TMB - Template Model Builder: A General Random Effect Tool Inspired by 'ADMB'
With this tool, a user should be able to quickly implement complex random effect models through simple C++ templates. The package combines 'CppAD' (C++ automatic differentiation), 'Eigen' (templated matrix-vector library) and 'CHOLMOD' (sparse matrix routines available from R) to obtain an efficient implementation of the applied Laplace approximation with exact derivatives. Key features are: Automatic sparseness detection, parallelism through 'BLAS' and parallel user templates.
Last updated
openblascppopenmp
15.62 score 195 stars 178 dependents 2.7k scripts 63k downloadsRTMB - 'R' Bindings for 'TMB'
Native 'R' interface to 'TMB' (Template Model Builder) so models can be written entirely in 'R' rather than 'C++'. Automatic differentiation, to any order, is available for a rich subset of 'R' features, including linear algebra for dense and sparse matrices, complex arithmetic, Fast Fourier Transform, probability distributions and special functions. See Kristensen et al. (2026) <DOI:10.1002/ece3.74572>.
Last updated
openblascpp
14.33 score 84 stars 70 dependents 900 scripts 47k downloadstmbstan - MCMC Sampling from 'TMB' Model Object using 'Stan'
Enables all 'rstan' functionality for a 'TMB' model object, in particular MCMC sampling and chain visualization. Sampling can be performed with or without Laplace approximation for the random effects. This is demonstrated in Monnahan & Kristensen (2018) <DOI:10.1371/journal.pone.0197954>.
Last updated
onetbbcpp
9.00 score 34 stars 2 dependents 293 scripts 1.4k downloads