I am using the decent tool for quantisation of Resnet model to 8 bit int.
The ouput is generated as deploy.prototxt and deploy.caffemodel.
The problem is that when I read weights in deploy.caffemodel, they are still float 32 bit. What does this means? I supposed the weights to be int 8 bit.
Also, when I see the size of input caffemodel(float.caffemodel) and output caffemodel(deploy.caffemodel), both are exactly of same size i.e. 102.1 MB.
@anz162112Can you tell us what board you are running this on, as well as advise if you are using the latest stretch image and latest version of DNNDK?
Does everything else appear to be as expected?
We are running this tutorial in v2.08 and everything is working as expected: https://github.com/jimheaton/Ultra96_ML_Embedded_Workshop