spalloc - Allocate space for sparse matrix - MATLAB (original) (raw)

Allocate space for sparse matrix

Syntax

Description

[S](#mw%5Fd20503bb-158e-44ac-9d2e-e06c2f8b2cf3) = spalloc([m](#mw%5F7d46c77a-953c-4565-8c3d-c16b71e936da),[n](#mw%5Fa66f1fe6-4939-4a29-a4f1-3c8636ce2a42),[nz](#mw%5F7fff94d1-afe2-4ffd-a8a1-041f77b19036)) creates an all-zero sparse matrix S of sizem-by-n with room to hold nz nonzero elements, where nz >= 1.

example

[S](#mw%5Fd20503bb-158e-44ac-9d2e-e06c2f8b2cf3) = spalloc([m](#mw%5F7d46c77a-953c-4565-8c3d-c16b71e936da),[n](#mw%5Fa66f1fe6-4939-4a29-a4f1-3c8636ce2a42),[nz](#mw%5F7fff94d1-afe2-4ffd-a8a1-041f77b19036),[typename](#mw%5F8dc31b04-4361-47dd-aae3-433c6230a347)) returns a sparse matrix of the specified data type. (since R2025a)

Examples

collapse all

Use spalloc to initialize a 10-by-10 all-zero sparse matrix with room for up to 20 nonzero elements.

Define several elements in the matrix.

S = 10×10 sparse double matrix (9 nonzeros) (1,1) 8 (2,1) 3 (3,1) 4 (1,2) 1 (2,2) 5 (3,2) 9 (1,3) 6 (2,3) 7 (3,3) 2

Show the number of nonzero elements in the matrix.

Show the amount of storage allocated for nonzero matrix elements.

Use spalloc to initialize a 20-by-20 all-zero sparse matrix with space for 100 nonzero elements.

n = 20; S = spalloc(n,n,5*n);

Then use a for loop to fill in the columns of S one at a time with an average of at most five nonzero elements per column.

for j = 1:n S(:,j) = [zeros(n-5,1); round(rand(5,1))]; end

Plot the sparsity pattern of matrix S. The dots represent the nonzero elements.

Figure contains an axes object. The axes object with xlabel nz = 54 contains a line object which displays its values using only markers.

Show the number of nonzero elements in the matrix.

Show the amount of storage allocated for nonzero matrix elements.

Input Arguments

collapse all

Number of matrix rows, specified as a nonnegative integer.

Number of matrix columns, specified as a nonnegative integer.

Storage allocation for nonzero elements, specified as a nonnegative integer. If you specify a value of 0 for nz, then spalloc instead sets the value of nz to 1.

Since R2025a

Output data type, specified as "double","single", or "logical".

Output Arguments

collapse all

Output matrix, returned as an all-zero sparse matrix.

Limitations

Both matrix dimensions, m and n, must be smaller than 2^31-1 on 32-bit platforms, or 2^48-1 on 64-bit platforms.

Tips

When you assign several times into a matrix you created withspalloc, the preallocated memory can prevent repeated reallocations. However, assigning into a sparse matrix is still a relatively expensive operation, which should usually be avoided if it can easily be replaced by one of the following:
- a one-time call to the sparse function
- a one-time call to the spdiags function
- a one-time concatenation of a set of matrices, which can be sparse, dense, or both

Extended Capabilities

Version History

Introduced before R2006a

expand all

You can specify the output data type by specifying the typename argument as "double", "single", or"logical".