Table 1. Hyperparameter tuning.

Parameter	Values tested	Optimal value
Learning rate ( $\eta$)	0.01, 0.001, 0.0005, 0.0001	0.001
Batch size	8, 16, 32	16
Activation function	ReLU, Leaky ReLU, Tanh	Leaky ReLU
Patch size	16 $\times$ 16, 32 $\times$ 32, 64 $\times$ 64	32 $\times$ 32
Dropout rate	0.1, 0.2, 0.3	0.1