Table 3.

The mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and area under curve (AUC) of receiver operating characteristics (ROC) of the applied machine learning (ML) regression methods in predicting the grain yield per unit area of the Lallemantia iberica ecotypes with the training of other traits, using the K-fold cross-validation as the data splitting method.

ML method	AUC-ROC	MAE	MSE	RMSE
Linear regression	0.9969	0.176	0.760	0.871
Support vector regression (SVR); linear kernel	0.9979	0.120	0.681	0.825
SVR; Gaussian kernel	0.8157	11.981	250.224	15.818
SVR; polynomial kernel	0.8404	11.068	243.275	15.597
SVR; sigmoid kernel	0.2229	22.008	829.985	28.809
Random forest regression	0.9540	2.537	15.648	3.955
Gradient boosting decision tree regression	0.9695	2.097	10.907	3.302