Skip to main content

View full-text article in PMC

. 2022 Nov 21;22(22):8997. doi: 10.3390/s22228997

Table 4.

Comparison of the prediction accuracy among different models.

Models	Calibration			Validation
Models	RMSEc	R²c	MAEc	RMSEv	R²v	MAEv
Ranger	1.669	0.834	1.335	*2.493*	*0.362*	*2.043*
Cubist	*1.88*	*0.693*	*1.518*	2.283	0.445	1.877
GBM	2.551	0.43	2.005	2.522	0.336	2.098
BRNN	2.712	0.323	2.121	2.589	0.282	2.058
RK-ranger	1.553	0.867	1.194	1.944	*0.718*	*1.604*
RK-Cubist	*1.646*	*0.794*	*1.256*	1.741	0.674	1.474
RK-GBM	2.050	0.734	1.556	1.906	0.724	1.608
RK-BRNN	2.369	0.569	1.797	2.066	0.625	1.632

Notes: the model with the highest accuracy is marked in bold; the model with the second highest accuracy is marked in bold italics.