default_per_channel_weight_observer

torch.quantization.observer.default_per_channel_weight_observer

Default per-channel weight observer, usually used on backends where per-channel weight quantization is supported, such as fbgemm.

alias of torch.ao.quantization.observer.functools.partial(<class 'torch.ao.quantization.observer.PerChannelMinMaxObserver'>, dtype=torch.qint8, qscheme=torch.per_channel_symmetric){}