Quantizes Weights and activations for convolutions. Refer to quantizable_layer.ToAqtConv. Args: w_name: Previously created w_name QWeight to quantize weight. act: The activation tensor to quantize. weight: The weight tensor to quantizes. w_feature_axis: axis corresp
(self,
w_name,
act,
weight,
w_feature_axis,
act_distribution=QDistribution.SYMMETRIC,
w_expected_scale_shape=None)
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected