大山的专栏

docker-compose.yaml文件version: '3.8'services: tensorrt-dev: build: context: .. dockerfile: docker/Dockerfile image: worker:v0.0.1 restart: always volumes: - /tmp/.X11-unix:/tmp/.X11-unix environment:

服务器端配置第一步，修改/etc/docker/daemon.json文件替换为：{"default-runtime": "nvidia","runtimes": {"nvidia": {"path": "/usr/bin/nvidia-container-runtime","runtimeArgs": []}},"exec-opts": ["native.cgroupdriver=systemd"]"log-driver": "json-file","log-opts": {"

ELRepo源提供了nvidia-detect命令，会自动帮你寻找合适的驱动，然后根据显示的结果，用yum就能完成安装，重启之后不会有任何问题，省时省力。步骤如下：导入公共密钥，执行命令：rpm --import https://www.elrepo.org/RPM-GPG-KEY-elrepo.org安装elrepo, 执行命令：yum install https://www.elrepo.org/elrepo-release-7.el7.elrepo.noarch.rpm安装nvidia驱动，

ubuntu安装多个版本的cuda切换sudo rm -rf cudasudo ln -s /usr/local/cuda-10.0 /usr/local/cuda

Each ICudaEngine object is bound to a specific GPU when it is instantiated, eitherby the builder or on deserialization. To select the GPU, use cudaSetDevice() beforecalling the builder or deserializing the engine. Each IExecutionContext is boundto the s

查看Nvidia JetPack版本信息命令：sudo pip3 install jetson-statssudo jtop

ubuntu 自动安装nvidia显卡驱动步骤如下：sudo apt-get remove --purge nvidia*ubuntu-drivers devicessudo ubuntu-drivers autoinstallcat /etc/modprobe.d/blacklist-nouveau.confblacklist nouveauoptions nouveau modeset=0sudo update-initramfs -usudo reboot...