Validation results of the machine-learning model for the testing set. The results of the three submodels corresponding to different subsets are presented in the left, middle, and right panels, respectively. The comparisons of the departure coefficients calculated by the simulation models and the machine-learning submodels are presented in the top panels. In the bottom panels, we show the comparisons of the Hnα line luminosities. L^m and L^s denote the line luminosities computed using the departure coefficients from the machine-learning model predictions and the original simulation models, respectively. The solid and dashed red lines in the bottom panels correspond to error levels of 10% and 20%, respectively.