Skip to content
Permalink
master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

pytorch-osx-build/BuildInstractions-1.0rc1.md

Go to file
 
 
Cannot retrieve contributors at this time
Build Instrctions of Pytorch 1.0rc1 on Mac OS X 10.13 Dependencies Clone Pytorch Repo Update Dependency.cmake Compilation Installation
106 lines (74 sloc) 3.19 KB
Raw Blame
Open in GitHub Desktop

Build Instrctions of Pytorch 1.0rc1 on Mac OS X 10.13

By TomHeaven @ 2018.11.30

If you only want to use pytorch on Mac with CUDA support, first go to release page: https://github.com/TomHeaven/pytorch-osx-build/releases and try to find proper wheel pakcages.

However, as their are mulitple versions of python (2.7, 3.6, 3.7) and CUDA (8.0, 9.0, 9.2, 10.0). It's more and more time consuming to support all combinations of them. If your combination of python and CUDA is not listed in the release page, you may need to compile a wheel package yourself.

This guide will help you compile a wheel package of pytorch-1.0rc1 on Mac OS X 10.13 with CUDA support.

Note compilation on or for 10.12 will encounter a strange error due to clang does not recognize -fopenmp at 98%. I have no idea how to fix it.

Dependencies

Mandatory dependencies:

  • Mac OS 10.13.
  • XCode 8.0. (XCode 9.0 requires CUDA 9.2 or above.)
  • CUDA 9.0 and CuDNN7.0 (CUDA 9.2 and CuDNN7.1 should also work, but not tested).
  • OpenMP for multi-thread computation. Install using Homebrew:
brew install libomp
brew link --overwrite libomp
  • Openblas for CPU accelerated matrix computation and so on.
brew install openblas
brew link --overwrite openblas

You can also use MKL from Intel to replace Openblas (Not tested).

Note at the time of writing this guide, Tensorflow can be compiled but does not work properly with CUDA 9.2. If you also need Tensorflow, use CUDA 9.0.

Clone Pytorch Repo

Install git and clone pytorch repo by

brew install git
git clone https://github.com/pytorch/pytorch
cd pytorch
git checkout v1.0rc1
git submodule update --init

Update Dependency.cmake

The default OpenMP detection will fail. So let's just manually set related build flags and skip the detection by modifying cmake/Dependency.cmake:448 as

# ---[ OpenMP
#if(USE_OPENMP)
  #find_package(OpenMP)
  
  set(OpenMP_C_FLAGS "-Xpreprocessor -fopenmp -I/usr/local/include -L/usr/local/lib -lomp -lgomp")
  set(OpenMP_CXX_FLAGS "-Xpreprocessor -fopenmp -I/usr/local/include -L/usr/local/lib -lomp -lgomp")
  set(OpenMP_EXE_LINKER_FLAGS "-Xpreprocessor -fopenmp -L/usr/local/lib -lomp -lgomp")
  caffe2_update_option(USE_OPENMP ON)

  #if(OPENMP_FOUND)
    message(STATUS "Adding " ${OpenMP_CXX_FLAGS})
    set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
    set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${OpenMP_EXE_LINKER_FLAGS}")
  #else()
  #  message(WARNING "Not compiling with OpenMP. Suppress this warning with -DUSE_OPENMP=OFF")
   # caffe2_update_option(USE_OPENMP OFF)
  #endif()
#endif()

Compilation

Compile using

# set CUDA_ARCH
export TORCH_CUDA_ARCH_LIST="3.0;3.5;5.2;6.1"
# compile for python2
pip install pyyaml
python setup.py bdist_wheel
# compile for python3
pip3 install pyyaml
python3 setup.py bdist_wheel

The wheel packages will be avaiable at dist folder if everything goes smoothly.

Installation

Just cd to dist and install wheel packages using pip.

cd dist
pip install pytorch-xxx-cp27-xxx.wheel
pip3 install pytorch-xxx-cp3x-xxx.wheel