This is a fast C++ backend for the PCMBase R-package.
The package needs a C++ 11 compiler and Rcpp to be installed in you R-environment. Once this is done, you can install the most recent version of the package from github:
devtools::install_github("venelin/PCMBaseCpp")
If you experience problems installing the package from github, you may try installing a possibly older version from CRAN:
install.packages("PCMBaseCpp")
Once the package is installed, use the function BenchmarkRvsCpp
to
evaluate the gain in speed of the likelihood calculation on your
machine, relative to the R implementation:
library(PCMBaseCpp)
library(data.table)
options(digits = 4)
# Depending on your use case, you can change the number of traits, as well as the
# other arguments:
benchRes <- BenchmarkRvsCpp(ks = 2, includeParallelMode = FALSE, verbose = TRUE)
# Example output:
# Performing benchmark for k: 2 ; optionSet: serial / 1D-multiv. ...
# k modelType N R mode logLik logLikCpp timeR timeCpp
# 1: 2 MGPM (A-F) 10 2 11 -7.416e+02 -7.416e+02 0.010 0.0007
# 2: 2 MGPM (A-F) 100 4 11 -4.294e+03 -4.294e+03 0.107 0.0016
# 3: 2 MGPM (A-F) 1000 11 11 -1.700e+05 -1.700e+05 1.221 0.0095
# 4: 2 MGPM (A-F) 10000 11 11 -1.210e+06 -1.210e+06 12.443 0.0795
# 5: 2 BM (B) 10 2 11 -4.451e+03 -4.451e+03 0.010 0.0003
# 6: 2 BM (B) 100 4 11 -8.427e+03 -8.427e+03 0.082 0.0008
# 7: 2 BM (B) 1000 11 11 -1.830e+04 -1.830e+04 0.847 0.0064
# 8: 2 BM (B) 10000 11 11 -6.574e+05 -6.574e+05 8.414 0.0663
# 9: 2 OU (E) 10 2 11 -1.126e+04 -1.126e+04 0.016 0.0006
# 10: 2 OU (E) 100 4 11 -8.486e+05 -8.486e+05 0.147 0.0015
# 11: 2 OU (E) 1000 11 11 -1.234e+06 -1.234e+06 1.505 0.0096
# 12: 2 OU (E) 10000 11 11 -1.058e+07 -1.058e+07 15.062 0.0854
For further examples, read the Getting started guide and the reference available on the package homepage.
To give credit to the PCMBase package in a publication, please cite the following articles:
Mitov, V., & Stadler, T. (2018). Parallel likelihood calculation for phylogenetic comparative models: The SPLITT C++ library. Methods in Ecology and Evolution, 2041–210X.13136. http://doi.org/10.1111/2041-210X.13136
Mitov, V., Bartoszek, K., Asimomitis, G., & Stadler, T. (2019). Fast likelihood calculation for multivariate Gaussian phylogenetic models with shifts. Theor. Popul. Biol. https://doi.org/10.1016/j.tpb.2019.11.005
The PCMBaseCpp R-package uses the following R-packages and C++ libraries:
Copyright 2016-2020 Venelin Mitov
Source code to PCMBaseCpp is made available under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. PCMBaseCpp is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.