Make user specify positive_class to BaselineClassifier.predict
#896
Labels
enhancement 💡
New feature or request
Comments
sibre28
changed the title
positive_class to BaselineClassifier.predictpositive_class to BaselineClassifier.predict
|
Why is this needed for |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Is your feature request related to a problem?
The positive class for BaselineClassifier Metrics is always set as the first value of the target column.
Desired solution
Add a non-optional parameter
positive_classtoBaselineClassifier.predictthat forces the user to specify the positive class.Possible alternatives (optional)
Don't always calculate all metrics, let the User specify the metric. This would eliminate the need to define a positive class, when only needing a baseline accuracy.
Screenshots (optional)
No response
Additional Context (optional)
No response
The text was updated successfully, but these errors were encountered: