Fancy indexing with integer arrays in NumPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is fancy indexing in NumPy?

Fancy indexing is a way to select elements from a NumPy array using integer arrays or lists as indices, allowing you to pick multiple elements at once in any order.

beginner

How does fancy indexing differ from slicing in NumPy?

Slicing selects a continuous range of elements, while fancy indexing lets you select arbitrary elements using arrays of indices, which can be non-continuous and repeated.

beginner

Given array arr = np.array([10, 20, 30, 40, 50]), what does arr[[1, 3, 4]] return?

It returns a new array with elements at positions 1, 3, and 4: [20, 40, 50].

intermediate

Can fancy indexing be used to modify elements in a NumPy array? How?

Yes. You can assign new values to elements selected by fancy indexing. For example, arr[[1, 3]] = [100, 200] changes elements at positions 1 and 3.

intermediate

What happens if you use repeated indices in fancy indexing, like arr[[2, 2, 0]]?

The output array will include repeated elements corresponding to the repeated indices. For example, it will return elements at positions 2, 2, and 0, repeating values accordingly.

What type of object can be used for fancy indexing in NumPy?

A Only slices

B Strings

C Integer arrays or lists

D Boolean values

Fancy indexing uses integer arrays or lists to select elements by their positions.

What does arr[[0, 2, 4]] do if arr = np.array([5, 10, 15, 20, 25])?

A Raises an error

B Returns <code>[10, 20]</code>

C Returns <code>[5, 10, 15, 20, 25]</code>

D Returns <code>[5, 15, 25]</code>

It selects elements at indices 0, 2, and 4.

Can fancy indexing select elements in any order?

A Yes, order can be arbitrary

B No, only ascending order

C No, only descending order

D Only continuous ranges

Fancy indexing allows arbitrary order and repetition of indices.

What happens if you assign values using fancy indexing with repeated indices?

A All assignments are summed

B The last assignment to a repeated index is kept

C An error occurs

D Values are averaged

When assigning, repeated indices get overwritten in order, so the last value sticks.

Which of these is NOT a use case for fancy indexing?

A Selecting a continuous slice of elements

B Duplicating elements in a new array

C Reordering elements arbitrarily

D Selecting specific rows from a 2D array

Continuous slices are done with slicing, not fancy indexing.

Explain fancy indexing with integer arrays in NumPy and give a simple example.

Think about how you pick specific items from a list using their positions.

Describe how you can modify elements in a NumPy array using fancy indexing.

Consider how you change values at certain positions in a list.