using quantization is slower

[liangzimei]

hello, i run the imagenet classification demo, i find it is slower when setting quantize=ture, is it normal?
using quantization
not using quantization

[mathmanu]

What we are doing is simulating quantization and measuring how much error would occur if you this kind of fixed point result. Thus you know what degradation to expect when you port your CNN algorithm to an embedded device with fixed point quantization. This is nit a fast fixed point implementation.

[liangzimei]

@mathmanu okay, thanks for your reply..