ML Foundations & Evaluation

What "learning" means once you write it down formally, and how a trained model is judged. Everything later in the course assumes these two things: the ⟨T, P, E⟩ framing of a learning task, and the confusion matrix the metrics are read off.

The pages

• Machine Learning Foundations - intuition, the two phases, the seven steps, the three broad categories, and how a learning system is designed.
• Performance Metrics - the confusion matrix and everything derived from it: accuracy, precision, recall, F1, and when each one lies.
• Cheat Sheet - the whole topic on one page.