Combining fancy and slice indexing in NumPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is fancy indexing in numpy?

Fancy indexing means selecting elements from an array using another array of indices. It allows picking specific elements in any order.

beginner

How does slice indexing differ from fancy indexing?

Slice indexing selects a continuous range of elements using start:stop:step syntax, while fancy indexing selects elements at specific positions given by an array of indices.

intermediate

What happens when you combine fancy and slice indexing in numpy?

You can select specific rows or columns using fancy indexing and then select a slice of elements within those rows or columns. This lets you pick complex subsets of data.

beginner

Example: Given a 2D array, how to select rows 1 and 3, and columns 2 to 4?

Use arr[[1, 3], 2:5] to select rows 1 and 3 with columns from 2 up to 4 (5 is exclusive).

intermediate

Why is the order of fancy and slice indexing important in numpy?

The order matters because numpy first applies fancy indexing to select specific rows or columns, then applies slicing on the result. Changing order can change the output shape and values.

What does arr[[0, 2], 1:3] select from a 2D numpy array arr?

A Rows 1 and 3, columns 0 and 2

B Columns 0 and 2, rows 1 and 3

C Rows 0 and 2, columns 1 and 2

D Rows 0 to 2, columns 1 to 3

Fancy indexing selects rows 0 and 2, then slice 1:3 selects columns 1 and 2.

Which indexing method allows selecting elements at arbitrary positions in numpy?

A Fancy indexing

B None of the above

C Boolean indexing

D Slice indexing

Fancy indexing uses arrays of indices to select arbitrary elements.

What is the result shape of arr[[1, 2], 0:2] if arr has shape (4, 5)?

A (4, 2)

B (2, 2)

C (2, 5)

D (4, 5)

Selecting 2 rows (1 and 2) and slicing 2 columns results in shape (2, 2).

Can you combine fancy indexing on rows with slice indexing on columns in numpy?

A Yes, like arr[[0, 3], 1:4]

B No, numpy does not allow mixing

C Only if array is 1D

D Only if array is 3D

Numpy supports combining fancy and slice indexing on different axes.

What happens if you reverse the order: arr[1:3, [0, 2]] instead of arr[[1, 3], 0:2]?

A Selects all rows and columns

B Selects rows 1 and 3, columns 0 and 1

C Raises an error

D Selects rows 1 and 2, columns 0 and 2

Slice selects rows 1 and 2, fancy indexing selects columns 0 and 2.

Explain how to combine fancy and slice indexing to select specific rows and a range of columns from a 2D numpy array.

Think about selecting rows first, then columns.

Describe the difference between fancy indexing and slice indexing and how they can be used together in numpy.

Compare how each indexing method selects elements.