24. Functions - advanced#

Author: Tue Nguyen

24.1. Outline#

We use def to define regular functions
We use lambda quickly define simple functions (normally in one line)
Sometimes, we want to quickly define a function without a name (thinking of a name may be time-consuming)
- They are called anonymous functions
- We define anonymous functions using lambda
- Anonymous functions are used to support other operations such as sorting and filtering

a) Ex 1: greeting

# Define the function
say_hi = lambda name: print(f"Hi {name}")

# Call
say_hi("Tom")

Hi Tom

b) Ex 2: add 2 numbers

# Define
add = lambda a, b: a + b

# Call
add(2, 3)

Ex 3: \(f(x) = x^2 + 2x + 5\)

# Define
f = lambda x: x**2 + 2*x + 5

# Call
f(1)

Ex 4: \(g(x, y) = x^2 + 2xy + 3y^2\)

# Define
g = lambda x, y: x**2 + 2*x*y + 3*y**2

# Call
g(2, 1)

Syntax: lambda params: code
An anonymous function cannot stand on its own but must be used in another operation such as sorting or filter

a) Ex 1: sorting

Suppose we have a list of students
- Each element of this list is a tuple of 2 entries
- The first entry is the student’s name
- The second entry is the student grade
We might want to sort the student by name or grade

# Init the list of students
students = [("John", 6), ("Alex", 5.5), ("Jane", 9), ("Peter", 8)]
students

[('John', 6), ('Alex', 5.5), ('Jane', 9), ('Peter', 8)]

# Case 1: Sort by name (ascending order)
sorted(students, key=lambda x: x[0])

[('Alex', 5.5), ('Jane', 9), ('John', 6), ('Peter', 8)]

Comments

The function sorted will sort the list students by key
key expects a function that does something to each element of the list and returns a value for that element
sorted will then sort students based on those returned values
In the above example, key=lambda x: x[0] means that
- Iterate through each element of the list students using variable x
- At each element, return the first entry of x (or x[0])
- Then sort the list students by those first entries
As a result, you can see the list students is sorted by name in the ascending order

# Case 2: Sort by name (descendin order)
# Just pass reverse=True
sorted(students, key=lambda x: x[0], reverse=True)

[('Peter', 8), ('John', 6), ('Jane', 9), ('Alex', 5.5)]

# Case 3: Sort by grade (ascending order)
# Very similar to Case 1
# Just need to change the expression to x[1]
sorted(students, key=lambda x: x[1])

[('Alex', 5.5), ('John', 6), ('Peter', 8), ('Jane', 9)]

Remarks

For now, the use of lambda functions is very limited including
- Quickly define short and simple functions
- Define anonymous function to use in sorted
However, anonymous lambda functions will be very convenient for real-life data manipulation when used with pandas (which will be discussed in a separate series)

To be updated

To be updated

To be updated