Hi @sanghoon
Thank you for your work and share.
I tried your script for ResNet50 to combine conv, batchnorm and scale layer into one conv layer. I have doubts regarding generated prototxt and caffemodel
generated prototxt combines only scale and batchnorm into one and keep conv layers as it is
before : conv - BN- scale - relu
after : conv - BN - scale - relu
generated caffemodel weight should also change as now number of parameters are less to be stored?
Hi @sanghoon Thank you for your work and share. I tried your script for ResNet50 to combine conv, batchnorm and scale layer into one conv layer. I have doubts regarding generated prototxt and caffemodel
generated prototxt combines only scale and batchnorm into one and keep conv layers as it is before : conv - BN- scale - relu after : conv - BN - scale - relu
generated caffemodel weight should also change as now number of parameters are less to be stored?