Jens Egholm Pedersen

> I did not claim that Norse is owned by Heidelberg. You wrote "Even though a lot of the original code I contributed has been refactored / slightly modified most...

Is this closed with #123 @cpehle ?

How about using the odd-firing model in https://github.com/norse/notebooks/blob/master/parameter-learning.ipynb? It's sufficiently complex to have to be learned and it shouldn't steal too many resources.

What's exactly the purpose of this? Performance? Usability?

One more thing; I had a look at the tests for Mac. For obvious reasons it doesn't seem to support CUDA. Could you add a quick test / try-except block...

> > One more thing; I had a look at the tests for Mac. For obvious reasons it doesn't seem to support CUDA. Could you add a quick test /...

Also agreed in principle. What does your comment mean for this PR @cpehle? Are you suggesting to rework it or would you be ok with a merge with a warning...

Thanks for pointing this out! We're in the process of releasing a new version of this. I'll push a fix for the task as soon as I can.

Alright, the PR should have fixed this. Could you kindly let me know if it's still not working? :-) Thanks for reporting!

Here's a paper summarising online learning in SNNs: https://arxiv.org/pdf/1907.02649.pdf