poncatime

Analysis of Ponca performances using atime

Fetch the sources

To compile this project, you need to retrieve the dependencies using git submodules:

git clone https://github.com/poncateam/poncatime.git
cd poncatime
git submodule update --init --recursive

Compile and test

First install R. If you don’t have the Rcpp package, you can install it in R via:

install.packages("Rcpp")

If you edit the *_interface functions in src/interface.cpp then you need to re-generate src/RcppExports.cpp and R/RcppExports.R:

R -e "Rcpp::compileAttributes('path/to/poncatime')"

After that, you can compile the C++ code in this package, and install it, by running

R CMD INSTALL path/to/poncatime

Finally, you can run tests via

R --vanilla < tests/testthat/test-CRAN.R

Or test under valgrind via

R -d valgrind --vanilla < tests/testthat/test-small.R

Benchmarking

performance testing

.ci/atime/tests.R contains test cases.

To run the performance testing, I had to make some modifications to atime to handle this use case. To install the updated version,

remotes::install_github("tdhock/atime@poncatime")

Then I run the performance test suite via

atime::atime_pkg("path/to/poncatime")

which creates result files in poncatime/.ci/atime.

comparative benchmarking

If there are other algorithms (baselines, state-of-the-art) that do a similar computation as asoCurveEstimation, we can add them to the atime code in the vignette, to compare performance.