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.
cusna is a self-contained native engine — a C ABI over CUDA kernels and
C++ host logic (libcusna) — callable from R without a Python runtime. It is
the native counterpart of the reticulate-based cusna R wrapper, modelled on the
R torch package:
nvcc) is
detected at configure time.cusna_has_cuda() reports which build you have.
| Family | Functions | Validated against |
|---|---|---|
| SAOM (RSiena) | saom_data(), cusna_effect(), mom_estimate(), cusna_fit methods; behavior co-evolution, composition change, mom_estimate_multinet(), cusna_fran() |
data/masks/targets bit-identical to the reference; estimates within simulation SE; RSiena targets to machine zero |
| ERGM | ergm_simulate() (TNT sampler), ergm_stats(), ergm_mple(), ergm_mcmle() |
sampler ≡ ergm::simulate; MLE matches ergm::ergm() |
| Temporal ERGM | tergm_mple() (+ block bootstrap), tergm_simulate(), stergm_cmle() |
matches btergm to machine precision; tergm CMLE within SE |
| ALAAM | alaam_mple(), alaam_mcmle(), alaam_simulate() |
MPLE ≡ glm; MLE recovers observed moments |
| Low-level | cusna_network_stats(), cusna_behavior_stats(), cusna_gof_distribution() |
RSiena Appendix B conventions, machine zero |
The underlying C ABI is bit-for-bit validated in native/test
(see native/VALIDATION.md).
# from a checkout of the monorepo (configure vendors ../../native sources):
install.packages("cpp11") # build-time only
# then, with a C++17 toolchain (Rtools on Windows):
# R CMD INSTALL Rpkg-native
library(cusna)
cusna_has_cuda() # FALSE on the CPU-only build
# a two-wave panel and a Method-of-Moments SAOM fit, all native:
set.seed(7)
w1 <- matrix(as.integer(runif(400) < 0.12), 20, 20); diag(w1) <- 0L
w2 <- w1; flip <- sample(400, 40); w2[flip] <- 1L - w2[flip]; diag(w2) <- 0L
fit <- mom_estimate(saom_data(list(w1, w2)),
effects = list(cusna_effect("density"), cusna_effect("recip")))
summary(fit)
# an ERGM maximum-likelihood fit on the same data:
ergm_mcmle(w1, list(ergm_term("edges"), ergm_term("mutual")), directed = TRUE)
See vignette("cusna") for the full tour (covariates, co-evolution,
multi-network models, TERGM/STERGM/ALAAM) and vignette("siena07-backend")
for driving RSiena's siena07() on the native simulator.
See BUILD.md.
In short: the CPU-only build needs a C++17 compiler
(Rtools on Windows); the GPU build additionally needs a CUDA 12.x toolkit.
The native sources are vendored from ../../native/ by configure.
MIT (see LICENSE). We compare outputs against RSiena but do not link its code.