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
Core ML vs MPS vs BNNS #fincwwdc
Search
shu223
July 24, 2017
Programming
3
5.9k
Core ML vs MPS vs BNNS #fincwwdc
2017.7.24 WWDC2017振り返り勉強会@FiNC
での発表資料です。
shu223
July 24, 2017
Tweet
Share
More Decks by shu223
See All by shu223
slidify-sample
shu223
1
1.5k
機械学習のブルーオーシャン Core ML by 堤 修一 #iOSDC Japan 2020
shu223
12
5.9k
エンジニアのための発信講座 #4 登壇しよう
shu223
0
150
エンジニアのための発信講座 #3
shu223
0
140
エンジニアのための発信講座 #2
shu223
2
260
エンジニアのための発信講座 #1
shu223
1
1k
今こそwatchOS #iosdc
shu223
5
11k
Depth in Depth #iOSDC
shu223
4
30k
Metal Acceleratedかどうかを調べる方法〜あなたのアプリのどこでMetalが暗躍しているのか #potatotips
shu223
1
1.7k
Other Decks in Programming
See All in Programming
S3ストレージクラスの「見える」「ある」「使える」は全部違う ─ 体験から見た、仕様の深淵を覗く
ya_ma23
0
1.2k
AI 開発合宿を通して得た学び
niftycorp
PRO
0
180
一度始めたらやめられない開発効率向上術 / Findy あなたのdotfilesを教えて!
k0kubun
3
2.5k
「効かない!」依存性注入(DI)を活用したAPI Platformのエラーハンドリング奮闘記
mkmk884
0
280
Symfony + NelmioApiDocBundle を使った スキーマ駆動開発 / Schema Driven Development with NelmioApiDocBundle
okashoi
0
250
PHP でエミュレータを自作して Ubuntu を動かそう
m3m0r7
PRO
2
150
How to stabilize UI tests using XCTest
akkeylab
0
150
今年もTECHSCOREブログを書き続けます!
hiraoku101
0
200
存在論的プログラミング: 時間と存在を記述する
koriym
5
580
Kubernetesでセルフホストが簡単なNewSQLを求めて / Seeking a NewSQL Database That's Simple to Self-Host on Kubernetes
nnaka2992
0
190
へんな働き方
yusukebe
6
2.9k
テレメトリーシグナルが導くパフォーマンス最適化 / Performance Optimization Driven by Telemetry Signals
seike460
PRO
2
200
Featured
See All Featured
The Mindset for Success: Future Career Progression
greggifford
PRO
0
290
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
tammyeverts
55
3.3k
Beyond borders and beyond the search box: How to win the global "messy middle" with AI-driven SEO
davidcarrasco
3
97
The Organizational Zoo: Understanding Human Behavior Agility Through Metaphoric Constructive Conversations (based on the works of Arthur Shelley, Ph.D)
kimpetersen
PRO
0
280
Art, The Web, and Tiny UX
lynnandtonic
304
21k
Embracing the Ebb and Flow
colly
88
5k
No one is an island. Learnings from fostering a developers community.
thoeni
21
3.6k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
49
9.9k
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
31
2.7k
Unsuck your backbone
ammeep
672
58k
Noah Learner - AI + Me: how we built a GSC Bulk Export data pipeline
techseoconnect
PRO
0
150
Being A Developer After 40
akosma
91
590k
Transcript
Core ML vs MPSCNN vs BNNS Shuichi Tsutsumi @shu223
!!!Core ML!!!
None
Core MLͷηογϣϯ Ұ൪େ͖͍ձͰ΄΅ຬһ
ਖ਼ࣜϦϦʔεલʹͯ͠طʹଟ͘ͷਓ͕ʹ
ͪΐɺͪΐͬͱͬͯ
ଞͷػցֶशπʔϧͰֶशͨ͠ϞσϧΛ iOSΞϓϦʹ͖࣋ͬͯͯʮਪʯΛߦ͏
ଞͷػցֶशπʔϧͰֶशͨ͠ϞσϧΛ iOSΞϓϦʹ͖࣋ͬͯͯʮਪʯΛߦ͏ → ͜Εࣗମ iOS 10 ͔ΒͰ͖ͨ
Metal Performance Shaders Accelerate (BNNS) Core ML Vision Your App
Metal Performance Shaders Accelerate (BNNS) Core ML Vision Your App
iOS 11
Metal Performance Shaders Accelerate (BNNS) Core ML Vision Your App
iOS 11 iOS 10
Metal Performance Shaders Accelerate (BNNS) Your App iOS 10
Metal Performance Shaders Accelerate (BNNS) Your App iOS 10
Metal Performance Shaders Accelerate (BNNS) Your App iOS 10 ͜ΕͰOK
Metal Performance Shaders Accelerate (BNNS) Your App iOS 10 •
iOS 10Ͱ͑Δ • વMetalͰॲཧ͞ΕΔʢMPSCNNͷ߹ʣ • ֶशଆɺTensorFlowͰKerasͰجຊతʹԿͰOK ͜ΕͰOK
͏ਖ਼ࣜϦϦʔε͔Β1ۙ͘ܦͭͷʹ ΄ͱΜͲʹͳͬͯͳ͍
ʢ୭͔ͷʣʮલ͔Β͋ΔͬͯͷΘ͔ͬͯ ·͢ΑɻͰMPSCNNͬͯػցֶशϑ ϨʔϜϫʔΫ͔Βॻ͖ग़ͨ͠Ϟσϧͷωο τϫʔΫͷߏ·ͰಡΜͰ͘Εͳ͍Ͱ͠ΐʁ ͰɺࣗͰͦ͜Λ࣮͢ΔΜͰ͠ΐʁͦΜ ͳͷΊΜͲͯͬͯ͘͘͞ΒΕͳ͍Ͱ͢Αʯ
͜Εͦͷ௨Γ
ͨͩɺ • ͜Ε΄Ͳͷͷ͕ࠩɺMPSCNNͷ۩ମతͳ͔ܽΒདྷ͍ͯΔͱࢥ͑ͳ͍ • ୯ʹʮMPSCNNͰʰͦΕʱ͕Ͱ͖Δʯͱ͍͏͜ͱ͕ΒΕ͍ͯͳ͔͚ͬͨͩͰʁ • ͔֬ʹɺใݯ͕Appleͷαϯϓϧ͙Β͍͔͠ͳ͔ͬͨ
࣮खॱ͕໌ࣔ͞ΕͯΕͬͱ͏ਓ͕ग़ͯ͘Δ ͷͰͳ͍͔ʁ ʢগͳ͘ͱCore MLʹڵຯ͕͋Δਓͪ͜Βݕ౼ͯ͠Έ͍͍ͯͣʣ ࣮खॱʢΘ͔ͬͯΈΕʣҙ֎ͱγϯϓϧ
MPSCNNΛ༻͍ͨػցֶशػೳͷ ࣮खॱ ̏εςοϓ
Step 1: ϞσϧΛͭ͘Δ
ֶशπʔϧɿͳΜͰ͍͍ Train ʢ˞$//ͷֶश͕Ͱ͖ͯύϥϝʔλΛॻ͖ग़ͤΔͷͰ͋Εʣ
ֶशπʔϧɿͳΜͰ͍͍ ϞσϧͷϑΥʔϚοτɿͳΜͰ͍͍ Train Model (Trained Params) dat or hdf5 •
χϡʔϥϧωοτϫʔΫͷʮॏΈʯ ͱʮόΠΞεʯͱ͍͏ύϥϝʔλ͕ ೖ͍ͬͯΔ • datͨͩͷόΠφϦϑΝΠϧ • iOSΞϓϦͰಡΈࠐΊ͑͢͞Εͳ ΜͰ͍͍ ʢ˞$//ͷֶश͕Ͱ͖ͯύϥϝʔλΛॻ͖ग़ͤΔͷͰ͋Εʣ
Step 2: ωοτϫʔΫΛ࣮͢Δ
None
MPSCNNConvolution MPSCNNFullyConnected MPSCNNPooling MPSCNNConvolution MPSCNNConvolution MPSCNNPooling
MPSCNNConvolution MPSCNNFullyConnected MPSCNNPooling MPSCNNConvolution MPSCNNConvolution MPSCNNPooling CNNͷ֤ʹରԠͨ͠Ϋϥε͕MPSʹ༻ҙ͞Ε͍ͯΔ
MPSCNNConvolution MPSCNNFullyConnected MPSCNNPooling MPSCNNConvolution MPSCNNConvolution MPSCNNPooling CNNͷ֤ʹରԠͨ͠Ϋϥε͕MPSʹ༻ҙ͞Ε͍ͯΔ • Convolution Layer
-> MPSCNNConvolution ͱ໊લͦͷ··ͳͷͰɺ࣮ʹͦΕ ΄Ͳ໎͏͜ͱͳ͍ • ͍͜͠ߦྻܭࢉɺGPUͰͷฒྻԋࢉͷ࠷దԽΛࣗͰҙࣝ͢Δඞཁͳ͍ • Step1Ͱֶͭͬͨ͘शࡁΈύϥϝʔλϑΝΠϧ͔ΒಡΜͰ͖ͯϙΠϯλΛ͢
Step 3: ਪͷ࣮ߦॲཧΛॻ͘
Input image MPSImage Result (UInt, etc.) CNN • Step 2Ͱ࣮ͨ͠ωοτϫʔΫ
• Step 1Ͱֶͭͬͨ͘शࡁΈϞσϧ ʢύϥϝʔλʣΛಡΈࠐΜͰ͍Δ Do something
؆୯ͦ͏Ͱ͠ΐʁ
MPSCNNͰ࣮ͨ͠ Swiftϩΰೝࣝ • MPSCNNͳͷͰiOS 10Ͱಈ͘
BUT
ϩΰೝࣝͷωοτϫʔΫ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ) ஈҧ͍ʹෳࡶ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ
GoogLeNet (Inception v3: Ұൠମೝࣝ༻ͷϞσϧ)) MPSCNNΛ༻͍ͨAppleʹΑΔެ࣮ࣜ 3,000ߦ" ʢωοτϫʔΫ෦͚ͩͰʣ
͜ΕҎ֎ʹ • ػցֶशπʔϧଆͰόΠφϦϑΝΠϧΛॻ͖ग़͢Α͏ʹ͢ΔͷଟগΊΜͲ͍͘͞ • ֶशࡁΈύϥϝʔλͷೖͬͨόΠφϦσʔλΛಡΈࠐΜͰϙΠϯλΛ͢ͱ͔Cͬ Ά͍࣮͕ඞཁ • ೖग़ྗͷࡍʹMPSImageͱ͍͏ΫϥεΛѻ͏ඞཁ͕͋Γɺೖྗը૾Λมͨ͠Γ ਪ݁ՌΛऔΓग़ͨ͠Γ͢ΔࡍʹMetalͷϨΠϠʔͰΞϨίϨ͢Δඞཁ͕͋Δ
͜ΕҎ֎ʹ • ػցֶशπʔϧଆͰόΠφϦϑΝΠϧΛॻ͖ग़͢Α͏ʹ͢ΔͷଟগΊΜͲ͍͘͞ • ֶशࡁΈύϥϝʔλͷೖͬͨόΠφϦσʔλΛಡΈࠐΜͰϙΠϯλΛ͢ͱ͔Cͬ Ά͍࣮͕ඞཁ • ೖग़ྗͷࡍʹMPSImageͱ͍͏ΫϥεΛѻ͏ඞཁ͕͋Γɺೖྗը૾Λมͨ͠Γ ਪ݁ՌΛऔΓग़ͨ͠Γ͢ΔࡍʹMetalͷϨΠϠʔͰΞϨίϨ͢Δඞཁ͕͋Δ →
֓೦γϯϓϧ͕ͩͬͺΓ৭ʑΊΜͲ͍͘͞
Core MLπʔϧͰมͯ͠Xcodeϓϩ δΣΫτʹυϥοάˍυϩοϓ͢Δ͚ͩʂ
VisionΛ͑CIImageܕͰೖྗͰ ͖ɺ݁ՌͷऔΓग़͠؆୯ → MetalϨΠϠʹҰλον͢Δඞཁͳ͠
ͦ͏͍͏Θ͚Ͱ
Metal Performance Shaders Accelerate (BNNS) Your App iOS 10
Metal Performance Shaders Accelerate (BNNS) Core ML Vision Your App
iOS 11 iOS 10
ͬͺΓ͋Γ͕͍ͨ###
·ͱΊ • Core ML͕͜Μͳʹܴ͞ΕͯΔͷʹʮࠓ͙͢ϦϦʔε൛Ͱ͑ ΔʯMPSCNN͕͋·Γʹͳͬͯͳ͍ - ใ͕গͳ͘ɺ͍ํ͕Θͬͯͳ͍͚ͩͰͳ͍͔ • MPSCNNΛར༻࣮ͨ͠खॱ3εςοϓ -
Ϟσϧ࡞ˠωοτϫʔΫ࣮ˠਪ࣮ߦ - େہతʹΈΕҙ֎ͱγϯϓϧʹ͑Δ - ͔͠͠ਖ਼ͳͱ͜Ζେมͳ෦͋Δ • ͦ͜Λܶతʹ؆୯ʹͨ͠ͷ͕Core ML & Vision
͝ਗ਼ௌ͋Γ͕ɾɾɾ
Core ML vs MPSCNN vs BNNS
Εͯͨɾɾɾ Metal Performance Shaders Accelerate (BNNS) Core ML Vision Your
App
CNN on CPU?$
WWDC17ͷMetalϥϘͰฉ͍ͯΈ·ͨ͠ ʮͲ͏ͬͯMPSCNNͱBNNSΛ͍ ͚ͨΒ͍͍ͷʁʯ
ʮجຊతʹMPSCNNͬͯΕ͍͍Αʯ %
ʮ͑ͬɺ͡Ό͋BNNS͍ͭ͏ͷʁʯ
“I don’t know.” %
• ͦͷޙɺʮ͋ɺͰApple WatchGPUͳ͍͔ΒɺwatchOSͱ͔ BNNSͷ͍Ͳ͜Ζ͔Ͷʯͱݴͬͯͨ • ಈ͘ɺͱ͍͕ͬͯ͢͞ʹ⌚ͰϚγϯύϫʔ͕͖͍ͭͷͰͳ͍͔ ͱʢ˞࣮ػͰࢼͨ͜͠ͱͳ͍Ͱ͢ʣ • খ͍͞ωοτϫʔΫͰBNNSͷํ͕͔ͬͨͱ͍͏هࣄݟͨ͜ͱ͕ ͋Δʢ+25%ʣ
• جຊతʹ BNNSͷ͜ͱΕͯΑͦ͞͏ʢݸਓత݁ʣ
͝ਗ਼ௌ͋Γ͕ͱ͏͍͟͝·ͨ͠ʂ