find

Find Non-zero Elements

Syntax

  • K = find(M)

  • [R, C] = find(M)

  • [R, C, V] = find(M)

  • K = find(M, N)

  • [R, C] = find(M, N)

  • [R, C, V] = find(M, N)

  • K = find(M, N, D)

  • [R, C] = find(M, N, D)

  • [R, C, V] = find(M, N, D)

Input argument

  • M - a scalar, vector, matrix, or multidimensional array.

  • N - positive integer scalar value: number of nonzeros to find.

  • D - direction: 'first' (default) or 'last'.

Output argument

  • K - indices to nonzero elements (vector).

  • R - row subscripts (vector).

  • C - column subscripts (vector).

  • V - nonzero elements of M (vector).

Description

K = find(M) returns a vector with the linear indices of each nonzero element of M.

Example

M = rand(4, 3, 5);
[R, C, V] = find(M > 0.9)
M(R(1),C(1),V(1))

See also

strfind.

History

Author

Allan CORNET

Last updated