Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add a pass to fuse fc+elementwise_add+layernorm #19776

Merged
merged 10 commits into from
Sep 19, 2019
Merged

Add a pass to fuse fc+elementwise_add+layernorm #19776

merged 10 commits into from
Sep 19, 2019

Conversation

Xreki
Copy link
Contributor

@Xreki Xreki commented Sep 12, 2019 •
edited

Example 1

  • Before
    image
  • After
    image

Example 2

  • Before
    image
  • After
    image

@Xreki Xreki mentioned this pull request Sep 12, 2019
Open
@Xreki Xreki closed this Sep 17, 2019
@Xreki Xreki reopened this Sep 17, 2019
zhaoyuchen2018
zhaoyuchen2018 approved these changes Sep 18, 2019
View reviewed changes
Copy link
Contributor

@zhaoyuchen2018 zhaoyuchen2018 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@Xreki Xreki merged commit 3cd985a into PaddlePaddle:develop Sep 19, 2019
mapingshuo pushed a commit to mapingshuo/Paddle that referenced this pull request Sep 20, 2019
@Xreki @mapingshuo
Add a pass to fuse fc+elementwise_add+layernorm (PaddlePaddle#19776)
16ed39d 
* Add fc_elementwise_layernorm_fuse pass and unittest.

* Add fused_fc_elementwise_layernorm op and its GPU kernel.
test=develop

* Apply fc_elementwise_layernorm_fuse_pass to GPU inference.

* Add the setting of attrs in the definition of binary_op.
test=develop

* Add comment.

* Implement the unittest.
test=develop

* Change the unittest name of layer_norm.
test=develop
@Xreki Xreki deleted the pass_fuse_fc_elementwise_layernorm branch October 29, 2019 00:38
seiriosPlus pushed a commit to seiriosPlus/Paddle that referenced this pull request Dec 9, 2019
@Xreki
Add a pass to fuse fc+elementwise_add+layernorm (PaddlePaddle#19776)
efa6e0e 
* Add fc_elementwise_layernorm_fuse pass and unittest.

* Add fused_fc_elementwise_layernorm op and its GPU kernel.
test=develop

* Apply fc_elementwise_layernorm_fuse_pass to GPU inference.

* Add the setting of attrs in the definition of binary_op.
test=develop

* Add comment.

* Implement the unittest.
test=develop

* Change the unittest name of layer_norm.
test=develop
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@zhaoyuchen2018 zhaoyuchen2018 zhaoyuchen2018 approved these changes

@tensor-tang tensor-tang Awaiting requested review from tensor-tang

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
2 participants
@Xreki @zhaoyuchen2018