Concerns on non-linear function

[HaoKun-Li]

Hello, thanks for your excellent paper and code!
I have some concerns on the non-linear functions that are used in the ResNet and Mobilenet. Could you please provide more details?
In the Section 4 of the paper, the authors claim that "they use RPReLU [32] as non-linear function":

However, in this code, it seems that you use PReLU for Resnet and ReLU6 for MobileNet. May this difference affect the accuracy of quantized models seriously?

[HaoKun-Li]

Hello, thanks for your excellent paper and code! I have some concerns on the non-linear functions that are used in the ResNet and Mobilenet. Could you please provide more details? In the Section 4 of the paper, the authors claim that "they use RPReLU [32] as non-linear function":

However, in this code, it seems that you use PReLU for Resnet and ReLU6 for MobileNet. May this difference affect the accuracy of quantized models seriously?

Ok, I got it. The authors use two additional LearnableBias parameters to implement the RPReLU as follows: