span></code>. The <code class="docutils literal notranslate"><span class="pre">training</span></code> directory contains training checkpoints as well as an HDF5 file <code class="docutils literal notranslate"><span class="pre">result.h5</span></code> that holds the local energies throughout the training, an exponential moving average of the training energy and the values of the wave function at every iteration:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">h5py</span>
<span class="gp">>>> </span><span class="k">with</span> <span class="n">h5py</span><span class="o">.</span><span class="n">File</span><span class="p">(</span><span class="s1">'runs/01/training/result.h5'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span> <span class="nb">print</span><span class="p">(</span><span class="n">f</span><span class="o">.</span><span class="n">keys</span><span class="p">(),</span><span class="n">f</span><span class="p">[</span><span class="s1">'local_energy'</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
<span class="go"><KeysViewHDF5 ['local_energy']> <KeysViewHDF5 ['max', 'mean', 'min', 'samples', 'std']></span>
</pre></div>
</div>
<p>Additional observables can also be computed and logged during a run by specifying the <code class="docutils literal notranslate"><span class="pre">observable_monitors</span></code> argument to <a class="reference internal" href="api.html#deepqmc.train.train" title="deepqmc.train.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a>.
For example, to evaluate the spin of the wave function during training one can use the <code class="xref py py-class docutils literal notranslate"><span class="pre">SpinMonitor</span></code> class:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">deepqmc.observable</span> <span class="kn">import</span> <span class="n">SpinMonitor</span>
<span class="n">observable_monitors</span> <span class="o">=</span> <span class="p">[</span><span class="n">SpinMonitor</span><span class="p">(</span><span class="n">period</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">save_samples</span><span class="o">=</span><span class="kc">True</span><span class="p">)]</span>
<span class="n">train</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">ansatz</span><span class="p">,</span> <span class="n">kfac</span><span class="p">,</span> <span class="n">sampler_factory</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">electron_batch_size</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">42</span><span class="p">,</span> <span class="n">workdir</span><span class="o">=</span><span class="s1">'runs/02'</span><span class="p">,</span> <span class="n">observable_monitors</span><span class="o">=</span><span class="n">observable_monitors</span><span class="p">)</span>
</pre></div>
</div>
<p>Then the <code class="docutils literal notranslate"><span class="pre">runs/02/training/result.h5</span></code> file will contain the keys <code class="docutils literal notranslate"><span class="pre">spin/samples</span></code>, <code class="docutils literal notranslate"><span class="pre">spin/mean</span></code>, and <code class="docutils literal notranslate"><span class="pre">spin/std</span></code>.
Note: since logging all samples can result in very large <code class="docutils literal notranslate"><span class="pre">result.h5</span></code> files, it may be useful to disable saving of individual samples by setting save_samples=False if not explicitly required.
See also the <code class="xref py py-mod docutils literal notranslate"><span class="pre">observable</span></code> submodule for more details.</p>
</section>
<section id="evaluate-the-energy">
<h2>Evaluate the energy<a class="headerlink" href="#evaluate-the-energy" title="Link to this heading">#</a></h2>
<p>A rough estimate of the expectation value of the energy of a trained wave function can be obtained already from the local energies of the training run. A rigorous estimation of the energy expectation value up to the statistical sampling error can be obtained when evaluating the energy expectation value of the trained wavefunction without further optimization. This is achieved by passing a training checkpoint to the <a class="reference internal" href="api.html#module-deepqmc.train" title="deepqmc.train"><code class="xref py py-func docutils literal notranslate"><span class="pre">train()</span></code></a> function, and specifying the optimizer to be <code class="docutils literal notranslate"><span class="pre">None</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">deepqmc.log</span> <span class="kn">import</span> <span class="n">CheckpointStore</span>
<span class="gp">>>> </span><span class="n">step</span><span class="p">,</span> <span class="n">train_state</span> <span class="o">=</span><span class="n">CheckpointStore</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s1">'runs/01/training/chkpt-10000.pt'</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">train</span><span class="p">(</span><span class="n">H</span><span class="p">,</span> <span class="n">ansatz</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="n">sampler_factory</span><span class="p">,</span> <span class="n">train_state</span><span class="o">=</span><span class="n">train_state</span><span class="p">,</span> <span class="n">steps</span><span class="o">=</span><span class="mi">500</span><span class="p">,</span> <span class="n">electron_batch_size</span><span class="o">=</span><span class="mi">2000</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="mi">42</span><span class="p">,</span> <span class="n">workdir</span><span class="o">=</span><span class="s1">'runs/01'</span><span class="p">)</span>
<span class="go">evaluating: 100%|█████████| 500/500 [01:20<00:00, 6.20it/s, E=-8.07000(19)]</span>
</pre></div>
</div>
<p>The evaluation generates the same type of logs as the training, but writes to <code class="docutils literal notranslate"><span class="pre">workdir/evaluation</span></code> instead. The final energy can be read from the progress bar, the Tensorboard event file or computed from the local energies logged to the <code class="docutils literal notranslate"><span class="pre">workdir/evaluation/result.h5</span></code> file.</p>
</section>
<section id="pseudopotentials">
<h2>Pseudopotentials<a class="headerlink" href="#pseudopotentials" title="Link to this heading">#</a></h2>
<p>DeepQMC currently supports <code class="docutils literal notranslate"><span class="pre">bfd</span></code> <a class="reference internal" href="refs.html#burkatzki07" id="id5"><span>[Burkatzki07]</span></a> and <code class="docutils literal notranslate"><span class="pre">ccECP</span></code> <a class="reference internal" href="refs.html#bennett17" id="id6"><span>[Bennett17]</span></a> pseudopotentials, which can be enabled by passing the <code class="docutils literal notranslate"><span class="pre">ecp_type</span></code> argument to the Hamiltonian definition. This replaces a certain number of core electrons with a pseudopotential, reducing the total number of electrons explicitly treated and thus decreasing the computational cost. The pseudopotentials for all nuclei heavier than He in the molecule will be used if the argument <code class="docutils literal notranslate"><span class="pre">ecp_type</span></code> is passed. They can be turned off or on for individual nuclei by specifying <code class="docutils literal notranslate"><span class="pre">pp_mask</span></code>, a boolean array with <code class="docutils literal notranslate"><span class="pre">True</span></code> (<code class="docutils literal notranslate"><span class="pre">False</span></code>) for each nucleus with pseudopotential turned on (off). The following example defines the Hamiltonian of a TiO molecule where the titanium core is replaced by a pseudopotential and the oxygen core is left unaffected:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">mol</span> <span class="o">=</span> <span class="n">Molecule</span><span class="p">(</span> <span class="c1"># TiO</span>
<span class="n">coords</span><span class="o">=</span><span class="p">[[</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">],</span> <span class="p">[</span><span class="mf">1.668</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">]],</span>
<span class="n">charges</span><span class="o">=</span><span class="p">[</span><span class="mi">22</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span>
<span class="n">charge</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
<span class="n">spin</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
<span class="n">unit</span><span class="o">=</span><span class="s1">'angstrom'</span><span class="p">,</span>
<span class="p">)</span>
<span class="n">H</span> <span class="o">=</span> <span class="n">MolecularHamiltonian</span><span class="p">(</span><span class="n">mol</span><span class="o">=</span><span class="n">mol</span><span class="p">,</span> <span class="n">ecp_type</span><span class="o">=</span><span class="s1">'ccECP'</span><span class="p">,</span> <span class="n">ecp_mask</span><span class="o">=</span><span class="p">[</span><span class="kc">True</span><span class="p">,</span><span class="kc">False</span><span class="p">],</span> <span class="n">elec_std</span><span class="o">=</span><span class="mf">0.1</span><span class="p">)</span>
</pre></div>
</div>
<p>Systems containing heavier atoms sometimes tend to produce NaN errors. To avoid these issues, it was found useful to use a smaller variance for the initial distribution of electrons around the nuclei (via the <code class="docutils literal notranslate"><span class="pre">elec_std</span></code> argument) and a larger decorrelation length for sampling.</p>
</section>
</section>
</article>
<footer class="prev-next-footer d-print-none">
<div class="prev-next-area">
<a class="left-prev"
href="cli.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">Command-line interface</p>
</div>
</a>
<a class="right-next"
href="api.html"
title="next page">
<div class="prev-next-info">
<p class="prev-next-subtitle">next</p>
<p class="prev-next-title">API</p>
</div>
<i class="fa-solid fa-angle-right"></i>
</a>
</div>
</footer>
</div>
</div>
<footer class="bd-footer-content">
</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">
<div class="bd-footer__inner bd-page-width">
<div class="footer-items__start">
<div class="footer-item">
<p class="copyright">
© Copyright 2019-2024, Frank Noé and collaborators.
<br/>
</p>
</div>
</div>
<div class="footer-items__end">
<div class="footer-item">
<p class="theme-version">
Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.15.4.
</p></div>
</div>
</div>
</footer>
</body>
</html>