#Course Project

script was tested on linuxmint17 and R 3.0.2

Script will download a zip file from https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip Script will unzip the file to a tmp directory- tempdir() You man need to modify this part if you are using a different OS

extract the feature list (columns of test/train data) from ./UCI HAR Dataset/features.txt This will be used to set the column names in the test/train data sets

Extract the test data from ./UCI HAR Dataset/test/X_test.txt Extract the activity column from ./UCI HAR Dataset/test/y_test.txt extract the subjects column from the ./UCI HAR Dataset/test/subjects.txt

Add these columns to the test dataframe

Extract the test data from ./UCI HAR Dataset/train/X_test.txt Extract the activity column from ./UCI HAR Dataset/train/y_test.txt extract the subjects column from the ./UCI HAR Dataset/train/subjects.txt

Add these columns to the train dataframe

using rbind

Activity column is currently a numeric there are descriptive names these are extracted : ./UCI HAR Dataset/activity_labels.txt I used a loop instead of hard coding in the event that there was an update to the descriptions in future version of the source data

I used grep for mean and std in the column names to extract the numerical indexes I subsetted the merged data set with these column names and subject, activity Then aggregated by the subject and activity columns with MEAN as the function

is written to working directory as output.txt