dominikrafacz / drake-presentation

Presentation about drake R package prepared for Gdańsk 2019 satRday

Who am I?

A few words describing me in a random order:

🧑‍🎓 Student. 👐 Extravert. ✝️ Catholic. 💃 Dancer. 🌈 Gay. 👨‍💻 Developer. 🎲 Boardgamer. 🇵🇱 Polish. 🐶 Dog person. ⛰ Highlander.

What am I able to do?

• Programming in R, Java, Python, C++, some JavaScript
• Especially, developing R packages and R Shiny apps
• Visualizing stuff and analyzing data
• Learning machines
• Singing Eminem's songs (but only selected ones)
• Working with frameworks like Vue.js etc.
• Tickling Amazon AWS
• Putting my foot behind my head
• Refactoring code (oh gosh, I really love doing it)

What am I working on?

• Developing GUI for HaDeX 2.0
• Writing some open-source R packages for turtletopia
• ... and writing some articles for the blog
• Finishing my master's thesis
• Learning Rust

What funny GitHub stats can I show?

Contact me!

drake-presentation's Issues

Permission to link to this presentation in the drake manual

@DominikRafacz, thank you for presenting drake at the a satRday. It really helps that you are spreading the word. May I link to your materials from this section of the manual?

Code and supporting data for the benchmarks

@DominikRafacz, it was helpful to see the parallel computing benchmarks on slide 9. Would it be possible to access the full code base, including phplE7q6h.csv or an equivalent input file, so I can reproduce them? I think this is a good opportunity to dive deeper into profiling and reduce drake's computational overhead.

Use of drake_config()

@DominikRafaczk, I would just like to point out a minor detail on slide 6. Contrary to intuition, problems can arise when you call

vis_drake_graph(drake_config(plan))

or

drake_config(plan) %>%
  vis_drake_graph()

rather than the more reliable

config <- drake_config(plan)
vis_drake_graph(config)

Details: ropensci/drake#163, ropensci/drake#335, ropensci/drake#527, ropensci/drake#874.