Some of the weird quirks of how pytorch modules and functions are called.
Scraping wikipedia to investigate historical density (Part 1).
Making it easier for others to use data sets by providing standardised ways of getting the data and loading the data with cookiecutter.
You can try out bootstrapping with my new shiny app.
Writing papers in Rmarkdown is easy when integrated with Zotero and the Knit-to-Word function.