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
Introduction to GPU Programming in Python
Search
Sponsored
·
Ship Features Fearlessly
Turn features on and off without deploys. Used by thousands of Ruby developers.
→
Yuki Ishikawa
April 21, 2018
Technology
160
0
Share
Introduction to GPU Programming in Python
PyData.Okinawa Meetup #34
https://pydataokinawa.connpass.com/event/82009/
Yuki Ishikawa
April 21, 2018
More Decks by Yuki Ishikawa
See All by Yuki Ishikawa
Snowflake Openflow さわってみた
hoto17296
0
520
第3回 Snowflake 中部ユーザ会- dbt × Snowflake ハンズオン
hoto17296
4
1.3k
ORM と向き合う
hoto17296
14
11k
明日業務で役立たない Web 開発 TIPS
hoto17296
0
230
クソ bot 実装ライブコーディング
hoto17296
0
250
DeepGBM 論文の紹介
hoto17296
0
710
試行錯誤のための Docker 活用術
hoto17296
4
3.2k
Hive 集計テクニック
hoto17296
0
540
データ分析と Docker / Data Analysis with Docker
hoto17296
0
400
Other Decks in Technology
See All in Technology
AIペネトレーションテスト・ セキュリティ検証「AgenticSec」ご紹介資料
laysakura
0
2.2k
2026年に相応しい 最先端プラグインホストの設計<del>と実装</del>
atsushieno
0
120
Introduction to Bill One Development Engineer
sansan33
PRO
0
410
明日からドヤれる!超マニアックなAWSセキュリティTips10連発 / 10 Ultra-Niche AWS Security Tips
yuj1osm
0
460
今年60歳のおっさんCBになる
kentapapa
2
390
NOSTR, réseau social et espace de liberté décentralisé
rlifchitz
0
180
サイバーフィジカル社会とは何か / What Is a Cyber-Physical Society?
ks91
PRO
0
190
AI時代に新卒採用、はじめました/junior-engineer-never-die
dmnlk
0
270
昔はシンプルだった_AmazonS3
kawaji_scratch
0
260
暗黙知について一歩踏み込んで考える - 暗黙知の4タイプと暗黙考・暗黙動へ
masayamoriofficial
0
1.7k
Zero Data Loss Autonomous Recovery Service サービス概要
oracle4engineer
PRO
5
14k
CloudSec JP #005 後締め ~ソフトウェアサプライチェーン攻撃から開発者のシークレットを守る~
lhazy
0
200
Featured
See All Featured
A Soul's Torment
seathinner
6
2.6k
How to Build an AI Search Optimization Roadmap - Criteria and Steps to Take #SEOIRL
aleyda
1
2k
Kristin Tynski - Automating Marketing Tasks With AI
techseoconnect
PRO
0
220
GitHub's CSS Performance
jonrohan
1032
470k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
Have SEOs Ruined the Internet? - User Awareness of SEO in 2025
akashhashmi
0
310
How People are Using Generative and Agentic AI to Supercharge Their Products, Projects, Services and Value Streams Today
helenjbeal
1
160
Ruling the World: When Life Gets Gamed
codingconduct
0
190
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.3k
BBQ
matthewcrist
89
10k
A Modern Web Designer's Workflow
chriscoyier
698
190k
Navigating Team Friction
lara
192
16k
Transcript
Introduction to GPU Programming in Python 2018.04.21 PyData.Okinawa #34 @hoto17296
None
None
@hoto17296 • ͪΎΒσʔλגࣜձࣾ • σʔλੳϚϯ • JavaScript / Python /
AWS • IPv6 ͕Θ͔Βͳ͍ࠓ͜ͷ͝Ζ
Ͳͷ͝Ոఉʹ͋Δ GPU Ϛγϯ
͍͜ͳͯ͠·͔͢ʁʁʁ
ࣗͷ GPU ͷΠϝʔδ Λ͘ಈ͔ͤΔͭʂ
͍ͬͨͳ͍
GPU Λ͍ͨ͢Ίʹ CUDA Λ৮ͬͯΈΑ͏
GPU ͱ • Graphics Processing Unit • άϥϑΟοΫॲཧͷͨΊͷϓϩηοα • ୯७ͳॲཧΛฒྻԋࢉͰ͖Δ
• άϥϑΟοΫҎ֎ʹ׆༻͞ΕΔ (GPGPU) • σΟʔϓϥʔχϯάͱ͔
(NVIDIA) GPU ͷத ετϦʔϛϯά Ϛϧνϓϩηοα ଞʹϝϞϦͱ͔ Ωϟογϡͱ͔ ৭ʑ͋Δ͚Ͳলུ CUDA ίΞ
CUDA ͱ • Compute Unified Device Architecture • NVIDIA GPU
ΛͬͯฒྻܭࢉΛ͢ΔͨΊͷ ϓϩάϥϛϯάϞσϧ
CUDA ༻ޠ • ϗετͱσόΠε • CPU ଆΛϗετ, GPU ଆΛσόΠε ͱݺͿ
• Χʔωϧؔ • GPU ଆͰ࣮ߦ͢ΔॲཧΛهड़ͨؔ͠
CUDA ϓϩάϥϛϯάϞσϧ • ฒྻ͕େʹͳΔ • εϨου, ϒϩοΫ, άϦου ͱ͍͏ ·ͱ·ΓͰฒྻॲཧΛߦ͏
ग़యɿ http://www.nvidia.co.jp/object/cuda_education_jp_old.html
CUDA ϓϩάϥϛϯάϞσϧ • ฒྻ = εϨου x ϒϩοΫ x άϦου
• 1ϒϩοΫ͋ͨΓ࠷େ 512 εϨου • 1άϦου͋ͨΓ࠷େ 65536^2 ϒϩοΫ 1GPU ͋ͨΓ࠷େ 2 ஹ ฒྻʂʂʂ ✌('ω'✌ )ࡾ✌('ω')✌ࡾ( ✌’ω')✌
CUDA ϓϩάϥϛϯά Γͨ͘ͳ͖ͬͯ·ͨ͠Ͷ ʁʁʁ
CUDA ϓϩάϥϛϯάͷํ๏ • C Έ͍ͨͳݴޠͰΧʔωϧؔΛ࣮ͯ͠ nvcc ͱ͍͏ίϯύΠϥͰίϯύΠϧ͢Δ • μϧ͍ •
GPU Λར༻ͨ͠ฒྻॲཧΛॻ͍ͯΈ͍͕ͨ C ͕ॻ͖͍ͨΘ͚Ͱͳ͍ • Python Ͱॻ͔ͤͯ͘Ε
Numba • JIT ίϯύΠϧͯ͠ Python ΛߴԽͰ͖Δͭ • GPU ར༻Ͱ͖Δ (Numba
CUDA)
Χʔωϧؔͷ ఆٛ σίϨʔλΛࢦఆ͢Δ͜ͱͰ Χʔωϧؔͱͯ͠ఆٛ ͜ͷΧʔωϧ͕ؔ ฒྻ࣮ߦ͞ΕΔ ݱࡏͲͷҐஔ (ʁ) ͷεϨουͰ ࣮ߦ͍ͯ͠Δ͔औಘͰ͖Δ
εϨουͷҐஔใ͔Β ॲཧ͢Δରཁૉͷ࠲ඪΛܭࢉ͢Δ ߦྻͷରཁૉʹରͯ͠ॲཧΛߦ͏
Χʔωϧؔͷ࣮ߦ ฒྻ (griddim, blockdim) Λࢦఆ͔ͯ͠Β࣮ߦ͢Δ
͏গ͠ෳࡶͳྫɿ ΈࠐΈԋࢉ ग़యɿ http://tecmemo.wpblog.jp/category/machine-learning/deep-learning/page/2/
ϥϓϥγΞϯϑΟϧλ • ΈࠐΈԋࢉͰը૾ͷΤοδݕग़Λߦ͏ख๏ • ը૾ͷ৭ͷඍʹ૬͢Δ • ͜ͷߦྻΛϑΟϧλͱͯ͠ ΈࠐΈԋࢉΛߦ͏ 1 1
1 1 -8 1 1 1 1
ΈࠐΈԋࢉΛߦ͏ ΧʔωϧؔΛ ࣮
None
ೖྗը૾ ϑΟϧλ݁Ռ
·ͱΊ • Python Ͱ GPU ϓϩάϥϛϯάͰ͖Δ • GPU Ͱ༡Ϳͷͨͷ͍͠ ✌('ω'✌
)ࡾ✌('ω')✌ࡾ( ✌’ω')✌