Open MatSci ML Toolkit is a framework for prototyping and scaling out deep learning models for materials discovery supporting widely used materials science datasets, and built on top of PyTorch Lightning, the Deep Graph Library, and PyTorch Geometric.
MIT License
155
stars
25
forks
source link
[Bug]: `FAENet` uses a different periodic boundary condition implementation #309
Models are expected to re-use the same pipeline for computing periodic boundary conditions/images - i.e. matsciml.datasets.transforms.PeriodicBoundaryConditions and subroutines.
Actual behavior
Per this line of code in FAENet.read_batch, it actually calls the non-pymatgen version of the PBC calculation.
Expected behavior
Models are expected to re-use the same pipeline for computing periodic boundary conditions/images - i.e.
matsciml.datasets.transforms.PeriodicBoundaryConditionsand subroutines.Actual behavior
Per this line of code in
FAENet.read_batch, it actually calls the non-pymatgen version of the PBC calculation.Steps to reproduce the problem
N/A
Specifications
N/A