The script can be run by typing:

source(./run_analisys.R)

The folder containing the data should be in the working directory.

The library needed for the script are:

• plyr
• reshape2 These are loaded at the beginning of the script.

The script goes through the following main steps:

• load the features and activity list
• load the test and train data
• merge the test and train data (x_measurement)
• get list of features that are either standard deviations or means using 'grep' (searching for 'std' or 'mean(' in features name)
• get measurement for only these features and transpose the data set using 'melt' (x)
• calculate the mean for each features per patient and activity using ddply (x_mean)
• write to a file (x_mean.csv)
• generate the codebook.md