Innovative deep learning artificial intelligence applications for predicting relationships between individual tree height and diameter at breast height

Forest Ecosystems

Table 8 The goodness-of-fit statistics of DLA in validation data set

The number of neuron alternatives and regression models	The number of hidden layers	RMSE	RMSE% (%)	AIC	BIC	FI	AAE	Max. AE	Bias	Bias% (%)
50 # neuron	3	0.8314	7.6684	−50.1267	277.8514	0.8907	0.6544	3.7280	−0.0905	−0.8351
90 # neuron	4	0.8426	7.7715	−46.0660	281.9122	0.8878	0.6546	3.3615	0.0377	0.3475
90 # neuron	5	0.7787	7.1825	−70.0252	257.9530	0.9042	0.6320	2.7641	−0.0126	−0.1162
100 # neuron	6	0.8631	7.9606	−38.7586	289.2195	0.8823	0.6878	3.1312	−0.0114	−0.1048
70 # neuron	7	0.7800	7.1940	−69.5381	258.4401	0.9038	0.6219	2.8616	−0.0406	−0.3740
90 # neuron	8	0.8985	8.2869	−26.5451	301.4330	0.8724	0.6741	5.5280	0.0256	0.2360
100 # neuron	9	0.8291	7.6466	−50.9918	276.9863	0.8914	0.6501	2.9782	−0.0481	−0.4433
80 # neuron	10	0.8613	7.9438	−39.4010	288.5772	0.8828	0.6808	2.9433	−0.0209	−0.1926