Week 9 Tutorial: Fundamentals of Python Programming II

POP77001 Computer Programming for Social Scientists

Indentation

  • Use tabs (1 tab) or spaces (4) consistently in your code
  • Python also permits mixing of different indentation levels, avoid it for legibility 📜
# Only for illustration purposes, do not do in practice!
# 2 spaces before print() statement
for i in range(5):
  print(i, end = ' ')
0 1 2 3 4 
# 4 spaces before print() statement
for i in range(5):
    print(i, end = ' ')
0 1 2 3 4

Indentation and Readability

  • Main rule, be consistent!
  • Think not just whether Python throws an error, but also readability
# This is semantically valid, but is badly styled
l = [0, 1, 1, 5]
for i in l:
  if i % 2 == 1: # 2 spaces
        print(i) # 6 spaces
print('End')
1
1
5
End

Iterables

  • An object is an iterable if it has __iter__ method that can be called with iter() function
x = 3
iter(x)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[4], line 2
      1 x = 3
----> 2 iter(x)

TypeError: 'int' object is not iterable
y = 'abc'
iter(y)
<str_ascii_iterator at 0x7b94caadc490>

Iteration over Dictionaries

  • items() method allows to iterate over keys and values in a dictionary
  • keys() method allows to iterate over just keys
  • values() method allow to iterate over just values
d = {'apple': 150.0, 'banana': 120.0, 'watermelon': 3000.0}
for k, v in d.items():
    print(k.upper(), int(v))
APPLE 150
BANANA 120
WATERMELON 3000
for k in d.keys():
    print(k.title())
Apple
Banana
Watermelon
for v in d.values():
    print(str(v/1000) + ' kg')
0.15 kg
0.12 kg
3.0 kg

List Comprehensions

  • The same iteration can often be implemented with forloop block or list comprehension
  • The choice is often between less typing, speed (🏎️) and legibility (📜)
[<expr> for <elem> in <iterable>]
[<expr> for <elem> in <iterable> if <test>]
[<expr> for <elem1> in <iterable1> for <elem2> in <iterable2>]

E.g. the following conditional statement:

l = [0, 'one', 1, 2]
for i in range(len(l)):
    if isinstance(x, int):
        l[i] = x * 2

Can be implemented as a list comprehension:

l = [0, 'one', 1, 2]
[x * 2 for x in l if isinstance(x, int)]
[0, 2, 4]

Exercise: List comprehensions and for loops

  • Consider a list of International vehicle registration codes below.
  • Suppose we want to create a list where each element is the length of each string in this list.
  • First, implement it using a for loop.
  • Now try doing the same using a list comprehension.
  • Finally, modify the list comprehension to keep only those elements that start with D.
  • You can use string method startswith for the last task.

l = ['D', 'DK', 'EST', 'F', 'IRL', 'MD', 'NL', 'S', 'UK']

Set and Dictionary Comprehensions

  • Analogous to list, sets and dictionaries have their own concise ways of iterating over them.
  • Note that iterating over them tends to be slower than over lists (🏎️).
{<expr> for <elem> in <iterable> if <test>}
{<key>: <value> for <elem1>, <elem2> in <iterable> if <test>}
o = {'apple', 'banana', 'watermelon'}
{e[0].title() + ' - ' + e for e in o}
{'A - apple', 'B - banana', 'W - watermelon'}
d = {'apple': 150.0, 'banana': 120.0, 'watermelon': 3000.0}
{k.upper(): int(v) for k, v in d.items()}
{'APPLE': 150, 'BANANA': 120, 'WATERMELON': 3000}

Note of Caution

  • Avoid modifying a sequence that you are iterating over.
  • As this can lead to unexpected results:
l = [x for x in range(1,11)]
l
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
for i in l:
    print('Element - ' + str(i))
    if i % 2 == 0:
        l.pop(i)
    print('Length = ' + str(len(l)))
Element - 1
Length = 10
Element - 2
Length = 9
Element - 4
Length = 8
Element - 5
Length = 8
Element - 7
Length = 8
Element - 8
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
Cell In[16], line 4
      2 print('Element - ' + str(i))
      3 if i % 2 == 0:
----> 4     l.pop(i)
      5 print('Length = ' + str(len(l)))

IndexError: pop index out of range

Docstring

  • Docstring provides a standardized way of documenting functionality.
  • It is defined as a first statement in module, function, class, or method definition.
  • Docstring is accessible with help() function.
  • It also creates a special __doc__ attribute.

Extra

Python documentation on docstring

One-line Docstrings

  • In the simplest case, docstrings can take just one line:
def <function_name>(arg_1, arg_2, ..., arg_n):
    """<docstring>"""
    <function_body>
def add_one(x):
    """Adds 1 to numeric input"""
    return x + 1
help(add_one)
Help on function add_one in module __main__:

add_one(x)
    Adds 1 to numeric input

add_one.__doc__
'Adds 1 to numeric input'

Multi-line Docstrings

  • A more elaborate docstring would consist of a single summary line, followed by a blank line, followed by a longer description of inputs, arguments and output:
def <function_name>(arg_1, arg_2, ..., arg_n):
    """<summary_docstring>
    
    <longer_description>
    """
    <function_body>
def even_or_odd(num):
    """Check whether the number is even or odd

    Takes an integer as input
    Returns the result as a string
    """
    if num % 2 == 0:
        return 'even'
    else:
        return 'odd'
help(even_or_odd)
Help on function even_or_odd in module __main__:

even_or_odd(num)
    Check whether the number is even or odd

    Takes an integer as input
    Returns the result as a string

Exercise: Functions and Docstrings

  • Most functions for calculating summary statistics would be available in separate packages (built-in statistics and external numpy).
  • But it is helpful to try programming some of those yourself to understand the internal working.
  • Modify the function definition below according to its docstring specification.
  • You can use function round for rounding.
  • Try your function with 0.1, 2.7, 3.5, 4, 5.98 supplied as arguments.
def calculate_mean():
    """
    Calculates mean
    
    Takes any number of numeric arguments as an input.
    Returns mean rounded to two decimal place.
    """
    pass

Week 9: Assignment 3

  • Python Fundamentals and Control Flow
  • Due by 12:00 on Monday, 11th November