🐙

Stanford CS 229 Cheatsheets

Open Source

Freemachine learningeducationstanfordcheatsheetscs229open source

This repository by Afshine Amidi contains beautifully designed cheatsheets covering the full curriculum of Stanford's CS 229 Machine Learning course. Topics span supervised learning, unsupervised learning, deep learning, reinforcement learning, probability theory, and algebra refreshers — all condensed into dense, visually organized reference sheets. With nearly 20,000 GitHub stars it is one of the most widely shared ML study resources.

What is Stanford CS 229 Cheatsheets?

This repository collects the VIP cheatsheets written by Afshine Amidi and Shervine Amidi for CS 229, the machine learning course at Stanford. They are reference sheets rather than a course: each one compresses a block of the syllabus into definitions, formulas, and diagrams laid out so you can find a result quickly. The set covers supervised learning, unsupervised learning, deep learning, and reinforcement learning, plus refreshers on probability and linear algebra for the math that the rest assumes. Everything is available as PDFs you can print or keep open next to your work, with translations contributed in a number of languages. The repository also holds the source used to produce the sheets. It is not interactive and there is nothing to install. The value is entirely in the density and the layout, which is why people keep them pinned long after the exam they studied for.

Who is Stanford CS 229 Cheatsheets for?

These are for revision and for lookup, not for first contact with a topic. If you are working through machine learning material and need to recall what regularization terms look like or how an algorithm is defined, the sheets get you there faster than rereading lecture notes. They suit interview preparation particularly well, since interviewers tend to probe exactly the definitions the sheets compress. Engineers who use models daily but rarely touch the underlying math find them a useful refresher before a deeper piece of work. They are a poor first teacher. The notation assumes you have seen the derivation once, so if a topic is new, learn it elsewhere and come back to these when you need the summary rather than the explanation.

How does Stanford CS 229 Cheatsheets compare?

AI Engineering Hub is oriented toward building things, collecting worked examples and projects for retrieval systems and agents, so it answers how do I ship this rather than what is the definition. LLM Wiki is reference material for the language model era specifically, which is where most current practice sits but not where the CS 229 fundamentals live. Awesome Design.md is a curated list in the same open source tradition, pointing at resources rather than containing the material itself. The CS 229 cheatsheets differ from all three by being finished artifacts: fixed, self contained documents covering classical machine learning theory, rather than a living collection of links. Use them for the groundwork and the others for what is being built on top of it.

Compare with: AI Engineering Hub, LLM Wiki, Awesome Design.md

Stanford CS 229 Cheatsheets pricing

The catalog records this as free, and it is a public GitHub repository, so nothing is charged and no account is needed. You can read the PDFs in the browser, download them, or print them. There is no hosted service, no sign up, and no usage limit to run into. The only cost is your time, and the sheets are deliberately built to reduce that. If you want the material offline, clone the repository once and you have the complete set.

Notes

No notes have been added for this tool yet.

Stanford CS 229 Cheatsheets FAQ

Do I need to be taking CS 229 to use these?
No. The sheets stand alone as summaries of standard machine learning material and are widely used outside Stanford. They follow the course structure, so someone taking it will recognize the order, but any reader with basic calculus, linear algebra, and probability can use them as reference.
What topics do the cheatsheets cover?
Supervised learning, unsupervised learning, deep learning, and reinforcement learning, along with refresher sheets on probability and linear algebra. The refreshers exist because the main sheets assume that background. Together they track the curriculum of a general machine learning course rather than any single framework or library.
Are these useful for interview preparation?
Yes, this is one of the most common uses. Machine learning interviews often test definitions, assumptions, and trade offs between algorithms, which is exactly what the sheets compress. They will not replace practice with real problems, but they make an efficient last pass before a technical screen.
Browse all Open Source