Khrystyna Faryna

icon company RadboudUMC icon location Nijmegen, Netherlands icon location PhD student at DIAG and Computational Pathology groups.

Results 3 issues of Khrystyna Faryna

CycleGAN discriminator activation fucnction

1 comment

I noticed that the discriminator does not have activation function in the output. The last layer is just a convolution. Why is that?

Interpreting Generator and Critic loss

1 comment

Dear @martinarjovsky, I am currently working on a project with MRI data. I was using WGAN -GP loss on 2D implementation, with hyperparameters proposed in WGAN-GP paper - everything worked...

Batch norm in WGAN discriminator

1 comment

Hi, I was wondering why did you guys decide to use batch norm in a critic of WGAN-GP. The paper on Improved training of WGAN(the one where gradient penalty is...