Data Science, Getting and Cleaning Data

This repo now is home to run_analysis.R. It's pretty straightforward. Just source it in RStudio or your favorite R IDE, make sure that the dir variable points to something sensible as a working directory, and fire it off.

The script is commented to describe what it does, but briefly, it:

• Downloads the prescribed dataset from Cloudfront
• Extracts the data
• Reads it all in
• Combines the training and test data sets into one dataframe
• Names all the columns sensibly, per the provided features_info.txt
• Fixes up the Activity values, per the provided activity_labels.txt
• Generates a tidy dataset of the means of each variable by each activity and subject.

For that last bit, run_analysis.R depends on the dplyr library from CRAN, so get that (I'm not sure if it's normal practice to arbitrarily install libraries from within an R script, as seems a little skeezy). If you hit errors on the group_by call, you'll want to install.packages("dplyr").

Have fun!