MACHINE LEARNING LAB 2

Data Structures, NumPy Arrays,

SciPy Functions
MUNADI SIAL

SCHOOL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE

NATIONAL UNIVERSITY OF SCIENCES AND TECHNOLOGY
Data Structures
Python offers 4 built-in data types for storing collections of data:

1. Lists: A collection of data which is ordered, changeable and allows duplicate

members

2. Tuples: A collection of data which is ordered, unchangeable and allows duplicate

members

3. Sets: A collection of data which is unordered, unchangeable and does not allow
duplicate members

4. Dictionaries: A collection of data which is stored in key:value pairs and is

ordered, changeable and does not allow duplicate members (Dictionaries were
unordered before Python 3.7)
Data Structures
Examples of the collection data are given below. Notice the bracket types.

x_list = [1, 2, 3, 4, 5]

x_tuple = (1, 2, 3, 4, 5)

x_set = {1, 2, 3, 4, 5}

x_dictionary = {"name" : "Ali"

"age" : 31
"result" : False}
Lists
• A List is a collection of data that is
o Ordered
o Changeable (Mutable)
o Allows Duplicate Members

• A List is a sequence of values (called items or elements)

• Whereas a string is a sequence of characters, a list can be a sequence of any data
type

• Lists are among the commonly used data types in python

• Lists are used extensively in ROS such as in reading laser data
• A List works similar to the arrays in C++
Lists - Creation
• To create a list, the simplest way is to use square brackets to enclose the items
and use commas to separate the items
x = [1, 2, 3, 4, 5]

print(x)

Output:

[1, 2, 3, 4, 5]
Lists
• Lists can hold different types of data:
a = [11, 2, 93, 401, 560]
b = [1.5, 6.6, 7.3, 8.9]
c = ["apple", "banana", "cherry"]
d = [True, True, False, True, False]
e = []
f = list("ROBOT")
Output:

print(a) [11, 2, 93, 401, 560]

print(b) [1.5, 6.6, 7.3, 8.9]
print(c) ['apple', 'banana', 'cherry']
print(d) [True, True, False, True, False]
print(e) []
print(f) ['R', 'O', 'B', 'O', 'T']
Lists
• The same list can also hold different types of data:
g = [31.5, "Robotics", 4, True, 9]

print(g)

Output:

[31.5, 'Robotics', 4, True, 9]

Lists
• Lists can nest other lists:
h = ["wheels", 35, 5.2, [10, 20], 50]

print(h)
print(h[3])
print(h[3][0])
print(h[3][1])
Output:

['wheels', 35, 5.2, [10, 20], 50]

[10, 20]
10
20
Lists
• Lists allow duplicates of data:
j = ["apple", "banana", "cherry", "apple", "watermelon"]

print(j)

Output:

['apple', 'banana', 'cherry',

'apple', 'watermelon']
Accessing Items in a List
• To access individual items in a list, square brackets are used which contain the
index number of the item
• The index of the first item starts at zero
• The index of the last item is (number of items – 1)

0 1 2 3 4 5 6 7

my_list = [14, 20, 93, 41, 56, 77, 38, 62]

print(my_list[0]) Output:
print(my_list[1])
print(my_list[5]) 14
print(my_list[7]) 20
77
62
Changing Items in a List
• The value of an item in a list can be changed by using its index
0 1 2 3 4 5 6 7

my_list = [14, 20, 93, 41, 56, 77, 38, 62]

print(my_list)

my_list[2] = 37
print(my_list)

Output:

[14, 20, 93, 41, 56, 77, 38, 62]

[14, 20, 37, 41, 56, 77, 38, 62]
Accessing Items with Negative Index
• Items can also be indexed from the end. This is done by using negative numbers
for the index
• The negative indexing starts from -1 (not zero)

0 1 2 3 4 5 6 7
my_list = [14, 20, 93, 41, 56, 77, 38, 62]
-8 -7 -6 -5 -4 -3 -2 --1

print(my_list[-1]) Output:
print(my_list[-2])
print(my_list[-3]) 62
print(my_list[-8]) 38
77
14
Changing Items with Negative Index
• Negative indexing can also be used to change the value of items
-8 -7 -6 -5 -4 -3 -2 --1

my_list = [14, 20, 93, 41, 56, 77, 38, 62]

print(my_list)

my_list[-5] = 500
print(my_list)

Output:

[14, 20, 93, 41, 56, 77, 38, 62]

[14, 20, 93, 500, 56, 77, 38, 62]
Accessing Range of Items
• The access a range of items, the colon (:) is used (slice operation)
• Note the post-colon number is NOT included in the index
0 1 2 3 4 5 6 7

my_list = [14, 20, 93, 41, 56, 77, 38, 62]

print( my_list[2:5] ) 0 1 2 3 4 5 6 7

print( my_list[2:] ) 0 1 2 3 4 5 6 7

print( my_list[:5] ) 0 1 2 3 4 5 6 7

Output: [93, 41, 56]

[93, 41, 56, 77, 38, 62]
[14, 20, 93, 41, 56]
 Changing Range of Items
• The range of items, accessed with the colon (:), can be changed

0 1 2 3 4

my_list = ["apple", "banana", "cherry", "orange", "mango"]

print( my_list)

my_list[2:4] = ["grapes", "melon"]

print( my_list)

Output: ['apple', 'banana', 'cherry', 'orange', 'mango']

['apple', 'banana', 'grapes', 'melon', 'mango']
 Changing Range of Items
• If you insert more items than you replace, the list will increase in length
• In the example below, banana and cherry are replaced by 3 items

0 1 2 3

my_list = ["apple", "banana", "cherry", "orange“]

print( my_list)

my_list[1:3] = ["carrot", "potato", "turnip"]

print( my_list)

Output: ['apple', 'banana', 'cherry', 'orange']

['apple', 'carrot', 'potato', 'turnip', 'orange']
 Changing Range of Items
• If you insert less items than you replace, the list will decrease in length
• In the example below, banana and cherry are replaced by 1 item

0 1 2 3

my_list = ["apple", "banana", "cherry", "orange"]

print( my_list)

my_list[1:3] = ["strawberry"]
print( my_list)

Output: ['apple', 'banana', 'cherry', 'orange']

['apple', 'strawberry', 'orange']
 Checking Items in List
• The in and not in keywords can used to check for items in a list:
my_list = ["apple", "banana", "cherry", "orange"]

print( "orange" in my_list )

print( "mango" in my_list )

print( "orange" not in my_list )

print( "mango" not in my_list )

Output: True
False
False
True
 Looping a List
• The for loop can used to iterate through the items of a list
• The number of times the loop executes is equal to the number of items
• The iterator (fruit) takes the value of each item of every iteration
my_list = ["apple", "banana", "cherry", "orange"]

for fruit in my_list:

print(fruit)

Output: apple
banana
cherry
orange
 Looping a List
• If the indices are needed in the loop, then the range and length functions can be
used

my_list = [1,2,3,4,5]

for i in range(len(my_list)):
my_list[i] = 2 * my_list[i]

print(my_list)

Output: [2, 4, 6, 8, 10]

List Functions
• List functions (also called methods) are useful functions that can be used with a list
• Python provides the following list functions:
o Append()
o Insert()
o Extend() Functions to Add Items
o Remove()
o Pop() Functions to Remove Items
o Count()
o Index()
o Reverse()
 Functions to Add Items
• To add items, Python provides 3 list functions:
• Append
• Insert
• Extend

• The append function adds an item to the end of a list:

listA = ["apple","banana","cherry"]
print(listA)

listA.append("mango")
print(listA)

Output: ['apple', 'banana', 'cherry']

['apple', 'banana', 'cherry', 'mango']
 Functions to Add Items
• The insert function adds an item to a certain index
• The items are readjusted and the list size increases
listA = ["apple","banana","cherry"]
print(listA)

listA.insert(1, "mango")
print(listA)

Output: ['apple', 'banana', 'cherry']

['apple', 'mango', 'banana', 'cherry']
 Functions to Add Items
• The extend function adds a list to another list
listA = ["apple","banana","cherry"]
print(listA)

listB = ["carrot", "onion"]

listA.extend(listB)
print(listA)

• Another way to do this is to use the addition operation to concatenate the lists
(listA = listA + listB)

Output: ['apple', 'banana', 'cherry']

['apple', 'banana', 'cherry', 'carrot', 'onion']
 Functions to Remove Items
• To remove items, Python provides 2 list functions:
• Pop
• Remove

• The remove function can remove a specified item:

listA = ["apple","banana","cherry", "mango"]

print(listA)

listA.remove("cherry")
print(listA)

Output: ['apple', 'banana', 'cherry', 'mango']

['apple', 'banana', 'mango']
 Functions to Remove Items
• The pop function can remove an item from a specific index:

listB = ["apple","banana","cherry", "mango"]

print(listB)

listB.pop(1)
print(listB)

Output: ['apple', 'banana', 'cherry', 'mango']

['apple', 'cherry', 'mango']
Other List Functions
• The count function returns the total number of a specified item
listC = [1,1,3,4,3,5,7,4,3,7,8,3,2]
number = listC.count(3)
print(number)

Output: 4

• The index function returns the index of the first occurrence of a specified item

index_val = listC.index(3)
print(index_val)

Output: 2
Other List Functions
• The reverse function reverses the order of the items in the list
listA = ["apple","banana","cherry", "mango"]
listA.reverse()
print(listA)

Output: ['mango', 'cherry', 'banana', 'apple']

2-D Lists
• The following is a 2-D list (a list of lists)
• The element of such lists can be accessed by multiple indexes
• The first index selects among the “items” of the outer list
• The second index selects among the items of the inner sub-list
• Remember that indexing in python starts from zero
my_2d_list = [[11, 22, 33, 44, 55],
[71, 27, 42, 99, 58],
[61, 62, 63, 64, 65],
[83, 85, 19, 24, 21]]

print( my_2d_list[2][3] )

Output: 64
Strings - Review
• Python supports the string data type which is an array of characters
h = “Manipulator”
print(h) Output: Manipulator

• You can get individual characters with square brackets

print(h[0]) Output: M
print(h[1]) a
print(h[8]) p

• You can get the number of characters with the len() function
g = len(h)
print(g) Output: 11
Strings - Review
• You can concatenate strings easily in python
c = “Computer”
d = “Vision” Output: ComputerVision
e = c + d
print(e)

f = c + “ “ + d
print(f) Output: Computer Vision

• You can check if a character is present in the string with the “in” keyword

print(“t” in c) Output: True

print(“s” in c) False
Strings - Review
• The for loop can go through a sequence of characters
• The iterable object will be a string variable in this case
for i in "PYTHON": Output:
print(i) P
Y
T
H
O
N
Dictionaries
• A dictionary is another data structure
• A dictionary is like a list but somewhat more general
• In a list, index positions have to be integers; in a dictionary, index positions can be
(almost) any type

• A Dictionary is a collection of data that is

o Ordered (as of Python 3.7)
o Changeable (Mutable)
o Does NOT Allow Duplicate Members
 Dictionaries - Creation
• A dictionary is a mapping of a set of indices (called keys) to a set of values
• Each key maps to a value
• Each key-value pair is an item of a dictionary
• To create a dictionary, key-value pairs are enclosed in braces

eng2span = {'one': 'uno',

'two': 'dos',
'three': 'tres'}

print(eng2span)

Output: {'one': 'uno', 'two': 'dos', 'three': 'tres'}

 Dictionaries - Accessing
• The following dictionary maps English (keys) to Spanish (values)
eng2span = {'one': 'uno',
'two': 'dos',
'three': 'tres'}

• The keys are used as indexes to look up the corresponding values

print(eng2span['one'])
print(eng2span['two'])
print(eng2span['three'])

uno
Output: dos
tres
 Dictionaries - Accessing
• The following dictionary maps English (keys) to Numbers (values)
eng2num = {'one': 1,
'two': 2,
'three': 3}

• The keys are used as indexes to look up the corresponding values

print(eng2num['one'])
print(eng2num['two'])
print(eng2num['three'])

1
Output: 2
3
 Dictionaries – Changing Value
• Consider the following dictionary which maps prices
prices = {'burger': 250,
'sandwich': 150,
'pizza': 400}

• The value can be changed by using its key:

prices['sandwich'] = 180
print(prices)

{'burger': 250, 'sandwich': 180, 'pizza': 400}

Output:
Importing Modules
• In python, we use the import keyword to use libraries (modules) which contain
functions and classes
• There are many libraries commonly used in python such as NumPy, Pandas,
OpenCV, Matplotlib, SciPy, Tensorflow, Keras, PySerial, Math and Datetime etc

• We will use the Math library as an example:

import math 8.0

var = math.sqrt(64)
print(var)
• We can also assign an alias for the library:

import math as mt 5.0

var = mt.sqrt(25)
print(var)
Importing Modules
• The math library contains some common mathematical functions:
print( math.sqrt(64)) 8.0
print( math.ceil(1.5) ) 2
print( math.floor(1.5) ) 1
print( math.pi ) 3.141592653589793
print( math.inf ) inf
print( math.sin(30) ) -0.9880316240928618
print( math.cos(30) ) 0.15425144988758405
print( math.tan(30) ) -6.405331196646276
print( math.log(5) ) 1.6094379124341003
print( math.log10(5) ) 0.6989700043360189
print( math.radians(180) ) 3.141592653589793
print( math.degrees(math.pi) ) 180.0
print( math.exp(4) ) 54.598150033144236
NumPy
• NumPy (Numerical Python) is a library used for creating computationally efficient
arrays for matrix operations
import numpy as np
A = np.array([[1, 2, 3],
[4, 5, 6]])

• Numpy offers wide range of functions on the arrays

A + B np.dot(A,B)
A - B np.sum(A, axis=0)
A*B np.mean(A, axis=0)
A/B np.std(A, axis=1)
SciPy
• SciPy (Scientific Python) is a library used to implement different scientific functions and
is compatible with numpy arrays
from scipy.optimize import root
def equation(x):
return x + x*sin(x)
output = root(equation, 0)
print(output.x)

• In machine learning, SciPy is used mostly for minimization functions, sparse

matrices, interpolation etc.
Lab Tasks

 Download the materials from LMS

 Perform the Lab Tasks given in the manual
 Convert the completed manual into .pdf and submit on LMS