Importing from the Standard Library

By the end of this lesson you can:

• Use three import shapes: import math, from math import sqrt, import statistics as st.
• Reach for the right standard-library module: math, statistics, random, time, datetime, os, sys, csv, json, re.
• Read the Python docs page for any module — the "index of useful functions".
• Avoid the from x import * trap.

Since PY-L1-11 you've been writing this:

import random
print(random.randint(1, 6))

That's an import. random is a module — a file of pre-written code that ships with Python. There are over 200 of these in the standard library. Today we look at the picture more broadly and meet three more.

Shape 1 · import module

The default. You access things inside the module by typing the module name first.

import math

print(math.sqrt(81))        # → 9.0
print(math.pi)              # → 3.141592653589793
print(math.floor(3.9))      # → 3
print(math.ceil(3.1))       # → 4

The math. prefix is friction, but it's also clarity — a future reader instantly knows where sqrt came from.

Shape 2 · from module import name

Pull a specific function or constant into your namespace. No prefix needed afterwards.

from math import sqrt, pi

print(sqrt(81))     # → 9.0   (no math. prefix)
print(pi)           # → 3.141592653589793

Good when you use the same name a lot. Less good if it's ambiguous — sqrt could be from math or cmath; the prefix makes it obvious.

Shape 3 · import module as alias

Rename a module on import. Used for either (a) clearer names, or (b) standard ecosystem conventions — like import numpy as np, which everyone in the data world writes.

import statistics as st

scores = [82, 91, 76, 65, 88]
print(st.mean(scores))      # → 80.4
print(st.median(scores))    # → 82
print(st.stdev(scores))     # → 10.21...

Avoid · from module import *

The star imports everything the module exports. Now you don't know where any of your functions came from — your open might be Python's built-in or might be requests.open.

from math import *      # ❌ avoid
# Now sqrt, pi, log, sin, ... all leak into your namespace.

Acceptable only in tiny throw-away scripts. Don't use it in real files.

Modules you'll meet

math         sqrt, floor, ceil, pi, log, sin, ...
statistics   mean, median, stdev, variance, mode
random       randint, choice, shuffle, sample, random, seed
time         time(), sleep(seconds)
datetime     date, time, datetime, timedelta
os           os.path.join, os.listdir, os.remove, os.makedirs
sys          sys.argv (command-line args), sys.exit
csv          csv.reader, csv.writer  (Level 4)
json         json.dump, json.load    (PY-L2-45)
re           re.findall, re.search   (PY-L2-42)

You'll touch most of these before Level 2 ends.

Where do imports live?

Convention: all imports at the top of the file. Standard-library first, third-party next, your own modules last. Three groups, separated by a blank line.

# stdlib
import os
import sys
import random
from datetime import date

# third-party (we won't use any until Level 4)
# import requests

# your own modules (tomorrow's lesson)
# from .my_helpers import safe_int

Build class_stats.py. It loads a file of test scores and uses statistics + math to produce a polished report.

Save marks.txt first — one score per line:

82
91
76
65
88
73
95
58
80
71

Code

# class_stats.py — use statistics + math + random

import statistics as st
import math
import random

def load_marks(path):
    with open(path, encoding="utf-8") as f:
        out = []
        for line in f:
            try:
                out.append(int(line.strip()))
            except ValueError:
                pass
        return out

marks = load_marks("marks.txt")
if not marks:
    print("No marks loaded.")
    raise SystemExit

print(f"Students   : {len(marks)}")
print(f"Mean       : {st.mean(marks):.2f}")
print(f"Median     : {st.median(marks)}")
print(f"Stdev      : {st.stdev(marks):.2f}")
print(f"Range      : {min(marks)} … {max(marks)}")

# Math: how many sit within 1 standard deviation of the mean?
mean = st.mean(marks)
sd   = st.stdev(marks)
within = [m for m in marks if abs(m - mean) <= sd]
print(f"Within 1σ  : {len(within)} of {len(marks)} ({len(within) / len(marks) * 100:.0f}%)")

# Random: pick a student at random to present today
random.seed()
print(f"Today's presenter index: {random.randrange(len(marks))}")

# Math: ceiling of average (round UP for "minimum passing year-group avg")
print(f"Ceil(mean) : {math.ceil(mean)}")

Sample output

Students   : 10
Mean       : 77.90
Median     : 78.0
Stdev      : 11.32
Range      : 58 … 95
Within 1σ  : 7 of 10 (70%)
Today's presenter index: 4
Ceil(mean) : 78

Read the diff

Six lines of computation. Three modules. We didn't write a single mean, median or standard-deviation function from scratch — they came free with Python. st is statistics renamed; random.randrange(n) picks an integer 0 to n-1; math.ceil rounds up.

The standard library is hundreds of free modules. import x brings the whole module in (you write x.func()). from x import y brings a specific name in. import x as y renames on the way in. Don't use from x import * — it makes your code hard to read. Put all imports at the top of the file, grouped. Reach for stdlib first before writing your own version of mean, sqrt, sleep or randint.