-
-
Notifications
You must be signed in to change notification settings - Fork 1.1k
feat: add number/float16/base/assert/is-almost-same-value
#9664
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: develop
Are you sure you want to change the base?
feat: add number/float16/base/assert/is-almost-same-value
#9664
Conversation
---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
- task: lint_filenames
status: passed
- task: lint_editorconfig
status: passed
- task: lint_markdown
status: passed
- task: lint_package_json
status: passed
- task: lint_repl_help
status: passed
- task: lint_javascript_src
status: passed
- task: lint_javascript_cli
status: na
- task: lint_javascript_examples
status: passed
- task: lint_javascript_tests
status: passed
- task: lint_javascript_benchmarks
status: passed
- task: lint_python
status: na
- task: lint_r
status: na
- task: lint_c_src
status: na
- task: lint_c_examples
status: na
- task: lint_c_benchmarks
status: na
- task: lint_c_tests_fixtures
status: na
- task: lint_shell
status: na
- task: lint_typescript_declarations
status: passed
- task: lint_typescript_tests
status: passed
- task: lint_license_headers
status: passed
---
Coverage Report
The above coverage report was generated for the changes in this PR. |
---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
- task: lint_filenames
status: passed
- task: lint_editorconfig
status: passed
- task: lint_markdown
status: na
- task: lint_package_json
status: na
- task: lint_repl_help
status: na
- task: lint_javascript_src
status: na
- task: lint_javascript_cli
status: na
- task: lint_javascript_examples
status: na
- task: lint_javascript_tests
status: na
- task: lint_javascript_benchmarks
status: na
- task: lint_python
status: na
- task: lint_r
status: na
- task: lint_c_src
status: na
- task: lint_c_examples
status: na
- task: lint_c_benchmarks
status: na
- task: lint_c_tests_fixtures
status: na
- task: lint_shell
status: na
- task: lint_typescript_declarations
status: passed
- task: lint_typescript_tests
status: na
- task: lint_license_headers
status: passed
---
Planeshifter
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Implementation differs in behavior from float32 and float64.
| // Handle signed zeros (-0 !== +0 in SameValue): | ||
| if ( isPositiveZero( af ) && isNegativeZero( bf ) ) { | ||
| return false; | ||
| } | ||
| if ( isNegativeZero( af ) && isPositiveZero( bf ) ) { | ||
| return false; | ||
| } |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This signed zero handling behaves differently from the float32 and float64 equivalents.
The float32 is-almost-same-value only distinguishes between +0 and -0 when maxULP === 0. When maxULP >= 1, it returns true for (0.0, -0.0) because they're 0 ULPs apart (per ulp-difference).
Currently this implementation always returns false for +0 vs -0 regardless of maxULP, which creates an inconsistency:
// float32 behavior:
isAlmostSameValue(0.0, -0.0, 0); // false
isAlmostSameValue(0.0, -0.0, 1); // true
// float16 behavior (this PR):
isAlmostSameValue(0.0, -0.0, 0); // false
isAlmostSameValue(0.0, -0.0, 1); // false (inconsistent!)Check the float32 and float64 implementations for what is expected.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Planeshifter I noticed that the float32 implementation uses isSameValue as a dependency:
Should I:
Implement is-same-value for float16 first, then update this PR to match the float32 pattern exactly? This would maintain consistency across the codebase.
Keep the current self-contained implementation that uses the 1/x approch directly for signed zeros without depending on is-same-value?
| tape( 'the function returns `false` if signed zeros are provided as input values irrespective of the specified number of ULPs', function test( t ) { | ||
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 0 ), false, 'returns expected value' ); | ||
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 0 ), false, 'returns expected value' ); | ||
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 1 ), false, 'returns expected value' ); | ||
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 1 ), false, 'returns expected value' ); | ||
| t.end(); | ||
| }); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The test description and expectations for maxULP >= 1 don't match the float32 equivalent's behavior. The float32 version returns true when comparing +0 and -0 with maxULP >= 1, because they're 0 ULPs apart.
If aligning with float32, these tests should be:
| tape( 'the function returns `false` if signed zeros are provided as input values irrespective of the specified number of ULPs', function test( t ) { | |
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 0 ), false, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 0 ), false, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 1 ), false, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 1 ), false, 'returns expected value' ); | |
| t.end(); | |
| }); | |
| tape( 'the function distinguishes between signed zeros when the specified number of ULPs is zero', function test( t ) { | |
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 0 ), false, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 0 ), false, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( 0.0, -0.0, 1 ), true, 'returns expected value' ); | |
| t.strictEqual( isAlmostSameValue( -0.0, 0.0, 1 ), true, 'returns expected value' ); | |
| t.end(); | |
| }); |
---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
- task: lint_filenames
status: passed
- task: lint_editorconfig
status: passed
- task: lint_markdown
status: na
- task: lint_package_json
status: na
- task: lint_repl_help
status: na
- task: lint_javascript_src
status: passed
- task: lint_javascript_cli
status: na
- task: lint_javascript_examples
status: na
- task: lint_javascript_tests
status: passed
- task: lint_javascript_benchmarks
status: na
- task: lint_python
status: na
- task: lint_r
status: na
- task: lint_c_src
status: na
- task: lint_c_examples
status: na
- task: lint_c_benchmarks
status: na
- task: lint_c_tests_fixtures
status: na
- task: lint_shell
status: na
- task: lint_typescript_declarations
status: passed
- task: lint_typescript_tests
status: na
- task: lint_license_headers
status: passed
---
---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
- task: lint_filenames
status: passed
- task: lint_editorconfig
status: passed
- task: lint_markdown
status: na
- task: lint_package_json
status: na
- task: lint_repl_help
status: na
- task: lint_javascript_src
status: passed
- task: lint_javascript_cli
status: na
- task: lint_javascript_examples
status: na
- task: lint_javascript_tests
status: na
- task: lint_javascript_benchmarks
status: na
- task: lint_python
status: na
- task: lint_r
status: na
- task: lint_c_src
status: na
- task: lint_c_examples
status: na
- task: lint_c_benchmarks
status: na
- task: lint_c_tests_fixtures
status: na
- task: lint_shell
status: na
- task: lint_typescript_declarations
status: passed
- task: lint_typescript_tests
status: na
- task: lint_license_headers
status: passed
---
---
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes.
report:
- task: lint_filenames
status: passed
- task: lint_editorconfig
status: passed
- task: lint_markdown
status: passed
- task: lint_package_json
status: na
- task: lint_repl_help
status: na
- task: lint_javascript_src
status: na
- task: lint_javascript_cli
status: na
- task: lint_javascript_examples
status: na
- task: lint_javascript_tests
status: na
- task: lint_javascript_benchmarks
status: na
- task: lint_python
status: na
- task: lint_r
status: na
- task: lint_c_src
status: na
- task: lint_c_examples
status: na
- task: lint_c_benchmarks
status: na
- task: lint_c_tests_fixtures
status: na
- task: lint_shell
status: na
- task: lint_typescript_declarations
status: passed
- task: lint_typescript_tests
status: na
- task: lint_license_headers
status: passed
---
| var isnan = require( '@stdlib/math/base/assert/is-nan' ); | ||
| var f16 = require( '@stdlib/number/float64/base/to-float16' ); | ||
| var ulpdiff = require( '@stdlib/number/float16/base/ulp-difference' ); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| var isnan = require( '@stdlib/math/base/assert/is-nan' ); | |
| var f16 = require( '@stdlib/number/float64/base/to-float16' ); | |
| var ulpdiff = require( '@stdlib/number/float16/base/ulp-difference' ); | |
| var isnan = require( '@stdlib/number/float16/base/assert/is-nan' ); | |
| var isSameValue = require( '@stdlib/number/float16/base/assert/is-same-value' ); | |
| var ulpdiff = require( '@stdlib/number/float16/base/ulp-difference' ); |
Handling -+0 and isnan should come form isSameValue
Should be resolved once #9677 is merged.
| var af; | ||
| var bf; | ||
|
|
||
| af = f16( a ); | ||
| bf = f16( b ); | ||
|
|
||
| // Check for NaN (NaN is considered equal to NaN in SameValue): | ||
| if ( isnan( af ) && isnan( bf ) ) { | ||
| return true; | ||
| } | ||
| // Only distinguish signed zeros when maxULP is 0: | ||
| if ( maxULP === 0 && af === 0.0 && bf === 0.0 ) { | ||
| return ( 1.0 / af === 1.0 / bf ); | ||
| } | ||
| // Check if exactly equal: | ||
| if ( af === bf ) { | ||
| return true; | ||
| } |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| var af; | |
| var bf; | |
| af = f16( a ); | |
| bf = f16( b ); | |
| // Check for NaN (NaN is considered equal to NaN in SameValue): | |
| if ( isnan( af ) && isnan( bf ) ) { | |
| return true; | |
| } | |
| // Only distinguish signed zeros when maxULP is 0: | |
| if ( maxULP === 0 && af === 0.0 && bf === 0.0 ) { | |
| return ( 1.0 / af === 1.0 / bf ); | |
| } | |
| // Check if exactly equal: | |
| if ( af === bf ) { | |
| return true; | |
| } | |
| if ( isnan( a ) || isnan( b ) || maxULP === 0 ) { | |
| return isSameValue( a, b ); | |
| } |
Better to use the explicit float16 version of isnan rather than converting to half-precision
| var bool = isAlmostSameValue( 1.0, 1.0+EPS, 1 ); | ||
| console.log( bool ); | ||
| // => true | ||
|
|
||
| bool = isAlmostSameValue( 1.0+EPS, 1.0, 1 ); | ||
| console.log( bool ); | ||
| // => true | ||
|
|
||
| bool = isAlmostSameValue( 1.0, 1.0+EPS+EPS, 1 ); | ||
| console.log( bool ); | ||
| // => false | ||
|
|
||
| bool = isAlmostSameValue( 1.0, 1.0+EPS, 0 ); | ||
| console.log( bool ); | ||
| // => false | ||
|
|
||
| bool = isAlmostSameValue( -0.0, 0.0, 0 ); | ||
| console.log( bool ); | ||
| // => false | ||
|
|
||
| bool = isAlmostSameValue( 1.0, NaN, 1 ); | ||
| console.log( bool ); | ||
| // => false | ||
|
|
||
| bool = isAlmostSameValue( NaN, NaN, 1 ); | ||
| console.log( bool ); | ||
| // => true | ||
| ``` |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| var bool = isAlmostSameValue( 1.0, 1.0+EPS, 1 ); | |
| console.log( bool ); | |
| // => true | |
| bool = isAlmostSameValue( 1.0+EPS, 1.0, 1 ); | |
| console.log( bool ); | |
| // => true | |
| bool = isAlmostSameValue( 1.0, 1.0+EPS+EPS, 1 ); | |
| console.log( bool ); | |
| // => false | |
| bool = isAlmostSameValue( 1.0, 1.0+EPS, 0 ); | |
| console.log( bool ); | |
| // => false | |
| bool = isAlmostSameValue( -0.0, 0.0, 0 ); | |
| console.log( bool ); | |
| // => false | |
| bool = isAlmostSameValue( 1.0, NaN, 1 ); | |
| console.log( bool ); | |
| // => false | |
| bool = isAlmostSameValue( NaN, NaN, 1 ); | |
| console.log( bool ); | |
| // => true | |
| ``` | |
| var bool = isAlmostSameValue( 1.0, 1.0+EPS, 1 ); | |
| // returns true | |
| bool = isAlmostSameValue( 1.0+EPS, 1.0, 1 ); | |
| // returns true | |
| bool = isAlmostSameValue( 1.0, 1.0+EPS+EPS, 1 ); | |
| // returns false | |
| bool = isAlmostSameValue( 1.0, 1.0+EPS, 0 ); | |
| // returns false | |
| bool = isAlmostSameValue( -0.0, 0.0, 0 ); | |
| // returns false | |
| bool = isAlmostSameValue( 1.0, NaN, 1 ); | |
| // returns false | |
| bool = isAlmostSameValue( NaN, NaN, 1 ); | |
| // returns true |
Let's remove the console.log in the markdown.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes. report:
Resolves none.
Description
This pull request:
number/float16/base/assert/is-almost-same-valueRelated Issues
This pull request has the following related issues:
Questions
No.
Other
No.
Checklist
AI Assistance
@stdlib-js/reviewers