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

Key hyperparameters to reproduce results? #96

Open
VickiCui opened this issue Jul 1, 2022 · 3 comments
Open

Key hyperparameters to reproduce results? #96

VickiCui opened this issue Jul 1, 2022 · 3 comments

Comments

@VickiCui
Copy link

VickiCui commented Jul 1, 2022

Thanks for your great work!
I'm having some trouble when reproducing the results. Although trained as in the tutorial, there is still a big gap between the model I trained and the results given in the paper. Can you provide the necessary hyperparameters in order to reproduce the results? Such as batch size, epoch, learning rate, beam size, and so on. Thx!!!

@tuetschek
Copy link
Collaborator

@VickiCui I don't think this is the correct place – the repo here is just for the metrics scripts, it doesn't include the baseline submissions. Perhaps @sebastianGehrmann or @yjernite would know more on the baselines?

@VickiCui
Copy link
Author

VickiCui commented Jul 11, 2022 via email

@sebastianGehrmann
Copy link
Contributor

Which task are we talking about? Most of the models were trained for https://openreview.net/forum?id=CSi1eu_2q96 and the hyperparameters from that paper should replicate results.

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

3 participants