Migrates PoS tagging infrastructure to a torch.utils.data.Dataset and implements PUNCT augmentation #1303
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
… implements PUNCT augmentation In order to ensure long-term ability to augment the PoS tagging data against various types of augmentation (so far, punctuation removal on some sentences), and in order to make the DataLoader more efficient and compatible with current PyTorch semantics, we rename models.pos.data.DataLoader to models.pos.data.Dataset which has a to_loader() function that yields a torch.utils.data.DataLoader. The data in this loader can dynamically be augmented per epoch (by changing the getitem function to perform any augmentation with a certain chance each time getitem is called), because the PyTorch DataLoader lazily fetches data only when it is accessed. batch_size is updated from counting words to counting sentences, since the torch DataLoader counts one sentence per __getitem__ A few tests are added to make sure the augmentation happens 0%, 100%, or a reasonable number of times
…__. This cuts down on the number of copies needed to make the collated batch
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Outdated suggestions cannot be applied.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.
In order to ensure long-term ability to augment the PoS tagging data against various types of augmentation (so far, punctuation removal on some sentences), and in order to make the DataLoader more efficient and compatible with current PyTorch semantics, we rename models.pos.data.DataLoader to models.pos.data.Dataset which has a to_loader() function that yields a torch.utils.data.DataLoader.
The data in this loader can dynamically be augmented per epoch (by changing the getitem function to perform any augmentation with a certain chance each time getitem is called), because the PyTorch DataLoader lazily fetches data only when it is accessed.
Debugged & squashed version of #1296