scipy.sparse.

diags#

scipy.sparse.diags(diagonals, offsets=0, shape=None, format=None, dtype=None)[source]#

Construct a sparse matrix from diagonals.

Warning

This function returns a sparse matrix – not a sparse array. You are encouraged to use diags_array to take advantage of the sparse array functionality.

Parameters:

diagonalssequence of array_like

Sequence of arrays containing the matrix diagonals, corresponding to offsets.

offsetssequence of int or an int, optional

Diagonals to set (repeated offsets are not allowed):

k = 0 the main diagonal (default)
k > 0 the kth upper diagonal
k < 0 the kth lower diagonal

shapetuple of int, optional

Shape of the result. If omitted, a square matrix large enough to contain the diagonals is returned.

format{“dia”, “csr”, “csc”, “lil”, …}, optional

Matrix format of the result. By default (format=None) an appropriate sparse matrix format is returned. This choice is subject to change.

dtypedtype, optional

Data type of the matrix.

See also

spdiags: construct matrix from diagonals
diags_array: construct sparse array instead of sparse matrix

Notes

Repeated diagonal offsets are disallowed.

The result from diags is the sparse equivalent of:

np.diag(diagonals[0], offsets[0])
+ ...
+ np.diag(diagonals[k], offsets[k])

diags differs from dia_matrix in the way it handles off-diagonals. Specifically, dia_matrix assumes the data input includes padding (ignored values) at the start/end of the rows for positive/negative offset, while diags` assumes the input data has no padding. Each value in the input ``diagonals is used.

Added in version 0.11.

Examples

>>> from scipy.sparse import diags
>>> diagonals = [[1, 2, 3, 4], [1, 2, 3], [1, 2]]
>>> diags(diagonals, [0, -1, 2]).toarray()
array([[1., 0., 1., 0.],
       [1., 2., 0., 2.],
       [0., 2., 3., 0.],
       [0., 0., 3., 4.]])

Broadcasting of scalars is supported (but shape needs to be specified):

>>> diags([1, -2, 1], [-1, 0, 1], shape=(4, 4)).toarray()
array([[-2.,  1.,  0.,  0.],
       [ 1., -2.,  1.,  0.],
       [ 0.,  1., -2.,  1.],
       [ 0.,  0.,  1., -2.]])

If only one diagonal is wanted (as in numpy.diag), the following works as well:

>>> diags([1, 2, 3], 1).toarray()
array([[ 0.,  1.,  0.,  0.],
       [ 0.,  0.,  2.,  0.],
       [ 0.,  0.,  0.,  3.],
       [ 0.,  0.,  0.,  0.]])