Skip to content

OPEN LEARNING / MACHINE LEARNING

Less mystery.
More “I get it.”

Learn how machines find patterns, make predictions and get things wrong. Small experiments. Clear explanations. A purpose beyond the model.

Free to explore · No code needed to start · Your pace

THE PATTERN LAB01 / 06
Can yesterday help us plan tomorrow?
Synthetic meal demand and a fitted regression line
↳

Move the line. Find the pattern.

Explore the regression lesson ↗

6 lessons ready to explore

12 lessons in the course plan

0 installs to get started

Built around original examples
for community impact

BEFORE THE ALGORITHMS

A useful question
comes first.

A community kitchen wants to plan enough meals without unnecessary waste. Should it predict a number, flag a busy day, or use a simple rule?

We use this fictional setting throughout the foundations. You’ll learn the modelling ideas and practise asking whether the result actually helps.

Find your starting point ↗

YOUR LEARNING PATH

One idea. One experiment.
One step forward.

Build understanding first. Open the technical layer when you’re ready.

0 of 6 lessons completed

Six foundations · About 85 minutes · No coding required

Start lesson 1 ↗

One learning path for everyone. Each lesson includes an optional Go deeper section with maths, Python and further reading.

↳

The first six lessons are available now. The second half is a proposed next stage. Progress is saved on this device. A lesson completes when you’ve tried its experiment and answered its questions.

HOW YOU’LL LEARN

Understanding you can put to work.

01 / MAKE A GUESS

Begin with your intuition.

Predict what will happen before you touch the controls. No specialist vocabulary needed.

02 / CHANGE SOMETHING

See the evidence move.

Adjust a model, inspect its mistakes and compare it with a simpler alternative.

03 / EXPLAIN WHY

Take away a useful idea.

Check your understanding, read the technical layer or try a small project.

PUT IT TOGETHER

One small project.
A complete explanation.

Turn the six lessons into a project brief, an evaluation plan and an honest model note. The worksheet runs in your browser; an optional original Python notebook takes the workflow from synthetic data to a saved pipeline.

The notebook uses its own 180 fictional days. Python setup is required; see the notebook guide. It has been tested and runs from start to finish.

QUESTIONS AND IDEAS WELCOME

Stuck, curious, or want to help?

There’s no such thing as a silly question. Ask us about anything in these lessons, suggest an improvement, or collaborate with us on a lesson or a real project for a charity or research team.

KEEP YOUR CURIOSITY GOING

Good material.
A clear next step.

Our original lessons are a starting point. Explore a curated shelf of courses, visual explanations, reference guides and practice data.

Find your next resource ↗