[Cherry-Pick][PHI] non-contiguous paddle.nansum accuracy matching#78296
Merged
sneaxiy merged 7 commits intoPaddlePaddle:release/3.3from Mar 16, 2026
Merged
[Cherry-Pick][PHI] non-contiguous paddle.nansum accuracy matching#78296sneaxiy merged 7 commits intoPaddlePaddle:release/3.3from
paddle.nansum accuracy matching#78296sneaxiy merged 7 commits intoPaddlePaddle:release/3.3from
Conversation
- GPU(CUDA/ROCm) and CPU only - Window compatible
- ODR rule protection: removing GPU registration, keep KPS.
|
你的PR提交成功,感谢你对开源项目的贡献! |
…ase3.3) (re-push)
Codecov Report❌ Patch coverage is
❌ Your patch status has failed because the patch coverage (79.24%) is below the target coverage (90.00%). You can increase the patch coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## release/3.3 #78296 +/- ##
==============================================
Coverage ? 79.24%
==============================================
Files ? 4
Lines ? 53
Branches ? 0
==============================================
Hits ? 42
Misses ? 11
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Contributor
Author
|
单个文件 Coverage 未覆盖的问题与 |
zyfncg
approved these changes
Mar 16, 2026
sneaxiy
approved these changes
Mar 16, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR Category
Operator Mechanism
PR Types
Improvements
Description
从主仓:
paddle.nansumaccuracy matching #78219Cherry-pick 而来
将组合算子
paddle.nansum改造为 PHI 算子,以对 PyTorch 做精度对齐。组合算子的paddle.where会强制把不连续 tensor 转为连续,导致累加顺序变化 (@baoqiwen)。复用phi::Reduce操作以及手动实现的NanSumOps可修正此问题。本 PR 需要考虑 non-contiguous layout,所以在 strided ops 里增加了 nansum。
注意! 本 PR 与 dev 分支也有一定的差别:
大 Tensor 精度对齐测试:
Torch 对齐精度测试:
是否引起精度变化
是
devPR:#78219
Pcard-89620