A bug may have been resolved.

[haiduo]

There is a bug here, I thought of a simple way to fix it, which is applicable to QAT of ViT.

elif 'input_quantizer.scale' in dict(_module.state_dict()).keys():
      _module.input_quantizer.set_fake_fused()  # 有bug, quant_state会来回切.
else:
     print("no_set_fake_fused:", _user.name, _module.input_quantizer_generated)

[PeiqinSun]

I think this issue is from your misunderstanding. Please check your code to verify it.

[haiduo]

Thanks for your Reply. First of all, I found the error by Vit_QAT training, and reporting: 'MatMul' object has no attribute 'input_quantizer', so I did this, to exclude this error by whether the operator has a scale. Otherwise, cannot continue model.init_QAT().

[KenRamzes]

Hi, did you successfully reproduce the result of ViT_QAT?