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

Chronicles #14

Open
7 of 9 tasks
eubinecto opened this issue Apr 15, 2022 · 3 comments
Open
7 of 9 tasks

Chronicles #14

eubinecto opened this issue Apr 15, 2022 · 3 comments

Comments

@eubinecto
Copy link
Owner

eubinecto commented Apr 15, 2022

v1.0

현재까지 구현한 것까지가 v1.0

v2.0

v3.0: Towards the-qualitative-transformer

점수에만 집착하지말고, 학습이 진행됨에 따라 모델의 내부가 어떻게 변화하는지 관측을 해보자.

향후 계획

  • 논문의 구현체와 다른 점이 무엇인지 표기하기 (tokenizer 라던지)
@eubinecto
Copy link
Owner Author

@eubinecto
Copy link
Owner Author

다음 방향성

각 구조의 당위성을 파악하는 실험을 하는 것이 주 목표. 물론 발전된 모델이 존재하기는 하나, 항상 중요한 것은 기본이다.
우선 SingleHead vs. MultiHead를 비교를 해보는 것으로 시작해보자. 이미 누군가 실험을 해보았을수도 ?

  1. 정량적인 분석
  2. 정성적인 분석

둘다 필요하다.

이와 함께, 왜 반드시 "scaled"가 필요한지, 그 당위성을 분석하는 것도 중요할 것.

어차피 ablation은 실험만을 위한 것이므로... main branch에는 지금 구현 것을 그대로 유지해두면 된다.
ablation실험은 git tag로 관리하자.

@eubinecto eubinecto changed the title Progress Chronicles May 24, 2022
@eubinecto
Copy link
Owner Author

더 연습해볼만한 것들

  • fastapi 구축해서 Ainize에서 호스팅하기 (도커활용)
  • streamlit으로 간단한 데모 만들어보기
  • wandb.restore도 구현할 수 있다면 좋을 것!

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