`min`

Smallest element, along a dimension, or the smaller of two arrays.

Category: Language Fundamentals > Data Types

Syntax

M = min(A)

Smallest element of a vector, or of each column of a matrix.

[M, I] = min(A)

Also returns the index of the smallest element; the first one when there are ties.

M = min(A, [], dim)

Smallest along dimension dim.

M = min(A, [], 'all')

Smallest of all elements.

M = min(A, B)

Element-wise smaller of A and B, with implicit expansion.

Description

min mirrors max. NaN values are ignored unless every value is NaN, so min([4 NaN 2]) is 2. Integer types keep their class, and the empty matrix gives [].

With one output the first minimum wins on ties; with the two-array form there is no index output.

Input arguments

A, B: Numeric, logical or char arrays, or datetime or duration arrays.
dim: Dimension to work along.
'omitnan' | 'includenan': Whether NaN values are skipped (the default) or make the result NaN.

Output arguments

M: The smallest values.
I: Indices of the minima along the working dimension.

Example

Minima of a matrix

A = [1 5; 7 2];
min(A)
min(A, [], 2)
[m, i] = min([3 1 2 1])
min([4 NaN 2])
min(int8([5 -3 9]))

Output

ans = 1×2

     1     2
ans = 2×1

     1
     2
m = 1
i = 2
ans = 2
ans = int8
   -3

min