What it does

FreeML is a hand-picked index of courses, textbooks, and video lectures covering machine learning, NLP, deep learning, reinforcement learning, systems, and interview prep. The author split it into two halves: what they actually studied over two years, and what they planned to tackle next. It is explicitly aimed at people outside CS who want a structured, mostly-free entry into data science.

The interesting bit

The value is not comprehensiveness — the author admits you can find bigger lists elsewhere. It is the curation: each section includes a short “Comments” note explaining which course to take first, which two are interchangeable, and which one is “so hard for me but covers almost everything.” That editorial voice turns a link dump into a plausible study plan.

Key highlights

• One-month crash course suggestion: Python, Stanford Statistical Learning or Andrew Ng’s Coursera course, Ng’s deep learning specialization, Keras, and a Stanford database course.
• Heavy Stanford bias: CS229, CS224n, CS231n, and Statistical Learning all feature prominently.
• Includes adjacent skills often missing from ML syllabi: Docker, Hadoop, Spark, analytics books, and quant/time-series sections (latter truncated in source).
• Interview section aggregates question lists from KDnuggets, Analytics Vidhya, and RPubs.
• Author notes which courses offer free certificates and which follow specific textbooks closely.

Caveats

• Several Udemy courses listed under “Systems” are not free; the “mostly free” claim is looser there.
• Some sections (Bayesian, time series, quant) were truncated in the README source, so their depth is unclear.
• Link rot is inevitable: a few YouTube playlists and course pages will age out without maintenance.

Verdict

Grab this if you are a non-CS background learner who wants a vetted, opinionated syllabus rather than an overwhelming awesome-list. Skip it if you already know which Hinton course to take and own a shelf of Hastie textbooks.