ucidata

[coatless]

Instead of doing the download.files() and read.csv() route, how about just embedding the necessary data sets within an R package ala ucidata. From here, the information could downloaded via CRAN or a GH only release.

I think the only data you needed was: autompg?

In the future, these different packages or routines can be made into "virtual" or self-contained STAT 420 package that includes all the relevant material (c.f. msos)

[coatless]

autompg is now in the ucidata package. This means there is built-in help support via ?autompg and there needn't be a cached version of autompg.csv in the build directory. For instructions as to the read.csv / read.table off of the remote ML link, you should consider linking to the ucidata's build script for it:

https://github.com/coatless/ucidata/blob/master/data-raw/autompg_build.R

[coatless]

@daviddalpiaz

Added more data to ucidata:

• abalone
• adult
• autoimports
• autompg
• Breast Cancer Wisconsin:
  • bcw_original (Breast Cancer Wisconsin Original)
• bike_sharing_daily
• bridges
• glass
• hepatitis
• wine

See the import scripts here:

https://github.com/coatless/ucidata/blob/master/data-raw/

Also, check out the uiucdata... Have enrollment by curriculum and year loaded. About to add graybook & non-graybook data + UIUC Grade Distributions.