A Network Data Reinforcement Method Based on the Multiclass Variational Autoencoder

Table 5

The metrics of ADMs in the end of epoch 5.

Dataset	Model	Accuracy (%)	Loss	Val_Accuracy (%)	Val_Loss
Raw data	ADM (MLP)	96.83	0.0700	99.67	1.1841e-04
	ADM (CNN)	99.67	0.0310	99.67	0.0203
Reinforced data	ADM (MLP)	98.98	0.0327	97.16	0.7763
	ADM (CNN)	99.99	0.0010	95.38	0.9995