Fix OOM in contrastive pair generation with streaming approach#627
Open
Wert1996 wants to merge 1 commit intohuggingface:mainfrom
Open
Fix OOM in contrastive pair generation with streaming approach#627Wert1996 wants to merge 1 commit intohuggingface:mainfrom
Wert1996 wants to merge 1 commit intohuggingface:mainfrom
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
This PR fixes out-of-memory issues when training SetFit on large datasets by replacing eager O(n²) pair generation with streaming pair generation.
Problem
Training with contrastive loss (e.g.,
CosineSimilarityLoss) on datasets with a large enough number of samples causes OOM before training even starts. The root cause is three layers of O(n²) memory allocation:shuffle_combinations()creates all pair indices upfrontContrastiveDatasetstores all pairs inpos_pairs/neg_pairslistsDataset.from_list(list(...))materializes the iterator againSolution
sampler.py:shuffle_combinations()with on-the-fly random pair samplingContrastiveDatasetnow stores onlylabel_to_indicesmapping (O(n))__iter__()generates pairs on-the-fly with set-based uniqueness trackingContrastiveDistillationDatasettrainer.py/trainer_distillation.py:IterableDataset.from_generator()instead ofDataset.from_list(list(...))max_stepsautomatically for IterableDataset compatibilityMemory Comparison
Breaking Changes
ContrastiveDataset.pos_pairsandneg_pairsattributes removedContrastiveDataset.len_pos_pairs/len_neg_pairsnow represent target counts, not stored countsestimated_num_pairsproperty for loggingTesting