Skip to content

Training Issue: Zero Loss Values and "NA" Metrics in SEI Model with Pig Data #27

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
zhuofanyan opened this issue Jun 18, 2024 · 0 comments

Comments

@zhuofanyan
Copy link

zhuofanyan commented Jun 18, 2024

Hi
I am training the Sei model with pig data. However, during the training process, I noticed that all the loss values in the output file "selene_sdk.train_model.validation.txt" are consistently zero. Additionally, the metrics "average_precision" and "roc_auc" are reported as "NA" for all samples.

Considering your expertise in developing the SEI model, I am reaching out to seek your insights on the following questions:

  1. Under what circumstances could all loss values be consistently zero during training? Could this indicate any specific issues or errors in the training process?
  2. What could be the potential reasons behind the "NA" values for average_precision and roc_auc? Are there any common pitfalls or prerequisites that I might have overlooked?
    Thanks for your help!
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

1 participant