MACE 0.3.14-foss-2023a-CUDA-12.1.1-ASE-3.24.0

Fast and accurate machine learning interatomic potentials with higher order equivariant message passing.

Accessing MACE 0.3.14-foss-2023a-CUDA-12.1.1-ASE-3.24.0

To load the module for MACE 0.3.14-foss-2023a-CUDA-12.1.1-ASE-3.24.0 please use this command on the BEAR systems (BlueBEAR and BEAR Cloud VMs):

📋 module load bear-apps/2023a
module load MACE/0.3.14-foss-2023a-CUDA-12.1.1-ASE-3.24.0

BEAR Apps Version

2023a

Architectures

EL8-icelake (GPUs: NVIDIA A100, NVIDIA A30)

The listed architectures consist of two parts: OS-CPU. The OS used is represented by EL and there are several different processor (CPU) types available on BlueBEAR. More information about the processor types on BlueBEAR is available on the BlueBEAR Job Submission page.

Extensions

• configargparse 1.7.1
• lightning-utilities-0.10.1
• lmdb 1.6.2
• mace_torch 0.3.14
• matscipy 1.1.1
• prettytable 3.9.0
• python-hostlist-2.2.1
• torch_ema 0.3
• torchmetrics 1.7.3

More Information

For more information visit the MACE website.

Dependencies

This version of MACE has a direct dependency on: ASE/3.24.0-gfbf-2023a CUDA/12.1.1 e3nn/0.4.4-foss-2023a-CUDA-12.1.1 foss/2023a GitPython/3.1.40-GCCcore-12.3.0 h5py/3.9.0-foss-2023a matplotlib/3.7.2-gfbf-2023a meson-python/0.13.2-GCCcore-12.3.0 orjson/3.9.15-GCCcore-12.3.0 Python/3.11.3-GCCcore-12.3.0 PyTorch-bundle/2.1.2-foss-2023a-CUDA-12.1.1 PyYAML/6.0-GCCcore-12.3.0 SciPy-bundle/2023.07-gfbf-2023a tqdm/4.66.1-GCCcore-12.3.0

Last modified on 8th September 2025