Generative Adversarial Networks Projects

Losses for Stage-I of StackGAN

There are two losses used in Stage-I of StackGAN as follows:

• Generator loss
• Discriminator loss

The discriminator loss  can be represented as follows:

                                                   

The preceding equation is pretty self-explanatory. It represents the loss function for the discriminator network, in which both networks are conditioned on the text embeddings. 

The generator loss  can be represented ...