added if statement to account for IterableDatasets doing distributed … #2151
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Outdated suggestions cannot be applied.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.
Motivation
Distributed training currently doesn't work if the dataset is an IterableDataset, due to always specifying the DistributedSampler if the distributed flag is on. IterableDatasets require no Sampler when creating a dataloader over them.
Modification
I added an if statement to the build_dataloader function to check for the case where the dataset is an IterableDataset, and if so, to use no sampler.
BC-breaking (Optional)
I don't believe it should, unless people were doing distributed training with IterableDatasets beforehand and somehow made the DistributedSampler work for them when pytorch itself doesn't support it.
Use cases (Optional)
Specific case of distributed training with IterableDataset
Checklist