This is an intro to the drake
R
package for data pipelining. It might be useful for you if you’re
interested in reproducibility after you make changes to your code.
The talk is split into two parts: slides, and a live coding walkthrough. The latter is meant to be stepped through in an R session rather than knit.
A version of this talk was given at NYR 2019.
…especially if the whole point of drake
is that you don’t need to burn
it all down and start from scratch each time?
This talk is motivated by the NYCFireWire Twitter account with an assist from Gritty at a burner account.
It relies on the rtweet
and
ggmap
packages, so to be able
to run it in full you’ll need a Twitter API access
token and Google Maps
Geocoding API
key.
-
All functions live in
didnt_start_it.R
. -
Feel free to use any and all of the data, including the raw and geocoded tweet motherlode.
- The best resources on
drake
remain thedrake
manual and Will Landau :)