Skip to content


Repository files navigation

Style-Based GAN in PyTorch

Update (2019/09/01)

I found bugs in the implementation thanks to @adambielski and @TropComplique! (#33, #34) I have fixed this and updated checkpoints

Update (2019/07/04)
  • Now trainer uses pre-resized lmdb dataset for more stable data loading and training.
  • Model architecture is now more closely matches with official implementation.

Implementation of A Style-Based Generator Architecture for Generative Adversarial Networks ( in PyTorch


You should prepare lmdb dataset

python --out LMDB_PATH --n_worker N_WORKER DATASET_PATH

This will convert images to jpeg and pre-resizes it. (For example, 8/16/32/64/128/256/512/1024) Then you can train StyleGAN.

for celebA

python --mixing LMDB_PATH

for FFHQ

python --mixing --loss r1 --sched LMDB_PATH

Resolution Model & Optimizer
256px Link
512px Link
1024px Link

Model & Optimizer checkpoints saved at the end of phases of each resolution. (that is, 512px checkpoint saved at the end of 512px training.)


Sample of the model trained on FFHQ Style mixing sample of the model trained on FFHQ

512px sample from the generator trained on FFHQ.

Old Checkpoints

Resolution Model & Optimizer Running average of generator
128px Link 100k iter Link
256px Link 140k iter Link
512px Link 180k iter Link

Old version of checkpoints. As gradient penalty and discriminator activations are different, it is better to use new checkpoints to do some training. But you can use these checkpoints to make samples as generator architecture is not changed.

Running average of generator is saved at the specified iterations. So these two are saved at different iterations. (Yes, this is my mistake.)


Implementation A Style-Based Generator Architecture for Generative Adversarial Networks in PyTorch







No releases published


No packages published