Deep Learning with PyTorch

Training the complete network

We looked at individual pieces of how a GAN is trained. Let's summarize them as follows and look at the complete code that will be used to train the GAN network we created:

• Train the discriminator network with real images
• Train the discriminator network with fake images
• Optimize the discriminator
• Train the generator based on the discriminator feedback
• Optimize the generator network alone

We will use the following code to train the network:

for epoch in range(niter):    for i, data in enumerate(dataloader, 0):        ############################        # (1) Update D network: maximize log(D(x)) + log(1 - D(G(z)))        ###########################        # train with real        netD.zero_grad()        real, _ = data        batch_size = real.size(0) if torch.cuda.is_available(): ...