Skip to content

Instantly share code, notes, and snippets.

@juliendehos
Created March 11, 2020 15:09
Show Gist options
  • Save juliendehos/d23d876b3c6c7d9d9e655aa535df361f to your computer and use it in GitHub Desktop.
Save juliendehos/d23d876b3c6c7d9d9e655aa535df361f to your computer and use it in GitHub Desktop.
Dockerfile for Polygames on Centos 7 (WIP)
FROM nvidia/cuda:10.2-cudnn7-devel-centos7
RUN yum update -y
RUN yum install -y git sudo wget which make cmake
RUN yum install -y centos-release-scl
RUN yum install -y devtoolset-8-gcc devtoolset-8-gcc-c++
RUN yum install -y epel-release
RUN yum install -y zeromq-devel
RUN useradd -ms /bin/bash -G wheel myuser
RUN echo "myuser ALL=(ALL) NOPASSWD:ALL" >> /etc/sudoers
USER myuser
ENV USER="myuser"
ENV HOME="/home/myuser"
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/rh/devtoolset-8/root/usr/lib/gcc/x86_64-redhat-linux/8/
WORKDIR $HOME
RUN wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
RUN bash Miniconda3-latest-Linux-x86_64.sh -b
ENV PATH="/home/myuser/miniconda3/bin:$PATH"
RUN source "/home/myuser/miniconda3/etc/profile.d/conda.sh"
RUN conda create --name myenv python=3.7 pip
RUN source activate myenv
RUN conda install numpy pyyaml mkl mkl-include setuptools cmake cffi typing
RUN conda install -c pytorch magma-cuda100
RUN conda install -c conda-forge tensorboardx
RUN pip install visdom
# RUN git clone --recursive https://github.com/pytorch/pytorch --branch=v1.1.0
# WORKDIR $HOME/pytorch
# ENV CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"}
# RUN scl enable devtoolset-8 -- python setup.py install
RUN scl enable devtoolset-8 -- conda install pytorch=1.1.0
WORKDIR $HOME
RUN git clone https://github.com/facebookincubator/Polygames.git
RUN mkdir $HOME/Polygames/build
WORKDIR $HOME/Polygames/build
RUN scl enable devtoolset-8 -- cmake ..
RUN scl enable devtoolset-8 -- make -j4
RUN mkdir $HOME/Polygames/tests/build
WORKDIR $HOME/Polygames/tests/build
RUN scl enable devtoolset-8 -- cmake ..
RUN scl enable devtoolset-8 -- make -j4
RUN ./polygames-tests
WORKDIR $HOME/Polygames/
# docker build -t polygames -f Dockerfile .
# docker run --rm -it polygames
# python -m pypolygames train --game_name Hex11 --device=cpu
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment