Develop MLIP for molecular liquids

Notebook MACE (Machine-learning interatomic potentials).

Repository structure

• notebooks/: Jupyter notebooks
• src/: helper modules used by notebooks
• data/: datasets
• outputs/: generated figures/logs/models

Setup

Option 1: conda

conda env create -f environment.yml
conda activate mace
python -m ipykernel install --user --name mace --display-name "Python (mace)"
jupyter lab