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

The arugment order of Rouge score might be wrong. #307

Open
zhaowei-wang-nlp opened this issue Dec 28, 2023 · 0 comments
Open

The arugment order of Rouge score might be wrong. #307

zhaowei-wang-nlp opened this issue Dec 28, 2023 · 0 comments

Comments

@zhaowei-wang-nlp
Copy link

zhaowei-wang-nlp commented Dec 28, 2023

I am reading the code in generate_instruction.py.
If you see the doc of Google's rouge_scorer, the input order of def _score_lcs(target_tokens, prediction_tokens): is target_tokens in the first.
image at https://github.com/google-research/google-research/blob/master/rouge/rouge_scorer.py#L186

In our case, should we take new instructions as the target_tokens? It seems unnatural.
image
https://github.com/tatsu-lab/stanford_alpaca/blob/main/generate_instruction.py#L189

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

1 participant