본문 바로가기 메뉴 바로가기

홍석쓰 블로그

프로필사진
  • 글쓰기
  • 관리
  • 태그
  • 방명록
  • RSS

홍석쓰 블로그

검색하기 폼
  • 분류 전체보기 (155)
    • Research (연구 관련) (151)
  • 방명록

backpropagation (1)
netD(fake.detach())의 이유

GAN을 공부하다가 tutorial 코드에 netD(fake.detach())가 어떤 원리인 지 이해가 안 갔다. 의도야 설명에 나온대로 netG에 backpropagation이 안되도록, 즉 첫번째 스텝에서는 netD만 학습하려는 것이라는 건 알겠다. (code: pytorch gan tutorial) ## Train with all-fake batch # Generate batch of latent vectors noise = torch.randn(b_size, nz, 1, 1, device=device) # Generate fake image batch with G fake = netG(noise) label.fill_(fake_label) # Classify all fake batch with D..

Research (연구 관련) 2019. 5. 11. 18:01
이전 1 다음
이전 다음
공지사항
최근에 올라온 글
최근에 달린 댓글
Total
Today
Yesterday
링크
TAG
  • 문경식
  • 머신러닝
  • pyrender
  • densepose
  • VAE
  • demo
  • pytorch
  • Pose2Mesh
  • spin
  • Docker
  • 에디톨로지
  • Generative model
  • focal length
  • Machine Learning
  • 2d pose
  • Transformation
  • part segmentation
  • 피트니스
  • nohup
  • 헬스
  • Interview
  • Virtual Camera
  • nerf
  • 컴퓨터비젼
  • 비전
  • 컴퓨터비전
  • world coordinate
  • 인터뷰
  • deep learning
  • camera coordinate
more
«   2025/07   »
일 월 화 수 목 금 토
1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31
글 보관함

Blog is powered by Tistory / Designed by Tistory

티스토리툴바