Assessment: ML Fundamentals

Explain

• What is the difference between supervised and unsupervised learning?
• Why do we need a validation set?
• When is accuracy a bad metric?
• What is data leakage?
• How do you detect overfitting?

Build

• Design a sales prediction pipeline from raw data to API.
• Choose metrics for regression and justify them.
• Choose a baseline and explain why it is useful.

Debug

• A model performs perfectly on validation but fails in production. What do you check?
• A model has high recall but low precision. What does that mean?
• A training set has missing values. What are safe ways to handle them?

Completion Gate

• Can explain all answers without notes.
• Can show code evidence.
• Can show evaluation evidence.