07 Numerical Computation

Question 1

One dimensional array

Answer 1

Collection of numbers which we can access by index

Question 2

What is the default type of NumPy array

Answer 2

Float

Question 3

Shape

Answer 3

Tells us size of the array in each direction

Question 4

arange

Answer 4

Creates an array with equally spaced values

Question 5

What is the downside of the flexibility in python lists?

Answer 5

performance
slow computation speed

Question 6

What are 3 ways to create arrays in python?

Answer 6

• np.array(x)
• numpy.arange(x)
• numpy.linspace(x)

Question 7

arange

Answer 7

creates an array with equally spaced values

Question 8

linspace

Answer 8

creates array with prescribed start and end values, all equally spaced

Question 9

do NumPy arrays support common arithmetic operations?

Answer 9

Yes

Question 10

Vectorisation

Answer 10

manipulation of arrays without indexing

Question 11

How does the performance of specialised functions compare to your own general implementations?

Answer 11

Specialised functions many orders of magnitude faster
e.g. NumPy sqrt vs computing it yourself

Question 12

How do we index a 2D array?

Answer 12

Using two indices, first for row index and second for column index

Question 13

How do we compute matrix multiplication

Answer 13

_= .dot()

Question 14

How do you compute the inverse of a matrix?

Answer 14

Ainv = np.linalg.inv(A)

Question 15

How do you compute the determinant for a matrix

Answer 15

Adet = np.linalg.det(A)

Question 16

Array slicing

Answer 16

Extract subsets of an array

Question 17

What is something you have to watch out for in array slicing

Answer 17

y = x[1:3] will get values 1 to 3 but NOT including 3

Question 18

How do we slice from the start of an array

Answer 18

e.g. y = x[:3]
start to (3)

Question 19

How do we slice from the end of an array

Answer 19

y = x[4:]