Skip to content

fix: use adaptive precision for moffat computation#181

Merged
beckermr merged 3 commits into
mainfrom
bessel-eps
Feb 9, 2026
Merged

fix: use adaptive precision for moffat computation#181
beckermr merged 3 commits into
mainfrom
bessel-eps

Conversation

@beckermr

@beckermr beckermr commented Feb 9, 2026

Copy link
Copy Markdown
Collaborator

This PR pulls the fix from #180 by @EiffL for the float precision issue in the Moffat bessel functions and adds a test. The goal in merging it is to at least fix the precision issue on main while we wait on the other improvements.

fixes #179
fixes #182

@beckermr

beckermr commented Feb 9, 2026

Copy link
Copy Markdown
Collaborator Author

I test the fix directly in float32. With out current testing setup, we can't force float32 without changes to the code configuration. I can try and make a testing option for this maybe.

@codspeed-hq

codspeed-hq Bot commented Feb 9, 2026

Copy link
Copy Markdown

Merging this PR will not alter performance

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

✅ 32 untouched benchmarks


Comparing bessel-eps (1b89eb8) with main (9fb8019)

Open in CodSpeed

@beckermr

beckermr commented Feb 9, 2026

Copy link
Copy Markdown
Collaborator Author

pre-commit.ci autofix

1 similar comment
@beckermr

beckermr commented Feb 9, 2026

Copy link
Copy Markdown
Collaborator Author

pre-commit.ci autofix

@beckermr beckermr merged commit 89b9d56 into main Feb 9, 2026
9 checks passed
@beckermr beckermr deleted the bessel-eps branch February 9, 2026 20:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

add pytest configuration to enable running specific tests in float32 Moffat profile drawing returns NaN (demo2.py)

1 participant