Use first $10$ rows as training, the next $4$ rows as testing set. Apply KNN Classifier to the new data table in part b$).$ In other words, build your KNN classifier by the following requirements based on the knowledge in the table, and then predict which class$/$label the object $($in red$)$ belongs to:

$$ Distance measures: Manhattan distance

$$ K $=1,3,5$

Find the best K value by examining accuracy on the test set. Finally apply the best setting and predict the label for the unseen data.