liam_sun

I will double check that. The experiments were done a few days ago, perhaps not basing on the recent version.

I just tried an asymmetric per-channel QAT model. It turns out that some ops really have different scale/zero point. For instance, the add ops you illustrated. But the inference results...

Yes, I also found -128 in some weights. I notice that you really set quant_min=-127 and quant_max=127 in quantizer.py. My understanding is it's to avoid the risk of overflow. So...

I'm on the way. Meanwhile, what's the meaning of 'bisecting', could you please explain it in more detail?

My experiment shows no -128 weights anymore for asymmetric per-channel case. Not sure if this is what you said "work out with that patch". Similar to former experiments, asymmetric per-channel...

OK. Just confirm 'bisecting' should be done against the original model or the quanatized (handled by QATQuantizer.quantize ) model ?

Before conducting the experiments, I have a few more questions since I am still not very clear about the point you mentioned > Suppose you have the following model description...

Sure, I will get that done.

Same question. Looks that the author does not maintain this project any more. :)

I met the same issue. Any solution or workaround for it?