Closed twmht closed 6 years ago
According to the paper,
Finally, we append a 3×3 convolution on each merged map togenerate the final feature map, which is to reduce the alias-ing effect of upsampling.
But in the network you defined, there is no such layer.
Why?
in my code, it is 1*1
According to the paper,
But in the network you defined, there is no such layer.
Why?