*** RuntimeError: CUDA error: out of memory.

CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

 

这个error 的原因是,当期指定的GPU的显存不足,可以关闭现有的process,或者重指定显卡编号。

device = torch.device("cuda:0")
 
 
posted @ 2021-09-14 21:24  呦呦南山  阅读(16188)  评论(0编辑  收藏  举报