Table 3.

Summary of neural network architectures used in model building.

Network family	Network architecture	Source package	Reference	Layers	Trainable weights (${10}^6$)
Residual learning	resnet18	torchvision	49	18	11
Squeeze and excitation	se_resnext50_32x4d	pretrainedmodels	50,51	50	25
Squeeze and excitation	se_resnext101_32x4d	pretrainedmodels	50,51	101	47
Inception	inceptionv4	pretrainedmodels	52	150	41
Inception	inceptionresnetv2	pretrainedmodels	52	245	54
Efficient-net	efficientnet-b3	efficientnet-pytorch	53	27	11
Efficient-net	efficientnet-b4	efficientnet-pytorch	53	33	17
Efficient-net	efficientnet-b5	efficientnet-pytorch	53	39	28