admin管理员组

文章数量:1531792

出现如下报错:

RuntimeError: CUDA error: invalid device ordinal
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.

原因:选择GPU设备时torch没有识别到。
解决方法:
加一句torch.cuda.device_count()代码

import torch
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "1"
torch.cuda.device_count()

本文标签: 报错debuggingCUDALAUNCHBLOCKINGpassing