Replication of Stabilizing GAN Training with Multiple Random Projections and extension including training with multi-objective training via hyper volume maximization (based on Multi-Objective Optimization for Self-Adjusting Weighted Gradient in Machine Learning Tasks)
- Python 3.6
- Pytorch 0.3.0
- Cropped and aligned version of CelebA
- Create a directory named 'celebA' and unzip img_align_celeba.zip inside it
python train.py --ndiscrimiators 12
Collaborators: Joao Monteiro, Breandan Considine