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
Data-centric MLOps(이정권)
Search
MLOpsKR
June 05, 2021
Programming
0
840
Data-centric MLOps(이정권)
MLOps KR(
https://www.facebook.com/groups/mlopskr)에서
주최한 1회 온라인 이벤트 발표 자료입니다
MLOpsKR
June 05, 2021
Tweet
Share
More Decks by MLOpsKR
See All by MLOpsKR
Ray: 대규모 ML인프라를 위한 분산 시스템 프레임워크(조상빈)
mlopskr
0
1.7k
JupyterFlow : 당신의 모델에 날개를 달아드립니다(유홍근)
mlopskr
0
930
모델을 데이터셋에 맞게 대량을 찍어내는 방법(only 파이썬)(김태영)
mlopskr
0
720
KRSH: 선언형 Kubeflow, Terraform처럼 파이프라인 관리하기(김완수)
mlopskr
0
780
MLOps 춘추 전국 시대 정리(변성윤)
mlopskr
0
9.6k
Other Decks in Programming
See All in Programming
オレオレkaggle開発環境に Formatter/Linter入れてみた
stgkrt
0
350
「プログラマーのためのCPU入門」は入り口として丁度よい!
forrep
25
22k
PHP で読む楽しいコアダンプ
sji
0
220
品質が高いコードって何?Rev2.1
ickx
1
390
Go1.22からの疑似乱数生成器について/go-122-pseudo-random-generator
convto
1
120
DDD, necessary but insufficient: physical design principles for microservices
cer
PRO
0
220
孤独のCTOグルメという やや奇抜な企画をやった目的と効果
shoheimitani
3
830
Static Analysis Automation for Hunting Vulnerable Kernel Drivers
takahiro_haruyama
1
1.4k
TokyoR#111_ANOVA
kilometer
1
710
So You Think You Know Git - Part 2
schacon
PRO
0
1.4k
上手な探索的テストとその上達方法について
matsu802
1
440
「Hono遍歴」と「HonoXでブログ作成」
yasu551
0
170
Featured
See All Featured
Into the Great Unknown - MozCon
thekraken
10
810
Building a Modern Day E-commerce SEO Strategy
aleyda
15
6.3k
[RailsConf 2023] Rails as a piece of cake
palkan
21
3.8k
The Invisible Side of Design
smashingmag
293
49k
GraphQLの誤解/rethinking-graphql
sonatard
48
9.1k
KATA
mclloyd
14
11k
How GitHub Uses GitHub to Build GitHub
holman
467
290k
Producing Creativity
orderedlist
PRO
335
39k
RailsConf 2023
tenderlove
0
500
A Modern Web Designer's Workflow
chriscoyier
689
190k
The World Runs on Bad Software
bkeepers
PRO
60
6.6k
No one is an island. Learnings from fostering a developers community.
thoeni
14
2k
Transcript
Data-centric MLOps : 데이터 중심 MLOps를 돕기 위한 작은 장치들
Superb AI 이정권
AI / ML = Model + Data
AI / ML = Model + Data Data centric?
Task Baseline: 70% accuracy Target Performance: 90% accuracy Should the
team improve the code or the data? : code(20%), data(80%) A Chat with Andrew on MLOps: From Model-centric to Data-centric AI
A Chat with Andrew on MLOps: From Model-centric to Data-centric
AI Improve AI → Improve the quality of the data: consistency error rate diversity coverage feedback frequency size ...
A Chat with Andrew on MLOps: From Model-centric to Data-centric
AI slide credit: A Chat with Andrew on MLOps: From Model-centric to Data-centric AI (https://www.youtube.com/watch?v=06-AZXmwHjo)
사실은, 늘 해오던 일 Project progress month 1 month 2
month 3 month 4 month 5 Code a model Build data Launch training job
사실은, 늘 해오던 일 Building the Software 2.0 Stack (Andrej
Karpathy, 2018)
Question: How many labeled images are needed to solve this
problem?
Answer: 100,000 images?
My Answer: I don’t know. Let’s start from 5,000 WHY?
여전히, 잘 모른다 → Data-centric MLOps Systematic & iterative way
to build Data for ML 단순히 지루한 작업을 자동화하는 과정이 아닌 ML 문제를 해결하기 위한 과정 저는 Superb AI라는 팀에서 이 문제를 풀고 있습니다.
<2달 <30명 <20,000 Images The Problem
The Meta Problem Design Data Spec Build Data Train a
model Deploy to service
Starting Point Labeling Tool Data Label
Reusable Data Spec { project_name: potato_detect_1 data_spec: good_potato: box: color:
red condition: ... bad_potato: box: } { project_name: potato_detect_2 data_spec: good_potato: polygon: color: red condition: ... bad_potato: box: }
Reusable Data Spec { project_name: potato_detect_13 data_spec: best_potato: polygon: direction:
options: ... good_potato: {} normal_potato: {} bad_potato: {} } Goal ≠ Task ALWAYS configured repeatedly name, color, type, conditions, options, property, ROI Info, ...
Support flexible pipeline 100 different problems, 100 different datasets, 100
different ways To support flexible pipeline Build Data Team Model WORKING SUBMITTED REVIEWED
Support flexible pipeline
Versioning Set 단위, 실험 당
ML Engineer를 위해 … ? Detailed Statistics & Report
Human in the loop ^ 2 Human in the loop
ML
Inside Human Labeling Data Human Labeling Service Model Data Labeling
Our Model ? Uncertain? Label-wise Confidence Overall Set Confidence User performance estimate Boost Labeling ... Human in the loop ^ 2
Keep labels consistent
Keep labels consistent
요약
Source data analysis, User analysis, Log, Task matching, etc 여전히
할일이 정말 많다. 마무리 SDK를 이용한 사용 예제!는 다음에 https://github.com/superb-AI-Suite/ Full-pipeline MLOps https://ai-infrastructure.org/