Floating point relational operator.

Numerical comparison with specified precision.

Version 1.2.0.0 (2.85 KB)
125 Downloads
(1)
1 Jun 2016

You are now following this Submission

This function provides comparison operation improvement for floating point numbers.
Singleton expansion, multidimensional arrays supported.
example:
>> A = [ 1, 1, 1, 100, 1, 4];
>> B = [ 1, 1.2, 1.01, 101, 4, 1];
>> A<B
ans = 0 1 1 1 1 0
>> numcmp(A,B,1,1,@lt)
ans = 0 1 0 0 1 0

Cite As

Jan Orwat (2026). Floating point relational operator. (https://in.mathworks.com/matlabcentral/fileexchange/57457-floating-point-relational-operator), MATLAB Central File Exchange. Retrieved .

Acknowledgements

Inspired by: numcmp.m

Categories

Find more on Numeric Types in Help Center and MATLAB Answers

General Information

MATLAB Release Compatibility

  • Compatible with any release

Platform Compatibility

  • Windows
  • macOS
  • Linux
Open in new tab
Version Published Release Notes
1.2.0.0

.
1.1.0.0

inf/nan errors solved. Please note that with numcmp infinities are not equal. inf == inf returns 1, while numcmp(inf,inf) returns 0. This is actually closer to reality:)
function rewritten, relative tolerance introduced.
bugs left to solve: eliminate errors when infs/nans are used.
1.0.0.0

Note, this function uses only absolute precision.