JinhuaSu / mpi_lsr

mpi_lsr

mpi4py examples and use mpi4py to implement the parallel least square regression

reference

• https://github.com/wangshusen/DeepLearning/blob/master/Slides/14_Parallel_1.pdf
• https://mpi4py.readthedocs.io/en/stable/tutorial.html

Installation

requirement: python3.x , build-essentials(gcc ..)

recommand OS: unix

• build mpi source: https://mpi4py.readthedocs.io/en/stable/appendix.html#building-mpi
  • or download pre-build: https://www.mpich.org/downloads/

  # sudo user
  sudo ./configure --enable-shared --prefix=/usr/local/mpich
  make
  make install
  

  • install distribution version by package management tools: for me(MacOS 13, older version can direct use brew install mpich), brew install --build-from-source mpich https://formulae.brew.sh/formula/mpich
• pip install mpi4py https://mpi4py.readthedocs.io/en/stable/install.html
• other common packages: numpy

Build Error

checking whether Fortran 77 and C objects are compatible... no checking for file... file checking for linker for Fortran main program... configure: error: Could not determine a way to link a Fortran test program!

Warning: You are using macOS 13. We do not provide support for this pre-release version. You will encounter build failures with some formulae. Please create pull requests instead of asking for help on Homebrew's GitHub, Twitter or any other official channels. You are responsible for resolving any issues you experience while you are running this pre-release version.

Error: texinfo: undefined method `on_monterey' for #Class:0x00007fab3e924bb8

Run mpi script

mpiexec -n 4 python xx.py

mpiexec -n 4 python src/mpi_demo.py