Skip to content

Conversation

VanyaBK
Copy link
Collaborator

@VanyaBK VanyaBK commented Dec 3, 2023

Description

Added explanation column to the dataset generation and modified the prompt templates to include explanations

References

Chain-of-thought prompting

Blocked by

  • NA

@VanyaBK VanyaBK requested review from neubig and viswavi December 3, 2023 02:57
Copy link
Collaborator

@neubig neubig left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for contributing this!

I think that using explanations should probably be optional though. It's not the same as what we did in the prompt2model paper, and I'm not convinced that it will be useful in all cases. I know this is a pretty big change to this PR (and I don't suggest this lightly), but could you please consider adding a flag that makes it possible to either use or not use explanations? Thank you!

"""An example from a dataset, containing input, explanation and output columns."""

input_col: str
explain_col: str
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would suggest that we just call this explanation_col to be consistent. Also, it's not clear what an explanation is. I'd suggest that you expand the docstring and describe what all the variables are.

Suggested change
explain_col: str
explanation_col: str

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

Successfully merging this pull request may close these issues.

2 participants