Free
Google's rebuilt ML Crash Course — regression, classification, neural networks, embeddings and LLMs, with interactive visualisations and per-module badges.
Intermediate10+ hours
Learn Data Science & Machine Learning with the Data Science & Machine Learning tutorials and online courses.
Two things sit under this heading and they aren't the same job. Data science is mostly questions, cleaning and communication; machine learning is mostly models. The overlap is real, and Python sits underneath both, which is why our shelf starts there rather than with an algorithm.
Six tutorials, five of them free. They span the whole range, from a three-hour notebook course to a deep learning curriculum that expects you to train something in week one.
Practical Deep Learning for Coders from fast.ai is top-down: Jeremy Howard has you training a working model in lesson one and explains the maths afterwards. The Machine Learning Specialization from Andrew Ng builds upward instead — regression, then neural networks, then trees, clustering and recommenders, with the reasoning first. Both work. Which suits you depends on whether being handed something that works before you understand it feels motivating or unbearable. Ng's is subscription-only on Coursera; fast.ai costs nothing.
Kaggle Learn: Intro to Machine Learning is six lessons in hosted notebooks — decision trees, validation, overfitting, random forests — and about three hours end to end. It's the cheapest way to find out whether you like this. Google's rebuilt Machine Learning Crash Course is the step up, now running through embeddings and LLMs with interactive visualisations. And keep StatQuest with Josh Starmer bookmarked for whenever a term stops making sense; it's a reference library rather than a course, and it's the best explanation of p-values you'll find.
Six entries is not a curriculum, and the gaps matter. Python for Everybody is doing all the data-wrangling work here on its own — we have no dedicated pandas course and no SQL course, and SQL is the single most requested skill in data job listings. Nothing on statistics as a subject in its own right, nothing on R, nothing on visualisation, and nothing on getting a model into production. fast.ai's part one was recorded in 2022, which in this field is a while ago.
Do Kaggle's three hours first. If you're still interested afterwards, pick fast.ai or Ng and commit to finishing — then go find a SQL course somewhere else, because we can't give you one yet.
Google's rebuilt ML Crash Course — regression, classification, neural networks, embeddings and LLMs, with interactive visualisations and per-module badges.
Six lessons, about three hours, in hosted notebooks — decision trees, validation, overfitting and random forests with pandas and scikit-learn.
DeepLearning.AI
Andrew Ng's rebuilt three-course ML specialization in Python — regression, neural networks, trees, clustering and recommenders. Subscription-only, not free.
Jeremy Howard's top-down deep learning course — train a working model in lesson one, PyTorch and fastai underneath. Part 1 was recorded in 2022.
Dr. Charles Severance
Dr. Chuck's data-first Python course — files, APIs, scraping, SQL and visualisation, with free autograded assignments and a gradebook at py4e.com.
Josh Starmer
Josh Starmer's drawn, deliberately goofy explanations of statistics and ML — p-values through transformers. A reference library, not a course.