DCGAN is initialized with random weights, so a random code plugged into the network would create a totally random impression. However, as you might imagine, the network has a lot of parameters that we can easily tweak, and also the aim is to find a environment of those parameters https://angelonnlkg.onzeblog.com/33952084/how-much-you-need-to-expect-you-ll-pay-for-a-good-neuralspot-features