DCGAN is initialized with random weights, so a random code plugged in to the network would produce a very random graphic. On the other hand, while you might imagine, the network has millions of parameters that we can easily tweak, plus the aim is to find a location of those parameters that makes samples created from random codes appear like the edu