You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello! I was looking into the issue and was considering working on a fix. It seems like the solution would involve checking if the URL is a parquet file and, if so, adding logic to convert it into json/csv/txt before proceeding with the regular processing. Before I begin, I wanted to check if it’s alright to take this on or if there’s already a plan to address it. Let me know your thoughts!
Hi @MoolmanM ! Thanks for expressing interest. You can absolutely get started on this. I would recommend checking out the pyrit.datasets module. We have functions in there to pull in datasets from Hugging Face. This particular one is also available on HuggingFace so you don't need to do anything special, just call HF. The not so straightforward part is deciding which columns to use for our SeedPrompt objects. I'm happy to point you in the right direction if you want help, but you can also give it a try first. LMK
Is your feature request related to a problem? Please describe.
This is brand new and not yet added:
https://svannie678.github.io/svannie678-red_team_repo_social_bias/
Describe the solution you'd like
Similar fetch function like all other datasets.
Describe alternatives you've considered, if relevant
Additional context
The text was updated successfully, but these errors were encountered: