Table 1.

ResNet50 structure.

Layer Type	Layer Setting	Output Shape	Trainable Parameters
Block_conv1	7 × 7, 64, stride 2	(112, 112, 64)	9508
Block_pool1	3 × 3, max pool stride 2	(56, 56, 64)	0
Block_conv2	$\left[\begin{array}{c}\begin{array}{c}1\times 1, 64\\ 3\times 3, 64\\ 1\times 1, 256\end{array}\end{array}\right]\times 3$	(56, 56, 128)	248, 832
Block_conv3	$\left[\begin{array}{c}\begin{array}{c}1\times 1, 128\\ 3\times 3, 128\\ 1\times 1, 512\end{array}\end{array}\right]\times 4$	(28, 28, 256)	1, 318, 144
Block_conv4	$\left[\begin{array}{c}\begin{array}{c}1\times 1, 128\\ 3\times 3, 128\\ 1\times 1, 512\end{array}\end{array}\right]\times 6$	(14, 14, 512)	7, 492, 096
Block_conv5	$\left[\begin{array}{c}\begin{array}{c}1\times 1, 128\\ 3\times 3, 128\\ 1\times 1, 512\end{array}\end{array}\right]\times 3$	(7, 7, 2048)	14, 439, 424
Block_pool2	average pool	(1, 1, 2048)	0
Drop out	0.5	(1, 1, 2048)	0