Gustav Šír
Gustav Šír
Nice! but adding new fields to the base neuron class just for this is an overkill, that's why I was reusing the depth field - there are millions of neurons...
Great! Yes, I agree, a wrapper or hashmap or something like that living in the local scope only. Thanks!
Hi, thanks for the question! For regression yes, absolutely - you can use real numbers (not just 0/1) in the target [queries](https://pyneuralogic.readthedocs.io/en/latest/custom_model.html#queries). However for uncertainty estimates no, as it is...
Hi, hmm, I'm not sure I get it, you mean like globally scaling all sample labels by some factor in the backend? That sounds very specific/ad-hoc, I guess it would...
ah, I see what you mean now - there's definitely no need to do that, you most likely just have a sigmoid at the model output - replace it with...
I can't find exactly with which backend commit this went wrong, but it seems it has been working fine for a while (backend since July 2024, which is a bit...