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
5.8k
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
1.4k
機械学習のブルーオーシャン Core ML by 堤 修一 #iOSDC Japan 2020
shu223
12
5.7k
エンジニアのための発信講座 #4 登壇しよう
shu223
0
130
エンジニアのための発信講座 #3
shu223
0
120
エンジニアのための発信講座 #2
shu223
2
250
エンジニアのための発信講座 #1
shu223
1
980
今こそwatchOS #iosdc
shu223
5
10k
Depth in Depth #iOSDC
shu223
4
29k
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
1.6k
Other Decks in Programming
See All in Programming
ktr0731/go-mcpでMCPサーバー作ってみた
takak2166
0
160
Parallel::Pipesの紹介
skaji
2
910
GoのWebAssembly活用パターン紹介
syumai
3
10k
Team topologies and the microservice architecture: a synergistic relationship
cer
PRO
0
110
Practical Tips and Tricks for Working with Compose Multiplatform Previews (mDevCamp 2025)
stewemetal
0
120
事業戦略を理解してソフトウェアを設計する
masuda220
PRO
22
6k
コードに語らせよう――自己ドキュメント化が内包する楽しさについて / Let the Code Speak
nrslib
6
1.4k
Haskell でアルゴリズムを抽象化する / 関数型言語で競技プログラミング
naoya
17
4.2k
関数型まつり2025登壇資料「関数プログラミングと再帰」
taisontsukada
2
800
Elixir で IoT 開発、 Nerves なら簡単にできる!?
pojiro
1
120
TypeScript LSP の今までとこれから
quramy
1
500
從零到一:搭建你的第一個 Observability 平台
blueswen
1
890
Featured
See All Featured
Practical Tips for Bootstrapping Information Extraction Pipelines
honnibal
PRO
20
1.3k
Being A Developer After 40
akosma
90
590k
Learning to Love Humans: Emotional Interface Design
aarron
273
40k
Fight the Zombie Pattern Library - RWD Summit 2016
marcelosomers
233
17k
VelocityConf: Rendering Performance Case Studies
addyosmani
329
24k
How to Think Like a Performance Engineer
csswizardry
24
1.7k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
10
900
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
48
5.4k
Visualization
eitanlees
146
16k
Done Done
chrislema
184
16k
Statistics for Hackers
jakevdp
799
220k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
50k
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