Python: Primer

Seong-Hwan Jun

Python on the cloud

We will primarily use Google Colab for classroom demos.
Basic features of colab
We may also run Python on JupyterLab via Center for Integrated Research Computing (CIRC).
To request account, speak to your advisor if he/she has an account with CIRC; otherwise, contact Prof. McCall (DBCB students). This is optional.

Python locally

Recommended to work locally for homeworks and projects.
Visual studio code is an integrated development environment (like RStudio) that is available free of use.
More details to come on setting up the local environment.

Language basics

Just like R, it is an interpreted language as opposed to being a compiled language.
Important distinction from R: indexing begins from 0 not 1!

Language basics

a = 3
b = 5
print(a+b)

Primitive types

Data types like boolean, integer, float, and strings (characters in R) are natively supported.

Print

print("hello")
age = 5
print("hello I am " + str(age) + " years old.")

hello
hello I am 5 years old.

Print

print("hello I am " + age + " years old.") # error: string concat with int

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[3], line 1
----> 1 print("hello I am " + age + " years old.")

TypeError: can only concatenate str (not "int") to str

String formatting

age = 5
print(f"hello I am {age} years old.")
pi = 3.141
print(f"hello pi is {pi:.2f} to 2 decimal places.")

hello I am 5 years old.
hello pi is 3.14 to 2 decimal places.

Data structures: list

Ordered, mutatble collection of items (similar to R’s list).

a = []
a.append(3)
a.append(5)
print(a)

[3, 5]

a[0] = 1.8
print(a)

[1.8, 5]

Data structures: list

b = [5, "hello"]
a.append(b)
print(a)

[1.8, 5, [5, 'hello']]

Data structures: list

c = ["ABC", 3.7]
d = b + c # concatenates b and c and assign as d
print(d)

[5, 'hello', 'ABC', 3.7]

b.extend(c) # modifies b by concat of c
print(b)

[5, 'hello', 'ABC', 3.7]

print(len(b)) # len() is a function that retrieves the length of the list.

Data structures: list

Indexing and slicing:

print(b)

[5, 'hello', 'ABC', 3.7]

print(b[0:2])

[5, 'hello']

print(b[-2]) # retrieve the second elements from the end

ABC

print(b[-3:]) # third last element to the end.

['hello', 'ABC', 3.7]

print(b[:3]) # first three elements.

[5, 'hello', 'ABC']

Data structures: list

aa = [1.5, 0.3, -5, 9]
aa.sort()
print(aa)

[-5, 0.3, 1.5, 9]

bb = ["hello", "ah!!!", "who?", "123", "?!@#!"]
bb.sort()
print(bb)

bb.reverse()
print(bb)

['123', '?!@#!', 'ah!!!', 'hello', 'who?']
['who?', 'hello', 'ah!!!', '?!@#!', '123']

Data structures: list

aa = [1.5, 0.3, -5, 9]
aa.remove(1.5)
print(aa)

[0.3, -5, 9]

aa.insert(1, 5) # Insert 5 at index 1.
print(aa)

[0.3, 5, -5, 9]

aa.append(10) # Insert at the end.
print(aa)

[0.3, 5, -5, 9, 10]

last_element = aa.pop() # Remove and return the last element.
print(aa)
print(last_element)

[0.3, 5, -5, 9]
10

element=aa.pop(1) # Remove and return the element at index 1.
print(element)
print(aa)

5
[0.3, -5, 9]

Data structures: list

print(aa)
aa.append(-5)
print(aa.count(-5))

[0.3, -5, 9]
2

print(aa.index(-5)) # returns the index corresponding to the first occurrence of -5.

aa.clear() # Remove all items.
print(aa)

[]

Data structures: tuples

Ordered, immutatble collection of items.

Faster and minimal storage requirements compared to list (e.g., accessing and unpacking, no overhead needed to handle dynamic size changes).
Often used by functions returning multiple objects.
Hashable. (what is this?)

Data structures: tuples

coord = (5.2, 1.8, -1.4)
print(coord)

(5.2, 1.8, -1.4)

coord[0] = 10.0 # Immutable so we cannot modify in place.
print(coord)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[27], line 1
----> 1 coord[0] = 10.0 # Immutable so we cannot modify in place.
      2 print(coord)

TypeError: 'tuple' object does not support item assignment

Data structures: dictionary

Stores Key-Value pair.
Key needs to be hashable: most primitive types are hashable.
Runtime is \(O(1)\) for access, insertion, removal, and search.
Cannot have duplicate keys.

Data structures: dictionary

a = {}
a["key1"] = 3.5
a["key2"] = 9.5
print(a)

{'key1': 3.5, 'key2': 9.5}

b = {"k1": "value1", "k2": "v2"}
b["k1"] = "x1" # Overwrites the value at k1
print(b)
print("k1" in b) # Check if key is in dict.

{'k1': 'x1', 'k2': 'v2'}
True

Data structures: set

Similar as set in mathematics: no duplicates.
\(O(1)\) for insertion, removal, and search.
\(O(n)\) for set operations: union, intersection, difference.

a = {1, 2, 3}
b = set()
b.add(1)
b.add(5)
print(1 in a) # Check if item in set

True

print(a.intersection(b))

{1}

print(a.union(b))

{1, 2, 3, 5}

print(a.difference(b))

{2, 3}

Control flows: if, elif, else

x = 3
if x > 5:
    print("x is greater than 5")
elif x == 5:
    print("x is equal to 5")
else:
    print("x is less than 5")

x is less than 5

Supports ternary operation:

ss = "x is greater than 5" if x > 5 else "x is less than equal to 5"
print(ss)

x is less than equal to 5

Control flows: for loop

for i in range(5): # exclusive so 5 is not included
    print(i)

for i in range(2, 5): # can specify begin index
    print(i)

2
3
4

a = ["item1", "item2", 3, 6]
for i in a:
    print(i)

item1
item2
3
6

Control flows: for loop

d = {1: "item1", 2: "item2", 3: "item3"}
for k, v in d.items():
    print("Key: " + str(k) + ", Value: " + v)
    if k == 2:
      break

Key: 1, Value: item1
Key: 2, Value: item2

a = {1, 2, 3}
for item in a:
    print(item)

1
2
3

zip

Combine two or more lists into a list of tuples for iteration.

names = ["Alice", "Bob"]
ages = [25, 30]
for name, age in zip(names, ages):
    print(f"{name} is {age} years old.")

Alice is 25 years old.
Bob is 30 years old.

names = ["Alice", "Bob"]
ages = [25, 30]
net_assets = [1000000, 1000, 400000]
for name, age, net_asset in zip(names, ages, net_assets): # zip truncates to the length of the shorter list
    print(f"{name} is {age} years old, and has ${net_asset}.")

Alice is 25 years old, and has $1000000.
Bob is 30 years old, and has $1000.

dictionary = dict(zip(names, ages)) # forms key:value pairs.
print(dictionary)

{'Alice': 25, 'Bob': 30}

enumerate

Iterate over the index and the item.

fruits = ["apple", "banana", "cherry"]
for index, fruit in enumerate(fruits): 
    print(f"{index}: {fruit}")

0: apple
1: banana
2: cherry

Comprehensions

Generate new collection by iteration and optionally applying conditions/operations.

squares = [x**2 for x in range(5) if x % 2 == 0]
print(squares)

[0, 4, 16]

squares = {x: x**2 for x in range(5)}
print(squares)  # Output: {0: 0, 1: 1, 2: 4, 3: 9, 4: 16}

{0: 0, 1: 1, 2: 4, 3: 9, 4: 16}

Functions

def foo(a, b):
  return(a + b, a-b, a*b, a/b) # Returning a tuple.

add, sub, mul, div = foo(3, 5)
print(add)
print(sub)
print(mul)
print(div)

Functions

def foo(a, b) -> float: # declare return type for clarity
  return(a + b)

print(foo(5, 2))

def bar(a, b) -> float: # in larger projects, this adds clarity
  return("hah!")

print(foo(5, 2))  # but interpreter ignores the return type

Lambda

Functions without names, often used where short, disposable functionality is needed. Similar to anonymous functions in R’s apply.

numbers = [1, 2, 3, 4]
squared = list(map(lambda x: x**2, numbers))
print(squared)

[1, 4, 9, 16]

Lambda

numbers = [1, 2, 3, 4, 5]
odd_numbers = list(filter(lambda x: x % 2 != 0, numbers))
print(odd_numbers)

[1, 3, 5]

Lambda

from functools import reduce
numbers = [1, 2, 3, 4]
initial_value = -5
total = reduce(lambda x, y: x + y, numbers, initial_value)
print(total)

Lambda

pairs = [(2, 'b'), (1, 'a'), (3, 'c')]
pairs.sort(key=lambda pair: pair[0])
print(pairs)

[(1, 'a'), (2, 'b'), (3, 'c')]

Lambda

def foo(x):
  bar = lambda x: x**2 # can assign it to a variable for re-use.
  y = bar(x)
  z = 2*bar(x)
  return(y, z)

print(foo(3))

(9, 18)

bar = lambda x: x**2 # can assign it to a variable if desired.
def foo(fn, z):
  return(fn(z))

print(foo(bar, 3))

Context managers

Starts with keyword with to declare a opening of a resource such as file.

with open("example.txt", "w") as f:
    f.write("Hello, Python!") 

# File is closed and any resources released automatically without explicitly calling f.close()

f.write("Bye Python!") # Error.

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[56], line 6
      2     f.write("Hello, Python!") 
      4 # File is closed and any resources released automatically without explicitly calling f.close()
----> 6 f.write("Bye Python!")

ValueError: I/O operation on closed file.

Context managers

with open("example.txt", "r") as f:
    print(f.readline())

print(f.readline())

Hello, Python!

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[57], line 4
      1 with open("example.txt", "r") as f:
      2     print(f.readline())
----> 4 print(f.readline())

ValueError: I/O operation on closed file.

Class

Class allows you to create your own data type made up of attributes (variables) and operations (functions).

Bundle attributes and operations together (Encapsulation).
Improved reusability: classes can be extended where deriving class implements its own specific attribute and methods while inheriting behaviors of parent class.
Functions defined for a class is referred to as methods.

Class

Instance: A specific object created from a class.
self: Refers to the instance on which the method is called.
__init__: A special method for initializing a new object, called constructor.
Attributes: Variables that store the state or data of an object.
Methods: Functions that define the behavior of an object. Every function of the class must contain self as first argument.

Class

class Animal:
  def __init__(self, name):
    print("Animal constructor.")
    self.name = name

  def speak(self):
    print(f"{self.name}'s default noise.")

  def print_name(self):
    print(self.name)

object = Animal("John")
object.print_name()
object.speak()

Animal constructor.
John
John's default noise.

Class

class Cat(Animal):
  def __init__(self, name, sound):
    print("Cat constructor.")
    super().__init__(name)
    self.sound = sound

  def speak(self): # overriding parent's method
    print(f"{self.name} {self.sound}")

  def print_name(self):
    print(self.name)

doug = Cat("Doug", "barks")
doug.print_name()
doug.speak()

Cat constructor.
Animal constructor.
Doug
Doug barks

Debugging

class TestClass:
  def __init__(self):
    print("Class constructor.")
    self.a = 5
    self.dict = dict()

  def add_to_dict(self, k, v):
    # Divide v by self.a, then the resulting value to self.dict[k]
    self.dict[k] = k / self.a
    print("Added to dictionary.")

  def get_item(self, k):
    return self.dict[k]

obj = TestClass()
obj.add_to_dict(1, 5)
expected_value = 5 / obj.a
realized_value = obj.get_item(1) 
print(f"Expected value: {expected_value}")
print(f"Realized value: {realized_value}") # Not what we expected.

Class constructor.
Added to dictionary.
Expected value: 1.0
Realized value: 0.2