Quick way to interpolate between matrices? (using numpy arrays)

Question

I now have a lot of matrices (m rows and n columns), I want to get the "intermediate matrix" of these matrices, and I want to do interpolation. I know how to write my program with a for loop, but it is too slow! Each matrix has about several million elements, and I want to interpolate over 20 matrices to get at least three times the original number of matrices among them. If I use a loop, the calculation speed is very slow. I want to know if there are any advanced methods.

Answer 1

Do you use vectorized forms? NUMPY is the most common approach. But you might use NUMBA or CUDA too, depending on your environment

import numpy as np

nm = 3  # number of matrices
ns = 3 # size of matrices

def generate_matrices(ns, nm):
    return np.array([(i*i)*np.ones((ns,ns)) for i in range(nm)])

#---- generate data matrices ----
A = generate_matrices(ns, nm)
print('A'); print(A);print()

#---- interpolate thee matrices -------
t0 = 0.4
t1 = 1 - t0

for jm in range(1,nm):
    B = t0*A[jm-1] + t1*A[jm]
    print('jm = ', jm); print(B); print()


A
[[[0. 0. 0.]
  [0. 0. 0.]
  [0. 0. 0.]]

 [[1. 1. 1.]
  [1. 1. 1.]
  [1. 1. 1.]]

 [[4. 4. 4.]
  [4. 4. 4.]
  [4. 4. 4.]]]
jm =  1
[[0.6 0.6 0.6]
 [0.6 0.6 0.6]
 [0.6 0.6 0.6]]

jm =  2
[[2.8 2.8 2.8]
 [2.8 2.8 2.8]
 [2.8 2.8 2.8]]