Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ML monitoring - Intro #120

Open
yeomko22 opened this issue Sep 20, 2021 · 1 comment
Open

ML monitoring - Intro #120

yeomko22 opened this issue Sep 20, 2021 · 1 comment

Comments

@yeomko22
Copy link
Owner

yeomko22 commented Sep 20, 2021

Deploying a model to production

스크린샷 2021-09-20 오후 6 30 24

  • 모델을 배포한다는 것은 live 환경에 모델을 올려놓고 live data를 받는다는 얘기다.
  • 더 이상 research environment에서 historical data로 모델을 학습시키지 않는다.
  • research environment와 production environment를 구분해야한다.

Scenarios we often encounter

  • deployment of first model
  • replacement of an existing model
  • editing deployed model (minor tweak)
    -> 두번째 케이스에 집중하는 코스가 될 것
@yeomko22
Copy link
Owner Author

yeomko22 commented Sep 21, 2021

ML system life cycle

스크린샷 2021-09-21 오후 7 47 56

  • unit test, integration test, system monitoring은 전통적으로 수행하는 테스트들이다.
  • ML Infrastructure test, model test, data test는 ML specific한 테스트 들이다.
  • skew test, data moniroting, prediction monitoring은 shadow deployment를 수행한 이후에 해야하는 테스트들이다.

CD4ML

스크린샷 2021-09-21 오후 7 50 15

  • model building, model evaluation & experiment는 research phase이다. jupyter notebook 환경에서 개발된다.
  • 그 이후 제품화를 하고, 이를 테스트하고 배포하고 모니터링하는 단계가 있다.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant