[oneDNN] Add 2 new patterns for layernorm fusion #61332
Merged
Conversation
This file contains 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
|
Hi @ezhulenev, Can you please take a look at this PR? Thank you! |
ezhulenev
approved these changes
Aug 24, 2023
google-ml-butler
bot
added
kokoro:force-run
copybara-service
bot
merged commit 28649cd
into
tensorflow:master
15 of 18 checks passed
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Outdated suggestions cannot be applied.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.
co-author: @ustcuna
Following pattern is seen in 3 models. It looks similar to InstanceNorm pattern but it is actually LayerNorm based on the reduction axis. Under right conditions, this pattern will be fused as LayerNorm to improve performance.
With this change we saw ~20% improvement in performance for the 3 models
These are the repo links for 2 of the 3 models
BERT_LARGE : https://github.com/mlperf/training/tree/master/language_model/tensorflow/bert
BERT_BASE : https://github.com/google-research/bert
The other pattern is seen in another customer model and brings in 10-20% improvement
