如何使用二维索引数组从二维数组中提取元素？

Question

I am trying to extract elements from a 2 dimensional array, a , and I am using a 2 dimensional array of indices, b , representing x/y coordinates. I have found something similar for a 1D array but I was unable to successfully apply it to a 2D array: Python - How to extract elements from an array based on an array of indices?

a = np.random.randn(10,7)
a = a.astype(int)
b = np.array([[1, 2], [2,3], [3,5], [2,7], [5,6]])

I have been using the bit of code below, however it returns a 3D matrix with values from the rows of each of the indices:

result2 = np.array(a)[b]

result2
Out[101]: 
array([[[ 0, -1,  0,  0,  0,  1,  0],
        [ 0, -1,  0,  0,  0,  0,  0]],

       [[ 0, -1,  0,  0,  0,  0,  0],
        [-1,  0,  0,  1,  0,  0,  0]],

       [[-1,  0,  0,  1,  0,  0,  0],
        [ 0,  0, -1, -2,  1,  0,  0]],

       [[ 0, -1,  0,  0,  0,  0,  0],
        [-1,  0,  0,  0,  0,  0,  1]],

       [[ 0,  0, -1, -2,  1,  0,  0],
        [ 1,  0,  0,  1,  0, -1,  0]]])

How can I modify b in order to index (column 1,row 2)... (column 2, row 3)... (column 3, row 5)... etc?

...

This is a minimal reproducible example and my actual data involves me indexing 500 cells in a 100x100 matrix (using an array of x/y coordinates/indices, size (500x2), similar to the above b ). Would it be best to use a for loop in this case? Something along the lines of...

for i in b:
    for j in b:
        result2 = np.array(a)[i,j]

Answer 1

I've encountered the same issue not long ago, and the answer is actually quite simple:

result = a[b[:,0], b[:,1]]