Add a rule about DLRM training data shuffling #441
Closed
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Outdated suggestions cannot be applied.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.
Shuffling rules about DLRM were not clear enough in the v0.7 round and they left a lot of room for interpretation. This update makes a clear rule that is easy to follow and should not impact convergence or performance of DLRM implementations.
This was actually part of #411, which we discussed, but I mistakenly closed that thinking it was only about packing, which we no longer are using that PR for. This is cleaner to break out data shuffling into its own PR, anyway.