Use absolute tolerance for sum_of_poly kernel #654
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Clayton Smith <[email protected]>
jdemel
approved these changes
Oct 14, 2023
There was a problem hiding this comment.
LGTM.
This issue is one of the reasons why we should think of a better test environment. However, implementing more manual tests is tedious work.
In any case, we should try to strike a balance between more automated tests and more tailoring where necessary.
My comments are obviously out of scope of this PR.
marcusmueller
approved these changes
Oct 20, 2023
Alesha72003
pushed a commit
to Alesha72003/volk
that referenced
this pull request
May 15, 2024
Use absolute tolerance for sum_of_poly kernel
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #650.
The "sum of poly" kernel output is the sum of many positive & negative values, and the final sum can end up being close to zero. This causes test failures, because the relative error becomes large when the sum is near zero. Here I've switched to an absolute tolerance. I tested the generic kernel against a double-precision implementation and found that the error can be as high as 120 (which isn't surprising, since 131071 * 5 = 655355 polynomial terms are summed), so I set the tolerance at 1000 to keep it a safe distance away.