brightlab.Made for curious mindsBrightlab
Foundation · Classification
Which basket will your shape go in? Change its size and try again.
Go to the investigation ↓
Investigate with Wattle the wombatA sorter poster with a named property and three tested shapes.
Which basket will your shape go in? Change its size and try again.
A sorting rule tells us which property to look at. This gate looks only at size. It does not use colour or decide which shape is nicer. When we move the gate, we change the rule, so a shape can change groups without changing itself. People decide which properties matter for a task.
Before this lesson: Compare two objects and say which is bigger. No prior AI lesson is required.
A sorting rule tells us which property to look at. This gate looks only at size. It does not use colour or decide which shape is nicer. When we move the gate, we change the rule, so a shape can change groups without changing itself. People decide which properties matter for a task.
Compare two objects and say which is bigger. No prior AI lesson is required.
Cut circles, squares and triangles in two colours and three sizes. Place two hoops or paper circles on a desk. Rehearse the browser gate; note that a size exactly on the gate goes to the large group.
ACARA V9 AI curriculum connection ↗ · Technologies ↗
A complete teaching sequence · 35 minutes
Children place paper shapes in two groups. Invite two different sensible arrangements.
Ask: “Could both groups be sensible?”
Listen for: “Yes, one uses colour and one uses size.”
Show the size number of one shape and the gate. Hide its destination until a child predicts.
Ask: “Does this gate look at blue?”
Listen for: “No; it looks at how big it is.”
Move the size gate in single steps. Count each group and follow the same highlighted shape. Record a before and after drawing.
Ask: “Did the shape grow when it crossed groups?”
Listen for: “No; the gate moved.”
Use two same-size shapes of different colours, including a size exactly equal to the gate.
Ask: “Where should an equal-size shape go? How do we know?”
Listen for: “We need a rule for equal; this gate says large.”
Ask pairs to invent a sorter for envelopes that fit a small box. Choose a threshold, test three objects and revise it once.
Ask: “Which property helps with fitting in the box?”
Listen for: “Size, not colour.”
Children complete a two-picture rule poster.
Ask: “What stayed the same in your experiment?”
Listen for: “The shapes; we changed the gate.”
An object has just one correct group.
Two equal-size shapes of different colours go to the same group. A boundary-size shape needs an explicit equality rule.
Set a size gate for a small-box sorter and test a shape on each side and exactly at the boundary.
Ask children to explain an unchanged object’s changed destination. Re-teach by physically moving a dividing string if they say the object changed size.
| Criterion | Beginning | Secure | Extending |
|---|---|---|---|
| Using a rule | Sorts by changing properties | Uses the size rule consistently | Explains the equality case |
| Explaining groups | Treats groups as permanent | Connects a changed group to a changed gate | Proposes another useful rule for a different task |
Use only three shapes with clearly different sizes; compare by overlaying paper shapes. Accept oral evidence.
Design a two-stage rule using size and shape. Find an object on which size-only and two-stage rules disagree.
Use objects, not children, as sorting examples. Do not sort classmates by bodies, ability or personal characteristics.
Shared device? Turn remembering off. A project file lets you continue on another device.
Your browser is the laboratory
Which basket will your shape go in? Change its size and try again.
Six fixed shapes have sizes 2, 3, 5, 5, 7 and 9. The test shape is size 5. A gate of 5 sends sizes below 5 to small and sizes equal to or above 5 to large. Colour is ignored.
The experiment opens after your prediction.
There is no penalty for being surprised.
Apple Silicon · PyTorch MPS
Evaluate a large synthetic shape population across every possible gate; display how many destinations change at each step.
Teacher-operated extension: students predict and interpret the visual report together.
The bundle contains lesson-specific working code, a configuration file, a reactive notebook, a deterministic CPU check and hardware setup instructions. Acceleration is reported only after a tensor operation and result read-back succeed.
Download Mac Studio investigation ↓unzip f-sort-mac-pathway.zip -d f-sort-mac cd f-sort-mac bash setup-mac.sh source .venv/bin/activate python experiment.py --device mps --output results marimo edit notebook.py
CPU and available-device execution status is recorded in the downloaded README and validation report. DGX Spark execution requires that hardware; static validation alone does not establish GPU compatibility or performance. The browser lesson remains fully available without this extension.
NVIDIA DGX Spark · PyTorch CUDA
Evaluate a large synthetic shape population across every possible gate; display how many destinations change at each step.
Teacher-operated extension: students predict and interpret the visual report together.
The bundle contains lesson-specific working code, a configuration file, a reactive notebook, a deterministic CPU check and hardware setup instructions. Acceleration is reported only after a tensor operation and result read-back succeed.
Download DGX Spark investigation ↓unzip f-sort-dgx-pathway.zip -d f-sort-dgx cd f-sort-dgx bash run-dgx.sh # Open the localhost notebook URL printed by the container.
CPU and available-device execution status is recorded in the downloaded README and validation report. DGX Spark execution requires that hardware; static validation alone does not establish GPU compatibility or performance. The browser lesson remains fully available without this extension.
A sorter poster with a named property and three tested shapes.
Download the editable handout →