Table 2.

Parameters of the models used in the evaluation.

Parameters	Networks
	Inception-v3	Capsule network
Depth	48	8
Input size	128 × 128	128 × 128
Optimizer	RMSProb	Adam
Loss function	Cross-entropy	MSE
Batch size	64	20
Learning rate	0.0001	0.007
Momentum	0.9	0.8
Gradient	L2 norm	L2 norm

Adam Adaptive momentum estimation, MSE mean squared error.