Skip to content

Fine-tune from checkpoint #10

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

Open
sandhiya18 opened this issue Feb 16, 2019 · 2 comments
Open

Fine-tune from checkpoint #10

sandhiya18 opened this issue Feb 16, 2019 · 2 comments

Comments

@sandhiya18
Copy link

Is it possible to do a transfer learning on a new similar QA dataset by using the checkpoint from Link as an initialization point?

@unilight
Copy link
Owner

Hi @sandhiya18,

Technically, yes. But notice that the form of the answers in SQuAD is that the answer is given as a span of the context, so you may only test on other datasets with the same form. In addition, some data preprocessing is requires. You need to look at preprocessing.py to see how I process SQuAD and probably preform the same procedure on the dataset you wish to test on.

@sandhiya18
Copy link
Author

@unilight Thanks for your response. How do I resume my training from the checkpoints that is provided in the link?

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