ganhacks
ganhacks copied to clipboard
how does trick 6 change ideal output activation/loss
Regarding Trick 6:
What activation do you recommend for the discriminator's real_or_generated
output layer? 'm using a sigmoid activation function but since the output can be as large as 1.2 I'm wondering if something like leaky relu would be better since sigmoid activation fcn is [0, 1.0]. And would you still recommend using binary_crossentropy
as loss for this output or something else now that using soft labels?
Or you can keep sigmoid, and do random uniform sampling in interval [0.7, 1.0] for real labels.