Improve the precision of abs() and sign() for large values #99550
Conversation
This file contains hidden or 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
@ev-br found in Quansight-Labs/numpy_pytorch_interop#117 (comment) that the precision of `abs()` for large values in the vectorised case is less-than-good. This PR fixes this issue. While doing that, we are able to comment out a few tests on extremal values. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/99550
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ No FailuresAs of commit 92cbe49: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
lezcano
added a commit
that referenced
this pull request
Apr 19, 2023
ev-br found in Quansight-Labs/numpy_pytorch_interop#117 (comment) that the precision of `abs()` for large values in the vectorised case is less-than-good. This PR fixes this issue. While doing that, we are able to comment out a few tests on extremal values. ghstack-source-id: 96a8b77 Pull Request resolved: #99550
ngimel
approved these changes
Apr 19, 2023
peterbell10
reviewed
Apr 19, 2023
ev-br found in Quansight-Labs/numpy_pytorch_interop#117 (comment) that the precision of `abs()` for large values in the vectorised case is less-than-good. This PR fixes this issue. While doing that, we are able to comment out a few tests on extremal values. cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
|
@peterbell10, addressed the review. I also cleaned |
ev-br found in Quansight-Labs/numpy_pytorch_interop#117 (comment) that the precision of `abs()` for large values in the vectorised case is less-than-good. This PR fixes this issue. While doing that, we are able to comment out a few tests on extremal values. Fixes #53958 #48486 cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
lezcano
added a commit
that referenced
this pull request
Apr 20, 2023
ev-br found in Quansight-Labs/numpy_pytorch_interop#117 (comment) that the precision of `abs()` for large values in the vectorised case is less-than-good. This PR fixes this issue. While doing that, we are able to comment out a few tests on extremal values. ghstack-source-id: c3ca417 Pull Request resolved: #99550
peterbell10
approved these changes
Apr 24, 2023
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
Unfortunately the VSX code hasn't been updated and produces NaNs where inf is expected for those test_reference_numerics_extremal__refs_abs_cpu_complex* tests |
|
Feel free to send a PR |
|
Done: #116859 |
pytorchmergebot
pushed a commit
to Flamefire/pytorch
that referenced
this pull request
Jan 5, 2024
Use a similar approach with Sleef as in pytorch#99550 to improve the precision and extremal value handling of the `abs` function for complex tensors. This fixes - test_reference_numerics_extremal__refs_abs_cpu_float64 - test_reference_numerics_extremal__refs_abs_cpu_float128 which failed on PPC.
pytorchmergebot
pushed a commit
that referenced
this pull request
Jan 5, 2024
Use a similar approach with Sleef as in #99550 to improve the precision and extremal value handling of the `abs` function for complex tensors. This fixes - test_reference_numerics_extremal__refs_abs_cpu_float64 - test_reference_numerics_extremal__refs_abs_cpu_float128 which failed on PPC. Pull Request resolved: #116859 Approved by: https://github.com/lezcano
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.
Stack from ghstack (oldest at bottom):
@ev-br found in
Quansight-Labs/numpy_pytorch_interop#117 (comment)
that the precision of
abs()for large values in the vectorised case is less-than-good.This PR fixes this issue. While doing that, we are able to comment out a
few tests on extremal values.
Fixes #53958 #48486
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10