-
Notifications
You must be signed in to change notification settings - Fork 5.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Need approval] Add AdamW-CPU FP32 JIT assembly kernel #42522
Conversation
你的PR提交成功,感谢你对开源项目的贡献! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@Aganlengzi In @chenwhql In the file |
hi, re-run will be ok. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@Aganlengzi Could you please help with the PR-CI-APPROVAL ?
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR types
New features
PR changes
OPs
Describe
This PR adds AdamW JIT assembly kernel. This feature was requested by #41950
Before in AdamW, all params were computed with Eigen
Paddle/paddle/phi/kernels/funcs/adam_functors.h
Line 426 in 34ac7b7
In this PR AdamW JIT implementation, params are calculated with vector instructions in the same loop where params and momentums are calculated.
Results for VGG training script "test_image_classification.py"
100 batches of 128 images
Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
Average kernel time [ms]