Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enable option to train on huggingface datasets. #17

Merged
merged 8 commits into from
Oct 17, 2023
Merged

Conversation

harrykeightley
Copy link
Contributor

Close #11

Features

  • Enable training on hf datasets rather than local datasets
  • Clean the datasets after they've been converted to a DatasetDict or IterableDatasetDict rather than in a preprocessing stage. This allows cleaning of huggingface datasets.
  • Add option not to perform the above cleaning (see do_clean in elpis.models.job.DataArguments)
  • Add a second example script hf_example.py for usage with hf datasets.
  • Allow building Jobs from dictionaries
  • Allow setting a seed.
  • Make parameter names on clean_text more general.
  • Pave the way for streaming datasets from huggingface instead of having to download them outright. Currently broken but will do in the future if needed. Allow for huggingface dataset streaming #16

@harrykeightley harrykeightley added the enhancement New feature or request label Oct 17, 2023
@harrykeightley harrykeightley self-assigned this Oct 17, 2023
@harrykeightley harrykeightley merged commit 577bb4f into main Oct 17, 2023
2 checks passed
@harrykeightley harrykeightley deleted the hf_datasets branch October 17, 2023 07:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Poetry tests in Github Actions fails due to pytorch issues w latest versions.
1 participant