This repository has been archived by the owner on Dec 16, 2022. It is now read-only.
implement manual distributed sharding for SNLI reader #89
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
dirkgr
approved these changes
Jul 14, 2020
| @@ -45,6 +46,7 @@ def __init__( | |||
| combine_input_fields: Optional[bool] = None, | |||
| **kwargs, | |||
| ) -> None: | |||
| kwargs["manual_distributed_sharding"] = True | |||
| super().__init__(**kwargs) | |||
There was a problem hiding this comment.
I think it would be more readable to call super().__init__(manual_distributed_sharding=True, **kwargs).
| filtered_example_iter = ( | ||
| example for example in example_iter if example["gold_label"] != "-" | ||
| ) | ||
| for example in itertools.islice(filtered_example_iter, start_index, None, step_size): |
There was a problem hiding this comment.
You could do the islice before the json.loads, so that it doesn't have to parse the JSON for lines that get discarded anyways. The problem is that then the filtering has to happen afterwards, which would mean different workers have different number of instances. This is not ideal, but not the end of the world.
I'm just leaving this here as a suggestion.
There was a problem hiding this comment.
I had it that way initially but it didn't make a significant impact on speed.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
This gives a huge speed-up to SNLI/MNLI dataset reading in distributed training.