script for peer project for Getting and Cleaning Data
offered on: Coursera.org
offered by: John Hopkins University, Bloomberg School of Public Health
Instructor: Jeff Leek, PhD, Roger D. Peng, PhD, Brian Caffo, PhD
Purpose: This script reads in a series of data files based on a collection set, of activity measurement obtain from a Samsun Galaxy S then parses the data creating a tidy data set, which is then analysed to produce a seperate tidy data set that is a collection of means for the points that are of interest.
To run this script the data structure below witht he data files must be direct attached to the working directory with this script in the working directory. The script is then sourced from within R with the command source(run_analysis)
Input file structure: The files must be in a data directory with two subdirectories called test and training. The data directory must contain the features.txt and activity lables.txt files. Each of the sub directories contains three text files (.txt extensions) and with names that end with "test" or "train" depending the sub directory.
The "x" file contains the observations.
The "y" file contains the activity codes
The "subject" file contains the subject ID.
The files are delinated by spaces.
output:
The script products two tidy data sets:
data: Is a data frame which contains the combined data set from the two sub-directories with the subject and activity identified for each collection set. The files feature names are also cleaned up and presented in a more readable form. The features are a sub set of the orginal collection set of mean and standard deviations reported.
data.new: which contains the average for each selected feature selected in the data data frame by subject and activity.
Details of how the script functions can be found in the run_analysis.md file in this repo. Details of the data, selection, naming, filtering and processing along with a high level description of how the code functions can be found in the CodeBook.md file in this repo.
This repro contain the following files:
| file | purpose |
|---|---|
| .gitignore | a file used by git to exclude certian files from the git repro. |
| CodeBook.md | A markdown file that descripes the data processing and data with this project. |
| Prompt.md | A markdown file that has the instructions for the project. |
| README.md | This file |
| run_analysis.R | the R Script for this project (commented) |
| run_analysis.Rmd | A RMarkdown file that explains the operation of the script that can be run on any computer that supports R and ideally RStudio. |
| run_analysis.md | A markdown file that shows the code in the script and explains the processing. Produced in RStudio from the run_analysis.Rmd file. |
| data directory | the source data for this analysis. |
| data_final.txt | A tab delinated copy of the final tidy data set produced. |