Replies: 1 comment
-
|
nvidia-smi 下的 GPU 编号默认使用 PCI_BUS_ID，而 PyTorch 代码默认情况下设备排序是 FASTEST_FIRST，可以看看这篇文章https://blog.csdn.net/sdnuwjw/article/details/111615052 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
为什么设置了set CUDA_VISIBLE_DEVICES=1 ，但并没有选择，标号为1的gpu进行训练
Beta Was this translation helpful? Give feedback.
All reactions