A PyTorch implementation of RelGAN: Multi-Domain Image-to-Image Translation via Relative Attributes
The paper is accepted to ICCV 2019. We also have the Keras version here.
- Python 3.6 or higher
- PyTorch 0.4.0 or higher
- All the dependencies
pip3 install -r requirements.
tensorboard --logdir runs
CUDA_VISIBLE_DEVICES=0 python3 train.py --data <PATH_TO_CELEBA-HQ> --gpu [--image_size 256]
CUDA_VISIBLE_DEVICES=0,1,2,3 python3 train.py --data <PATH_TO_CELEBA-HQ> --multi_gpu [--image_size 256]
CUDA_VISIBLE_DEVICES=0 python3 train.py --config config.yaml