GAN is one of my main interests and research cases.
The main purpose of the repository is to implement the models I have read on the papers for deeper understanding and more practice on GAN and famous frameworks like Pytorch or Tensorflow. But of course, model architectures will not exactly the same as proposed in the papers.

• ThisAnimeCharacterDoesntExist
• ThisPersonDoesntExist
• ThisCatDoesntExist
• NumberGenerator
• MonetToPhoto

Deep Convolutional Generative Adversarial Network

DCGAN implementation on anime-faces dataset. I've implemented a model that generates 64 * 64 anime faces.
Some examples that are generated from this model:

[Paper] [Code] [Html_Code]

Deep Convolutional Generative Adversarial Network

DCGAN implementation on CelebA dataset. I've implemented a model that generates 64 * 64 faces.
Some examples that are generated from this model:

[Paper] [Code] [Html_Code]

Deep Convolutional Generative Adversarial Network

DCGAN implementation on cats dataset. I've implemented a model that generates 64 * 64 images of cats.

[Paper] [Code] [Html_Code]

Deep Convolutional Generative Adversarial Network

DCGAN implementation on mnist dataset. I've implemented a model that generates 32 * 32 images of numbers.

[Paper] [Code] [Html_Code]

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

CycleGAN implementation on monet painting dataset. In this project we have to model that the first one convert monet painting to real images and the other one, convert images into monet painting space.

[Paper] [Code] [Html_Code]