Skip to content

Latest commit

 

History

History

BigGAN-PyTorch

README.md

How To Use This Code

You will need:

  • PyTorch, version 1.6.0
  • tqdm, numpy, scipy, and h5py

Before running SSGAN-LA on Tiny-ImageNet, you need to manually download the Tiny-ImageNet dataset and store it into data folder with the following structure.

tiny_imagenet
    train
        cls0
            img0
            img1
            ...
        cls1
            img0
            ...
        ...
    val
        cls0
            img0
            ...
        cls1
            img0
            ...
        ...

Before first running methods on a dataset named DATASET={C10,S10,TINY}, please run the following command to prepare the statistics for calculating FID.

python calculate_inception_moments.py --dataset DATASET --data_root data
  • run SSGAN-LA on CIFAR-10
sh scripts/launch_cifar10_ema_ssganla.sh
  • run SSGAN-LA on STL-10
sh scripts/launch_stl10_ema_ssganla.sh
  • run SSGAN-LA on Tiny-ImageNet
sh scripts/launch_tiny_ema_ssganla.sh

Acknowledgements

This code is developed based on BigGAN-PyTorch.