What is the fastest way to slice a scipy.sparse matrix?

python numpy scipy sparse-matrix
todpole3 picture todpole3 · Dec 12, 2012 · Viewed 17.3k times · Source

I normally use

matrix[:, i:]

It seems not work as fast as I expected.

Answer

Saullo G. P. Castro picture Saullo G. P. Castro · Jun 2, 2013

If you want to obtain a sparse matrix as output the fastest way to do row slicing is to have a csr type, and for columns slicing csc, as detailed here. In both cases you just have to do what you are currently doing:

matrix[l1:l2,c1:c2]

If you want another type as output there maybe faster ways. In this other answer it is explained many methods for slicing a matrix and their different timings compared. For example, if you want a ndarray as output the fastest slicing is:

matrix.A[l1:l2,c1:c2]

or:

matrix.toarray()[l1:l2,c1:c2]

much faster than:

matrix[l1:l2,c1:c2].A #or .toarray()

Related questions

How to transform numpy.matrix or array to scipy sparse matrix

For SciPy sparse matrix, one can use todense() or toarray() to transform to NumPy matrix or array. What are the functions to do the inverse? I searched, but got no idea what keywords should be the right hit.

Read More
Generating a dense matrix from a sparse matrix in numpy python

I have a Sqlite database that contains following type of schema: termcount(doc_num, term , count) This table contains terms with their respective counts in the document. like (doc1 , term1 ,12) (doc1, term 22, 2) . . (docn,term1 , 10) This matrix can be considered as …

Read More
How do I transform a "SciPy sparse matrix" to a "NumPy matrix"?

I am using a python function called "incidence_matrix(G)", which returns the incident matrix of graph. It is from Networkx package. The problem that I am facing is the return type of this function is "Scipy Sparse Matrix". I …

Read More