Week 7: Decision trees and hyperparameter tuning

Learning Objectives

By the end of this week, students will be able to:

• Train and visualize a decision tree classifier or regressor
• Explain how a tree splits data using Gini impurity or information gain
• Identify overfitting in trees and control it with depth and leaf-size parameters
• Use cross-validation and grid search to tune hyperparameters systematically

Perspectival Reading

Reading: TBD

Reflection Questions

1. Decision trees are often called “interpretable” — is a tree with 50 nodes still interpretable? By whom?
2. Hyperparameter tuning optimizes a metric. What gets optimized away in the process?
3. Cross-validation gives an estimate of generalization. Generalization to what population?

Slides

View slides

Notebook Demo

Open in Google Colab (link TBD)

Lab Assignment

Week 7 Lab — GitHub Classroom (link TBD)