The table below shows a list of published neuroevolution potential (NEP) models.
The entries in the first column are linked to NEP model files in nep.txt format.
The second column links to the publication in which the respective model has been constructed and that should be cited when using this model. Please also refer to the original publication for supporting information, which is typically provided on zenodo and includes, e.g., reference data and supporting scripts.
The third column contains links to structure files that represent typical structures associated with this model. These structures can, e.g., be used for testing the model file. This can be done with ease using the calorine package as follows:
from ase.io import read
from calorine.calculators import CPUNEP
calculator = CPUNEP('nep.txt')
structure = read('structure.xyz')
structure.calc = calculator
print(structure.get_potential_energy())