Learn Python

A free, open-source way to learn Python quickly and thoroughly, with no sign-up and nothing to pay for.

• In your browser: short interactive lessons with a built-in code editor. Python runs right in the page, so there is nothing to install. Live site: https://adutchguy.github.io/Learn-Python/
• On your own computer: the same lessons as ordinary Python files, checked with pytest.

Both tracks come from the same lesson files, so they always match.

Status: all planned features are built and the site is live. The lessons are still awaiting full review and testing report anything that is wrong or confusing. 72 core lessons take you from Start Here through Pythonic Python (iterators, generators, decorators, functools, type hints, async) and Testing & Quality (pytest, fixtures, linting, Git, packaging), with five projects: a number-guessing game, a quiz engine, an expense tracker, a bank system and a capstone where you build, test and publish a small package. Three optional tracks add 16 more lessons: Data Analysis (NumPy, pandas, charts), Automation & scripting, and Web & APIs. There is a placement test that suggests where to start, a fast track for people who already code in another language, and a spaced-review queue that brings finished lessons back at growing gaps. The site works on phones (no sideways scrolling, 44 px tap targets), has had an accessibility pass (keyboard use, screen-reader labels, WCAG AA colour contrast in light and dark mode, reduced-motion support), and can be installed as an app and used offline. See docs/PLAN.md for the roadmap.

Optional tracks

After the core path (or any time you want a practical detour) there are three optional tracks. They are not part of the core progress count, the placement test or the fast track.

• Data Analysis: NumPy, pandas (selecting, grouping, cleaning), a small sales-analysis project and Matplotlib charts.
• Automation & scripting: command-line programs with argparse, files and folders in bulk, subprocess, retries and scheduling, and a file-organiser project.
• Web & APIs: how HTTP works, JSON APIs, calling a web service and building a tiny JSON API with only the standard library.

In the browser, the Data Analysis lessons download NumPy and pandas the first time (a few megabytes). Lessons that need real network, subprocesses or a chart you can see are marked as needing your own computer.

Use it offline

On the home page, Download for offline saves Python itself (about 15 MB with NumPy and pandas). After that the whole course, including running and checking code, works without a connection, and your browser can install it as an app. Your progress always stays in your browser.

How learning works here

Every lesson takes 10 to 15 minutes: a short explanation, a worked example, then an exercise that is checked automatically with plain-language feedback. Hints unlock one at a time, and the solution unlocks after a few honest attempts. The ideas behind the design (active practice, spaced review, early error-message skills, projects) are described in docs/PLAN.md.

Work on your own computer

You need Python 3.11 or newer.

git clone https://github.com/Adutchguy/Learn-Python.git
cd Learn-Python

python -m venv .venv
# macOS / Linux:
source .venv/bin/activate
# Windows (PowerShell):
.venv\Scripts\Activate.ps1

pip install -r requirements.txt

# Only if you want the optional Data Analysis track (NumPy, pandas, Matplotlib):
pip install -r requirements-tracks.txt

Each lesson is a folder under exercises/:

exercises/01-foundations/01-hello-world/
├── LESSON.md          <- read this first
├── exercise.py        <- edit this file
└── test_exercise.py   <- the checks (read-only)

The learn helper keeps track of where you are, so you never have to hunt for the next lesson. Run it from the repository root:

python -m learn                 # what to do next
python -m learn check           # run the checks for your current lesson (a pass marks it complete)
python -m learn hint            # reveal one hint at a time
python -m learn status          # your progress, module by module
python -m learn list            # every lesson, with the finished ones ticked
python -m learn tracks          # the optional tracks; "python -m learn next data" continues one

check and hint also accept a lesson number from list, or a word from its name (python -m learn check variables). Your progress is saved in .learn-progress.json, which Git ignores.

Prefer plain pytest? That works too:

pytest exercises/01-foundations/01-hello-world

Tip: work on your own branch (git switch -c my-answers) or in a fork, so your answers stay yours when you pull new lessons.

Repository layout

Folder	What it is
content/	Source of truth. One folder per lesson: lesson.md, starter.py, solution.py, test_exercise.py. Also placement/questions.json (placement-test questions) and review/questions.json (spaced-review questions)
exercises/	Local track, generated from content/ (do not edit by hand, except conftest.py)
web/	The website (Astro) and the in-browser Python runner (Pyodide)
learn/	The python -m learn command-line guide for the local track
tools/	Python scripts that validate lessons, generate exercises/ and scaffold new lessons (new_lesson.py)
docs/	Plan, content style guide

Contributing

Lessons, fixes and ideas are welcome. See CONTRIBUTING.md and the content style guide.

Quick start for contributors:

pip install -r requirements.txt -r requirements-tracks.txt
python tools/new_lesson.py "Title" --module 03-data-structures   # start a new lesson from a template
python tools/validate_content.py          # are all lessons well formed? (also runs the placement and review questions to verify their answers)
python tools/build_exercises.py           # regenerate exercises/ after editing content/
python tools/check_solutions.py           # every solution passes, every starter fails
python -m pytest tools/tests              # tests for the tooling itself

cd web
npm ci
npm test                                  # runs every lesson through the browser runner on real Pyodide
npm run dev                               # local website at http://localhost:4321/Learn-Python/
npm run build

Web development needs Node.js 22.23.4 or newer.

Publishing the website

The site deploys automatically from main using GitHub Actions. One-time setup: in the repository on GitHub, open Settings → Pages → Build and deployment and set Source to GitHub Actions. The site then appears at https://adutchguy.github.io/Learn-Python/ (it is live now).

Licenses

• Code: MIT
• Lessons and other learning content (content/, and the lesson files generated from it in exercises/): CC BY-SA 4.0