Kashif Rasul

@chenxiaodanhit these keys come from the gluonts' transformations, in particular their splitters. Hope that helps!

thanks for letting me know.. perhaps i screwed something up while refactoring, I'll check and get back to you

@turmeric-blend which parameters did you give the model? I remember that when I set the `beta_end` to be high I got `nan`s...

thanks! I just re-ran it again on my machine and it all worked out... very strange... 🤔

hmm not sure... perhaps in the downloaded zip folder do: `pip install .` and then try?

also which version of pytorch to you use? I am using pytorch 1.7.1 here

I can try sure.. I just re-ran it again with the parameters from the paper and checked in the notebook, I also fixed the "cuda" device name...

ok so setting a seed e.g. via: ```python np.random.seed(123456) torch.manual_seed(123456) ``` also works for me and I get no `nan` in training... can you try to perhaps train with less...

sorry to hear that... i will try to reproduce on a clean env as well!

Thanks @rruizdeaustri for your interest. Surprisingly ([Nalisnick et al., 2019](https://openreview.net/forum?id=H1xwNhCcYm)) showed that likelihood-based models like flows etc. assign higher likelihoods to Out-of-Distribution (OOD) data, so I would say this model...