CODEWITHSUNDEEP
pythonmatrix
Sebastian
Sebastian

Reputation: 39

Inserting matrix into another matrix (specific positions)

I have question about matrix addition. I have a matrix A = np.ones([10,10]) and a matrix B = np.array([[2,2,2],[2,2,2],[2,2,2]]). Now I want to add matrix B into A but on specific positions, rows 2,6,7 and columns 2,6,7.

How should I do to get the following matrix:

[[1,1,1,1,1,1,1,1,1,1],
 [1,1,1,1,1,1,1,1,1,1],
 [1,1,3,1,1,1,3,3,1,1],
 [1,1,1,1,1,1,1,1,1,1],
 [1,1,1,1,1,1,1,1,1,1],
 [1,1,1,1,1,1,1,1,1,1],
 [1,1,3,1,1,1,3,3,1,1],
 [1,1,3,1,1,1,3,3,1,1],
 [1,1,1,1,1,1,1,1,1,1],
 [1,1,1,1,1,1,1,1,1,1]]

I'm more used to Matlab and there, it would look something like this: A((3,7,8),(3,7,8)) = A((3,7,8),(3,7,8)) + B. I tried something similar in Python but the dimensions didnt match.

Upvotes: 1

Views: 470

Answers (1)

Mahathi Vempati
Mahathi Vempati

Reputation: 1408

Here is one way to do it:

Multidimensional indexing in Python requires you to explicitly index every cell. So you need to first create the indexing, and then use it as follows:

ind = np.array([[2,6,7]])   # Notice the 2D array
rows = np.broadcast_to(ind.transpose(), (3,3))
cols = np.broadcast_to(ind, (3,3))
A[rows, cols]+=B  # A cell from rows matrix and a corresponding cell in cols matrix together form one cell index.

Output:

array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 3, 1, 1, 1, 3, 3, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 3, 1, 1, 1, 3, 3, 1, 1],
   [1, 1, 3, 1, 1, 1, 3, 3, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
   [1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])

Do read: https://docs.scipy.org/doc/numpy-1.13.0/user/basics.indexing.html

For some reason, while the following does pick out the right matrix from A, assignment to it does not work:

ind_1 = np.array([2,6,7])
A[ind_1,:][:, ind_1] = B # No error, but assignment does not take place

Upvotes: 2

Related Questions