suzbert / gacd-activity-recognition-using-smartphones

GACD-Activity-Recognition-Using-Smartphones

an exercise using R to create tidy data

#The R script called "run_analysis.R" executes the following steps: Test data are run through the scripted commands first, then training data.

1. Reads activity labels from "activity_labels.txt".

2. Reads column names from "features.txt". Reads the data (x, y and subject). For test data: from 'test/X_test.txt', 'test/y_test.txt', "test/subject_test.txt". For training data: from 'train/X_train.txt', 'train/y_train.txt', "train/subject_train.txt".

3. For the test and training data: Identifies data to subset out from features: Extract_features gets only column names containing "mean" or "std"

4. Uses "features.txt" to supply column names for the test and training datasets

5. Adds a new column "activity_label" next to activity IDs in the test and training datasets, then populates the column with the labels from "activity_labels.txt".

6. Pulls all test data together, column-wise, and puts "subject" into first column.

7. Adds new columns "subject", "activity_id", "activity_label" into the test and training datasets.

8. Merges test and training data.

9. Melts data so that each subject has one line of data.

10. Stores tidy_data in plain text file called "tidy_data.txt".