default_fake_quant

torch.quantization.fake_quantize.default_fake_quant

alias of functools.partial(<class ‘torch.ao.quantization.fake_quantize.FakeQuantize’>, observer=<class ‘torch.ao.quantization.observer.MovingAverageMinMaxObserver’>, quant_min=0, quant_max=255, dtype=torch.quint8, qscheme=torch.per_tensor_affine, reduce_range=True){}