ttcg: Three-Term Conjugate Gradient for Unconstrained Optimization

Some accelerated three-term conjugate gradient algorithms implemented purely in R with the same user interface as optim(). The search directions and acceleration scheme are described in Andrei, N. (2013) <doi:10.1016/j.amc.2012.11.097>, Andrei, N. (2013) <doi:10.1016/j.cam.2012.10.002>, and Andrei, N (2015) <doi:10.1007/s11075-014-9845-9>. Line search is done by a hybrid algorithm incorporating the ideas in Oliveia and Takahashi (2020) <doi:10.1145/3423597> and More and Thuente (1994) <doi:10.1145/192115.192132>.

Version: 1.0.1
Depends: R (≥ 3.3.0)
Imports: numDeriv
Suggests: testthat
Published: 2021-10-05
Author: Hao Chi Kiang [cre, aut]
Maintainer: Hao Chi Kiang <hello at hckiang.com>
License: GPL-3
URL: https://git.sr.ht/~hckiang/ttcg
NeedsCompilation: no
CRAN checks: ttcg results

Documentation:

Reference manual: ttcg.pdf

Downloads:

Package source: ttcg_1.0.1.tar.gz
Windows binaries: r-devel: ttcg_1.0.1.zip, r-release: ttcg_1.0.1.zip, r-oldrel: ttcg_1.0.1.zip
macOS binaries: r-release (arm64): ttcg_1.0.1.tgz, r-oldrel (arm64): ttcg_1.0.1.tgz, r-release (x86_64): ttcg_1.0.1.tgz, r-oldrel (x86_64): ttcg_1.0.1.tgz

Linking:

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