Skip to content

1. Meet Teachable Machine

Teachable Machine is a free, browser-based tool that lets you train a real machine learning model by example — the exact process from Unit 1, Lesson 3, done by hand.

Three project types

Project Classifies
Image Pictures, into categories you define
Audio Short sound clips, into categories you define
Pose Body positions from a camera feed, into categories you define

This unit uses Image projects, with uploaded files rather than live capture — see the warning below.

The core workflow, every time

  1. Create classes — the categories you want the model to tell apart.
  2. Give each class several example files.
  3. Click Train Model — Teachable Machine trains it right there in your browser.
  4. Test it with new examples it hasn't seen.

That's the entire loop. Every later lesson in this unit is a variation on these four steps.

Use uploaded files, not live webcam capture

This lab's remote desktops don't support webcam passthrough at all (verified directly against the installed software), and separately, photos of yourself or classmates are personally identifiable information. For every project in this unit, click Upload to add existing image files rather than using the webcam button — and choose non-identifying subjects: objects, hand gestures framed to exclude faces, drawings, household items.

Try it

Open Teachable Machine and start a new Image Project, just to see the interface — don't train anything yet.

Tip

You don't need an account or to sign in to use Teachable Machine at all — it's free with zero setup.

Next: Training an Image Classifier →