CODEWITHSUNDEEP
pythonscipysparse-matrix
silvado
silvado

Reputation: 18167

Multiplication operator for scipy sparse matrices

I'm a little bit confused by the behavior of the multiplication operator * when scipy sparse matrices are involved. It seems the operator implements matrix multiplication, not component-wise multiplication as it would with numpy arrays.

Some code to check this:

from scipy.sparse import lil_matrix
A = lil_matrix(-numpy.eye(2))
b = lil_matrix(numpy.ones((2,2)))
print (A * B).toarray()

results in:

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

The documentation of the scipy.sparse module does not really go into details on this, and I wonder whether there is a clear specification of the multiplication behavior somewhere?

Furthermore, are there some clearly defined rules for multiplication operator with scipy sparse matrices and numpy matrices or arrays?

Upvotes: 4

Views: 2972

Answers (2)

Femkemilene
Femkemilene

Reputation: 305

Documentation is indeed scarse. If you are looking for component-wise multiplication you can use A.multiply(b), where b can be an element, vector or matrix:

https://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.sparse.csr_matrix.multiply.html

Upvotes: 2

pv.
pv.

Reputation: 35125

Yes, it is matrix multiplication in all cases.

Upvotes: 1

Related Questions