Thanks for your reply!
I'm still confused about some detail in calculating the crop_param. It seems the crop_offset has nothing to do with the exact spatial size of input?
Besides, i feel it's a little inconvenient to use the python interface to determine the network. For example,
(a)How to combine train.prototxt and val.prototxt into a single trainval.prototxt?
(b)What if a layer name has a dash mark? it becomes a minus sign in python.
(c)How to set mean_value for each channel in Data layer?
Currently, I need to use the python interface to get the crop_param, and then 'refine the prototxt' by myself, it's a little cumbersome.
在 2016年4月26日星期二 UTC+8上午1:46:25,Evan Shelhamer写道: