Table 2.

$\text{cv}{R}^2$ for the five ML methods and for BLUP across 10 resampling runs applied to the wheat dataset. The best performance for each trait is in boldface. The average ranks for computation of the Friedman test are on the bottom line

Trait/method	Lasso	Ridge	BLUP	GBM	RF	SVM
Yield (drought)	0.023	0.060	0.217	0.051	0.172	0.219
Yield (irrigated)	0.084	0.162	0.253	0.132	0.184	0.258
TKW	0.172	0.240	0.277	0.218	0.242	0.304
DTH	0.292	0.325	0.381	0.325	0.358	0.394
Average rank	6.00	4.12	2.00	4.88	3.00	1.00