This repo includes a script run_analysis.R which assumes that the UCI HAR Dataset folder is in the same directory as the script. The script performs five steps in total:
- merge training and test data
- only keep variables that measure the mean and standard deviation
- renames the activities to text
- renames the features to their respective names
- creates a new dataset from the data in step 4 by aggregating the mean for each subject and activity
The script will produce a file datasetMeanBySubjectAndActivity.txt which contains the aggregated dataset from step 5. It might be required to install the dplyr library.
Please see the file Codebook.md.