-
Notifications
You must be signed in to change notification settings - Fork 22
Closed
Description
The default metric name f1 in benchmark.py is not consistent with metric options in the modeler, which should be F1 Macro.
Cardea/cardea/benchmark/benchmark.py
Lines 329 to 330 in e0461e2
| scores = modeler.evaluate(features, target, | |
| tune=optimize, scoring='f1', metrics=metrics, max_evals=10) |
Reactions are currently unavailable
6136
Metadata
Metadata
Assignees
Labels
bugSomething isn't workingSomething isn't working