Question

I’m implementing Strassen’s Matrix Multiplication using python. In divide step, we divide a larger matrix into smaller sub-matrices. Is there a built-in numpy function to split a matrix?

Answer 1

Not exactly, but using array slicing notation you should be able to do it yourself pretty easily.

>>> A = np.linspace(0,24,25).reshape([5,5,])
>>> A
array([[  0.,   1.,   2.,   3.,   4.],
       [  5.,   6.,   7.,   8.,   9.],
       [ 10.,  11.,  12.,  13.,  14.],
       [ 15.,  16.,  17.,  18.,  19.],
       [ 20.,  21.,  22.,  23.,  24.]])

Make B the top-left 2×2 in A:

>>> B = A[0:2,0:2]

Note that B is a view, it shares data with A

>>> B[1,1] = 60
>>> print A
[[  0.   1.   2.   3.   4.]
 [  5.  60.   7.   8.   9.]
 [ 10.  11.  12.  13.  14.]
 [ 15.  16.  17.  18.  19.]
 [ 20.  21.  22.  23.  24.]]

If you need to copy the data from A, use the copy method:

>>> B = A[0:2,0:2].copy()
>>> B
array([[  0.,   1.],
       [  5.,  60.]])
>>> B[1,1] = 600
>>> B
array([[   0.,    1.],
       [   5.,  600.]])
>>> A
array([[  0.,   1.,   2.,   3.,   4.],
       [  5.,  60.,   7.,   8.,   9.],
       [ 10.,  11.,  12.,  13.,  14.],
       [ 15.,  16.,  17.,  18.,  19.],
       [ 20.,  21.,  22.,  23.,  24.]])