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
野球エンジニアの72万球 #BPStudy
Search
Shinichi Nakagawa
PRO
March 29, 2018
Research
0
2.6k
野球エンジニアの72万球 #BPStudy
Baseballsavantを例とした可視化と簡単な分析事例です
Shinichi Nakagawa
PRO
March 29, 2018
Tweet
Share
More Decks by Shinichi Nakagawa
See All by Shinichi Nakagawa
WBCの解説は生成AIにやらせよう - 生成AIで野球解説者AI Agentを実現する / Baseball Commentator AI Agent for Gemini
shinyorke
PRO
1
360
自らを強いエンジニアにするための3つの習慣 2025/ Fitter happier more productive
shinyorke
PRO
0
280
生成AI時代におけるSREの進化とキャリア戦略 / Building an Embedded SRE team and my career
shinyorke
PRO
0
150
生成AIを活用した野球データ分析 - メジャーリーグ編 / Baseball Analytics for Gen AI
shinyorke
PRO
1
6k
ゼロから始めるSREの事業貢献 - 生成AI時代のSRE成長戦略と実践 / Starting SRE from Day One
shinyorke
PRO
3
7.3k
AI・LLM事業部のSREとタスクの自動運転
shinyorke
PRO
0
520
実践Dash - 手を抜きながら本気で作るデータApplicationの基本と応用 / Dash for Python and Baseball
shinyorke
PRO
2
4.2k
Terraform, GitHub Actions, Cloud Buildでデータ基盤をProvisioningする / Data Platform provisioning for Google Cloud and Terraform
shinyorke
PRO
2
3.6k
Cloud RunとCloud PubSubでサーバレスなデータ基盤2024 with Terraform / Cloud Run and PubSub with Terraform
shinyorke
PRO
9
4.4k
Other Decks in Research
See All in Research
視覚から身体性を持つAIへ: 巧緻な動作の3次元理解
tkhkaeio
1
210
YOLO26_ Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection
satai
3
100
大規模言語モデルにおけるData-Centric AIと合成データの活用 / Data-Centric AI and Synthetic Data in Large Language Models
tsurubee
1
520
第二言語習得研究における 明示的・暗示的知識の再検討:この分類は何に役に立つか,何に役に立たないか
tam07pb915
0
1.7k
Upgrading Multi-Agent Pathfinding for the Real World
kei18
0
430
業界横断 副業コンプライアンス調査 三者(副業者・本業先・発注者)におけるトラブル認知ギャップの構造分析
fkske
0
120
ForestCast: Forecasting Deforestation Risk at Scale with Deep Learning
satai
3
520
生成的情報検索時代におけるAI利用と認知バイアス
trycycle
PRO
0
380
LLM-jp-3 and beyond: Training Large Language Models
odashi
1
780
SkySense V2: A Unified Foundation Model for Multi-modal Remote Sensing
satai
3
620
LLMアプリケーションの透明性について
fufufukakaka
0
190
[IBIS 2025] 深層基盤モデルのための強化学習驚きから理論にもとづく納得へ
akifumi_wachi
20
9.8k
Featured
See All Featured
HU Berlin: Industrial-Strength Natural Language Processing with spaCy and Prodigy
inesmontani
PRO
0
260
From π to Pie charts
rasagy
0
150
How to Grow Your eCommerce with AI & Automation
katarinadahlin
PRO
1
140
Building Experiences: Design Systems, User Experience, and Full Site Editing
marktimemedia
0
440
SEO Brein meetup: CTRL+C is not how to scale international SEO
lindahogenes
1
2.4k
What's in a price? How to price your products and services
michaelherold
247
13k
職位にかかわらず全員がリーダーシップを発揮するチーム作り / Building a team where everyone can demonstrate leadership regardless of position
madoxten
61
52k
From Legacy to Launchpad: Building Startup-Ready Communities
dugsong
0
170
Writing Fast Ruby
sferik
630
63k
The B2B funnel & how to create a winning content strategy
katarinadahlin
PRO
1
300
Building AI with AI
inesmontani
PRO
1
780
Visualization
eitanlees
150
17k
Transcript
ٿΤϯδχΞͷ72ສٿ τϥοΩϯάɾσʔλ͔ΒޠΔٿϊϯϑΟΫγϣϯ Shinichi Nakagawa@shinyorke
ٿΤϯδχΞis୭?
ʲʳϫΠͰ͢ • Shinichi Nakagawa(த৳Ұ) • ωΫετϕʔε CTO/ٿΤϯδχΞ • #ηΠόʔϝτϦΫε #Python
#σʔλੳ • Baseball Play Study ։͔࢝࣌Βৗ࿈(2014ʙ) • Baseball Play Study͔Βϗϯτʹٿքʹདྷ·ͨ͠
ʁʁʁʮ72ສٿ͛ͨΒݞග͕(ryʯ ※͛ͯͳ͍Ͱ͢w
72ສٿ=MLBͷ1γʔζϯٿ 2017ͷ࣮,ϨΪϡϥʔγʔζϯͷΈ. ϓϨʔΦϑΛؚΊΔͱ73ສٿͪΐͬͱʹͳΔ.
Ͳ͜ʹσʔλ͋Δͷ? • MLBެࣜʮBaseballsavantʯͱ͍͏αΠτͰ ୭ͰೖखͰ͖Δ • https://baseballsavant.mlb.com/ statcast_search • τϥοΫϚϯɾStatcastͰهͨ͠ τϥοΩϯάɾσʔλ͕ݩʹͳ͍ͬͯΔ
τϥοΫϚϯ=ٿɾଧٿͷܭଌػث ͘Θ͘͠ʮBaseball GeeksʯͷղઆΛͲ͏ͧʂ https://www.baseballgeeks.jp/?p=3551
ࠓͷςʔϚʮଧٿʯ • 72ສٿ͔Βબग़ͨ͠ʮҹతͳଧٿʯΛհ • ຊͱ͍,ϝδϟʔͷϨδΣϯυ͞Μ • ࠓ͔ΒೋྲྀͰߦ͘ਓ…ͷಉ྅ • งғؾΛ௫ΜͰ͘ΕΔͱ͋Γ͕͍ͨͰ͢
128,945 / 718,917(ٿ) ※શσʔλͷ18%Λ༻(͓͓Αͦ100MB͘Β͍)
ʲਤʳશଧٿσʔλͷ݁Ռ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯
ʲਤʳશଧٿσʔλͷ݁Ռ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ଧκʔϯ
ʲਤʳશଧٿσʔλͷ݁Ռ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ଧκʔϯ ୯ଧκʔϯ
ʲਤʳશଧٿσʔλͷ݁Ռ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ଧκʔϯ ୯ଧκʔϯ खͷ͓ࣄκʔϯ
ʲਤʳશଧٿσʔλͷ݁Ռ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ଧκʔϯ ୯ଧκʔϯ ্͕Γ͗͌͢ खͷ͓ࣄκʔϯ
֮͑ͯ΄͍͜͠ͱ • ͍͍ײ͡ͷʮଧٿʯʮඈᠳ֯ʯͰඈͿଧٿϗʔϜϥϯɾଧʹͳΔՄೳੑ͕ߴ͍ • ҆ • 187km/h / 8~50 •
161km/h / 24~33 • 158km/h / 26~30 • ͜ΕΛʮόϨϧκʔϯʯͱ͍͍·͢ • ʁʁʁʮڈϑϥΠϘʔϧɾϨϘϦϡʔγϣϯ͕͋ͬͨ͡Όͳ͍ɺͦΕ(ryʯ ˠਖ਼ղʂͦ͏͍͏͜ͱͰ͢ • ʲࢀߟจݙʳ https://www.baseballgeeks.jp/?p=1342 ※Baseball GeeksΑΓҾ༻
ೋਓͷଧऀʹ͍ͭͯ • ຊͱ͍,ϝδϟʔͷϨδΣϯυ͞Μ • ࠓ͔ΒೋྲྀͰߦ͘ਓ…ͷಉ྅ • ͜ͷೋਓͷଧٿΛݟͯΈΑ͏
ҰਓʮIchiro Suzukiʯ ϚϦφʔζ෮ؼ͓ΊͰͱ͏͍͟͝·͢ʂ ը૾ɿ https://commons.wikimedia.org/wiki/File:Ichiro_Suzuki_2010.jpg
ʲਤʳIchiro Suzukiબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯
ʲਤʳIchiro Suzukiબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ͜ͷล͕ ଧκʔϯ
ʲਤʳIchiro Suzukiબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ͜ͷล͕ ଧκʔϯ όϨϧ ൃݟʂ
ΠνϩʔબखͷόϨϧ • 2017/8/22 ϑΟϦʔζઓ(ఢͰͷࢼ߹) • ୈ3߸ιϩ,ઌൃͷϊϥ͔ΒҰൃ • 160.48 km/h, 28
• શ3ຊͷΞʔνத,όϨϧೖΓ͜ͷ1ຊͷΈ …Ͱ͚͢Ͳ,͜Ε͕40ͱ͔ා͍(ଚܟͷ؟ࠩ͠)
ೋਓʮMike Troutʯ େ୩ᠳฏ(ΤϯδΣϧε)ͷಉ྅͔ͭεʔύʔελʔ ը૾ɿ https://commons.wikimedia.org/wiki/File:Los_Angeles_Angels_center_fielder_Mike_Trout_(27)_(5972457428).jpg
Mike Trout #ͱ ※೦ͷҝ • ϝδϟʔΛද͢ΔελʔͷҰਓ • ϩαϯθϧεɾΤϯθϧεͷ֎ख(ηϯλʔ) • ӈ͛ӈଧͪ,26ࡀ,ϝδϟʔ8
• ߈कࡾഥࢠ͕ʮຊʹʯἧ໊ͬͨબख • ௨ࢉOPS .976ɹ˞Ϊʔλ(ιϑτόϯΫ).946
ʲਤʳMike Troutબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯
ʲਤʳMike Troutબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ͜ͷล͕ଧκʔϯ ˠϗʔϜϥϯଟ͗͌͢
ʲਤʳMike Troutબखͷଧٿ(2017) X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯ ͜ͷล͕ଧκʔϯ ˠϗʔϜϥϯଟ͗͌͢ όϨϧ͚ͩͲ Ξτͩͱʁ
Ξτʹͳͬͨଧٿͷৄࡉ X(ԣ)ɿଧٿ, Y(ॎ)ɿଧٿ֯
͜Ε͍͢͝ϓϨʔͳͷͰʁ • ͱࢥ͍,ࢼ߹݁ՌΛνΣοΫ • هɿηϯλʔϑϥΠ • ϑΝΠϯϓϨʔͱ͍͏هͳ͘ • ී௨ͷଧٿͱͯ͠ͱΒΕ͍ͯͨ •
ϝονϟྑ͍͋ͨΓͷਅਖ਼໘ͩͬͨʁʁʁ ;ʔΜ(ಡΈ)
·ͱΊ • ϝδϟʔϦʔάଧٿɾٿͷσʔλ͕ϑΝϯͰ͑Δ • ଧٿͱ֯ʹண͢Δ͚ͩͰ৭ʑͳࢹ͕Ͱ͖Δ • Πνϩʔબख·ͩ·͔ͩͬͱͤΔ (ελΠϧม͑ͯ͘Εͳ͍͔ͳ͋ʁ) • େ୩ᠳฏ͕͛Δͱ͖τϥτʹͯ͠Ͷ
• ࢸͬͯී௨ͷϑϥΠ࣮ී௨͡Όͳ͍Մೳੑ͕
τϥοΩϯάɾσʔλ ָ͘͠ͳ͖͔ͬͯͨͳʁ
Baseball GeeksͰͬͱָ͘͠! • τϥοΩϯάɾσʔλΛ׆༻ͨ͠ٿͷ৽͍͠ݟํɾࢹΛհͯ͠·͢ • σʔλɾεϙʔπՊֶͰ໌Β͔ʹͳͬͨ͜ͱΛʮΘ͔Γ͘͢ʯ͑Δ • ΈΜͳಡΜͰͶ&ϒΫϚΑΖ͘͠ʂ https://www.baseballgeeks.jp/
ϓϨΠϘʔϧʂ ࠓٿͰྑ͍ҰΛʂ͝ਗ਼ௌ͋Γ͕ͱ͏͍͟͝·ͨ͠⽁ Shinichi Nakagawa(Twitter/Facebook/etc… @shinyorke)