Simple attempt to implement https://github.com/lucidrains/lightweight-gan in a docker image. Inspired by https://www.docker.com/blog/containerized-python-development-part-1/ and https://github.com/jeffheaton/docker-stylegan2-ada/blob/main/Dockerfile.
Requirements:
- nvidia-docker https://github.com/NVIDIA/nvidia-docker
- Data somewhere on your computer in a folder to use somewhere for training.
Usage:
- Clone the repo
- docker build . -t lightgan
- sudo nvidia-docker run -it -v $path_to_images:/mnt --ipc=host lightgan:latest /bin/bash
- lightweight_gan --data /mnt --aug-prob 0.25 --aug-types [translation,cutout,color] --num-train-steps 5000
- exit
- sudo docker ps -a
- sudo docker cp $container_name:/results .