최종 계층은 선형이므로, 입력에 대한 최종 계층의 그레이디언트는 해당 계층의 가중치와 같습
니다. 하지만 더 깊은 신경망 내부 계층의 그레이디언트는 그렇지 않습니다. 따라서 그레이디
언트를 계산해야만 합니다. 파이토치는 역전파 과정에서 모든 계층의 경사도를 계산하지만, 저
장하지는 않습니다(
requires
_
grad
가
False
일 때). 하지만 역전파 과정에 훅을 달아두면, 파
이토치가 계산한 그레이디언트를 저장할 수 있겠죠. 이를 위해
Hook
과 유사한
HookBwd
클래
스를 만듭니다. 단 이번에는 활성 대신 그레이디언트를 저장합니다.
classHookBwd():
def__init__(self,m):
self.hook=m.register_backward_hook ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month, and much more.
O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.
Julian F.
Head of Cybersecurity
I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.
Addison B.
Field Engineer
I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.
Amir M.
Data Platform Tech Lead
I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.