News for hackers. Stuff that matters.
Re:Streamz: Python pipelines to manage continuous streams of... by betterwithbacon on Friday March 27, 2020 @01:14PM (#22704722)
PyTorch supports both `__cuda_array_interface__` and dlpack which allows effectively sharing the GPU pointer without having to go through a host numpy array. Tensorflow is actively working on adding support for dlpack here: https://github.com/tensorflow/tensorflow/issues/24453

There are no comments on this comment yet.