Installing Caffe with CUDA in Conda

The following guide shows you how to install install Caffe with CUDA under the Conda virtual environment.

Assumptions

• Ubuntu OS
• NVIDIA GPU with CUDA support
• Conda (see installation instructions here)
• CUDA (installed by system admin)

Specifications

This guide is written for the following specs:

• Ubuntu 16.04
• Python 2.7
• CUDA 8
• cuDNN v7.1
• Miniconda 2
• OpenCV3

Guide

First, get cuDNN by following this cuDNN Guide.

Let’s create a virtual Conda environment called “caffe”:

conda create -n caffe python=2.7

You many of course use a different environment name, just be sure to adjust accordingly for the rest of this guide

After it prepares the environment and installs the default packages, activate the virtual environment via:

conda activate caffe
# to deactivate: conda deactivate caffe

Now let’s install the necessary dependencies in our current caffe environment:

conda install lmdb openblas glog gflags hdf5 protobuf leveldb boost opencv cmake numpy=1.15 -y
conda install -c conda-forge doxygen -y

Let’s clone caffe’s repo and its submodules into our home directory.

cd ~
git clone --recursive https://github.com/BVLC/caffe.git
cd caffe
git submodule update --init --recursive

We shall avoid polluting the caffe source tree by building within a build folder

mkdir build && cd build
pwd #=> ~/caffe/build

We shall now build the package using CMake with the following flags

cmake -DBLAS=open -DCUDNN_INCLUDE=$CUDA_HOME/include/ -DCUDNN_LIBRARY=$CUDA_HOME/lib64/libcudnn.so -DCMAKE_PREFIX_PATH=$CONDA_PREFIX -DCMAKE_INSTALL_PREFIX=$CONDA_PREFIX -DCMAKE_CXX_FLAGS="-std=c++11" ..

• CMake variable BLAS=open indicates that we would like use OpenBLAS instead of the default which is ATLAS
• CMake variable CUDNN_INCLUDE indicates where to find the include directory for your cuDNN
• CMake variable CUDNN_LIBRARY indicates where to find the library path for your cuDNN
• CMake variable CMAKE_PREFIX_PATH tells CMake to look for packages in your conda environment before looking in system install locations (like /usr/local)
• CMake variable CMAKE_INSTALL_PREFIX indicates where to install Caffe binaries
• CMake variable DCMAKE_CXX_FLAGS indicates which C++ compiler version to use
• CMake variable CPU indicates whether or not to use CPU-only installation
• Also see the list of the available make flags and their default values

Let’s find out how many cores your machine has

cat /proc/cpuinfo | grep processor | wc -l
#=> n

Let’s make the package efficiently by maximising the number of jobs for it. General rule of thumb is to use 1 + n number of jobs where n is the output from the previous command. i.e. number of cores. Mine was 24 so I run the following

make all -j 25
make pycaffe

After make is completed, we are now finally ready to install

make install

Now we run test

make runtest

You’d think we’re done, but not quite! We have to point the $PYTHONPATH environment variable to our build folder like so

export PYTHONPATH=$HOME/caffe/python:$PYTHONPATH

However it will be tedious to type that everytime we activate our environment. You may append that line to .bash_profile or .bashrc but some variables such as $PYTHONPATH are potentially used in many environments and it could lead to python import errors when the paths contain different modules sharing the same name. For instance, both caffe and caffe2 contain a module named ‘caffe’.

The solution to overcome this is to write a script to save our environment variables within our environemnt so that they get loaded automatically every time we activate our environment and get unset automatically when we deactivate our environment. The following steps are an adaptation of this guide stated in the official Conda documentation.

Let’s enter our environment directory and do the following

cd $CONDA_PREFIX
mkdir -p ./etc/conda/activate.d
mkdir -p ./etc/conda/deactivate.d
touch ./etc/conda/activate.d/env_vars.sh
touch ./etc/conda/deactivate.d/env_vars.sh

Edit ./etc/conda/activate.d/env_vars.sh as follows:

#!/bin/sh

export PYTHONPATH=$HOME/caffe/python:$PYTHONPATH

Edit ./etc/conda/deactivate.d/env_vars.sh as follows:

#!/bin/sh

unset PYTHONPATH

Now let’s reload the current environment to reflect the variables

conda activate caffe

We are now ready to test if caffe has been installed correctly with CUDA

cd ~
# To check if Caffe build was successful
python -c 'import caffe; caffe.set_mode_gpu()' 2>/dev/null && echo "Success" || echo "Failure"
#=> Success