Open fferroni opened 7 years ago
Yeah I think same. I think it should be
y = np.concatenate(np.ones(10,1,64,64), np.zeros(10,1,64,64))
Furthermore, what is the reason of input the y in line 194. Since in the loss function we do not take into account y but we use the labels provided from the loss function itself. Isn't pointless? Secondly, why the y is a vector of size 20x1x64x64? What is the reason for the 64?
https://github.com/r0nn13/conditional-dcgan-keras/blob/master/conditional_gan.py#L193
This creates an array of all zeros. Surely half should be zero and the other half ones?