Week 6 · Day 4 — Hyperparameter Tuning: Making Models Better Systematically

Week 6: "The Scout" | Day 4 of 5 Theme: GridSearchCV and systematic optimization — finding the best version of your model

What Are Hyperparameters?

Every ML model has settings you choose before training. These are hyperparameters.

Same as setting up your formation before kickoff:

• How deep should your defensive line be? (like max_depth in a Decision Tree)
• How many players in midfield? (like n_estimators in a Random Forest)
• How aggressive is the press? (like C in Logistic Regression)

Parameters = what the model learns from data (weights, thresholds)
Hyperparameters = what you set before training begins

Today you learn to search for the best hyperparameter settings systematically instead of guessing.