Improved GANs—Conditional GAN
Learn about an improved version of GANs, the conditional GAN, and how it can be implemented.
We'll cover the following
GANs are powerful systems that can generate realistic samples from the domain of training. Architectures like the vanilla GAN and DCGAN have also been used to generate samples that resemble human faces and even real-world items (from training on CIFAR-10 and so on). However, they miss out on the ability to control the samples we would like to generate.
Architecture
In simple words, we can use a trained generator to generate any number of samples required, yet we cannot control it to generate a specific type of example. Conditional
In their work, conditional adversarial networks, Mirza et al. pointed toward using class labels as additional conditioning input to both generator and discriminator models. We denote the conditioning input as
Get hands-on with 1400+ tech skills courses.