Densely Connected Convolutional Networks, In CVPR 2017 (Best Paper Award).
Geek Repo:Geek Repo
Github PK Tool:Github PK Tool
xmyqsh opened this issue 7 years ago · comments
I got it. The pool of torch use floor((imsize + 2*pad - kernel) / stride) + 1
floor((imsize + 2*pad - kernel) / stride) + 1