Rewriting numpy function to handle 2d and 3d inputs

Question

I am trying to rewrite an numpy function such that it can deal with 2d and 3d inputs. Consider the following code:

import numpy as np
def adstock_geometric(x: np.array, theta: np.array):
    x_decayed = np.zeros_like(x)
    x_decayed[0] = x[0]

    for xi in range(1, len(x_decayed)):
        x_decayed[xi] = x[xi] + theta * x_decayed[xi - 1]

    return x_decayed

def testrun():
    rand3d = np.random.randint(0, 10, size=(4, 1000, 1)) / 10
    rand2d = np.random.randint(0, 10, size=(1000, 1)) / 10

    x = np.ones(10)

    output1d = adstock_geometric(x=x, theta=0.5)  # works fine
    output3d = adstock_geometric(x=x, theta=rand3d)
    output2d = adstock_geometric(x=x, theta=rand2d)
    
if __name__ == '__main__':
    testrun()

As you can see, it works fine for the 1d case but does not work for the 2d and 3d cases. You can imagine that the thetas are stacked in the third dimension. Expected output shape 2d: (1000, 10) Expected output shape 3d: (4, 1000, 10)

The most obvious is just to iterate through all dimensions but this is really slow.

[0] indexes the first dimension. And for 1d produces a scalar. x[...,0] would index the last, but may yield a 1 or 2d array. — hpaulj
– hpaulj, Commented Nov 20, 2023 at 17:16
I am not sure what the output should look like. Just giving the dimension of the output is not enough. Consider using a smaller example for theta, just like you used for x. ie use theta with size (4, 2), for 2d and show the expected results, and also show the results for (3,2,1) for 3d and show the results. Also notice that you can avoid the loops completely — Onyambu
– Onyambu, Commented Nov 20, 2023 at 18:34

Onyambu · Accepted Answer · 2023-11-20 18:54:15Z

1

Here is one way to approach the problem:

def addstock_geom(x, theta):
    axis = n if (n := len(np.array(theta).shape) - 1) > 0 else None
    return ((theta**np.arange(x.size)) * x).cumsum(axis)

Try using the following

x = np.ones(10)
rand1d = 0.5
rand2d = np.random.randint(0, 10, size=(5, 1)) / 10
rand3d = np.random.randint(0, 10, size=(4, 2, 1)) / 10
addstock_geom(x, rand1d)
addstock_geom(x, rand2d)
addstock_geom(x, rand3d)

answered Nov 20, 2023 at 18:54

Onyambu

80.3k3 gold badges29 silver badges65 bronze badges

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

Collectives™ on Stack Overflow

Rewriting numpy function to handle 2d and 3d inputs

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related