Question why do we divide our data into a training and test set? Should we use our test se to find the best possible number of neighbors for a k-NN classifer? What is the point of a test set and why do we only want to use the test set once? Explain. Write your answer here, replacing this text. Question 8 Why do we use an odd-numbered k in k-NN? Explain. Write your answer here, replacing this text