“…The results of the previously described machine-learning models for E%, F%, T%, and K/S, respectively, are summarized in Tables 2 , 3 , 4 , and 5 for convenience. In this work, a model's performance was quantified using three widely used error metrics, including RMSE, MAE, and R2, to exclude any possibility of bias in the evaluation of the model's performance 44 . Tables 2 , 3 , 4 , and 5 show that the training data's error metrics are RMSE 1 , MAE 1 , , whereas the testing data's error metrics are RMSE 1 , MAE 1 , and .…”