Refeshments - Talk/demo - Lunch
Schedule
11:30 - 12:00 Refreshments (in-person)
12:00 - 13:00 Talk/demo (hybrid)
13:00 - 14:00 Lunch (in-person)
Abstract
Reproducibility of scientific research enables others including your future self to validate, extend, and build upon analytical results. This in turn, helps build confidence in the results of scientific analyses. However, reproducibility is not a binary concept, rather, there is a scale from less reproducible to more reproducible, where various tools and practices can help enhance it.
The R package {targets}, developed and maintained by Will Landau, is a workflow management package designed to increase reproducibility in R based data analysis. The major features of {targets} include automation of workflows, caching of intermediate steps, batch creation of workflow steps, and parallelisation at the level of the workflow. These features not only help to reproduce scientific analysis, but also help you to tackle several other challenges in your research workflows. For example, it supports you to return to a project after working on something else and still be able to immediately pick up where you left off without confusion or trying to remember what you were doing. If you change the workflow, then you only have to re-run the parts that are affected by the change. It is also possible to scale up the workflow, to say, handle large datasets, without changing the underlying individual functions.
This session will provide an introduction to using {targets} for reproducible data analysis in R. Participants will learn how to structure their workflows to make their data analysis more reproducible in R.
