CODEWITHSUNDEEP
pythonarraysnumpyloops
Thomas
Thomas

Reputation: 51

How would I perform an operation on each element in an array and add the new value to a new array in Python?

I have an array of ten random values generated using Numpy. I want to perform an operation on each element in the array by looping over all ten elements and then add the result of each to a new array. The first part of looping over the array but I am stuck on how to add the result to a new array.

So far I have tried something along the lines of:

import numpy as np

array = np.random.rand(10)

empty_array = np.zeros(10)

for elem in array:

    new_val = elem**2

where empty_array is an array of 10 elements set to zero, my logic being similar to initialising an empty list to add elements to when looping over another list or array. I am stuck on how to replace the corresponding elements of empty_array with the square of elements in 'array'.

Any help on how to do this or a better way of doing this would be greatly appreciated!

Upvotes: 0

Views: 1860

Answers (3)

AJH
AJH

Reputation: 799

This is a case where NumPy vectorisation comes in quite handy:

import numpy as np

array = np.random.rand(10)
array2 = array*array

And if you were dead set on calculating things in a loop (which I wouldn't recommend, but each to their own), you would add the new value to empty_array like this:

import numpy as np

array = np.random.rand(10)
empty_array = np.zeros(10)

for i in range(array.shape[0]):

    # Get element and calculate square.
    # Note that using elem**2 is in some cases less efficient than using elem*elem.
    elem = array[i]
    new_val = elem**2

    empty_array[i] = new_val

Upvotes: 0

chepner
chepner

Reputation: 530930

One of the reasons to use Numpy is its expressiveness for cases like these.

>>> import numpy as np
>>> a = np.random.rand(10)
>>> a
array([0.94797029, 0.39628409, 0.64609633, 0.44994779, 0.23083464,
       0.60191075, 0.71651581, 0.78152364, 0.05516691, 0.22452054])
>>> a **2
array([0.89864768, 0.15704108, 0.41744046, 0.20245301, 0.05328463,
       0.36229656, 0.5133949 , 0.6107792 , 0.00304339, 0.05040947])

You don't have to iterate over the original array and build up the new array step by step: you just "square" the array itself.

Upvotes: 2

K9.dev
K9.dev

Reputation: 369

You can append the new value like:

import numpy as np

array = np.random.rand(10)
new_array = []
for elem in array:
    new_array.append(elem ** 2)

Or using the list comprehension:

import numpy as np

array = np.random.rand(10)
new_array = [e**2 for e in array]

Upvotes: 0

Related Questions