This repository contains the R code run_analysis.R that perform the folowing tasks:
1. Downloads data sets from the Samsung's Human Activity Recognition project.
2. Stores the data into the directory `./data` in the working directory.
3. Merges the trainig and testing data sets.
4. Exteacts only the mean and standard deviation variables.
5. Replaces the activity codes with descriptive names.
6. Creates a data set with the average of each variable extracted in step 4.
7. Stores the resulting data set in the text file `./data/tidy_data.txt`.
The data sets were obtained from the Samsung's Human Activity Recognition Datasets, which can be daownloaded from the url: https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip
The package reshape2 was needed to process the raw data.
The tidy data consists in a separated by commas dataset with 81 variables and 180 observations.