小行星

热爱平淡,向往未知

  • Home
  • Categories
  • Archives
  • About
  • top
  • Search
  • Table of Contents
  • Overview

fangyh

最爱的是那苍穹之外的浩渺宇宙
597 posts
4 categories
66 tags
0%

Loss.backward() raises error ‘grad can be implicitly created only for scalar outputs’

Posted on 2018-08-Mon Views: Views:

输出的loss需要是一个值
loss.sum().backward()

subprocess.call 使用非sh
pytorch retain_graph=True
© 2022 fangyh
Powered by Hexo & NexT.Muse