You can define functions to provide the required functionality. Here are simple rules to define a function in Python:
An example of a function that adds on to the parameter a prints and returns the output as b:
In [1]:
def add(a):
"""
add 1 to a
"""
b=a+1;
print(a, "if you add one" ,b)
return(b)
The figure below illustrates the terminology:
We can obtain help about a function :
In [2]:
help(add)
We can call the function:
In [3]:
add(1)
Out[3]:
If we call the function with a new input we get a new result:
In [4]:
add(2)
Out[4]:
We can create different functions. For example, we can create a function that multiplies two numbers. The numbers will be represented by the variables a and b:
In [5]:
def Mult(a,b):
c=a*b
return(c)
The same function can be used for different data types. For example, we can multiply two integers:
In [6]:
Mult(2,3)
Out[6]:
Two Floats:
In [7]:
Mult(10,3.14)
Out[7]:
We can even replicate a string by multiplying with an integer:
In [8]:
Mult(2,"Michael Jackson ")
Out[8]:
In [9]:
def divide_values(a, b):
return a / b
The input to a function is called a formal parameter.
A variable that is declared inside a function is called a local variable. The parameter only exists within the function (i.e. the point where the function starts and stops).
A variable that is declared outside a function definition is a global variable, and its value is accessible and modifiable throughout the program. We will discuss more about global variables at the end of the lab.
In [10]:
#Function Definition
def square(a):
"""Square the input and add one
"""
#Local variable
b=1
c=a*a+b;
print(a, "if you square +1 ",c)
return(c)
The labels are displayed in the figure:
We can call the function with an input of 3:
In [11]:
#Initializes Global variable
x=3
#Makes function call and return function a y
z=square(x)
z
Out[11]:
We can call the function with an input of 2 in a different manner:
In [12]:
square(2)
Out[12]:
If there is no return statement, the function returns None. The following two functions are equivalent:
In [13]:
def MJ():
print('Michael Jackson')
def MJ1():
print('Michael Jackson')
return(None)
In [14]:
MJ()
In [15]:
MJ1()
Printing the function after a call reveals a None is the default return statement:
In [16]:
print(MJ())
print(MJ1())
In [17]:
def con(a,b):
return(a+b)
In [18]:
print(con(1, 2))
print(con('1', '2'))
In [20]:
print(con([1], [2]))
There are many pre-defined functions in Python, so let's start with the simple ones.
The print() function:
In [21]:
album_ratings = [10.0,8.5,9.5,7.0,7.0,9.5,9.0,9.5]
print(album_ratings)
The sum() function adds all the elements in a list or tuple:
In [22]:
sum(album_ratings)
Out[22]:
The length function returns the length of a list or tuple:
In [23]:
len(album_ratings)
Out[23]:
Consider the two lines of code in Block 1 and Block 2: the procedure for each block is identical. The only thing that is different is the variable names and values.
In [24]:
a1=4;
b1=5;
c1=a1+b1+2*a1*b1-1
if(c1<0):
c1=0;
else:
c1=5;
c1
Out[24]:
In [25]:
a2=0;
b2=0;
c2=a2+b2+2*a2*b2-1
if(c2<0):
c2=0;
else:
c2=5;
c2
Out[25]:
We can replace the lines of code with a function. A function combines many instructions into a single line of code. Once a function is defined, it can be used repeatedly. You can invoke the same function many times in your program. You can save your function and use it in another program or use someone else’s function. The lines of code in code block 1 and code block 2 can be replaced by the following function:
In [26]:
def Equation(a,b):
c=a+b+2*a*b-1
if(c<0):
c=0
else:
c=5
return(c)
This function takes two inputs, a and b, then applies several operations to return c. We simply define the function, replace the instructions with the function, and input the new values of a1,b1 and a2,b2 as inputs. The entire process is demonstrated in the figure:
Code Blocks 1 and Block 2 can now be replaced with code Block 3 and code Block 4.
In [27]:
a1=4;
b1=5;
c1=Equation(a1,b1)
c1
Out[27]:
In [28]:
a2=0;
b2=0;
c2=Equation(a2,b2)
c2
Out[28]:
In [29]:
def type_of_album(artist,album,year_released):
if year_released > 1980:
print(artist,album,year_released)
return "Modern"
else:
print(artist,album,year_released)
return "Oldie"
x = type_of_album("Michael Jackson","Thriller",1980)
print(x)
We can use a loop in a function. For example, we can print out each element in a list:
In [30]:
def PrintList(the_list):
for element in the_list:
print(element)
In [31]:
PrintList(['1',1,'the man',"abc"])
You can set a default value for arguments in your function. For example, in the isGoodRating() function, what if we wanted to create a threshold for what we consider to be a good rating? Perhaps by default, we should have a default rating of 4:
In [32]:
def isGoodRating(rating=4):
if(rating < 7):
print("this album sucks it's rating is",rating)
else:
print("this album is good its rating is",rating)
In [33]:
isGoodRating()
isGoodRating(10)
In [34]:
artist = "Michael Jackson"
def printer1(artist):
internal_var = artist
print(artist,"is an artist")
printer1(artist)
If we print internal_var we get an error.
We got a Name Error: name 'internal_var' is not defined. Why?
It's because all the variables we create in the function is a local variable, meaning that the variable assignment does not persist outside the function.
But there is a way to create global variables from within a function as follows:
In [35]:
artist = "Michael Jackson"
def printer(artist):
global internal_var
internal_var= "Whitney Houston"
print(artist,"is an artist")
printer(artist)
printer(internal_var)
The scope of a variable is the part of that program where that variable is accessible. Variables that are declared outside of all function definitions, such as the myFavouriteBand variable in the code shown here, are accessible from anywhere within the program. As a result, such variables are said to have global scope, and are known as global variables. myFavouriteBand is a global variable, so it is accessible from within the getBandRating function, and we can use it to determine a band's rating. We can also use it outside of the function, such as when we pass it to the print function to display it:
In [36]:
myFavouriteBand = "AC/DC"
def getBandRating(bandname):
if bandname == myFavouriteBand:
return 10.0
else:
return 0.0
print("AC/DC's rating is:", getBandRating("AC/DC"))
print("Deep Purple's rating is:",getBandRating("Deep Purple"))
print("My favourite band is:", myFavouriteBand)
Take a look at this modified version of our code. Now the myFavouriteBand variable is defined within the getBandRating function. A variable that is defined within a function is said to be a local variable of that function. That means that it is only accessible from within the function in which it is defined. Our getBandRating function will still work, because myFavouriteBand is still defined within the function. However, we can no longer print myFavouriteBand outside our function, because it is a local variable of our getBandRating function; it is only defined within the getBandRating function:
In [37]:
def getBandRating(bandname):
myFavouriteBand = "AC/DC"
if bandname == myFavouriteBand:
return 10.0
else:
return 0.0
print("AC/DC's rating is: ", getBandRating("AC/DC"))
print("Deep Purple's rating is: ", getBandRating("Deep Purple"))
print("My favourite band is", myFavouriteBand)
Finally, take a look at this example. We now have two myFavouriteBand variable definitions. The first one of these has a global scope, and the second of them is a local variable within the getBandRating function. Within the getBandRating function, the local variable takes precedence. Deep Purple will receive a rating of 10.0 when passed to the getBandRating function. However, outside of the getBandRating function, the getBandRating s local variable is not defined, so the myFavouriteBand variable we print is the global variable, which has a value of AC/DC:
In [38]:
myFavouriteBand = "AC/DC"
def getBandRating(bandname):
myFavouriteBand = "Deep Purple"
if bandname == myFavouriteBand:
return 10.0
else:
return 0.0
print("AC/DC's rating is:",getBandRating("AC/DC"))
print("Deep Purple's rating is: ",getBandRating("Deep Purple"))
print("My favourite band is:",myFavouriteBand)
Joseph Santarcangelo has a PhD in Electrical Engineering, his research focused on using machine learning, signal processing, and computer vision to determine how videos impact human cognition. Joseph has been working for IBM since he completed his PhD.
James Reeve James Reeves is a Software Engineering intern at IBM.