torch.ao.quantization.observer.default_observer

torch.ao.quantization.observer.default_observer = functools.partial(<class 'torch.ao.quantization.observer.MinMaxObserver'>, quant_min=0, quant_max=127){}

Default observer for static quantization, usually used for debugging.

上一页

torch.ao.quantization.observer.default_float_qparams_observer_4bit

下一页

torch.ao.quantization.observer.default_per_channel_weight_observer