0.3 Free compute: Colab, Kaggle, HF Spaces
You can get a free GPU when your laptop is not enough.
Before:nothing requiredUnlocks:01. Mathematics for Machine Learning02. Python — Basics to Advanced
Deep learning eventually wants a GPU, and Colab, Kaggle and Hugging Face Spaces hand one over free — each with limits worth knowing before they bite: sessions die, quotas reset weekly. This early topic exists so hardware never becomes the reason to stop. The judgement it plants is when to leave notebooks: they are for exploration, and part of growing through this course is noticing the moment a project deserves real files instead.
Work through these
Colab runtimes, GPU/TPU limits and session death
Free hosted notebooks give you a graphics processor, with limits on session length and availability that will interrupt you eventually. Knowing the limits lets you save your state before they bite.
Kaggle notebooks and weekly GPU quota
A second free provider with a different allowance, which is worth having as an alternative when the first is unavailable. Between them most learning workloads are covered without spending anything.
Mounting data; when to stop using notebooks
Getting your data into a hosted session, and recognising the point where a notebook has outgrown its usefulness and the work belongs in a repository. That transition is a real turning point in this subject.
Sign in to keep your progress.
Free resources
We haven't checked most of these for screen reader use yet.
- FreeHarvard CS50 · CourseCS50's Introduction to Programming with Python
- FreeGitHub Skills · TutorialIntroduction to GitHub
- FreePython.org · DocsThe Python Tutorial
Using a screen reader? 1 resource
Links last checked 29 Aug 2026.
Stuck here?
Ask a mentor. A real person answers, and they can see exactly which topic you're on. Usually within a couple of working days.
Checking your session…
Topics shown in module order.