QATModule

class QATModule(**kwargs)

Base class of quantized-float related Module, basically for QAT and Calibration.

Use from_float_module to generate a instance from float Module. Or use quantize_qat to do it recursively and automatically.

Can also be converted to QuantizedModule for deployment using quantize further.

apply_quant_activation(target)

Apply weight’s observer and fake_quant from qconfig on target.

apply_quant_bias(target, inp, w_qat)

Use fake_quant_bias to process target. Only valid when act_fake_quant and weight_fake_quant are both enabled.

apply_quant_weight(target)

Apply weight’s observer and fake_quant from qconfig on target.

abstract classmethod from_float_module(float_module)

Return a QATModule instance converted from a float Module instance.

get_activation_dtype()

Get activation’s quantization dtype as the method from qconfig.

get_activation_qparams()

Get activation’s quantization parameters.

get_weight_dtype()

Get weight’s quantization dtype as the method from qconfig.

get_weight_qparams()

Get weight’s quantization parameters.

set_qconfig(qconfig)

Set quantization related configs with qconfig, including observer and fake_quant for weight and activation.