Skip to content

Confusion with result calculations #27

@shaswatpatel123

Description

@shaswatpatel123

While observing few predictions, I came across few examples where the model predicted [0,0,0,0]. In such cases, the micro-average F1-score is 0.75. For such cases how do we calculate FPR(False Positive Rate) and TPR(True Positive Rate).
Suppose correct label is "Unknown" and the predicted outcome is "[]"(basically, the model predicted no outcome). So what do we imply from such cases?

Kindly please reply as soon as possible.
Thank you.

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