Block 80: Mini-Project: Complete ML Notebook

Concepts

• Problem definition → data → preprocessing → modeling → evaluation → insights
• Comparing multiple models
• Communicating results clearly

Code Examples

See exercise below.

Exercise

Choose a dataset (Titanic, diabetes, or similar). Perform: EDA (3 plots) → preprocessing (Pipeline) → train 3 models → evaluate with cross-validation → report best model with all metrics. Write a 'model card' cell: describe what the model does, its limitations, and how it should be used.

Homework

Reflection: Write your ML project checklist — the 10 steps you'll always follow.