gpalmer1190 / peerassessment4 Goto Github PK

First, the code reads the files in the UCI HAR Dataset folder.

Then, the column names are added to the test dataset.

An empty matrix is created to store the activity labels when they are added inside the matrix.

Once the for loop substitutes the number for the appropriate label, the subject and activity columns are labeled.

Lastly for the test matrix, the columns and dataset are combined into test_dataset.

Next, the training files are read into variables with column names.

Another empty matrix is made to substitute the activity labels for the training series of activity numbers.

Then, another for loop to substitute numbers for labels.

The train_dataset contains the combination of subject and activity columns with the training dataset.

The first dataset generated is the overall dataset with test and training data combined.

The second dataset finds the columns with mean or std of the features. Then, the dataset is made of just those columns with the subject and activity columns.

The third dataset splits just the test and training datasets by subject groups. Then, calculates the mean of those columns vs. subject groups.

The portion underneath the function call states the output of the three datasets: UCIdataset, UCIpull, and UCI_mean.