Skip to content

Latest commit

 

History

History
35 lines (24 loc) · 1.04 KB

README.md

File metadata and controls

35 lines (24 loc) · 1.04 KB
Raw

pybmix: Bayesian Mixture Models in Python

A Python interface to bayesmix

Installation

  1. install cmake>=3.21.0 from source following the official documentation

  2. install libtbb-dev=2020.1-2 (e.g. for Linux sudo apt install libtbb-dev=2020.1-2)

  3. clone the repository

git clone --recurse-submodules git@github.com:bayesmix-dev/pybmix.git
  1. setup a python environment with python=3.9 and install the following packages with pip
pip3 install numpy==1.22.4 scipy==1.7.3 matplotlib==3.5.2 2to3==1.0

  1. install protobuf==3.14.0 and libprotobuf=3.14.0 (e.g. conda install protobuf==3.14.0 will install both)

  2. add the path for 2to3 to the PATH environment variable, e.g.

export PYTHONPATH="path/to/2to3/"
  1. finally, to build the library
./build_pybmix.sh build

Note that, the argument build substitutes mkdir build, thus you can skip it in subsequent builds if only the new changes need to be compiled.