Machine Learning: final assessment

This covers the whole course. Take it after the last module.

Final quiz

1. Machine learning turns data and answers into...
2. Which task is classification?
3. A z-score of 2 means the value is...
4. Which data is used once, at the very end?
5. What is the Euclidean distance between (1, 1) and (4, 5)?
6. Why does k-NN need scaled features?
7. What makes the log loss large?
8. Raising a classifier's threshold usually...
9. Laplace smoothing in Naive Bayes prevents...
10. Which gain does a decision tree use to pick a split?
11. Random forests mainly reduce...
12. Gradient descent on (w − 5)² from w = 0 with learning rate 0.5 reaches 5 in...
13. Scaling with the mean and σ of the whole dataset before splitting is an example of...
14. Why does k-means always stop?
15. What does a hidden layer let a network do that one neuron cannot?
16. An ensemble's members all make the same mistakes. A majority vote will...

Practical assessment: An end-to-end classifier

Without hints or libraries, write a program that trains and evaluates a classifier on a small table of data you create, from raw rows to a final test score.

Final project

A spam classifier, compared three ways: Build a spam filter for short messages with three different methods and decide which one you would deploy, with evidence.

How it is evaluated

To complete this course