Table 1.

The training, validation and hold-out dataset test results in terms of MSE and R² for two data-splitting schemes: 6-fold cross-validation (with no holdout dataset test) and 5-fold cross-validation with a holdout test dataset.

	6-Fold cross validation				5-Fold cross validation with hold-out dataset test
	MSE (training)	R2 (training)	MSE (validation)	R2 (validation)	MSE (training)	R2 (training)	MSE (validation)	R2 (validation)	MSE (test)	R2 (test)
Basic linear	0.01054	0.50619	0.01062	0.50194	0.01042	0.51559	0.01053	0.51013	0.01120	0.45445
Ridge	0.01054	0.50616	0.01062	0.50196	0.01042	0.51553	0.01053	0.51016	0.01118	0.45520
Lasso	0.01054	0.50618	0.01062	0.50194	0.01042	0.51557	0.01053	0.51013	0.01119	0.45482
LARS	0.01117	0.47690	0.01120	0.47468	0.01101	0.48821	0.01106	0.48538	0.01133	0.44781
ENR	0.01066	0.50039	0.01073	0.49697	0.01056	0.50903	0.01064	0.50481	0.01114	0.45728
KRR	0.00196	0.90795	0.00617	0.71054	0.00179	0.91684	0.00610	0.71586	0.00728	0.64537
BRR	0.01054	0.50607	0.01062	0.50190	0.01042	0.51542	0.01053	0.51009	0.01117	0.45567
ARD	0.01054	0.50609	0.01062	0.50183	0.01042	0.51545	0.01053	0.50999	0.01118	0.45536
RF	0.00142	0.93353	0.00591	0.72277	0.00133	0.93827	0.00587	0.72657	0.00692	0.66267
Ada Boost	0.00960	0.55009	0.00987	0.53697	0.00950	0.55828	0.00983	0.54263	0.01017	0.50440
Gradient Boost	0.00561	0.73716	0.00682	0.68001	0.00532	0.75283	0.00670	0.68811	0.00768	0.62585
XG Boost	0.00136	0.93624	0.00608	0.71474	0.00114	0.94720	0.00604	0.71883	0.00723	0.64757
SVR	0.00493	0.76910	0.00716	0.66437	0.00485	0.77447	0.00717	0.66626	0.00821	0.60003
KNN	0.00000	1.00000	0.00624	0.70733	0.00000	1.00000	0.00622	0.71028	0.00727	0.64595
PLS	0.01171	0.45137	0.01174	0.44938	0.01165	0.45821	0.01170	0.45573	0.01198	0.41626
GPR	0.00001	0.99976	0.00724	0.66042	5.06E-06	0.99976	0.00702	0.67268	0.00805	0.60791