Machine learning you can see and change

Teach a model. Watch it learn.

Start with data and first principles, then build models you can explain, evaluate and deploy. Change the inputs, run an algorithm, and see how the prediction moves.

Try the interactive labsBrowse the course
14 guided modules3 model simulationsPython from the start
model / neural-networklive model
Interactive neural network preview
Inputs2 featuresHidden layer3 neuronsPrediction0.50
Interactive model bench

Change the data. Run it. Learn.

Each simulation runs in your browser. Move a point, change a parameter, or advance the algorithm one step at a time.

Neural network

Forward pass · sigmoid activation

Change two input features and follow their signals through a small network to its prediction.

Neural network diagram
Hidden unit 1
—
Hidden unit 2
—
Output score
—
Adjust the features, then run the network.

k-nearest neighbours

Classification · distance

Move the query point. Its nearest labelled examples vote on which class it belongs to.

k-nearest neighbours decision map
Predicted class
—
Neighbours
—
Class votes
—

Tap or drag inside the chart to move the query point.

The shaded areas show the current k-NN decision regions.

k-means clustering

Unsupervised learning · centroids

Assign points to the nearest centre, recalculate each centre, and watch the clusters settle.

k-means clustering plot
Clusters k
3
Iteration
0
Centre movement
—

Step through the assignment and update cycle.

Three unlabelled groups are ready to cluster.
Understand before you deploy

From the first dataset to models you can trust.

Learn the maths and Python together, then use your knowledge on practical problems in vision, text, forecasting and more.

01 / SEE IT

Build intuition first

Watch data, distance, weights and loss shape the model’s decisions before you dive into code.

02 / CODE IT

Work with real Python

Use NumPy, pandas, scikit-learn and small neural networks in lessons you can edit and run.

03 / CHECK IT

Know when to trust it

Measure performance, spot leakage and bias, and monitor what changes after deployment.

Your learning path

Fourteen steps to applied machine learning.

Start with the lessons
  1. 01Python foundations for ML
  2. 02Math & statistics for ML
  3. 03Data analysis with NumPy & pandas
  4. 04Understanding machine learning
  5. 05Supervised learning: regression
  6. 06Supervised learning: classification
  7. 07Unsupervised learning
  8. 08Model evaluation & optimisation
  9. 09Advanced machine learning
  10. 10Deep learning
  11. 11Applied ML projects
  12. 12Deploying models
  13. 13Responsible AI & data privacy
  14. 14Monitoring & MLOps
Connect a model to the physical world

Ready to work with sensors, circuits and edge AI?

Take your model beyond the notebook. Explore electronics, Arduino, control systems and smart devices.

Explore electronics & hardware