You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is due to the padding.
In the paper, we explained this.
Notably, for a K×K conv that zero-pads the input, Eq. 8
does not hold because F (2) does not convolve on the result
of I ~ F (1) + REP(b(1)) (but an additional circle of zero
pixels). The solution is to either A) configure the first conv
with padding and the second without, or B) pad by b(1). An
efficient implementation of the latter is customizing the first
BN to 1) batch-normalize the input as usual, 2) calculate
b(1) (Eq. 6), 3) pad the batch-normalized result with b(1),
i.e., pad every channel j with a circle of b(1) j instead of 0.
In this repo, BNAndPadLayer is designed to solve this problem.
Hi,
I verified like this:
And the output is 0.11, which is much too great. Have you noticed this ?
The text was updated successfully, but these errors were encountered: