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

New option to TextEncoder to prefix with the column name #1206

Open
GaelVaroquaux opened this issue Dec 25, 2024 · 0 comments
Open

New option to TextEncoder to prefix with the column name #1206

GaelVaroquaux opened this issue Dec 25, 2024 · 0 comments
Labels
enhancement New feature or request

Comments

@GaelVaroquaux
Copy link
Member

GaelVaroquaux commented Dec 25, 2024

Language models use very well the context, and the column name can give useful context. For instance the value "mouse" should probably be embedded differently if the column name is "peripheral" versus "animal".

My suggestion is to have an option to embed entries with "{column_name}: {entry}" to enable exploiting this piece of context.

Of course, ideally we should benchmark this option, to see how / when it brings value.

The benefit would be to disambiguate "State: New York" and "City: New York"

@GaelVaroquaux GaelVaroquaux added the enhancement New feature or request label Dec 25, 2024
@jeromedockes jeromedockes added the good first issue Good for newcomers label Jan 13, 2025
@jeromedockes jeromedockes changed the title New option to table encoder to prefix with the column name New option to TextEncoder to prefix with the column name Jan 13, 2025
@jeromedockes jeromedockes removed the good first issue Good for newcomers label Jan 13, 2025
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

No branches or pull requests

2 participants