loss.out

This files contains the terms that enter the loss function for every 100-th generation.

If a potential model is trained, each row contains the following fields:

gen L_t L_1 L_2 L_e_train L_f_train L_v_train L_e_test L_f_test L_v_test

where

• gen is the current generation.

• L_t is the total loss function.

• L_1 is the loss function related to the \(\mathcal{L}_1\) regularization.

• L_2 is the loss function related to the \(\mathcal{L}_2\) regularization.

• L_e_train is the energy RMSE (in units of eV/atom) for the training set.

• L_f_train is the force RMSE (in units of eV/Å) for the training set.

• L_v_train is the virial RMSE (in units of eV/atom) for the training set.

• L_e_test is the energy RMSE (in units of eV/atom) for the test set.

• L_f_test is the force RMSE (in units of eV/Å) for the test set.

• L_v_test is the virial RMSE (in units of eV/atom) for the test set.

If a dipole model is trained, each row contains the following fields:

gen L_t L_1 L_2 L_mu_train L_mu_test

where

• L_mu_train is the dipole RMSE (per atom) for the training set.

• L_mu_test is the dipole RMSE (per atom) for the test set.

If a polarizability model is trained, each row contains the following fields:

gen L_t L_1 L_2 L_alpha_train L_alpha_test

where

• L_alpha_train is the polarizability RMSE (per atom) for the training set.

• L_alpha_test is the polarizability RMSE (per atom) for the test set.