SciPy Recipes

How to do it…

Lets get into the details.

1. Import the relevant packages:

from scipy.sparse import lil_matrixfrom scipy.sparse.linalgimport spsolvefrom numpy.linalg import solve, normfrom numpy.random import rand 

2. Initialize a matrix:

A = lil_matrix((10000, 10000))

In the preceding snippet of code, we have initialized a linked list matrix that is 10,000 x 10,000 in size. 

3. Impute very few values with random numbers:

A[0, :100] = rand(100)A[1, 100:200] = A[0, :100] 

In the preceding snippet of code, we have imputed the first 100 columns of the first row of the matrix, and the next 100 columns of the second row of the matrix with random numbers. 

4. Set the diagonal of the original matrix:

A.setdiag(rand(10000)) 

In the preceding set of code, ...