Homework 2

Classification

Learning Objectives

• Derive the gradients needed to fit a logistic regression model.

• Translate mathematical quantities into code by implementing logistic regression and loss function evaluations.

• Learn about dataset documentation for ML transparency and accountability.

• Practice with a classification task for a real-world classification problem.

• Perform error analysis in the context of ML-supported decision-making.

Tip

I recommend that you test your functions in Part 1 with the Gradescope autograder before moving on to Part 2. That way you’ll have more certainty that your functions are working correctly for Part 2.

If you are uncertain or have questions about any portion of the homework, please do come to my office hours, Bhargavi’s TA hours, or reach out on Ed!

Note

This homework is due Wednesday 3/4 at 11:59pm.

• Part 1: Foundations

• Part 2: Models