Machine Learning from Zero: Intuition, Plain-Words Maths and Code

Learn machine learning the way it is actually done, from first principles to a shipped model. Every idea starts with an everyday example and a diagram, the maths is explained in plain words with numbers you can check, and every program was run for real so the output on the page is the output it produced. Regression, classification, trees, forests, clustering, PCA, neural networks and deployment, ending in a full capstone.

Who it is for

Developers, analysts and students who can write a little Python and want to understand machine learning properly (not just call a library), including the maths, explained without jargon. No calculus or statistics background is assumed.

Before you start

• Basic Python (variables, lists, functions, running a script)
• A laptop with Python 3.10 or newer (the setup page walks you through it, no GPU needed)
• School-level arithmetic; every piece of further maths is taught in the course

What you will be able to do

• Explain in plain words how a machine learns from examples, and why it can overfit
• Use vectors, averages, probability and slopes (derivatives) confidently, and see exactly where each appears in a model
• Build and evaluate regression and classification models (linear, logistic, kNN, naive Bayes, trees, forests, boosting, SVM) with honest validation
• Prepare real messy data with pipelines, avoid data leakage, and choose metrics and thresholds from business costs
• Find groups and compress features with clustering and PCA, and build a neural network with backpropagation from scratch
• Tune, save, serve and monitor a model, and deliver a complete end-to-end project