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

Vocabulary #10

Closed
antoyang opened this issue Nov 11, 2020 · 1 comment
Closed

Vocabulary #10

antoyang opened this issue Nov 11, 2020 · 1 comment

Comments

@antoyang
Copy link

Hi,
It is written in your AAAI paper "To generate the answer vocabulary, we choose the top 1,000 most frequent answers in the train split as the answer vocabulary, which covers 84.7% / 86.2% / 85.6% of the train/val/test answers, respectively".
However, choosing the same vocabulary, I obtain significantly lower coverage ( about 78% for test and 80% for val). Do you have an idea why? Do you process further the answers?

Best,
Antoine Yang

@datar001
Copy link

datar001 commented Jul 1, 2021

Same questions. I got 84.7% / 82.2% / 78.6% for the train/val/test answers. Did you find the reason? Thx.
微信截图_20210701110508

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