forked from hpc2n/R_for_HPC
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Profiling.html
504 lines (394 loc) · 20.1 KB
/
Profiling.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
<!DOCTYPE html>
<html>
<head>
<title>Profiling your code</title>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="chrome=1">
<meta name="generator" content="pandoc" />
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta name="apple-mobile-web-app-capable" content="yes">
<base target="_blank">
<script type="text/javascript">
var SLIDE_CONFIG = {
// Slide settings
settings: {
title: 'Profiling your code',
useBuilds: true,
usePrettify: true,
enableSlideAreas: true,
enableTouch: true,
favIcon: 'Profiling_files/logo.png',
},
// Author information
presenters: [
{
name: 'Pedro Ojeda' ,
company: '',
gplus: '',
twitter: '',
www: '',
github: ''
},
]
};
</script>
<script src="site_libs/header-attrs-2.7/header-attrs.js"></script>
<link href="site_libs/ioslides-13.5.1/fonts/fonts.css" rel="stylesheet" />
<link href="site_libs/ioslides-13.5.1/theme/css/default.css" rel="stylesheet" />
<link href="site_libs/ioslides-13.5.1/theme/css/phone.css" rel="stylesheet" />
<script src="site_libs/ioslides-13.5.1/js/modernizr.custom.45394.js"></script>
<script src="site_libs/ioslides-13.5.1/js/prettify/prettify.js"></script>
<script src="site_libs/ioslides-13.5.1/js/prettify/lang-r.js"></script>
<script src="site_libs/ioslides-13.5.1/js/prettify/lang-yaml.js"></script>
<script src="site_libs/ioslides-13.5.1/js/hammer.js"></script>
<script src="site_libs/ioslides-13.5.1/js/slide-controller.js"></script>
<script src="site_libs/ioslides-13.5.1/js/slide-deck.js"></script>
<style type="text/css">
b, strong {
font-weight: bold;
}
em {
font-style: italic;
}
summary {
display: list-item;
}
slides > slide {
-webkit-transition: all 0.4s ease-in-out;
-moz-transition: all 0.4s ease-in-out;
-o-transition: all 0.4s ease-in-out;
transition: all 0.4s ease-in-out;
}
.auto-fadein {
-webkit-transition: opacity 0.6s ease-in;
-webkit-transition-delay: 0.4s;
-moz-transition: opacity 0.6s ease-in 0.4s;
-o-transition: opacity 0.6s ease-in 0.4s;
transition: opacity 0.6s ease-in 0.4s;
opacity: 0;
}
/* https://github.com/ropensci/plotly/pull/524#issuecomment-468142578 */
slide:not(.current) .plotly.html-widget{
display: block;
}
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
span.underline{text-decoration: underline;}
div.column{display: inline-block; vertical-align: top; width: 50%;}
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
ul.task-list{list-style: none;}
pre > code.sourceCode { white-space: pre; position: relative; }
pre > code.sourceCode > span { display: inline-block; line-height: 1.25; }
pre > code.sourceCode > span:empty { height: 1.2em; }
code.sourceCode > span { color: inherit; text-decoration: inherit; }
div.sourceCode { margin: 1em 0; }
pre.sourceCode { margin: 0; }
@media screen {
div.sourceCode { overflow: auto; }
}
@media print {
pre > code.sourceCode { white-space: pre-wrap; }
pre > code.sourceCode > span { text-indent: -5em; padding-left: 5em; }
}
pre.numberSource code
{ counter-reset: source-line 0; }
pre.numberSource code > span
{ position: relative; left: -4em; counter-increment: source-line; }
pre.numberSource code > span > a:first-child::before
{ content: counter(source-line);
position: relative; left: -1em; text-align: right; vertical-align: baseline;
border: none; display: inline-block;
-webkit-touch-callout: none; -webkit-user-select: none;
-khtml-user-select: none; -moz-user-select: none;
-ms-user-select: none; user-select: none;
padding: 0 4px; width: 4em;
color: #aaaaaa;
}
pre.numberSource { margin-left: 3em; border-left: 1px solid #aaaaaa; padding-left: 4px; }
div.sourceCode
{ }
@media screen {
pre > code.sourceCode > span > a:first-child::before { text-decoration: underline; }
}
code span.al { color: #ff0000; font-weight: bold; } /* Alert */
code span.an { color: #60a0b0; font-weight: bold; font-style: italic; } /* Annotation */
code span.at { color: #7d9029; } /* Attribute */
code span.bn { color: #40a070; } /* BaseN */
code span.bu { } /* BuiltIn */
code span.cf { color: #007020; font-weight: bold; } /* ControlFlow */
code span.ch { color: #4070a0; } /* Char */
code span.cn { color: #880000; } /* Constant */
code span.co { color: #60a0b0; font-style: italic; } /* Comment */
code span.cv { color: #60a0b0; font-weight: bold; font-style: italic; } /* CommentVar */
code span.do { color: #ba2121; font-style: italic; } /* Documentation */
code span.dt { color: #902000; } /* DataType */
code span.dv { color: #40a070; } /* DecVal */
code span.er { color: #ff0000; font-weight: bold; } /* Error */
code span.ex { } /* Extension */
code span.fl { color: #40a070; } /* Float */
code span.fu { color: #06287e; } /* Function */
code span.im { } /* Import */
code span.in { color: #60a0b0; font-weight: bold; font-style: italic; } /* Information */
code span.kw { color: #007020; font-weight: bold; } /* Keyword */
code span.op { color: #666666; } /* Operator */
code span.ot { color: #007020; } /* Other */
code span.pp { color: #bc7a00; } /* Preprocessor */
code span.sc { color: #4070a0; } /* SpecialChar */
code span.ss { color: #bb6688; } /* SpecialString */
code span.st { color: #4070a0; } /* String */
code span.va { color: #19177c; } /* Variable */
code span.vs { color: #4070a0; } /* VerbatimString */
code span.wa { color: #60a0b0; font-weight: bold; font-style: italic; } /* Warning */
slides > slide:not(.nobackground):before {
font-size: 12pt;
content: "";
position: absolute;
bottom: 20px;
left: 60px;
background: url(Profiling_files/logo.png) no-repeat 0 50%;
-webkit-background-size: 30px 30px;
-moz-background-size: 30px 30px;
-o-background-size: 30px 30px;
background-size: 30px 30px;
padding-left: 40px;
height: 30px;
line-height: 1.9;
}
</style>
<link rel="stylesheet" href="custom.css" type="text/css" />
</head>
<body style="opacity: 0">
<slides class="layout-widescreen">
<slide class="title-slide segue nobackground">
<aside class="gdbar"><img src="Profiling_files/logo.png"></aside>
<!-- The content of this hgroup is replaced programmatically through the slide_config.json. -->
<hgroup class="auto-fadein">
<h1 data-config-title><!-- populated from slide_config.json --></h1>
<p data-config-presenter><!-- populated from slide_config.json --></p>
<p style="margin-top: 6px; margin-left: -2px;">Feb., 2021</p>
</hgroup>
</slide>
<slide class=""><hgroup><h2>Profiling</h2></hgroup><article id="profiling">
<ul>
<li>summary of the times spent in different function calls</li>
<li>memory usage report</li>
</ul>
</article></slide><slide class=""><hgroup><h2>Pi calculation</h2></hgroup><article id="pi-calculation">
<img width="100px" src='Images/MC1.png' title=''/>
<p>\(\textrm{Surface circle} = \left ( \frac{\textrm{Surface circle}}{\textrm{Surface square}} \right ) * (\textrm{Surface square})\)</p>
<p>is always valid. Knowing that \(\textrm{Surface circle} = \pi * r^2\), \(\pi\) can be computed as:</p>
<p>\(\pi = \frac{1}{r^2} \left ( \frac{\textrm{Surface circle}}{\textrm{Surface square}} \right ) * (\textrm{Surface square})\)</p>
<p>the ratio in parentheses is approximated with a Monte Carlo process throwing random points</p>
</article></slide><slide class=""><hgroup><h2>Pi calculation</h2></hgroup><article id="pi-calculation-1">
<div style="float: left; width: 50%;">
<img width="250px" src='Images/MC2.png' title='fig:'/><p class='caption'>Surface ratio</p>
<ul>
<li>The R function to compute Pi is:</li>
</ul></div>
<div style="float: right; width: 50%;">
<pre class = 'prettyprint lang-r'>sim <- function(l) {
c <- rep(0,l)
hits <- 0
pow2 <- function(x) {
x2 <- sqrt( x[1]*x[1]+x[2]*x[2] )
return(x2)
}
for(i in 1:l){
x = runif(2,-1,1)
if( pow2(x) <=1 ){
hits <- hits + 1
}
dens <- hits/i
pi_partial = dens*4
c[i] = pi_partial
}
return(c)
}</pre></div>
</article></slide><slide class=""><hgroup><h2>Pi calculation</h2></hgroup><article id="pi-calculation-2">
<p>The accuracy of the calculation increases with the number of iterations</p>
<pre class = 'prettyprint lang-r'>size <- 100000
res <- sim(size)
plot(res[1:size],type='l', xlab="Nr. iterations", ylab="Pi")
lines(rep(pi,size)[1:size], col = 'red')</pre>
<p><img src="Profiling_files/figure-html/unnamed-chunk-2-1.png" width="480" style="display: block; margin: auto;" /></p>
</article></slide><slide class=""><hgroup><h2>Monitoring the execution time</h2></hgroup><article id="monitoring-the-execution-time">
<h3>System.time</h3>
<p>This function is included in R by default</p>
<pre class = 'prettyprint lang-r'>size <- 500000
system.time(
res <- sim(size)
)</pre>
<pre >## user system elapsed
## 1.75 0.02 1.83</pre>
</article></slide><slide class=""><hgroup><h2>Monitoring the execution time</h2></hgroup><article id="monitoring-the-execution-time-1">
<h3>Tic toc</h3>
<p>Another way to obtain execution times is by using the tictoc package:</p>
<pre class = 'prettyprint lang-r'>install.packages("tictoc")</pre>
<p>one can nest tic and toc calls and save the outputs to a log file:</p>
</article></slide><slide class=""><hgroup><h2>Monitoring the execution time</h2></hgroup><article id="monitoring-the-execution-time-2">
<h3>Tic toc</h3>
<pre class = 'prettyprint lang-r'>library("tictoc")
size <- 1000000
sim2 <- function(l) {
c <- rep(0,l)
hits <- 0
pow2 <- function(x) { x2 <- sqrt( x[1]*x[1]+x[2]*x[2] ); return(x2) }
tic("only for-loop")
for(i in 1:l){
x = runif(2,-1,1)
if( pow2(x) <=1 ){ hits <- hits + 1 }
dens <- hits/i; pi_partial = dens*4; c[i] = pi_partial
}
toc(log = TRUE)
return(c)
}</pre>
</article></slide><slide class=""><hgroup><h2>Monitoring the execution time</h2></hgroup><article id="monitoring-the-execution-time-3">
<h3>Tic toc</h3>
<pre class = 'prettyprint lang-r'>tic("Total execution time")
res <- sim2(size)</pre>
<pre >## only for-loop: 3.32 sec elapsed</pre>
<pre class = 'prettyprint lang-r'>toc(log = TRUE)</pre>
<pre >## Total execution time: 3.36 sec elapsed</pre>
</article></slide><slide class=""><hgroup><h2>Monitoring the execution time</h2></hgroup><article id="monitoring-the-execution-time-4">
<h3>Tic toc</h3>
<pre class = 'prettyprint lang-r'>tic.log()</pre>
<pre >## [[1]]
## [1] "only for-loop: 3.32 sec elapsed"
##
## [[2]]
## [1] "Total execution time: 3.36 sec elapsed"</pre>
<pre class = 'prettyprint lang-r'>tic.clearlog()</pre>
</article></slide><slide class=""><hgroup><h2>Rprof</h2></hgroup><article id="rprof">
<p>Rprof should be present in your R installation. For a graphical analysis, we will use <em>proftools</em> package. One needs to install this package in case it is not already installed. For R versions < 3.5 the instructions are:</p>
<pre class = 'prettyprint lang-r'>install.packages("proftools")
source("http://bioconductor.org/biocLite.R")
biocLite(c("graph","Rgraphviz"))</pre>
<p>while for R > 3.5 one needs to do</p>
<pre class = 'prettyprint lang-r'>install.packages("proftools")
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install()
BiocManager::install(c("graph","Rgraphviz"))</pre>
</article></slide><slide class=""><hgroup><h2>Rprof</h2></hgroup><article id="rprof-1">
<p>the profiling is performed with the following lines:</p>
<pre class = 'prettyprint lang-r'>size <- 500000
Rprof("Rprof.out")
res <- sim(size)
Rprof(NULL)</pre>
</article></slide><slide class=""><hgroup><h2>Rprof</h2></hgroup><article id="rprof-2">
<p>the profiling is performed with the following lines:</p>
<pre class = 'prettyprint lang-r'>summaryRprof("Rprof.out") </pre>
<pre >## $by.self
## self.time self.pct total.time total.pct
## "runif" 0.80 50.63 0.80 50.63
## "sim" 0.52 32.91 1.58 100.00
## "pow2" 0.26 16.46 0.26 16.46
##
## $by.total
## total.time total.pct self.time self.pct
## "sim" 1.58 100.00 0.52 32.91
## "block_exec" 1.58 100.00 0.00 0.00
## "call_block" 1.58 100.00 0.00 0.00
## "eval" 1.58 100.00 0.00 0.00
## "evaluate" 1.58 100.00 0.00 0.00
## "evaluate::evaluate" 1.58 100.00 0.00 0.00
## "evaluate_call" 1.58 100.00 0.00 0.00
## "FUN" 1.58 100.00 0.00 0.00
## "generator$render" 1.58 100.00 0.00 0.00
## "handle" 1.58 100.00 0.00 0.00
## "in_dir" 1.58 100.00 0.00 0.00
## "knitr::knit" 1.58 100.00 0.00 0.00
## "lapply" 1.58 100.00 0.00 0.00
## "process_file" 1.58 100.00 0.00 0.00
## "process_group" 1.58 100.00 0.00 0.00
## "process_group.block" 1.58 100.00 0.00 0.00
## "render" 1.58 100.00 0.00 0.00
## "render_one" 1.58 100.00 0.00 0.00
## "rmarkdown::render" 1.58 100.00 0.00 0.00
## "rmarkdown::render_site" 1.58 100.00 0.00 0.00
## "sapply" 1.58 100.00 0.00 0.00
## "suppressMessages" 1.58 100.00 0.00 0.00
## "timing_fn" 1.58 100.00 0.00 0.00
## "withCallingHandlers" 1.58 100.00 0.00 0.00
## "withVisible" 1.58 100.00 0.00 0.00
## "runif" 0.80 50.63 0.80 50.63
## "pow2" 0.26 16.46 0.26 16.46
##
## $sample.interval
## [1] 0.02
##
## $sampling.time
## [1] 1.58</pre>
</article></slide><slide class=""><hgroup><h2>Rprof</h2></hgroup><article id="rprof-3">
<p>here you can see that the functions <em>runif</em> and <em>pow2</em> are the most expensive parts in our code. A graphical output can be obtained through the <em>proftools</em> package:</p>
<pre class = 'prettyprint lang-r'>library(proftools)
p <- readProfileData(filename = "Rprof.out")</pre>
</article></slide><slide class=""><hgroup><h2>Rprof</h2></hgroup><article id="rprof-4">
<pre class = 'prettyprint lang-r'>plotProfileCallGraph(p, style=google.style, score="total")</pre>
<p><img src="Profiling_files/figure-html/unnamed-chunk-13-1.png" width="720" /></p>
</article></slide><slide class=""><hgroup><h2>Rbenchmark</h2></hgroup><article id="rbenchmark">
<p>One most probably needs to install this package as it is not included by default in R installations:</p>
<pre class = 'prettyprint lang-r'>install.packages("rbenchmark")</pre>
<p>then we can benchmark our function <em>sim()</em></p>
<pre class = 'prettyprint lang-r'>library(rbenchmark)
size <- 500000
bench <- benchmark(sim(size), replications=10)</pre>
</article></slide><slide class=""><hgroup><h2>Rbenchmark</h2></hgroup><article id="rbenchmark-1">
<pre class = 'prettyprint lang-r'>bench </pre>
<pre >## test replications elapsed relative user.self sys.self user.child sys.child
## 1 sim(size) 10 15.69 1 15.65 0 NA NA</pre>
<p>the elapsed time is an average over the 10 replications we especified in the benchmark function.</p>
</article></slide><slide class=""><hgroup><h2>Microbenchmark</h2></hgroup><article id="microbenchmark">
<p>If this package is not installed, do as usual:</p>
<pre class = 'prettyprint lang-r'>install.packages("microbenchmark")</pre>
<p>and do the benchmarking with:</p>
<pre class = 'prettyprint lang-r'>library(microbenchmark)
bench2 <- microbenchmark(sim(size), times=10)</pre>
</article></slide><slide class=""><hgroup><h2>Microbenchmark</h2></hgroup><article id="microbenchmark-1">
<pre class = 'prettyprint lang-r'>bench2 </pre>
<pre >## Unit: seconds
## expr min lq mean median uq max neval
## sim(size) 1.548863 1.567254 1.573243 1.575009 1.581044 1.601857 10</pre>
<p>in this case we obtain more statistics of the benchmarking process like the <em>mean</em>, <em>min</em>, <em>max</em>, …</p>
</article></slide><slide class=""><hgroup><h2>Summary</h2></hgroup><article id="summary">
<ul>
<li><p>Timing your R code is useful to see what parts require optimization or a better package.</p></li>
<li><p><strong>system.time</strong> and <strong>tic-toc</strong> will give you a single evaluation of the time taken by some R code</p></li>
<li><p><strong>rbenchmark</strong>, <strong>microbenchmark</strong> functions will give statistics over independent replicas of the code</p></li>
<li><p>More useful information from profiling functions will be obtained if one uses functions to enclose independent tasks in your code (remember <em>pow2</em>, <em>runif</em> in the Pi calculation)</p></li>
<li><p>Once you know what are the bottlenecks of your code, working on a few of the most expensive ones could be more effective than working on many less significative functions</p></li>
</ul>
</article></slide><slide class=""><hgroup><h2>References</h2></hgroup><article id="references">
<ul>
<li><a href='https://swcarpentry.github.io/r-novice-inflammation/' title=''>https://swcarpentry.github.io/r-novice-inflammation/</a></li>
<li><a href='https://www.tutorialspoint.com/r/index.htm' title=''>https://www.tutorialspoint.com/r/index.htm</a></li>
<li>R High Performance Programming. Aloysius, Lim; William, Tjhi. Packt Publishing, 2015.</li>
<li><a href='https://www.r-bloggers.com/estimation-of-the-number-pi-a-monte-carlo-simulation/' title=''>Pi calculation</a></li>
</ul>
<p><a href='index.html' title=''>Return to Index</a></p></article></slide>
<slide class="backdrop"></slide>
</slides>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
var script = document.createElement("script");
script.type = "text/javascript";
script.src = "https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML";
document.getElementsByTagName("head")[0].appendChild(script);
})();
</script>
<!-- map slide visiblity events into shiny -->
<script>
(function() {
if (window.jQuery) {
window.jQuery(document).on('slideleave', function(e) {
window.jQuery(e.target).trigger('hidden');
});
window.jQuery(document).on('slideenter', function(e) {
window.jQuery(e.target).trigger('shown');
});
}
})();
</script>
</body>
</html>