Table 2.

Details about the image sizes at various stages of the proposed model

Layers	Output Size	Details
Input Layer	$224\times 224\times 3$
Convolution Layer	$112\times 112\times 64$	$7\times 7$ conv, stride =2
Max Pooling	$56\times 56\times 64$	$3\times 3$ Max pool, stride =2
Average Pooling	$56\times 56\times 64$	$3\times 3$ Avg pool, stride =2
Hybrid Pooling	$56\times 56\times 64$	[ Both max & avg pooling]
Dense Block 1	$56\times 56\times 256$	[$1\times 1$ conv, $3\times 3$ conv] $\times \mathbf{6}$
Attention Block 1	$56\times 56\times 256$	GAP, Dense layers
Transition Layer 1	$56\times 56\times 128$	$1\times 1$ conv
	$28\times 28\times 128$	$2\times 2$ Avg pool, stride =2
Dense Block 2	$28\times 28\times 512$	[$1\times 1$ conv, $3\times 3$ conv] $\times \mathbf{12}$
Attention Block 2	$28\times 28\times 512$	GAP, Dense layers
Transition Layer 2	$28\times 28\times 256$	$1\times 1$ conv
	$14\times 14\times 256$	$2\times 2$ Avg pool, stride =2
Dense Block 3	$14\times 14\times 1792$	[$1\times 1$ conv, $3\times 3$ conv] $\times \mathbf{48}$
Attention Block 3	$14\times 14\times 1792$	GAP, Dense layers
Transition Layer 3	$14\times 14\times 896$	$1\times 1$ conv
	$7\times 7\times 896$	$2\times 2$ Avg pool, stride =2
Dense Block 4	$7\times 7\times 1920$	[$1\times 1$ conv, $3\times 3$ conv] $\times \mathbf{32}$
Global Average pooling	$1\times 1\times 1920$	$7\times 7$ global average pooling