TABLE 4.

RMSE of internal and external testing set compared with ground-truth.

	${\mathit{I}}_{\mathit{k}\mathit{h}}$	${\mathit{I}}_{\mathit{k}\mathit{h}-\mathit{H}\mathit{S}}$	${\mathit{I}}_{\mathit{k}\mathit{h}-\mathit{C}\mathit{L}\mathit{A}\mathit{H}\mathit{E}}$	$I_{\mathit{k}\mathit{h}-\mathit{H}\mathit{S}-\mathit{C}\mathit{L}\mathit{A}\mathit{H}\mathit{E}}$	${\mathit{I}}_{\mathit{k}\mathit{5}\mathit{m}}$	${\mathit{I}}_{\mathit{k}\mathit{5}\mathit{m}-\mathit{C}\mathit{L}\mathit{A}\mathit{H}\mathit{E}}$
RMSE	0.12	0.12	0.13	0.11	0.10	0.09

The bold value here means the smallest RMSE and the best performance.