cppDE/DESCRIPTION at master · dModverse/cppDE · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Package: cppDE
Title: 'C++' Code Generation for ODEs with Sensitivity Propagation
Version: 0.10.2
Authors@R: c(
    person("Simon", "Beyer", , "simon.beyer@fdm.uni-freiburg.de", role = c("aut", "cre")),
    person("Lawrence Livermore National Security", role = "cph",
           comment = "Portions of cppde_multistepper(_controller).hpp and cppde_newton.hpp derived from SUNDIALS/CVODE(S); see inst/COPYRIGHTS"),
    person("Southern Methodist University", role = "cph",
           comment = "SUNDIALS/CVODE(S) co-copyright holder; see inst/COPYRIGHTS"),
    person("Karsten", "Ahnert", role = "cph",
           comment = "Boost.Odeint Rosenbrock4 base implementation; see inst/COPYRIGHTS"),
    person("Mario", "Mulansky", role = "cph",
           comment = "Boost.Odeint Rosenbrock4 co-copyright holder; see inst/COPYRIGHTS"),
    person("Christoph", "Koke", role = "cph",
           comment = "Boost.Odeint Rosenbrock4 co-copyright holder; see inst/COPYRIGHTS"))
Description: Generates, compiles and runs 'C++' code for models given as
    ordinary differential equations (ODEs) or as algebraic equations. The
    code generator builds an expression graph of the model and derives
    Jacobians and derivative contractions from it by automatic
    differentiation (Griewank and Walther, 2008,
    <doi:10.1137/1.9780898717761>). The integrators are BDF/NDF and
    Adams-Moulton multistep methods, a Rosenbrock method derived from
    'Boost.Numeric.Odeint' and the Tsitouras 5(4) Runge-Kutta pair, with
    dense or sparse ('KLU') linear algebra; an optional backend compiles
    against 'SUNDIALS' CVODE/CVODES (Hindmarsh et al., 2005,
    <doi:10.1145/1089014.1089020>). Sensitivities are propagated forward
    on dual numbers, to first and second order, and backward by a discrete
    adjoint, which yields gradients at a cost independent of the number of
    parameters and, forward over reverse, Hessian-vector products. Events
    triggered by time or state, forcings, and batches of conditions solved
    in parallel are supported.
Language: en-GB
License: MIT + file LICENSE
URL: https://github.com/dModverse/cppDE
BugReports: https://github.com/dModverse/cppDE/issues
SystemRequirements: C++17; Python (>= 3.9) with SymPy. Two further
    features are gated on external libraries that ./configure detects at
    install time: the cvode() backend and the sparse Jacobian path.
    Neither is required to install or use the package, and neither is
    listed here for that reason: when one is absent, only the
    corresponding feature is disabled and reports how to enable it. The
    libraries can be installed as system packages or built from source
    by cppDE itself; see ?cppDE::install_libs and the README.
Encoding: UTF-8
Roxygen: list(markdown = TRUE)
Depends:
    R (>= 4.3.0)
Imports:
    parallel,
    reticulate,
    tools,
    utils
Suggests:
    spelling,
    testthat,
    knitr,
    rmarkdown,
    R.rsp,
    ggplot2,
    tidyr,
    dplyr
VignetteBuilder: R.rsp
Config/roxygen2/version: 8.1.0
Config/testthat/edition: 3
Config/testthat/parallel: true