torch.ao.quantization.observer._ObserverBase#

class torch.ao.quantization.observer._ObserverBase(dtype=torch.quint8, qscheme=torch.per_tensor_affine, reduce_range=False, quant_min=None, quant_max=None, factory_kwargs=None)[源代码]#

Internal common base for all qint/quint8 observers.

This base is for commonly used parameters used internally. Users should use ~torch.ao.quantization.observer.ObserverBase as a base class for custom observers.

参数

  • dtype – Quantized data type.

  • qscheme – Quantization scheme to be used.

  • reduce_range – Reduces the range of the quantized data type by 1 bit. This is sometimes required to avoid instruction overflow.

  • quant_min – Minimum quantization value. If unspecified, it will follow the 8-bit setup.

  • quant_max – Maximum quantization value. If unspecified, it will follow the 8-bit setup.

警告

dtype can only take torch.qint8 or torch.quint8.

警告

qscheme can only take one of the following options:

  • torch.per_tensor_affine

  • torch.per_tensor_symmetric

  • torch.per_channel_affine

  • torch.per_channel_symmetric

上一页

torch.ao.quantization.observer.ObserverBase

下一页

torch.ao.quantization.observer.MinMaxObserver