Debian logoDebian Screenshots >

python3-pycuda

Python 3 module to access Nvidia‘s CUDA parallel computation API

Description

PyCUDA lets you access Nvidia‘s 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 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.
 * 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.

This package contains Python 3 modules.

Upload more screenshots

Please help extend the collection of screenshots. Just make a screenshot and upload it here. You don't need to register or anything.


Homepage

http://mathema.tician.de/software/pycuda


Statistics

3731 other people were interested in this package here. The newest known version of this software is 2018.1.1-3+b1 (Information last updated about 22 hours ago.)