cusna: Native GPU-Accelerated Simulation and Estimation of Network Models

A self-contained native engine (a C interface over 'CUDA' kernels and C++ host logic) for stochastic actor-oriented models (the model family of 'RSiena'), exponential random graph models (cross-sectional, temporal, and separable temporal), and models for binary actor attributes, callable from R without a Python runtime. Modelled on the 'torch' package: the CRAN build is CPU-only from source; the GPU path is compiled from source when a 'CUDA' toolkit is detected at configure time. The data preparation, host statistics ('RSiena' Appendix B conventions), and moment targets are validated bit-for-bit against the reference implementation and reproduce 'RSiena' targets on public datasets to machine precision; the estimators match 'RSiena', 'ergm', 'btergm', and 'tergm' on public benchmark models.

Version:	0.1.0
Imports:	stats, utils
LinkingTo:	cpp11
Suggests:	testthat (≥ 3.0.0), jsonlite, litedown, RSiena (≥ 1.4)
Published:	2026-07-15
DOI:	10.32614/CRAN.package.cusna (may not be active yet)
Author:	Artem Maltsev [aut, cre]
Maintainer:	Artem Maltsev <MaltsevSNA at proton.me>
BugReports:	https://github.com/artemmaltsev74-techcom/cusna/issues
License:	MIT + file LICENSE
URL:	https://github.com/artemmaltsev74-techcom/cusna
NeedsCompilation:	yes
SystemRequirements:	C++17; optionally a CUDA 12.x toolkit (nvcc) for the GPU path
Materials:	README, NEWS
CRAN checks:	cusna results

Documentation:

Reference manual:	cusna.html , cusna.pdf
Vignettes:	Introduction to cusna (source, R code) Accelerating siena07() with cusna (source, R code)

Downloads:

Package source:	cusna_0.1.0.tar.gz
Windows binaries:	r-devel: not available, r-release: cusna_0.1.0.zip, r-oldrel: not available
macOS binaries:	r-release (arm64): cusna_0.1.0.tgz, r-oldrel (arm64): cusna_0.1.0.tgz, r-release (x86_64): cusna_0.1.0.tgz, r-oldrel (x86_64): cusna_0.1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=cusna to link to this page.