###That file describes how the run_analysis.R script works
####1 The first part of the script sets all files into objects in R. The objects are:
- activity_label as label
- features.txt as clabel
- flabel.txt file flabel
- X_test.txt as testx
- y_test.txt as testy
- X_train.txt as trainx
- y_train.txt as trainy
####2 The second part of the script takes the objects of test and train data and merges x test with y test and x train with y train.
Before merge the script creates a id variable for each object.
- creation of id variable
- test is a merge of testx with testy
- train is a merge of trainx with trainy
Than the script takes a sqldf function to join test with label (activity_label), and train with label. I used the sqldf function, becouse I'm confortable with SQL language. I know I could use merge(), or another function to do it.
The result of sqldf function was:
- testt
- trainn
Later the script takes testt and trainn objects and join each one with clabel (object with colnames). So we got the testt and trainn object with colnames and id variable and activity names as well.
3 The third part of the script split testt and trainn variables taking only means and std variables. The script put new variables names based on file:flabel.txt.
And at the end, of the script creates a final merged dataset called tidydata and also a final out table called tidy.data.csv.
#####OBS: In the last part of the run_analysis script there is a code commented. If you want see a final dataset organized by test and train you can uncomment and run this part of code. It will create a final out file called: tity-dataset.csv summarized by test and train.