PyCUDA: Pythonic Access to CUDA, with Arrays and Algorithms
.. image:: https://gitlab.tiker.net/inducer/pycuda/badges/main/pipeline.svg
:alt: Gitlab Build Status
:target: https://gitlab.tiker.net/inducer/pycuda/commits/main
.. image:: https://badge.fury.io/py/pycuda.png
:target: https://pypi.org/project/pycuda
.. image:: https://zenodo.org/badge/1575319.svg
:alt: Zenodo DOI for latest release
:target: https://zenodo.org/badge/latestdoi/1575319
PyCUDA lets you access Nvidia <https://nvidia.com>
's CUDA <https://nvidia.com/cuda/>
parallel computation API from Python.
Several wrappers of the CUDA API already exist-so what's so special
about PyCUDA?
-
Object cleanup tied to lifetime of objects. This idiom, often
called
RAII <https://en.wikipedia.org/wiki/Resource_Acquisition_Is_Initialization>
_
in C++, makes it much easier to write correct, leak- and
crash-free code. PyCUDA knows about dependencies, too, so (for
example) it won't detach from a context before all memory
allocated in it is also freed.
-
Convenience. Abstractions like pycuda.driver.SourceModule and
pycuda.gpuarray.GPUArray make CUDA programming even more
convenient than with Nvidia's C-based runtime.
-
Completeness. PyCUDA puts the full power of CUDA's driver API at
your disposal, if you wish. It also includes code for
interoperability with OpenGL.
-
Automatic Error Checking. All CUDA errors are automatically
translated into Python exceptions.
-
Speed. PyCUDA's base layer is written in C++, so all the niceties
above are virtually free.
-
Helpful Documentation <https://documen.tician.de/pycuda>
_.
Relatedly, like-minded computing goodness for OpenCL <https://www.khronos.org/registry/OpenCL/>
_
is provided by PyCUDA's sister project PyOpenCL <https://pypi.org/project/pyopencl>
_.