Sorting 2D Arrays

As you can see, simply passing our 2D array to the sort() function sorts each 1D array along the axis 1 (which is the default option in a 2D array). Setting axis=0 sorts each 1D array along the axis 0 (every column).

Setting axis=None returns a contiguous sorted 1D array of all the elements of the 2D array.


              12345678
            
import numpy as np
array_2d = np.array([[2, 9, 3], [1, 6, 4], [5, 7, 8]])
# Sorting a 2D array along axis 1
print(np.sort(array_2d))
# Sorting a 2D array along axis 0
print(np.sort(array_2d, axis=0))
# Creating a 1D sorted array out of the elements of array_2d
print(np.sort(array_2d, axis=None))

Sorting 2D Arrays in Descending Order

When sorting 2D arrays in descending order along a given axis, you need to use two slices: one full slice ([:]) and another with a negative step ([::-1]). The position of the slice with the negative step should correspond to the axis along which you are sorting.

Note

When sorting along axis 0, you can use only a slice with a negative step, as it already indexes along this axis.


              12345678
            
import numpy as np
array_2d = np.array([[2, 9, 3], [1, 6, 4], [5, 7, 8]])
# Sorting a 2D array along axis 1 in descending order
print(np.sort(array_2d)[:, ::-1])
# Sorting a 2D array along axis 0 in descending order
print(np.sort(array_2d, axis=0)[::-1])
# Creating a 1D sorted array out of the elements of array_2d in descending order
print(np.sort(array_2d, axis=None)[::-1])

Task

Swipe to show code editor

You have a 2D array named exam_scores containing the scores for each exam from a certain subject. Each column represents a specific subject, and each row represents an individual student. Thus, a specific row displays the scores of that student for each exam.

Create a 2D NumPy array named top_scores_subject based on exam_scores where each column, representing a certain subject, is sorted by scores in descending order.
Create a 1D NumPy array named sorted_scores based on exam_scores, containing all scores sorted in ascending order.

By doing this, you can easily identify the highest scores for each exam and the lowest scores across all exams.

Switch to desktop for real-world practiceContinue from where you are using one of the options below

Everything was clear?

Thanks for your feedback!

Section 3. Chapter 2

Sorting 2D Arrays

Setting axis=None returns a contiguous sorted 1D array of all the elements of the 2D array.


              12345678
            
import numpy as np
array_2d = np.array([[2, 9, 3], [1, 6, 4], [5, 7, 8]])
# Sorting a 2D array along axis 1
print(np.sort(array_2d))
# Sorting a 2D array along axis 0
print(np.sort(array_2d, axis=0))
# Creating a 1D sorted array out of the elements of array_2d
print(np.sort(array_2d, axis=None))

Sorting 2D Arrays in Descending Order

Note

When sorting along axis 0, you can use only a slice with a negative step, as it already indexes along this axis.


              12345678
            
import numpy as np
array_2d = np.array([[2, 9, 3], [1, 6, 4], [5, 7, 8]])
# Sorting a 2D array along axis 1 in descending order
print(np.sort(array_2d)[:, ::-1])
# Sorting a 2D array along axis 0 in descending order
print(np.sort(array_2d, axis=0)[::-1])
# Creating a 1D sorted array out of the elements of array_2d in descending order
print(np.sort(array_2d, axis=None)[::-1])

Task

Swipe to show code editor

Create a 2D NumPy array named top_scores_subject based on exam_scores where each column, representing a certain subject, is sorted by scores in descending order.
Create a 1D NumPy array named sorted_scores based on exam_scores, containing all scores sorted in ascending order.

By doing this, you can easily identify the highest scores for each exam and the lowest scores across all exams.

Switch to desktop for real-world practiceContinue from where you are using one of the options below

Everything was clear?

Thanks for your feedback!

Section 3. Chapter 2

Switch to desktop for real-world practiceContinue from where you are using one of the options below