We’ve looked at dictionaries as able to represent what something is. For example, this dictionary represents a student:

my_student = {
  'name': 'Rolf Smith',
  'grades': [70, 88, 90, 99]

If we want to calculate the average grade of the student, we could create a function to do so:

def average_grade(student):
  return sum(student['grades']) / len(student['grades'])

However, there is a flaw with this. This function is separate and unrelated from the student (e.g. in a large program, they could even be in different files), but it depends on the student variable having a particular structure:

  • The student must be a dictionary; and
  • There must be a grades key that must be a list or tuple, so that we can use sum() and len() on them.

It would be great if we could have something inside our dictionary that would return the average grade. That means the function would live in the same place as the data, and then it’s easier to see whether the data we require has changed or not.

Something like this:

my_student = {
  'name': 'Rolf Smith',
  'grades': [70, 88, 90, 99],
  'average': # something here

It would be fantastic if we could do this, and naturally the 'average' would have to change when then 'grades changes. It must be a function.

There’s no way to do this in a dictionary.


We must use objects for this. We can begin by thinking of objects as things that can store both data and functions that relate to that data.

Here’s that dictionary in object format:

class Student:
  def __init__(self, new_name, new_grades):
    self.name = new_name
    self.grades = new_grades

  def average(self):
    return sum(self.grades) / len(self.grades)

Scary syntax! Don’t worry—what it does is close to the same.

When you have that class, you can create objects using it. Let’s do that first and then explain exactly what is happening:

student_one = Student('Rolf Smith', [70, 88, 90, 99])
student_two = Student('Jose', [50, 60, 99, 100])

To create a new object, we use the class name as if it were a function call: Student().

Inside the brackets, we put arguments that will map to the __init__ method in the Student class.

Student('Rolf Smith', [70, 88, 90, 99]) maps to __init__(new_name, new_grades).

What you end up with is a thing that has two properties, name and grades.


Inside the __init__ method, we use self.name and self.grades. self is the current object, so when we assign values we modify only the “current object”. Notice how we don't pass anything that maps to the self parameter, because Python will do that for us.

In all these function calls, Python will treat self as the object that you are referring to (more about this in just a moment!).

Student('Rolf Smith', [70, 88, 90, 99])

def __init__(self, new_name, grades):
  self.name = new_name
  self.grades = new_grades

When you do this, self is the new object you are creating. You can assign it to a variable:

student_one = Student('Rolf Smith', [70, 88, 90, 99])

As you do that more, every object is a different self, with differently assigned properties depending on what you passed to the Student() constructor call.


Cool, so now we have the objects, both of which have different properties:

student_one = Student('Rolf Smith', [70, 88, 90, 99])
student_two = Student('Jose', [50, 60, 99, 100])

These are similar to our dictionaries, in that the dictionaries also store values:

d_student_one = {
  'name': 'Rolf Smith',
  'grades': [70, 88, 90, 99]
d_student_two = {
  'name': 'Jose',
  'grades': [50, 60, 99, 100]

To access them:






A method is a function which lives in a class.

The average() method in the Student class also has access to self, the current object. When we call the method:


What is really happening in the background is:


As you can see, student_one is passed as the first argument (and that is what self is in the method definition):

def average(self):
  return sum(self.grades) / len(self.grades)

So again, because self is student_one, self.grades is student_one.grades.


  • The sum of self.grades is the sum of [70, 88, 90, 99]: 347.
  • The length of self.grades is 4.

The result will be 86.75.


Just to recap, the class is very similar to the dictionary but it allows us to include methods as well that have access to the properties of the object we created.

Classes also gives us a bunch more functionality, we’ll look at that in the coming posts!