onwn / C2N

MIT License
36 stars 3 forks source link

Training loss #7

Open FrauSong opened 2 years ago

FrauSong commented 2 years ago

Hi,

Thanks for sharing the code of this excellent paper. I am a little bit confused by Eq 4. Should this loss be backpropagated only to the discriminator? The second term of this equation differs from the original WGAN-GP. Could you be so kind as to give some hints on how to understand it?

Many thanks!