Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
異常検知ライブラリを作った話
Search
tsurubee
November 18, 2017
Technology
1.1k
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
異常検知ライブラリを作った話
ライブラリの特徴:特異スペクトル変換による変化点検知
tsurubee
November 18, 2017
More Decks by tsurubee
See All by tsurubee
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
670
言語モデルによるAI創薬の進展 / Advancements in AI-Driven Drug Discovery Using Language Models
tsurubee
2
670
AIトップカンファレンスからみるData-Centric AIの研究動向 / Research Trends in Data-Centric AI: Insights from Top AI Conferences
tsurubee
3
3.5k
DeepCrysTet: A Deep Learning Approach Using Tetrahedral Mesh for Predicting Properties of Crystalline Materials
tsurubee
0
1.4k
3次元メッシュで表現した結晶構造を用いた材料物性の予測に向けた深層学習モデルの設計 / Design of Deep Learning Model for Predicting Material Properties Using Crystal Structure Represented by Three-Dimensional Mesh
tsurubee
1
2.6k
分散システムの性能異常に対する機械学習の解釈性に基づく原因診断手法 / A Method for Diagnosing the Causes of Performance Issues in Distributed Systems Based on the Interpretability of Machine Learning
tsurubee
0
1.8k
機械学習の解釈性に関する研究動向とシステム運用への応用 / A Survey on Interpretable Machine Learning and Its Application for System Operation
tsurubee
0
400
機械学習モデルの局所的な解釈に着目したシステムにおける異常の原因診断手法の構想
tsurubee
0
8.2k
アニーリングマシンを活用したエッジAIにおける 生成モデルの学習効率化のためのアーキテクチャ
tsurubee
0
1.7k
Other Decks in Technology
See All in Technology
不可逆な意思決定を避けながら10億超の顧客データを AI Ready にするデータ基盤設計
baseballyama
1
320
触りたくないデータに、どう触るか AI時代のRBACアクセスガバナンス
ivry_presentationmaterials
0
220
LambdaWebFunctionsを試してみた / exploring-lambda-web-functions
koukihosaka
0
290
Local CI × GitHub Actions
gogutan
2
140
初海外で注意すべきこと12選!!!安全快適なre:invent生活にするために
yuuuuuuu168
1
110
Kiro Crewをはじめよう
kizawa2020
1
150
大型犬好きによるPugの紹介
yamanoku
1
150
パッチ更新でterraform initが落ちた日
grimoh
1
140
DGX Spark でどこまでハッキングができるのか?
kash1064
3
1.3k
TinyGoでオブザーバビリティ?OpenTelemetryはTinyGoで使えるのか / Is OpenTelemetry available in TinyGo environment?
ymotongpoo
2
140
10年後、Engineering 行為を剥奪されるとして、わたしたちは、どう、何を創ればいいだろうか。 / If Engineering Is Taken Away in 10 Years, How and What Should We Create?
bitkey
PRO
1
180
How to Design Configuration for Rails Applications
willnet
1
680
Featured
See All Featured
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.6k
The Illustrated Children's Guide to Kubernetes
chrisshort
51
53k
Mind Mapping
helmedeiros
1
400
What's in a price? How to price your products and services
michaelherold
247
13k
[Rails World 2026] Durable orchestration on Rails: from continuation to workflow
palkan
1
740
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
52
6.1k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
3k
How to Grow Your eCommerce with AI & Automation
katarinadahlin
PRO
2
330
Accessibility Awareness
sabderemane
1
250
From π to Pie charts
rasagy
1
400
Introduction to Domain-Driven Design and Collaborative software design
baasie
1
1k
A better future with KSS
kneath
240
18k
Transcript
ҟৗݕϥΠϒϥϦ Λ࡞ͬͨ भ %BUB4DJFOUJTU.FFU6Q
ࣗݾհ ా തจ ʢ!UTVSVCFFʣ ΤϯδχΞྺɿ ͪΐ͍ େֶͷઐ߈Խֶ લ৬ফ࢜ झຯےτϨ ʢಛʹϕϯνϓϨεʣ
ࠓճͷ༰ɾɾ લʹθϩ͔ΒϓϩάϥϛϯάΛ࢝Ίͨࢲ͕ɺ ϊϦͱؾ߹͍ͰҟৗݕϥΠϒϥϦΛ࡞ͬͨ ɾҟৗݕʹ͍ͭͯ ɾϥΠϒϥϦΛ࡞ͬͨഎܠ ɾ࣮ΞϧΰϦζϜ ɾϦΞϧλΠϜҟৗࢹͷσϞ ͳͲͳͲɾɾ
ҟৗݕͱʁ
େଟͷσʔλͱৼΔ͍͕ ҟͳΔσʔλΛݕग़͢Δٕज़ σʔλϚΠχϯά نଇੑ ҟৗ σʔλͷࢁ
ҟৗݕͷԠ༻ྫ ίϯϐϡʔλΠϧε%PT߈ܸͷૣظൃݟ ηΩϡϦςΟ ނো༧ஹݕ ػց ྲྀߦͷݕɾ৽τϐοΫͷൃݟɾ ϢʔβߦಈͷมԽݕ ϚʔέςΟϯά
ҟৗσʔλྫ ʙͦͷ̍
ҟৗσʔλྫ ʙͦͷ̍ ҟৗʂ
ҟৗʂ if value > 120: print('ERROR!') JGจͰݕͰ͖ͦ͏ʂ ҟৗσʔλྫ ʙͦͷ̍
ҟৗσʔλྫ ʙͦͷ̎
Կ͔ҟৗ͕ى͖ͯΔ ҟৗʹ͍Ζ͍Ζ͋Δ ҟৗσʔλྫ ʙͦͷ̎
ҟৗݕͷྨ ֎Εݕ มԽݕ ҟৗݕ ٸܹͳৼΔ͍ͷมԽΛݕ ࣌ܥྻϞσϧ ଞͱେ͖͘ҟͳΔΛݕ ಠཱϞσϧ
ҟৗݕͷྨ ֎Εݕ มԽݕ ҟৗݕ ࠓճίονͷʂ ٸܹͳৼΔ͍ͷมԽΛݕ ࣌ܥྻϞσϧ ଞͱେ͖͘ҟͳΔΛݕ ಠཱϞσϧ
ҟৗݕϥΠϒϥϦ ʹ͍ͭͯ
ͳͥ࡞͔ͬͨ ͍ͭͷ͔ɺٕज़Λ͍͜ͳ͢ଆͰͳ͘ ࡞ΔଆʹͳΓ͍ͨ ϚΠϯυతഎܠ σʔλαΠΤϯεతഎܠ ϏδωεͷݱʹᷓΕΔ࣌ܥྻσʔλ ʢྫ͑ɺച্σʔλɾΞΫηεϩάͳͲʣ ͷมԽͷஹީΛ͍ͪૣ͘ݕ͍ͨ͠
࡞ͬͨͷ ಛʢڧΈʣԿʁ #BOQFJʢ൪ฌʣ ɿ1ZUIPOҟৗݕύοέʔδ "ಛҟεϖΫτϧมʹΑΔ࣌ܥྻσʔλͷมԽݕ ಛఆͷ֬ΛԾఆ͍ͯ͠ͳ͍ͨΊɺ ᶃଟ༷ͳมԽʹؤڧʹରԠͰ͖ɺ ᶄύϥϝʔλνϡʔχϯά͕༰қͳ มԽݕ͕Մೳ
ಛҟεϖΫτϧม ࣌ࠁ ཤྺߦྻ" ςετߦྻ# ओ෦ۭؒ ",''', ( ",''', ( มԽ
શମΠϝʔδ ʮೖػցֶशʹΑΔҟৗݕ ʕ3ʹΑΔ࣮ફΨΠυʯͷਤ Λࢀߟʹ࡞ աڈଆͱݱࡏଆͷߦྻಉ࢜ͷ৯͍ҧ͍ͷେ͖͞ΛఆྔԽ
มԽͷఆٛ ಛҟղʹΑΔಛύλʔϯͷநग़ 9 6 7 ಛҟ্ҐͷࠨಛҟϕΫτϧΛ NຊऔΓग़͢ աڈଆ ݱࡏଆ
( ͷ࠷େಛҟ) 4 ֊SΛͭNºOߦྻ9 N O S S O S มԽ
#BOQFJͷ͍ํ ҟৗݕ ϋΠύʔύϥϝʔλʢXʣɿεϥΠυ૭ͷαΠζ ΠϯϓοτʢEBUBʣϦετܕ/VN1Z BSSBZͳͲͷσʔλྻ ΞτϓοτʢSFTVMUTʣΠϯϓοτͱಉ͡αΠζͷ/VN1Z BSSBZ git clone https://github.com/tsurubee/banpei.git
cd banpei pip install . Πϯετʔϧ import banpei model = banpei.SST(w=50) results = model.detect(data) ˞ͨͬͨ̏ߦʂ
#BOQFJʹΑΔपҟৗݕ
#PLFIͱͷ࿈ܞʹΑΔ ϦΞϧλΠϜҟৗࢹ ͷઓ
#PLFIͱʁ IUUQTCPLFIQZEBUBPSHFOMBUFTU ɾ*OUFSBDUJWFWJTVBMJ[FUJPO ɾ/PWFMHSBQIJDT ɾ4USFBNJOH EZOBNJD MBSHFEBUB ɾ/POFFEUPXSJUF+BWBTDSJQU ରܕՄࢹԽڥΛఏڙ͢Δ1ZUIPOϥΠϒϥϦ IUUQTXXXTMJEFTIBSFOFUDPOUJOVVNJPIBTTMFGSFFEBUBTDJFODFBQQTXJUICPLFIXFCJOBS
σ Ϟ
• ʮಛҟεϖΫτϧมʹΑΔपҟৗݕʯ IUUQTZPVUVCFF'7/,"7U/1 • ʮಛҟεϖΫτϧมʹΑΔ͖ͷมԽݕʯ IUUQTZPVUVCF@XPVC-"I9L :PV5VCFʹσϞಈըΛެ։͍ͯ͠·͢ʂ
(JU)VCʹެ։͍ͯ͠·͢ʂ IUUQTHJUIVCDPNUTVSVCFFCBOQFJ
ϒϩάॻ͍͍ͯ·͢ʂ IUUQTHJUIVCDPNUTVSVCFFCBOQFJ
ࠓޙͷల l ಛҟεϖΫτϧมͷߴԽʹΑΔߋͳΔ ϦΞϧλΠϜੑͷٻ l ҟৗ௨ػೳͷ࣮ʹΑΔ࣮༻ੑͷ্ ͋ͱɺ৽ͨͳΞϧΰϦζϜ࣮͍ͨ͠ɾɾ
1ZUIPOίϛϡχςΟ ʹ͍ͭͯ
1Z'VLVPLBͬͯ·͢ʂ ݄ʹ-5ΠϕϯτΛ։࠵༧ఆͰ͢ʂ
1Z$PO ,ZVTIVΓ·͢ʂ ελοϑืूதʂ ڵຯ͕͋Δํ࠙ձͰ͓͔͚͍ͩ͘͞ʂ IUUQLZVTIVQZDPOKQ
͝ਗ਼ௌ͋Γ͕ͱ͏ ͍͟͝·ͨ͠ʂ