deb_control_files:
- control
- md5sums
deb_fields:
Architecture: all
Depends: fonts-mathjax, libjs-mathjax, libjs-sphinxdoc (>= 7.4)
Description: "module to access Nvidia\u2018s CUDA computation API (documentation)\n\
\ PyCUDA lets you access Nvidia\u2018s CUDA parallel computation API from Python.\n\
\ Several wrappers of the CUDA API already exist\u2013so what\u2019s so special\
\ about\n PyCUDA?\n * Object cleanup tied to lifetime of objects. This idiom,\
\ often called\n RAII in C++, makes it much easier to write correct, leak-\
\ and crash-free\n code. PyCUDA knows about dependencies, too, so (for example)\
\ it won\u2019t\n detach from a context before all memory allocated in it is\
\ also freed.\n * Convenience. Abstractions like pycuda.driver.SourceModule and\n\
\ pycuda.gpuarray.GPUArray make CUDA programming even more convenient than\n\
\ with Nvidia\u2019s C-based runtime.\n * Completeness. PyCUDA puts the full\
\ power of CUDA\u2019s driver API at your\n disposal, if you wish.\n * Automatic\
\ Error Checking. All CUDA errors are automatically translated\n into Python\
\ exceptions.\n * Speed. PyCUDA\u2019s base layer is written in C++, so all the\
\ niceties\n above are virtually free.\n * Helpful Documentation.\n .\n This\
\ package contains HTML documentation and example scripts."
Homepage: http://mathema.tician.de/software/pycuda
Installed-Size: '1522'
Maintainer: Debian NVIDIA Maintainers <pkg-nvidia-devel@lists.alioth.debian.org>
Multi-Arch: foreign
Package: python-pycuda-doc
Priority: optional
Recommends: nvidia-cuda-doc, python-mako-doc, python3-doc
Section: contrib/doc
Source: pycuda
Suggests: python3-pycuda
Version: 2024.1.2~dfsg-1
srcpkg_name: pycuda
srcpkg_version: 2024.1.2~dfsg-1