Journal of Wood Science

Official Journal of the Japan Wood Research Society

Table 3 Performance of random forest models in predicting carbon content, O/C, and H/C predictions

From: Feature importance measures from random forest regressor using near-infrared spectra for predicting carbonization characteristics of kraft lignin-derived hydrochar

Output variable	RF parameters		Training set		Test set
Output variable	n_feature	n_tree	R²	RMSE	R²	RMSE
C (wt%)	‘log2’ = 7	43	0.997	0.083	0.989	0.254
O/C	‘sqrt’ = 13	43	0.996	0.001	0.988	0.003
H/C	‘sqrt’ = 13	16	0.998	0.002	0.985	0.008

n_feature, number of features; n_tree, number of decision trees; all, all features (input variables); sqrt, square root of n_feature; R², coefficient of determination; RMSE, root mean square error

Back to article page