Reproducible Research (JHU Coursera, Course 5)
The fifth course in the data science specialization, “Exploratory Data Analysis” was an okay course. It is definitely frustrating not starting any modeling as it hasn’t been covered in the first 5 courses, but making code reproducible is a valuable skill so I continued along the specialization. As always, the code for the quizzes and assignments is located on my github.
Week 1 Review: No real coding this week. Just going over what makes code producible. The week 1 quiz can be seen below.
Week 2 Review: Week 2 went over R Markdown and Knitr which is pretty useful as gists (the things showing the quizzes and projects on this blog post) can be made with R Markdown and Knitr. The quiz questions this week weren’t that useful though.
This week had a project which was basically just reinforcing R Markdown knowledge by analyzing Fitbit-like data. The raw R Markdown version of the project is located here.
Week 3 Review: Fantastic week. There was one major important lesson emphasized this week which was don’t do things by hand. Anyways, there was no quiz in week 3 so that was a welcome relief.
Week 4 Review: The lecture was more of the same from week 3. The project was on analyzing severe weather events in the United States and their health and economic consequences. Practical project in the sense that it forces people to download unzip, do some cleaning and aggregation on the data, and make a pretty html file (you can find mine here). However, I said something very similar for the previous course.
Please let me know if you have any questions or if you have any tips on how I can improve my coding! Reproducible Research was a bit of a disappointment, but I can’t wait to see what course 6 has to offer.