1 · First steps
Hello world, variables, and a peek at where this track is heading.
●Track 1 · live
Python Mastery
From your first print statement to NumPy and Pandas — the language ML is written in.
A complete Python track designed as the foundation for everything that follows. Syntax, data structures, functions, OOP, the standard library, and the scientific stack (NumPy, Pandas, Matplotlib, Seaborn). Each lesson runs Python in your browser via Pyodide and offers Arena challenges for stdin/stdout practice.
59 / 60 lessons17 chapters
2 · Numbers, strings, booleans
The four primitive types in detail — the building blocks every Python expression rests on.
3 · Control flow
Branching with if/elif/else and iterating with for/while — the loops every training script depends on.
4 · Data structures
Lists, tuples, sets, dictionaries — the four containers every Python program uses to organise data.
5 · Functions
Build your own abstractions — functions, lambdas, map/filter, closures, decorators. The toolkit every ML codebase relies on.
6 · Modules & standard library
Import, package, and lean on Python's batteries — datetime, pathlib, json, itertools, functools, random.
7 · Files & exceptions
Read and write files with pathlib, then handle errors gracefully — every data pipeline needs both.
8 · Object-oriented Python
Classes, inheritance, polymorphism, encapsulation, abstraction, magic methods — the way every framework you'll touch is built.
9 · Iterators, generators, advanced decorators
Lazy evaluation and meta-programming — the patterns that let you stream gigabytes of data and add cross-cutting behaviour without touching the function body.
10 · NumPy
The packed-array library that turns Python into a numerical powerhouse. Arrays, broadcasting, and the linear algebra every ML model depends on.
11 · Pandas
Tables in Python — Series, DataFrames, groupby, merge. The library every ML notebook reaches for in line one.
12 · Visualization
Matplotlib and Seaborn — the plotting libraries every ML notebook leans on for loss curves, scatter plots, and correlation heatmaps.
13 · SQLite & data persistence
Persist data in a single file with sqlite3 — and round-trip DataFrames in one line with Pandas.
14 · Logging
Replace print() with structured, level-aware logging — the bare minimum for any pipeline that runs longer than 30 seconds.
15 · Concurrency & parallelism
Threads, processes, executors — when to reach for each, and the GIL gotchas every Pythonista learns the hard way.
16 · Memory & idiomatic Python
Reference counting, the cyclic GC, __slots__, and the small habits that separate beginner Python from production Python.
17 · The scikit-learn toolkit
The estimator API, preprocessing, pipelines, model selection, and persistence — the bridge from Python into every ML track.