Train-test split

6 practice questionsData ScienceStep-by-step solutions

Try a few

A common ratio for splitting data into training and test sets is:

\frac{80}{20}

\frac{100}{0}

\frac{1}{99}

\frac{50}{50}

only

Why is data split into separate training and test sets?

A.To make the dataset smaller

B.To estimate how the model performs on unseen data

C.To remove missing values

D.To increase the number of features

The main reason to hold out a test set is to:

A.Speed up training

B.Estimate how the model generalizes to unseen data

C.Increase the training data

D.Remove outliers

All 6 questions with worked solutions, an AI tutor that explains every step, and games that make the drilling stick. Free to start.

No card needed · 10 free AI questions daily