GPU arrays (via CuPy) when possible, otherwise CPU arrays (via NumPy)
Project description
HyRay
GPU ? CuPy : NumPy
GPU arrays (via CuPy) when possible, otherwise CPU arrays (via NumPy)
The hyray
library provides a single interface to use both NumPy and CuPy, depending on the availability of CuPy, CUDA, or sufficient memory. With hyray
a single implementation can run with CuPy when CuPy is available (on a local machine with a GPU), and still work with NumPy when on GPU or insufficient memory is available (such as on continuous integration servers).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
hyray-0.1.0.tar.gz
(17.9 kB
view hashes)
Built Distribution
hyray-0.1.0-py3-none-any.whl
(17.2 kB
view hashes)