-
Notifications
You must be signed in to change notification settings - Fork 26
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
Is there a pre-trained model trained on coQA available #79
Comments
Sorry, I haven't released trained CoQA model to public yet |
No Worries, I was able to train the model using your code |
Great! |
what accuracy were you able to achieve? |
Hi! I’m wondering if you’vr released trained CoQa model by now? |
I was able to achieve an F1 score of 87. If you need my help, email me at [email protected] |
I'm trying to train XLNET on COQA using your 'run_coqa' script, but my Google Colab notebooks keep crashing. I tried using both CPU and GPU notebook instances and changing batch_size to 1, but they still keep crashing and showing the: 'Allocation of 16777216 exceeds 10% of system memory.' error. Did you face any similar memory usage errors and did you manage to fix them? |
There is bug in the code which causes an infinite loop which causes crashing. Try to fix that. |
Hmm, do you mean in my code or in the xlnet library?
…On Wed, 9 Sep 2020 at 05:29, abhishek-jha13 ***@***.***> wrote:
There is bug in the code which causes an infinite loop which causes
crashing. Try to fix that.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#79 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AQQZJ3FDV3P3AFNBQMD6TW3SE3ZA3ANCNFSM4KNBDRXQ>
.
|
There is a bug in the run_coqa.py |
I’m really struggling to find it. Could you make a pull-request on this
repo, so that this can be patched?
…On Wed, 9 Sep 2020 at 08:28, abhishek-jha13 ***@***.***> wrote:
There is a bug in the run_coqa.py
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#79 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AQQZJ3HHEGKURBSLUVYJF7DSE4OCLANCNFSM4KNBDRXQ>
.
|
Please feel free to send out a PR for fixing the bug, thanks! |
I want to use pre-trained question answering model trained on coQA like hugging face. Do we have something available?
The text was updated successfully, but these errors were encountered: