Inquiry Regarding Random Perturbations in Diamond Silicon Experiments

[LiMahappy]

Hi,@mjwen I am currently attempting to train a neural network potential for PtRh, but the training results are not satisfactory. I have reviewed the attachment of your published paper "KLIFF: A framework to develop physics-based and machine learning interatomic potentials" and noticed that you have compressed and stretched diamond silicon with strains ranging from -10% to 10%, and applied random perturbations to the atoms at each strain level. Could you please share how you applied these random perturbations? Additionally, if you could provide the corresponding VASP input files, it would be greatly appreciated. I am looking forward to your reply. Best regards, LiMa