Vectorized computation on Numpy array

Question

Currently, I am using the loops to do the task below. It is quite similar to my previous question: Dot product two 4D Numpy array

Just wondering if anyone can help me to vectorize the computation without the loops.

A1 = np.random.rand(2, 320, 320)
A2 = np.random.rand(15, 2, 320, 320)
B   = np.zeros((2, 320, 320))

for row in range (320):
    for col in range (320):
        C = np.diag(A1[:, row, col]) - (np.dot(A2[:, :, row, col].T, A2[:, :, row, col]) / 15) # the shape of array C is (2, 2)
        C = np.diag(C)                
        B[:, row, col] = C

Any help or suggestions would be greatly appreciated!

apparently, just the new A1 from which the previous output needs to be subtracted. I am not sure if it qualifies as a duplicate, almost does. — Akshay Sehgal
– Akshay Sehgal, Commented Jan 15, 2021 at 4:11

Akshay Sehgal · Accepted Answer · 2021-01-15 04:07:43Z

1

Similar to the last solution -

output = A1-(A2*A2).sum(0)/15
np.allclose(output,B)

True

answered Jan 15, 2021 at 4:07

Akshay Sehgal

19.4k3 gold badges26 silver badges57 bronze badges

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

Collectives™ on Stack Overflow

Vectorized computation on Numpy array

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related