• 'README.md': Overall description of the contents of this data.
• 'CodeBook.md': Description of data fields in the sample data.
• 'fitnessData.txt: Fitness tracking accelerometer data means and standard deviations.
• 'fitnessDataAverageByActivityAndSubject.txt: Fitness tracking accelerometer data means and standard deviations, averaged for each grouping of activity type and subject.
• 'run_analysis.R': An R script that downloads the original data, unpacks it, and transforms it to produce 'fitnessData.txt and 'fitnessDataAverageByActivityAndSubject.txt.

The provided script performs a full conversion from the original UCI_HAR_Dataset to the fitnessData.txt and fitnessDataAverageByActivityAndSubject.txt datasets.

A full description of the UCI_HAR_Dataset can be found at their website.

The following steps are performed on on the UCI_HAR_Dataset to create the new dataset.

1. Merge the training and test datasets for:
2. subjects - UCI HAR Dataset/test/subject_test.txt - UCI HAR Dataset/train/subject_train.txt
3. features - UCI HAR Dataset/test/X_test.txt - UCI HAR Dataset/train/X_train.txt
4. activities - UCI HAR Dataset/test/y_test.txt - UCI HAR Dataset/train/y_train.txt
5. The columns for the data are labeled as follows:
6. subjects: One column, named 'subject_id'.
7. features: Column names taken from UCI HAR Dataset/features_info.txt.
8. activities: The original activity id is in 'activity', and this value is mapped to a name taken from UCI HAR Dataset/activity_labels.txt and placed in the column 'activity_label'.
9. The columns for the featuers that are not an average, i.e. mean(), or standard deviation, i.e. std(), are stripped from the data.
10. The fitnessData.txt dataset is created by putting together the columns of subjects, features, and activities.
11. The fitnessDataAverageByActivityAndSubject.txt is created as follows:
12. Rows that share the same activity and subject_id are grouped together.
13. Each group of rows is replaced with a single row that contains the average value for every feature in features.
14. The features columns are renamed by prefixing average_ to their name.

The full description of the columns and their units can be found in CodeBook.md.

1. First start R.

$ R
>

2. Source the provided script, which will produce an output like the following:

> source("run_analysis.R")
Looking for 'UCI_HAR_Dataset.zip'.
 ... File found.
Looking for extracted data in 'UCI HAR Dataset'.
 ... extracting data in UCI HAR Dataset
Reading feature data...
Labeling feature data...
Trimming feature data columns...
Reading activity data...
Labeling activity data...
Reading subject data...
Labeling subject data...
Writing tidy data to 'fitnessData.txt.
Writing average data grouped by activity and subject to 'fitnessDataAverageByActivityAndSubject.txt.
>

3. The data is ready to eat in 'fitnessData.txt and 'fitnessDataAverageByActivityAndSubject.txt.