Skip to content

Releases: SparseLinearAlgebra/spbla

SPbLA: The Library of GPGPU-powered Sparse Boolean Linear Algebra Operations

19 Aug 11:09
@EgorOrachyov EgorOrachyov
v1.0.0+zenodo
6773ceb
Compare
Choose a tag to compare
SPbLA: The Library of GPGPU-powered Sparse Boolean Linear Algebra Operations Latest
Latest

Project source code archive for JOSS publication SPbLA: The Library of GPGPU-powered Sparse Boolean Linear Algebra Operations

Authors

  • name: Egor Orachev
    orcid: 0000-0002-0424-4059
    affiliations: 1, 3
  • name: Maria Karpenko
    affiliations: 2
  • name: Pavel Alimov
    affiliations: 1
  • name: Semyon Grigorev
    affiliations: 1, 3
    orcid: 0000-0002-7966-0698

Affiliations

  • name: Saint Petersburg State University
  • name: ITMO University
  • name: JetBrains Research

Tags

  • Python
  • C
  • C++
  • sparse-matrix
  • linear-algebra
  • graph-analysis
  • graph-algorithms
  • nvidia-cuda
  • opencl

Note

spbla project source code for the version 1.0.0+ (Zenodo archive).

Assets 2
Loading

spbla 1.0.0

08 Jun 12:08
@EgorOrachyov EgorOrachyov
v1.0.0
699f351
Compare
Choose a tag to compare
spbla 1.0.0

spbla project source code for the version 1.0.0.

Summary

spbla is a sparse linear Boolean algebra for Nvidia Cuda, OpenCL and CPU computations.
Library provides C compatible API, the core of the library is written on C++ with
CUDA C/C++, CUDA Thrust and OpenCL for actual backend implementation.
Library supports CPU backend as fallback for debugging, prototyping and running on low-end PC.

pyspbla Python-package sources are shipped with the library source code.
It provides high-level safe and efficient access to the library within Python runtime.

Python-package for Linux-based OSs is published in PyPI and available at link.

Features

  • Cuda backend
  • OpenCL backend
  • Sequential (CPU) backend
  • Sparse matrix support
  • Matrix creation (empty, from data, with random data)
  • Matrix-matrix operations (multiplication, element-wise addition, kronecker product)
  • Matrix operations (equality, transpose, reduce to vector, extract sub-matrix)
  • Matrix data extraction (as lists, as list of pairs)
  • Matrix syntax sugar (pretty string printing, slicing, iterating through non-zero values)
  • IO (import/export matrix from/to .mtx file format)
  • GraphViz (export single matrix or set of matrices as a graph with custom color and label settings)
  • Debug (matrix string debug markers, logging, operations time profiling)
Assets 2
Loading