8.5 Lesson 4 - Training and Test Data

What do machine learning models learn by finding in training data?

• Random numbers

• Patterns

• Colours

• Distances

What does training data include to help the model learn?

• Incorrect answers

• The correct answers

• Only questions

• Only images

Why is test data used after training a model?

• To confuse the model

• To check how well the model performs on new examples

• To make the model slower

• To delete the model

What is the main purpose of using test data?

• To train the model further

• To measure the accuracy of the model

• To make the model forget

• To increase the size of the model

What might happen if the training data is too small?

• The model will always be correct

• The model might make mistakes with new data

• The model will run faster

• The model will use less memory

If the training data is too similar, what is a possible result?

• The model will be perfect

• The model might not perform well on new data

• The model will use more electricity

• The model will be more colourful

What is the role of correct answers in training data?

• To confuse the model

• To help the model learn what to recognise

• To make the data larger

• To slow down the training

What do we call the data used to check the model’s performance after training?

• Training data

• Test data

• Practice data

• Sample data

Why is it important for test data to include examples the model has not seen before?

• To make the test easier

• To properly measure how well the model can generalise

• To make the model faster

• To reduce the size of the data

What does measuring the accuracy of a model help us understand?

• How colourful the data is

• How well the model works

• How much the model weighs

• How fast the model runs