Python Machine Learning camp plan

Use this page when running Python Machine Learning as a 5-day camp.

At a glance

Day	Blocks	Goal
1	Setup, Stage 1, Stage 2	Colab opens and students understand images as data.
2	Stage 3, Stage 4	Students prepare data and build the CNN structure.
3	Stage 5, Stage 6, Stage 7	Students train, evaluate, and inspect mistakes.
4	Stage 8, Stage 9, catch-up	Students improve the experiment and test new images.
5	Stage 10, rehearsal, parent demo	Students present an honest AI demo with limitations.

Minimum viable finish

A student should leave with a Colab notebook that trains, reports accuracy, and shows at least one prediction on a new image.

Coach triage

• If Colab runtime setup is slow, keep one room demo notebook ready.
• If training is slow, reduce repeated runs and focus on reading the results.
• If accuracy is low, treat it as evidence to discuss, not as failure.
• If a student is ahead, ask them to compare two experiment changes and explain the tradeoff.

Common stuck points

• Students may confuse training accuracy with real-world reliability.
• Runtime disconnects can reset state.
• Uploaded images need to be found in the notebook's file browser before prediction code can use them.