-
Notifications
You must be signed in to change notification settings - Fork 3k
Support tpdp-ep sharding reshard #10568
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
Support tpdp-ep sharding reshard #10568
Conversation
Thanks for your contribution! |
7985e41
to
d447d0c
Compare
Codecov ReportAttention: Patch coverage is
❌ Your patch status has failed because the patch coverage (13.58%) is below the target coverage (80.00%). You can increase the patch coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## incubate/paddlenlp-fleety_20250421 #10568 +/- ##
======================================================================
+ Coverage 48.94% 48.96% +0.02%
======================================================================
Files 768 765 -3
Lines 126279 126281 +2
======================================================================
+ Hits 61806 61833 +27
+ Misses 64473 64448 -25 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
d447d0c
to
c2c0537
Compare
@@ -1230,6 +1230,9 @@ def __post_init__(self): | |||
|
|||
if expert_parallel_degree > 1: | |||
moe_sharding_parallel_degree = world_size // (pipeline_parallel_degree * expert_parallel_degree) | |||
assert ( | |||
self.expert_tensor_parallel_degree <= 1 |
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.
expert_tensor_parallel_degree 现在这几种 ep 的模式有空,最好统一一下,之前 gexiao 合入的时候说这个 expert_tensor_parallel_degree 好像一般不启用。
4213c30
into
PaddlePaddle:incubate/paddlenlp-fleety_20250421
Before submitting
tests
folder. If there are codecov issues, please add tests cases first.PR types
New features
PR changes
APIs
Description
Support tpdp-ep sharding reshard.