Philipp Kats, May 2015
##Summary This course project aims to merge and tidy train and test parts for the Human Activity Recognition Using Smartphones Data Set.
##Instructions Script requires data to be unziped and stored in /data folder. It also requires plyr package.
####0. Setup and read data At this point we read 3 files (data, labels and subjects) for each dataset (train and test). Also, we read the file with all variable names and attach those names to columns in both datasets
####1. Merges the training and the test sets to create one data set Here we merge train ad test datasets into one.
####2.Extracts only the measurements on the mean and standard deviation for each measurement Now we define a checkName function, which answers whatever chosen column is needed, depending on column name - we keep "subject", "labels" and all other columns with "mean()" and "std()", skipping all others. Then we apply this function to filter, creating new, reduced dataset.
####3. Describtive variables - we can skip this step as columns are already renamed with discribtive labels in step 1
####4. Uses descriptive activity names to name the activities in the data set Now we transform both subject and labels to factor levels. then, using mapvalues function from plyr package we rename labels in labels with the given vector of names (I hardwrited labels as ther is only six of them)
####5. creates an independent tidy data set with the average of each variable for each activity and each subject Now we aggregate averages for each activity type and subject using aggregate function.
####6. Saving file now lets save new dataframe