Supplementary Data: NEP Machine Learning Potential Training for C24 Monolayers of qHP and qTP

This folder contains all input and output files for training the C24-NEP potential function. The combined files of C24 monolayers for qHP and qTP include 325 qHP monolayer structures and 175 qTP monolayer structures, totaling 500 structures. Among these, 400 structures (38,400 atoms) were randomly selected as the training dataset and stored in the file train.xyz, while the remaining 100 structures (9,600 atoms) serve as the test dataset in the file test.xyz. The file nep.in contains hyperparameter settings related to NEP training. The file nep.txt is the potential function file output during the training process; additionally, we performed a total of 300,000 training steps, with a potential function file generated every 100,000 steps. Furthermore, the output files for energy, force, and virial of the training and test sets are energy\_train.out, energy\_test.out, force\_train.out, force\_test.out, virial\_train.out, and virial\_test.out respectively. The file loss.out records the loss function values throughout the training.