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

Label Encoding for fact verifacation over TabFact #12

Closed
vikxoxo opened this issue Jan 27, 2022 · 3 comments
Closed

Label Encoding for fact verifacation over TabFact #12

vikxoxo opened this issue Jan 27, 2022 · 3 comments

Comments

@vikxoxo
Copy link

vikxoxo commented Jan 27, 2022

The Trained Tapex large model (especially availaible at Huggingface) has 0 label corresponding to Entailed class or 0 label corresponding to Refute class?
As by default, the TabFact when loaded through huggingface dataset has 0 corresponding to Refute class.
Please tell us this.

@SivilTaram
Copy link
Collaborator

@vikxoxo Thanks for your interest on our work! 0 corresponds to Refute and 1 corresponds to Entailed, following the same with TabFact.

@SivilTaram
Copy link
Collaborator

@vikxoxo Sorry for the previous message. I have just realized that the output is reversed, i.e., the current output of tapex-large-finetuned-tabfact means that 0 corresponds to Entailed.

@SivilTaram
Copy link
Collaborator

@vikxoxo Hi, I have created the official tapex-large-finetuned-tabafct under the microsoft namespace and its label space is correct! Try it if you're interested. :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants