bayesiains / nflows

Normalizing flows in PyTorch
MIT License
845 stars 118 forks source link

Transforms Summary #18

Closed ttaa9 closed 4 years ago

ttaa9 commented 4 years ago

Sorry if it is already here, but do you have a summary somewhere of the set of transforms (i.e., bijective mapping layers) that you have implemented? A short list of them, perhaps along with the papers they're from and how to use/stack them in practice, would be quite helpful.

I suppose this is related to issue #3.

This looks like a fantastic repo. Thanks for putting it up!

arturbekasov commented 4 years ago

Hi @ttaa9,

We don't have a good summary of the set of implemented transforms, unfortunately. You're right in that this should be one of the deliverables of the wider effort to improve documentation (which indeed starts with #3). Contributions are always welcome! 🙂

If it helps, we've tried to keep all the classes in nsf.transforms well-documented, pointing to papers whenever possible.

Thanks,

Artur