caffe-docker.md

From: http://tleyden.github.io/blog/2014/10/25/running-caffe-on-aws-gpu-instance-via-docker/

ls -la /dev | grep nvidia

You should see:

crw-rw-rw-  1 root root    195,   0 Oct 25 19:37 nvidia0
crw-rw-rw-  1 root root    195, 255 Oct 25 19:37 nvidiactl
crw-rw-rw-  1 root root    251,   0 Oct 25 19:37 nvidia-uvm

You’ll have to adapt the DOCKER_NVIDIA_DEVICES variable below to match your particular devices. Here’s how to start the docker container:

$ DOCKER_NVIDIA_DEVICES="--device /dev/nvidia0:/dev/nvidia0 --device /dev/nvidiactl:/dev/nvidiactl --device /dev/nvidia-uvm:/dev/nvidia-uvm"
$ sudo docker run -ti $DOCKER_NVIDIA_DEVICES tleyden5iwx/caffe-gpu-master /bin/bash

Test:

$ cd /opt/caffe
$ make test && make runtest

caffe-gpu-master.dockerfile

# Don't edit this file directly, since it was generated from a template, 
# and you're changes will be *clobbered*.  Edit the template instead.

 
FROM tleyden5iwx/ubuntu-cuda


ENV PYTHONPATH /opt/caffe/python

# Add caffe binaries to path
ENV PATH $PATH:/opt/caffe/.build_release/tools

# Get dependencies
RUN apt-get update && apt-get install -y \
  bc \ 
  cmake \ 
  curl \ 
  gcc-4.6 \ 
  g++-4.6 \ 
  gcc-4.6-multilib \  
  g++-4.6-multilib \ 
  gfortran \ 
  git \ 
  libprotobuf-dev \
  libleveldb-dev \
  libsnappy-dev \
  libopencv-dev \
  libboost-all-dev \ 
  libhdf5-serial-dev \ 
  liblmdb-dev \  
  libjpeg62 \ 
  libfreeimage-dev \  
  libatlas-base-dev \  
  pkgconf \
  protobuf-compiler \ 
  python-dev \  
  python-pip \ 
  unzip \
  wget 

# Use gcc 4.6
RUN update-alternatives --install /usr/bin/cc cc /usr/bin/gcc-4.6 30 && \
  update-alternatives --install /usr/bin/c++ c++ /usr/bin/g++-4.6 30 && \ 
  update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-4.6 30 && \
  update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-4.6 30


# Allow it to find CUDA libs
RUN echo "/usr/local/cuda/lib64" > /etc/ld.so.conf.d/cuda.conf && \
  ldconfig 


# Clone the Caffe repo 
RUN cd /opt && git clone https://github.com/BVLC/caffe.git


# Glog 
RUN cd /opt && wget https://google-glog.googlecode.com/files/glog-0.3.3.tar.gz && \
  tar zxvf glog-0.3.3.tar.gz && \
  cd /opt/glog-0.3.3 && \
  ./configure && \
  make && \
  make install

# Workaround for error loading libglog: 
#   error while loading shared libraries: libglog.so.0: cannot open shared object file
# The system already has /usr/local/lib listed in /etc/ld.so.conf.d/libc.conf, so
# running `ldconfig` fixes the problem (which is simpler than using $LD_LIBRARY_PATH)
# TODO: looks like this needs to be run _every_ time a new docker instance is run,
#       so maybe LD_LIBRARY_PATh is a better approach (or add call to ldconfig in ~/.bashrc)
RUN ldconfig

# Gflags
RUN cd /opt && \
  wget https://github.com/schuhschuh/gflags/archive/master.zip && \
  unzip master.zip && \
  cd /opt/gflags-master && \
  mkdir build && \
  cd /opt/gflags-master/build && \
  export CXXFLAGS="-fPIC" && \
  cmake .. && \ 
  make VERBOSE=1 && \
  make && \
  make install

# Build Caffe core
RUN cd /opt/caffe && \
  cp Makefile.config.example Makefile.config && \
   \ 
  echo "CXX := /usr/bin/g++-4.6" >> Makefile.config && \
  sed -i 's/CXX :=/CXX ?=/' Makefile && \
  make all

# Add ld-so.conf so it can find libcaffe.so
ADD caffe-ld-so.conf /etc/ld.so.conf.d/

# Run ldconfig again (not sure if needed)
RUN ldconfig

# Install python deps
RUN cd /opt/caffe && \
  (pip install -r python/requirements.txt; easy_install numpy; pip install -r python/requirements.txt) && \
  easy_install pillow

# Numpy include path hack - github.com/BVLC/caffe/wiki/Setting-up-Caffe-on-Ubuntu-14.04
RUN NUMPY_EGG=`ls /usr/local/lib/python2.7/dist-packages | grep -i numpy` && \
  ln -s /usr/local/lib/python2.7/dist-packages/$NUMPY_EGG/numpy/core/include/numpy /usr/include/python2.7/numpy

# Build Caffe python bindings
RUN cd /opt/caffe && make pycaffe

ubuntu-cuda.dockerfile

FROM ubuntu:14.04
MAINTAINER Traun Leyden <traun.leyden@gmail.com>

# A docker container with the Nvidia kernel module and CUDA drivers installed

ENV CUDA_RUN http://developer.download.nvidia.com/compute/cuda/6_5/rel/installers/cuda_6.5.14_linux_64.run

RUN apt-get update && apt-get install -q -y \
  wget \
  build-essential 

RUN cd /opt && \
  wget $CUDA_RUN && \
  chmod +x *.run && \
  mkdir nvidia_installers && \
  ./cuda_6.5.14_linux_64.run -extract=`pwd`/nvidia_installers && \
  cd nvidia_installers && \
  ./NVIDIA-Linux-x86_64-340.29.run -s -N --no-kernel-module

RUN cd /opt/nvidia_installers && \
  ./cuda-linux64-rel-6.5.14-18749181.run -noprompt

# Ensure the CUDA libs and binaries are in the correct environment variables
ENV LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-6.5/lib64
ENV PATH=$PATH:/usr/local/cuda-6.5/bin