nvidia 驱动使用的一些运维记录：解决常见问题的完整指南

一：nvidia 查看显卡

nvidia-smi --query-gpu=gpu_name,gpu_bus_id --format=csv

二：nvidia-container-toolkit

##

distribution=$(. /etc/os-release;echo $ID$VERSION_ID)

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

yum install -y nvidia-container-toolkit

yum install -y nvidia-docker2

systemctl restart docker

###

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html#id5

distribution=$(. /etc/os-release;echo $ID$VERSION_ID)

&& curl -s -L https://nvidia.github.io/libnvidia-container/gpgkey | sudo apt-key add -

&& curl -s -L https://nvidia.github.io/libnvidia-container/$distribution/libnvidia-container.list | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

sudo apt-get update

&& sudo apt-get install -y nvidia-container-toolkit

验证

sudo ctr image pull docker.io/nvidia/cuda:11.6.2-base-ubuntu20.04

sudo ctr run --rm -t \ --runc-binary=/usr/bin/nvidia-container-runtime \ --env NVIDIA_VISIBLE_DEVICES=all \ docker.io/nvidia/cuda:11.6.2-base-ubuntu20.04 \ cuda-11.6.2-base-ubuntu20.04 nvidia-smi

三：GPU使用技巧

Q3：下载tensorflow 对应版本镜像

#tensorflow 镜像

https://hub.docker.com/r/tensorflow/tensorflow/

docker pull tensorflow/tensorflow:1.5.0

docker pull tensorflow/tensorflow:1.5.0-gpu

Q2：验证GPU

lspci | grep -i nvidia

docker run --gpus all --rm nvidia/cuda:12.2.0-base-ubuntu22.04 nvidia-smi

Q1：安装指定tensorflow版本

lscpu 查看CPU是否支持avx 指令，1.6.0版本以上需要avx指令

https://tensorflow.google.cn/install/pip?hl=zh-cn

https://pypi.org/project/tensorflow/#files

https://tensorflow.google.cn/install/pip?skip_cache=true&hl=de

tensorflow-gpu==1.15：支持 GPU 的版本（适用于 Ubuntu 和 Windows）

pip3 install tensorflow-gpu==1.5.0

#tensorflow 镜像下载

https://hub.docker.com/r/tensorflow/tensorflow/

https://tensorflow.google.cn/install/docker?hl=zh-cn

https://tensorflow.google.cn/install/docker?hl=de&skip_cache=true#gpu_support

如：

docker pull tensorflow/tensorflow:2.1.0-gpu