NOT an issue: It works amazingly well. However, use Spectral Normalization will converge VERY fast and nice !

[thusinh1969]

Your codes works amazingly well. However, use Spectral Normalization will converge VERY fast and nice !

That is all. I changed 2 things: (1) Added Spectral Normalization to Conv2D layers of Discriminator network (except the first Conv2D layer) and (2) Replace VGG19 by VGGface for my problem of scale-up faces those are not very clear for a better face recognition/triplet loss.

Thanks mate.
Steve

nguyen@hatto.com

[weizhiliang0520]

Your codes works amazingly well. However, use Spectral Normalization will converge VERY fast and nice !

That is all. I changed 2 things: (1) Added Spectral Normalization to Conv2D layers of Discriminator network (except the first Conv2D layer) and (2) Replace VGG19 by VGGface for my problem of scale-up faces those are not very clear for a better face recognition/triplet loss.

Thanks mate. Steve

nguyen@hatto.com

Hello, could you share your code? I would like to try to change the part of the module for example try to add the SN module you mentioned. Thanks!