u10-slides-numpy

Question 1

What is the main purpose of NumPy?

Answer 1

It’s the go-to module for numerical computations in Python, providing optimized functions for scientific computations and array data handling

Question 2

Why is NumPy faster than standard Python?

Answer 2

NumPy functions are highly optimized and implemented in C, and arrays store elements in contiguous memory blocks

Question 3

How are elements stored in NumPy arrays vs Python lists?

Answer 3

In NumPy arrays, elements are stored as direct bit patterns in contiguous memory; in Python lists, elements are stored as references to objects

Question 4

What is a key difference in data types between NumPy arrays and Python lists?

Answer 4

NumPy arrays have fixed data types for all elements, while Python lists can contain elements of different types

Question 5

What is meant by ‘contiguous memory’ in NumPy arrays?

Answer 5

Elements are stored as one block with consecutive memory addresses, making access faster

Question 6

What is row-major order in NumPy?

Answer 6

The default storage method where consecutive elements of a row reside next to each other in memory

Question 7

What is column-major order in NumPy?

Answer 7

A storage method where consecutive elements of a column reside next to each other in memory

Question 8

How do you access elements in a multi-dimensional NumPy array?

Answer 8

Using comma-separated indices, e.g., my_array[row, col]

Question 9

What is the shape of a NumPy array?

Answer 9

It defines the multi-dimensionality of an array, specifying the length of each dimension

Question 10

What is an axis in NumPy?

Answer 10

A specific dimension of an array; a 2D array has 2 axes, a 3D array has 3 axes, etc.

Question 11

How are 2D arrays stored in NumPy internally?

Answer 11

They’re stored as 1D arrays in a flat manner, with a mapping system to maintain the dimensional structure

Question 12

Why is slicing fast in NumPy arrays?

Answer 12

Because elements are stored consecutively in memory, making it efficient to access continuous sections

Question 13

What are the indexing options available in NumPy?

Answer 13

Integer indexing, slicing, multi-dimensional indexing, boolean masking, and list of indices

Question 14

How is memory calculated for accessing elements in a 2D array?

Answer 14

For position [r,c] in an array with n columns: index = n * r + c

Question 15

What is the main advantage of NumPy’s storage method?

Answer 15

Operations on elements can be optimized and are faster due to fixed data types and contiguous memory storage

Question 16

What is the primary data structure in NumPy?

Answer 16

The numpy.ndarray object (N-dimensional array)

Question 17

Why does NumPy require all elements in an array to have the same data type?

Answer 17

For optimization purposes - operations can be performed faster when the data type is fixed and known

Question 18

What happens when you reshape a NumPy array?

Answer 18

The same data is viewed in a different dimensional structure, without changing the underlying memory layout

Question 19

How does NumPy’s memory efficiency compare to Python lists?

Answer 19

NumPy is more memory efficient as it stores values directly rather than object references

Question 20

What is the relationship between dimensions and axes in NumPy?

Answer 20

They are equivalent - the number of dimensions equals the number of axes