Tutorial-ACE4LDSim is the repository for the tutorial paper "Lattice Dynamics Modeling of Thermal Transport in Solids Using Machine-Learned Atomic Cluster Expansion Potentials: A Tutorial" (J. Appl. Phys. 137, 081101 (2025); doi: 10.1063/5.0251119). It includes all the data, codes and scripts necessary to reproduce the results presented in the tutorial. Some precalculated data of this repository can be accessed at zenodo, which can be fetched with the codes in code_available.ipynb
We recommend creating a new environment with Python 3.9.
conda create -n tut python=3.9.18
conda activate tut
Firstly, we deal with the installation of pacemaker, which is detailed in the origin document. For convenience, here to quote some pieces:
pip install tensorflow==2.8.0
git clone https://github.com/ICAMS/TensorPotential.git
cd TensorPotential
pip install --upgrade .
git clone https://github.com/ICAMS/python-ace.git
cd python-ace
pip install --upgrade .
Secondly, we deal with other packages:
pip install hiphive
pip install ase==3.22.1
Finally, we install phono3py (see origin document):
conda install -c conda-forge phono3py
The requirements.txt is also provided.
In addition, to use ACE potential in LAMMPS, please compilate LAMMPS detailed in here. To use kaldo package, it has to be installed in a new environment (refer to origin document) due to the conflict with pacemaker.
Start with code_available.ipynb right now!