Today, we’ll be learning how to train and fine-tune CycleGAN. Here, it’s in action, transforming a photograph into a painting.
Source: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
CycleGAN Installation
- Clone the repository:
git clone https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix
CycleGAN Train
- Download a dataset; we’ll be using the datasets linked in references. But feel free to use whatever suits your needs.
- Arrange the training dataset,
pytorch-CycleGAN-and-pix2pix: