dmx.compressor.advanced_recipe.DmxQuantizerCalibrationHyperparams

class dmx.compressor.advanced_recipe.DmxQuantizerCalibrationHyperparams(observer_cls: ~torch.ao.quantization.observer.ObserverBase = <class 'dmx.compressor.numerical.observer.HistogramObserver'>, qscheme_to_overload: ~torch.qscheme = torch.per_tensor_symmetric, group_size: int | None = None, ch_axis: int | None = None)

Fake quantizer (i.e. CastTo) calibration hyperparameters with default values

__init__(observer_cls: ~torch.ao.quantization.observer.ObserverBase = <class 'dmx.compressor.numerical.observer.HistogramObserver'>, qscheme_to_overload: ~torch.qscheme = torch.per_tensor_symmetric, group_size: int | None = None, ch_axis: int | None = None) → None

Methods

__init__([observer_cls, ...])

Attributes

`ch_axis`
`group_size`
`qscheme_to_overload`