Using step 1 and step 2, for every MNIST digit, we maintain a list of (score, feature) pairs. We composed a map that relates each MNIST digit to this list of pairs. In the end, we just have to sort it to find the best/worst cases.

Also, we used the score() function to calculate the reconstruction error:

double score = net.score(new DataSet(example,example));

During the evaluation, we reconstruct the test features and measure how much it differs from actual feature values. A high reconstruction error indicates the presence of a high percentage of outliers.

After step 4, we should see JFrame visualization for reconstruction errors, as shown here:

Visualization is JFrame dependent. Basically, what we do is take the N best/worst pairs from the previously created map in step 1. We make a list of best/worst data and pass it to our JFrame visualization logic to display the outlier in the JFrame window. The JFrame window on the right side represents the outlier data. We are leaving the JFrame implementation aside as it is beyond the scope for this book. For the complete JFrame implementation, refer to GitHub source mentioned in the Technical requirements section.