Why indexing matters in NumPy - Challenge Your Understanding

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of slicing vs. fancy indexing in NumPy

What is the output of the following code snippet?

NumPy

import numpy as np
arr = np.array([10, 20, 30, 40, 50])
slice_view = arr[1:4]
slice_view[0] = 99
print(arr)
fancy_index = arr[[1, 2, 3]]
fancy_index[0] = 77
print(arr)

A [10 20 30 40 50]\n[10 20 30 40 50]

B [10 20 30 40 50]\n[10 77 30 40 50]

C [10 99 30 40 50]\n[10 99 30 40 50]

D [10 99 30 40 50]\n[10 20 30 40 50]

💡 Hint

Remember that slicing returns a view, while fancy indexing returns a copy.

Slicing in NumPy returns a view of the original array, so modifying the slice changes the original array. Fancy indexing returns a copy, so modifying it does not affect the original array.

❓ data_output

intermediate

Number of elements selected by boolean indexing

Given the array and boolean mask below, how many elements are selected?

NumPy

import numpy as np
arr = np.array([5, 10, 15, 20, 25, 30])
mask = arr > 15
selected = arr[mask]
print(len(selected))

A 3

B 4

C 2

D 5

💡 Hint

Count how many values in arr are greater than 15.

The values greater than 15 are 20, 25, and 30, so 3 elements are selected.

🔧 Debug

advanced

Why does this indexing code raise an error?

What error does the following code raise and why?

NumPy

import numpy as np
arr = np.array([[1, 2], [3, 4], [5, 6]])
print(arr[[0, 2], 1, 0])

A IndexError: too many indices for array

B TypeError: unhashable type: 'list'

C ValueError: shape mismatch

D No error, prints [2 6]

💡 Hint

Check how many indices are used for a 2D array.

The array is 2D, but three indices are used. This causes an IndexError for too many indices.

🚀 Application

advanced

Selecting rows with multiple conditions using indexing

Which code correctly selects rows from the array where the first column is > 2 and the second column is < 5?

NumPy

import numpy as np
arr = np.array([[1, 4], [3, 2], [5, 6], [4, 3]])

A arr[arr[:, 0] > 2, arr[:, 1] < 5]

B arr[arr[:, 0] > 2 and arr[:, 1] < 5]

C arr[(arr[:, 0] > 2) & (arr[:, 1] < 5)]

D arr[(arr[:, 0] > 2) | (arr[:, 1] < 5)]

💡 Hint

Use bitwise operators (&, |) for element-wise logical operations in NumPy.

Option C uses element-wise logical AND with & and parentheses, which is correct. Option C uses OR, which selects more rows. Option C uses Python 'and' which is invalid for arrays. Option C uses wrong indexing syntax.

🧠 Conceptual

expert

Why does modifying a slice affect the original array but modifying a fancy index does not?

Choose the best explanation for why modifying a slice of a NumPy array changes the original array, but modifying a fancy indexed array does not.

A Slices create copies of the data; fancy indexing returns views that share memory.

B Slices return views that share memory with the original array; fancy indexing returns a copy with separate memory.

C Both slices and fancy indexing return copies, but slices are linked to the original array by reference.

D Fancy indexing modifies the original array only if the array is 1D; slices always modify the original.

💡 Hint

Think about memory sharing between the original array and the indexed result.

Slices in NumPy return views, which means they share the same memory as the original array. Changes to the slice affect the original. Fancy indexing returns a new array copy, so changes do not affect the original.