why you use sampled_label for the combined model instead of labels and didn't set "self.discriminator.trainable = True" for training the discriminator?

[GussailRaat]

Hello Erik, (1) I just want to understand why you use sampled_label for the combined model instead of labels? g_loss = self.combined.train_on_batch([noise, sampled_labels], valid).

(2) you set "self.discriminator.trainable = False" to train the generator but for training the discriminator why did you not set "self.discriminator.trainable = Ture"?

Thank you.

[TheGuywithTheHat]

regarding (2), see: https://github.com/eriklindernoren/Keras-GAN/issues/22#issuecomment-379553290

Essentially, it is important that trainable = false is set between discriminator.compile() and combined.compile().

[hiteshnitetc]

For which gan you are talking about.