train_disc.py

#To train the Discriminator
output_d_real = discriminator(real_images)
d_real_loss = criterion(output_d_real, real_labels)

z = torch.randn(batch_size, random_size).to(device)
fake_images = generator(z)
output_d_fake = discriminator(fake_images)
d_fake_loss = criterion(output_d_fake, fake_labels)
d_loss = d_real_loss + d_fake_loss

optimizer_d.zero_grad()
d_loss.backward()
optimizer_d.step()#this is going to update only parameters of discriminator