Hi,
I am quite new in caffe and neuronal networks.
I saw that for example in the imagenet example of caffe you have to specify the mean file trhough a binaryproto file.
I want to implement a LeNet network but I am not sure if it is needed in that case.
When is needed to use the binaryproto mean file and why?
Thanks in advance!
Fran