. 2021 Apr 28;7(5):81. doi: 10.3390/jimaging7050081

Table 2.

Data dimensions in CORONA-Net (the decoder network).

Layers	Kernel Size	Stride	Padding	Output Channel
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 2 × 1	S = 1 × 1	P = 0	O = 6
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 2 × 2	S = 2 × 1	P = 0	O = 8
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 2 × 2	S = 2 × 2	P = 0	O = 10
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 1 × 1	P = 0	O = 14
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 4 × 4	S =4 × 4	P = 0	O = 16
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 3 × 3	P = 0	O = 18
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 2 × 2	P = 0	O = 22
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 1 × 1	P = 4	O = 24
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 1 × 1	P = 4	O = 32
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 3 × 3	S = 1 × 1	P = 4	O = 3
$C o n v o l u t i o n T r a n s p o s e$ $(B a t c h N o r m + T a n h)$	K = 4 × 4	S = 1 × 1	P = 3	O = 3
$O u t p u t$	Tensor (50, 3, 224, 224)	-	-	-