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

fix(xlnet): Set training mode to False and set dropout to zero #47

Closed
wants to merge 3 commits into from

Conversation

chiragjn
Copy link
Collaborator

@chiragjn chiragjn commented May 6, 2020

Attempts to fix #45

Taken directly from zihangdai/xlnet@d1d0ff4

Todo:

  • Test manually
  • Add a test for C/I if possible

@amansrivastava17
Copy link
Owner

@chiragjn LGTM, should we merge this?

@chiragjn
Copy link
Collaborator Author

chiragjn commented Aug 8, 2020

unfortunately, this still does not solve the problem, without going component by component it is hard to tell what is still causing the variance :(

@chiragjn chiragjn closed this Feb 28, 2022
@chiragjn chiragjn reopened this Feb 28, 2022
@chiragjn
Copy link
Collaborator Author

Closing this for now, this still remains a issue 🙃

@chiragjn chiragjn closed this Feb 28, 2022
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.

XLNet generates random word embeddings
2 participants