| Type: | Package |
| Title: | Reliable Comparison of Floating Point Numbers |
| Version: | 0.2.6 |
| Description: | Comparisons of floating point numbers are problematic due to errors associated with the binary representation of decimal numbers. Despite being aware of these problems, people still use numerical methods that fail to account for these and other rounding errors (this pitfall is the first to be highlighted in Circle 1 of Burns (2012) 'The R Inferno' https://www.burns-stat.com/pages/Tutor/R_inferno.pdf). This package provides new relational operators useful for performing floating point number comparisons with a set tolerance. |
| URL: | https://fpcompare.predictiveecology.org, https://github.com/PredictiveEcology/fpCompare |
| BugReports: | https://github.com/PredictiveEcology/fpCompare/issues |
| License: | GPL-3 |
| Depends: | R (≥ 4.1) |
| Suggests: | knitr, rmarkdown, testthat (≥ 3.0.0), withr |
| Encoding: | UTF-8 |
| Language: | en-CA |
| VignetteBuilder: | knitr |
| ByteCompile: | yes |
| Config/Needs/coverage: | covr |
| Config/Needs/website: | pkgdown |
| Config/testthat/edition: | 3 |
| Config/roxygen2/version: | 8.0.0 |
| NeedsCompilation: | no |
| Packaged: | 2026-05-26 02:05:41 UTC; achubaty |
| Author: | Alex M Chubaty 
[aut, cre],
His Majesty the King in Right of Canada, as represented by the Minister
of Natural Resources Canada [cph] |
| Maintainer: | Alex M Chubaty <achubaty@for-cast.ca> |
| Repository: | CRAN |
| Date/Publication: | 2026-05-26 05:10:02 UTC |
fpCompare: Reliable Comparison of Floating Point Numbers
Description
Comparisons of floating point numbers are problematic due to errors associated with the binary representation of decimal numbers. Despite being aware of these problems, people still use numerical methods that fail to account for these and other rounding errors (this pitfall is the first to be highlighted in Circle 1 of Burns (2012) 'The R Inferno' https://www.burns-stat.com/pages/Tutor/R_inferno.pdf). This package provides new relational operators useful for performing floating point number comparisons with a set tolerance.
Author(s)
Maintainer: Alex M Chubaty achubaty@for-cast.ca (ORCID)
Authors:
Alex M Chubaty achubaty@for-cast.ca (ORCID)
Other contributors:
His Majesty the King in Right of Canada, as represented by the Minister of Natural Resources Canada [copyright holder]
See Also
Useful links:
Report bugs at https://github.com/PredictiveEcology/fpCompare/issues
Relational operators with tolerance
Description
Binary operators which allow the comparison of values in numeric vectors.
Usage
x %>=% y
x %>>% y
x %<=% y
x %<<% y
x %==% y
x %!=% y
Arguments
x |
Any numeric object. |
y |
Any numeric object. |
Details
These are similar to their counterparts in base, except a tolerance
fpCompare.tolerance can be specified via options to account for floating
point rounding errors:
fpCompare | base |
%>=% | >= |
%>>% | > |
%<=% | <= |
%<<% | < |
%==% | == |
%!=% | !=
|
Inspired by R FAQ 7.31 (https://cran.r-project.org/doc/FAQ/R-FAQ.html#Why-doesn_0027t-R-think-these-numbers-are-equal_003f) and this post (https://stackoverflow.com/a/2769618/1380598).
Value
A logical vector indicating the result of the element-by-element comparison. The elements of shorter vectors are recycled as necessary.
Author(s)
Alex Chubaty
See Also
Examples
x1 <- 0.5 - 0.3
x2 <- 0.3 - 0.1
x1 == x2 ## FALSE on most machines
x1 %==% x2 ## TRUE everywhere
identical(all.equal(x1, x2), TRUE) ## TRUE everywhere
set.seed(123)
a <- 1:6
b <- jitter(1:6, 1e-7)
print(rbind(a, b), digits = 16)
b %<=% a
b %<<% a
b %>=% a
b %>>% a
b %==% a
b %!=% a