add get_fused_optimizers #558
Closed
Conversation
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
Summary: Experimental API for now Differential Revision: https://www.internalfb.com/diff/D38190544?entry_point=27 fbshipit-source-id: 350c00202a4da2516e927539ca445acf5e5493ff
Summary: Replaces the nn.Module APIs in ShardedEmbeddingBagCollection, and replaces it with the torch.nn.Module() representation as proposed in meta-pytorch#528 We're putting these modules/tests in a new modular_* path as per https://docs.google.com/document/d/15r5XxMTeVC90kA0u8t006-KLLeh9MZmL9DqX_F4VLxU/edit# We also enforce the separation of sharding and compute_kernel/fused_params at the sharder level (but still keeping it at the ShardedEmbeddingBagcollection for simplicity) A couple of differences: 1. named_parameters() returns ShardedTensors 2. data_parallel tensors have DDP wrapped around the tables Differential Revision: https://www.internalfb.com/diff/D38190302?entry_point=27 fbshipit-source-id: 7060968324dd9024676d66e7d233c5005f5ac915
Summary: Since we no longer rely on DistributedModelParallel (for composability piece), we need an alternative way of getting the fused optimizer. get_fused_optimizer implements this, logically it's the same as the recursive call in DMP Differential Revision: D37505229 fbshipit-source-id: a158eaba34638130e7a312aeda620c4fc7382d0c
facebook-github-bot
added
CLA Signed
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D37505229 |
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Outdated suggestions cannot be applied.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.
Summary:
Since we no longer rely on DistributedModelParallel (for composability piece), we need an alternative way of getting the fused optimizer.
get_fused_optimizer implements this, logically it's the same as the recursive call in DMP
Differential Revision: D37505229