Pytorch implementation of A Style-Based Generator Architecture for Generative Adversarial Network
- Python3
- Pytorch >= 1.0.0
- TensorBoardX
- fire
- apex [optional]
- pyyaml
train
python main.py
--config_file=path_to_config_file
--checkpoint=path_to_config_file[default='']
inference
python main.py
--config_file=path_to_config_file
--run_type=inference
Default configuration file is located in config directory.
- Progressive method
- Tuning
- Add mapping and styles
- Remove traditional input
- Add noise inputs
- Mixing regularization
There seems to be no difference in the score.
There is a problem with R1 regularization, so training does not work properly. This also affects image samples. It would be better not to use it now.
There seems to be a clear speed difference depending on the precision, but it seems to be meaningless because the mixed precision training isn't done properly.
