Data, Features and Labels
Every machine learning model starts with a table of examples. The columns the model looks at are called features, and the column it tries to predict is called the label. Choosing good clues is half the job.
New words
- Dataset
- A table of examples that a model learns from.
- Row
- One example in the table, such as one fruit.
- Feature
- A clue the model looks at, such as colour or weight.
- Label
- The correct answer the model must learn to predict.
- Leakage
- A clue that secretly gives away the answer.
- Pixel
- One tiny dot of a picture, stored as a number.
What you will learn
- A spreadsheet is all you need
- Rows, features and labels
- What makes a good feature?
- Try it: choose the clues
- Garbage in, garbage out
- But what about text and pictures?
- Where you see it every day
- Explain it to a friend
- Key takeaways
10 quiz questions at the end.
The full lesson, with interactive demos and the quiz, opens after you sign in with Google. It is free.