Prediction of Mg Alloy Corrosion Based on Machine Learning Models

Table 4

Prediction accuracy of each algorithm (the R² and MAE calculation results for each optimized model in the training set and testing set).

Models	Ecorr				Icorr
	R^2		MAE		R^2		MAE
	Training set	Testing set	Training set	Testing set	Training set	Testing set	Training set	Testing set
RF	0.900	0.790	0.025	0.042	0.850	0.730	0.029	0.041
MLR	0.430	0.140	0.079	0.093	0.160	0.040	0.092	0.043
SVR	0.590	0.500	0.067	0.078	0.600	0.400	0.068	0.090
XGBoost	0.970	0.740	0.013	0.049	0.950	0.690	0.020	0.053