Table 2.

Model evaluation on the training, validation, and testing sets, showing the median and 1$\sigma$ uncertainty for $r^2$, root-mean-square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE)

Data set	$r^2$ [%]	RMSE	MAE	MAPE
Training	94.0${}^{\pm 1.0}$	0.107${}^{\pm 0.009}$	0.089${}^{\pm 0.009}$	7.3${}^{\pm 0.6}$
Validation	92.9${}^{\pm 1.1}$	0.128${}^{\pm 0.008}$	0.105${}^{\pm 0.009}$	8.3${}^{\pm 0.7}$
Testing	92.0${}^{\pm 1.0}$	0.134${}^{\pm 0.009}$	0.105${}^{\pm 0.009}$	8.3${}^{\pm 0.7}$