Numpy Session 02: Class Activities

Dr. Muhammad Abbas Abbasi

07-11-2023

Task 1:
In this task, you will work with NumPy arrays and practice various slicing operations. Make sure to write your Python code
using NumPy to solve the following sub-parts. Use the provided NumPy arrays for your operations.

1. Perform basic slicing on the NumPy array:

 
arr = 10 20 30 40 50

Extract elements from index 2 to 4. Store the result in a new array and display it.
2. Create a 2D NumPy array:  
1 2 3
matrix = 4 5 6
7 8 9
Slice it to obtain the second column as a 1D array. Display the result.

3. Given a 2D NumPy array:  

11 12 13 14
mat = 21 22 23 24
31 32 33 34
Extract a sub-matrix containing the second and third rows and the first and third columns. Display the sub-matrix.
4. You have a 3D NumPy array tensor with shape (3, 4, 5). The array is defined as follows:
     
[1 2 3 4 5] [21 22 23 24 25] [41 42 43 44 45]
 [6 7 8 9 10]  [26 27 28 29 30] [46 47 48 49 50] 
[11 12 13 14 15] , [31 32 33 34 35] , [51 52 53
tensor =      
54 55]
[16 17 18 19 20] [36 37 38 39 40] [56 57 58 59 60]

Slice this tensor to get a 2D array with the shape (4, 5) by selecting the second 2D array along the first dimension.
Display the result.

5. Suppose you have a NumPy array:

 
data = 1 2 3 4 5 6 7 8 9

Extract all even numbers from data using boolean indexing and store them in a new array. Display the resulting array.

1
Task 2:
Write Python code using NumPy to solve each sub-part. Use the provided NumPy arrays for your operations.

1. You are working with a grayscale image represented as a 2D NumPy array. The image is defined as follows:
 
[10 20 30 40 50]
 [60 70 80 90 100]
 
image = [110 120 130 140 150]

[160 170 180 190 200]
[210 220 230 240 250]

Slice a specific region of interest (ROI) from the image (e.g., from row 2 to 4 and from column 2 to 4) and display only
that part.

2. Imagine you have a dataset of historical weather temperature records for a city stored in a 2D NumPy array. The dataset
is defined as follows:  
[75.2 78.3 81.5 79.0 82.1]
[71.6 73.8 76.2 72.4 77.5]
temperature data = [69.3 70.7 73.2 71.1 74.8]


[74.8 76.2 78.4 76.0 80.5]

Slice the data to extract the temperature records for the third day (row 2) and the time range from the second hour to the
fourth hour (columns 1 to 3).
3. Consider a financial dataset stored in a 2D NumPy array. The dataset contains the daily stock prices of three different
stocks. It is defined as follows:
 
[50.0 51.2 49.5 48.9 50.3]
stock data = [62.1 60.5 63.2 59.7 61.0]
[90.5 88.7 91.2 92.0 89.5]

Slice the data to extract the stock prices for the second stock (row 1) and the time range from the first day to the third
day (columns 0 to 2).
4. You have a dataset of student exam scores stored in a 2D NumPy array, where each row represents a different student,
and each column is a different exam score. The dataset is defined as follows:
 
[85 92 78]
[90 88 76]
student scores =  [78 85 92]


[95 91 89]

Slice the data to obtain the scores of the third student (row 2) and calculate their overall performance.

Task 3:
In this task, you will work with 3D NumPy arrays and practice slicing and indexing operations. Write Python code using
NumPy to solve the following sub-parts. Use the provided 3D NumPy arrays for your operations.

a) You have a 3D NumPy array cube with shape (3, 4, 5). The array is defined as follows:
     
[1 2 3 4 5] [21 22 23 24 25] [41 42 43 44 45]
 [6 7 8 9 10] [26 27 28 29 30] [46 47 48 49 50] 
cube = 
[11
, , 
12 13 14 15] [31 32 33 34 35] [51 52 53 54 55]
[16 17 18 19 20] [36 37 38 39 40] [56 57 58 59 60]

Perform the following operations:

2
 • Extract the element at coordinates (2, 3, 4).
• Extract the second 2D array along the first dimension (at depth 1).
• Extract the first row of the second 2D array (at depth 1).
• Extract a 2x3 sub-matrix from the third 2D array (at depth 2) starting from the coordinates (1, 1).
b) You have a 3D NumPy array data cube with shape (4, 3, 3). The array is defined as follows:
       
[10 20 30] [110 120 130] [210 220 230] [310 320 330]
data cube = [40 50 60] , [140 150 160] , [240 250 260] , [340 350 360]
[70 80 90] [170 180 190] [270 280 290] [370 380 390]
Perform the following tasks:
• Extract the element at coordinates 2, 1, 2.
• Extract the entire 2D array at depth 3.
• Extract the last column of the first 2D array at depth 1.
• Extract a 2x2 sub-matrix from the third 2D array at depth 2 starting from the coordinates (0, 1).
c) You have a 3D NumPy array color cube with shape (2, 3, 4) representing a 3D RGB color image. The array contains
RGB values for a 2x3 grid of pixels for each of the 4 depth levels. The array is defined as follows:
   
[255 0 0] [128 128 0]
color cube =  [0 255 0]  ,  [0 128 128]
[0 0 255] [128 0 128]
• Extract the RGB value at coordinates (1, 2, 3).
• Extract the entire 2D array at depth 1.
• Extract the green channel (middle value) from the 2D array at depth 0.
• Extract a 2x2 sub-matrix from the first 2D array at depth 0 starting from the coordinates (0, 1).

Task 4:
In this task, you will work with NumPy arrays and practice boolean indexing operations. Write Python code using NumPy to
solve the following sub-parts. Use the provided arrays for your operations.
1. You have a 2D NumPy array matrix with shape (4, 4). The array is defined as follows:
 
1 2 3 4
5 6 7 8
matrix =  
 9 10 11 12
13 14 15 16
a) Find all elements in the 2D array matrix that are greater than 8.
b) Extract a sub-array from the 2D array matrix containing all elements that are divisible by 3.
2. You have a 3D NumPy array data cube with shape (4, 3, 3). The array is defined as follows:
     
10 20 30 110 120 130 210 220 230 310 320 330
data cube = 40 50 60 , 140 150 160 240 250 260 340 350 360
70 80 90 170 180 190 270 280 290 370 380 390
a) Find all elements in the 3D array data cube that are greater than 100.
b) Extract a sub-array from the 3D array data cube containing all elements that are divisible by 20.
c) Create a boolean mask for elements in the 3D array data cube greater than 250 and less than 350, and then use
the mask to extract the corresponding elements.

3
Task 5:
In this task, you will work with NumPy arrays and explore the concept of broadcasting. Write Python code using NumPy to
solve the following sub-parts.

Part a: Broadcasting with 1D Arrays

You have two NumPy arrays:

array1 = [10, 20, 30, 40]

array2 = [5]

1. Perform an addition operation using broadcasting to add array2 to array1.

2. Perform a multiplication operation using broadcasting to multiply array1 by array2.

3. Create a new NumPy array, array3, that is a result of adding a scalar value of 15 to array1 using broadcasting.
4. Create a new NumPy array, array4, that is a result of multiplying array1 by a scalar value of 3 using broadcasting.
5. Perform a subtraction operation using broadcasting to subtract array2 from array1.

Part b: Broadcasting with 2D Arrays

You have two NumPy arrays:
 
1 2
matrix1 = 3 4
5 6

array3 = [10, 20]

1. Perform an addition operation using broadcasting to add array3 to matrix1.

2. Perform a multiplication operation using broadcasting to multiply each column of matrix1 by array3.

3. Create a new NumPy array, matrix2, that is a result of adding a scalar value of 5 to matrix1 using broadcasting.
4. Create a new NumPy array, matrix3, that is a result of multiplying matrix1 by a scalar value of 2 using broadcasting.
5. Perform a division operation using broadcasting to divide each element of matrix1 by the corresponding element of
array3.