Home

Sharing my findings

With the current vocabulary, I will say that I am a “data scientist”, i.e. a scientist who plays with data. This website is my modest participation in sharing science… I write blog articles with R scripts in it. I used to mainly write about maps and spatial data, but I am now more focused on making development tools accessible to a larger audience.

Open-source projects

I have some personal open-source projects.
Today, I mainly develop R packages to improve development workflows and help everyone to have access to great tools allowing to share their nice work more easily. This is driven by ‘Rmd first’ method, package development and git management tools like:

  • {fusen}
  • {attachment}
  • {gitlabr}

As a former researcher, I decided to share my R-scripts as R libraries. I cannot say these respect all packages development standards but at least the code is available. Have a look at the following packages:

  • {SDMSelect}: 1. Covariate selection procedures on GLM and GAM. 2. Species distribution modelling.
  • {GeoDist}: 1. Calculate distances between points when there are obstacles. 2. Modifications of {geoR} functions to allow for custom distances.

I also use Hugo for this website, which made me work on my own template:

Other projects are available on my Github account.

Who am I?

I am a business unit director at ThinkR, a company where we answer all your questions with R, offering training, development, installation and consultancy in everything around R.

Deep inside, I am a scientist. Indeed, I have a master in agronomy and a PhD in marine biology. But data analysis and R diverted me a little from my original objectives. I left the world of marine biology research to join ThinkR. As a result, I continue to share my knowledge through trainings and open-source development. I still do data analysis and visualization, mostly through consultancy, while sharing development good practices. And I love it! I play with the cards regularly, but I don’t really adjust patterns anymore. That said, I still have material to fill this blog and share my findings and other tips with you. On R mainly!
If you are curious about what we can do with spatial tools on non-geographical data, have a look at my category “geohacking”.

Enjoy your reading!

Sébastien Rochette

3D visualisation of a meristem of Arabidopsis thaliana rendered with {rayshader}

Figure 1: 3D visualisation of a meristem of Arabidopsis thaliana rendered with {rayshader}

Jeff Leek, on simplystatistics.org

If every method in every stats journal was implemented in a corresponding R package (easy), was required to have a companion document that was a tutorial on how to use the software (easy), included a reference to how to cite the paper if you used the software (easy) and the paper/tutorial was posted to the relevant message boards for the communities of interest (easy) that journal would see a dramatic bump in its impact factor.