Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
bench
Search
Jim Hester
September 19, 2018
Programming
1
1.8k
bench
The bench package, a new take in benchmarking for R
Jim Hester
September 19, 2018
Tweet
Share
More Decks by Jim Hester
See All by Jim Hester
cpp11 - welding R and C++
jimhester
0
2.5k
GitHub Actions for R
jimhester
10
5.3k
vroom
jimhester
1
6.1k
it depends
jimhester
5
3.8k
Extending R Syntax: in package space
jimhester
1
980
Glue strings to data with glue
jimhester
0
230
You can make a package in 20 minutes
jimhester
1
98
Other Decks in Programming
See All in Programming
The Past, Present, and Future of Enterprise Java
ivargrimstad
0
330
AIに仕事を丸投げしたら、本当に楽になれるのか
dip_tech
PRO
0
170
15年目のiOSアプリを1から作り直す技術
teakun
0
560
LangChain4jとは一味違うLangChain4j-CDI
kazumura
1
120
izumin5210のプロポーザルのネタ探し #tskaigi_msup
izumin5210
1
490
new(1.26) ← これすき / kamakura.go #8
utgwkk
0
160
「ブロックテーマでは再現できない」は本当か?
inc2734
0
1.1k
Rails Girls Tokyo 18th GMO Pepabo Sponsor Talk
yutokyokutyo
0
180
クライアントワークでSREをするということ。あるいは事業会社におけるSREと同じこと・違うこと
nnaka2992
0
100
生成AIを使ったコードレビューで定性的に品質カバー
chiilog
1
310
NOT A HOTEL - 建築や人と融合し、自由を創り出すソフトウェア
not_a_hokuts
2
470
AIによる開発の民主化を支える コンテキスト管理のこれまでとこれから
mulyu
3
2.1k
Featured
See All Featured
Automating Front-end Workflow
addyosmani
1371
200k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
21
1.4k
Leveraging Curiosity to Care for An Aging Population
cassininazir
1
180
Building Better People: How to give real-time feedback that sticks.
wjessup
370
20k
We Have a Design System, Now What?
morganepeng
55
8k
What Being in a Rock Band Can Teach Us About Real World SEO
427marketing
0
180
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
2.8k
Optimising Largest Contentful Paint
csswizardry
37
3.6k
Practical Orchestrator
shlominoach
191
11k
Designing for humans not robots
tammielis
254
26k
Skip the Path - Find Your Career Trail
mkilby
0
69
Navigating Team Friction
lara
192
16k
Transcript
CC by RStudio bench ! September 19th 2018 " bench.r-lib.org
Jim Hester CC-BY-4.0 A new take on benchmarking
Why benchmark?
"premature optimization is the root of all evil." - Donald
Knuth
"A good programmer… will be wise to look carefully at
the critical code; but only a!er that code has been identified." - Donald Knuth
"All (tools) should be designed to provide (us) with feedback
indicating what parts of (our) programs are costing the most" - Donald Knuth
bench::mark()
CC by RStudio bench::mark() 1. High precision timers (ns) 2.
Memory allocations 3. Number and type of R garbage collections 4. Result equality 5. Adaptive stopping 6. Statistics filter GC effects
CC by RStudio bench::mark() 1. Human readable times 2. Human
readable memory sizes 3. Full results / timings / GCs in list- cols
bench::press()
CC by RStudio bench::press()
bench::plot()
ggplot2::autoplot(results) CC by RStudio
CC by RStudio
bench.r-lib.org install.packages("bench") jimhester_ jimhester
[email protected]