A PyTorch implementation of Fader Networks: Manipulating Images by Sliding Attributes by Lample et al. by Guillaume Lample, Neil Zeghidour, Nicolas Usunier, Antoine Bordes, Ludovic Denoyer, and Marc'Aurelio Ranzato
I made this implementation because I was looking for a real project that I could use to learn PyTorch. It's possible that there are still bugs in this implementation, because I have not yet been able to reproduce the results from the paper. However, I also haven't implemented the Model Selection step from Section 4. I'm releasing the code in its current state in the hope that it's useful to someone else. Feedback and constructive criticism are welcome, and the same goes for pull requests.
You can download the CelebA dataset here.