hidden bcast_params call in dist train #11575
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
typhoonzero
reviewed
Jun 20, 2018
| @@ -131,6 +132,11 @@ def __init__(self, | |||
| main = main_program | |||
| main = main if main else framework.default_main_program() | |||
| scope = executor.global_scope() | |||
| # FIXME(Yancey1989): it's a temporary approach to determinate the distribute | |||
| # train program, call self.bcast_param() at the end of each mini-batch. | |||
| self.is_dist = True if "recv" in [ | |||
There was a problem hiding this comment.
Can use num_trainers to test if it's distributed training
There was a problem hiding this comment.
The num_trainers looks a bit suspicious now. Is it only for nccl? I found some pserver training script working without setting num_trainers...
There was a problem hiding this comment.
It's only a temporary fix, we can insert bcast after recv op in PrallelExecuotor.
There was a problem hiding this comment.
And i think @panyx0718 is right, pserver update does not set num_trainers.
…bcast_params_call
panyx0718
approved these changes
Jun 20, 2018
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Outdated suggestions cannot be applied.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.
Fixed #11572