Add configuration option to allow users to specify a custom Dataset class #781
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
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Outdated suggestions cannot be applied.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.
solution to #780
Adds an optional field to the data configuration to allow users to specify an arbitrary class to use
as their underlying dataset to be wrapped in the OLMo
IterableDataset. Allows users to also specify their own collation function (assuming that it will output fields that OLMo proper can work with) or to specify mappings from fields in their dataset to fields expected by OLMo's collation function.Unit tests have been added, and integration tests have been run confirming that no currently existing OLMo configurations will need to be changed and that both standard OLMo dataset based training runs and custom dataset runs are able to run to completion and be restarted successfully from checkpoints. For example use, please see the included example config in this PR.
Edit: Updated PR to wrap custom dataset in an
IterableDataset, updated the PR description to reflect this