错误

RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.
毫无疑问，要想使用GPU，必须要安装显卡驱动，这样nvidia docker才能正常运行。


安装docker,安装nvidia

curl -s -L https://nvidia.github.io/nvidia-docker/centos7/x86_64/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo


如果报错多执行几次

下载 nvidia-docker2

查找可安装的nvidia docker版本

yum search --showduplicates nvidia-docker

安装​​nvidia-docker​​

yum install nvidia-docker-1.0.1-1.x86_64


配置docker

vim /etc/docker/daemon.json

{
    "data-root": "/docker",
    "default-runtime": "nvidia",
    "runtimes": {
        "nvidia": {
            "path": "nvidia-container-runtime",
            "runtimeArgs": []
        }
    }
}


添加一行 "default-runtime": "nvidia",

重启docker

systemctl restart docker

查看是否安装

yum list installed | grep nvidia   或者 rpm -qa | grep  nvidia

docker 部署gpu 服务 提示RuntimeError: Attempting_github