先决条件
docker
安装docker的官网教程
https://docs.docker.com/install
有需要的话可以再装一个docker-compose
https://docs.docker.com/compose/install/#install-compose
nvidia-docker
github上安装nvidia-docker的教程
https://github.com/NVIDIA/nvidia-docker
tensorflow
tensorflow官网对docker的支持已经非常完善，只要照着官网进行操作即可
https://www.tensorflow.org/install/docker
实际运行时的命令如下
docker run --runtime=nvidia -it --rm -p 8888:8888 -v /home/tian/docker/docker-volumes/tensorflow-projects:/notebooks/projects/ --name tensorflow --net=host --env HTTP_PROXY="http://127.0.0.1:8118" --env HTTPS_PROXY="https://127.0.0.1:8118" tensorflow/tensorflow:latest-gpu-py3
其中指定了映射的目录卷，容器名字，以及本地ss+privoxy搭建的代理等信息。
这个docker镜像是tensorflow官方提供的，集成了常用的py3的包、jupyter等，使用非常方便。
pytorch
pytorch也是常用的类似tensorflow的工具，由于其使用简单、方便debug等特点而颇受青睐，在其官网的安装教程中使用cuda、conda之类的安装步骤比较详尽，但对于docker构建镜像却只字未提。所以想用docker搭建一个像tensorflow官方提供的镜像那样方便好用的镜像还是比较费劲的，我根据网上搜到的极为前辈的经验，整合了三个不同的pytorch镜像放在了dockerhub上，下面是具体的dockerfile，读者可根据自己需要修改和编译。
https://cloud.docker.com/repository/docker/tianhaoo/pytorch-jupyter-gpu
- pytorch-gpu
FROM nvidia/cuda:9.0-base-ubuntu16.04
# Install some basic utilities
RUN apt-get update && apt-get install -y \
curl \
ca-certificates \
sudo \
git \
bzip2 \
libx11-6 \
&& rm -rf /var/lib/apt/lists/*
# Create a working directory
RUN mkdir /app
WORKDIR /app
# Create a non-root user and switch to it
RUN adduser --disabled-password --gecos '' --shell /bin/bash user \
&& chown -R user:user /app
RUN echo "user ALL=(ALL) NOPASSWD:ALL" > /etc/sudoers.d/90-user
USER user
# All users can use /home/user as their home directory
ENV HOME=/home/user
RUN chmod 777 /home/user
# Install Miniconda
RUN curl -so ~/miniconda.sh https://repo.continuum.io/miniconda/Miniconda3-4.5.1-Linux-x86_64.sh \
&& chmod +x ~/miniconda.sh \
&& ~/miniconda.sh -b -p ~/miniconda \
&& rm ~/miniconda.sh
ENV PATH=/home/user/miniconda/bin:$PATH
ENV CONDA_AUTO_UPDATE_CONDA=false
# Create a Python 3.6 environment
RUN /home/user/miniconda/bin/conda install conda-build \
&& /home/user/miniconda/bin/conda create -y --name py36 python=3.6.5 \
&& /home/user/miniconda/bin/conda clean -ya
ENV CONDA_DEFAULT_ENV=py36
ENV CONDA_PREFIX=/home/user/miniconda/envs/$CONDA_DEFAULT_ENV
ENV PATH=$CONDA_PREFIX/bin:$PATH
# CUDA 9.0-specific steps
RUN conda install -y -c pytorch \
cuda90=1.0 \
magma-cuda90=2.3.0 \
"pytorch=0.4.1=py36_cuda9.0.176_cudnn7.1.2_1" \
torchvision=0.2.1 \
&& conda clean -ya
# Install HDF5 Python bindings
RUN conda install -y h5py=2.8.0 \
&& conda clean -ya
RUN pip install h5py-cache==1.0
# Install Torchnet, a high-level framework for PyTorch
RUN pip install torchnet==0.0.4
# Install Requests, a Python library for making HTTP requests
RUN conda install -y requests=2.19.1 \
&& conda clean -ya
# Install Graphviz
RUN conda install -y graphviz=2.38.0 \
&& conda clean -ya
RUN pip install graphviz==0.8.4
# Install OpenCV3 Python bindings
RUN sudo apt-get update && sudo apt-get install -y --no-install-recommends \
libgtk2.0-0 \
libcanberra-gtk-module \
&& sudo rm -rf /var/lib/apt/lists/*
RUN conda install -y -c menpo opencv3=3.1.0 \
&& conda clean -ya
# Set the default command to python3
CMD ["python3"]
- pytorch-jupyter-cpu
FROM jupyter/datascience-notebook:latest
RUN conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ \
&& conda config --set show_channel_urls yes \
&& conda install pytorch torchvision -c pytorch \
&& conda install pytorch torchvision -c pytorch \
&& conda install pytorch torchvision -c pytorch \
&& conda install pytorch torchvision -c pytorch \
&& conda clean -ya
- pytorch-jupyter-gpu
# from https://github.com/anibali/docker-pytorch cuda-9.0
FROM tianhaoo/pytorch-gpu:1.0
RUN conda install -y jupyter \
&& conda install -y jupyter \
&& conda install -y jupyter \
&& conda clean -ya
CMD ["python3"]
