avoid "divided by zero error" in logging_outputs when --use-bmuf is e… #812
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
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@nayansinghal has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
Contributor
|
@nayansinghal merged this pull request in b3864b2. |
yzpang
pushed a commit
to yzpang/gold-off-policy-text-gen-iclr21
that referenced
this pull request
Feb 19, 2021
…… (#812) Summary: … enabled. When doing multi-gpu training with --use-bmuf turned on and --global-sync-iter > 1, each replica may not sync with other replicas at each iteration. So logging_outputs only has stats of their own. On the other hand, logging_outputs may be empty at the end of an epoch after "a dummy iteration" because the number of replicas does not divide the number of batches of the training data. If this happens, sample_size and ntokens would be 0 for some replica and cause "divided by 0" error. This fix sets *loss to 0 if sample_size/ntokens is 0. Pull Request resolved: facebookresearch/fairseq#812 Reviewed By: myleott, yqwangustc Differential Revision: D15908614 Pulled By: nayansinghal fbshipit-source-id: c92e8e095f012bdb4ef753a3c627fd215afa215d
yzpang
pushed a commit
to yzpang/gold-off-policy-text-gen-iclr21
that referenced
this pull request
Feb 19, 2021
…… (#812) Summary: … enabled. When doing multi-gpu training with --use-bmuf turned on and --global-sync-iter > 1, each replica may not sync with other replicas at each iteration. So logging_outputs only has stats of their own. On the other hand, logging_outputs may be empty at the end of an epoch after "a dummy iteration" because the number of replicas does not divide the number of batches of the training data. If this happens, sample_size and ntokens would be 0 for some replica and cause "divided by 0" error. This fix sets *loss to 0 if sample_size/ntokens is 0. Pull Request resolved: facebookresearch/fairseq#812 Reviewed By: myleott, yqwangustc Differential Revision: D15908614 Pulled By: nayansinghal fbshipit-source-id: c92e8e095f012bdb4ef753a3c627fd215afa215d
yfyeung
pushed a commit
to yfyeung/fairseq
that referenced
this pull request
Dec 6, 2023
* add README to docs * update documents for distillation * upload png files
Harleen8118
pushed a commit
to Harleen8118/IBERT
that referenced
this pull request
Jun 26, 2025
…… (#812) Summary: … enabled. When doing multi-gpu training with --use-bmuf turned on and --global-sync-iter > 1, each replica may not sync with other replicas at each iteration. So logging_outputs only has stats of their own. On the other hand, logging_outputs may be empty at the end of an epoch after "a dummy iteration" because the number of replicas does not divide the number of batches of the training data. If this happens, sample_size and ntokens would be 0 for some replica and cause "divided by 0" error. This fix sets *loss to 0 if sample_size/ntokens is 0. Pull Request resolved: facebookresearch/fairseq#812 Reviewed By: myleott, yqwangustc Differential Revision: D15908614 Pulled By: nayansinghal fbshipit-source-id: c92e8e095f012bdb4ef753a3c627fd215afa215d
caltia
pushed a commit
to caltia/fairseq
that referenced
this pull request
Jul 8, 2025
…… (#812) Summary: … enabled. When doing multi-gpu training with --use-bmuf turned on and --global-sync-iter > 1, each replica may not sync with other replicas at each iteration. So logging_outputs only has stats of their own. On the other hand, logging_outputs may be empty at the end of an epoch after "a dummy iteration" because the number of replicas does not divide the number of batches of the training data. If this happens, sample_size and ntokens would be 0 for some replica and cause "divided by 0" error. This fix sets *loss to 0 if sample_size/ntokens is 0. Pull Request resolved: facebookresearch/fairseq#812 Reviewed By: myleott, yqwangustc Differential Revision: D15908614 Pulled By: nayansinghal fbshipit-source-id: c92e8e095f012bdb4ef753a3c627fd215afa215d
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
… enabled.
When doing multi-gpu training with --use-bmuf turned on and --global-sync-iter > 1, each replica may not sync with other replicas at each iteration. So logging_outputs only has stats of their own. On the other hand, logging_outputs may be empty at the end of an epoch after "a dummy iteration" because the number of replicas does not divide the number of batches of the training data. If this happens, sample_size and ntokens would be 0 for some replica and cause "divided by 0" error. This fix sets *loss to 0 if sample_size/ntokens is 0.