Divide each element of an array by each element of an other array

Question

I'm coding a function that tells which numbers are prime numbers in a given array. I would also like to do it without a for loop if it is possible

The numpy.divide function and division symbole (/) do not work as they throw a ValueError because the two arrays do not have the same shape

numbers = np.array([2, 3, 6])

def prime(numbers):
    biggest = np.max(numbers)
    division_array = np.arange(2, ceil(biggest / 2) + 1)
    # division_array = [2, 3]
    return numbers / division_array

print(numbers)

Expected output:

[[1, 0.6],
 [1.5, 1],
 [3, 2]]

Output:

ValueError: operands could not be broadcast together with shapes (4,) (3,)

So, by your (apparent) reasoning, dividing a 3-element array by another should produce a 3x3 array of results? — Scott Hunter
– Scott Hunter, Commented Jul 29, 2019 at 17:37
Add a new axis to numbers to use broadcasting: return numbers[:, np.newaxis] / division_array — javidcf
– javidcf, Commented Jul 29, 2019 at 17:37
@ScottHunter No but a 3-element array by a 2-element array should produce a 3x2 array, not a 3x3. — Docaro
– Docaro, Commented Jul 29, 2019 at 17:40
@jdehesa Thanks it work, could you put it as an answer so that i accept it. — Docaro
– Docaro, Commented Jul 29, 2019 at 17:43

javidcf · Accepted Answer · 2019-07-29 17:46:35Z

2

Since you want an "all-to-all" operation, the best thing is to broadcast the arrays. Maybe the easiest way is to add a dimension to numbers with np.newaxis and divide:

import numpy as np

numbers = np.array([2, 3, 6])

def prime(numbers):
    biggest = np.max(numbers)
    division_array = np.arange(2, np.ceil(biggest / 2) + 1)
    # division_array = [2, 3]
    return numbers[:, np.newaxis] / division_array

print(prime(numbers))
# [[1.         0.66666667]
#  [1.5        1.        ]
#  [3.         2.        ]]

answered Jul 29, 2019 at 17:46

javidcf

59.9k7 gold badges87 silver badges134 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Docaro Over a year ago

Thanks, you just reminded me of how I did it : I reshaped it to a Nx1 array

Sheshank S. · Accepted Answer · 2019-07-29 17:47:58Z

1

For anyone who doesn't want to use numpy, I created a plain python function for this.

I created this function that you could use to get what you need: https://repl.it/repls/ImpartialVainWebsites

array1 = [2, 3, 6]
array2 = [2, 3]

def divide_array(array1, array2):
  new_array = list()
  for element in array1:
    array1_list = list()
    for element2 in array2:
      array1_list.append(round(element/element2, 3))
    new_array.append(array1_list)
  return new_array

print(divide_array(array1, array2))

Returns:

[[1.0, 0.667], [1.5, 1.0], [3.0, 2.0]]

Or in a more compact form:

array1 = [2, 3, 6]
array2 = [2, 3]

def divide_array(array1, array2):
  new_array = list()
  for element in array1:
    new_array.append([round(element/element2, 3) for element2 in array2])
  return new_array

print(divide_array(array1, array2))

edited Jul 29, 2019 at 17:47

answered Jul 29, 2019 at 17:40

Sheshank S.

3,3184 gold badges24 silver badges42 bronze badges

5 Comments

Docaro Over a year ago

Is it possible to do it without a for loop ? I know i did it before but i forgot how...

Sheshank S. Over a year ago

@Docaro I changed it to use list comprehension, would that work better for you?

Docaro Over a year ago

Not really as it is still a for loop so for really big lists it will take a long time

Sheshank S. Over a year ago

Okay I guess, I'll just keep the answer as it as for people in the future who might want to do it without numpy

Docaro Over a year ago

Thanks anyway for helping :)

Collectives™ on Stack Overflow

Divide each element of an array by each element of an other array

2 Answers 2

1 Comment

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related