Imports and the standard library tour

The module on ending a loop at the end of input used import sys to read every line. Python ships with many more modules like sys, each a ready-made toolbox for one kind of task. This lesson tours four you will reach for often, and shows how to bring in and read about any of them.

Bringing in a module

import math makes every name inside the math module available as math.something. from math import gcd instead brings just gcd into your program, so you write gcd(a, b) without the math. in front. Both forms load the same code. Pick whichever reads more clearly for the names you use.

An import runs where it appears in the file, the same as any other line. Put every import at the top of the file anyway, so it runs before the code that needs it. Anyone reading the file can then see every tool it uses in one place.

Import only what you plan to use. An unused import does no harm to your program's output. It clutters the file, though, and makes it harder to tell, at a glance, which tools the program relies on.

You will meet many more standard-library modules over the rest of this course, each one introduced in its own module when the topic it supports comes up. This lesson is only a first look at four of them. Each solves a task you already recognize well: reducing a ratio, counting a group, tallying values, and listing every possible pair from a small set.

math: gcd and comb

math.gcd(a, b) returns the largest number that divides both a and b with no remainder. On this course's Python, math.gcd always takes exactly two numbers. A newer Python accepts more, which the module on the language boundary covers. math.comb(n, k) returns the number of ways to choose k items from n, with order not mattering. Both n and k must be given by position, not by name: math.comb(n=6, k=2) fails, but math.comb(6, 2) works.

1import math2
3red_plates = 184blue_plates = 245shared_factor = math.gcd(red_plates, blue_plates)6
7print(red_plates // shared_factor, blue_plates // shared_factor)8print(math.comb(6, 2))

3 4
15

Eighteen red plates to twenty-four blue plates reduces to three to four, because math.gcd(18, 24) is 6, and dividing both counts by 6 gives the smallest whole-number ratio with the same proportion. Separately, choosing two people for a committee out of six candidates has math.comb(6, 2), which is 15, possible pairs. Writing either of these by hand takes a loop or nested loops. The module functions replace both with one call.

collections.Counter: tallying values

Counter builds a count of how often each value appears in a collection you can loop over, such as a list or a string. Once tally has been built this way, tally["red"] gives the count for one value. Asking it for a value it never saw gives 0, not an error. .most_common(k) gives the k most frequent values, paired with their counts.

1from collections import Counter2
3plate_colors = ["red", "blue", "red", "green", "red", "blue"]4tally = Counter(plate_colors)5
6print(tally["red"])7print(tally.most_common(1))

3
[('red', 3)]

Six plates come off the shelf in the order red, blue, red, green, red, blue. tally["red"] reports 3. tally.most_common(1) reports [('red', 3)], the single most common colour with its count. Writing this by hand means a dictionary and a manual loop that checks and updates a count for each item. Counter does the same job in one line.

itertools.combinations: every pair or group

combinations(items, r) produces every way of choosing r items from items, keeping their original order and never repeating a choice. It works on any small collection you already have, not only numbers.

1from itertools import combinations2
3players = ["Amir", "Bo", "Chen"]4
5for pair in combinations(players, 2):6    print(pair)

('Amir', 'Bo')
('Amir', 'Chen')
('Bo', 'Chen')

Three players, Amir, Bo and Chen, give exactly three possible pairs for a doubles team. combinations produces each one once, in the order the players were listed. A hand-written version needs two nested loops with an index check to avoid pairing a player with themselves or repeating a pair in reverse. combinations handles that bookkeeping for you.

Reach for combinations whenever a problem asks for every group of a fixed size from a small list, rather than writing the nested loops and the index check yourself.

Reading a function before you use it

The official Python documentation lists every module's functions with their exact signature. Reading it before you call a function tells you what to pass and what comes back, without guessing or testing it first. Always read the signature on the 3.8 page of the docs, docs.python.org/3.8/, since a newer page can describe a newer, different signature for the same name. This is exactly what happened above: a newer docs page shows math.gcd(*integers), accepting any number of arguments, but that page also marks the change with a note reading "Changed in version 3.9". The 3.8 page shows the two-argument form this lesson teaches, with no such note, because two arguments is how math.gcd has always worked up to and including 3.8. The module on the language boundary covers reading these version notes in full, and the module on searching the docs under contest conditions covers finding the right page quickly.

Not every package you might have seen elsewhere is available. The standard librarystandard libraryThe collection of modules that ships with Python itself, without installing anything extra.In the glossary ships with Python itself, but third-party packages installed separately, such as numpy, are not available on the contest grader. Stick to modules from the standard library.

The rest of the toolbox

This lesson worked through four names closely. The standard library has many more worth knowing by name now, even before you meet each one in full:

Knowing a module's name and rough purpose is enough for now. Each one gets its own full treatment when the topic it supports comes up.

Misspelling a module name

1import maths2
3print(maths.sqrt(16))

Traceback (most recent call last):
  File "typo_import.py", line 1, in <module>
    import maths
ModuleNotFoundError: No module named 'maths'

maths is not a Python module. There is nothing by that name to import, so Python stops immediately with ModuleNotFoundError, before your program does anything else. When you see this error, check the module name against the docs and check your spelling.

Recap

import and from ... import bring a module's names into your program. Do this once, near the top of the file. math, collections.Counter and itertools.combinations each replace a hand-written loop with a single call. The official docs tell you exactly what each one takes and returns. A misspelled module name fails immediately with ModuleNotFoundError, and a package outside the standard library is not available at all.