https://github.com/jcuda/jcuda-matrix-utils
Utility classes for dense and sparse matrices in JCuda
https://github.com/jcuda/jcuda-matrix-utils
Last synced: 7 months ago
JSON representation
Utility classes for dense and sparse matrices in JCuda
- Host: GitHub
- URL: https://github.com/jcuda/jcuda-matrix-utils
- Owner: jcuda
- License: mit
- Created: 2016-03-09T15:27:04.000Z (almost 10 years ago)
- Default Branch: master
- Last Pushed: 2019-03-08T18:41:35.000Z (almost 7 years ago)
- Last Synced: 2025-04-10T20:31:53.513Z (9 months ago)
- Language: Java
- Size: 36.1 KB
- Stars: 11
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# jcuda-matrix-utils
Utility classes for dense and sparse matrices in JCuda.
# NOTE:
These classes should not be considered to be a stable, general matrix library.
Originally, these classes had been created for one of the
[JCuda samples](http://www.jcuda.org/samples), namely the example
implementation of a CG solver based on JCublas and JCusparse.
The goal of these classes was to ease the conversion between dense and
sparse matrices (in CSR format) and for handling the conversion between
matrices in host-and device memory.
The [samples package](https://github.com/jcuda/jcuda-matrix-utils/blob/master/JCudaMatrixUtils/src/test/java/org/jcuda/matrix/samples)
contains basic examples showing the implementation of CG solvers, and how
to perform a DGEMM operation with sparse matrices using JCusparse.