Alessandro Flati

Also, `./tools/build_pip_package.sh simple` didn't work since it requires `python3.9`. Modifying the bash script with 3.8 would have been fine, except from the fact I had a custom built TensorFlow. In...

No problem, I know too well how hard it takes to retrieve such "structured" feedback on non-standard environments. :) Maybe you have any suggestion for the missing symbol in the...

The answer is yes at all questions :grin: Actually, I have a custom built TF2.7.0, because of Cuda Capabilities 3.0 which were not natively supported (they actually were, but they...

I think they'll need some more information here. How big is your dataset? For small samples, the speeds could be comparable because of the overhead spent in swapping memory from...

There's also another problem in creating `sp2` later on. I'll open a more complete PR soon.

This should be fixed with https://github.com/KindXiaoming/pykan/pull/98 (please try)

Could you please take the time to review https://github.com/KindXiaoming/pykan/pull/98? Because it does exactly what you mentioned and more. Probably you could have saved some migraine!

Yeah we already noticed the slowness of CUDA for small datasets. I didn't have time to test it on big datasets yet, but I think it's actually due to a...

This issue is solved by https://github.com/KindXiaoming/pykan/pull/98

You forgot to add the error message. Also, did you pull latest master version?