README.md

In erickim/GA: Genetic Algorithms for Regression

Genetic Algorithms

By: Rui Chen | Hangyu Huang | Eric Kim | Shuyu Zhao

In order to install this package, make sure you have devtools and then run devtools::install_github("erickim/GA"). If not, please run the following.

install.packages("devtools")
library(devtools)
devtools::install_github("erickim/GA")

Once you've installed the package, you can load it in via

library(GA)

If the documentations on the help pages fail to load because of a "corruption" error, please restart your R process. It should work then.

For a basic usage of the genetic algorithm for feature selection in linear regression, run

Y <- MASS::Boston[,1]
X <- MASS::Boston[,-1]

select(Y, X)

For more detailed control over the code, check help("select") or help("GA::select") for extra details on the arguments and default values.

To run test code, make sure to have the testthat library installed and run test_package("GA"). If not, please run

install.packages("testthat")
library(testthat)
test_package("GA")

There is one last hidden .internal directory. Should you choose to look into that directory, you will find a simplified spec for this project in spec.md that we used to split up some of the tasks as well as an early design document in design.md. Our code does not strictly adhere to the design document, but it provides a good idea of our early directions. Lastly, explor.md has some more test code and example usages that we came up with when initially implementing the functions.

erickim/GA documentation built on May 28, 2019, 7:15 a.m.