Frédéric Bastien

@denera As this PR is removed some old API not used, we need to have that documented in the next release. Is there a place we need to add them...

cublas from CUDA 12 already as this fix. We do not have yet a public cudnn for CUDA 12. So JAX can't be easily build with CUDA 12 yet. This...

The new cudnn 8.8 release is done. So all should be good now for a full CUDA 12 stack. https://docs.nvidia.com/deeplearning/cudnn/release-notes/index.html#rel-880

yes.

I found some code change in the example that make it work on Windows. I forgot where it is. Look in others issue in this repo or on Theano mailing...

Look in other issue in this repo or Theano mailing list. It worked with a small patch on Windows. Le sam. 8 avr. 2017 05:10, goffredogiordano a écrit : >...

I think we need a no-optimization region mechanism. This will happens in other cases too. What do you think of that?

Was this merged manually? At least part of it seems merged.

I think it is still not implemented. We would also like to have this done. So +1 :) thanks for the good plugin.

@abergeron, I think this is done, can you confirm?