nonzeros - Nonzero matrix elements - MATLAB (original) (raw)
Syntax
Description
[v](#mw%5F07596ee6-4a1f-4610-a920-87f9fcc44c88) = nonzeros([A](#mw%5F562860bf-a007-4b7d-8ee6-8ba72a2c061c)) returns a full column vector of the nonzero elements in A. The elements inv are ordered by columns.
Examples
Use nonzeros to return the nonzero elements in a sparse matrix.
Create a 10-by-10 sparse matrix that contains a few nonzero elements. The typical display of sparse matrices shows a list of the nonzero values and their locations.
A = sparse([1 3 2 1],[1 1 2 3],1:4,10,10)
A = 10×10 sparse double matrix (4 nonzeros) (1,1) 1 (3,1) 2 (2,2) 3 (1,3) 4
Find the values of the nonzero elements.
Use nonzeros, nnz, and find to locate and count nonzero matrix elements.
Create a 10-by-10 random sparse matrix with 7% density of nonzeros.
Use nonzeros to find the values of the nonzero elements.
v = 7×1
0.9595
0.4218
0.7922
0.8003
0.1419
0.9157
0.6557Use nnz to count the number of nonzeros.
Use find to get the indices and values of the nonzeros.
j = 7×1
2
5
6
10
10
10
10v = 7×1
0.9595
0.4218
0.7922
0.8003
0.1419
0.9157
0.6557Input Arguments
Input array, specified as a vector, matrix, or multidimensional array.A can be full or sparse.
Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | logical | char
Complex Number Support: Yes
Output Arguments
Nonzero elements, returned as a column vector. v is returned in full-storage regardless of whether A is full or sparse. The elements in v are ordered first by column subscript and then by row subscript.
nonzeros gives the v, but not the indicesi and j, from [i,j,v] = find(A). Generally,
length(v) = nnz(A) <= nzmax(A) <= prod(size(A))
Extended Capabilities
The nonzeros function fully supports GPU arrays. To run the function on a GPU, specify the input data as a gpuArray (Parallel Computing Toolbox). For more information, see Run MATLAB Functions on a GPU (Parallel Computing Toolbox).
Version History
Introduced before R2006a