Add parameter broadcasting to PJRT examples.#3836
Merged
Conversation
will-cromar
requested changes
Aug 5, 2022
Collaborator
will-cromar
left a comment
will-cromar
left a comment
There was a problem hiding this comment.
Thanks! Just a few minor suggestions
darisoy
force-pushed
the
pjrt-param-broadcast
branch
2 times, most recently
from
39ca5ed to
9528d15
Compare
darisoy
force-pushed
the
pjrt-param-broadcast
branch
from
9528d15 to
9d1355a
Compare
JackCaoG
approved these changes
Aug 8, 2022
Collaborator
JackCaoG
left a comment
JackCaoG
left a comment
There was a problem hiding this comment.
LGTM. Does pjrt.run_multiprocess supports single core training? In the old xmp.spawn world, we can pass the num_process as 1 and thins still works.
Collaborator
|
That's possible on v4-8. You'll need to pick one chip and set the process bounds (e.g. |
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 the this snippet to
torch_xlato initialize all PJRT process with the same parameters.Update the PJRT examples to use this change and add unit test for the functionality.
The unit test, mnist and imagenet tests have been tested on both v3-8 and v4-8 TPU VMs.