Learn ANN first. A living ML wiki next to the teaching wiki: algorithms, neural nets, language models, and the Python I study with — written as I learn it, for anyone who wants to follow.
A path you can walk today, using pages that already exist. Evolution articles and most classroom labs are still to be written — I will not pretend they are here.
The decision / error / optimisation loop, and the four learning paradigms.
Linear regression — still a short definition page, and the seed for the other classic algorithms.
The strongest ML article already on this wiki: structure, types, and how a network learns.
The former home page: the formula, videos, and executable Python (spiking neuron, XOR perceptron).
How deep networks, GANs and transformers sit inside machine learning.
A comparison table is written; the child pages (LLM, GPT, Transformer) are honest outlines, not finished articles.
Named, not yet taught. Start with the section outline; the RAG page is in progress at its original URL.
Counted from this repository on 10 November 2026. The page total includes new section outlines written so every agreed section has a home, plus its own Labs and Resources pages. It is not 71 finished articles.
Existing articles stay at their URLs. New sections begin with an honest outline rather than empty boilerplate. About is not a section here — it lives on the teaching wiki.
What machine learning is, and the Helsinki Python Programming MOOC 2024 track. Python lives here, not in a side menu.
Today this is the Big Data overview (the five Vs). Cleaning, splits and pipelines are still to be written.
One overview of algorithm types, then nine short definition pages: linear and logistic regression, naïve Bayes, SVM, trees, k-NN, forests, boosting, k-means.
How well a model does, and by which measure. An outline only: splits, metrics, overfitting, cross-validation.
ANN is the first chapter: introduction, mathematical model, and the relocated home (formula, videos, Python demos). Then a deep-learning overview.
A written comparison of generative models. The LLM, GPT and Transformer child pages are outlines, ready to be filled.
Retrieve, then generate — and later the agent loop. Named in the generative table; not yet taught.
Using a model inside a real program. No articles yet, only an honest outline.
How the code is structured, how a run is repeated, train versus serve. Outline only.
Mirrors Training on albertprofe.dev. No live AI course is documented here yet — no invented sprint or calendar.
The teaching wiki this site sits next to: Java SE, Spring Boot, React, DevOps, Linux, Scrum, labs and course editions.
If you have thirty minutes, start with the relocated home article: the formula, the videos, and the Python demos that used to greet every visitor.
Go to ANN: model and demos →