filterfunction in Python
When discussing conditionals, we saw how an
if statement could be combined with a for loop to return a subset of a list that meets certain conditions. We did so by iterating through a list of elements and adding the element to a new list when an element met certain conditions. For example, imagine we want to write a function that selects even numbers, how would we do this?
Before determining how to categorize all elements in a list, let's just answer the question of whether one element is even. We can do so by making use of the modulo operator,
%. The % returns the remainder resulting from dividing a number by another. For example:
7 % 3
Seven divided by three, is two with a remainder of one. So the modulo operator returns the remainder, one. Let's look at some other examples. Six divides into three two times leaving a remainder of zero. So the operator returns zero.
6 % 3
And four divided by two also brings a remainder of zero.
4 % 2
Note that the above line effectively asks (and answers) whether 4 is even. This is because the statement
4 % 2 returns a zero, which means that four divided by two has a remainder of zero, and as we know, any number that is divisible by two with no remainder leftover is an even number.
Similarly, if a number is odd, then dividing by the number two results in a remainder of one. Ok so now let's write a function that checks if a number is even.
def is_even(number): return number % 2 == 0
print(is_even(3)) # False print(is_even(100)) # True
Now we are ready to write our function that selects just even numbers. We can iterate through the numbers one by one, and for each number, check if that number is even. If it's even then append the element to a new list of even numbers.
def select_even(elements): selected =  for element in elements: if is_even(element): selected.append(element) return selected
numbers = list(range(0, 11)) numbers
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
[0, 2, 4, 6, 8, 10]
Returning a subset of elements that meet a specific criteria is something commonly done in Python. And the procedure looks pretty much the same regardless of what we are selecting. For example, let's now select words that end with
def ends_ing(word): return word.endswith('ing') def select_ing(elements): selected =  for element in elements: if ends_ing(element): selected.append(element) return selected words = ['camping', 'biking', 'sailed', 'swam'] select_ing(words)
Notice that our two functions
select_even share a lot of similarity. Below, let's just highlight the differences.
def select_ing(elements): # selected =  # for element in elements: if ends_ing(element): # selected.append(element) # return selected def select_even(elements): # selected =  # for element in elements: if is_even(element): # selected.append(element) # return selected
Essentially, the only thing different between the functions is the criteria of how we are selecting. The
filter() function, allows us to filter for specific elements, so long as it knows the criteria and the elements.
filter()function returns an iterator where the items are filtered through a function to test if the item is accepted or not.
The general syntax of a
filter() function is given as:
filter ( function, iterable )
Let's apply this to filter the even numbers as we did above with a condition and loop.
<filter at 0x10642fda0>
filter returns a filter object, which isn't much use to us. What does a
filter object do? Not much. So we coerce it to a list and to see just the even numbers.
[0, 2, 4, 6, 8, 10]
Also notice that the
filter function takes two arguments, the first is the function itself. The function goes through each element, and passes that element into the criteria function. If that criteria function returns a truthy value, the element is selected. If not, the element is not selected.
Note that when passing through the function, no parentheses are added at the end of the function. This is important. If we pass through the parentheses the filtering will not occur.
Filter()without a filter function !
So what happens if pass a list to the
filter without supply the first argument i.e. a filtering function. Let's try it.
random_list = [0,'0','python', 2, True, 'flatiron', False, 38.9]
random_list above contains a number of different data types. Let's pass this list through
function() and use
None for filter function.
['0', 'python', 2, True, 'flatiron', 38.9]
So we see that with filter function as
None, the function defaults to Identity function i.e. each element in
random_list is checked to see if it is true or not. So in the output, we only get the elements which are true. Numbers and strings are always true (this includes '0' as a string), whereas 0 and False return a False value and hence are not included in the output.
In this section, we learned about the
filter function, which selects the elements in a list that match a specific criteria. We learned that
filter takes in two arguments. The first is the function that specifies the criteria specifying which elements to select, and the second argument is the list of elements to filter. Each element in the list is passed to the function one by one and if the function returns true the element is selected and put into a new array, which will be returned in the Filter object at the end.