-
Download the data folder from the link on the course website, https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip.
-
Change the setwd() line of code at the beginning of the R script to match your working directory.
-
When you source the code, it will automatically place the tidy dataset in your working directory.
-
Pulls in the test and training datasets (X_test.txt and X_train.txt)
-
Adds descriptive variable names to both datasets (features.txt)
-
Pulls in and adds columns for subject and test name (subject_test.txt, y_test.txt, subject_train.txt, y_train.txt)
-
Converts the test numbers into descriptive labels (activity_labels.txt)
-
Consolidates the test and training datasets into 1 dataset
-
Limits to columns that only contain "mean()" or "std()"
-
Finds the average of the remaining variables for each subject/activity combination