Table 6. Complexities of the CNN models with transfer learning.

Model	Learnable parameters (millions)	Prediction time per image (ms)		Training time per epoch (s)
		Original dataset	Augmented dataset	Original dataset	Augmented Dataset
AlexNet	44.56	0.2894	0.2451	21	57
GoogleNet	6.04	0.4335	0.3796	27	90
Inceptionv3	21.91	0.9965	0.9386	115	432
ResNet18	11.21	0.3067	0.2417	22	80
VGG16	113.57	1.6758	1.6971	109	541
Xception	20.94	2.2419	1.8831	142	815