template.project

R Project Template for Analytic Applications such as Machine Learning Competitions

---

Overview

Using this template reduces:
* Unnecessary variance between projects configurations; and
* Development time spent on making a barebone project working for the first time.
This is possible as the boilerplate comes with:
* Fully configured test-suite, including code-coverage; and
* Fully configured continuous-integration (CI) script for Travis.

Useage

1. Create a new repo on GitHub.
2. Use the git-flow approach in your development cycle.
3. Create a new release named inception.
4. Copy template.project content to the new reposetory.
5. Change the template.project.Rproj file to <package-name>.Rproj.
6. Open the DESCRIPTION file, and edit the following fields:
7. Package modify the package name while using the tidylab. prefix.
8. Title modify the package title; use uppercase words with no period (‘.’).
9. URL modify the package URL such that it leads to its GitHub repo.
10. BugReports edit the URL such that it leads to the package issue page.
11. Description modify the package decription.
12. In README.Rmd delete the Useage Section.
13. Render README.Rmd by clicking the Knit button.
14. Push changed on the inception branch.
15. Go to Travis website, add the project and enable its integration.
16. Decide if you would need binder – an RStudio Server that lets you demonstrate the package. If you do then:

• Uncomment build-binder under .travis.yml; and
• Uncomment Launch Rstudio Binder from README.Rmd.

Installation

You can install template.project by using:

install.packages("devtools")
devtools::install_github("data-science-competitions/template.project")

Function Dependencies

Datasets