G4-5 AI with SCRATCH Session 4

Choose the options that are important to make sure the training data is diverse

• Take very few pictures

• Take many pictures

• Take very few types of pictures

• Take many types of pictures

After training, you can give the teachable machine model new, unseen inputs and it will then "predict" which category it belongs to. 

• True

• False

You cannot train Teachable Machine to recognise sounds.

• True

• False

What is the correct sequence in which Teachable Machine can be trained?

• Train Teachable Machine > Collect Diverse Input Data for all classes > Test the model with new data

• Collect Diverse Input Data for all classes > Train Teachable Machine > Test the model with new data

• Test the model with new data > Collect Diverse Input Data for all classes > Train Teachable Machine

Diverse training data leads to ____________________.

• Lower prediction accuracy

• Higher prediction accuracy

• The same prediction accuracy