The original data was downloaded from https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip, on 2014-05-14 20:07:28 BRT, into the ./data folder of my working directory. I named it as proj_dataset.zip. Then, it was extracted into the ./data/UCI HAR Dataset folder. The following script did this to me:
if ( !file.exists("./data") )
dir.create("./data/")
fileDataset <- "./data/proj_dataset.zip"
if ( !file.exists("./data/proj_dataset.zip") ) {
fileUrl <- "https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip"
download.file(fileUrl, file.dataset, method="curl")
dateDownloaded <- Sys.time()
fileDate <- "./data/download_date.txt"
if ( file.exists(file.date) ) {
invisible(file.remove(fileDate))
} else {
invisible(file.create(fileDate))
}
writeLines(paste("Original data downloaded on:", dateDownloaded, "BRT"), fileDate)
}
unzip(fileDataset, list=T) # many files! let's extract them...
unzip(fileDataset, exdir="./data")
list.files("./data")
list.files("./data/UCI HAR Dataset")
After that, the run_analysis.R script was run to generate my analysis. It is not needed to run the download script before the analysis, as long as the dataset is already unziped on the ./data/UCI HAR Dataset folder, relative to the working directory.
The Code Book contains info on the datasets - the original ones and the one generated by my script and my analysis.
A full description of the original study can be found here: http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones