default_weight_fake_quant

torch.ao.quantization.fake_quantize.default_weight_fake_quant

Default fake_quant for weights. Observer is memoryless since averaging_constant is 1.