FINETUNA: Fine-Tuning Accelerated Molecular Simulations

Are you doing structural optimizations with DFT or other electronic structure codes?? Try 🧐🐟 FINETUNA for accurate but 90% faster relaxation!

FINETUNA accelerates atomistic simulations by fine-tuning a pre-trained graph model in an active learning framework.

Installation is easy:

conda env create -f env.cpu.yml
conda activate finetuna
cd finetuna
pip install -e .
git clone https://github.com/Open-Catalyst-Project/ocp.git
cd ocp
pip install -e .
pip install git+https://github.com/ulissigroup/vasp-interactive.git

All pre-trained machine learning model checkpoint can be found here. We recommend to download the GemNet-dT all model. click here to download.

You are all set! Now in your VASP input folder, run the calculation by: finetuna_wrap.py -c /path/to/the/checkpoint.

Usage

If you have an ASE atoms object, see example 1 and 2.

If you have VASP input files (INCAR, KPOINTS, POTCAR, and POSCAR), see example 3.