Docker利用DockerFile创建部署NVIDIA+PyTorch容器的详细过程

mkdir mycode
cd mycode/
touch Dockerfile

# 设置基础映像
FROM nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
# 定义构建参数
# 例如ARG USER=test为USER变量设置默认值"test"。
ARG USER=test
ARG PASSWORD=${USER}123$
# 处理讨厌的 Python 3 编码问题
ENV LANG C.UTF-8
ENV DEBIAN_FRONTEND noninteractive
ENV MPLLOCALFREETYPE 1
# 更新软件包列表并安装软件属性通用包
RUN apt-get update && apt-get install -y software-properties-common
# 添加Python ppa，以便后续安装Python版本
RUN add-apt-repository ppa:deadsnakes/ppa
# 安装Ubuntu的常用软件包，包括wget、vim、curl、zip、unzip等
RUN apt-get update && apt-get install -y \
    build-essential \
    git \
    wget \
    vim \
    curl \
    zip \
    zlib1g-dev \
    unzip \
    pkg-config \
    libgl-dev \
    libblas-dev \
    liblapack-dev \
    python3-tk \
    python3-wheel \
    graphviz \
    libhdf5-dev \
    python3.9 \
    python3.9-dev \
    python3.9-distutils \
    openssh-server \
    swig \
    apt-transport-https \
    lsb-release \
    libpng-dev \
    ca-certificates &&\
    apt-get clean &&\
    ln -s /usr/bin/python3.9 /usr/local/bin/python &&\
    ln -s /usr/bin/python3.9 /usr/local/bin/python3 &&\
    curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py &&\
    python3 get-pip.py &&\
    rm get-pip.py &&\
    # 清理APT缓存以减小Docker镜像大小
    rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
# 设置时区
ENV TZ=Asia/Seoul
RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
# 为应用程序创建一个用户
RUN useradd --create-home --shell /bin/bash --groups sudo ${USER}
RUN echo ${USER}:${PASSWORD} | chpasswd
USER ${USER}
ENV HOME /home/${USER}
WORKDIR $HOME
# 安装一些Python库，例如numpy、matplotlib、scipy等
RUN python3 -m pip --no-cache-dir install \
    blackcellmagic\
    pytest \
    pytest-cov \
    numpy \
    matplotlib \
    scipy \
    pandas \
    jupyter \
    scikit-learn \
    scikit-image \
    seaborn \
    graphviz \
    gpustat \
    h5py \
    gitpython \
    ptvsd \
    Pillow==6.1.0 \
    opencv-python
# 安装PyTorch和DataJoint等其他库
# 其中torch==1.13.1表示安装版本为1.13.1的PyTorch
RUN python3 -m pip --no-cache-dir install -i https://pypi.doubanio.com/simple/ \
    torch==1.13.1 \
    torchvision==0.14.1 \
    torchaudio==0.13.1 \
    'jupyterlab>=2'
RUN python3 -m pip --no-cache-dir install datajoint==0.12.9
# 设置环境变量LD_LIBRARY_PATH，以便支持性能分析库
ENV LD_LIBRARY_PATH /usr/local/cuda/extras/CUPTI/lib64:${LD_LIBRARY_PATH}
# 启动ssh服务器，打开22号端口
USER root
RUN mkdir /var/run/sshd
EXPOSE 22
CMD ["/usr/sbin/sshd", "-D"]

$ docker build -t jboss-eap:v1 .