Deep potential model and datasets used in the publication "Phase behavior of a machine-learning potential trained on stress-strain curves: The case of superionic water ice"

The files training_data_for_deepmd.zip and MLIP-model.zip contains the training data set and the deep potential model (MLP), respectively, for superionic and insulator ices X and XVIII used in the publication "Phase behavior of a machine-learning potential trained on stress-strain curves: The case of superionic water ice". The deep potential (DP) neural-network model was generated using the DeepMD-kit package. The training set was generated using AIMD/CP2K calculations for both undeformed and deformed states of the fcc and bcc superionic and insulator phase at temperatures of 1000, 1500, 2000, 2500, 3000, 3500 and 4000 K. Inputs for the CP2K calculations are available in the file cp2k_inputs.zip.