-
Notifications
You must be signed in to change notification settings - Fork 3k
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
Add Pipeline Parallel for PPO training and support generation with InferenceModel #7953
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #7953 +/- ##
===========================================
+ Coverage 54.29% 54.41% +0.12%
===========================================
Files 617 632 +15
Lines 96339 99476 +3137
===========================================
+ Hits 52310 54134 +1824
- Misses 44029 45342 +1313 ☔ View full report in Codecov by Sentry. |
… from main from_pratrined.
PPP 4d/support uc
Add ppo 4d test
… ppo-4d-reshard
…ain/eval to align.
# and sampling, and then broadcast to avoid broadcast logits. | ||
if hasattr(self, "pp_group"): | ||
paddle.distributed.broadcast( | ||
next_tokens, src=self.pp_group.ranks[0], group=self.pp_group # use rank 0 for same seed to check |
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.
我看CI这里,self.pp_group有可能为None,导致.ranks[0]报错
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
Others
Description
Add Pipeline Parallel for PPO training and support generation with InferenceModel