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

关于显存占用过大问题 #3687

Open
onlywl9598 opened this issue Apr 3, 2024 · 1 comment
Open

关于显存占用过大问题 #3687

onlywl9598 opened this issue Apr 3, 2024 · 1 comment
Assignees
Labels
bug Something isn't working

Comments

@onlywl9598
Copy link

onlywl9598 commented Apr 3, 2024

问题确认 Search before asking

我想咨询一下,如果我训练自己的数据集,建议使用的显存大小是多少,目前,我的设备是12G显存,我使用pp-liteseg模型,设置batch size为4都出现OOM

环境 Environment

  • OS: Linux
  • paddlepaddle-gpu: 2.4.0rc0.post111
  • PaddleSeg: 2.9.0
  • python: 3.7.16
  • CUDA: 12.2
  • CUDNN: 8.1.0
  • GCC: 7.5.0
  • 显卡: NVIDIA GeForce RTX 4070
  • 显存: 12G
@onlywl9598 onlywl9598 added the bug Something isn't working label Apr 3, 2024
@onlywl9598 onlywl9598 changed the title SignalInfo: *** SIGABRT (@0x3e80030c6bc) received by PID 3196604 (TID 0x7f7d7989b3c0) from PID 3196604,问题是因为显存占用过大导致 关于显存占用过大问题 Apr 3, 2024
@onlywl9598 onlywl9598 reopened this Apr 3, 2024
@shiyutang
Copy link
Collaborator

你好,可以尝试减小图片输入的尺度,不过这个也许会对精度产生一定影响,12G显存不小,但也对于单卡训练来说也不是很大。其他降低显存方法可以采用fp16训练等。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants