Centos安装invidia-docker2
<h4 id="显卡驱动安装">显卡驱动安装</h4><p>如果是普通用户，命令前面加上sudo</p>
<pre><code class="language-bash">yum -y install gcc gcc-c++ wget
yum install gcc kernel-devel-($uname -r) kernel-headers
</code></pre>
<p>显卡驱动安装</p>
<p><strong>error:</strong> yum安装提示“没有可用软件包”</p>
<pre><code class="language-bash">yum install -y epel-release
# yum install -y kernel-headers kernel-devel dkms
yum install -y dkms
</code></pre>
<p>测试安装</p>
<pre><code class="language-bash">nvidia-smi
</code></pre>
<h4 id="安装docker">安装Docker</h4>
<p>安装docker</p>
<pre><code class="language-bash">yum-config-manager \
--add-repo \
https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
sed -i 's/download.docker.com/mirrors.aliyun.com\/docker-ce/g' /etc/yum.repos.d/docker-ce.repo
yum install docker-ce docker-ce-cli containerd.io
</code></pre>
<h4 id="启动docker">启动Docker</h4>
<pre><code class="language-bash">systemctl enable docker
systemctl start docker
</code></pre>
<h4 id="安装nvidia-docker2">安装nvidia-docker2</h4>
<p>更新仓库<br>
CentOS 7 安装 nvidia-docker2</p>
<pre><code class="language-bash">distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | \
sudo tee /etc/yum.repos.d/nvidia-docker.repo
DIST=$(sed -n 's/releasever=//p' /etc/yum.conf)
DIST=${DIST:-$(. /etc/os-release; echo $VERSION_ID)}
sudo yum makecache
</code></pre>
<pre><code class="language-bash"># 拉取镜像
docker pull nvidia/cuda:10.2-cudnn7-runtime-centos7
docker pull nvidia/cuda:10.2-cudnn7-runtime-ubuntu16.04
# 测试
docker run --runtime=nvidia --rm nvidia/cuda:10.2-cudnn7-runtime-centos7 nvidia-smi
# bash运行
docker run -it --runtime=nvidia --rm nvidia/cuda:10.2-cudnn7-runtime-centos7 bash
</code></pre>
<h4 id="docker操作">Docker操作</h4>
<p>Docker教程-使用</p>
<p><strong>镜像操作：</strong></p>
<pre><code class="language-bash"># 查看所有镜像
docker images
# 停止运行
docker stop $(docker ps -a -q)
# 删除镜像
docker rmi 镜像Id
#把容器打包成镜像
docker commit 0bd244689ed2 ubuntu-vim
</code></pre>
<p><strong>容器操作：</strong></p>
<pre><code class="language-bash"># 查看容器的详细信息
docker ps -a
# 后台运行容器，返回容器Id
docker run -i -t -d ubuntu /bin/bash
# 退出容器，容器的运行也会终止
docker attach 容器id
# 进入这个运行中的容器，但是随着容器的退出，容器运行不会终止
docker exec -it 容器id /bin/bash
# 退出
exit
# 删除指定容器
docker rm -f 容器id
# 停用全部运行中的容器
docker stop $(docker ps -q)
# 删除全部容器
docker rm $(docker ps -aq)
</code></pre>
<p><strong>运行命令：</strong></p>
<pre><code class="language-bash">docker run -itd -p 50001:22 ubuntu:18.04
docker run -it -v /home/dock/Downloads:/usr/Downloads ubuntu64 /bin/bash
docker run --runtime=nvidia -itd -p 50002:22 -v /home/lyq/shareWithDocker:/user/Downloads 56b5278bbbf0
</code></pre>
<p>删除容器</p>
<h4 id="vscode连接docker">VScode连接Docker</h4>
<p>搭建深度学习环境之二：远程连接Docker容器</p>
<pre><code class="language-bash"># 查看linux版本
cat /etc/issue
# 启动容器
docker run -itd -p 50001:22 ubuntu:18.04
# 进入容器
docker exec -it 7338ded2d7e6 /bin/bash
</code></pre>
<pre><code class="language-bash"># 安装python3
apt-get update
apt-get install -y software-properties-common
add-apt-repository ppa:deadsnakes/ppa
apt-get update
apt-get install -y python3.6 python3.6-dev python3-pip
ln -sfn /usr/bin/python3.6 /usr/bin/python3
ln -sfn /usr/bin/python3 /usr/bin/python
ln -sfn /usr/bin/pip3 /usr/bin/pip
pip install --upgrade pip
# 安装pytorch
pip3 install torch torchvision torchaudio
# 安装kernel
pip3 install ipykernel
/usr/bin/python -m pip install -U notebook
</code></pre>
<h4 id="git配置">git配置</h4>
<pre><code class="language-bash">sudo apt-get install git
git config --global user.name "docker"
git config --global user.email "docker"
</code></pre><br><br>
来源：https://www.cnblogs.com/wuu02/p/15759290.html
頁:
[1]