Test functions are often used to test computer code. They are used in optimization to test algorithms and in metamodeling to evaluate model predictions. This package provides test functions that can be used for any purpose. Some functions are taken from < https://www.sfu.ca/~ssurjano>, but their R code is not used.
This R package provides functions for various surfaces that can be used for testing of models, algorithms, etc. For example, if you want to compare how well a neural network does compared to a random forest, you can take data from one of these test functions to train and test them.
Each of the test functions is called like any other function. The first argument,
x, should be a vector representing one point or a matrix that has points in its rows. This can cause problems if you are using a 1-dimensional function and pass in a vector of values. Instead you should pass them in as a matrix with a single column, or vectorize the function.
set.seed(0)library(TestFunctions)branin(runif(2))branin(matrix(runif(20), ncol=2))#>  24.119600 71.180268 18.374071 9.839029 36.607437 72.884496#>  196.302169 25.185022 13.059216 27.129463
Install it like any other package with
install.packages("TestFunctions") # Or the the development version from GitHub: # install.packages("devtools") devtools::install_github("CollinErickson/TestFunctions")
Add many new test functions. Added some function modifiers, such as add_zoom.
Releasing an R package that provides generic test functions to be used for any purpose.
Accepted to CRAN on 9/26/2016.