Prediction of Soil Moisture-Holding Capacity with Support Vector Machines in Dry Subhumid Tropics

Table 6

Performance indicators for SVR models with different predictors.

Model	Inputs	ME (cm^3·cm^−3)	RMSE (cm^3·cm^−3)	R^2
FC1	Sand + clay	5.53 × 10^−3	0.042	0.562
FC2	Clay + sand + BD	2.61 × 10^−3	0.038	0.643
FC3	Clay + sand + BD + OC	5.00 × 10^−4	0.037	0.663
FC4	Sand + BD	2.62 × 10^−3	0.038	0.645
WP1	Sand + clay	6.96 × 10^−3	0.045	0.546
WP2	Clay + sand + BD	3.74 × 10^−3	0.037	0.668
WP3	Clay + sand + BD + OC	2.99 × 10^−3	0.037	0.677
WP4	Sand + BD + OC	2.47 × 10^−3	0.037	0.679