Course Project
Getting and Cleaning Data
Author: Stephen Wade Date: 26/07/2015
This is the repository for the project for Getting and Cleaning Data, a course run by John Hopkins University via coursera.
The output is generated by running the script run_analysis.R in Rstudio.
This will;
- download the UCI HAR data into the
./data/sub-directory; - extract the contents of the archive;
- merge the test and training data into a single tidy dataset, extracting only the mean and standard deviation components of the feature vector; and
- generate a summary (tidy) dataset which contains the average of the mean and standard deviation variables grouped by subject and activity.
To run, the dplyr library (version 0.2 or greater) is required.
The extracted data (without averaging) is stored in output/UCI_HAR_ms_tidy.txt
while the summary (averaged) data is found in output/UCI_HAR_ms_averages.txt,
both are generated using the write.table() function in R.
codebook.mddescribes the original data and the tidy dataset produced by this projectrun_analysis' is a bash script which simply runsrun_analysis.R` using Rscript.download_UCI_data.Rcontains a function for downloading the data, and is invoked byrun_analysis.Rprocess_UCI_data.Rcontains functions for processing/cleaning the UCI HAR datasets into a tidy datasets (ready for merging/summary), is invoked byrun_analysis.R
The MD5 sums of the output files (for verification of procedure) are
UCI_HAR_ms_tidy.txt858320ed6cd349fb51b49e837a5bbe93UCI_HAR_ms_summary.txt2d1141575bef34b7d85b1065c49f521e