Version control your analysis with git
Overview
Teaching: 45 min
Exercises: 15 minQuestions
What is version control?
WHow can I be sure of the script and data I used to generate my manuscript figure?
What are popular version control systems?
Objectives
Understand the importance of version control for tracability of results.
Be able to display the differences between two (small) datasets or two workflows.
Be able to upload a dataset to one of the open-access repositories.
1. Table of contents
1. Introduction
1.1 What is version control?
1.2 How is it valuable for my research and Research Data Management?
1.3 What is the git software?
2. Version control your next publication
2.1 Exercise 1: keeping track of a figure.
Make a folder. Respect the previous folder structure and naming convention that you’ve choosen before.
Add a dataset (link)
Open RStudio and create a new script. Add a comment (using the # sign followed by your comment.)
Open the Shell and navigate (cd) into the folder that you have created previously.
Type git init
Type git status
Create a plot:
- read the data
- make a back and white plot with ggplot.
- save your figure as a
.png
Type git status
(untracked files)
git add
git commit
git tree
You are not so happy with this figure. You could perhaps add some color.
git add
git commit
git tree
2.2 Tracking a processed dataset
Your figure is produced from a relatively small files (X megabytes). You have noticed that some numbers need to be corrected because this measurement was mixed up.
3. Moving online
3.1 GitHub
Provides hosting for software development and version control using Git. Free for Open Source project.
3.2 GitLab
3.3. BitBucket
Key Points
Version control allows you to retrieve the complete history of your processed data and analytical workflows.