Kevin Eales

I have experienced this problem on systems that have ``OpenMP`` as a prerequisite that are built with ``cuda/NVPTX`` extensions such as LLVM. importing cv2 first then scikit-learn results in: ``ImportError:...

This is because OpenCV was compiled with fast_math enabled.

It would also be prudent to allow specification of a preferred single GPU other than device 0.

@swenkel I am attempting this procedure as well, I feel adding a second ``cuda_array_interface`` would likely be the most straight forward method; however, how would one go about the type...

It is quite possible to use a shared memory space to bidirectionally move data from cupy torch OpenCV pycuda ; however, it essential to be aware of the runtime contexts...

See here for some working examples https://github.com/manbehindthemadness/blood-magic Take note that new developments from @cudawarped on the cv2 codebase in relation to memory pointer exchange will likely provide more robust options...

This is very exciting, I will build against the AGX Orin and put it through the paces during my next round of development.

+1

@dingtine there is some mention of an abnormality regarding an end/termination token mentioned here: https://x.com/bartowski1182/status/1782206933719515467?s=46&t=iIhAbXdfE1VCk7vAgMnlRQ as this came out just now, it might affect your results.

Fantastic!