Knowledge check

Completed

Check your knowledge

1.

Decision trees are classification models that…

Typically have excellent generalizability to test sets

Can only make binary (true/false) predictions

Typically overfit their training data

2.

What does selecting model architectures mean?

Picking a model’s parameter values before training

Deciding how a model is structured

Selecting a 32-bit or 64-bit compilation option

3.

Hyperparameter selection can help give optimal models and…

Refers to selecting values that change how the learning process works

Often requires experimentation to get right

All of the above

You must answer all questions before checking your work.