What a model actually learns from data, the three main kinds of problem, and why a model that looks perfect often isn't.
A plain-language introduction to machine learning for anyone who works alongside it. No maths beyond averages and no code: just the ideas you need to ask good questions of a model, a vendor or a data team.
By the end you will be able to describe any machine learning project in terms of its data, its task and how it was tested.
4 lessons · 56 min
Rules written by people versus patterns found in examples.
Classification, regression and clustering, with real datasets as examples.
Why you always keep some data back, and what each split is for.
When a model learns the noise instead of the pattern.
No ratings yet.
Sign in to leave a rating. Sign in
Use @handle to bring somebody in, and #tag to file it.
Nothing here yet. Ask a question, or say how you used it.
Sign in to join the discussion. Sign in
2 min read
2 min read
2 min read
2 min read
Turning things into numbers that carry meaning, and what you can do with them.
Attention, tokens and the architecture behind current language models — enough to reason about behaviour.
Model, parameter, inference, context, temperature — the terms, used correctly.