Original paper: Jiang, Yifan, Shiyu Chang, and Zhangyang Wang. "Transgan: Two transformers can make one strong gan." arXiv preprint arXiv:2102.07074 (2021).
TransGAN: Two Transformers Can Make One Strong GAN
Code inspired from TransGAN
-
GELU activation is only approximated through:
to speed up computations done at GPU.
-
MLP is formed by Fully Connected => GELU ==> Fully Connected => GELU
pip install -r requirements.txt
Python version >= 3.6
python train.py
Use Untitled.ipynb to easily modify training parameters