search

shekkizh / WassersteinGAN.tensorflow

Tensorflow implementation of Wasserstein GAN - arxiv: https://arxiv.org/abs/1701.07875
MIT License
416 stars 130 forks source link

WGAN loss is right? #9

Closed kikyou123 closed 7 years ago

kikyou123 commented 7 years ago

I think your WGAN loss if not right, it should be: self.discriminator_loss = tf.reduce_mean(logits_fake - logits_real) self.gen_loss = tf.reduce_mean(-logits_fake) ans your WGAN discriminator output shape is [batch_size, channel=1, height=2, width=2],it is a mistake?

shekkizh commented 7 years ago

The discriminator output is of shape [batch_size, 1]. The last layer of discriminator is a fc layer. The loss term has been previously discussed in issues #6 and #7

Cheers.