Machine LearningLesson 11Free preview

ML Testing

This lesson is part of A.I. · All levels

Summary

After training, the model must be tested with unprecedented data to measure the actual capability. The test results are usually expressed as a percentage of accuracy, which is the number of correct predictions compared to the total experiment. Honest testing uses data testing separate from data training.

Key points

Testing measures the capabilities of a model on unprecedented data.
Accuracy = the number of correct predictions / total predictions × 100%.
Data testing has to be separate from training data so the results are honest.
High training accuracy but low testing means overfitting.

Full lesson

Code examples and quizzes

Full lessons, code examples, and quizzes are available to subscribers. 5 questions.