mafda / generative_adversarial_networks_101

Keras implementations of Generative Adversarial Networks. GANs, DCGAN, CGAN, CCGAN, WGAN and LSGAN models with MNIST and CIFAR-10 datasets.

mafda/generative_adversarial_networks_101 Issues

why we always use img_real to train generator weights in ccGAN？
Closed 6 months ago1
why we use discriminator.trainable = False before d_g.train_on_batch(x=z, y=real) in 01_GAN_MNIST_original?
Closed 4 years ago2
DCGAN and CGAN suffer from mode collapse
Closed 5 years ago1
why random labels are used for training of d_g?
Closed 5 years ago2
LSGAN loss function
Closed 5 years ago1