Table 1.

EfficientNet-based model, the stage i with ${\widehat{L}}_i$ layers, with input resolution $〈{\widehat{H}}_I, {\widehat{W}}_i〉$ and output channels ${\widehat{C}}_i$.

stage $\mathit{i}$	Operator ${\widehat{\mathit{F}}}_{\mathit{i}}$.	Resolution ${\widehat{\mathit{H}}}_{\mathit{I}}\times {\widehat{\mathit{W}}}_{\mathit{i}}$	#Channels ${\widehat{\mathit{C}}}_{\mathit{i}}$	#Layers ${\widehat{\mathit{L}}}_{\mathit{i}}$
1	Conv3$\times$3	224$\times$192	32	1
2	MBConv1, k3$\times$3	112$\times$96	16	1
3	MBConv6, k3$\times$3	112$\times$96	24	2
4	MBConv6, k5$\times$5	56$\times$48	40	2
5	MBConv6, k3$\times$3	28$\times$24	80	3
6	MBConv6, k5$\times$5	28$\times$24	112	3
7	MBConv6, k5$\times$5	14$\times$12	192	4
8	MBConv6, k3$\times$3	7$\times$6	320	1
9	Conv1$\times$1 & Flatten & FC	7$\times$6	1280	1