qkeras
qkeras copied to clipboard
How can I get the scale of the QAdaptiveActivation layer
As shown in the comment, the QAdaptiveActivation layer can be used to calculate the EMA of min and max of the activation values, i.e., the quantization range of the activations tensor.
After we have the quantization range and the number of quantization levels (e.g., 256 for INT8), we can get the scale.
However, I did not find the scale attribute in the QAdaptiveActivation layer.
So how can I get the scale of the QAdaptiveActivation layer?