TABLE 1.

ResNet-18’s input and output for each layer.

Layer name	Operation	Input	Output
Conv1	7 × 7, 64, stride2	512 × 512 × 1	256 × 265 × 64
Max pool	3 × 3, max_pooling, stride2	256 × 265 × 64	128 × 128 × 64
Conv2_x	$\left[\begin{array}{cc}3\times 3& 64\\ 3\times 3& 64\end{array}\right]\times 2$	128 × 128 × 64	128 × 128 × 64
Conv3_x	$\left[\begin{array}{cc}3\times 3& 128\\ 3\times 3& 128\end{array}\right]\times 2$	128 × 128 × 64	64 × 64 × 128
Conv4_x	$\left[\begin{array}{cc}3\times 3& 256\\ 3\times 3& 256\end{array}\right]\times 2$	64 × 64 × 128	32 × 32 × 256
Conv5_x	$\left[\begin{array}{cc}3\times 3& 512\\ 3\times 3& 512\end{array}\right]\times 2$	32 × 32 × 256	16 × 16 × 512
Average pool	avg_pooling	16 × 16 × 512	2 × 2 × 512
FC	1,000–d fc + softmax	2 × 2 × 512	1 × 2