Table 3.

Training Options and Hyperparameters.

Options and Hyper-parameters	Four Classes		Seven Classes
	ResNet18	ResNet50	ResNet18	ResNet50
Initial Learning Rate	0.0005	0.0001	0.0001	0.0001
Learning Rate's Drop Factor	0.05	0.05	0.05	0.05
Learning Rate's Drop Period (Epochs)	2	3	3	3
Batch Size	128	64	128	64
L2 – Regularisation	0.4	0.75	0.3	0.3
Epochs	15	12	15	12
Environment	Multi-GPU	Multi-GPU	Multi-GPU	Multi-GPU
Optimiser	Adam	Adam	Adam	Adam
Loss Function	Cross-Entropy	Cross-Entropy	Cross-Entropy	Cross-Entropy