Add an option to cache NNX traversals in PEFT trainer.#928
Merged
copybara-service[bot] merged 1 commit intomainfrom Dec 22, 2025
Merged
Add an option to cache NNX traversals in PEFT trainer.#928copybara-service[bot] merged 1 commit intomainfrom
copybara-service[bot] merged 1 commit intomainfrom
Conversation
copybara-service
bot
requested review from
abheesht17,
hgao327,
jiangyangmu,
lc5211,
sizhit2,
tianshub and
wang2yn84
as code owners
copybara-service
bot
force-pushed
the
test_847406317
branch
from
0e6ada0 to
a7394e3
Compare
copybara-service
bot
force-pushed
the
test_847406317
branch
from
a7394e3 to
4b1bb16
Compare
copybara-service
bot
force-pushed
the
test_847406317
branch
from
4b1bb16 to
07c92a1
Compare
copybara-service
bot
force-pushed
the
test_847406317
branch
from
07c92a1 to
93cce6c
Compare
copybara-service
bot
changed the title
PiperOrigin-RevId: 847827483
copybara-service
bot
force-pushed
the
test_847406317
branch
from
93cce6c to
c948ebe
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add an option to cache NNX traversals in PEFT trainer.
In peft_trainer.py, wrap
train_stepandeval_stepwithnnx.cached_partialto cache NNX graph traversals for performance as documented in https://flax.readthedocs.io/en/latest/guides/performance.html#caching-graph-node-traversalsThis was show to significantly reduce the time spent in Python traversing the graph, which can be important for smaller models (e.g. Gemma3 1B) that may not spend that much accelerator time per step.