gpu命令cuda命令

weixin_ry5219775

于 2019-09-05 16:54:00 发布

阅读量918

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/kyle1314608/article/details/100595276

# device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
os.environ["CUDA_VISIBLE_DEVICES"]="0,1"
# test=torch.cuda.is_available()
print(torch.cuda.device_count())
print(torch.cuda.get_device_name(1))
print(torch.cuda.current_device())
torch.cuda.set_device(0)
https://developer.nvidia.com/cuda-toolkit-archivecuda 版本列表
显卡算力查看命令   nvidia-smicuda下载

https://developer.nvidia.com/cuda-10.0-download-archive?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal

cudnn 下载

https://developer.nvidia.com/compute/machine-learning/cudnn/secure/v7.3.0/prod/10.0_2018920/cudnn-10.0-windows10-x64-v7.3.0.29