WGAN

Code is taken from https://github.com/martinarjovsky/WassersteinGAN and modified for our needs

if you run it in maxwell node (CPU) :

(Assuming you are in /beegfs/desy/user/)

docker run -it --rm -v $PWD:/home engineren/ml:pytorch python main.py --imageSize 16 --batchSize 512 \ --workers 3 --experiment test \

if you run with GPUs in maxwell :

1. singularity instance start --bind $PWD:/home --nv docker://engineren/ml:pytorch testPy
2. singularity run instance://testPy python main.py --imageSize 16 --batchSize 512 --experiment test --cuda