BASICS: pix2pix & cycleGAN - jamahun/fakelab.studio GitHub Wiki

Follow the docs

https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix

Using cycleGAN

  • To view training results and loss plots, run python -m visdom.server and click the URL http://localhost:8097

Train

To train a model split your dataset into 2 folders trainA & trainB and move your dataset directory to ./datasets/<directoryname> once your files are in the datasets folder then use:

python train.py --dataroot ./datasets/<directoryname> --name <modelname>_cyclegan --model cycle_gan --preprocess scale_width_and_crop --load_size 1024 --crop_size 360
⚠️ **GitHub.com Fallback** ⚠️