Turning a 1-D numpy array into a N-d numpy array

Question

I'm not sure if this is called broadcasting or something else, but supposed I'm given a 1-D numpy array of size L. I want to turn this into an n-dimensional numpy array, where one of the dimensions is of size L and the other dimensions are arbitrary, but greater than 0.

e.g., suppose L = 2 and I want to turn this into a n=2-dimensional array of size 2x3 (2 rows, 3 columns), in this case, each row will consist of the original 1-D array.

Is there a function that will allow me to do this in numpy?

First reshape it to (2,1). Expanding to (2,3) will be then be easier (and easier to visualize). — hpaulj
– hpaulj, Commented Feb 18, 2022 at 3:56

Ben Grossmann · Accepted Answer · 2022-02-18 03:07:04Z

1

One approach is to use broadcast_to. You could also use the tile function to a similar effect.

import numpy as np

a = np.arange(10)
print('Array before:')
print(a)
print('Array after:')
np.broadcast_to(a,(3,10))

Result:

Array before:
[0 1 2 3 4 5 6 7 8 9]
Array after:

array([[0, 1, 2, 3, 4, 5, 6, 7, 8, 9],
       [0, 1, 2, 3, 4, 5, 6, 7, 8, 9],
       [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]])

answered Feb 18, 2022 at 3:07

Ben Grossmann

5,0471 gold badge15 silver badges22 bronze badges

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

2 Comments

roulette01 Over a year ago

Ah I think this is it, but why can't I do np.broadcast_to(a,(10,3))? This gives the error

ValueError: operands could not be broadcast together with remapped shapes [original->remapped]: (10,)  and requested shape (10,3)

hpaulj Over a year ago

By the rules of broadcasting, new leading dimensions are automatic, but trailing dimensions have to be explicit. np.broadcast_to(a[:,None],(10,3)). expands a (10,1) to (10,3). Ben's example expands a (10,) to (1,10) to (3,10).

Collectives™ on Stack Overflow

Turning a 1-D numpy array into a N-d numpy array

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related