David W. Romero

You can solve this in other python versions by adding `from typing import Tuple` and use `Tuple` instead of `tuple` in the definition of types in functions. Same holds for...

No need to apologize at all! :) happy to be of help! Please let us know if you have any other questions! Cheers, David

Hi @zzw-zwzhang, sorry for my late response. We treat omega_0 as a hyperparameter. In other words, we select it by trying out different values and checking the validation accuracy. Cheers,...

Hi @DiuDiu-1123 I am not sure I understand your question. Do you mean how to show the logits on the test dataset?

Hi @BenjiKCF, thank you for your question and apologies for my late answer. Yes it is. In the current version of the code this is unfortunately not available. We will...

HI liminyu-web, I apologize for the long delay. I am afraid missed your issue. Unfortunately, the memory requirements of self-attention are quadratic. That means there are huge! To give you...

Hi James, sorry for my last response. Yes, indeed. That should be sufficient. Could it perhaps be that you are trying to predict the current value given the previous steps,...

Yes, although it is not necessary, it is practical to do it this way to vectorize operations across all input samples :) (sorry for the long wait, I do not...

Hi Linwei, Thank you for your interest in our work :) The concept of bandwidth is actually analogous to the concept of bandwidth in signal processing. In particular, we refer...

@edsaac Could you please help us out? Has somebody solved the problem?