Skip to content

Question regarding test prediction submission csv file. #2

Description

@ccleveland33

Great work on this repo!

My submissions are no better than 0.5 AUC despite the training & validation going smoothly (~0.86 to 0.88). This seems to imply that the id and prediction are not matching up. However, when I check submission.csv the ids are in the correct order which indicated the predictions are being scambled in some way.

Is there a difference in tensflow/keras flatten or .predict() that is version dependent?
Or is something else going on?

Thanks!
Screenshot 2022-11-29 at 2 09 14 PM

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions