In Generative Adversarial Networks, the Generator takes noise vector as input and feeds it forward to create an image. The noise vector consists of random numbers sampled from the normal distribution. In several examples that I've encountered, the noise vector had 100 numbers (implementation 1, implementation 2). Is there a reason this number is used? How does noise size affect the generation image?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.