-
-
Notifications
You must be signed in to change notification settings - Fork 2.1k
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
[Bug]: Custom trained model not correctly loaded when using FlairEmbeddings #3459
Labels
bug
Something isn't working
Comments
See here the relevant logs:
|
Hi @petermartens1992 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
The F-scores shown after training a SequenceTagger model with FlairEmbeddings are way higher than when loading the model from file and doing an evaluation on it. Also the predictions from a loaded model seem way off compared to the initial F-scores.
The issue doesn't appear when initializing a new model and only loading the state from file:
Also, when removing the FlairEmbeddings the F-scores between training and evaluation correctly align again.
See here the code to reproduce the issue (using flair 0.13.1):
To Reproduce
Expected behavior
.
Logs and Stack traces
No response
Screenshots
No response
Additional Context
No response
Environment
.
The text was updated successfully, but these errors were encountered: