Skip to content

README.md

Latest commit

 

History

History
29 lines (22 loc) · 1.05 KB

README.md

File metadata and controls

29 lines (22 loc) · 1.05 KB

BayesianSelection

Installation

Prerequisites

  • Python 3.9 - 3.11
  • Julia 1.10
  • An MPI implementation compatible with mpi4py (e.g. OpenMPI)

Julia Installation

To use ACE models and descriptor features, the code requires Julia, with the ACEpotentials (v0.6), ASE, and JuLIP packages installed.

A suitable Julia environment is provided via the Project.toml and Manifest.toml files, which contain a minimal number of required packages. Switching Julia to this project can be achieved by setting the environment variable export JULIA_PROJECT=., which is required so that importing the julia package from Python uses the correct project.

The packages can then be installed from a julia repl using:

using Pkg
Pkg.instantiate()

Python installation

The Python requirements can be installed via:

pip install .
python -c "import julia; julia.install()"

Reproducing the publication

The tools required to reproduce the publication are hosted in the Si_Surfaces directory. See the specific documentation here