GNU Make paper published in JSS

Tags

GNU Make

Using GNU Make to Manage the Workflow of Data Analysis Projects

It’s been a long time since my last post about UseR2019! in Toulouse. Since then there’s been very little time for thinking about research, never mind blogging about it. Work, in the form of re-designing courses, coordinating large courses, and teaching, took over when I got back home (apart from a great 10 day trip to Namibia just before the COVID19 lock down hit).

On 30 August “Using GNU Make to Manage the Workflow of Data Analysis Projects” was finally publised in the Journal of Statistical Software: doi: 10.18637/jss.v094.c01

From now on, I’m hoping to get back into thinking, coding and writing but I’m always an optimist.

UseR!2019 talk available here

Tags

GNU Make

It looks like the direct link to talk on gitlab is blocked. Thanks Ray for letting me know!

The talk may be downloaded here.

If you want to obtain the R Markdown files for the talk, then you may be able to reach the gitlab site from the link on the top left hand side of this page. If you are trying to install `gnumaker` package and don’t have the graphics packages installed then you may need to install one or more graphics packages via `bioconductor`.

I’ll fix the links and installation instructions when I get back to Australia.

In the mean time, I’m having a lovely holiday in France and will be back online in 10 days or so.

Peter

Paper on using GNU Make for data analysis workflow accepted by JSS

Tags

GNU Make

My recently accepted and revised paper (Code Snippet) Using GNU Make to Manage the Workflow of Data Analysis Projects is now in the publication queue at the Journal of Statistical Software. A final draft version, which is subject to change, may be dowloaded here.

The article describes GNU Make pattern rules for R, Sweave, R Markdown, SAS, Stata, Perl and Python to streamline management of data analysis and reporting projects. Rules are used by adding a single line to project Makefiles. Additional flexibility is available for modifying standard program options. An overall strategy is outlined for Makefile construction and illustrated via simple and complex examples.

The GNU Makefile rules described in the paper may be found at https://github.com/petebaker/r-makefile-definitions

Alpha version of codebookr for R is now on github

Tags

make

Alpha version of codebookr now on github

Version 0.0.0.9003 of the R package codebookr is now available. codebookr aids cleaning, checking and formatting datasets using codebook metadata.

It is very preliminary but currently reads codebooks supplied as spreadsheet and also can

• check limits (continuous variables)
• check categories (categorical variables/factors)

Check it out at http://github.com/petebaker/codebookr

useR! 2015 Aalborg Tutorial

Efficient statistical consulting using R: Workflow for data analysis projects
Tutorial Tuesday, June 30 at 9:00-12:00

Pre-tutorial:

Please install the dryworkflow package from github using these commands in R:

library(devtools) # available on CRAN (or github) devtools::install_github(“petebaker/dryworkflow“, dependencies = TRUE)

You should also install GNU make and git. Git should be installed if you have RStudio installed. Make will be installed on linux systems. Windows users should install Rtools and for MACOSX  please install XCode. For more details please see https://github.com/petebaker/dryworkflow

dryworkflow package 0.1.9016 available on github

The dryworkflow package produces a project skeleton for data analysis including R syntax files, report and Makefiles. Given data files and documents, the skeleton is generated with initial directories, template log files, template R syntax for data checking and initial analysis, makefiles and a git repository is initialised.

Further details and installation instructions are available at https://github.com/petebaker/dryworkflow

Initial Post

This site is my first foray into blogging. It’s bound to change.

Peter