Lesson 1.2

2 · Variables & primitive types

Names that hold values, and the four primitive types every Python program uses.

⏱ ~9 min· intro· python / variables / types

Big idea

A variable is a label pointing at a value — assign with `=`, never `==`.

Without this:

Without variables, you cannot reuse a value, name an intermediate result, or build anything beyond a one-shot script.

Python is dynamically typed: a variable's type is the type of whatever value it currently holds. Bind a value to a name with =.

Python (in browser)

Four variables, four different primitive types. Run it and check the type() output.

main.pypython

Loading editor…

Ready — click Run to execute in your browser.

Python runs entirely in your browser via Pyodide (~6 MB on first Run, cached after).

f-strings are the modern way to interpolate values into text. They start with f and embed expressions in {}.

Python (in browser)

main.pypython

Loading editor…

Ready — click Run to execute in your browser.

Python runs entirely in your browser via Pyodide (~6 MB on first Run, cached after).

Shebang line + module docstring + a top-level CONSTANT in UPPER_SNAKE_CASE. The conventions you'll see in real codebases.

Mental model: a variable is a label pointing at a value. — Reassignment moves the arrow; the old value gets garbage-collected when nothing points at it.

🧮

Foundations

Why Math for Machine Learning

If 'function' and 'value' feel hand-wavy, read the MML opener — it grounds the mental model.

Key takeaways

Variables are bindings (name → value), not boxes. Reassignment changes what the name points at.
The four primitives — `int`, `float`, `str`, `bool` — cover most simple values you'll see.

Where this shows up in ML

Hyperparameters are variables (`learning_rate = 0.01`); model state is held in variables (`model = LinearRegression()`).

If you remove it: Without named state, every experiment becomes a wall of literal numbers — unreadable and untunable.

Lesson 1.2

2 · Variables & primitive types

Names that hold values, and the four primitive types every Python program uses.

⏱ ~9 min· intro· python / variables / types

Big idea

A variable is a label pointing at a value — assign with `=`, never `==`.

Without this:

Without variables, you cannot reuse a value, name an intermediate result, or build anything beyond a one-shot script.

Python is dynamically typed: a variable's type is the type of whatever value it currently holds. Bind a value to a name with =.

Python (in browser)

Four variables, four different primitive types. Run it and check the type() output.

main.pypython

Loading editor…

Ready — click Run to execute in your browser.

Python runs entirely in your browser via Pyodide (~6 MB on first Run, cached after).

f-strings are the modern way to interpolate values into text. They start with f and embed expressions in {}.

Python (in browser)

main.pypython

Loading editor…

Ready — click Run to execute in your browser.

Python runs entirely in your browser via Pyodide (~6 MB on first Run, cached after).

Shebang line + module docstring + a top-level CONSTANT in UPPER_SNAKE_CASE. The conventions you'll see in real codebases.

🧮

Foundations

Why Math for Machine Learning

If 'function' and 'value' feel hand-wavy, read the MML opener — it grounds the mental model.

Key takeaways

Variables are bindings (name → value), not boxes. Reassignment changes what the name points at.
The four primitives — `int`, `float`, `str`, `bool` — cover most simple values you'll see.

Where this shows up in ML

Hyperparameters are variables (`learning_rate = 0.01`); model state is held in variables (`model = LinearRegression()`).

If you remove it: Without named state, every experiment becomes a wall of literal numbers — unreadable and untunable.