Explicit Wasserstein Minimization

Code for A Gradual, Semi-Discrete Approach to Generative Network Training via Explicit Wasserstein Minimization (Chen et al., ICML 2019)

Notes

• Compilation of W1_extension requires PyTorch >= 1.0.0