I’m trying to figure out the training process of a conditional GAN.

For example, consider a dataset like MNIST. I give the conditional vector to produce only the number 7 for both the generator and discriminator. In the following scenarios, the discriminator will classify which one is fake and which one is real:

  1. The generator produces realistic numbers other than 7, such as a realistic number 9 ?
  2. The samples from the MNIST dataset that are not the number 7 (i.e., other numbers) ?

Thanks for your help!