You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Have questions about stdlib? Need help fixing a bug, figuring out what to do next, or just looking for feedback? Join our weekly office hours to connect with project maintainers, stay updated on the latest project news, and chat with other community members. This is a great opportunity to ask questions, share ideas, and engage directly with the stdlib team.
Specifically, I'm using @stdlib/assert/is-almost-same-value with 10 ULPs for most test values, but in the zero-crossing region around x≈3.38 (where Ci(x) approaches zero), the implementation shows errors up to ~4100 ULPs.
Question: Are 4100 ULPs acceptable for test tolerances in this zero-crossing region?
The implementation now uses accurate auxiliary function methods for all large values rather than simplified approximations. I wanted to confirm if this level of tolerance is acceptable or if something else should be done.
I have made a PR implementing dlartgp routine in LAPACK, would appreciate a review.
Whilst working on other LAPACK routines, I had to manually search whether a PR already existed for the given routine. Would it be a good idea for me to create a list of all LAPACK routines, in working order, and track their implementation status?
I am aiming to continue the LAPACK project for GSoC'26. Apart from LAPACK, what other areas or issues in stdlib should I explore to gain a broader understanding of the project?
Improve doctests for complex number instances in ndarray/base/unary #8999
Status
I’ve addressed all the issues that were reviewed and pointed out to me.
I wanted to check if any of these still need further changes.
I’d appreciate guidance on which issues I should work on next.
Question
While running make lint, I’m also seeing filename-related lint errors in several scaffold packages under @stdlib/_tools/scaffold. These files currently exist with a .txt extension (e.g., README_md.txt, Makefile.txt).
I wanted to confirm whether these .txt filenames are intentional and part of the stdlib scaffold conventions, or if this is an issue that needs to be addressed by renaming them to their expected forms (e.g., README.md, Makefile).
If this is a real issue, I’d be happy to raise an issue for it and work on fixing it.
Activity
navyansh007 commentedon Jan 13, 2026
1. Review Request: Half-Normal CDF Implementation
I would like to request a review of my PR for the half-normal CDF implementation:
PR: stdlib-js/stdlib#9429
2. Question: ULP Tolerance for SiCi C Implementation
I have a question regarding the acceptable ULP tolerance in the C implementation of SiCi functions:
PR: stdlib-js/stdlib#8187
Specifically, I'm using
@stdlib/assert/is-almost-same-valuewith 10 ULPs for most test values, but in the zero-crossing region around x≈3.38 (where Ci(x) approaches zero), the implementation shows errors up to ~4100 ULPs.Question: Are 4100 ULPs acceptable for test tolerances in this zero-crossing region?
The implementation now uses accurate auxiliary function methods for all large values rather than simplified approximations. I wanted to confirm if this level of tolerance is acceptable or if something else should be done.
prajjwalbajpai commentedon Jan 13, 2026
dlartgproutine in LAPACK, would appreciate a review.manit2004 commentedon Jan 13, 2026
aryan7071 commentedon Jan 13, 2026
I’ve raised the following pull requests in the past week
math/base/special
stats/base/ndarray
constants/float16
Documentation (Good First Issue)
Status
I’ve addressed all the issues that were reviewed and pointed out to me.
I wanted to check if any of these still need further changes.
I’d appreciate guidance on which issues I should work on next.
Question
While running make lint, I’m also seeing filename-related lint errors in several scaffold packages under @stdlib/_tools/scaffold. These files currently exist with a .txt extension (e.g., README_md.txt, Makefile.txt).
I wanted to confirm whether these .txt filenames are intentional and part of the stdlib scaffold conventions, or if this is an issue that needs to be addressed by renaming them to their expected forms (e.g., README.md, Makefile).
If this is a real issue, I’d be happy to raise an issue for it and work on fixing it.
DhruvArvindSingh commentedon Jan 13, 2026
Comment on this LINK
anoofmhd commentedon Jan 13, 2026
can you please review the PR #8502 which resolves the issue #5592