-
Notifications
You must be signed in to change notification settings - Fork 284
intermediate/forward_ad_usage.py 번역 #1031
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
base: master
Are you sure you want to change the base?
intermediate/forward_ad_usage.py 번역 #1031
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
번역하시느라 고생 많으셨습니다. 코멘트 드린 내용 검토 부탁드려요.
| This tutorial demonstrates how to use forward-mode AD to compute | ||
| directional derivatives (or equivalently, Jacobian-vector products). | ||
| 이 튜토리얼은 순전파 모드 자동 미분(Forward-mode AD)을 사용하여 방향성 도함수(directional derivative) 또는 야코비안-벡터 곱(Jacobian-vector product)을 계산하는 방법을 보여줍니다. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
자동 미분에 대한 약어 표기가 이 문장에서 처음 등장하는데, 전체 명칭을 작성해주면 이해가 더 쉬울 것 같습니다.
순전파 모드 자동 미분(Forward-mode AD) -> 순전파 모드 자동 미분(Forward-mode Automatic Differentiation)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
반영하겠습니다!
| Also note that forward-mode AD is currently in beta. The API is | ||
| subject to change and operator coverage is still incomplete. | ||
| 또한, 순전파 모드 AD는 현재 베타 버전입니다. 따라서 API가 변경될 수 있으며, 아직 일부 연산자는 지원되지 않을 수 있습니다. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
글 전반에서 AD라고 작성하지 않고, 자동 미분이라고 작성해도 괜찮을 것 같습니다.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
전반에 설명이 있으니 자동 미분으로 바꾸겠습니다
|
|
||
| ###################################################################### | ||
| # Custom autograd Function | ||
| # 사용자 정의 autograd Function |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Function은 함수라고 번역해도 좋을 것 같습니다.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
반영하겠습니다!
| # ``functorch.jvp`` requires every primal to be associated with a tangent. | ||
| # If we only want to associate certain inputs to `fn` with tangents, | ||
| # then we'll need to create a new function that captures inputs without tangents: | ||
| # ``functorch.jvp`` 는 모든 원시가 탄젠트와 연결될 것을 요구합니다. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
본문에서 "primal" 이라는 변수명을 지속적으로 사용하고 있기 때문에 "원시"라는 표현 대신, "primal"을 그대로 사용하는 것이 더 좋을 것 같습니다.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
반영하겠습니다!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
용어집에서 확인하실 수 있듯이 tensor는 별도 번역 없이 영문 표현 그대로 사용하고 있습니다.
"텐서"라고 작성해주신 부분에 대해 수정 요청드립니다.
|
반영했습니다! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
전반적으로 잘되어있으나 사소한 수정이 필요할거 같습니다
| # Another way to use ``nn.Module`` with forward AD is to utilize | ||
| # the functional Module API (also known as the stateless Module API). | ||
| # ``nn.Module`` 을 순전파 자동 미분과 함께 사용하는 또 다른 방법은 | ||
| # 함수형 모듈 API를 활용하는 것입니다. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
(also known as the stateless Module API) 에 대한 부분이 빠져있습니다.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
반영하겠습니다!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good 👍
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
라이선스 동의
변경해주시는 내용에 BSD 3항 라이선스가 적용됨을 동의해주셔야 합니다. 더 자세한 내용은 기여하기 문서를 참고해주세요. 동의하시면 아래
[ ]를[x]로 만들어주세요.관련 이슈 번호
이 Pull Request와 관련있는 이슈 번호를 적어주세요. 이슈 또는 PR 번호 앞에 #을 붙이시면 제목을 바로 확인하실 수 있습니다. (예. #999 )
PR 종류
이 PR에 해당되는 종류 앞의
[ ]을[x]로 변경해주세요.PR 설명
이 PR로 무엇이 달라지는지 대략적으로 알려주세요.
intermediate/forward_ad_usage.py 통번역 했습니다