Quantize network stops at "Network initialization done." step

Hello,

I trained and i try to quantize the network but it stops at “Network initialization done.” step.

The commands is :
python2 nn_quantizer.py --gpu --model models/lettre_O_V1/lettre_O_train_test.prototxt --weights models/lettre_O_V1/lettre_O_iter_1300.caffemodel --save models/lettre_O_V1/lettre_O.pkl

I join the log.txt of the command and lettre_O_train_test.prototxt network achitecture.
lettre_O_iter_1300.caffemodel trained network is too big (40.2Mo).

Help would be great,
Thanks
files.zip (3.19 KB)

Hi, that script isn’t robust if the network is bad when you trained it. The error is happening before the conversion. What accuracy did you get on training the net and testing it? If they were bogus you cannot convert the network.