[Precision Depth Alignment] fix eps of paddle.logit from float to double #75816

zhengshengning · 2025-10-13T09:46:58Z

PR Category

Operator Mechanism

PR Types

New features

Description

主要修改：
本Pr主要将paddle.logit 的 eps参数的类型从float改为double，这样就可以完全与torch的精度对齐。
参考pr：#75426
参考文档：https://ku.baidu-int.com/knowledge/HFVrC7hq1Q/pKzJfZczuc/RqSZ7wzMmI/PYmTSAOJppi_li

测试情况：

paddle.logit 对齐全部case（共 9 个）

pcard-93269

paddle-bot · 2025-10-13T09:47:04Z

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

codecov-commenter · 2025-10-14T09:10:11Z

Codecov Report

❌ Patch coverage is 93.10345% with 2 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@be8a65c). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
...addle/fluid/ir_adaptor/translator/op_translator.cc	92.85%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop   #75816   +/-   ##
==========================================
  Coverage           ?   93.10%           
==========================================
  Files              ?        2           
  Lines              ?       29           
  Branches           ?        0           
==========================================
  Hits               ?       27           
  Misses             ?        2           
  Partials           ?        0

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

changeyoung98 · 2025-10-15T03:05:59Z

paddle/fluid/pir/serialize_deserialize/0.yaml

+      - action : modify_attr
+        object : eps
+        type : pir::DoubleAttribute
+        data : 1e-6


0.yaml应该在/patch目录下，与之前的yaml文件在一起

wanghuancoder · 2025-10-15T03:11:41Z

paddle/fluid/pir/serialize_deserialize/0.yaml

      - action : modify_attr
        object : fuse_alpha
        type : pir::DoubleAttribute
-        data : 0.0


这里是不是误删了？

志杨说可以修改attribute可以不用添加data，新增的时候才需要

… accuracy_stable_logit

wanghuancoder

LGTM

XiaoguangHu01

LGTM

…ble (PaddlePaddle#75816) * accuracy_stable_logit * add LogitOpTranscriber * fix coverage * fix 0yaml

* CallScalarFunction uses the dtype of 'self' as the type of 'other' when opotype is 'div'(#75237) * LinspaceKernel uses the dtype of 'self' as the type of 'step' when tensor is floating (#75238) * align LinspaceKernel * update meta * update gpu kernel * fix LinspaceKernelInner * improve kernel * fix CudaSigmoidGradFunctor and CudaSiluGradFunctor (#75341) * Softplus accuracy and torch alignment 1 (#75363) * [Precision Depth Alignment] paddle.tan reverse calculation: dx = dout *(1 + tan(x)^2) (#75335) * Tan reverse calculation: dx = dout *(1 + tan(x)^2) * [Precision Depth Alignment] Add support for CUDNN to paddle.nn.functional.grid_sample to align with torch accuracy. (#75355) * accuracy_stable_grid_sample * fix * correlation supports big tensor (#75383) * fix * fix test * fix * paddle.tanh Grad and torch alignment (float16) (#75454) * [Precision Depth Alignment] paddle.sin and paddle.cos aligns with torch precision. (#75503) * accuracy_stable_sin * accuracy_stable_cos * [深度对齐]Divide (#75379) * fix * fix * fix * fix * fix * [Precision Depth Alignment] fix precision for float16 of paddle.tan backward (#75525) * fix precision for float16 of paddle.tan backward * fix else branch of CudaTanGradFunctor * [Precision Depth Alignment] fix precision for paddle.expm1 (#75549) * accuracy_stable_expm1 * fix * Bigtensor排查修复[Paddle/paddle/phi/kernels/funcs] (#75523) * fix * fix * [Precision Depth Alignment] fix beta and threshold of paddle.nn.functional.softplus to double (#75426) * fix beta and threshold of Softplus to double * fix test_softplus_activation_fuse_pass v1 * fix test_activation_zero * fix flaot of SoftplusDoubleGradKernel to double * add op_patches for softplus * add yaml for ops/yaml/legacy * fix infershape/operator for FLOAT64 * fix * add SoftPlusOpTranscriber * fix * fix * fix1 * fix2 * fix coverage * fix coverage2 * fix (#75605) * [深度对齐] dot (#75717) * fix * fix * fix dcu * [Precision Depth Alignment] paddle.log aligns with torch precision (#75799) * accuracy_stable_log * accuracy_stable_log * fix * fix * fix * fix * fix5 * [Precision Depth Alignment] fix eps of paddle.logit from float to double (#75816) * accuracy_stable_logit * add LogitOpTranscriber * fix coverage * fix 0yaml * [Precision Depth Alignment] paddle.log_sigmoid (#75898) * accuracy_stable_log_sigmoid * fix test_activation_stride_op.py * [Precision Depth Alignment] Modify the negative_slope parameter of the paddle.nn.functional.leaky_relu API to double (#75547) * [big tensor] Paddle/paddle/phi/kernels/funcs gpuBigtensor (#75856) * fix funcs * gpu * fix * fix * 修改PADDLE_ENFORCE信息 * fix cpu error * fix dcu * fix dcu * fix * [Fix] log sigmoid complex (#75953) * feature: Add specialized LogSigmoidFunctor and CudaLogSigmoidFunctor for complex numbers This commit introduces specialized implementations of LogSigmoidFunctor and CudaLogSigmoidFunctor to handle complex number inputs. The new implementations utilize direct formulas for improved accuracy and stability in calculations involving complex types. * refactor: Optimize LogSigmoidFunctor and CudaLogSigmoidFunctor for complex types by caching exp(-x) to reduce redundant computations. This change enhances performance while maintaining accuracy in calculations. * refactor: modified the formula in LogSigmoidFunctor to make it numerical stable --------- Co-authored-by: Zhan Rongrui <46243324+zrr1999@users.noreply.github.com> Co-authored-by: 正在学习 <62892980+cszdrg@users.noreply.github.com> Co-authored-by: Bvicii <98971614+scyyh11@users.noreply.github.com>

accuracy_stable_logit

d3312da

zhengshengning requested review from ForFishes, LiYuRio and zhiqiu as code owners October 13, 2025 09:46

zhengshengning added 2 commits October 13, 2025 12:58

merge develop

29af85e

add LogitOpTranscriber

cbbb9da

fix coverage

686decd

zhengshengning requested review from XiaoguangHu01, changeyoung98, wanghuancoder and zyfncg and removed request for ForFishes, LiYuRio and zhiqiu October 15, 2025 02:55

changeyoung98 reviewed Oct 15, 2025

View reviewed changes

wanghuancoder reviewed Oct 15, 2025

View reviewed changes

fix 0yaml

5edda69

zhengshengning force-pushed the accuracy_stable_logit branch from f352ac2 to 5edda69 Compare October 15, 2025 03:18

Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…

002e07b

… accuracy_stable_logit

changeyoung98 approved these changes Oct 16, 2025

View reviewed changes

wanghuancoder approved these changes Oct 16, 2025

View reviewed changes

XiaoguangHu01 approved these changes Oct 16, 2025

View reviewed changes

zyfncg approved these changes Oct 17, 2025

View reviewed changes

zhengshengning merged commit d2f4afd into PaddlePaddle:develop Oct 17, 2025
106 of 112 checks passed

zhengshengning added a commit to zhengshengning/Paddle that referenced this pull request Oct 24, 2025

[Precision Depth Alignment] fix eps of paddle.logit from float to dou…

fecf083

…ble (PaddlePaddle#75816) * accuracy_stable_logit * add LogitOpTranscriber * fix coverage * fix 0yaml

wanghuancoder mentioned this pull request Oct 24, 2025

[Cherry-pick Fleety_12] Bigtensor and api precision #76023

Closed

zhengshengning added a commit to zhengshengning/Paddle that referenced this pull request Oct 24, 2025

[Precision Depth Alignment] fix eps of paddle.logit from float to dou…

e69958f

…ble (PaddlePaddle#75816) * accuracy_stable_logit * add LogitOpTranscriber * fix coverage * fix 0yaml

zhengshengning mentioned this pull request Oct 24, 2025

[Cherry-pick Fleety_12] Bigtensor and api precision #76028

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[Precision Depth Alignment] fix eps of paddle.logit from float to double #75816

[Precision Depth Alignment] fix eps of paddle.logit from float to double #75816

Uh oh!

zhengshengning commented Oct 13, 2025

Uh oh!

paddle-bot bot commented Oct 13, 2025

Uh oh!

codecov-commenter commented Oct 14, 2025 •

edited

Loading

Uh oh!

changeyoung98 Oct 15, 2025

Uh oh!

wanghuancoder Oct 15, 2025

Uh oh!

zhengshengning Oct 15, 2025

Uh oh!

wanghuancoder left a comment

Uh oh!

XiaoguangHu01 left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

[Precision Depth Alignment] fix eps of paddle.logit from float to double #75816

[Precision Depth Alignment] fix eps of paddle.logit from float to double #75816

Uh oh!

Conversation

zhengshengning commented Oct 13, 2025

PR Category

PR Types

Description

Uh oh!

paddle-bot bot commented Oct 13, 2025

Uh oh!

codecov-commenter commented Oct 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

changeyoung98 Oct 15, 2025

Choose a reason for hiding this comment

Uh oh!

wanghuancoder Oct 15, 2025

Choose a reason for hiding this comment

Uh oh!

zhengshengning Oct 15, 2025

Choose a reason for hiding this comment

Uh oh!

wanghuancoder left a comment

Choose a reason for hiding this comment

Uh oh!

XiaoguangHu01 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

codecov-commenter commented Oct 14, 2025 •

edited

Loading