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.

Sign up for GitHub

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

Jump to bottom

Thoughts on Hits@1 #12

Open
bkj opened this issue Apr 17, 2019 · 0 comments
Open

Thoughts on Hits@1 #12

bkj opened this issue Apr 17, 2019 · 0 comments

Comments

@bkj
Copy link

bkj commented Apr 17, 2019

Thanks for releasing the code!

I'm wondering whether you have any thoughts on why your Hits@1 is substantially lower than many of the other teams in the ConvAI2 competition, while the scores from the human evaluation are the highest overall.

From the slides you presented, it looks like you didn't actually optimize a classification objective (just PPL, smoothed PPL, and an F1 risk objective). Do you think adding a classification objective might bump your Hits@1 up significantly?
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@bkj