How to Train StyleGAN to Generate Realistic Faces

Why StyleGAN outperforms other models for face generation and how to train your own StyleGAN

Kayo Yin
Towards Data Science
8 min readNov 18, 2019

--

Sample output during my training of StyleGAN

Generative Adversarial Networks (GAN) is an architecture introduced by Ian Goodfellow and his colleagues in 2014 for generative modeling, which is using a model to generate new samples that imitate an existing dataset. It is…

--

--